Wine Geology Soil

4Geology and Soil:

effects on wine quality

Cra viticoltura_libro 1_capitolo 4.indd 1 03/06/10 15:51

TERROIR, CLIMAT ET SOL

Jacques FANET Syndicat Coteaux du Languedoc

3 chemin des Combes d’Arlenques 34 800 ASPIRAN, FRANCE.

Email : [email protected] RESUME Le sol et le climat occupent une place prépondérante dans le concept de terroir, pour lequel l’OIV s’apprête à adopter une définition internationale. Les travaux de recherche qui ont été menés depuis une trentaine d’années sur ces thèmes et qui ont été, pour les plus importants, présentés dans les 7 premiers Congrès Internationaux des Terroirs Viticoles ont considérablement modifié les connaissances sur le fonctionnement des terroirs viticoles dans le monde et le comportement des consommateurs avertis par rapport aux vins de terroirs. Ce Congrès pourrait être l’occasion de réfléchir à de nouvelles orientations en matière de recherche sur ces thèmes. Notamment, l’élargissement de l’étude des terroirs à d’autres disciplines pourraient être étudiées, en particulier la microbiologie pour l’étude des sols et les mesures à prendre pour s’adapter au changement climatique dans les zones viticoles traditionnelles MOTS-CLES Terroir, sol, climat, nouvelles orientations, changement climatique, adaptation. INTRODUCTION A l’initiative du groupe de travail « Environnement Viticole et changement climatique » et des colloques internationaux sur les terroirs viticoles et notamment celui de Bordeaux-Montpellier qui s’est tenu en 2006, l’Organisation Internationale de la Vigne et du Vin est sur le point d’approuver une « définition internationale du terroir vitivinicole » dont le texte serait le suivant : « Le « terroir » vitivinicole est un concept qui se réfère à un espace sur lequel se développe un savoir collectif des interactions entre un milieu physique et biologique identifiable et les pratiques vitivinicoles appliquées, qui confèrent des caractéristiques distinctives aux produits originaires de cet espace. Le « terroir » inclut des caractéristiques spécifiques du sol, de la topographie, du climat, du paysage et de la biodiversité. » Cette définition fait apparaître des éléments clefs qui sont les suivants :

- un espace géographique - un savoir collectif des interactions entre des facteurs naturels et des facteurs humains qui a

été progressivement et plus ou moins rapidement acquis sur cet espace - un produit vitivinicole qui, sur cet espace présente des caractéristiques distinctives dues à ces

interactions. Dans son dernier alinéa, cette définition précise quelques éléments, non exhaustifs, qui contribuent à caractériser un terroir :

- des éléments strictement relatifs aux facteurs naturels (sol topographie, climat). - des éléments où les facteurs humains prennent une place plus ou moins importante (paysage,

biodiversité)


TERROIR, CLIMAT ET SOL

Jacques FANET Syndicat Coteaux du Languedoc

3 chemin des Combes d’Arlenques 34 800 ASPIRAN, FRANCE.

Email : [email protected] RESUME Le sol et le climat occupent une place prépondérante dans le concept de terroir, pour lequel l’OIV s’apprête à adopter une définition internationale. Les travaux de recherche qui ont été menés depuis une trentaine d’années sur ces thèmes et qui ont été, pour les plus importants, présentés dans les 7 premiers Congrès Internationaux des Terroirs Viticoles ont considérablement modifié les connaissances sur le fonctionnement des terroirs viticoles dans le monde et le comportement des consommateurs avertis par rapport aux vins de terroirs. Ce Congrès pourrait être l’occasion de réfléchir à de nouvelles orientations en matière de recherche sur ces thèmes. Notamment, l’élargissement de l’étude des terroirs à d’autres disciplines pourraient être étudiées, en particulier la microbiologie pour l’étude des sols et les mesures à prendre pour s’adapter au changement climatique dans les zones viticoles traditionnelles MOTS-CLES Terroir, sol, climat, nouvelles orientations, changement climatique, adaptation. INTRODUCTION A l’initiative du groupe de travail « Environnement Viticole et changement climatique » et des colloques internationaux sur les terroirs viticoles et notamment celui de Bordeaux-Montpellier qui s’est tenu en 2006, l’Organisation Internationale de la Vigne et du Vin est sur le point d’approuver une « définition internationale du terroir vitivinicole » dont le texte serait le suivant : « Le « terroir » vitivinicole est un concept qui se réfère à un espace sur lequel se développe un savoir collectif des interactions entre un milieu physique et biologique identifiable et les pratiques vitivinicoles appliquées, qui confèrent des caractéristiques distinctives aux produits originaires de cet espace. Le « terroir » inclut des caractéristiques spécifiques du sol, de la topographie, du climat, du paysage et de la biodiversité. » Cette définition fait apparaître des éléments clefs qui sont les suivants :

- un espace géographique - un savoir collectif des interactions entre des facteurs naturels et des facteurs humains qui a

été progressivement et plus ou moins rapidement acquis sur cet espace - un produit vitivinicole qui, sur cet espace présente des caractéristiques distinctives dues à ces

interactions. Dans son dernier alinéa, cette définition précise quelques éléments, non exhaustifs, qui contribuent à caractériser un terroir :

- des éléments strictement relatifs aux facteurs naturels (sol topographie, climat). - des éléments où les facteurs humains prennent une place plus ou moins importante (paysage,

biodiversité)

4 - 3

VIII INTERNATIONAL TERROIR CONGRESS


Comme on peut le constater à la lecture de cette définition, le sol et le climat qui font l’objet de cette deuxième journée (session 3 et 4) occupent une place importante dans la notion de terroir. Pour s’en convaincre, il suffit de reprendre la liste des quelques 500 communications présentées au cours des 7 précédents Congrès Internationaux des Terroirs Viticoles. Près des trois-quarts des communications traitent de sujets relatifs au sol, au climat ou aux deux à la fois. On peut faire le même constat en observant les présentations qui sont faites aujourd’hui de pratiquement toutes les régions viticoles dans le monde. Que les documents concernés soient à caractère technique, commercial, publicitaire ou touristique, la description d’une région viticole commence invariablement par la présentation de la géologie, des sols et du climat du lieu. LE ROLE DU SOL Ce sont les pays européens et notamment la France qui ont lancé depuis près d’un siècle cette mise en avant des caractères du milieu physique avec une certaine préférence pour le rôle des sols en tant que responsable des caractéristiques organoleptiques spécifiques et de la typicité des vins d’Appellation d’Origine Contrôlées. Jusque dans les années 70 – 80, sans qu’aucune démonstration scientifique ne vienne étayer ou démontrer ces écrits, on n’hésitait pas à affirmer que le goût ou les caractéristiques gustatives de tel ou tel vin était à coup sûr, du à telle ou telle caractéristique physique ou chimique du sol. Ne disait-on pas que le caractère si particulier des vins de Pomerol étaient du à la « crasse de fer », concrétions de fer et de manganèse, formés au contact des molasses et des alluvions fluviatiles du confluent de l’Isle et de la Dordogne. Ne disait-on pas encore que tel vin blanc avait un caractère iodé du à des petites huitres fossiles présentes dans le sol alors que le vignoble se trouve très éloigné des côtes atlantiques ou méditerranéennes. On était beaucoup plus proche de la poésie que de la science, mais sur le plan marketing, il s’agissait là de messages qui ont été longtemps très porteurs. Il suffisait que le présentateur des vins ait une élocution relativement aisée et le consommateur peu averti était persuadé d’avoir reçu un message à caractère scientifique indiscutable. Cette importance supposée du sol et de la géologie a même parfois poussé les hommes au-delà des limites du raisonnable. Les vignobles de l’Appellation Chablis s’étendent presqu’exclusivement sur les formations jurassiques du portlandien et du kimméridgien, cette dernière supportant les grands crus de Chablis. En 1904, le professeur Georges Chappaz enseignait : « L’étude des terres du vignoble de Chablis est en quelque sorte l’étude des sols et sous-sols d’origine kimméridgienne ». S’appuyant sur cette affirmation, les viticulteurs du secteur des grands crus feront réserver, par l’INAO en 1938, l’appellation Chablis aux seuls sols sur kimméridgien. Excluant ainsi les vignobles implantés sur portlandien, ils se débarrassaient d’une certaine concurrence locale, ce qui, bien entendu, n’était pas du goût des viticulteurs exclus. Il fallut près de quarante ans et l’intervention du Ministre de l’Agriculture français pour faire cesser cette guerre fratricide. Châteauneuf du Pape a vécu une aventure similaire. Les sols mythiques de Châteauneuf sont les fameux galets roulés des alluvions fluviatiles que le Rhône a arrachés aux Alpes en surrection tout au long de son cours et déposés là dans des parties plus planes de sa basse vallée. Dans l’époque la plus récente de l’ère quaternaire, ces alluvions ont été protégées des crues destructrices du fleuve par une grosse colline de calcaire urgonien qui émergeait au milieu de la vallée du Rhône. Lorsque les sols de galets roulés furent entièrement plantés de vignes, certains vignerons étendirent dans les années 70 leurs plantations sur les calcaires urgoniens qui ont, il faut le reconnaître, un régime hydrique tout à fait différent de celui des terrasses quaternaires caillouteuses. Il s’agissait là d’un véritable sacrilège qui, là encore, divisa les hommes du terroir dans une guerre dont les cicatrices sont encore sensibles. Mais, que l’on ne s’y trompe pas. Ces batailles internes à certains terroirs ne sont pas le lot de la seule vieille Europe. Car en Australie, également, les hommes se sont battus autour du rôle du sol dans la définition d’une zone viticole. Le Coonawarra, petite bande de terra rossa en forme de cigare de 15 kilomètres de long sur 3 de large, a fait l’objet de combats juridiques farouches entre les partisans de

l’utilisation du nom « Coonawarra » par les seuls viticulteurs installés sur la terra rossa et ceux qui, bien qu’implantés sur les terres noires profondes qui entourent le « cigare », avaient depuis des décades des usages d’utilisation du nom de Coonawarra. Il ne fallut pas moins d’une dizaine d’années de débats acharnés devant les tribunaux australiens avant d’aboutir à un accord sur les limites de la zone à l’intérieur de laquelle les vignerons auront désormais l’exclusivité de l’utilisation du nom de Coonawara pour designer les vins qu’ils y produisent. Les années 70-80 marquèrent un virage décisif dans cette connaissance du rôle du sol dans les terroirs viticoles et dans la relation plus ou moins directe entre les caractéristiques du milieu physique et les qualités organoleptiques des vins qui y sont produits. Ce furent d’abord les travaux du professeur Gérard Seguin à Bordeaux qui seront poursuivis et prolongés par ceux de l’équipe de Kees Van Leeuwen. En essayant de comprendre ce qui pouvait expliquer le niveau de qualité des grands crus de la région de Bordeaux, ces travaux ont montré le rôle fondamental de l’alimentation hydrique dans le fonctionnement de la plante dans des situations géo-pédologiques aussi diverses que les terrasses quaternaires caillouteuses du Haut-Médoc, les calcaires à astéries de Saint Emilion ou les argiles gonflantes (montmorillonites) de Petrus. Quelques années plus tard ont aboutis les travaux de l’équipe INRA d’Angers dirigés par Christian Asselin et René Morlat sur le comportement de deux des cépages caractéristiques des terroirs de l’Anjou, dans la vallée de la Loire, le chenin dans les coteaux du Layon et le cabernet franc à Saumur Champigny, Bourgueil et Chinon. Ils confirmèrent les travaux de Gérard Seguin en apportant de nouveaux éléments complémentaires dans la compréhension du rôle du sol dans l’élaboration des grands vins comme la capacité des sols à se réchauffer, par exemple. En mettant à mal certaines affirmations ou croyances largement répandues dans les vignobles ou les rédactions de journaux spécialisés dans le domaine du vin, ces travaux ont incontestablement ouvert une nouvelle ère, celle de la connaissance scientifique de la relation sol-climat-vin.et en conséquence d’une nouvelle approche du vin par le consommateur. C’est d’ailleurs ainsi qu’a été mis en place le premier des Congrès Internationaux des Terroirs Viticoles, qui s‘est tenu à Angers en 1996. Il s’agissait pour l’Institut National de la Recherche Agronomique de communiquer précisément sur ces tout nouveaux résultats scientifiques avec l’appui de l’Institut National des Appellations d’Origine et de l’Office National des Vins français. Ce lancement s’est fait dans un contexte de grande méfiance notamment d’une partie du nouveau monde qui ne voulait pas entendre parler de ce mot de terroir. Après une phase strictement européenne, nos collègues sud-africains ont ouvert la voie d’un élargissement de cette notion à l’ensemble du monde et on ne peut que se réjouir de voir ce Congrès accueilli par nos collègues américains en 2012. Ce Congrès d’Angers fut le point de départ d’une formidable série de travaux dans le monde entier qui ont beaucoup aidé à la connaissance et au développement de ces produits de terroir sur les cinq continents. Ce fut notamment le cas sur la connaissance du rôle du sol qui a souvent été contesté, dans le nouveau monde par les grands industriels du vin ou une certaine presse internationale, mais aussi au sein de pays comme la France ou il existe une école de géographes du vin très forte avec comme chef de file Roger Dion qui, dans son remarquable ouvrage « Histoire de la vigne et du vin des origines au 19ème siècle », avait fortement contesté le rôle du sous-sol et du sol dans la localisation des vignobles français. Il attribuait en effet cette localisation à des facteurs totalement différents comme les grands courants de commercialisation et les routes commerciales, les vignobles étant principalement installés le long des fleuves Loire , Seine, Rhône, etc… parce qu’ils représentaient de voies de circulation très pratiques, ou autour des grands ports d’expédition des vins comme Bordeaux ou Porto. C’est ici que les travaux très complets qui ont été présentés au cours de nos 8 congrès internationaux prennent toute leur valeur. Ils ont largement contribué à modifier ces idées et les comportements des viticulteurs dans tout le monde viticole et à améliorer la connaissance des consommateurs sur les origines de la qualité des grands vins partout dans le monde. En tant que participant de la première heure à la mise en place de ce système de diffusion des connaissances scientifiques avec la complicité

4 - 4



Comme on peut le constater à la lecture de cette définition, le sol et le climat qui font l’objet de cette deuxième journée (session 3 et 4) occupent une place importante dans la notion de terroir. Pour s’en convaincre, il suffit de reprendre la liste des quelques 500 communications présentées au cours des 7 précédents Congrès Internationaux des Terroirs Viticoles. Près des trois-quarts des communications traitent de sujets relatifs au sol, au climat ou aux deux à la fois. On peut faire le même constat en observant les présentations qui sont faites aujourd’hui de pratiquement toutes les régions viticoles dans le monde. Que les documents concernés soient à caractère technique, commercial, publicitaire ou touristique, la description d’une région viticole commence invariablement par la présentation de la géologie, des sols et du climat du lieu. LE ROLE DU SOL Ce sont les pays européens et notamment la France qui ont lancé depuis près d’un siècle cette mise en avant des caractères du milieu physique avec une certaine préférence pour le rôle des sols en tant que responsable des caractéristiques organoleptiques spécifiques et de la typicité des vins d’Appellation d’Origine Contrôlées. Jusque dans les années 70 – 80, sans qu’aucune démonstration scientifique ne vienne étayer ou démontrer ces écrits, on n’hésitait pas à affirmer que le goût ou les caractéristiques gustatives de tel ou tel vin était à coup sûr, du à telle ou telle caractéristique physique ou chimique du sol. Ne disait-on pas que le caractère si particulier des vins de Pomerol étaient du à la « crasse de fer », concrétions de fer et de manganèse, formés au contact des molasses et des alluvions fluviatiles du confluent de l’Isle et de la Dordogne. Ne disait-on pas encore que tel vin blanc avait un caractère iodé du à des petites huitres fossiles présentes dans le sol alors que le vignoble se trouve très éloigné des côtes atlantiques ou méditerranéennes. On était beaucoup plus proche de la poésie que de la science, mais sur le plan marketing, il s’agissait là de messages qui ont été longtemps très porteurs. Il suffisait que le présentateur des vins ait une élocution relativement aisée et le consommateur peu averti était persuadé d’avoir reçu un message à caractère scientifique indiscutable. Cette importance supposée du sol et de la géologie a même parfois poussé les hommes au-delà des limites du raisonnable. Les vignobles de l’Appellation Chablis s’étendent presqu’exclusivement sur les formations jurassiques du portlandien et du kimméridgien, cette dernière supportant les grands crus de Chablis. En 1904, le professeur Georges Chappaz enseignait : « L’étude des terres du vignoble de Chablis est en quelque sorte l’étude des sols et sous-sols d’origine kimméridgienne ». S’appuyant sur cette affirmation, les viticulteurs du secteur des grands crus feront réserver, par l’INAO en 1938, l’appellation Chablis aux seuls sols sur kimméridgien. Excluant ainsi les vignobles implantés sur portlandien, ils se débarrassaient d’une certaine concurrence locale, ce qui, bien entendu, n’était pas du goût des viticulteurs exclus. Il fallut près de quarante ans et l’intervention du Ministre de l’Agriculture français pour faire cesser cette guerre fratricide. Châteauneuf du Pape a vécu une aventure similaire. Les sols mythiques de Châteauneuf sont les fameux galets roulés des alluvions fluviatiles que le Rhône a arrachés aux Alpes en surrection tout au long de son cours et déposés là dans des parties plus planes de sa basse vallée. Dans l’époque la plus récente de l’ère quaternaire, ces alluvions ont été protégées des crues destructrices du fleuve par une grosse colline de calcaire urgonien qui émergeait au milieu de la vallée du Rhône. Lorsque les sols de galets roulés furent entièrement plantés de vignes, certains vignerons étendirent dans les années 70 leurs plantations sur les calcaires urgoniens qui ont, il faut le reconnaître, un régime hydrique tout à fait différent de celui des terrasses quaternaires caillouteuses. Il s’agissait là d’un véritable sacrilège qui, là encore, divisa les hommes du terroir dans une guerre dont les cicatrices sont encore sensibles. Mais, que l’on ne s’y trompe pas. Ces batailles internes à certains terroirs ne sont pas le lot de la seule vieille Europe. Car en Australie, également, les hommes se sont battus autour du rôle du sol dans la définition d’une zone viticole. Le Coonawarra, petite bande de terra rossa en forme de cigare de 15 kilomètres de long sur 3 de large, a fait l’objet de combats juridiques farouches entre les partisans de

l’utilisation du nom « Coonawarra » par les seuls viticulteurs installés sur la terra rossa et ceux qui, bien qu’implantés sur les terres noires profondes qui entourent le « cigare », avaient depuis des décades des usages d’utilisation du nom de Coonawarra. Il ne fallut pas moins d’une dizaine d’années de débats acharnés devant les tribunaux australiens avant d’aboutir à un accord sur les limites de la zone à l’intérieur de laquelle les vignerons auront désormais l’exclusivité de l’utilisation du nom de Coonawara pour designer les vins qu’ils y produisent. Les années 70-80 marquèrent un virage décisif dans cette connaissance du rôle du sol dans les terroirs viticoles et dans la relation plus ou moins directe entre les caractéristiques du milieu physique et les qualités organoleptiques des vins qui y sont produits. Ce furent d’abord les travaux du professeur Gérard Seguin à Bordeaux qui seront poursuivis et prolongés par ceux de l’équipe de Kees Van Leeuwen. En essayant de comprendre ce qui pouvait expliquer le niveau de qualité des grands crus de la région de Bordeaux, ces travaux ont montré le rôle fondamental de l’alimentation hydrique dans le fonctionnement de la plante dans des situations géo-pédologiques aussi diverses que les terrasses quaternaires caillouteuses du Haut-Médoc, les calcaires à astéries de Saint Emilion ou les argiles gonflantes (montmorillonites) de Petrus. Quelques années plus tard ont aboutis les travaux de l’équipe INRA d’Angers dirigés par Christian Asselin et René Morlat sur le comportement de deux des cépages caractéristiques des terroirs de l’Anjou, dans la vallée de la Loire, le chenin dans les coteaux du Layon et le cabernet franc à Saumur Champigny, Bourgueil et Chinon. Ils confirmèrent les travaux de Gérard Seguin en apportant de nouveaux éléments complémentaires dans la compréhension du rôle du sol dans l’élaboration des grands vins comme la capacité des sols à se réchauffer, par exemple. En mettant à mal certaines affirmations ou croyances largement répandues dans les vignobles ou les rédactions de journaux spécialisés dans le domaine du vin, ces travaux ont incontestablement ouvert une nouvelle ère, celle de la connaissance scientifique de la relation sol-climat-vin.et en conséquence d’une nouvelle approche du vin par le consommateur. C’est d’ailleurs ainsi qu’a été mis en place le premier des Congrès Internationaux des Terroirs Viticoles, qui s‘est tenu à Angers en 1996. Il s’agissait pour l’Institut National de la Recherche Agronomique de communiquer précisément sur ces tout nouveaux résultats scientifiques avec l’appui de l’Institut National des Appellations d’Origine et de l’Office National des Vins français. Ce lancement s’est fait dans un contexte de grande méfiance notamment d’une partie du nouveau monde qui ne voulait pas entendre parler de ce mot de terroir. Après une phase strictement européenne, nos collègues sud-africains ont ouvert la voie d’un élargissement de cette notion à l’ensemble du monde et on ne peut que se réjouir de voir ce Congrès accueilli par nos collègues américains en 2012. Ce Congrès d’Angers fut le point de départ d’une formidable série de travaux dans le monde entier qui ont beaucoup aidé à la connaissance et au développement de ces produits de terroir sur les cinq continents. Ce fut notamment le cas sur la connaissance du rôle du sol qui a souvent été contesté, dans le nouveau monde par les grands industriels du vin ou une certaine presse internationale, mais aussi au sein de pays comme la France ou il existe une école de géographes du vin très forte avec comme chef de file Roger Dion qui, dans son remarquable ouvrage « Histoire de la vigne et du vin des origines au 19ème siècle », avait fortement contesté le rôle du sous-sol et du sol dans la localisation des vignobles français. Il attribuait en effet cette localisation à des facteurs totalement différents comme les grands courants de commercialisation et les routes commerciales, les vignobles étant principalement installés le long des fleuves Loire , Seine, Rhône, etc… parce qu’ils représentaient de voies de circulation très pratiques, ou autour des grands ports d’expédition des vins comme Bordeaux ou Porto. C’est ici que les travaux très complets qui ont été présentés au cours de nos 8 congrès internationaux prennent toute leur valeur. Ils ont largement contribué à modifier ces idées et les comportements des viticulteurs dans tout le monde viticole et à améliorer la connaissance des consommateurs sur les origines de la qualité des grands vins partout dans le monde. En tant que participant de la première heure à la mise en place de ce système de diffusion des connaissances scientifiques avec la complicité

4 - 5



d’un certain nombre de collègues dont la plupart sont ici présents, je me dis parfois qu’il est peut-être tant de regarder le chemin parcouru depuis 1996, année qui a également coïncidé avec la mise en place à l’OIV d’une groupe de travail sur le terroir et le zonage. L’examen des quelques 500 communications présentées en une petite quinzaine d’années illustre parfaitement ce chemin parcouru dans la foulée des premières explications scientifiques de la relation sol-vigne-vin. Des avancées considérables ont notamment été réalisées :

- dans le domaine de l’amélioration constante des méthodologies d’étude des terroirs. D’ailleurs, à l’initiative du groupe de travail « environnement viticole et changement climatique » , un projet de résolution intitulé « méthodologie de zonage vitivinicole au niveau du sol » est actuellement à l’étude au sein de l’OIV.

- Dans le domaine de la caractérisation des différentes zones viticoles partout dans le monde. Ce sont les sujets qui ont le plus souvent faits l’objet de communication au cours de nos Congrès. On peut peut-être regretter que des travaux de recherche fondamentale sur les explications scientifiques de la relation sol-vigne-vin à l’image de ceux réalisés par les équipes d’Angers et de Bordeaux n’aient pas été développés d’une manière plus généralisées comme on aurait pu l’espérer. Certes, ils nécessitent des protocoles de recherche particulièrement lourds en moyen humains et financiers. Mais ils peuvent apporter des informations très importantes pour les nouveaux vignobles notamment et c’est certainement le rôle de ces Congrès Internationaux de puiser dans l’expérience des vignobles traditionnels pour aider au développement de nouveaux vignobles ou de vignobles dont l’évolution qualitative est récente. Par ailleurs, il serait utile de réfléchir, après une quinzaine d’années de réflexion essentiellement concentrées sur des sciences comme la géologie, la pédologie et le fonctionnement de la vigne, à l’ouverture du champ scientifique à des matières nouvelles comme par exemple, la microbiologie du sol, que nous connaissons en fait relativement mal dans le rôle qu’elle peut jouer dans la typicité des produits. A un moment où la viticulture s’oriente de manière très sensible vers des approches beaucoup plus respectueuses de l’environnement sous la poussée des consommateurs, cette ouverture semble indispensable. LE ROLE DU CLIMAT Le rôle du climat dans le système terroir a fait l’objet de communications beaucoup moins nombreuses au cours des précédents Congrès. L’essentiel des communications a tourné autour des problématiques d’indices climatiques et de zonages sur la base du climat même, si les différentes échelles depuis le niveau mondial jusqu’à celui de la plante ont toutes été étudiées. L’étude du rôle du climat dans les caractéristiques organoleptiques et la typicité des vins de terroir est en effet, beaucoup plus complexe à mettre en œuvre que dans le cas du sol et nécessite des moyens encore plus importants notamment en matière de points d’observation des différents phénomènes climatiques. C’est probablement ce qui explique la plus grande rareté d’études dans ce domaine mise à part quelques exemples comme celui du rôle de la brise de mer dans le vignoble d’Afrique du Sud qui a été présenté par Victoria Carey. Toutefois le champ d’investigation dans le domaine de la compréhension du rôle du climat dans le système terroir reste immense et beaucoup de travail reste à y faire. Dans des régions comme la Bourgogne et notamment les Côtes de Nuits et de Beaune qui connaissent une relative homogénéité des couches géologiques, le climat a très certainement un rôle important dans la mise en place de la multitude de terroirs et de crus différents. Il y a là un travail de recherche intéressant Le changement climatique qui est l’un des thèmes étudiés au cours de cette journée n’avait jamais fait l’objet de communication au sein des Congrès Terroir jusqu’à aujourd’hui. Il sera donc très intéressant de suivre cette nouvelle thématique. Les publications de Gregory Jones et de Hans Schulz au cours des

dernières années qui s’insèrent dans le débat mondial sur l’évolution du climat de la planète, ne laissent aujourd’hui plus guère de doute sur les changements importants que la viticulture mondiale va connaître dans les prochaines décades. Il reste à connaître à quelle vitesse ces changements vont se produire et quelle sera l’amplitude de ces évolutions. Aujourd’hui, nous en sommes au niveau des constations que ce changement opère sur les différentes viticultures du monde. Mais là encore un champ très vaste de recherche s’ouvre sur les moyens de lutter contre ces changements. Il ne s’agit pas ici des mesures à prendre au niveau de la planète qui ont fait récemment l’objet du sommet de Copenhague, mais bien des mesures à prendre par les vignerons de toutes les régions viticoles du monde. En effet, si l’on constate l’extension de certaines zones viticoles en direction des pôles qui conduit à l’implantation de nouveaux vignobles dans des zones jusqu’alors considérées comme impropres à la culture de la vigne, le changement climatique va modifier de manière plus ou moins sensible les modes de conduite de la vigne dans les zones viticoles traditionnelles. Les scénarios qui envisagent des déplacements géographiques de vignobles sont très peu probables. Il est évident que dans ces régions de longue tradition viticole, les vignerons devront trouver sur place des solutions pour contrer les effets du réchauffement climatique et dans ce domaine le champ de recherche est très vaste mais des solutions devront être trouvées très rapidement notamment quand elles se traduiront par des modifications des habitudes de plantation de la vigne.

4 - 6



d’un certain nombre de collègues dont la plupart sont ici présents, je me dis parfois qu’il est peut-être tant de regarder le chemin parcouru depuis 1996, année qui a également coïncidé avec la mise en place à l’OIV d’une groupe de travail sur le terroir et le zonage. L’examen des quelques 500 communications présentées en une petite quinzaine d’années illustre parfaitement ce chemin parcouru dans la foulée des premières explications scientifiques de la relation sol-vigne-vin. Des avancées considérables ont notamment été réalisées :

- dans le domaine de l’amélioration constante des méthodologies d’étude des terroirs. D’ailleurs, à l’initiative du groupe de travail « environnement viticole et changement climatique » , un projet de résolution intitulé « méthodologie de zonage vitivinicole au niveau du sol » est actuellement à l’étude au sein de l’OIV.

- Dans le domaine de la caractérisation des différentes zones viticoles partout dans le monde. Ce sont les sujets qui ont le plus souvent faits l’objet de communication au cours de nos Congrès. On peut peut-être regretter que des travaux de recherche fondamentale sur les explications scientifiques de la relation sol-vigne-vin à l’image de ceux réalisés par les équipes d’Angers et de Bordeaux n’aient pas été développés d’une manière plus généralisées comme on aurait pu l’espérer. Certes, ils nécessitent des protocoles de recherche particulièrement lourds en moyen humains et financiers. Mais ils peuvent apporter des informations très importantes pour les nouveaux vignobles notamment et c’est certainement le rôle de ces Congrès Internationaux de puiser dans l’expérience des vignobles traditionnels pour aider au développement de nouveaux vignobles ou de vignobles dont l’évolution qualitative est récente. Par ailleurs, il serait utile de réfléchir, après une quinzaine d’années de réflexion essentiellement concentrées sur des sciences comme la géologie, la pédologie et le fonctionnement de la vigne, à l’ouverture du champ scientifique à des matières nouvelles comme par exemple, la microbiologie du sol, que nous connaissons en fait relativement mal dans le rôle qu’elle peut jouer dans la typicité des produits. A un moment où la viticulture s’oriente de manière très sensible vers des approches beaucoup plus respectueuses de l’environnement sous la poussée des consommateurs, cette ouverture semble indispensable. LE ROLE DU CLIMAT Le rôle du climat dans le système terroir a fait l’objet de communications beaucoup moins nombreuses au cours des précédents Congrès. L’essentiel des communications a tourné autour des problématiques d’indices climatiques et de zonages sur la base du climat même, si les différentes échelles depuis le niveau mondial jusqu’à celui de la plante ont toutes été étudiées. L’étude du rôle du climat dans les caractéristiques organoleptiques et la typicité des vins de terroir est en effet, beaucoup plus complexe à mettre en œuvre que dans le cas du sol et nécessite des moyens encore plus importants notamment en matière de points d’observation des différents phénomènes climatiques. C’est probablement ce qui explique la plus grande rareté d’études dans ce domaine mise à part quelques exemples comme celui du rôle de la brise de mer dans le vignoble d’Afrique du Sud qui a été présenté par Victoria Carey. Toutefois le champ d’investigation dans le domaine de la compréhension du rôle du climat dans le système terroir reste immense et beaucoup de travail reste à y faire. Dans des régions comme la Bourgogne et notamment les Côtes de Nuits et de Beaune qui connaissent une relative homogénéité des couches géologiques, le climat a très certainement un rôle important dans la mise en place de la multitude de terroirs et de crus différents. Il y a là un travail de recherche intéressant Le changement climatique qui est l’un des thèmes étudiés au cours de cette journée n’avait jamais fait l’objet de communication au sein des Congrès Terroir jusqu’à aujourd’hui. Il sera donc très intéressant de suivre cette nouvelle thématique. Les publications de Gregory Jones et de Hans Schulz au cours des

dernières années qui s’insèrent dans le débat mondial sur l’évolution du climat de la planète, ne laissent aujourd’hui plus guère de doute sur les changements importants que la viticulture mondiale va connaître dans les prochaines décades. Il reste à connaître à quelle vitesse ces changements vont se produire et quelle sera l’amplitude de ces évolutions. Aujourd’hui, nous en sommes au niveau des constations que ce changement opère sur les différentes viticultures du monde. Mais là encore un champ très vaste de recherche s’ouvre sur les moyens de lutter contre ces changements. Il ne s’agit pas ici des mesures à prendre au niveau de la planète qui ont fait récemment l’objet du sommet de Copenhague, mais bien des mesures à prendre par les vignerons de toutes les régions viticoles du monde. En effet, si l’on constate l’extension de certaines zones viticoles en direction des pôles qui conduit à l’implantation de nouveaux vignobles dans des zones jusqu’alors considérées comme impropres à la culture de la vigne, le changement climatique va modifier de manière plus ou moins sensible les modes de conduite de la vigne dans les zones viticoles traditionnelles. Les scénarios qui envisagent des déplacements géographiques de vignobles sont très peu probables. Il est évident que dans ces régions de longue tradition viticole, les vignerons devront trouver sur place des solutions pour contrer les effets du réchauffement climatique et dans ce domaine le champ de recherche est très vaste mais des solutions devront être trouvées très rapidement notamment quand elles se traduiront par des modifications des habitudes de plantation de la vigne.

4 - 7



1

THE GEOLOGICAL AND GEOMORPHOLOGICAL EVENTS THATDETERMINE THE SOIL FUNCTIONAL CHARACTERS OF A TERROIR

E. A.C. Costantini (1) , P. Bucelli (1) , S. Priori (1)

(1)Agricultural Research Council, Research centre for Agrobiology and Pedology, p. D’Azeglio 30, Firenze, [email protected] geology of a region is deemed to be an important component of terroir, as it influences the shape ofthe landscape and the climate of vineyard. The nature of rock and the geomorphological history of aterroir affect soil physical and chemical composition through a dynamic interplay with the changes ofclimate, vegetation and other living organisms, as well as with man activities.This work is aimed at demonstrating that the soil functional characters which differentiate the terroirs ofa denomination of origin area are products and witnesses of the geological and geomorphologicalevents, natural and human induced, which occurred in that trait of land. The final scope being enhancingthe awareness of stakeholders about the possible environmental and economic losses that can derivefrom an irrational soil management, which can lead to the worsening or loss of irreproducible soilfunctional characters of the best terroirs.The work makes reference to the denomination of origin ”Vino Nobile di Montepulciano”, where a fouryears research was conducted on the relationships between soil characteristics and the viticultural andoenological behaviour of Sangiovese vine. The soils of the Montepulciano vineyard range notably infertility conditions and functional characters, also when formed on the same kind of sediments, inparticular as for water and oxygen availability. The grape production at vintage, as well as theorganoleptic characteristics of the wine, resulted strictly interactive with the different soils. The winesobtained on a first group of soils had a good structure and typicity, but the stability of wine sensorialprofile during the years was low. A second group exhibited good structure, typicity, and a good stabilityof wine sensorial profile. A third group had low structure, low typicity, and high astringency all theyears of trial.The oldest soils of the Montepulciano vineyard started their formation during the Pleistocene. Duringthe medium Holocene, man deeply influenced pedogenesis, but it is during the last 50 years that theintensity of the man action reached its maximum. Pre-plantation activities of the new specializedvineyards upset the land, leaving very different effects on soil functional characters. Where the soilsbefore vine plantation were deep and rather homogeneous, soil functional characters remained the same,whereas they changed significantly where soils were shallower. Shallow soils on marine clays, inparticular, resulted very vulnerable.Best soils for the Nobile di Montepulciano wine production, that is, those belonging to the second group,were old soils, formed as a consequence of particular natural and human induced geomorphologicalevents. Therefore they should be considered cultural heritages.KEY WORDSClimate change – cultural heritage – wine – quality – Sangiovese – Vino Nobile di Montepulciano

2

INTRODUCTIONSoil is considered a major component of terroir, although most evidences that relate wine with thespecific soil conditions of the vineyard are empirical (Van Leeuwen et al., 2004). It has been largelydemonstrated, for instance, that best terroirs for red wine production are often placed where some soillimiting factors reduce vine vigour and berry size (Van Leeuwen and Seguin 2006), so that grapes ripencompletely but slowly. In these soils, high quality wines are obtained every year, in spite of climaticvariations (Seguin, 1986).In addition to empirical evidence, a few mechanisms have been understood. In particular, it has beenproved that soil water availability influences the hormonal equilibrium of each vine variety, which inturn regulates the expression of the genotype (Van Leeuwen and Seguin, 1997). Similarly, nitrogen andwater supply controls the biosynthesis of flavonols, through the activation of the enzyme Phenylalanineammonia lyase, which diverts phenylalanine from the pathway that relate carbohydrates to the synthesisof proteins (Kao et al. 2002).As a whole, nitrogen nutrition and water supply during certain phases of the vegetative cycle of the vineare considered essential factors of wine quality. Their role in determining the terroir effect has beenexperienced in many wine producing areas and with several varieties, among others, in France, withCabernet Sauvignon (Choné et al. 2001), Merlot (Trégoat et al. 2002), and Sauvignon Blanc (Peyrot desGachons et al., 2005), in Australia (White et al., 2007) with Sauvignon Blanc, in Hungary withKékfrankos (Zsófi et al., 2009), in USA with Cabernet Sauvignon and Chardonnay (Chapman et al.,2005; Deluc et al., 2009).Furthermore, there is an array of other empirical relationships that prove terroir dependence from soilcharacters. Among the most renown there is the effect of soil colour. Colour is one of the main soilcharacteristics. It can differ widely from bright white, as for some calcareous soils, to red, as in “terrarossa” soils, or black, in soils from slate. Soil colour affects the quality and quantity of light reflectedinto the bunch zone and grapevine canopy, thus influencing grapevine performance. The colour of lightreflected from the soil surfaces appears to be used by the grapevine natural growth regulatory system toalter vegetative growth (Witbooi et al., 2008 a). In South Africa, with Cabernet sauvignon, grey soilsurface resulted in higher grape colour at 520 nm, while the potassium content of the pulp was thehighest on red soil surface treatments (Witbooi et al., 2008 b). Stony soils reflect heat if they are pale-coloured. Well known examples are the white cobbles of Chateauneuf-du-Pape, the pebbles at Sancerre(France), and at Monsant (Spain). In contrast, the metamorphic rocks and grey limestone of theFranconia (Germany) provide dark-coloured soils that warm relatively quickly and store heat, thuspromoting ripening in this region (Maltman, 2008). Large thermal effects on soil surface temperatureand on berry skin temperature were found in Geisenheim (Germany) for the vine Riesling and Pinot noir(Stoll et al., 2008). In the north Willamette Valley (Oregon, USA) basalt-derived surfaces enhancecytokinin synthesis through spreading the diurnal heat load (Nikolaou et al., 2000); similar effects arisefurther north in parts of the Walla Walla Valley, Washington, USA (Meinert and Busacca, 2000).Another important soil quality is water drainage, which is considered a major terroir characteristic of themoraine deposits of Fanciacorta (Northern Italy) as well as of many other territories (Panont et al.,1997). A rapid soil water drainage has been found to affect significantly the precocity of bud breakingand the intensity of summer stress.

4 - 8



1

THE GEOLOGICAL AND GEOMORPHOLOGICAL EVENTS THATDETERMINE THE SOIL FUNCTIONAL CHARACTERS OF A TERROIR

E. A.C. Costantini (1) , P. Bucelli (1) , S. Priori (1)

(1)Agricultural Research Council, Research centre for Agrobiology and Pedology, p. D’Azeglio 30, Firenze, [email protected] geology of a region is deemed to be an important component of terroir, as it influences the shape ofthe landscape and the climate of vineyard. The nature of rock and the geomorphological history of aterroir affect soil physical and chemical composition through a dynamic interplay with the changes ofclimate, vegetation and other living organisms, as well as with man activities.This work is aimed at demonstrating that the soil functional characters which differentiate the terroirs ofa denomination of origin area are products and witnesses of the geological and geomorphologicalevents, natural and human induced, which occurred in that trait of land. The final scope being enhancingthe awareness of stakeholders about the possible environmental and economic losses that can derivefrom an irrational soil management, which can lead to the worsening or loss of irreproducible soilfunctional characters of the best terroirs.The work makes reference to the denomination of origin ”Vino Nobile di Montepulciano”, where a fouryears research was conducted on the relationships between soil characteristics and the viticultural andoenological behaviour of Sangiovese vine. The soils of the Montepulciano vineyard range notably infertility conditions and functional characters, also when formed on the same kind of sediments, inparticular as for water and oxygen availability. The grape production at vintage, as well as theorganoleptic characteristics of the wine, resulted strictly interactive with the different soils. The winesobtained on a first group of soils had a good structure and typicity, but the stability of wine sensorialprofile during the years was low. A second group exhibited good structure, typicity, and a good stabilityof wine sensorial profile. A third group had low structure, low typicity, and high astringency all theyears of trial.The oldest soils of the Montepulciano vineyard started their formation during the Pleistocene. Duringthe medium Holocene, man deeply influenced pedogenesis, but it is during the last 50 years that theintensity of the man action reached its maximum. Pre-plantation activities of the new specializedvineyards upset the land, leaving very different effects on soil functional characters. Where the soilsbefore vine plantation were deep and rather homogeneous, soil functional characters remained the same,whereas they changed significantly where soils were shallower. Shallow soils on marine clays, inparticular, resulted very vulnerable.Best soils for the Nobile di Montepulciano wine production, that is, those belonging to the second group,were old soils, formed as a consequence of particular natural and human induced geomorphologicalevents. Therefore they should be considered cultural heritages.KEY WORDSClimate change – cultural heritage – wine – quality – Sangiovese – Vino Nobile di Montepulciano

2

INTRODUCTIONSoil is considered a major component of terroir, although most evidences that relate wine with thespecific soil conditions of the vineyard are empirical (Van Leeuwen et al., 2004). It has been largelydemonstrated, for instance, that best terroirs for red wine production are often placed where some soillimiting factors reduce vine vigour and berry size (Van Leeuwen and Seguin 2006), so that grapes ripencompletely but slowly. In these soils, high quality wines are obtained every year, in spite of climaticvariations (Seguin, 1986).In addition to empirical evidence, a few mechanisms have been understood. In particular, it has beenproved that soil water availability influences the hormonal equilibrium of each vine variety, which inturn regulates the expression of the genotype (Van Leeuwen and Seguin, 1997). Similarly, nitrogen andwater supply controls the biosynthesis of flavonols, through the activation of the enzyme Phenylalanineammonia lyase, which diverts phenylalanine from the pathway that relate carbohydrates to the synthesisof proteins (Kao et al. 2002).As a whole, nitrogen nutrition and water supply during certain phases of the vegetative cycle of the vineare considered essential factors of wine quality. Their role in determining the terroir effect has beenexperienced in many wine producing areas and with several varieties, among others, in France, withCabernet Sauvignon (Choné et al. 2001), Merlot (Trégoat et al. 2002), and Sauvignon Blanc (Peyrot desGachons et al., 2005), in Australia (White et al., 2007) with Sauvignon Blanc, in Hungary withKékfrankos (Zsófi et al., 2009), in USA with Cabernet Sauvignon and Chardonnay (Chapman et al.,2005; Deluc et al., 2009).Furthermore, there is an array of other empirical relationships that prove terroir dependence from soilcharacters. Among the most renown there is the effect of soil colour. Colour is one of the main soilcharacteristics. It can differ widely from bright white, as for some calcareous soils, to red, as in “terrarossa” soils, or black, in soils from slate. Soil colour affects the quality and quantity of light reflectedinto the bunch zone and grapevine canopy, thus influencing grapevine performance. The colour of lightreflected from the soil surfaces appears to be used by the grapevine natural growth regulatory system toalter vegetative growth (Witbooi et al., 2008 a). In South Africa, with Cabernet sauvignon, grey soilsurface resulted in higher grape colour at 520 nm, while the potassium content of the pulp was thehighest on red soil surface treatments (Witbooi et al., 2008 b). Stony soils reflect heat if they are pale-coloured. Well known examples are the white cobbles of Chateauneuf-du-Pape, the pebbles at Sancerre(France), and at Monsant (Spain). In contrast, the metamorphic rocks and grey limestone of theFranconia (Germany) provide dark-coloured soils that warm relatively quickly and store heat, thuspromoting ripening in this region (Maltman, 2008). Large thermal effects on soil surface temperatureand on berry skin temperature were found in Geisenheim (Germany) for the vine Riesling and Pinot noir(Stoll et al., 2008). In the north Willamette Valley (Oregon, USA) basalt-derived surfaces enhancecytokinin synthesis through spreading the diurnal heat load (Nikolaou et al., 2000); similar effects arisefurther north in parts of the Walla Walla Valley, Washington, USA (Meinert and Busacca, 2000).Another important soil quality is water drainage, which is considered a major terroir characteristic of themoraine deposits of Fanciacorta (Northern Italy) as well as of many other territories (Panont et al.,1997). A rapid soil water drainage has been found to affect significantly the precocity of bud breakingand the intensity of summer stress.

4 - 9



3

The results of a vine zoning in Emilia-Romagna (Italy) highlighted the relationship between limecontent in soil and wine colour, structure and perfume intensity. As a matter of fact, in soils with absentor little lime, grape had lower sugar degree, wine colour intensity and structure. On the contrary, in soilswith high active lime, grapes had a larger sugar and polyphenols content, and wines were full-bodiedand with high colour intensity (Scotti, 2006). Also studies carried out in Alto-Adige (Northern Italy)with Schiava vine demonstrated that total polyphenols of grape increased with the increase of activelime in soil (Fregoni, 2005).The potassium content of soil can have a strong effect on must acidity. In particular, the vine responds toan over potassium absorption synthesizing malic acid, to neutralize the surplus of K+ ions. The reactiondetermines the decrease of acidity and the increase of pH (Hale, 1977; Calò et al., 2002). Berrypotassium content has been also related to the pH value of must and some Authors have suggested thatany factor that reduces the photosynthetic activity of leaves could increase potassium accumulation inberries (Freeman et al, 1982). Such factors can be water stress, wind exposure, and excessive shading ofthe canopy, like with Cabernet Sauvignon in South Africa (Carey et al., 2008).High soil salinity strongly affects vine performance. Salt excess has both an osmotic and a toxic effect,causing reduction in yield, shoot growth and berry weight (Lanyon et al., 2004). However Costantini etal. (2009a) demonstrated a better performance of Sangiovese in the Chianti area when cultivated on veryfertile but moderately saline soils, when the salinity was confined to the deep soil horizons.The geology of a region is also deemed to be an important component of terroir (Vaudour, 2003;Maltman, 2008). Geology influences the shape of the landscape, conferring morphology, typical spacesand articulations that characterize a production district. Geology also influences the morphology of aterritory and thus the climate of vineyard, through the altitude, the aspect of the slope, the vicinity towater bodies, the exposure to dominant winds. At Stellenbosch (South Africa) in particular, it has beendemonstrated that site differences in wind exposure have stronger effect on Sauvignon blanc thanseasonal climatic differences (Carey et al., 2008).The nature of rock governs deep drainage, but also the quality of groundwater and irrigation water. Arelevant characteristic of rock is the degree of its resistance to root penetration. This property derivesfrom rock type, presence of planes of weakness, their spacing and orientation (Myburgh et al., 1996).For instance, some of the best grapes produced in the Upper Douro area of Portugal, as well as in Prioratin Catalunya (Spain), Languedoc Roussillon (South France) and in Chianti (Central Italy) are obtainedfrom shallow soils on clay schist (“galestro” in Italian). The foliation of schist provides surfaces forroots penetration in an otherwise impenetrable material.Furthermore, there is a common assumption that the kind of rock or sediment determines the soilphysical and chemical composition of the vineyard (White, 2003). As a matter of fact, the linkagebetween rock and the overlying soil is in most cases weak, due to the frequent presence of allochthonousmaterial, like aeolian dust, volcanic ash, colluvium, or human transported material and, above all,because of the transformations caused by pedogenesis. In fact, the geological and geomorphologicalhistory of a territory dynamically interacts with the changes of climate, vegetation and other livingorganisms, as well as with man activities, leading to soil formation. The soil of a vineyard is themetastable complex system coming out from a succession of rhexistasy and biostasy periods, duringwhich rocks and sediments are weathered, transformed and translocated, leached and depleted, erodedand accumulated, mixed with organic particles,

4

and organized in micro, meso, and macrostructures. The time scale of the biorhexistasy periods canrange from millions of years, in very stable landscapes, to few years, in heavily anthropized territories.Aim of this work is to demonstrate that the soil functional characters which differentiate the terroirs of adenomination of origin area are at the same time products and witnesses of the geological andgeomorphological events, natural as well as human induced, which occurred in that trait of land.Knowing the geological and geomorphological history of a terroir is relevant, because it can enhance theawareness of stakeholders about the possible environmental and economic losses that can derive from anirrational soil management, which can lead to the worsening or loss of the soil functional characters ofthe best terroirs.

MATERIALS AND METHODSThe work makes reference to the denomination of origin ”Vino Nobile di Montepulciano”. On 1st July1980 the Vino Nobile di Montepulciano became the first Italian red wine to get the D.O.C.G.(guaranteed and controlled denomination of origin), which places it among the most prestigious wines inItaly and the world. “Prugnolo gentile”, a biotype of Sangiovese, is the basic vine-variety used for theproduction of the Vino Nobile and the most important winegrape in the classic wine territory. TheNobile is a wine of elevated structure and longevity, thanks to the rich supply of anthocyanins andpolyphenols, capable of ensuring a positive evolution over a length of time. The considerable delicacy ofthe bouquet, distinguished by the pleasant scents of violet and woodland fruits, with elegant hints ofspices, dried fruits and dried vegetable, completes the Nobile's profile of high quality and originality.In the Montepulciano territory, a research was conducted on the relationships between soil characteristicsand the viticultural and oenological behaviour of Sangiovese (Costantini et al., 1996; Campostrini et al.,1997). The study was carried out during four years. On the basis of a soil map of the Montepulciano hillat 1:25,000 scale, 54 experimental not irrigated plots, homogeneous in soil, were selected (9 soil typesper 6 replications, tab. 1). Experimental soils were evaluated as for their physical, chemical, hydrologicaland biological properties. In addition, temperature and rainfall were monitored in 5 experimental sites, aswell as soil temperature and moisture content. The mean rainfall of the studied years was lower than inthe long term (1926- 1992: 685 vs. 728 mm). Also the year average temperature remained lower than thelong term (1928-1992: 12.9 vs. 13.9°C; tab. 2). The value of Winkler's index fitted the requirements forSangiovese (Fregoni, 2005). The number of days when the soil moisture control section is dry matchedthe requirements of the "ustic" soil moisture regime (USDA, 2010).

Table 1 - Topographic, physical, hydrological, and chemical characteristics of the experimental soils.

Parameters Mean values and (standard deviation)Elevation (m asl) 344(67)Slope gradient (%) 6.6(3.6)Aspect (°) 140(101)Solar radiation (MJ cm2 yr-1) 0.917(0.047)Rooting depth (cm) 106(28)Stoniness (% v/v) 0Rock fragments (% v/v) 0.9 (1.9)Bulk density (g cm-3) 1.53 (0.08)Clay (%) 35.1 (15.1)Sand (%) 27.6 (20.9)Available Water Capacity (mm m-1) 148 (44)Organic C (%) 0.64 (0.20)pH (H2O) 8.4 (0.18)Electrical conductivity (dS m-1) 0.24 (0.19)CEC (cmol(+) kg-1) 15.7 (4.8)Total CaCO3 (%) 11.9 (7.0)

4 - 10



3

The results of a vine zoning in Emilia-Romagna (Italy) highlighted the relationship between limecontent in soil and wine colour, structure and perfume intensity. As a matter of fact, in soils with absentor little lime, grape had lower sugar degree, wine colour intensity and structure. On the contrary, in soilswith high active lime, grapes had a larger sugar and polyphenols content, and wines were full-bodiedand with high colour intensity (Scotti, 2006). Also studies carried out in Alto-Adige (Northern Italy)with Schiava vine demonstrated that total polyphenols of grape increased with the increase of activelime in soil (Fregoni, 2005).The potassium content of soil can have a strong effect on must acidity. In particular, the vine responds toan over potassium absorption synthesizing malic acid, to neutralize the surplus of K+ ions. The reactiondetermines the decrease of acidity and the increase of pH (Hale, 1977; Calò et al., 2002). Berrypotassium content has been also related to the pH value of must and some Authors have suggested thatany factor that reduces the photosynthetic activity of leaves could increase potassium accumulation inberries (Freeman et al, 1982). Such factors can be water stress, wind exposure, and excessive shading ofthe canopy, like with Cabernet Sauvignon in South Africa (Carey et al., 2008).High soil salinity strongly affects vine performance. Salt excess has both an osmotic and a toxic effect,causing reduction in yield, shoot growth and berry weight (Lanyon et al., 2004). However Costantini etal. (2009a) demonstrated a better performance of Sangiovese in the Chianti area when cultivated on veryfertile but moderately saline soils, when the salinity was confined to the deep soil horizons.The geology of a region is also deemed to be an important component of terroir (Vaudour, 2003;Maltman, 2008). Geology influences the shape of the landscape, conferring morphology, typical spacesand articulations that characterize a production district. Geology also influences the morphology of aterritory and thus the climate of vineyard, through the altitude, the aspect of the slope, the vicinity towater bodies, the exposure to dominant winds. At Stellenbosch (South Africa) in particular, it has beendemonstrated that site differences in wind exposure have stronger effect on Sauvignon blanc thanseasonal climatic differences (Carey et al., 2008).The nature of rock governs deep drainage, but also the quality of groundwater and irrigation water. Arelevant characteristic of rock is the degree of its resistance to root penetration. This property derivesfrom rock type, presence of planes of weakness, their spacing and orientation (Myburgh et al., 1996).For instance, some of the best grapes produced in the Upper Douro area of Portugal, as well as in Prioratin Catalunya (Spain), Languedoc Roussillon (South France) and in Chianti (Central Italy) are obtainedfrom shallow soils on clay schist (“galestro” in Italian). The foliation of schist provides surfaces forroots penetration in an otherwise impenetrable material.Furthermore, there is a common assumption that the kind of rock or sediment determines the soilphysical and chemical composition of the vineyard (White, 2003). As a matter of fact, the linkagebetween rock and the overlying soil is in most cases weak, due to the frequent presence of allochthonousmaterial, like aeolian dust, volcanic ash, colluvium, or human transported material and, above all,because of the transformations caused by pedogenesis. In fact, the geological and geomorphologicalhistory of a territory dynamically interacts with the changes of climate, vegetation and other livingorganisms, as well as with man activities, leading to soil formation. The soil of a vineyard is themetastable complex system coming out from a succession of rhexistasy and biostasy periods, duringwhich rocks and sediments are weathered, transformed and translocated, leached and depleted, erodedand accumulated, mixed with organic particles,

4

and organized in micro, meso, and macrostructures. The time scale of the biorhexistasy periods canrange from millions of years, in very stable landscapes, to few years, in heavily anthropized territories.Aim of this work is to demonstrate that the soil functional characters which differentiate the terroirs of adenomination of origin area are at the same time products and witnesses of the geological andgeomorphological events, natural as well as human induced, which occurred in that trait of land.Knowing the geological and geomorphological history of a terroir is relevant, because it can enhance theawareness of stakeholders about the possible environmental and economic losses that can derive from anirrational soil management, which can lead to the worsening or loss of the soil functional characters ofthe best terroirs.

MATERIALS AND METHODSThe work makes reference to the denomination of origin ”Vino Nobile di Montepulciano”. On 1st July1980 the Vino Nobile di Montepulciano became the first Italian red wine to get the D.O.C.G.(guaranteed and controlled denomination of origin), which places it among the most prestigious wines inItaly and the world. “Prugnolo gentile”, a biotype of Sangiovese, is the basic vine-variety used for theproduction of the Vino Nobile and the most important winegrape in the classic wine territory. TheNobile is a wine of elevated structure and longevity, thanks to the rich supply of anthocyanins andpolyphenols, capable of ensuring a positive evolution over a length of time. The considerable delicacy ofthe bouquet, distinguished by the pleasant scents of violet and woodland fruits, with elegant hints ofspices, dried fruits and dried vegetable, completes the Nobile's profile of high quality and originality.In the Montepulciano territory, a research was conducted on the relationships between soil characteristicsand the viticultural and oenological behaviour of Sangiovese (Costantini et al., 1996; Campostrini et al.,1997). The study was carried out during four years. On the basis of a soil map of the Montepulciano hillat 1:25,000 scale, 54 experimental not irrigated plots, homogeneous in soil, were selected (9 soil typesper 6 replications, tab. 1). Experimental soils were evaluated as for their physical, chemical, hydrologicaland biological properties. In addition, temperature and rainfall were monitored in 5 experimental sites, aswell as soil temperature and moisture content. The mean rainfall of the studied years was lower than inthe long term (1926- 1992: 685 vs. 728 mm). Also the year average temperature remained lower than thelong term (1928-1992: 12.9 vs. 13.9°C; tab. 2). The value of Winkler's index fitted the requirements forSangiovese (Fregoni, 2005). The number of days when the soil moisture control section is dry matchedthe requirements of the "ustic" soil moisture regime (USDA, 2010).

Table 1 - Topographic, physical, hydrological, and chemical characteristics of the experimental soils.

Parameters Mean values and (standard deviation)Elevation (m asl) 344(67)Slope gradient (%) 6.6(3.6)Aspect (°) 140(101)Solar radiation (MJ cm2 yr-1) 0.917(0.047)Rooting depth (cm) 106(28)Stoniness (% v/v) 0Rock fragments (% v/v) 0.9 (1.9)Bulk density (g cm-3) 1.53 (0.08)Clay (%) 35.1 (15.1)Sand (%) 27.6 (20.9)Available Water Capacity (mm m-1) 148 (44)Organic C (%) 0.64 (0.20)pH (H2O) 8.4 (0.18)Electrical conductivity (dS m-1) 0.24 (0.19)CEC (cmol(+) kg-1) 15.7 (4.8)Total CaCO3 (%) 11.9 (7.0)

4 - 11



5

Table 2 - Mean climatic and pedoclimatic parameters of the experimental plots during the trial.

Parameter Mean values and (standard deviation)Annual rainfall (mm) 685 (114)Mean air temperature (°C) 12.9 (0.7)Mean soil temperature (°C) 15.3 (1.4)Winkler’s index 1815 (134)Number of dry days of the soil moisture control section 83 (15)

Soil types had different functional characters. The characters were considered functional when they werestatistically related to some viticultural parameters and when they actually discriminated between soiltypes at the reference scale (Costantini et al., 2008). For instance, the organic matter content did notresulted functional, since all the soils on trial were periodically cultivated during the growing season andwere poor in organic matter (around 1%). Similarly, all experimental soils were base saturated and theircation exchange capacity was always between 10 and 20 cmol(+) kg-1. On the other hand, the soilcharacteristics which resulted statistically different between soil types (ANOVA), as well as statisticallyrelated to Sangiovese parameters (PCA), where those that could provide a physical or chemical limitationto vine growing.Quanti-qualitative traits at vintage (yield per vine, cluster number, mean cluster weight) and sugaraccumulation rate in berries were recorded. The grapes of each experimental plot were analysed at vintagefor total soluble solids, titrable acidity, pH, malic and tartaric acid, and potassium. At ripeness 50-kgsamples of grapes were collected and processed by the standard techniques for small-lot wine making.Descriptor terms were defined after several taste sessions, and the relevant terminology underwentnormalisation. Through the application of the statistical analysis of PCA, the wines were grouped ingroups.The reconstruction of the geological events was based on the works of Costantini et al. (2009), Capezzuoliet al. (2009), Priori et al. (2008), Costantini and Lizio Bruno (1996), Ferrari and Magaldi (1978), andLosacco (1944). The statistical analysis of PCA and ANOVA were conducted with Statistica®. Geologicalsketches were created with a soil geodatabase and thesoftware ArcGIS® and Surfer®.

RESULTS AND DISCUSSION

Soil functional charactersThe soils of the Montepulciano vineyard range notably in fertility conditions and functional characters,also when formed on the same kind of sediments (tab. 3). On Pliocene marine sand, Cusona soils are verysandy and abundant in primary macroporosity, conferring them great air capacity and rapid drainage; theyare poorly structured, have medium total and active lime content and show low salinity. Strada and SanGimignano soils instead are either coarse or fineloamy, their relatively high air capacity is the consequenceof both particle size and aggregate formation. Their equilibrated texture and good structure confer a highavailable water holding capacity. Lime and salinity are reduced and do not cause any limitation to vinecultivation.Monte, San Quirico and Quercia soils characterize vineyards placed on Pliocene marine clay. Monte soilsare fine, poorly differentiated from the substrate, and show very low air capacity and prominentredoximorphic features. They have low available water capacity but are rich in lime and other salts, so thattheir electrical conductivity, averaged for the entire profile, is moderately high. San Quirico soils are fine-silty and show hydromorphy, although their air capacity is higher than Monte soils. Available waterholding capacity is high and they are moderately rich in lime, but their averaged electrical conductivity is

6

lower than Monte soils. Quercia soils are fine and show “vertic” properties (cracking soils), their aircapacity is slightly lower and available water higher than San Quirico soils, the other properties beingsimilar.On the Pleistocene fluvial-lacustrine clay there are three main soil types, Valiano and Valiano aquicformed in the surroundings of the Valiano town, while Poggio Golo soils are close to Torrita. Both Valianosoils are similar, except for air capacity and internal drainage, which are worse in Valiano aquic. Availablewater capacity is also somewhat less in Valiano aquic than in Valiano, though reaching a rather highabsolute mean value. The two Valiano soils have carbonates but do not show salinity. Poggio Golo soilsare also fine and show hydromorphy, as a consequence of limited air capacity, but have rather high waterholding capacity. Although very deep, they are “duplex” soils, that is, they show a marked change inpropertied between the surface and the subsurface horizons, with a sharp increase in clay and firmness, andparallel decrease in air capacity and hydraulic conductivity, with depth. The “duplex” characteristicstrongly limits root penetration. They have very low carbonates and other salts, so that the electricalconductivity is very low.The hydrological monitoring permitted to group the experimental soils in terms of water or oxygen deficitas follow:- soils without water and oxygen deficit during the whole vine growing season (San Gimignano and Stradasoils),- soils with reduced soil oxygen availability during early spring and moderate summer water deficit (SanQuirico, Quercia and Poggio Golo soils).- soils with pronounced summer water deficit (Cusona soils),- soils with pronounced summer water deficit, and reduced soil oxygen availability during early spring(Monte soils).Valiano and Valiano aquic soils were not monitored, but their behaviour was comparable to that of SanGimignano and San Quirico soils, respectively.

4 - 12



5

Table 2 - Mean climatic and pedoclimatic parameters of the experimental plots during the trial.

Parameter Mean values and (standard deviation)Annual rainfall (mm) 685 (114)Mean air temperature (°C) 12.9 (0.7)Mean soil temperature (°C) 15.3 (1.4)Winkler’s index 1815 (134)Number of dry days of the soil moisture control section 83 (15)

Soil types had different functional characters. The characters were considered functional when they werestatistically related to some viticultural parameters and when they actually discriminated between soiltypes at the reference scale (Costantini et al., 2008). For instance, the organic matter content did notresulted functional, since all the soils on trial were periodically cultivated during the growing season andwere poor in organic matter (around 1%). Similarly, all experimental soils were base saturated and theircation exchange capacity was always between 10 and 20 cmol(+) kg-1. On the other hand, the soilcharacteristics which resulted statistically different between soil types (ANOVA), as well as statisticallyrelated to Sangiovese parameters (PCA), where those that could provide a physical or chemical limitationto vine growing.Quanti-qualitative traits at vintage (yield per vine, cluster number, mean cluster weight) and sugaraccumulation rate in berries were recorded. The grapes of each experimental plot were analysed at vintagefor total soluble solids, titrable acidity, pH, malic and tartaric acid, and potassium. At ripeness 50-kgsamples of grapes were collected and processed by the standard techniques for small-lot wine making.Descriptor terms were defined after several taste sessions, and the relevant terminology underwentnormalisation. Through the application of the statistical analysis of PCA, the wines were grouped ingroups.The reconstruction of the geological events was based on the works of Costantini et al. (2009), Capezzuoliet al. (2009), Priori et al. (2008), Costantini and Lizio Bruno (1996), Ferrari and Magaldi (1978), andLosacco (1944). The statistical analysis of PCA and ANOVA were conducted with Statistica®. Geologicalsketches were created with a soil geodatabase and thesoftware ArcGIS® and Surfer®.


Soil functional charactersThe soils of the Montepulciano vineyard range notably in fertility conditions and functional characters,also when formed on the same kind of sediments (tab. 3). On Pliocene marine sand, Cusona soils are verysandy and abundant in primary macroporosity, conferring them great air capacity and rapid drainage; theyare poorly structured, have medium total and active lime content and show low salinity. Strada and SanGimignano soils instead are either coarse or fineloamy, their relatively high air capacity is the consequenceof both particle size and aggregate formation. Their equilibrated texture and good structure confer a highavailable water holding capacity. Lime and salinity are reduced and do not cause any limitation to vinecultivation.Monte, San Quirico and Quercia soils characterize vineyards placed on Pliocene marine clay. Monte soilsare fine, poorly differentiated from the substrate, and show very low air capacity and prominentredoximorphic features. They have low available water capacity but are rich in lime and other salts, so thattheir electrical conductivity, averaged for the entire profile, is moderately high. San Quirico soils are fine-silty and show hydromorphy, although their air capacity is higher than Monte soils. Available waterholding capacity is high and they are moderately rich in lime, but their averaged electrical conductivity is

6

lower than Monte soils. Quercia soils are fine and show “vertic” properties (cracking soils), their aircapacity is slightly lower and available water higher than San Quirico soils, the other properties beingsimilar.On the Pleistocene fluvial-lacustrine clay there are three main soil types, Valiano and Valiano aquicformed in the surroundings of the Valiano town, while Poggio Golo soils are close to Torrita. Both Valianosoils are similar, except for air capacity and internal drainage, which are worse in Valiano aquic. Availablewater capacity is also somewhat less in Valiano aquic than in Valiano, though reaching a rather highabsolute mean value. The two Valiano soils have carbonates but do not show salinity. Poggio Golo soilsare also fine and show hydromorphy, as a consequence of limited air capacity, but have rather high waterholding capacity. Although very deep, they are “duplex” soils, that is, they show a marked change inpropertied between the surface and the subsurface horizons, with a sharp increase in clay and firmness, andparallel decrease in air capacity and hydraulic conductivity, with depth. The “duplex” characteristicstrongly limits root penetration. They have very low carbonates and other salts, so that the electricalconductivity is very low.The hydrological monitoring permitted to group the experimental soils in terms of water or oxygen deficitas follow:- soils without water and oxygen deficit during the whole vine growing season (San Gimignano and Stradasoils),- soils with reduced soil oxygen availability during early spring and moderate summer water deficit (SanQuirico, Quercia and Poggio Golo soils).- soils with pronounced summer water deficit (Cusona soils),- soils with pronounced summer water deficit, and reduced soil oxygen availability during early spring(Monte soils).Valiano and Valiano aquic soils were not monitored, but their behaviour was comparable to that of SanGimignano and San Quirico soils, respectively.

4 - 13



7

Table 3 - Classification and main functional characters of the experimental soils. Values are the average of allhorizons of two soil profiles of each soil type, but O.M. is that of the plough layer.

Total Active¥ ElectricalGeology Soils Classification^ Air capacity t Oa‡ CaCO3 CaCO3 conductivity

(1:2.5) dS(%) (mm) (%ww-1) m-1

Pliocenemarinesand Cusona

Pliocenemarinesand Strada

TypicUstipsamments 16.2 83 10.9 1.0 0.100

Typic Haplustepts,coarse-loamy 10.9 157 10.7 1.1 0.117

Typic Haplustepts,fine-loamy 10.6 195 16.9 2.2 0.133

Aquic Ustorthents,fine 1.4 75 18.9 9.7 0.941

Aquic Haplustepts,fine-silty 3.8 146 16.1 6.8 0.541

Vertic Haplustepts,fine 3.0 165 20.8 7.0 0.453

Aquic Haplustepts,

fine 5.5 142 9.7 2.9 0.287

Pleistocenefluvial-lacustrine

clay Valiano

Pleistocenefluvial- Poggio Gololacustrine clay

Typic Haplustepts,fine 9.7 169 10.4 3.1 0.172

Aquic Haplustalf,fine 2.3 139 1.7 0.6 0.166

Pliocenemarine San

sand Gimignano

Pliocenemarineclay Monte

Pliocenemarineclay San Quirico

Pliocenemarineclay Quercia

Pleistocenefluvial-

lacustrine Valianoclay aquic

8

^Classification= Soil Taxonomy (USDA, 2010).tAir capacity = difference between total porosity (from bulk density of undisturbed samples) andvolumetric water content at field capacity.

‡Oa = available water holding capacity (difference between water content at field capacity andwilting point).¥Active CaCO3is the lime soluble in ammonium oxalate.

Viticultural and oenological resultsThe grape production at vintage was strictly interactive with the different soils. In particular, the yield pervine, the mean cluster weight, the 100 berries' weight were deeply influenced by the soil type. The numberof shoots in the various soils was significantly different, because of the adaptation of pruning to the differentwater capacity of soils. Also the sugar accumulation rate in berries underlined the importance of the soil fora minimum sugar level of grape, necessary to winemaking as Vino Nobile (tab. 4). In fact, vineyards onCusona soils did not reach the minimum sugar content every year, but only in the more rainy ones. On theother hand, grapes on San Gimignano and Poggio Golo soils reached the minimum sugar content only at latetime (tab. 5).

Table 4 - Mean values of viticultural parameters obtained from the experimental soils. Standard deviation inbrackets.

Soils

Grapeyield/vine

kg

Clusternumber

n°

Meanclusterweight

g

100berriesweight

g

Sugarcontent

°Brix

Sugaraccumulation

rate°Brix/day

Totalacidity

g/L

Cusona2.26(1.4)

12.5(8.4)

185(59.7)

167(17.4)

21.5(1.0) 0.31 (0.03)

7.64(1.6)

Strada 4.57(1.5)

15.2(3.0)

322(122.3)

196(29.3)

20.8(1.2) 0.31 (0.02)

7.28(0.5)

SanGimignano

4.38(1.5)

11.7(3.4)

401(153.2)

211(26.8)

20.8(0.9) 0.29 (0.02)

8.49(1.0)

Monte2.70(0.4)

14.8(0.7)

196(36.8)

130(0.5)

20.3(0.8) 0.31 (0.02)

7.28(0.1)

SanQuirico

3.87(1.3)

12.3(4.5)

321(154.8)

166(29.8)

21.3(1.4) 0.31 (0.02)

7.78(0.6)

Quercia 3.45(1.3)

12.7(5.6)

312(96.9)

160(28.0)

20.9(1.8) 0.31 (0.02) 7.73

(1.0)Valianoaquic

5.07(3.7)

18.6(5.6)

245(145.0)

178(47.4)

21.7(1.1) 0.30 (0.02)

7.77(0.6)

Valiano7.01(0.8)

15.0(1.0)

468(23.1)

202(16.7)

20.0(0.6) 0.27 (0.01)

8.14(1.0)

PoggioGolo

3.94(0.9)

14.6(5.4)

236(41.0)

176(18.8)

20.7(0.9) 0.31 (0.02)

7.53(0.9)

4 - 14



7

Table 3 - Classification and main functional characters of the experimental soils. Values are the average of allhorizons of two soil profiles of each soil type, but O.M. is that of the plough layer.

Total Active¥ ElectricalGeology Soils Classification^ Air capacity t Oa‡ CaCO3 CaCO3 conductivity

(1:2.5) dS(%) (mm) (%ww-1) m-1

Pliocenemarinesand Cusona

Pliocenemarinesand Strada

TypicUstipsamments 16.2 83 10.9 1.0 0.100

Typic Haplustepts,coarse-loamy 10.9 157 10.7 1.1 0.117

Typic Haplustepts,fine-loamy 10.6 195 16.9 2.2 0.133

Aquic Ustorthents,fine 1.4 75 18.9 9.7 0.941

Aquic Haplustepts,fine-silty 3.8 146 16.1 6.8 0.541

Vertic Haplustepts,fine 3.0 165 20.8 7.0 0.453

Aquic Haplustepts,

fine 5.5 142 9.7 2.9 0.287

Pleistocenefluvial-lacustrine

clay Valiano

Pleistocenefluvial- Poggio Gololacustrine clay

Typic Haplustepts,fine 9.7 169 10.4 3.1 0.172

Aquic Haplustalf,fine 2.3 139 1.7 0.6 0.166

Pliocenemarine San

sand Gimignano

Pliocenemarineclay Monte

Pliocenemarineclay San Quirico

Pliocenemarineclay Quercia

Pleistocenefluvial-

lacustrine Valianoclay aquic

8

^Classification= Soil Taxonomy (USDA, 2010).tAir capacity = difference between total porosity (from bulk density of undisturbed samples) andvolumetric water content at field capacity.

‡Oa = available water holding capacity (difference between water content at field capacity andwilting point).¥Active CaCO3is the lime soluble in ammonium oxalate.

Viticultural and oenological resultsThe grape production at vintage was strictly interactive with the different soils. In particular, the yield pervine, the mean cluster weight, the 100 berries' weight were deeply influenced by the soil type. The numberof shoots in the various soils was significantly different, because of the adaptation of pruning to the differentwater capacity of soils. Also the sugar accumulation rate in berries underlined the importance of the soil fora minimum sugar level of grape, necessary to winemaking as Vino Nobile (tab. 4). In fact, vineyards onCusona soils did not reach the minimum sugar content every year, but only in the more rainy ones. On theother hand, grapes on San Gimignano and Poggio Golo soils reached the minimum sugar content only at latetime (tab. 5).

Table 4 - Mean values of viticultural parameters obtained from the experimental soils. Standard deviation inbrackets.

Soils

Grapeyield/vine

kg

Clusternumber

n°

Meanclusterweight

g

100berriesweight

g

Sugarcontent

°Brix

Sugaraccumulation

rate°Brix/day

Totalacidity

g/L

Cusona2.26(1.4)

12.5(8.4)

185(59.7)

167(17.4)

21.5(1.0) 0.31 (0.03)

7.64(1.6)

Strada 4.57(1.5)

15.2(3.0)

322(122.3)

196(29.3)

20.8(1.2) 0.31 (0.02)

7.28(0.5)

SanGimignano

4.38(1.5)

11.7(3.4)

401(153.2)

211(26.8)

20.8(0.9) 0.29 (0.02)

8.49(1.0)

Monte2.70(0.4)

14.8(0.7)

196(36.8)

130(0.5)

20.3(0.8) 0.31 (0.02)

7.28(0.1)

SanQuirico

3.87(1.3)

12.3(4.5)

321(154.8)

166(29.8)

21.3(1.4) 0.31 (0.02)

7.78(0.6)

Quercia 3.45(1.3)

12.7(5.6)

312(96.9)

160(28.0)

20.9(1.8) 0.31 (0.02) 7.73

(1.0)Valianoaquic

5.07(3.7)

18.6(5.6)

245(145.0)

178(47.4)

21.7(1.1) 0.30 (0.02)

7.77(0.6)

Valiano7.01(0.8)

15.0(1.0)

468(23.1)

202(16.7)

20.0(0.6) 0.27 (0.01)

8.14(1.0)

PoggioGolo

3.94(0.9)

14.6(5.4)

236(41.0)

176(18.8)

20.7(0.9) 0.31 (0.02)

7.53(0.9)

4 - 15



9

The wines obtained on a first group of soils (Valiano aquic, Cusona and Monte) had a good structure, a goodtypicity with high cherry and berry fruity. Unfortunately, the stability of wine sensorial profile during theyears was low. A second group (Quercia, Poggio Golo and San Quirico soils) exhibited good structure,typicity, with medium level of cherry and berry fruit and a good stability of wine sensorial profile. The wineobtained on the Strada soils showed medium structure, medium astringency, low cherry and berry fruity witha good stability of the sensorial profile. Finally, the wine of San Gimignano and Valiano soils had a lowstructure, high astringency, low typicity, all the years of trial (tab. 5).

Table 5 - Nobile di Montepulciano grape production and wine quality according to soil and soil group.The number of stars reflects the quantity (grape) or quality of the organoleptic evaluation (othervariables).

GroupSoils

NameGrape

productionHarvest

dateWine

structureWine

harmonyWine

typicityYear

stability

1 Monte and Cusona * **** ** ** ** *1 Valiano aquic **** *** ** ** ** *2 Quercia ** *** ** ** ** **2 Poggio Golo ** ** ** ** ** **2 San Quirico ** **** ** ** ** **3 Strada *** ** */** */** * **3 San Gimignano *** * * * * **3 Valiano ***** * * * * **

The interplay between geological and geomorphological events and the formation of soil functionalcharactersThe territory of Montepulciano municipality was almost all completely covered by the sea during EarlyPliocene. Only small islands, corresponding to the highest relives on Mesozoic rocks of the Chianti-Cetonamount ridge, were emerging from the sea. During Middle Pliocene, about four million years ago, the searegressed and left a succession of deposits, some hundred meters thick, which were rather sandy at their top(Fig. 1). Pliocene was a time of intense pedogenesis, leading to the formation of very deep and red soils,similar to those currently widespread in the subtropics. There are no more remnants of these kind of soils atMontepulciano, since here, like in most part of Italy, they have been completely eroded, as a consequence ofthe successive uplift of the area and climate changes. However, a significant example of Pliocene soil,currently preserved and cultivated with vine, has been reported in the not far away DOCG of Montalcino(Costantini and Priori, 2007).The Villafranchian (Late Pliocene and Early Pleistocene) saw the incision of the Pliocene deposits and thefilling up with lacustrine sediments of the territory comprised between the

10

Apennines and the Chianti-Cetona mount ridge (Fig. 2). During Middle and Late Pleistocene theVillafranchian lacustrine sediments were eroded and filled up with new fluvial and lacustrine deposits (Fig.3). Several fluvial terraces formed along the “Val di Chiana” paleo valley, which was crossed at that time bya southward streaming river, the “paleo Arno”. The deposits on the terraces and hilly lands underwent arather intense and prolonged pedogenesis, leading to the formation of deep and well developed soils, withcontrasted horizons. Alternating biorhexistasy phases, in dependence of both tectonic movements andclimate changes, caused the partial erosion of soils on slopes and the formation of deep colluvial soils onrelatively more stable morphological positions.During Early Holocene the morphology of the Montepulciano territory was very similar to the present (Fig.4). The progressive filling up of the valley caused a further swamping of the lowlands, while tectonicmovements led to the reversion of the drainage of the paleo Arno river. Poggio Golo soils dominated on thefluvial terraces formed in the previous geological era. The sharp contrast in texture between the surface andsubsurface horizons of Poggio Golo soils was created as a result of the prolonged time of pedogenesis,during which carbonates and other bases were leached throughout the profile, clay particles deflocculatedand were transported along the profile, where they progressively accumulated. Many macropores weretherefore plugged by clay particles, so that firmness of the subsurface horizon increased, whereas aircapacity and hydraulic conductivity decreased. Soil rooting depth and rootability (mass of soil that can beactually penetrated by roots, between and inside aggregates) were consequently constrained. The breakdownand weathering of rock fragments augmented the proportion of fine particles and released a great deal ofelements, in particular metals. The elements were sequestrated by clays, accumulated, and iron painted thesoils with characteristics reddish brown colours. Subsurface horizons with somewhat poor internal drainagecaused the formation of perched temporary water tables during rainy seasons, thus the alternating reductionand oxidation of iron and manganese, their mobilization along fissures and root channels, and formation ofcharacteristics bleached streaks, reddish and blackish mottles, and nodules of iron and manganese .Neolithic/Early Bronze age was a time of rhexistasy and severe erosion affected soils on slopes (Fig. 5).Human settlements developed and forests were cleared. The reduced soil cover, combined with a climatedeterioration, with dry spells and strong winds, led to a sharp increase of water and wind erosion. Colluvialsediments accumulated on the lower parts of slopes, leading to the formation of very thick soils, like someSan Gimignano soils. Some aeolian sediments accumulated on soils placed on stable morphologicalpositions. It was the case of Poggio Golo soils, where the aeolian contribution increased the difference intexture between the surface and subsurface horizons, but also somehow changed the chemistry of the soil, asthe bases brought with the aeolian dusts saturated again the cation exchange complex and raised pH, haltingor reducing notably clay lessivage. During the successive bronze ages and Etruscan and Roman civilizations,the climatic conditions improved, and so did the land management.Pedogenesis of San Gimignano, Strada, San Quirico, Quercia, Valiano and Valiano aquic soils took placemainly during that time. A more or less developed structural subsurface horizon formed in all these soils, butwith very different functional characteristics. San Gimignano and Strada soils formed on marine sands,rather permeable and quite easily weatherable, which promoted the leaching of marine salts and the genesisof a thick horizon with subangular blocky aggregates, lined by organic matter and iron oxides, with a goodmacroporosity. Below the combined action of vegetation and climate, sands could be easily broken down;hence the firm

4 - 16



9

The wines obtained on a first group of soils (Valiano aquic, Cusona and Monte) had a good structure, a goodtypicity with high cherry and berry fruity. Unfortunately, the stability of wine sensorial profile during theyears was low. A second group (Quercia, Poggio Golo and San Quirico soils) exhibited good structure,typicity, with medium level of cherry and berry fruit and a good stability of wine sensorial profile. The wineobtained on the Strada soils showed medium structure, medium astringency, low cherry and berry fruity witha good stability of the sensorial profile. Finally, the wine of San Gimignano and Valiano soils had a lowstructure, high astringency, low typicity, all the years of trial (tab. 5).

Table 5 - Nobile di Montepulciano grape production and wine quality according to soil and soil group.The number of stars reflects the quantity (grape) or quality of the organoleptic evaluation (othervariables).

GroupSoils

NameGrape

productionHarvest

dateWine

structureWine

harmonyWine

typicityYear

stability

1 Monte and Cusona * **** ** ** ** *1 Valiano aquic **** *** ** ** ** *2 Quercia ** *** ** ** ** **2 Poggio Golo ** ** ** ** ** **2 San Quirico ** **** ** ** ** **3 Strada *** ** */** */** * **3 San Gimignano *** * * * * **3 Valiano ***** * * * * **

The interplay between geological and geomorphological events and the formation of soil functionalcharactersThe territory of Montepulciano municipality was almost all completely covered by the sea during EarlyPliocene. Only small islands, corresponding to the highest relives on Mesozoic rocks of the Chianti-Cetonamount ridge, were emerging from the sea. During Middle Pliocene, about four million years ago, the searegressed and left a succession of deposits, some hundred meters thick, which were rather sandy at their top(Fig. 1). Pliocene was a time of intense pedogenesis, leading to the formation of very deep and red soils,similar to those currently widespread in the subtropics. There are no more remnants of these kind of soils atMontepulciano, since here, like in most part of Italy, they have been completely eroded, as a consequence ofthe successive uplift of the area and climate changes. However, a significant example of Pliocene soil,currently preserved and cultivated with vine, has been reported in the not far away DOCG of Montalcino(Costantini and Priori, 2007).The Villafranchian (Late Pliocene and Early Pleistocene) saw the incision of the Pliocene deposits and thefilling up with lacustrine sediments of the territory comprised between the

10

Apennines and the Chianti-Cetona mount ridge (Fig. 2). During Middle and Late Pleistocene theVillafranchian lacustrine sediments were eroded and filled up with new fluvial and lacustrine deposits (Fig.3). Several fluvial terraces formed along the “Val di Chiana” paleo valley, which was crossed at that time bya southward streaming river, the “paleo Arno”. The deposits on the terraces and hilly lands underwent arather intense and prolonged pedogenesis, leading to the formation of deep and well developed soils, withcontrasted horizons. Alternating biorhexistasy phases, in dependence of both tectonic movements andclimate changes, caused the partial erosion of soils on slopes and the formation of deep colluvial soils onrelatively more stable morphological positions.During Early Holocene the morphology of the Montepulciano territory was very similar to the present (Fig.4). The progressive filling up of the valley caused a further swamping of the lowlands, while tectonicmovements led to the reversion of the drainage of the paleo Arno river. Poggio Golo soils dominated on thefluvial terraces formed in the previous geological era. The sharp contrast in texture between the surface andsubsurface horizons of Poggio Golo soils was created as a result of the prolonged time of pedogenesis,during which carbonates and other bases were leached throughout the profile, clay particles deflocculatedand were transported along the profile, where they progressively accumulated. Many macropores weretherefore plugged by clay particles, so that firmness of the subsurface horizon increased, whereas aircapacity and hydraulic conductivity decreased. Soil rooting depth and rootability (mass of soil that can beactually penetrated by roots, between and inside aggregates) were consequently constrained. The breakdownand weathering of rock fragments augmented the proportion of fine particles and released a great deal ofelements, in particular metals. The elements were sequestrated by clays, accumulated, and iron painted thesoils with characteristics reddish brown colours. Subsurface horizons with somewhat poor internal drainagecaused the formation of perched temporary water tables during rainy seasons, thus the alternating reductionand oxidation of iron and manganese, their mobilization along fissures and root channels, and formation ofcharacteristics bleached streaks, reddish and blackish mottles, and nodules of iron and manganese .Neolithic/Early Bronze age was a time of rhexistasy and severe erosion affected soils on slopes (Fig. 5).Human settlements developed and forests were cleared. The reduced soil cover, combined with a climatedeterioration, with dry spells and strong winds, led to a sharp increase of water and wind erosion. Colluvialsediments accumulated on the lower parts of slopes, leading to the formation of very thick soils, like someSan Gimignano soils. Some aeolian sediments accumulated on soils placed on stable morphologicalpositions. It was the case of Poggio Golo soils, where the aeolian contribution increased the difference intexture between the surface and subsurface horizons, but also somehow changed the chemistry of the soil, asthe bases brought with the aeolian dusts saturated again the cation exchange complex and raised pH, haltingor reducing notably clay lessivage. During the successive bronze ages and Etruscan and Roman civilizations,the climatic conditions improved, and so did the land management.Pedogenesis of San Gimignano, Strada, San Quirico, Quercia, Valiano and Valiano aquic soils took placemainly during that time. A more or less developed structural subsurface horizon formed in all these soils, butwith very different functional characteristics. San Gimignano and Strada soils formed on marine sands,rather permeable and quite easily weatherable, which promoted the leaching of marine salts and the genesisof a thick horizon with subangular blocky aggregates, lined by organic matter and iron oxides, with a goodmacroporosity. Below the combined action of vegetation and climate, sands could be easily broken down;hence the firm

4 - 17



11

sandy soil parent material easily transformed into friable soil, with excellent rootability. Weathering ofprimary minerals, aggregate dynamic, and mixing of marine layers of different particle size, operated byerosion and colluviation, favoured the accumulation of fine material and the genesis of the loamy texture.San Quirico and Quercia soils developed on marine clays and silts, almost impermeable and slowlyweatherable. The formation of the structured horizon was restricted to the depth of about one, one and halfmeter. The aggregates were mainly cemented by clay particles and took an angular blocky or prismaticshape, as a consequence of the wetting and dry cycles. Shrinking and swelling of clay aggregates wereparticularly intense in Quercia soils, which acquired vertic properties (cracking soils). As a consequence,macroporosity remained low, consistence firm, and permeability very low. Soil resistance to root penetrationwas high and restricted root development. Rootability was further limited by salinity of the unleached,massive deep soil horizons.Valiano and Valiano aquic soils formed both on Pleistocene fluvial-lacustrine clay. The continental insteadof marine origin of sediments was the main cause of their much lower salt content. Low salinity and lack ofsediment consolidation favoured drainage of soil parent material and pedogenesis. Valiano soils, inparticular, although formed from clays, developed thick deep horizons with well structured angular blockyaggregates, which did not limited root penetration. On the other hand, Valiano aquic formed on moreinstable slopes, so that pedogenesis was continuously offset by water erosion. As a consequence, aggregatesremained coarser and poorly developed. Microporosity dominated and internal drainage was constrained tothe planes between macroaggregates.The last 40 years of the Holocene has been a period of strong but localized rhexistasy (Fig. 6). Huge landlevelling and earth movements by bulldozing caused soil scalping and outcropping of almost unweatheredsediments, which started a new cycle of pedogenesis, strongly ruled by man. This was particularly the caseof many vineyard soils. Cusona and Monte are the most common examples of these kind of soils atMontepulciano, as well as in the province of Siena and in many other parts of Italy (Costantini and Barbetti,2008). Below the plough layer, their characteristic are very likely to the substratum. Since both kind of soilslack of a subsurface structured horizon, available water holding capacity is always very low, leaving vinewater uptake depending to a great extent on agricultural practices and on meteorology of the year. On theother hand, Cusona and Monte soils greatly differ as for air capacity and salinity, in dependence of thedifferent nature of the substrate, so that Monte soils are the most limiting environments for vine atMontepulciano.

CONCLUSIONSThe soils of the Montepulciano terroirs formed during different geological eras. In some cases theirdevelopment was a very long lasting process, which started during the Pleistocene, that is dozens ofthousands of years ago. It is important to underline that best soils for the Nobile di Montepulciano wineproduction formed as a consequence of particular natural and human induced geomorphological events,which are no more reproducible. Therefore soil functional characters of best Montepulciano terroirs areprecious and unique. Best terroirs not only fit the concept of “cru” (Seguin, 1986), but they can be alsoconsidered a soil heritage and a legacy (Costantini and L’Abate, 2009).

12

Man deeply influenced pedogenesis at Montepulciano since the medium Holocene, that is, since about 5,000years BP. While also in the past the effect of man on the environment was dramatic, comparable to that ofmajor geological events, it is during the last 50 years that the intensity of the man action has reached itsmaximum. With the advent of heavy mechanization and bulldozing, the pre-plantation activities of the newspecialized vineyards upset the land, leaving very different effects on soil functional characters, in function ofthe original soil type. Where the soils before vine plantation were deep and rather homogeneous, like SanGimignano, Poggio Golo, Valiano and Valiano aquic, soil functional characters remained the same, whereasthey changed significantly where soils were shallower. Soils on marine clays, in particular, due to the slowpedogenesis, were often shallow and vulnerable. Therefore San Quirico and Quercia soils could be easilyconverted into Monte soils, having worse functional characters for the Nobile di Montepulciano wine.Nevertheless, not always the consequences were negative. Cusona soils, for instance, which originated fromStrada soils, had better functional qualities for Sangiovese wine.As a final remark, it seems worthwhile to stress that the impact of land levelling and earth movements on soilfunctional characters should always be considered more carefully than currently done at Montepulciano, aswell as in many other viticultural territories. On top of that, it is recommended that soils of the best terroirs,formed in an irreproducible past, will be treated as cultural heritage, and therefore protected from destruction.

Figure 1 – The sea left the Montepulciano territory during the Middle Pliocene, leaving a thicksuccession of sediments, mainly sandy at the top. The Mesozoic isolated reliefs are part of theChianti – Centona mount ridge.

4 - 18



11

sandy soil parent material easily transformed into friable soil, with excellent rootability. Weathering ofprimary minerals, aggregate dynamic, and mixing of marine layers of different particle size, operated byerosion and colluviation, favoured the accumulation of fine material and the genesis of the loamy texture.San Quirico and Quercia soils developed on marine clays and silts, almost impermeable and slowlyweatherable. The formation of the structured horizon was restricted to the depth of about one, one and halfmeter. The aggregates were mainly cemented by clay particles and took an angular blocky or prismaticshape, as a consequence of the wetting and dry cycles. Shrinking and swelling of clay aggregates wereparticularly intense in Quercia soils, which acquired vertic properties (cracking soils). As a consequence,macroporosity remained low, consistence firm, and permeability very low. Soil resistance to root penetrationwas high and restricted root development. Rootability was further limited by salinity of the unleached,massive deep soil horizons.Valiano and Valiano aquic soils formed both on Pleistocene fluvial-lacustrine clay. The continental insteadof marine origin of sediments was the main cause of their much lower salt content. Low salinity and lack ofsediment consolidation favoured drainage of soil parent material and pedogenesis. Valiano soils, inparticular, although formed from clays, developed thick deep horizons with well structured angular blockyaggregates, which did not limited root penetration. On the other hand, Valiano aquic formed on moreinstable slopes, so that pedogenesis was continuously offset by water erosion. As a consequence, aggregatesremained coarser and poorly developed. Microporosity dominated and internal drainage was constrained tothe planes between macroaggregates.The last 40 years of the Holocene has been a period of strong but localized rhexistasy (Fig. 6). Huge landlevelling and earth movements by bulldozing caused soil scalping and outcropping of almost unweatheredsediments, which started a new cycle of pedogenesis, strongly ruled by man. This was particularly the caseof many vineyard soils. Cusona and Monte are the most common examples of these kind of soils atMontepulciano, as well as in the province of Siena and in many other parts of Italy (Costantini and Barbetti,2008). Below the plough layer, their characteristic are very likely to the substratum. Since both kind of soilslack of a subsurface structured horizon, available water holding capacity is always very low, leaving vinewater uptake depending to a great extent on agricultural practices and on meteorology of the year. On theother hand, Cusona and Monte soils greatly differ as for air capacity and salinity, in dependence of thedifferent nature of the substrate, so that Monte soils are the most limiting environments for vine atMontepulciano.

CONCLUSIONSThe soils of the Montepulciano terroirs formed during different geological eras. In some cases theirdevelopment was a very long lasting process, which started during the Pleistocene, that is dozens ofthousands of years ago. It is important to underline that best soils for the Nobile di Montepulciano wineproduction formed as a consequence of particular natural and human induced geomorphological events,which are no more reproducible. Therefore soil functional characters of best Montepulciano terroirs areprecious and unique. Best terroirs not only fit the concept of “cru” (Seguin, 1986), but they can be alsoconsidered a soil heritage and a legacy (Costantini and L’Abate, 2009).

12

Man deeply influenced pedogenesis at Montepulciano since the medium Holocene, that is, since about 5,000years BP. While also in the past the effect of man on the environment was dramatic, comparable to that ofmajor geological events, it is during the last 50 years that the intensity of the man action has reached itsmaximum. With the advent of heavy mechanization and bulldozing, the pre-plantation activities of the newspecialized vineyards upset the land, leaving very different effects on soil functional characters, in function ofthe original soil type. Where the soils before vine plantation were deep and rather homogeneous, like SanGimignano, Poggio Golo, Valiano and Valiano aquic, soil functional characters remained the same, whereasthey changed significantly where soils were shallower. Soils on marine clays, in particular, due to the slowpedogenesis, were often shallow and vulnerable. Therefore San Quirico and Quercia soils could be easilyconverted into Monte soils, having worse functional characters for the Nobile di Montepulciano wine.Nevertheless, not always the consequences were negative. Cusona soils, for instance, which originated fromStrada soils, had better functional qualities for Sangiovese wine.As a final remark, it seems worthwhile to stress that the impact of land levelling and earth movements on soilfunctional characters should always be considered more carefully than currently done at Montepulciano, aswell as in many other viticultural territories. On top of that, it is recommended that soils of the best terroirs,formed in an irreproducible past, will be treated as cultural heritage, and therefore protected from destruction.

Figure 1 – The sea left the Montepulciano territory during the Middle Pliocene, leaving a thicksuccession of sediments, mainly sandy at the top. The Mesozoic isolated reliefs are part of theChianti – Centona mount ridge.

4 - 19



13

Figure 2 – During Late Pliocene and Early Pleistocene (Villafranchian) the Montepulcianoterritory was first incised and then filled up with lacustrine deposits.

Figure 3 – During Middle and Late Pleistocene the Villafranchian lacustrine sediments wereeroded and filled up with new fluvial and lacustrine deposits. Several fluvial terraces formedalong the “Val di Chiana” paleo valley, which was at that time crossed by a southward streamingriver, the “paleo Arno”. The deposits on the terraces and hilly lands underwent a rather intenseand prolonged pedogenesis, leading to the formation of well developed soils.

14

Figure 4 – During Early Holocene the morphology of the Montepulciano territory was verysimilar to the present. The progressive filling up of the valley caused a further swamping of thelowlands, while tectonic movements led to the reversion of the drainage of the paleo Arno river.Poggio Golo soils dominated on the fluvial terraces formed in the previous geological era.

Figure 5 – Neolithic/Early Bronze age was a time of severe erosion of soils on slopes. Humansettlements developed and forests were cleared. The reduced soil cover, combined with a climatedeterioration, with dry spells and strong winds, led to a sharp increase of water and wind erosion.Colluvial sediments accumulated on the lower parts of slopes, leading to the formation of verythick soils, like some San Gimignano soils. Some aeolian sediments accumulated on soils placedon stable morphological positions. During the successive bronze ages and Etruscan and Romancivilizations, the climatic conditions improved, and so did the land management. Pedogenesis ofValiano, San Gimignano, Strada, San Quirico, and Quercia soils took place during that time.

4 - 20



13

Figure 2 – During Late Pliocene and Early Pleistocene (Villafranchian) the Montepulcianoterritory was first incised and then filled up with lacustrine deposits.

Figure 3 – During Middle and Late Pleistocene the Villafranchian lacustrine sediments wereeroded and filled up with new fluvial and lacustrine deposits. Several fluvial terraces formedalong the “Val di Chiana” paleo valley, which was at that time crossed by a southward streamingriver, the “paleo Arno”. The deposits on the terraces and hilly lands underwent a rather intenseand prolonged pedogenesis, leading to the formation of well developed soils.

14

Figure 4 – During Early Holocene the morphology of the Montepulciano territory was verysimilar to the present. The progressive filling up of the valley caused a further swamping of thelowlands, while tectonic movements led to the reversion of the drainage of the paleo Arno river.Poggio Golo soils dominated on the fluvial terraces formed in the previous geological era.

Figure 5 – Neolithic/Early Bronze age was a time of severe erosion of soils on slopes. Humansettlements developed and forests were cleared. The reduced soil cover, combined with a climatedeterioration, with dry spells and strong winds, led to a sharp increase of water and wind erosion.Colluvial sediments accumulated on the lower parts of slopes, leading to the formation of verythick soils, like some San Gimignano soils. Some aeolian sediments accumulated on soils placedon stable morphological positions. During the successive bronze ages and Etruscan and Romancivilizations, the climatic conditions improved, and so did the land management. Pedogenesis ofValiano, San Gimignano, Strada, San Quirico, and Quercia soils took place during that time.

4 - 21



15

Figure 6 – The last 40 years of the Holocene has been a period of strong but localizedrhexistasy. Huge land levelling and earth movements by bulldozing caused soil scalping andoutcropping of almost unweathered sediments, which started a new pedogenesis. This wasparticularly the case of vineyard soils. Cusona and Monte are the most common example of thesekind of soils.

BIBLIOGRAPHYCampostrini F., Costantini E.A.C., Mattivi F., Nicolini G., 1997. Effect of “Terroir” on quanti-

qualitative paramethers of “Vino Nobile di Montepulciano”. In: 1er colloque international“les terroirs viticoles”. Angers, France: INRA. 461-468.Consorzio Tutela Vini Soave e Recioto di Soave. Soave (VR), Italy.

Capezzuoli E., Priori S., Costantini E.A.C., Sandrelli F., 2009. Stratigraphic andpaleopedological aspects from the Middle Pleistocene continental deposits of the southernValdelsa Basin. Ital. i Geosci. (Boll. Soc. Geo. It.). 128, 2:395-406.

Carey V.A., Archer E., Barbeau G., Saayman D., 2008. Viticultural terroirs in Stellenbosch,South Africa. II. The interaction of Cabernet-Sauvignon and Sauvignon blanc withenvironment i Int. Sci. Vigne Vin, 42, 4:185-201.

Chapman D.M., Roby G., Ebeler S.E., Guinard J.X., Matthews, M.A., 2005. Sensory attributes ofCabernet Sauvignon wines made from vines with different water status. Aust. i GrapeWine Res., 11:339–347.

16

Choné, X., Van Leeuwen C., Chéry P., Ribéreau-Gayon P., 2001.Terroir influence on waterstatus and nitrogen status of non-irrigated Cabernet Sauvignon (Vitis vinifera): Vegetativedevelopment, must and wine composition. S. Af. J. Enol. Vitic., 22(1):8-15.

Costantini E.A.C., 1992. Study of the relationships between soil suitability for vine cultivation,wine quality and soil erosion through a territorial approach. Geoökoplus. III:1-14.

Costantini E.A.C., Barbetti R. 2008. Environmental and Visual Impact Analysis of Viticultureand Olive Tree Cultivation in the Province of Siena (Italy). Europ. J. Agronomy 28:412–426.

Costantini E.A.C., Barbetti R., Bucelli P., L’Abate G. Pellegrini S., Storchi P., 2008. Scaledependence of soil and climate functional characteristics for qualitative Sangiovese vineproduction. Proc. 31^ OIV congress Verona, CD-rom computer file. Org. Int. vigne et vin,Paris, France

Costantini E.A.C., Campostrini F., Arcara P.G., Cherubini P., Storchi P., Pierucci M., 1996. Soiland climate functional characters for grape ripening and wine quality of "Vino Nobile diMontepulciano". Acta Hort. 427 ISHS,:45-55.

Costantini E.A.C., L’Abate G., 2009. The soil cultural heritage of Italy: Geodatabase, maps, andpedodiversity evaluation. Quaternary International, 209:142-153.

Costantini E.A.C., Lizio-Bruno F., 1996. I suoli del comprensorio vitivinicolo di Montepulciano.Le loro caratteristiche, gli ambienti, i caratteri funzionali per la produzione di Vino Nobiledi Montepulciano. In: "Vino Nobile di Montepulciano: zonazione e valorizzazione dellerisorse naturali del territorio".F. Campostrini and E.A.C. Costantini. Firenze:RegioneToscana. 47-74.

Costantini E.A.C., Priori S., 2007. Pedogenesis of plinthite during early Pliocene in theMediterranean environment. Case study of a buried paleosol at Podere Renieri, central Italy.Catena, 71: 425–443.

Costantini E. A. C., Pellegrini S., Bucelli P., Storchi P., Vignozzi N., Barbetti R., Campagnolo S.,2009a. Relevance of the Lin’s and Host hydropedological models to predict grape yieldand wine quality. Hydrology and Earth System Sciences, 13:1635-1648

Costantini E.A.C., Priori S., Urban B., Hilgers A., Sauer D., Protano G., Trombino L., 2009b.Multidisciplinary characterization of the middle Holocene eolian deposits of the Elsa Riverbasin (central Italy). Quaternary International, 209:107-130

Deluc L.G., Quilici D.R., Decendit A., Grimplet J., Wheatley M.D., Schlauch K.A., MérillonJ.M., (...), Cramer G.R., 2009. Water deficit alters differentially metabolic pathwaysaffecting important flavor and quality traits in grape berries of Cabernet Sauvignon andChardonnay. BMC Genomics, 10, art. no. 212.

Ferrari G. A., Magaldi D., 1978. Sedimentologia e micropedologia dei paleosuoli sul terrazzoprincipale della Valdichiana (Arezzo). Geogr. Fis. e Dinam. Quat., 1:63-75.

Freeman B.M., Kliewer W.M., Stern P., 1982. Influence of windbreaks and climatic region ondiurnal fluctuation of leafwater potential, stomatal conductance, and leaf temperature ofgrapevines. Am. J. Enol. Vitic., 33:233-236.

Fregoni M., 2005. Viticoltura di qualità. Affi (VR):Phytoline.Hale C.R., 1977. Relation between potassium and the malate and tartrate contents of grape

berries. Vitis, 16:9-19.IUSS Working Group WRB, 2006. World reference base for soil resources 2006. World Soil

Resources Reports 103. 2nd edition. Rome: FAO.

4 - 22



15

Figure 6 – The last 40 years of the Holocene has been a period of strong but localizedrhexistasy. Huge land levelling and earth movements by bulldozing caused soil scalping andoutcropping of almost unweathered sediments, which started a new pedogenesis. This wasparticularly the case of vineyard soils. Cusona and Monte are the most common example of thesekind of soils.

BIBLIOGRAPHYCampostrini F., Costantini E.A.C., Mattivi F., Nicolini G., 1997. Effect of “Terroir” on quanti-

qualitative paramethers of “Vino Nobile di Montepulciano”. In: 1er colloque international“les terroirs viticoles”. Angers, France: INRA. 461-468.Consorzio Tutela Vini Soave e Recioto di Soave. Soave (VR), Italy.

Capezzuoli E., Priori S., Costantini E.A.C., Sandrelli F., 2009. Stratigraphic andpaleopedological aspects from the Middle Pleistocene continental deposits of the southernValdelsa Basin. Ital. i Geosci. (Boll. Soc. Geo. It.). 128, 2:395-406.

Carey V.A., Archer E., Barbeau G., Saayman D., 2008. Viticultural terroirs in Stellenbosch,South Africa. II. The interaction of Cabernet-Sauvignon and Sauvignon blanc withenvironment i Int. Sci. Vigne Vin, 42, 4:185-201.

Chapman D.M., Roby G., Ebeler S.E., Guinard J.X., Matthews, M.A., 2005. Sensory attributes ofCabernet Sauvignon wines made from vines with different water status. Aust. i GrapeWine Res., 11:339–347.

16

Choné, X., Van Leeuwen C., Chéry P., Ribéreau-Gayon P., 2001.Terroir influence on waterstatus and nitrogen status of non-irrigated Cabernet Sauvignon (Vitis vinifera): Vegetativedevelopment, must and wine composition. S. Af. J. Enol. Vitic., 22(1):8-15.

Costantini E.A.C., 1992. Study of the relationships between soil suitability for vine cultivation,wine quality and soil erosion through a territorial approach. Geoökoplus. III:1-14.

Costantini E.A.C., Barbetti R. 2008. Environmental and Visual Impact Analysis of Viticultureand Olive Tree Cultivation in the Province of Siena (Italy). Europ. J. Agronomy 28:412–426.

Costantini E.A.C., Barbetti R., Bucelli P., L’Abate G. Pellegrini S., Storchi P., 2008. Scaledependence of soil and climate functional characteristics for qualitative Sangiovese vineproduction. Proc. 31^ OIV congress Verona, CD-rom computer file. Org. Int. vigne et vin,Paris, France

Costantini E.A.C., Campostrini F., Arcara P.G., Cherubini P., Storchi P., Pierucci M., 1996. Soiland climate functional characters for grape ripening and wine quality of "Vino Nobile diMontepulciano". Acta Hort. 427 ISHS,:45-55.

Costantini E.A.C., L’Abate G., 2009. The soil cultural heritage of Italy: Geodatabase, maps, andpedodiversity evaluation. Quaternary International, 209:142-153.

Costantini E.A.C., Lizio-Bruno F., 1996. I suoli del comprensorio vitivinicolo di Montepulciano.Le loro caratteristiche, gli ambienti, i caratteri funzionali per la produzione di Vino Nobiledi Montepulciano. In: "Vino Nobile di Montepulciano: zonazione e valorizzazione dellerisorse naturali del territorio".F. Campostrini and E.A.C. Costantini. Firenze:RegioneToscana. 47-74.

Costantini E.A.C., Priori S., 2007. Pedogenesis of plinthite during early Pliocene in theMediterranean environment. Case study of a buried paleosol at Podere Renieri, central Italy.Catena, 71: 425–443.

Costantini E. A. C., Pellegrini S., Bucelli P., Storchi P., Vignozzi N., Barbetti R., Campagnolo S.,2009a. Relevance of the Lin’s and Host hydropedological models to predict grape yieldand wine quality. Hydrology and Earth System Sciences, 13:1635-1648

Costantini E.A.C., Priori S., Urban B., Hilgers A., Sauer D., Protano G., Trombino L., 2009b.Multidisciplinary characterization of the middle Holocene eolian deposits of the Elsa Riverbasin (central Italy). Quaternary International, 209:107-130

Deluc L.G., Quilici D.R., Decendit A., Grimplet J., Wheatley M.D., Schlauch K.A., MérillonJ.M., (...), Cramer G.R., 2009. Water deficit alters differentially metabolic pathwaysaffecting important flavor and quality traits in grape berries of Cabernet Sauvignon andChardonnay. BMC Genomics, 10, art. no. 212.

Ferrari G. A., Magaldi D., 1978. Sedimentologia e micropedologia dei paleosuoli sul terrazzoprincipale della Valdichiana (Arezzo). Geogr. Fis. e Dinam. Quat., 1:63-75.

Freeman B.M., Kliewer W.M., Stern P., 1982. Influence of windbreaks and climatic region ondiurnal fluctuation of leafwater potential, stomatal conductance, and leaf temperature ofgrapevines. Am. J. Enol. Vitic., 33:233-236.

Fregoni M., 2005. Viticoltura di qualità. Affi (VR):Phytoline.Hale C.R., 1977. Relation between potassium and the malate and tartrate contents of grape

berries. Vitis, 16:9-19.IUSS Working Group WRB, 2006. World reference base for soil resources 2006. World Soil

Resources Reports 103. 2nd edition. Rome: FAO.

4 - 23



17

Kao YY, Harding SA, Tsai CJ, 2002. Differential expression of two distinct phenylalanineammonia-lyase genes in condensed tannin-accumulating and lignifying cells of quakingaspen. Plant Physiology. 130:756-760.

Losacco U., 1944. Il bacino postpliocenico della Valdichiana. L'Universo, 2:45-71.Lanyon D.M, Cass A., Hansen. D., 2004. The effect of soil properties on vine performance.

CSIRO Land and Water Technical Report No. 34/04Maltman A., 2008. The Role of Vineyard Geology in Wine Typicity. J. Wine Research., 19,1:1-17.Meinert L.D., Busacca A.J., 2000. Geology and Wine 3: Terroirs of the Walla Walla Valley

appellation, southeastern Washington State, USA. Geoscience Canada, 27:149–170.Myburgh, P.A., Van Zuyl, J.L., Conradie, W.J., 1996. Effect of soil depth on growth and water

consumption of young vitis vinifera L. cv. Pinot Noir. South African J. Enol. Vitic., 17: 53–62.

Nikolaou N., Magdalini N., Koukourikou A., Karagiannidis N., 2000. Effects of variousrootstocks on xylem exudates cytokinin content, nutrient uptake and growth patterns ofgrapevine Vitis vinifera L. cv. Thompson seedless. Agronomie, 20:363–373.

Panont C.A., Bogoni M., Montoldi A., Scienza A., 1997. Improvement of sparkling winesproduction by a zoning approach in Franciacorta (Lombardy, Italy). In: Acts Colloqueinternational “Les terroirs viticoles”, Angers, 17-18 juillet 1996, 454-460.

Peyrot des Gachons C., Van Leeuwen C., Tominaga T., Soyer J.P., Gaudillère J.P., DubourdieuD., 2005. Influence of water and nitrogen deficit on fruit ripening and aroma potential ofVitis vinifera L. cv Sauvignon blanc in field conditions. J. Sci. Food Agric., 85:73–85.

Priori S., Costantini E.A.C., Capezzuoli E., Protano G., Hilgers A., Sauer D., Sandrelli F. (2008).Pedostratigraphy of Terra Rossa and Quaternary geological evolution of a lacustrinelimestone plateau in central Italy. J. Plant Nutr. Soil Sci. 171:509-523.

Seguin G., 1986. “Terroirs” and pedology of vine growing. Experientia, 42:861-873.Scotti C., 2006. Emilia-Romagna: dalla conoscenza del suolo alla qualità del vino. Il suolo, 1-3Stoll M., Stuebinger M., Lafontaine M., Schultz H. R., 2008. Radiative and thermal effects on

fruit ripening induced by differences in soil colour. VII International terroir Congress. NyonTrégoat O., Gaudillère J.P., Choné X., Van Leeuwen C., 2002.The assessment of vine water and

nitrogen uptake by means of physiological indicators. Influence on vine development andberry potential. (Vitis vinifera L. cv. Merlot, 2000, Bordeaux). J. Int. Sci.Vigne Vin,

36,(3):133-142.USDA United States Department of Agriculture, 2010. Keys to Soil Taxonomy (eleventh

edition). 338.Van Leeuwen C., Seguin G. 1997. Incidence de la nature du sol et du cépage sur la maturation du

raisin, à Saint emilion, en 1995.In: Colloque international “Les terroirs viticoles”, 17-18juillet 1996. Angers, 154-157.

Van Leeuwen C., Friant P.,Choné X., Tregoat O., Koundouras S., Dubourdieu D., 2004.Influence of Climate, Soil, and Cultivar on Terroir. Am. J. Enol. Vitic. 55,3: 207-217.

Van Leeuwen C., Seguin G., 2006. 'The concept of terroir in viticulture', J. Wine Research,17, 1 :1- 10

Vaudour E., 2003. Les terroirs viticoles. Définitions, caractérisation et protection. Ed. Dunod,Paris, France.

White R. E. 2003. Soils for fine wines. New York: Oxford University Press.

18

White R., Balachandra L., Edis R., Chen D., 2007. The soil component of terroir. i Int. Sci.Vigne Vin, 41, 1:9-18

Witbooi, E.H., V.A. Carey, J.E. Hoffman & A.E. Strever. 2008. The relationship between soilsurface colour and the performance of Vitis vinifera L. Cv. Cabernet Sauvignon inStellenbosch Wine of Origin District. I. Vegetative growth. In: Thirty First World Congressof Vine and Wine and the Sixth General Assembly of the OIV, Verona, Italy.

Witbooi, E.H., V.A. Carey, J.E. Hoffman & A.E. Strever. 2008. The relationship between soilsurface colour and the performance of Vitis vinifera L. Cv. Cabernet Sauvignon inStellenbosch Wine of Origin District. II. Yield, berry – and wine composition. In: ThirtyFirst World Congress of Vine and Wine and the Sixth General Assembly of the OIV,Verona, Italy.

Zsófi Zs., Gál L., Szilágyi Z., SzLIcs E., Marschall M., Nagy Z., Bálo B., 2009. Use of stomatalconductance and pre-dawn water potential to classify terroir for the grape varietyKékfrankos. Australian Journal of Grape and Wine Research, 15 (1):36-47

4 - 24



17

Kao YY, Harding SA, Tsai CJ, 2002. Differential expression of two distinct phenylalanineammonia-lyase genes in condensed tannin-accumulating and lignifying cells of quakingaspen. Plant Physiology. 130:756-760.

Losacco U., 1944. Il bacino postpliocenico della Valdichiana. L'Universo, 2:45-71.Lanyon D.M, Cass A., Hansen. D., 2004. The effect of soil properties on vine performance.

CSIRO Land and Water Technical Report No. 34/04Maltman A., 2008. The Role of Vineyard Geology in Wine Typicity. J. Wine Research., 19,1:1-17.Meinert L.D., Busacca A.J., 2000. Geology and Wine 3: Terroirs of the Walla Walla Valley

appellation, southeastern Washington State, USA. Geoscience Canada, 27:149–170.Myburgh, P.A., Van Zuyl, J.L., Conradie, W.J., 1996. Effect of soil depth on growth and water

consumption of young vitis vinifera L. cv. Pinot Noir. South African J. Enol. Vitic., 17: 53–62.

Nikolaou N., Magdalini N., Koukourikou A., Karagiannidis N., 2000. Effects of variousrootstocks on xylem exudates cytokinin content, nutrient uptake and growth patterns ofgrapevine Vitis vinifera L. cv. Thompson seedless. Agronomie, 20:363–373.

Panont C.A., Bogoni M., Montoldi A., Scienza A., 1997. Improvement of sparkling winesproduction by a zoning approach in Franciacorta (Lombardy, Italy). In: Acts Colloqueinternational “Les terroirs viticoles”, Angers, 17-18 juillet 1996, 454-460.

Peyrot des Gachons C., Van Leeuwen C., Tominaga T., Soyer J.P., Gaudillère J.P., DubourdieuD., 2005. Influence of water and nitrogen deficit on fruit ripening and aroma potential ofVitis vinifera L. cv Sauvignon blanc in field conditions. J. Sci. Food Agric., 85:73–85.

Priori S., Costantini E.A.C., Capezzuoli E., Protano G., Hilgers A., Sauer D., Sandrelli F. (2008).Pedostratigraphy of Terra Rossa and Quaternary geological evolution of a lacustrinelimestone plateau in central Italy. J. Plant Nutr. Soil Sci. 171:509-523.

Seguin G., 1986. “Terroirs” and pedology of vine growing. Experientia, 42:861-873.Scotti C., 2006. Emilia-Romagna: dalla conoscenza del suolo alla qualità del vino. Il suolo, 1-3Stoll M., Stuebinger M., Lafontaine M., Schultz H. R., 2008. Radiative and thermal effects on

fruit ripening induced by differences in soil colour. VII International terroir Congress. NyonTrégoat O., Gaudillère J.P., Choné X., Van Leeuwen C., 2002.The assessment of vine water and

nitrogen uptake by means of physiological indicators. Influence on vine development andberry potential. (Vitis vinifera L. cv. Merlot, 2000, Bordeaux). J. Int. Sci.Vigne Vin,

36,(3):133-142.USDA United States Department of Agriculture, 2010. Keys to Soil Taxonomy (eleventh

edition). 338.Van Leeuwen C., Seguin G. 1997. Incidence de la nature du sol et du cépage sur la maturation du

raisin, à Saint emilion, en 1995.In: Colloque international “Les terroirs viticoles”, 17-18juillet 1996. Angers, 154-157.

Van Leeuwen C., Friant P.,Choné X., Tregoat O., Koundouras S., Dubourdieu D., 2004.Influence of Climate, Soil, and Cultivar on Terroir. Am. J. Enol. Vitic. 55,3: 207-217.

Van Leeuwen C., Seguin G., 2006. 'The concept of terroir in viticulture', J. Wine Research,17, 1 :1- 10

Vaudour E., 2003. Les terroirs viticoles. Définitions, caractérisation et protection. Ed. Dunod,Paris, France.

White R. E. 2003. Soils for fine wines. New York: Oxford University Press.

18

White R., Balachandra L., Edis R., Chen D., 2007. The soil component of terroir. i Int. Sci.Vigne Vin, 41, 1:9-18

Witbooi, E.H., V.A. Carey, J.E. Hoffman & A.E. Strever. 2008. The relationship between soilsurface colour and the performance of Vitis vinifera L. Cv. Cabernet Sauvignon inStellenbosch Wine of Origin District. I. Vegetative growth. In: Thirty First World Congressof Vine and Wine and the Sixth General Assembly of the OIV, Verona, Italy.

Witbooi, E.H., V.A. Carey, J.E. Hoffman & A.E. Strever. 2008. The relationship between soilsurface colour and the performance of Vitis vinifera L. Cv. Cabernet Sauvignon inStellenbosch Wine of Origin District. II. Yield, berry – and wine composition. In: ThirtyFirst World Congress of Vine and Wine and the Sixth General Assembly of the OIV,Verona, Italy.

Zsófi Zs., Gál L., Szilágyi Z., SzLIcs E., Marschall M., Nagy Z., Bálo B., 2009. Use of stomatalconductance and pre-dawn water potential to classify terroir for the grape varietyKékfrankos. Australian Journal of Grape and Wine Research, 15 (1):36-47

4 - 25



INFLUENCE OF SOIL CHARACTERISTICS ON VINE GROWTH,

PLANT NUTRIENT LEVELS AND JUICE PROPERTIES:

A MULTI-YEAR ANALYSIS

J.-J. Lambert*1, J. Fujita1, C. Gruenwald1, R.A. Dahlgren2, H. Heymann1,

and J.A. Wolpert1,3

1Department of Viticulture and Enology, 2Department of Land, Air and Water Resources, UC Davis, and 3UC

Cooperative Extension, University of California at Davis, One Shields Avenue, Davis CA 95616 USA.

*Corresponding Author: [email protected]

ABSTRACT

Soil physical and chemical properties affect vine nutrition, as indicated by leaf and petiole

nutrient content, in a way that may directly impact wine properties. The goal of this multi-year

project is to study the relationship between vineyard soils and the wines produced on them

using a variety of biogeochemical and mineral analyses, coupled with an analysis of vine

properties and juice characteristics. This study examines leaf and petiole nutrient levels, as

well as fruit and juice characteristics, of own-rooted Cabernet Sauvignon vines grown on four

distinct soil types in the same Paso Robles vineyard. The soils were classified as Palexeralfs,

Haploxeralfs, Haploxerolls and Haploxererts. The four soils exhibited important

morphological differences in color, coarse fragment content, texture, water holding capacity,

and hydraulic conductivity. The soils also showed important differences in chemical

characteristics and nutrient availability. The soils covered contiguous vineyard patches

planted with the same cultivar, on its own roots. The vineyard was irrigated and fertilized.

Mesoclimatic conditions and slope aspect were similar. Soils were analyzed for physical and

chemical differences to determine the influence of the four contrasting soil types on

differences in vine growth, water stress and plant nutrient levels. Differences in cation

exchange capacity and cationic balance in the soil solution appeared to affect nutrient

availability to the vines, and likely contributed to the observed differences in the plant and

fruit characteristics. Berries harvested on the four blocks exhibited different sensory attributes,

as determined by a tasting panel. In an analysis of data from three consecutive growing

seasons, many of the observed differences in plant vigor between vineyard blocks were

consistent from year to year, as were differences in fruit yield and juice properties. Taken

together, these findings support a role for soil texture, water and nutrient availability on vine

and fruit parameters, and emphasize that differences in soil properties within a single vineyard

may require site-specific management practices.

KEYWORDS

Soil – Biogeochemistry – Nutrients – Leaf – Petiole - Management

INTRODUCTION

The goal of this multi-year project is to study the relationship between vineyard soil

properties (i.e., mineralogy, nutrient levels, water availability), vine growth characteristics,

juice and wine properties. To date, although much speculation has been devoted to this topic

in the popular wine press, few studies have systematically evaluated the relationship between

soil characteristics, vine vigor and fruit or juice properties (Andrés-de-Prado et al., 2007;

1

Tomasi et al., 2006). Here we present results of an ongoing, multi-year study performed with

the cooperation of J. Lohr Vineyards, Paso Robles, CA. The company determined that soils

with different chemical and physical properties existed in a contiguous field of Cabernet

Sauvignon. The vines in this vineyard were planted at the same time, on the same rootstock,

and received similar management practices. Mesoclimatic conditions, as determined by

elevation and slope aspect, were also similar. Upon detailed analysis, the four soils were

found to be significantly different, and in an informal tasting, small lot wines prepared from

vines growing on each of the four sites were also perceived to have different sensory

properties. Field observations and laboratory analyses over three growing seasons revealed

consistent trends in vine vigor as well as leaf and petiole nutrient levels between sites.

MATERIALS AND METHODS

Soil Analyses

Soil Sampling. Vineyard soils were sampled at four sites designated as Blocks 52, 53, 56

and 57. Two soil pits were excavated in each block, for a total of eight pits. Soil horizons were

described in the field following the National Cooperative Soil Survey field description manual

(Soil Survey Staff, 1993). A Trimble GeoXH GPS was used to georeference the pit locations

and vines, allowing for precise mapping of soil variability within the vineyard.

Soil Physical and Chemical Analysis. Solid-phase soil characterization was performed for

replicate samples from each site. Soil texture was analyzed by laser granulometry and by the

hydrometer method. Soil pH and electrical conductivity were measured in the laboratory using

a 1:1 soil:water paste. Soil samples were processed by passage through a 2 mm sieve to

separate coarse fragments from the fine earth fraction. Soil chemical analyses were performed

in the UC Davis DANR Analytical Laboratory for exchangeable cations (Ca, Mg, K, Na),

CEC, pH, EC, total N, NO3, NH4, P, S, Zn, Cu, Mn, Fe, Si, and B.

Soil Solution Chemistry. Soil solutions were collected in situ beginning in August, 2007

using implanted suction devices located at depths of 12, 24 and 36 inches under drip emitters.

Samples were collected twice: at harvest time and at two months post-harvest during Year 1,

and at monthly intervals during Year 2. The following parameters are being measured: pH,

EC, K, NO3, NH4, Si, B, CO3, SO4, Cl, Mg, Ca, and K; these analyses are ongoing.

Plant Analyses

Leaf Petiole and Blade Sampling. Leaf petioles and blades were collected from 3 sets of 10

replicate vines from 2 sampling sites within each of the four soil types, for a total of 240

vines. All vines were marked with identification tags, and vine locations were georeferenced.

Petiole and blade sampling was repeated at three phenological stages in 2007 and 2008:

bloom, veraison, and harvest. Bloom samples consisted of leaves located opposite the basal-

most cluster, while the most recent fully expanded leaves were collected at veraison and

harvest. At each sampling date, leaves and petioles were separated, air-dried at 60°C, ground

at 60-mesh in a Wiley mill, and sent to the DANR Analytical Laboratory at UC Davis for

analysis of total N, NO3, NH4, P, K, Ca, Mg, S, Zn, Cu, Mn, Fe, Si, and B.

Plant and Soil Water Status. Mid-day plant water potentials (Ψ) were measured manually

with a pressure bomb at bloom, veraison and harvest on the same vines within each vineyard

block. Soil moisture was measured at 30, 60 and 90 cm using TDR probes in embedded in soil

pit walls at four of the sampling sites. Canopy temperature sensors were installed at each site.

Temperature and moisture data are recorded at 30-minute intervals. Soil moisture content was

determined by gravimetric measurements in the laboratory, and the results compared to those

obtained using the TDR probe.

24 - 26



INFLUENCE OF SOIL CHARACTERISTICS ON VINE GROWTH,

PLANT NUTRIENT LEVELS AND JUICE PROPERTIES:

A MULTI-YEAR ANALYSIS

J.-J. Lambert*1, J. Fujita1, C. Gruenwald1, R.A. Dahlgren2, H. Heymann1,

and J.A. Wolpert1,3

1Department of Viticulture and Enology, 2Department of Land, Air and Water Resources, UC Davis, and 3UC

Cooperative Extension, University of California at Davis, One Shields Avenue, Davis CA 95616 USA.

*Corresponding Author: [email protected]

ABSTRACT

Soil physical and chemical properties affect vine nutrition, as indicated by leaf and petiole

nutrient content, in a way that may directly impact wine properties. The goal of this multi-year

project is to study the relationship between vineyard soils and the wines produced on them

using a variety of biogeochemical and mineral analyses, coupled with an analysis of vine

properties and juice characteristics. This study examines leaf and petiole nutrient levels, as

well as fruit and juice characteristics, of own-rooted Cabernet Sauvignon vines grown on four

distinct soil types in the same Paso Robles vineyard. The soils were classified as Palexeralfs,

Haploxeralfs, Haploxerolls and Haploxererts. The four soils exhibited important

morphological differences in color, coarse fragment content, texture, water holding capacity,

and hydraulic conductivity. The soils also showed important differences in chemical

characteristics and nutrient availability. The soils covered contiguous vineyard patches

planted with the same cultivar, on its own roots. The vineyard was irrigated and fertilized.

Mesoclimatic conditions and slope aspect were similar. Soils were analyzed for physical and

chemical differences to determine the influence of the four contrasting soil types on

differences in vine growth, water stress and plant nutrient levels. Differences in cation

exchange capacity and cationic balance in the soil solution appeared to affect nutrient

availability to the vines, and likely contributed to the observed differences in the plant and

fruit characteristics. Berries harvested on the four blocks exhibited different sensory attributes,

as determined by a tasting panel. In an analysis of data from three consecutive growing

seasons, many of the observed differences in plant vigor between vineyard blocks were

consistent from year to year, as were differences in fruit yield and juice properties. Taken

together, these findings support a role for soil texture, water and nutrient availability on vine

and fruit parameters, and emphasize that differences in soil properties within a single vineyard

may require site-specific management practices.

KEYWORDS

Soil – Biogeochemistry – Nutrients – Leaf – Petiole - Management

INTRODUCTION

The goal of this multi-year project is to study the relationship between vineyard soil

properties (i.e., mineralogy, nutrient levels, water availability), vine growth characteristics,

juice and wine properties. To date, although much speculation has been devoted to this topic

in the popular wine press, few studies have systematically evaluated the relationship between

soil characteristics, vine vigor and fruit or juice properties (Andrés-de-Prado et al., 2007;

1

Tomasi et al., 2006). Here we present results of an ongoing, multi-year study performed with

the cooperation of J. Lohr Vineyards, Paso Robles, CA. The company determined that soils

with different chemical and physical properties existed in a contiguous field of Cabernet

Sauvignon. The vines in this vineyard were planted at the same time, on the same rootstock,

and received similar management practices. Mesoclimatic conditions, as determined by

elevation and slope aspect, were also similar. Upon detailed analysis, the four soils were

found to be significantly different, and in an informal tasting, small lot wines prepared from

vines growing on each of the four sites were also perceived to have different sensory

properties. Field observations and laboratory analyses over three growing seasons revealed

consistent trends in vine vigor as well as leaf and petiole nutrient levels between sites.


Soil Analyses

Soil Sampling. Vineyard soils were sampled at four sites designated as Blocks 52, 53, 56

and 57. Two soil pits were excavated in each block, for a total of eight pits. Soil horizons were

described in the field following the National Cooperative Soil Survey field description manual

(Soil Survey Staff, 1993). A Trimble GeoXH GPS was used to georeference the pit locations

and vines, allowing for precise mapping of soil variability within the vineyard.

Soil Physical and Chemical Analysis. Solid-phase soil characterization was performed for

replicate samples from each site. Soil texture was analyzed by laser granulometry and by the

hydrometer method. Soil pH and electrical conductivity were measured in the laboratory using

a 1:1 soil:water paste. Soil samples were processed by passage through a 2 mm sieve to

separate coarse fragments from the fine earth fraction. Soil chemical analyses were performed

in the UC Davis DANR Analytical Laboratory for exchangeable cations (Ca, Mg, K, Na),

CEC, pH, EC, total N, NO3, NH4, P, S, Zn, Cu, Mn, Fe, Si, and B.

Soil Solution Chemistry. Soil solutions were collected in situ beginning in August, 2007

using implanted suction devices located at depths of 12, 24 and 36 inches under drip emitters.

Samples were collected twice: at harvest time and at two months post-harvest during Year 1,

and at monthly intervals during Year 2. The following parameters are being measured: pH,

EC, K, NO3, NH4, Si, B, CO3, SO4, Cl, Mg, Ca, and K; these analyses are ongoing.

Plant Analyses

Leaf Petiole and Blade Sampling. Leaf petioles and blades were collected from 3 sets of 10

replicate vines from 2 sampling sites within each of the four soil types, for a total of 240

vines. All vines were marked with identification tags, and vine locations were georeferenced.

Petiole and blade sampling was repeated at three phenological stages in 2007 and 2008:

bloom, veraison, and harvest. Bloom samples consisted of leaves located opposite the basal-

most cluster, while the most recent fully expanded leaves were collected at veraison and

harvest. At each sampling date, leaves and petioles were separated, air-dried at 60°C, ground

at 60-mesh in a Wiley mill, and sent to the DANR Analytical Laboratory at UC Davis for

analysis of total N, NO3, NH4, P, K, Ca, Mg, S, Zn, Cu, Mn, Fe, Si, and B.

Plant and Soil Water Status. Mid-day plant water potentials (Ψ) were measured manually

with a pressure bomb at bloom, veraison and harvest on the same vines within each vineyard

block. Soil moisture was measured at 30, 60 and 90 cm using TDR probes in embedded in soil

pit walls at four of the sampling sites. Canopy temperature sensors were installed at each site.

Temperature and moisture data are recorded at 30-minute intervals. Soil moisture content was

determined by gravimetric measurements in the laboratory, and the results compared to those

obtained using the TDR probe.

24 - 27



Vine Growth and Fruit Production. Vine trunk diameters were measured at heights of 25

and 50 cm. Root counts were determined in the field by hand counting root intercepts using a

10x10 cm counting grid. Vine canopy density was measured using a metering system,

developed by Dr. Mark Battany, UC Cooperative Extension, based on photovoltaic panels. At

harvest, fruit yield was determined by weighing the harvest and dividing by the number of

vines at each site. Berry clusters were counted and weighed. The number of berries per cluster

was counted for 20 to 25 clusters per group of observation vines. Pruning weights were

determined in late December 2007 and in January 2009 for all 240 tagged vines; this included

three groups of ten vines per soil sampling site.

Juice and Wine Analyses

At harvest, juice samples were analyzed for pH, sugar content (°Brix), Total Acidity, Yeast

Available Nitrogen, and Free Amino Nitrogen (NOPA). Sensory analysis of berries is ongoing

in collaboration with Dr. Hildegarde Heymann, UC Davis. Sensory analysis panels have been

created using volunteers; 3 replicate tasting events have been held with the same tasting panel

comprised of 8 volunteers. Volunteers are blinded to the identity of the samples. Berries were

sampled from vines in each of the areas surrounding the 8 individual soil pits and

cryopreserved at -80ºC prior to tasting.


Soil characterization

The soils in the four vineyard plots were sampled and analyzed during the first year of the

study (Lambert et al., 2008). The four vineyard plots and the eight sampling sites (two per

plot) are shown in Fig. 1. The soils situated in the four vineyard blocks differed significantly

in chemical and physical properties. The soils in Blocks 56 and 57 were classified as distinct,

yet related Alfisols: the soil in Block 57 was a fine, smectitic, thermic Typic Palexeralfs and

the soil in Block 56 was a fine-loamy, mixed, superactive, thermic Typic Haploxeralfs. The

soils in Block 53 were typical of Vertisols: fine, smectitic, thermic Haploxererts with greater

than 30% clay content and a tendency to ‘shrink/swell’ behavior. Finally, the soils in Block 52

had calcareous seams, laminar lime concretions and an angular, blocky structure in the

subsoil. These soils were characterized as Mollisols: fine-loamy, mixed, superactive, thermic

Calcic Haploxerolls (Lambert et al., 2008). Soil chemical analyses revealed several striking

differences between sites. Soil extract Nitrogen and Phosphorous were comparatively low in

the Mollisols. In addition, both the Mollisols and Vertisols had low K+ availability throughout

the profiles. Potassium levels were higher in the Alfisols, but only in the superficial horizons.

Electrical conductivity was particularly high in the Ca-rich Mollisols and increased with

depth.

Plant tissue nutrient levels

Plant tissue (petiole and blade) levels of phosphorus (P), potassium (K) and magnesium

(Mg) varied consistently between sites over the three-year study period. Vines grown on the

Haploxerolls (Mollisols) had consistently low levels of petiole and blade P at all three

phonological stages, approaching the threshold (0.1%) considered as deficient at harvest time

(Klein et al., 2000), as shown in Fig. 2. Vines grown on the Mollisols and on the Vertisols had

higher levels of petiole K at veraison and harvest than vines grown on Alfisols, as shown in

Fig. 3. This was also reflected in a high K/Mg ratio in petioles of vines grown on Mollisols

and Vertisols, suggesting Mg deficiency (data not shown) (Delas, 1996). Conversely, petiole

Mg levels were highest at veraison and harvest in vines grown on the Alfisols. Petiole Mn

3

levels were consistently low in vines grown in the calcic Mollisols, as explained by the

insolubility of Mn in calcareous soils with pH 7.5-8 (data not shown). Petiole N levels showed

no significant variation between sites.

Figure 1. Four contrasting soil types in a Paso Robles Cabernet Sauvignon vineyard. Soil

types and block numbers are shown. Numbered white circles indicate the locations of soil

sampling pits.

Figure 2. Petiole and Blade P levels are consistently low in plants grown on Calcic

Mollisols. Data shown are three-year averages (2007, 08, 09). Numbers indicate pit sampling

sites as seen in Fig. 1; sites 3 and 7 were located in Block 52 (Mollisol).

Plant vigor and fruit yields

Plant root counts varied considerably with soil type. In the Alfisols, the presence of a

compacted layer at depth prevented significant root penetration beyond 40-50 cm. In the

Mollisols, root density was greatest at depths below 50 cm, likely due to the high salt content

and electrical conductivity at the surface horizon. The Vertisols were characterized by good

44 - 28



Vine Growth and Fruit Production. Vine trunk diameters were measured at heights of 25

and 50 cm. Root counts were determined in the field by hand counting root intercepts using a

10x10 cm counting grid. Vine canopy density was measured using a metering system,

developed by Dr. Mark Battany, UC Cooperative Extension, based on photovoltaic panels. At

harvest, fruit yield was determined by weighing the harvest and dividing by the number of

vines at each site. Berry clusters were counted and weighed. The number of berries per cluster

was counted for 20 to 25 clusters per group of observation vines. Pruning weights were

determined in late December 2007 and in January 2009 for all 240 tagged vines; this included

three groups of ten vines per soil sampling site.

Juice and Wine Analyses

At harvest, juice samples were analyzed for pH, sugar content (°Brix), Total Acidity, Yeast

Available Nitrogen, and Free Amino Nitrogen (NOPA). Sensory analysis of berries is ongoing

in collaboration with Dr. Hildegarde Heymann, UC Davis. Sensory analysis panels have been

created using volunteers; 3 replicate tasting events have been held with the same tasting panel

comprised of 8 volunteers. Volunteers are blinded to the identity of the samples. Berries were

sampled from vines in each of the areas surrounding the 8 individual soil pits and

cryopreserved at -80ºC prior to tasting.


Soil characterization

The soils in the four vineyard plots were sampled and analyzed during the first year of the

study (Lambert et al., 2008). The four vineyard plots and the eight sampling sites (two per

plot) are shown in Fig. 1. The soils situated in the four vineyard blocks differed significantly

in chemical and physical properties. The soils in Blocks 56 and 57 were classified as distinct,

yet related Alfisols: the soil in Block 57 was a fine, smectitic, thermic Typic Palexeralfs and

the soil in Block 56 was a fine-loamy, mixed, superactive, thermic Typic Haploxeralfs. The

soils in Block 53 were typical of Vertisols: fine, smectitic, thermic Haploxererts with greater

than 30% clay content and a tendency to ‘shrink/swell’ behavior. Finally, the soils in Block 52

had calcareous seams, laminar lime concretions and an angular, blocky structure in the

subsoil. These soils were characterized as Mollisols: fine-loamy, mixed, superactive, thermic

Calcic Haploxerolls (Lambert et al., 2008). Soil chemical analyses revealed several striking

differences between sites. Soil extract Nitrogen and Phosphorous were comparatively low in

the Mollisols. In addition, both the Mollisols and Vertisols had low K+ availability throughout

the profiles. Potassium levels were higher in the Alfisols, but only in the superficial horizons.

Electrical conductivity was particularly high in the Ca-rich Mollisols and increased with

depth.

Plant tissue nutrient levels

Plant tissue (petiole and blade) levels of phosphorus (P), potassium (K) and magnesium

(Mg) varied consistently between sites over the three-year study period. Vines grown on the

Haploxerolls (Mollisols) had consistently low levels of petiole and blade P at all three

phonological stages, approaching the threshold (0.1%) considered as deficient at harvest time

(Klein et al., 2000), as shown in Fig. 2. Vines grown on the Mollisols and on the Vertisols had

higher levels of petiole K at veraison and harvest than vines grown on Alfisols, as shown in

Fig. 3. This was also reflected in a high K/Mg ratio in petioles of vines grown on Mollisols

and Vertisols, suggesting Mg deficiency (data not shown) (Delas, 1996). Conversely, petiole

Mg levels were highest at veraison and harvest in vines grown on the Alfisols. Petiole Mn

3

levels were consistently low in vines grown in the calcic Mollisols, as explained by the

insolubility of Mn in calcareous soils with pH 7.5-8 (data not shown). Petiole N levels showed

no significant variation between sites.

Figure 1. Four contrasting soil types in a Paso Robles Cabernet Sauvignon vineyard. Soil

types and block numbers are shown. Numbered white circles indicate the locations of soil

sampling pits.

Figure 2. Petiole and Blade P levels are consistently low in plants grown on Calcic

Mollisols. Data shown are three-year averages (2007, 08, 09). Numbers indicate pit sampling

sites as seen in Fig. 1; sites 3 and 7 were located in Block 52 (Mollisol).

Plant vigor and fruit yields

Plant root counts varied considerably with soil type. In the Alfisols, the presence of a

compacted layer at depth prevented significant root penetration beyond 40-50 cm. In the

Mollisols, root density was greatest at depths below 50 cm, likely due to the high salt content

and electrical conductivity at the surface horizon. The Vertisols were characterized by good

44 - 29



water distribution throughout the profile, and the presence of macropores allowed root

penetration to depths below 100 cm.

Differences in fruit harvest weights were subtle between sites, with vines grown on the

Mollisols having lower yields than those grown on other soil types (Lambert et al., 2008).

Although this trend was consistent from year to year, it only reached statistical significance in

the 2008 season. Differences in cluster and berry weight between sites were also subtle, but a

similar trend was apparent, with weights generally lowest in vines grown on the Mollisols

(data not shown).

Figure 3. Petiole and Blade K levels cluster by soil type. This trend was most apparent at

veraison and harvest, when Petiole K levels were highest in vines grown on Mollisols or

Vertisols, and lowest in vines grown on Alfisols. Data shown are three-year averages (2007,

08, 09). Numbers indicate pit sampling sites as seen in Fig. 1.

Berry flavor components

Preliminary analysis of results from sensory analysis of berries revealed clustering of flavor

components with soil types, with vegetal notes and sourness attributed to wines prepared on

the Mollisols (data not shown). These analyses are still in progress along with chemical

analyses of small-lot wines prepared from each of the four vineyard plots.

CONCLUSIONS

Detailed characterization of soils on the four vineyard plots revealed four distinct soil types.

Blocks 56 and 57 contained two related Alfisols. The soils in block 57 had loamy/sandy loam

topsoil and clayey subsoil with an abrupt textural change. Block 56 contained shallower, less

developed Alfisols characterized as Haploxeralfs. Block 53 contained Vertisols, characterized

by greater than 30% clay content and a tendency to ‘shrink/swell’ behavior. Lastly, the soils in

Block 52 were characterized as Mollisols: fine-loamy, mixed, superactive, thermic Calcic

5

Bloom K

1.0 1.1 1.2 1.3 1.4 1.5

2.00

2.25

2.50

2.75

3.00

7

3

Blade K (%)

Veraison K

0.8 0.9 1.0 1.1 1.23.5

4.0

4.5

5.0

5.5

6.0

6.5

6

7

2

3

Alfisols

Blade K (%)

Harvest K

0.0 0.2 0.4 0.6 0.8 1.00

1

2

3

4

5

6

7

6

7

2

3

Alfisols

Blade K (%)

0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.20

1

2

3

4

5

6

7

SUMMARY

Blade K (%)

Haploxerolls. These soils had calcareous seams, laminar lime concretions and an angular

blocky structure in the subsoil.

Analysis of plant tissue nutrient levels revealed consistent trends over the three-year study

period. Vines grown on the Mollisols had consistently low levels of petiole and blade P at

bloom, veraison and harvest. Petiole and Blade P levels were closely correlated. Vines grown

on the Mollisols and Vertisols had higher levels of petiole K at veraison and harvest than

vines grown on Alfisols. This was also reflected in a high K/Mg ratio in petioles of vines

grown on Mollisols and Vertisols, suggesting Mg deficiency. Some nutrients, such as N,

showed no significant variation between soil types.

As reported previously (Lambert et al., 2008), the Alfisol in Block 57-5 and the Mollisol in

Block 52-3 gave contrasting results in terms of vine, fruit and juice characteristics. Vines

grown in the Alfisol had average to high diameters, and the highest fruit yield per vine in

terms of weight and cluster number. Juice from these vines also had the highest °Brix and

lowest total acidity during the first two years of the study. In contrast, vines grown in the

Mollisol had the lowest vine diameters in the study, the highest root density at depth, the

lowest fruit yield per vine, and the lowest cluster weights. Juice from these vines had the

lowest °Brix and among the highest total acidity values. Vines grown on the two other soils

showed intermediate characteristics.

Thus, in this study comparing Cabernet Sauvignon grapes of a single clone, on its own

roots, grown in four distinct soil types within a single vineyard, vines grown on contrasting

soil types had different growth characteristics that were reflected in differences in plant

nutrient levels and differences in fruit yield and juice properties. Additional chemical and

sensory analyses of grape juice and small lot wines are underway.

ACKNOWLEDGMENTS

The authors thank the J. Lohr Winery, Paso Robles, CA, for field assistance and continuous

support. We thank Anji Perry and Kim Adams, Viticulturists, J. Lohr Winery, for assistance

in the field. This project is supported by the American Vineyard Foundation (AVF).

BIBLIOGRAPHY

Andrés-de-Prado, R., M. Yste-Rojas, X. Sort, C. Andrés-Lacueva, M. Torres and R.M.

Lamuela-Raventos. 2007. Effect of soil type on wines produced from vitis vinifera L. Cv.

Grenache in commercial vineyards. J. Agric. Food Chem. 55:779-786.

Delas, J., 2000. Fertilisation de la vigne. First Edition. Bordeaux: Editions Féret.

Klein, I., Strime, M., Fanberstein, L., and Mani, Y. 2000. Irrigation and fertigation effects on

phosphorus and potassium nutrition of winegrapes. Vitis 39:55-62.

Lambert, J.J., McElrone, A., Battany, M., Dahlgren, R.A., and Wolpert, J.A. 2008. Influence

of soil type and changes in soil solution chemistry on vine growth parameters and grape and

wine quality in a central coast California Vineyard. In: Proceedings of the VIIth International

Terroir Congress, C. Van Leuwen, Ed., Nyon, Switzerland: International Organization of

Vineyards and Wines (OIV). Vol. 1:38-44.

Soil Survey Staff. 1993. Soil Survey Manual, Soil Conservation Service, U.S. Department of

Agriculture Handbook 18. Washington, DC: United States Department of Agriculture.

Tomasi, D., P. Belvini, G. Pascarella, P. Sivilotti, and C. Giulivo. 2006. L'effetto del suolo

sulla resa e sulla qualita dei vitigni Cabernet Sauvignon, Cabernet franc e Merlot. VigniVini

33:59-65.

64 - 30



water distribution throughout the profile, and the presence of macropores allowed root

penetration to depths below 100 cm.

Differences in fruit harvest weights were subtle between sites, with vines grown on the

Mollisols having lower yields than those grown on other soil types (Lambert et al., 2008).

Although this trend was consistent from year to year, it only reached statistical significance in

the 2008 season. Differences in cluster and berry weight between sites were also subtle, but a

similar trend was apparent, with weights generally lowest in vines grown on the Mollisols

(data not shown).

Figure 3. Petiole and Blade K levels cluster by soil type. This trend was most apparent at

veraison and harvest, when Petiole K levels were highest in vines grown on Mollisols or

Vertisols, and lowest in vines grown on Alfisols. Data shown are three-year averages (2007,

08, 09). Numbers indicate pit sampling sites as seen in Fig. 1.

Berry flavor components

Preliminary analysis of results from sensory analysis of berries revealed clustering of flavor

components with soil types, with vegetal notes and sourness attributed to wines prepared on

the Mollisols (data not shown). These analyses are still in progress along with chemical

analyses of small-lot wines prepared from each of the four vineyard plots.

CONCLUSIONS

Detailed characterization of soils on the four vineyard plots revealed four distinct soil types.

Blocks 56 and 57 contained two related Alfisols. The soils in block 57 had loamy/sandy loam

topsoil and clayey subsoil with an abrupt textural change. Block 56 contained shallower, less

developed Alfisols characterized as Haploxeralfs. Block 53 contained Vertisols, characterized

by greater than 30% clay content and a tendency to ‘shrink/swell’ behavior. Lastly, the soils in

Block 52 were characterized as Mollisols: fine-loamy, mixed, superactive, thermic Calcic

5

Bloom K

1.0 1.1 1.2 1.3 1.4 1.5

2.00

2.25

2.50

2.75

3.00

7

3

Blade K (%)

Veraison K

0.8 0.9 1.0 1.1 1.23.5

4.0

4.5

5.0

5.5

6.0

6.5

6

7

2

3

Alfisols

Blade K (%)

Harvest K

0.0 0.2 0.4 0.6 0.8 1.00

1

2

3

4

5

6

7

6

7

2

3

Alfisols

Blade K (%)

0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.20

1

2

3

4

5

6

7

SUMMARY

Blade K (%)

Haploxerolls. These soils had calcareous seams, laminar lime concretions and an angular

blocky structure in the subsoil.

Analysis of plant tissue nutrient levels revealed consistent trends over the three-year study

period. Vines grown on the Mollisols had consistently low levels of petiole and blade P at

bloom, veraison and harvest. Petiole and Blade P levels were closely correlated. Vines grown

on the Mollisols and Vertisols had higher levels of petiole K at veraison and harvest than

vines grown on Alfisols. This was also reflected in a high K/Mg ratio in petioles of vines

grown on Mollisols and Vertisols, suggesting Mg deficiency. Some nutrients, such as N,

showed no significant variation between soil types.

As reported previously (Lambert et al., 2008), the Alfisol in Block 57-5 and the Mollisol in

Block 52-3 gave contrasting results in terms of vine, fruit and juice characteristics. Vines

grown in the Alfisol had average to high diameters, and the highest fruit yield per vine in

terms of weight and cluster number. Juice from these vines also had the highest °Brix and

lowest total acidity during the first two years of the study. In contrast, vines grown in the

Mollisol had the lowest vine diameters in the study, the highest root density at depth, the

lowest fruit yield per vine, and the lowest cluster weights. Juice from these vines had the

lowest °Brix and among the highest total acidity values. Vines grown on the two other soils

showed intermediate characteristics.

Thus, in this study comparing Cabernet Sauvignon grapes of a single clone, on its own

roots, grown in four distinct soil types within a single vineyard, vines grown on contrasting

soil types had different growth characteristics that were reflected in differences in plant

nutrient levels and differences in fruit yield and juice properties. Additional chemical and

sensory analyses of grape juice and small lot wines are underway.

ACKNOWLEDGMENTS

The authors thank the J. Lohr Winery, Paso Robles, CA, for field assistance and continuous

support. We thank Anji Perry and Kim Adams, Viticulturists, J. Lohr Winery, for assistance

in the field. This project is supported by the American Vineyard Foundation (AVF).

BIBLIOGRAPHY

Andrés-de-Prado, R., M. Yste-Rojas, X. Sort, C. Andrés-Lacueva, M. Torres and R.M.

Lamuela-Raventos. 2007. Effect of soil type on wines produced from vitis vinifera L. Cv.

Grenache in commercial vineyards. J. Agric. Food Chem. 55:779-786.

Delas, J., 2000. Fertilisation de la vigne. First Edition. Bordeaux: Editions Féret.

Klein, I., Strime, M., Fanberstein, L., and Mani, Y. 2000. Irrigation and fertigation effects on

phosphorus and potassium nutrition of winegrapes. Vitis 39:55-62.

Lambert, J.J., McElrone, A., Battany, M., Dahlgren, R.A., and Wolpert, J.A. 2008. Influence

of soil type and changes in soil solution chemistry on vine growth parameters and grape and

wine quality in a central coast California Vineyard. In: Proceedings of the VIIth International

Terroir Congress, C. Van Leuwen, Ed., Nyon, Switzerland: International Organization of

Vineyards and Wines (OIV). Vol. 1:38-44.

Soil Survey Staff. 1993. Soil Survey Manual, Soil Conservation Service, U.S. Department of

Agriculture Handbook 18. Washington, DC: United States Department of Agriculture.

Tomasi, D., P. Belvini, G. Pascarella, P. Sivilotti, and C. Giulivo. 2006. L'effetto del suolo

sulla resa e sulla qualita dei vitigni Cabernet Sauvignon, Cabernet franc e Merlot. VigniVini

33:59-65.

64 - 31



PEDO-GEOLOGICAL ANALYSIS OF GAILLAC “PREMIÈRESCÔTES” TOPOSEQUENCES (TARN, SW FRANCE)

CONSEQUENCES ON MICRO-TERROIRS CHARACTERIZATION

Courjault-Radé Pierre (1), Papa Marine (1), Oliva Priscia (1), Munoz Marguerite (1) andCazottes Alain (2)

(1) Université de Toulouse – LMTG / UMR 5563 – 14, Avenue E. Belin 31400 Toulouse (France). Phone :Email : [email protected]

(2) Domaine des Terrisses, 81600 Gaillac (France)[email protected]

ABSTRACTDrill-holes performed on a Gaillac appellation vineyard together with mineralogical andchemical analyses on soils samples have given evidence of a micro-scale pedo-geologicalvariability leading to characterization of distinct micro-terroirs. This variability results fromthe strong interdependence of water regime patterns and chemical element availability fromwhich pedo-geological wine typicity would emerge.

KEY-WORDSGeochemistry - Microterroir – Mineralogy – Pedogenesis – Solifluction – Water regime

INTRODUCTIONThe concept of terroir integrates all factors that work together to define region with specificcharacteristics to match the needs of wine grapes that will produce high quality wine. Thesefactors start with the rocks and resulting soils through complex pedo-geological processes,continue with climate and vineyard management practices and end with the winemaker’s art.Herein, we consider the key pedo-geological factors of terroir: the parent rocks and the soilsand their subsequent physical-chemical evolutions.Our purpose is to illustrate how parent rock characteristics guides the definition of micro-terroirs on the basis of preliminary pedo-geological results acquired from vineyard plotslocated in the Gaillac appellation in south-western France (Figure 1).Till now, 3 planting areas have been empirically defined by the winegrower’s in the studiedvineyard. From top to base of slope (see location on Figure 2), grapevine varieties wereplanted following especially local-scale climatic parameters as follows:- sweet white wines area (Mauzac and Loin de l’oeil grapevine varieties) situated at the top ofthe valley slope characterized by the warmest microclimatic conditions,- red wines area (Duras, Fer Servadou and Syrah grapevine varieties) in mid-part of the slope,- dry white wines area (Loin de l’œil and Cabernet grapevine varieties) situated at the base ofthe valley slope characterized by the coolest microclimatic conditions.Rootstocks were used according to the soil surface carbonate contents (3309C or 41B).Several pedological drill-holes have been carried out on two geologically distincttoposequences. Mineralogical and chemical analyses of representative soil samples have beenperformed. Finally, the association of pedo-geological data with topography andmicroclimatic zoning lead to characterize new micro-terroirs.

GEOLOGICAL AND MORPHOLOGICAL DATALocalisation of studied areaGaillac appellation area is located in SW France, north of Toulouse city (Figure 1a). Thestudied vineyard is situated on the “Right bank coteaux” area which is one of the four mainterroirs of the appellation area (Figure 1b). Precisely, the Terrisses vineyard is located on thefirst hills of the right Tarn river bank, informally named “Première Côtes” (Figure 1c).Geologically, it is composed of an Oligocene molassic sandy-clayey substratum overlain bydetrital material originated from late-Würmian solifluction phase and subsequent Holocenecolluviation, principally developed at the base of hillside slopes and in valleys (Figure 2).

Figure 1 / (a): localisation of the Gaillac appellation area (b): the main terroirs of the appellation area andlocalisation of the studied vineyard - (symbols: 1, “Calcareous Plateau Cordais”; 2,“Right bank molassic

coteaux”; 3,“Left bank alluvial terraces”; 4,Tarn alluvial plain); (c): mesoscale geological setting – (symbols:1,present-day alluvium;2, Holocene alluvium; 3, soliflued and colluvial deposits; 4, Oligocene molassic

basement).

Description of study toposequencesTwo NE-SW oriented toposequences have been investigated (Figure 2). The T1 toposequenceis located on a small valley covered by displaced soliflued materials and the T2 toposequenceis situated on a hill composed of molassic substratum material (Figures 2a and 2b). Both havesouth-facing weak to moderate-angle slopes (< 15°) and represent about 250 m in length forT1 and 195 m for T2 (Figure 2b).

Figure 2/ a: Geological setting with localisation of studied toposequences (symbols: 1, soliflued/colluvialdeposits; 2, Oligocene molassic basement; 3, small rivers); b: local morphologic setting. T1 and T2

toposequences are plotted with samples location (symbols: 1, soliflued/colluvial deposits; 2; Oligocene molassicbasement)

4 - 32



PEDO-GEOLOGICAL ANALYSIS OF GAILLAC “PREMIÈRESCÔTES” TOPOSEQUENCES (TARN, SW FRANCE)

CONSEQUENCES ON MICRO-TERROIRS CHARACTERIZATION

Courjault-Radé Pierre (1), Papa Marine (1), Oliva Priscia (1), Munoz Marguerite (1) andCazottes Alain (2)

(1) Université de Toulouse – LMTG / UMR 5563 – 14, Avenue E. Belin 31400 Toulouse (France). Phone :Email : [email protected]

(2) Domaine des Terrisses, 81600 Gaillac (France)[email protected]

ABSTRACTDrill-holes performed on a Gaillac appellation vineyard together with mineralogical andchemical analyses on soils samples have given evidence of a micro-scale pedo-geologicalvariability leading to characterization of distinct micro-terroirs. This variability results fromthe strong interdependence of water regime patterns and chemical element availability fromwhich pedo-geological wine typicity would emerge.

KEY-WORDSGeochemistry - Microterroir – Mineralogy – Pedogenesis – Solifluction – Water regime

INTRODUCTIONThe concept of terroir integrates all factors that work together to define region with specificcharacteristics to match the needs of wine grapes that will produce high quality wine. Thesefactors start with the rocks and resulting soils through complex pedo-geological processes,continue with climate and vineyard management practices and end with the winemaker’s art.Herein, we consider the key pedo-geological factors of terroir: the parent rocks and the soilsand their subsequent physical-chemical evolutions.Our purpose is to illustrate how parent rock characteristics guides the definition of micro-terroirs on the basis of preliminary pedo-geological results acquired from vineyard plotslocated in the Gaillac appellation in south-western France (Figure 1).Till now, 3 planting areas have been empirically defined by the winegrower’s in the studiedvineyard. From top to base of slope (see location on Figure 2), grapevine varieties wereplanted following especially local-scale climatic parameters as follows:- sweet white wines area (Mauzac and Loin de l’oeil grapevine varieties) situated at the top ofthe valley slope characterized by the warmest microclimatic conditions,- red wines area (Duras, Fer Servadou and Syrah grapevine varieties) in mid-part of the slope,- dry white wines area (Loin de l’œil and Cabernet grapevine varieties) situated at the base ofthe valley slope characterized by the coolest microclimatic conditions.Rootstocks were used according to the soil surface carbonate contents (3309C or 41B).Several pedological drill-holes have been carried out on two geologically distincttoposequences. Mineralogical and chemical analyses of representative soil samples have beenperformed. Finally, the association of pedo-geological data with topography andmicroclimatic zoning lead to characterize new micro-terroirs.

GEOLOGICAL AND MORPHOLOGICAL DATALocalisation of studied areaGaillac appellation area is located in SW France, north of Toulouse city (Figure 1a). Thestudied vineyard is situated on the “Right bank coteaux” area which is one of the four mainterroirs of the appellation area (Figure 1b). Precisely, the Terrisses vineyard is located on thefirst hills of the right Tarn river bank, informally named “Première Côtes” (Figure 1c).Geologically, it is composed of an Oligocene molassic sandy-clayey substratum overlain bydetrital material originated from late-Würmian solifluction phase and subsequent Holocenecolluviation, principally developed at the base of hillside slopes and in valleys (Figure 2).

Figure 1 / (a): localisation of the Gaillac appellation area (b): the main terroirs of the appellation area andlocalisation of the studied vineyard - (symbols: 1, “Calcareous Plateau Cordais”; 2,“Right bank molassic

coteaux”; 3,“Left bank alluvial terraces”; 4,Tarn alluvial plain); (c): mesoscale geological setting – (symbols:1,present-day alluvium;2, Holocene alluvium; 3, soliflued and colluvial deposits; 4, Oligocene molassic

basement).

Description of study toposequencesTwo NE-SW oriented toposequences have been investigated (Figure 2). The T1 toposequenceis located on a small valley covered by displaced soliflued materials and the T2 toposequenceis situated on a hill composed of molassic substratum material (Figures 2a and 2b). Both havesouth-facing weak to moderate-angle slopes (< 15°) and represent about 250 m in length forT1 and 195 m for T2 (Figure 2b).

Figure 2/ a: Geological setting with localisation of studied toposequences (symbols: 1, soliflued/colluvialdeposits; 2, Oligocene molassic basement; 3, small rivers); b: local morphologic setting. T1 and T2

toposequences are plotted with samples location (symbols: 1, soliflued/colluvial deposits; 2; Oligocene molassicbasement)

4 - 33



The molassic substratum of the studied toposequences is composed of clays, silts, and sandstogether with scattered gravels and pebbles occurrences. Patchily distributed sandycarbonated lenses and/or layers may also occur.In the case of valleys (toposequence 1), the molassic basement is overlaid by a thin sequence(up to 2-3m thick) of detrital material inherited from the main late-Würmian (ca. 13,000 yearsBP) solifluction phase processes. The later has occurred when soil and bedrock were affectedby alternate freezing and melting in peri-glacial climate. These mass movements havegenerated allochtonous sediment lobes on slopes, composed of mixed bedrock and alteritematerials concealing the underlying in-situ basement outcrops. Subsequently, this material hassuffered from pedogenetic alteration and was transported on short distance, producing clayey-silty colluvial deposits at the base of the valleys.

PEDOLOGICAL AND PEDOGENETIC DATAIdentification of soilsSoils of the 2 toposequences have been identified according to the French soil classification(Figure 3 below).

Figure 3 / Synthesis of the different soils sequences along T1 and T2 toposequences (not at scale) - Soil horizonsare noted using French classification nomenclature (symbols: 1, bed-rock fragments; 2, sand; 3, sand with clay;4, clay with sand; 5, silt; 6 clay and silt; 7, clay; (a), manganese nodules; (b), occurrence of CaCO3; (c), oxydo-

reduction spots; (d), carbonate nodules; (e), estimated soil/altered basement boundary) .

- Toposequence 1 is composed of calcisols at the top (n° 1.1) and clayey colluviosols at thebase (n° 1.4.). Complex soils, inherited from solifluction mass movements, are situated inmedium part of the slope (n°s 1.2 and 1.3). They are composed of allochthonous luvisols

overlying the autochthonous altered molassic basement. The limit between the soil sequenceand the altered bedrock (= C horizons) varies from less than 1m at the top of thetoposequence, about 2.50m in medium part and more than 4-5m at the base.- Toposequence 2 is composed of luvisols at the top (n°s 2.1 and 2.2.), calcosols in mediumpart (n°s 2.3., 2.4 and 2.5) and clayey colluviosols at the base (n° 2.6). The roof of the alteredmolassic bedrock (C horizons) varies from 0.60m at the top 1.50m in medium part to morethan 4m at the base of the toposequence. It is worth noting that no indication of solifluedmaterial has been evidenced.Mineralogical and geochemical resultsMineralogical investigations (XRD) have pointed out the occurrence of quartz, kaolinite,feldspaths, illite, goethite, smectite and minor calcite amount in all the samples. Thisspectrum is in agreement with the composition of the molassic basement from which the soilsoriginate as the globally similar REES flat patterns for the soil samples confirm (Figure 4).

Figure 4: Chondrite-normalized REE patterns of deep-soil horizons (= C horizons).

Measurements of pH and carbonate-content were performed on grinded soil samples and ICP-MS chemical analyses were performed to calculate chemical elements contents (Tab. 1).

Tab. 1 - Principal chemical and mineralogical results of T1 and T2 toposequencesSymbols /: no analysed; -: no detected; tr: traces; +: minor; ++: major; +++: dominant

SOILSAMPLES

Altitu-de(m)

Hori-zons

z(cm)

pH CaCO3%

Cag/kg

Hydro-Morphis

m

Smectite

T 1.1Calcisol 225

ASC

103570

7.497.827.07

1.20.1-

5.05.42.4

--

Tr

++++++

T 1.2Soliflued

soil207

ABt

ScaC

1060

120150

7.967.788.168.44

1.70.26.127.8

10.35.3

92.622.9

----

+++++

+++T 1.3

Solifluedsoil

202ABtC

50180230

7.468.098.32

0.90.812.5

2.76.7

39.6

-+-

++++++

T 1.4Colluviosol 190 A 120 7.87 0.2 3.9 ++ ++

4 - 34



The molassic substratum of the studied toposequences is composed of clays, silts, and sandstogether with scattered gravels and pebbles occurrences. Patchily distributed sandycarbonated lenses and/or layers may also occur.In the case of valleys (toposequence 1), the molassic basement is overlaid by a thin sequence(up to 2-3m thick) of detrital material inherited from the main late-Würmian (ca. 13,000 yearsBP) solifluction phase processes. The later has occurred when soil and bedrock were affectedby alternate freezing and melting in peri-glacial climate. These mass movements havegenerated allochtonous sediment lobes on slopes, composed of mixed bedrock and alteritematerials concealing the underlying in-situ basement outcrops. Subsequently, this material hassuffered from pedogenetic alteration and was transported on short distance, producing clayey-silty colluvial deposits at the base of the valleys.

PEDOLOGICAL AND PEDOGENETIC DATAIdentification of soilsSoils of the 2 toposequences have been identified according to the French soil classification(Figure 3 below).

Figure 3 / Synthesis of the different soils sequences along T1 and T2 toposequences (not at scale) - Soil horizonsare noted using French classification nomenclature (symbols: 1, bed-rock fragments; 2, sand; 3, sand with clay;4, clay with sand; 5, silt; 6 clay and silt; 7, clay; (a), manganese nodules; (b), occurrence of CaCO3; (c), oxydo-

reduction spots; (d), carbonate nodules; (e), estimated soil/altered basement boundary) .

- Toposequence 1 is composed of calcisols at the top (n° 1.1) and clayey colluviosols at thebase (n° 1.4.). Complex soils, inherited from solifluction mass movements, are situated inmedium part of the slope (n°s 1.2 and 1.3). They are composed of allochthonous luvisols

overlying the autochthonous altered molassic basement. The limit between the soil sequenceand the altered bedrock (= C horizons) varies from less than 1m at the top of thetoposequence, about 2.50m in medium part and more than 4-5m at the base.- Toposequence 2 is composed of luvisols at the top (n°s 2.1 and 2.2.), calcosols in mediumpart (n°s 2.3., 2.4 and 2.5) and clayey colluviosols at the base (n° 2.6). The roof of the alteredmolassic bedrock (C horizons) varies from 0.60m at the top 1.50m in medium part to morethan 4m at the base of the toposequence. It is worth noting that no indication of solifluedmaterial has been evidenced.Mineralogical and geochemical resultsMineralogical investigations (XRD) have pointed out the occurrence of quartz, kaolinite,feldspaths, illite, goethite, smectite and minor calcite amount in all the samples. Thisspectrum is in agreement with the composition of the molassic basement from which the soilsoriginate as the globally similar REES flat patterns for the soil samples confirm (Figure 4).

Figure 4: Chondrite-normalized REE patterns of deep-soil horizons (= C horizons).

Measurements of pH and carbonate-content were performed on grinded soil samples and ICP-MS chemical analyses were performed to calculate chemical elements contents (Tab. 1).

Tab. 1 - Principal chemical and mineralogical results of T1 and T2 toposequencesSymbols /: no analysed; -: no detected; tr: traces; +: minor; ++: major; +++: dominant

SOILSAMPLES

Altitu-de(m)

Hori-zons

z(cm)

pH CaCO3%

Cag/kg

Hydro-Morphis

m

Smectite

T 1.1Calcisol 225

ASC

103570

7.497.827.07

1.20.1-

5.05.42.4

--

Tr

++++++

T 1.2Soliflued

soil207

ABt

ScaC

1060

120150

7.967.788.168.44

1.70.26.127.8

10.35.3

92.622.9

----

+++++

+++T 1.3

Solifluedsoil

202ABtC

50180230

7.468.098.32

0.90.812.5

2.76.7

39.6

-+-

++++++

T 1.4Colluviosol 190 A 120 7.87 0.2 3.9 ++ ++

4 - 35



T 2.1Luvisol 232

ABtC

204060

7.006.595.25

---

///

-++

++++

T 2.2Luvisol

230AEBtC

204080

100

7.467.427.337.90

0.2--

1.0

3.51.93.53.0

--+-

++

++++

T 2.3Calcosol 227

AScaC

205060

7.958.188.30

5.218.622.0

18.955.367.0

-+++

++++++

T 2.4Calcosol

219A

Sca 1Sca 2Sca 3

C

550

100120150

8.088.168.128.068.03

16.215.417.916.822.5

/////

-trtrtr-

+++++++

++

T 2.5Calcosol

206 AScaC

2070

130

7.848.258.38

11.129.436.5

34.766.7110

-++-

++++++

T 2.6Colluviosol

196A1A2AcaAg

1040

100140

7.917.978.178.23

1.60.19.16.2

7.22.8

27.519.5

--

++++

++++

The above mineralogical/geochemical soil patterns result from a complex sequence ofpedogenetic processes essentially related to different water regimes as follows:- pH values vary from neutral to basic in agreement with the occurrence of calcite in the bulkof the soil/bedrock sequences. The neutral values correspond to the most decarbonatedhorizons (calcisols and luvisols). In contrast, the more basic values correspond to calcosolsand altered molassic carbonated bedrock horizons (= C horizons),- hydromorphism associated with conditions of reduction/oxidation of Fe and Mn is of weakamplitude and occurs at depth (≥ ~ 1m) in the case of the toposequence 1; it is of strongeramplitude and occurs near the surface in the case of the toposequence 2; in both cases, it ismoderate and only follows temporary state of water saturation (no occurrence of redoxichorizons),- clay eluviation is the main process affecting the soils situated at the nearly flat area at the topof toposequence 2; it is also observed in the allochtonous soliflued luvisols situated in themid-part of the toposequence 1,- decarbonatation and re-precipitation is a widespread process under the regional temperateclimatic conditions affecting almost all soil samples,- smectite occurs in relatively large quantities in all the soil samples. It regulates waterresource by swelling (vs. shrinking) processes.

CHARACTERIZATION OF MICROTERROIRSThe above soil patterns associated with geological and topographic data together with thesuperimposition of micro-scale climatic zoning, lead to characterize different microterroirs(Figure 6). In both toposequences, the bulk of the soil/bedrock sequence is composed ofmolassic detrital material either in situ originating from alteration of the molassic bedrock or“ex-situ”, displaced by mass transport (solifluction) and/or by erosion (colluvium). This broaddivision may be regarded as a first order pedo-geological terroir classification.

Figure 5/ Characterization of the microterroirs (A to E) integrating topographic, pedo-geological and microscaleclimatic zoning data (not at scale). Symbols: 1, molassic basement; 2, altered molassic basement (a: clay

dominant; b, sand dominant); 3, solifluction lobes.

Micro-terroirs on allochthonous colluvial deposits- The “Colluvial micro-terroir” (A): a weak sloping area composed of thick colluvialrecarbonated (or not) clayey deposits marked by temporary hydromorphism at the surface ordeeper (>1m) and characterized by the coolest and dampest microclimate conditions.Micro-terroirs on allochtonous soliflued material- The “Soliflued micro-terroir” (B): a moderately sloping area with soliflued luvisols forminglobes onto the altered carbonated molassic basement; hydromorphism is negligible.Micro-terroirs on in situ molassic basement- The “Altered carbonated micro-terroir” (C): a moderately sloping area with superficialhydromorphism which may be of moderate amplitude.- The “Leached micro-terroir” (D): a nearly flat area composed of leached soil sequence(luvisols) marked by moderate superficial hydromorphism.- The “Decarbonated micro-terroir” (E): a gently sloping area constituted of decarbonatedmolassic basement without traces of any hydromorphism and characterized by the warmestand driest microclimate conditions.

CONCLUSIONSThe above preliminary results show a great pedo-geological variability despite the relativehomogeneity of the original molassic material. Actually, the studied plots of the Terrissesvineyard appears as a mosaic of, at least, 5 micro-terroirs instead of the 3 empirically definedusing micro-climate variances. This microscale variability appears to results mainly from thestrong interdependence of water regime patterns and chemical element availability. Therefore,the pedo-geological wine typicity should be regarded as an emergent result of this highlycomplex interaction. The next step of the study will be to follow the fate of distinctive soilchemical elements from the vine crop to the wine glass for each defined microterroirs intaking into account characteristics of each grapevine varieties and rootstocks.

4 - 36



T 2.1Luvisol 232

ABtC

204060

7.006.595.25

---

///

-++

++++

T 2.2Luvisol

230AEBtC

204080

100

7.467.427.337.90

0.2--

1.0

3.51.93.53.0

--+-

++

++++

T 2.3Calcosol 227

AScaC

205060

7.958.188.30

5.218.622.0

18.955.367.0

-+++

++++++

T 2.4Calcosol

219A

Sca 1Sca 2Sca 3

C

550

100120150

8.088.168.128.068.03

16.215.417.916.822.5

/////

-trtrtr-

+++++++

++

T 2.5Calcosol

206 AScaC

2070

130

7.848.258.38

11.129.436.5

34.766.7110

-++-

++++++

T 2.6Colluviosol

196A1A2AcaAg

1040

100140

7.917.978.178.23

1.60.19.16.2

7.22.8

27.519.5

--

++++

++++

The above mineralogical/geochemical soil patterns result from a complex sequence ofpedogenetic processes essentially related to different water regimes as follows:- pH values vary from neutral to basic in agreement with the occurrence of calcite in the bulkof the soil/bedrock sequences. The neutral values correspond to the most decarbonatedhorizons (calcisols and luvisols). In contrast, the more basic values correspond to calcosolsand altered molassic carbonated bedrock horizons (= C horizons),- hydromorphism associated with conditions of reduction/oxidation of Fe and Mn is of weakamplitude and occurs at depth (≥ ~ 1m) in the case of the toposequence 1; it is of strongeramplitude and occurs near the surface in the case of the toposequence 2; in both cases, it ismoderate and only follows temporary state of water saturation (no occurrence of redoxichorizons),- clay eluviation is the main process affecting the soils situated at the nearly flat area at the topof toposequence 2; it is also observed in the allochtonous soliflued luvisols situated in themid-part of the toposequence 1,- decarbonatation and re-precipitation is a widespread process under the regional temperateclimatic conditions affecting almost all soil samples,- smectite occurs in relatively large quantities in all the soil samples. It regulates waterresource by swelling (vs. shrinking) processes.

CHARACTERIZATION OF MICROTERROIRSThe above soil patterns associated with geological and topographic data together with thesuperimposition of micro-scale climatic zoning, lead to characterize different microterroirs(Figure 6). In both toposequences, the bulk of the soil/bedrock sequence is composed ofmolassic detrital material either in situ originating from alteration of the molassic bedrock or“ex-situ”, displaced by mass transport (solifluction) and/or by erosion (colluvium). This broaddivision may be regarded as a first order pedo-geological terroir classification.

Figure 5/ Characterization of the microterroirs (A to E) integrating topographic, pedo-geological and microscaleclimatic zoning data (not at scale). Symbols: 1, molassic basement; 2, altered molassic basement (a: clay

dominant; b, sand dominant); 3, solifluction lobes.

Micro-terroirs on allochthonous colluvial deposits- The “Colluvial micro-terroir” (A): a weak sloping area composed of thick colluvialrecarbonated (or not) clayey deposits marked by temporary hydromorphism at the surface ordeeper (>1m) and characterized by the coolest and dampest microclimate conditions.Micro-terroirs on allochtonous soliflued material- The “Soliflued micro-terroir” (B): a moderately sloping area with soliflued luvisols forminglobes onto the altered carbonated molassic basement; hydromorphism is negligible.Micro-terroirs on in situ molassic basement- The “Altered carbonated micro-terroir” (C): a moderately sloping area with superficialhydromorphism which may be of moderate amplitude.- The “Leached micro-terroir” (D): a nearly flat area composed of leached soil sequence(luvisols) marked by moderate superficial hydromorphism.- The “Decarbonated micro-terroir” (E): a gently sloping area constituted of decarbonatedmolassic basement without traces of any hydromorphism and characterized by the warmestand driest microclimate conditions.

CONCLUSIONSThe above preliminary results show a great pedo-geological variability despite the relativehomogeneity of the original molassic material. Actually, the studied plots of the Terrissesvineyard appears as a mosaic of, at least, 5 micro-terroirs instead of the 3 empirically definedusing micro-climate variances. This microscale variability appears to results mainly from thestrong interdependence of water regime patterns and chemical element availability. Therefore,the pedo-geological wine typicity should be regarded as an emergent result of this highlycomplex interaction. The next step of the study will be to follow the fate of distinctive soilchemical elements from the vine crop to the wine glass for each defined microterroirs intaking into account characteristics of each grapevine varieties and rootstocks.

4 - 37



THREE PROXIMAL SENSORS TO ESTIMATE TEXTURE, SKELETON AND SOIL WATER STORAGE IN VINEYARDS

S. Priori (1), E.A.C. Costantini (1), A. Agnelli (1), S. Pellegrini (1), E. Martini (2)

(1) C.R.A.-A.B.P., Research Center for Agrobiology and Pedology, Piazza M.D’Azeglio, 30, 50121, Firenze, Italy. (2) University of Turin, Earth Science Department, Turin, Italy.

Corresponding author: [email protected]

ABSTRACT Proximal sensors are becoming widely used in precision viticulture, due to the quick, easy and

non-invasive identification of soil spatial variability. The apparent soil electrical conductivity (ECa) is the main parameter measured by sensors, which is correlated to many factors, like soil water content, salinity, clay content and mineralogy, rock fragments, bulk density, and porosity. This study compares three different sensors to delineate soil boundaries and estimate clay, skeleton content and available water (AWC) in a vineyard of the Chianti region (Central Italy). All three sensors produced ECa maps with similar pattern. Although the correlations between ECa, clay and skeleton content were usually moderate, the correlations between ECa and some important hydrological parameters, namely field capacity (FC), wilting point (WP) and available water capacity (AWC), was very high.

KEYWORD Soil – precision viticulture – geophysics – EMI sensors – apparent electrical conductivity. INTRODUCTION The use of instruments like GPS, GIS, remote sensing and soil monitoring technologies in

precision viticulture is becoming common for the most important farms (Proffitt et al., 2006). In particular, the knowledge of the spatial variability of soil hydrological parameters is crucial for a proper crop management, aimed at maximizing income and reducing environmental impacts of agriculture activities. In precision viticulture, it is very important to know the hydropedological variability (Morari et al., 2009; Costantini et al., 2009) of the vineyard to plan drainage, irrigation, tillage, fertilization etc., as well as to improve the quality of grapes and wine.

A rapid, non-invasive and relatively cheap mapping of the soil apparent electric conductivity (ECa) can be a very useful tool for identifying important soil map units and properties, in particular, clay (Morari et al., 2009), water content (Davies R., 2004; Tromp-van Meerveld and McDonnell, 2009; Costantini et al., 2009), bulk density and salinity (Doolittle et al., 2001). The relationships between apparent electric conductivity (ECa) and soil hydrological parameters are still under investigation (Cousin et al., 2009; Doussan and Ruy, 2009).

The goal of this work was to test the suitability of three different proximal sensors in a vineyard on skeletal soils, and to relate the measured ECa with the clay content, skeleton and hydrological parameters, namely field capacity (FC), wilting point (WP) and available water (AWC).

MATERIALS AND METHODS The studied vineyard, located in the Chianti area (Central Italy), was only 4 ha in size, but

heterogeneous in soils. All soils were difficult to be surveyed with the traditional hand auger, since they were rather clayey (clay content of the fine earth ranging from 28 to 56%) and stony (from 10 to 50 %). All soils were not saline.

The sensors used for this work (Fig.1) were: a) a single-frequency Electro-Magnetic Induction sensor (Geonics EM38-DD), b) a multi-frequency EMI sensor (GSSI Profiler EMP- 400) and c) a geoelectric system (ARP-Automatic Resistivity Profiling). The EM38-DD is an EMI sensor composed by two EM38 sensors, coupled in perpendicular position (Fig.1a). Each sensor has an intercoil spacing of 1 m and operates at a frequency of 14,600 Hz. The depths of the magnetic field penetration are about 0.75 m and 1.5 m, respectively for the horizontal (HDP) and vertical (VDP) dipoles modes (Geonics Limited, 1998). The instruments sensitivity varies as a non-linear function of depth (McNeil, 1990).

The GSSI Profiler EMP-400 (Fig.1b) is a multifrequency EMI sensor, which can operate to measure simultaneously up to 3 frequencies between 1,000 Hz and 16,000 Hz, with intercoil spacing of 1.2 m. For this study we operated at 8, 10 and 15 kHz. The instrument can be used in vertical dipole mode (VDP) or in horizontal dipole mode (HDP), but the instruments sensitivity in function of depth is not still studied. The output of both the EM38-DD and Profiler is the apparent electric conductivity (ECa), measured in mS m-1. Both the EMI sensors were supplied with a DGPS.

The ARP© device (Fig. 1c) was conceived by Geocarta, spin-off society of C.N.R.S. (National Scientific Research Center, France). The system, similar to a disc plough, consists of a couple of teeth discs operating as injection electrodes and of three couples of teeth discs, functioning as receivers and measuring the difference of electrical potential. The distance between each couple of receivers was conceived and calibrated to investigate three soil depths, about 0-50, 0-100 and 0-170 cm. The system, supplied with a DGPS, was pulled by a quad-bike.

Figure 1: The three proximal sensors used for this work. a) Geonics EM38-DD, b) GSSI Profiler EMP-400, c) Geocarta ARP©.

Resistivity values (ER), in Ω.m, were obtained from the intensity of the injected current and

from the differences in electrical potential. These values can be easily transformed in ECa (mS m-1) by the formula:

4 - 38



THREE PROXIMAL SENSORS TO ESTIMATE TEXTURE, SKELETON AND SOIL WATER STORAGE IN VINEYARDS

S. Priori (1), E.A.C. Costantini (1), A. Agnelli (1), S. Pellegrini (1), E. Martini (2)

(1) C.R.A.-A.B.P., Research Center for Agrobiology and Pedology, Piazza M.D’Azeglio, 30, 50121, Firenze, Italy. (2) University of Turin, Earth Science Department, Turin, Italy.

Corresponding author: [email protected]

ABSTRACT Proximal sensors are becoming widely used in precision viticulture, due to the quick, easy and

non-invasive identification of soil spatial variability. The apparent soil electrical conductivity (ECa) is the main parameter measured by sensors, which is correlated to many factors, like soil water content, salinity, clay content and mineralogy, rock fragments, bulk density, and porosity. This study compares three different sensors to delineate soil boundaries and estimate clay, skeleton content and available water (AWC) in a vineyard of the Chianti region (Central Italy). All three sensors produced ECa maps with similar pattern. Although the correlations between ECa, clay and skeleton content were usually moderate, the correlations between ECa and some important hydrological parameters, namely field capacity (FC), wilting point (WP) and available water capacity (AWC), was very high.

KEYWORD Soil – precision viticulture – geophysics – EMI sensors – apparent electrical conductivity. INTRODUCTION The use of instruments like GPS, GIS, remote sensing and soil monitoring technologies in

precision viticulture is becoming common for the most important farms (Proffitt et al., 2006). In particular, the knowledge of the spatial variability of soil hydrological parameters is crucial for a proper crop management, aimed at maximizing income and reducing environmental impacts of agriculture activities. In precision viticulture, it is very important to know the hydropedological variability (Morari et al., 2009; Costantini et al., 2009) of the vineyard to plan drainage, irrigation, tillage, fertilization etc., as well as to improve the quality of grapes and wine.

A rapid, non-invasive and relatively cheap mapping of the soil apparent electric conductivity (ECa) can be a very useful tool for identifying important soil map units and properties, in particular, clay (Morari et al., 2009), water content (Davies R., 2004; Tromp-van Meerveld and McDonnell, 2009; Costantini et al., 2009), bulk density and salinity (Doolittle et al., 2001). The relationships between apparent electric conductivity (ECa) and soil hydrological parameters are still under investigation (Cousin et al., 2009; Doussan and Ruy, 2009).

The goal of this work was to test the suitability of three different proximal sensors in a vineyard on skeletal soils, and to relate the measured ECa with the clay content, skeleton and hydrological parameters, namely field capacity (FC), wilting point (WP) and available water (AWC).

MATERIALS AND METHODS The studied vineyard, located in the Chianti area (Central Italy), was only 4 ha in size, but

heterogeneous in soils. All soils were difficult to be surveyed with the traditional hand auger, since they were rather clayey (clay content of the fine earth ranging from 28 to 56%) and stony (from 10 to 50 %). All soils were not saline.

The sensors used for this work (Fig.1) were: a) a single-frequency Electro-Magnetic Induction sensor (Geonics EM38-DD), b) a multi-frequency EMI sensor (GSSI Profiler EMP- 400) and c) a geoelectric system (ARP-Automatic Resistivity Profiling). The EM38-DD is an EMI sensor composed by two EM38 sensors, coupled in perpendicular position (Fig.1a). Each sensor has an intercoil spacing of 1 m and operates at a frequency of 14,600 Hz. The depths of the magnetic field penetration are about 0.75 m and 1.5 m, respectively for the horizontal (HDP) and vertical (VDP) dipoles modes (Geonics Limited, 1998). The instruments sensitivity varies as a non-linear function of depth (McNeil, 1990).

The GSSI Profiler EMP-400 (Fig.1b) is a multifrequency EMI sensor, which can operate to measure simultaneously up to 3 frequencies between 1,000 Hz and 16,000 Hz, with intercoil spacing of 1.2 m. For this study we operated at 8, 10 and 15 kHz. The instrument can be used in vertical dipole mode (VDP) or in horizontal dipole mode (HDP), but the instruments sensitivity in function of depth is not still studied. The output of both the EM38-DD and Profiler is the apparent electric conductivity (ECa), measured in mS m-1. Both the EMI sensors were supplied with a DGPS.

The ARP© device (Fig. 1c) was conceived by Geocarta, spin-off society of C.N.R.S. (National Scientific Research Center, France). The system, similar to a disc plough, consists of a couple of teeth discs operating as injection electrodes and of three couples of teeth discs, functioning as receivers and measuring the difference of electrical potential. The distance between each couple of receivers was conceived and calibrated to investigate three soil depths, about 0-50, 0-100 and 0-170 cm. The system, supplied with a DGPS, was pulled by a quad-bike.

Figure 1: The three proximal sensors used for this work. a) Geonics EM38-DD, b) GSSI Profiler EMP-400, c) Geocarta ARP©.

Resistivity values (ER), in Ω.m, were obtained from the intensity of the injected current and

from the differences in electrical potential. These values can be easily transformed in ECa (mS m-1) by the formula:

4 - 39



000,11

ER

ECa

The survey with the EM38-DD and the Profiler EMP-400 was performed on the same day in

August, when soils were dry on surface, whereas the survey with the ARP was carried out in May, when soils were moister and the contact soils-electrodes better. For this work, we did not consider the temperature and the moisture content of the soils, but the textural features only. 13 points were chosen for soil sampling and texture analysis on the basis of the ECa values. The samples were “tout venant” of some kilograms because of the measurement of skeleton content. Laboratory determination of the water content at FC and WP (v/v) was carried out by pressure plate apparatus at -33 and -1,500 kPa matric potential, respectively (Kassel and Nielsen, 1986). Each soil horizons was analyzed in triplicate, and the corresponding bulk density values were used to convert -33 and -1,500 kPa gravimetric water content to a volumetric basis. AWC was determined as the difference between water content at FC and WP.

RESULTS AND DISCUSSION The three instruments produced similar spatial patterns (Fig.2). During the proximal survey, the

EMI sensors (EM38-DD and Profiler EMP-400) in the HDP orientation registered negative or very low values in some vineyard areas. This was probably due to the interference of the iron wires of the vineyard rows or other iron materials with the magnetic field. Therefore, these wrong data (negative, or very close to 0) measured in HDP orientation were deleted before data interpolation.

As a whole, the ECa values measured by ARP device were higher, whereas the values measured by Profiler and EM-38 were similar (Tab.1).

Table 1: Descriptive statistics of ECa (mS m-1) measured with the three devices. The negative and very low values were not considered.

Profiler VDP Profiler HDP EM38 ARP

15kHz 10kHz 8kHz 15kHz 10kHz 8kHz VDP HDP 50 100 170

Mean 31.8 32.2 34.6 19.2 20.0 21.9 19.5 28.2 35.4 40.1 37.4

Median 26.7 26.3 29.0 15.8 16.4 18.2 19.1 23.6 32.3 33.3 29.4

Mode 20.4 17.6 23.8 14.2 14.0 16.2 19.1 15.3 28.6 34.5 100.0

Minimum 9.9 10.1 12.2 2.0 6.9 8.5 0.9 8.8 16.4 7.1 1.9

Maximum 78.8 79.4 81.6 46.2 46.7 49.0 47.9 71.6 83.3 142.9 111.1

Standard dev.

14.3 15.0 15.3 9.3 8.9 9.2 7.3 12.8 13.8 24.5 25.4

The most significant correlations between the different instruments were: ARP-50 and

EM38_HDP, ARP-170 and EM38_VDP, ARP-170 and Profiler, EM38_VDP and Profiler in all the configurations (Tab. 2).

Moderate or not significant correlations resulted between clay content at 0-50 cm and ECa of all the configurations for the EM38 and the Profiler, while a better correlations resulted with the ARP-50 (Tab.3). Clay content at 50-100 cm correlated either moderately with the ECa of Profiler and EM38, or well with the ECa obtained from ARP-100. On the other hand, highly significant correlations resulted between ECa and moisture content (in mm) at FC, WP and AWC.

Table 2: Pearson correlation coefficients (r) of the three sensors (n = 99, p < 0.01). The most significant correlations between the different sensors are in bold. Correlation coefficient between the same device in different configurations are in italic.

Prof_VDP15

Prof_VDP15 - Prof_VDP10

Prof_VDP10 0.999 - Prof_VDP8

Prof_VDP8 0.998 0.999 - Prof_HDP15

Prof_HDP15 0.978 0.979 0.979 - Prof_HDP10

Prof_HDP10 0.977 0.979 0.979 0.997 - Prof_HDP8

Prof_HDP8 0.977 0.981 0.982 0.996 0.997 - EM38_HDP

EM38_HDP 0.647 0.638 0.633 0.710 0.716 0.693 - EM38_VDP

EM38_VDP 0.924 0.924 0.924 0.919 0.923 0.922 0.764 - ARP-50

ARP-50 0.580 0.564 0.555 0.617 0.607 0.587 0.774 0.654 - ARP-100

ARP-100 0.699 0.688 0.683 0.717 0.710 0.699 0.774 0.753 0.920 - ARP-170

ARP-170 0.764 0.762 0.758 0.783 0.783 0.775 0.766 0.805 0.801 0.872 -

Table 3: Pearson correlation coefficients (r) between clay, skeleton, FC, WP, AWC and the different sensors (n = 13). Bold: p < 0.05; bold underlined: p < 0.01; normal: not significant.



clay 0-50 cm 0.532 0.517 0.503 0.493 0.507 0.482 0.418 0.363 0.682 0.647 0.608

clay 50-100 cm 0.610 0.585 0.569 0.568 0.578 0.553 0.530 0.587 0.808 0.815 0.653

skeleton content -0.663 -0.659 -0.654 -0.723 -0.713 -0.697 -0.661 -0.390 -0.702 -0.631 -0.659

FC 0-100 cm 0.904 0.890 0.879 0.920 0.921 0.910 0.899 0.773 0.908 0.962 0.925

WP 0-100 cm 0.902 0.884 0.870 0.870 0.866 0.850 0.885 0.748 0.937 0.970 0.850

AWC 0-100 cm 0.923 0.918 0.913 0.925 0.942 0.934 0.937 0.874 0.834 0.978 0.749

4 - 40



000,11

ER

ECa

The survey with the EM38-DD and the Profiler EMP-400 was performed on the same day in

August, when soils were dry on surface, whereas the survey with the ARP was carried out in May, when soils were moister and the contact soils-electrodes better. For this work, we did not consider the temperature and the moisture content of the soils, but the textural features only. 13 points were chosen for soil sampling and texture analysis on the basis of the ECa values. The samples were “tout venant” of some kilograms because of the measurement of skeleton content. Laboratory determination of the water content at FC and WP (v/v) was carried out by pressure plate apparatus at -33 and -1,500 kPa matric potential, respectively (Kassel and Nielsen, 1986). Each soil horizons was analyzed in triplicate, and the corresponding bulk density values were used to convert -33 and -1,500 kPa gravimetric water content to a volumetric basis. AWC was determined as the difference between water content at FC and WP.

RESULTS AND DISCUSSION The three instruments produced similar spatial patterns (Fig.2). During the proximal survey, the

EMI sensors (EM38-DD and Profiler EMP-400) in the HDP orientation registered negative or very low values in some vineyard areas. This was probably due to the interference of the iron wires of the vineyard rows or other iron materials with the magnetic field. Therefore, these wrong data (negative, or very close to 0) measured in HDP orientation were deleted before data interpolation.

As a whole, the ECa values measured by ARP device were higher, whereas the values measured by Profiler and EM-38 were similar (Tab.1).

Table 1: Descriptive statistics of ECa (mS m-1) measured with the three devices. The negative and very low values were not considered.



Mean 31.8 32.2 34.6 19.2 20.0 21.9 19.5 28.2 35.4 40.1 37.4

Median 26.7 26.3 29.0 15.8 16.4 18.2 19.1 23.6 32.3 33.3 29.4

Mode 20.4 17.6 23.8 14.2 14.0 16.2 19.1 15.3 28.6 34.5 100.0

Minimum 9.9 10.1 12.2 2.0 6.9 8.5 0.9 8.8 16.4 7.1 1.9

Maximum 78.8 79.4 81.6 46.2 46.7 49.0 47.9 71.6 83.3 142.9 111.1

Standard dev.

14.3 15.0 15.3 9.3 8.9 9.2 7.3 12.8 13.8 24.5 25.4

The most significant correlations between the different instruments were: ARP-50 and

EM38_HDP, ARP-170 and EM38_VDP, ARP-170 and Profiler, EM38_VDP and Profiler in all the configurations (Tab. 2).

Moderate or not significant correlations resulted between clay content at 0-50 cm and ECa of all the configurations for the EM38 and the Profiler, while a better correlations resulted with the ARP-50 (Tab.3). Clay content at 50-100 cm correlated either moderately with the ECa of Profiler and EM38, or well with the ECa obtained from ARP-100. On the other hand, highly significant correlations resulted between ECa and moisture content (in mm) at FC, WP and AWC.

Table 2: Pearson correlation coefficients (r) of the three sensors (n = 99, p < 0.01). The most significant correlations between the different sensors are in bold. Correlation coefficient between the same device in different configurations are in italic.

Prof_VDP15

Prof_VDP15 - Prof_VDP10

Prof_VDP10 0.999 - Prof_VDP8

Prof_VDP8 0.998 0.999 - Prof_HDP15

Prof_HDP15 0.978 0.979 0.979 - Prof_HDP10

Prof_HDP10 0.977 0.979 0.979 0.997 - Prof_HDP8

Prof_HDP8 0.977 0.981 0.982 0.996 0.997 - EM38_HDP

EM38_HDP 0.647 0.638 0.633 0.710 0.716 0.693 - EM38_VDP

EM38_VDP 0.924 0.924 0.924 0.919 0.923 0.922 0.764 - ARP-50

ARP-50 0.580 0.564 0.555 0.617 0.607 0.587 0.774 0.654 - ARP-100

ARP-100 0.699 0.688 0.683 0.717 0.710 0.699 0.774 0.753 0.920 - ARP-170

ARP-170 0.764 0.762 0.758 0.783 0.783 0.775 0.766 0.805 0.801 0.872 -

Table 3: Pearson correlation coefficients (r) between clay, skeleton, FC, WP, AWC and the different sensors (n = 13). Bold: p < 0.05; bold underlined: p < 0.01; normal: not significant.



clay 0-50 cm 0.532 0.517 0.503 0.493 0.507 0.482 0.418 0.363 0.682 0.647 0.608

clay 50-100 cm 0.610 0.585 0.569 0.568 0.578 0.553 0.530 0.587 0.808 0.815 0.653

skeleton content -0.663 -0.659 -0.654 -0.723 -0.713 -0.697 -0.661 -0.390 -0.702 -0.631 -0.659

FC 0-100 cm 0.904 0.890 0.879 0.920 0.921 0.910 0.899 0.773 0.908 0.962 0.925

WP 0-100 cm 0.902 0.884 0.870 0.870 0.866 0.850 0.885 0.748 0.937 0.970 0.850

AWC 0-100 cm 0.923 0.918 0.913 0.925 0.942 0.934 0.937 0.874 0.834 0.978 0.749

4 - 41



Fig.2: ECa maps of the vineyard, obtained with the three different sensors. The variogram parameters for the interpolation were the same for all the maps (ordinary kriging).

CONCLUSIONS The ECa pattern obtained with the three sensors was similar (Fig.2). The instruments could

provide for a rapid, non invasive and relatively cheap soil survey in a difficult environment, like that of a vineyard on clayey and stony soils, although iron materials in the vineyard sometimes interfered with the magnetic fields of the EMI sensors in the HDP configuration. The cumulative response of Profiler did not change at different frequencies and was very similar to the EM38_VDP response. On top of that, both sensors were strongly correlated with ARP-170, except for EM38_HDP, which was better correlated with ARP-50 and ARP-100. Correlations between ECa and hydrological parameters namely FC, WP and AWC resulted highly significant.

In conclusion, the proximal survey performed by any of these instruments can complement the traditional soil survey methods and allow a high quality predictive mapping of important soil hydrological properties like FC, WP and AWC.

ACKNOWLEDGMENTS The authors are grateful to Annamaria Castrignanò, Donatello Sollitto and Daniela De

Benedetto (CRA-SCA, Bari, Italy) for the loan, the advices and the assistance with the EM38-DD, as well as to the SO.IN.G and the Geostudi Astier (Livorno) for the loan and the collaboration with the ARP and the Profiler EMP-400, respectively. A special thank is for the farm “Barone Ricasoli s.p.a.”, Gaiole in Chianti, Siena (Italy) for the economic support to the research and for granting the access to its vineyards.

BIBLIOGRAPHY

Cousin I., Besson A., Bourennane H., Pasquier C., Nicoullaud B., King D., Richard G., 2009. From spatial-continuous electrical resistivity measurements to the soil hydraulic functioning at the field scale. C.R. Geoscience, 341: 859-867.

Costantini, E. A. C., Pellegrini, S., Bucelli, P., Storchi, P., Vignozzi, N., Barbetti, R., Campagnolo, S., 2009. Relevance of the Lin’s and Host hydropedological models to predict grape yield and wine quality. Hydrol. Earth Syst. Sci., 13: 1635-1648.

Doussan C., Ruy S., 2009. Prediction of unsaturated soil hydraulic conductivity with electrical conductivity. Water Resources Research, 45, W10408.

Davies R., 2004. Mapping soil properties for irrigation development in the Riverland of South Australia using EM38. SuperSoil 2004, 3d Australian and New Zealand Soils Conference, 5-9 December 2004, University of Sydney, Australia, published on CD_ROM, website http://www.regional.org.au/au/asssi/supersoil2004/s5/oral.

Doolittle J., Petersen M., Wheeler T., 2001. Comparison of two electromagnetic induction tools in salinity appraisals. Journal of Soil and Water Conservation, 56: 257-262.

Geonics Limited, 1998. EM38 ground conductivity meter operating manual. Mississangua, Ontario, Canada. www.geonics.com

McNeil J.D., 1990. Geonics EM38 Ground Conductivity Meter: EM38 Operating Manual, Geonics Limited, Ontario, Canada.

Morari F., Castrignanò A., Pagliarin C., 2009. Application of multivariate geostatistics in delineating management zones within a gravelly vineyard using geo-electrical sensors. Computers and Electronics in Agriculture, 68: 97-107.

Proffitt T., Bramley R., Lamb D., Winter E., 2006. Precision Viticulture. http://www.pvaustralia.com.au/book.html, pp. 96.

Tromp-van Meerveld H.J., McDonnell J.J., 2009. Assessment of multi-frequency electromagnetic induction for determining soil moisture patterns at hill slope scale. Journal of Hydrology, 368: 56-67.

4 - 42



Fig.2: ECa maps of the vineyard, obtained with the three different sensors. The variogram parameters for the interpolation were the same for all the maps (ordinary kriging).

CONCLUSIONS The ECa pattern obtained with the three sensors was similar (Fig.2). The instruments could

provide for a rapid, non invasive and relatively cheap soil survey in a difficult environment, like that of a vineyard on clayey and stony soils, although iron materials in the vineyard sometimes interfered with the magnetic fields of the EMI sensors in the HDP configuration. The cumulative response of Profiler did not change at different frequencies and was very similar to the EM38_VDP response. On top of that, both sensors were strongly correlated with ARP-170, except for EM38_HDP, which was better correlated with ARP-50 and ARP-100. Correlations between ECa and hydrological parameters namely FC, WP and AWC resulted highly significant.

In conclusion, the proximal survey performed by any of these instruments can complement the traditional soil survey methods and allow a high quality predictive mapping of important soil hydrological properties like FC, WP and AWC.

ACKNOWLEDGMENTS The authors are grateful to Annamaria Castrignanò, Donatello Sollitto and Daniela De

Benedetto (CRA-SCA, Bari, Italy) for the loan, the advices and the assistance with the EM38-DD, as well as to the SO.IN.G and the Geostudi Astier (Livorno) for the loan and the collaboration with the ARP and the Profiler EMP-400, respectively. A special thank is for the farm “Barone Ricasoli s.p.a.”, Gaiole in Chianti, Siena (Italy) for the economic support to the research and for granting the access to its vineyards.

BIBLIOGRAPHY

Cousin I., Besson A., Bourennane H., Pasquier C., Nicoullaud B., King D., Richard G., 2009. From spatial-continuous electrical resistivity measurements to the soil hydraulic functioning at the field scale. C.R. Geoscience, 341: 859-867.

Costantini, E. A. C., Pellegrini, S., Bucelli, P., Storchi, P., Vignozzi, N., Barbetti, R., Campagnolo, S., 2009. Relevance of the Lin’s and Host hydropedological models to predict grape yield and wine quality. Hydrol. Earth Syst. Sci., 13: 1635-1648.

Doussan C., Ruy S., 2009. Prediction of unsaturated soil hydraulic conductivity with electrical conductivity. Water Resources Research, 45, W10408.

Davies R., 2004. Mapping soil properties for irrigation development in the Riverland of South Australia using EM38. SuperSoil 2004, 3d Australian and New Zealand Soils Conference, 5-9 December 2004, University of Sydney, Australia, published on CD_ROM, website http://www.regional.org.au/au/asssi/supersoil2004/s5/oral.

Doolittle J., Petersen M., Wheeler T., 2001. Comparison of two electromagnetic induction tools in salinity appraisals. Journal of Soil and Water Conservation, 56: 257-262.

Geonics Limited, 1998. EM38 ground conductivity meter operating manual. Mississangua, Ontario, Canada. www.geonics.com

McNeil J.D., 1990. Geonics EM38 Ground Conductivity Meter: EM38 Operating Manual, Geonics Limited, Ontario, Canada.

Morari F., Castrignanò A., Pagliarin C., 2009. Application of multivariate geostatistics in delineating management zones within a gravelly vineyard using geo-electrical sensors. Computers and Electronics in Agriculture, 68: 97-107.

Proffitt T., Bramley R., Lamb D., Winter E., 2006. Precision Viticulture. http://www.pvaustralia.com.au/book.html, pp. 96.

Tromp-van Meerveld H.J., McDonnell J.J., 2009. Assessment of multi-frequency electromagnetic induction for determining soil moisture patterns at hill slope scale. Journal of Hydrology, 368: 56-67.

4 - 43



GEOLOGIC AND GEOMORPHOLOGIC FEATURES APPLIED FORIDENTIFICATION OF WINE TERROIR UNITS BY DIGITAL IMAGEPROCESSING, SPECTRORADIOMETRIC AND GIS TECHNIQUES IN

ENCRUZILHADA DO SUL, RS, BRAZIL

Rosemary Hoff(1), Jorge Ricardo Ducati(2), Magda Bergmann(3)(1) Embrapa Uva e Vinho/CNPUV

Empresa Brasileira de Pesquisa AgropecuáriaRua Livramento, 515 - 95700-000 - Bento Gonçalves – RS - Brasil

[email protected] (2) Centro Estadual de Pesquisas em Sensoriamento Remoto e Meteorologia/CEPSRM

Universidade Federal do Rio Grande do SulAv. Bento Gonçalves, 9500 – 91501-970 - Porto Alegre – RS - Brasil

[email protected](3) Companhia de Pesquisa de Recursos Minerais/CPRM – Serviço Geológico do Brasil

Rua Banco da Província, 105 – CEP 90840-030 - Porto Alegre – [email protected]

ABSTRACTResults in the characterization of a new wine terroir unit in south Brazil are reported.

Presently, several areas in Brazil are being studied, in an effort to define new wine terroirsand improve the quality of Brazilian wines. This paper reports what is being done, byEmbrapa (Brazilian Agricultural Research Corporation) and its partners Remote Sensing andMeteorological Research Center (CEPSRM/UFRGS) and Brazilian Geological Survey(CPRM), in the Encruzilhada do Sul region, at Rio Grande do Sul State, that is part of theSerra do Sudeste viticultural region. Satellite images from several sources (SRTM, ASTER,ALOS) were used, together with field data (rock samples). Digital elevation models were builtand used to define areas with slopes and solar expositions adequate to vine growing, withaltitudes above 350 m. Spectroradiometry of rock samples was performed, to identify severalminerals (montmorilonite, illite, pyrophilite and kaolinite). Geologic maps were used to locaterock types to collected in field trips; those rocks had their spectral response extracted fromradiometry, and fitted to the six bands of ASTER SWIR subsystem, resulting in a map of thedistribution of these rocks in some areas of interest. Two wineries were more closely studied.The first area produces wine from 35 hectares of Cabernet Sauvignon, Merlot, Nebbiolo,Pinot Noir and Chardonnay. The other winery has 61 hectares and produces Pinot Noir andChardonnay grapes for sparkling wines. The study concludes that the use of remote sensingresources and associated geotechnologies are effective to terroir studies.

KEYWORDBrazilian wines; geology; geomorphology; spectroradiometry; geographical information

system.

INTRODUCTIONIn Brazil, spectroradiometric studies have been developed to characterize new viticultural

geographic appellations, and for this effort, geotechnologies have been used by Embrapa andEmbrapa Grape and Wine Research Center (Centro Nacional de Pesquisas em Uva e

Vinho/CNPUV) is located in Bento Gonçalves, at Rio Grande do Sul State, at the mainBrazilian wine production region. One of the more promising new studied areas is locatednear Encruzilhada do Sul city, in Serra do Sudeste viticultural region (Figure 1). This area isdominated by extensive grasslands with minor forest patches developed over old Precambrianterrains represented mainly by granitic-gnaissic rocks. Soils are mainly cambisolos andargisolos, poor in organic matter, well drained and occasionally gravely. The whole region ispart of the Pampa Biome (Embrapa 2008), which also covers large areas in neighboringUruguay and Argentina.

Figure 1. Location and vineyards in Encruzilhada do Sul, Brazil.

Vineyards for fine wines (Vitis vinifera) were first introduced around 1970, and presentlythere are more than 300 hectares with grapes such as Barbera, Cabernet Franc, CabernetSauvignon, Merlot, Periquita, Teroldego, Marselan, Pinot Noir, Ancelota, Malbec, TourigaNacional, Gamay, Arinarnoa, Alicante Bouschet, Chardonnay, Gewürztraminer, Malvasia deCândia, Sauvignon Blanc, Riesling, and others. Such a large diversity of vineyards proves anongoing search for grape varieties that leads to a terroir identity. This identity, in fact, isalready suggested by Cemin and Ducati (2008), which detected spectral differences betweengrape varieties in Encruzilhada do Sul, when compared with French and Chilean vineyards.Further studies on the area were performed by Hoff et al. (2007), comparing geology andrelief units in the region, while Hoff et al. (2009) studied relief from two sets of orbital data;Bergmann et al. (2009) linked digital elevation models (DEM) and geological surveys tovineyards data. This work presents additional studies on this region, relating physical data asrock and landscape, which give specificity to wines, and thus define a potential terroir.

4 - 44



GEOLOGIC AND GEOMORPHOLOGIC FEATURES APPLIED FORIDENTIFICATION OF WINE TERROIR UNITS BY DIGITAL IMAGEPROCESSING, SPECTRORADIOMETRIC AND GIS TECHNIQUES IN

ENCRUZILHADA DO SUL, RS, BRAZIL

Rosemary Hoff(1), Jorge Ricardo Ducati(2), Magda Bergmann(3)(1) Embrapa Uva e Vinho/CNPUV

Empresa Brasileira de Pesquisa AgropecuáriaRua Livramento, 515 - 95700-000 - Bento Gonçalves – RS - Brasil

[email protected] (2) Centro Estadual de Pesquisas em Sensoriamento Remoto e Meteorologia/CEPSRM

Universidade Federal do Rio Grande do SulAv. Bento Gonçalves, 9500 – 91501-970 - Porto Alegre – RS - Brasil

[email protected](3) Companhia de Pesquisa de Recursos Minerais/CPRM – Serviço Geológico do Brasil

Rua Banco da Província, 105 – CEP 90840-030 - Porto Alegre – [email protected]

ABSTRACTResults in the characterization of a new wine terroir unit in south Brazil are reported.

Presently, several areas in Brazil are being studied, in an effort to define new wine terroirsand improve the quality of Brazilian wines. This paper reports what is being done, byEmbrapa (Brazilian Agricultural Research Corporation) and its partners Remote Sensing andMeteorological Research Center (CEPSRM/UFRGS) and Brazilian Geological Survey(CPRM), in the Encruzilhada do Sul region, at Rio Grande do Sul State, that is part of theSerra do Sudeste viticultural region. Satellite images from several sources (SRTM, ASTER,ALOS) were used, together with field data (rock samples). Digital elevation models were builtand used to define areas with slopes and solar expositions adequate to vine growing, withaltitudes above 350 m. Spectroradiometry of rock samples was performed, to identify severalminerals (montmorilonite, illite, pyrophilite and kaolinite). Geologic maps were used to locaterock types to collected in field trips; those rocks had their spectral response extracted fromradiometry, and fitted to the six bands of ASTER SWIR subsystem, resulting in a map of thedistribution of these rocks in some areas of interest. Two wineries were more closely studied.The first area produces wine from 35 hectares of Cabernet Sauvignon, Merlot, Nebbiolo,Pinot Noir and Chardonnay. The other winery has 61 hectares and produces Pinot Noir andChardonnay grapes for sparkling wines. The study concludes that the use of remote sensingresources and associated geotechnologies are effective to terroir studies.

KEYWORDBrazilian wines; geology; geomorphology; spectroradiometry; geographical information

system.

INTRODUCTIONIn Brazil, spectroradiometric studies have been developed to characterize new viticultural

geographic appellations, and for this effort, geotechnologies have been used by Embrapa andEmbrapa Grape and Wine Research Center (Centro Nacional de Pesquisas em Uva e

Vinho/CNPUV) is located in Bento Gonçalves, at Rio Grande do Sul State, at the mainBrazilian wine production region. One of the more promising new studied areas is locatednear Encruzilhada do Sul city, in Serra do Sudeste viticultural region (Figure 1). This area isdominated by extensive grasslands with minor forest patches developed over old Precambrianterrains represented mainly by granitic-gnaissic rocks. Soils are mainly cambisolos andargisolos, poor in organic matter, well drained and occasionally gravely. The whole region ispart of the Pampa Biome (Embrapa 2008), which also covers large areas in neighboringUruguay and Argentina.

Figure 1. Location and vineyards in Encruzilhada do Sul, Brazil.

Vineyards for fine wines (Vitis vinifera) were first introduced around 1970, and presentlythere are more than 300 hectares with grapes such as Barbera, Cabernet Franc, CabernetSauvignon, Merlot, Periquita, Teroldego, Marselan, Pinot Noir, Ancelota, Malbec, TourigaNacional, Gamay, Arinarnoa, Alicante Bouschet, Chardonnay, Gewürztraminer, Malvasia deCândia, Sauvignon Blanc, Riesling, and others. Such a large diversity of vineyards proves anongoing search for grape varieties that leads to a terroir identity. This identity, in fact, isalready suggested by Cemin and Ducati (2008), which detected spectral differences betweengrape varieties in Encruzilhada do Sul, when compared with French and Chilean vineyards.Further studies on the area were performed by Hoff et al. (2007), comparing geology andrelief units in the region, while Hoff et al. (2009) studied relief from two sets of orbital data;Bergmann et al. (2009) linked digital elevation models (DEM) and geological surveys tovineyards data. This work presents additional studies on this region, relating physical data asrock and landscape, which give specificity to wines, and thus define a potential terroir.

4 - 45



MATERIALS AND METHODSCartography was based on SH-22-Y-A-VI-2 map 1:50,000 scale, from the Brazilian Army.

Digital elevation models (DEM) were generated from three orbital sources: the Space ShuttleDEM (SRTM 2008) to characterize regional features on low resolution; the ASTER imager(Abrams and Hook 2002) to study the areas on medium resolution; and ALOS satellite(ALOS 2009) to characterize the vineyards on high resolution. Soil and geological maps wereproduced by Embrapa and CPRM, respectively.

Thematic relief and landscape maps providing information were produced byCEPSRM/UFRGS through digital image processing. Those products were grouped in aGeographic Information System (GIS) environment, using ground and satellite data, toproduce a suitable agricultural zoning, searching for potential terroir units.

Digital elevation models produced slope, elevation, and solar orientation maps, which wereintegrated, to be used in wine production planning.

Rock samples were collected in field trips, and analyzed with spectroradiometric techniquesby POSAM - Portable Spectroradiometer for Mineral identification and MISO - MineralIdentification Software (Dowa, 2003) at CPRM. The spectral signatures lead to theidentification of the main species of minerals and their spectra was degraded to match the sixbands of ASTER SWIR subsystem, thus allowing a comparison with satellite data, leading tothe identification of rock types on images.

RESULTS AND DISCUSSIONFigure 2A shows the elevation map. Geomorphologic features like areas above 350 m are

highlighted, since these higher places tend to yield better wines. Relief features as slope andsolar orientation are important factors to wine quality. These features were crossed with areasabove than 350 m to show the best places to grape cropping (Figure 2B).

Figure 2. A: DEM generated from SRTM data, where shaded relief shows areas above 350 m;B: from ASTER data, areas above 350 m, flat relief and northern exposition (Hoff et al.

2007). Dashed lines are roads.

Figure 3A shows the geologic map of the region. Figure 3B shows the reflectance spectra offive rock types, expressed in the six bands of ASTER SWIR subsystem. These signatureswere used to detect these five rocks in the terrain imaged by ASTER, producing a classifiedimage for areas above 350 m (Figure 3C). Spectral radiometric techniques identified mineralssuch as montmorilonite, illite, pyrophilite and kaolinite.

Figure 3. Geological map (A), adapted from CPRM (2008), showing location of wineriesLídio Carraro (1) and Chandon do Brasil (2) in Encruzilhada do Sul; spectral signature of

rocks (B); classified image for geology (C), from Hoff et al. (2009).

Two vineyard areas were more closely studied. Figure 4 shows the Lidio Carraro winery,with 35 ha of Cabernet Sauvignon, Merlot, Nebbiolo, Pinot Noir and Chardonnay grapes .The digital elevation model showed maximum elevation of 360 m, and slopes mainly in 3-8%and 8-30% classes; solar exposition is mostly North and Northeast.

The second area is Chandon do Brasil winery, which produces sparkling wine from PinotNoir and Chardonnay cropped over 61 ha. DEM indicated an elevation varying from 350 to380 m, slopes from 3-8% to 8-30%, and aspect (solar exposition) between North, Northwestand West (Figure 5).

4 - 46



MATERIALS AND METHODSCartography was based on SH-22-Y-A-VI-2 map 1:50,000 scale, from the Brazilian Army.

Digital elevation models (DEM) were generated from three orbital sources: the Space ShuttleDEM (SRTM 2008) to characterize regional features on low resolution; the ASTER imager(Abrams and Hook 2002) to study the areas on medium resolution; and ALOS satellite(ALOS 2009) to characterize the vineyards on high resolution. Soil and geological maps wereproduced by Embrapa and CPRM, respectively.

Thematic relief and landscape maps providing information were produced byCEPSRM/UFRGS through digital image processing. Those products were grouped in aGeographic Information System (GIS) environment, using ground and satellite data, toproduce a suitable agricultural zoning, searching for potential terroir units.

Digital elevation models produced slope, elevation, and solar orientation maps, which wereintegrated, to be used in wine production planning.

Rock samples were collected in field trips, and analyzed with spectroradiometric techniquesby POSAM - Portable Spectroradiometer for Mineral identification and MISO - MineralIdentification Software (Dowa, 2003) at CPRM. The spectral signatures lead to theidentification of the main species of minerals and their spectra was degraded to match the sixbands of ASTER SWIR subsystem, thus allowing a comparison with satellite data, leading tothe identification of rock types on images.

RESULTS AND DISCUSSIONFigure 2A shows the elevation map. Geomorphologic features like areas above 350 m are

highlighted, since these higher places tend to yield better wines. Relief features as slope andsolar orientation are important factors to wine quality. These features were crossed with areasabove than 350 m to show the best places to grape cropping (Figure 2B).

Figure 2. A: DEM generated from SRTM data, where shaded relief shows areas above 350 m;B: from ASTER data, areas above 350 m, flat relief and northern exposition (Hoff et al.

2007). Dashed lines are roads.

Figure 3A shows the geologic map of the region. Figure 3B shows the reflectance spectra offive rock types, expressed in the six bands of ASTER SWIR subsystem. These signatureswere used to detect these five rocks in the terrain imaged by ASTER, producing a classifiedimage for areas above 350 m (Figure 3C). Spectral radiometric techniques identified mineralssuch as montmorilonite, illite, pyrophilite and kaolinite.

Figure 3. Geological map (A), adapted from CPRM (2008), showing location of wineriesLídio Carraro (1) and Chandon do Brasil (2) in Encruzilhada do Sul; spectral signature of

rocks (B); classified image for geology (C), from Hoff et al. (2009).

Two vineyard areas were more closely studied. Figure 4 shows the Lidio Carraro winery,with 35 ha of Cabernet Sauvignon, Merlot, Nebbiolo, Pinot Noir and Chardonnay grapes .The digital elevation model showed maximum elevation of 360 m, and slopes mainly in 3-8%and 8-30% classes; solar exposition is mostly North and Northeast.

The second area is Chandon do Brasil winery, which produces sparkling wine from PinotNoir and Chardonnay cropped over 61 ha. DEM indicated an elevation varying from 350 to380 m, slopes from 3-8% to 8-30%, and aspect (solar exposition) between North, Northwestand West (Figure 5).

4 - 47



Figure 4 – Vinícola Lídio Carraro, Encruzilhada do Sul, Brazil.

Figure 5 – Chandon do Brasil, Encruzilhada do Sul, Brazil.

CONCLUSIONSThis study showed how the integrated use of geotechnologies and spectral analysis from

radiometry, supported by field data, can contribute effectively to viticulture studies aiming toterroir characterization. Satellite images from several sources, such as low (SRTM), medium(ASTER) and high (ALOS) resolutions, are effective to produce information on geological,

topographical, and solar orientation data, important to establish criteria to identify potentialterroir units, to agricultural zoning, and to strategies in vineyard and wine production.

ACKNOWLEDGMENTSThis study was partly supported by the Brazilian Conselho Nacional de Desenvolvimento

Científico e Tecnológico/CNPq, in a project coordinated by Embrapa Uva e Vinho, inpartnership with CEPSRM and CPRM.

BIBLIOGRAPHYAbrams, M., Hook, S. 2002. ASTER User Handbook: Advanced Spaceborne Thermal

Emission and Reflection Radiometer. USA: NASA/Jet Propulsion Laboratory CaliforniaInstitute of Technology, v.2, 2002, 135p. Available online inhttp://asterweb.jpl.nasa.gov/content/03_data/04_Documents/aster_user_guide_v2.pdf;accessed in 2 apr. 2010.

ALOS - Advanced Land Observing Satellite. 2009. Available online inhttps://ursa.aadn.alaska.edu/cgi-bin/login/guest/; accessed in 2 apr. 2010.

Bergmann, M.; Hoff, R.; Ducati, J.R.; Bombassaro, M.G.; Costa, G.L. 2009. Geologia evinho: Um novo enfoque para terroir vitícola na Região de Encruzilhada do Sul, Brasil. In:XII Congreso Latinoamericano de Viticultura y Enología, Montevidéu. Anales XIICLAVE (CD). Montevideo: PROMOVER, 2009.

Cemin, G.; Ducati, J. R. 2008. On the Stability of Spectral Features of Four Vine Varieties inBrazil, Chile and France. In: VIIth International Terroir Congress, 2008, Nyon.Proceedings of the VIIth International Terroir Congress. Nyon: Agroscope ChanginsWädenswil, 2008. v. 1. p. 475-480.

CPRM – Companhia de Pesquisa de Recursos Minerais 2008. Mapa Geológico do RioGrande do Sul. Escala 1:750.000. CPRM/MME. Available online inhttp://www.cprm.gov.br/

DOWA Engineering Co., Ltd. 2003. POSAM POrtable SpectrorAdiometer for Mineralidentification and MISO Ver.1.0 Mineral Identification Software User's Manual, 22 pp.Tokyo.

Embrapa - Empresa Brasileira de Pesquisa Agropécuária. 2009. A Embrapa nos biomasbrasileiros: Avanços no Manejo Sustentável dos Recursos Naturais. MAPA, Brasília.Available: http://www.embrapa.br/publicacoes/institucionais/laminas-biomas.pdf/view

Hoff, R.; Ducati, J. R.; Bergmann, M. 2009. Comparação de dados de modelo digital deelevação - MDE: ASTER e SRTM por processamento digital de imagem para identificaçãode terroir vitivinícola na Folha Encruzilhada do Sul, RS, Brasil. In: XIV SimpósioBrasileiro de Sensoriamento Remoto, 2009, Natal. Anais XIV Simpósio Brasileiro deSensoriamento Remoto. S. J. dos Campos : INPE, v.1. p.1-8. Available online inhttp://marte.dpi.inpe.br/col/dpi.inpe.br/sbsr@80/2008/11.18.02.00.46/doc/215-222.pdf

Hoff, R.; Ducati, J. R.; Flores, C. A.; Iglesias, C M F. 2007. Aspectos geológicos egeomorfológicos da identificação de critérios para estabelecimento de terroirs na MetadeSul (Rio Grande Do Sul, Brasil) pela aplicação de processamento digital de imagemASTER. In: V Congreso Uruguayo de Geología, 2007, Montevidéo. Resumenes VCongreso Uruguayo de Geología. Montevidéo : SUGEOLOGIA, 2007. v. 1. p. 120-120.

SRTM - Shuttle Radar Topography Mission. 2008. Available: http://www2.jpl.nasa.gov/srtm/

4 - 48



Figure 4 – Vinícola Lídio Carraro, Encruzilhada do Sul, Brazil.

Figure 5 – Chandon do Brasil, Encruzilhada do Sul, Brazil.

CONCLUSIONSThis study showed how the integrated use of geotechnologies and spectral analysis from

radiometry, supported by field data, can contribute effectively to viticulture studies aiming toterroir characterization. Satellite images from several sources, such as low (SRTM), medium(ASTER) and high (ALOS) resolutions, are effective to produce information on geological,

topographical, and solar orientation data, important to establish criteria to identify potentialterroir units, to agricultural zoning, and to strategies in vineyard and wine production.

ACKNOWLEDGMENTSThis study was partly supported by the Brazilian Conselho Nacional de Desenvolvimento

Científico e Tecnológico/CNPq, in a project coordinated by Embrapa Uva e Vinho, inpartnership with CEPSRM and CPRM.

BIBLIOGRAPHYAbrams, M., Hook, S. 2002. ASTER User Handbook: Advanced Spaceborne Thermal

Emission and Reflection Radiometer. USA: NASA/Jet Propulsion Laboratory CaliforniaInstitute of Technology, v.2, 2002, 135p. Available online inhttp://asterweb.jpl.nasa.gov/content/03_data/04_Documents/aster_user_guide_v2.pdf;accessed in 2 apr. 2010.

ALOS - Advanced Land Observing Satellite. 2009. Available online inhttps://ursa.aadn.alaska.edu/cgi-bin/login/guest/; accessed in 2 apr. 2010.

Bergmann, M.; Hoff, R.; Ducati, J.R.; Bombassaro, M.G.; Costa, G.L. 2009. Geologia evinho: Um novo enfoque para terroir vitícola na Região de Encruzilhada do Sul, Brasil. In:XII Congreso Latinoamericano de Viticultura y Enología, Montevidéu. Anales XIICLAVE (CD). Montevideo: PROMOVER, 2009.

Cemin, G.; Ducati, J. R. 2008. On the Stability of Spectral Features of Four Vine Varieties inBrazil, Chile and France. In: VIIth International Terroir Congress, 2008, Nyon.Proceedings of the VIIth International Terroir Congress. Nyon: Agroscope ChanginsWädenswil, 2008. v. 1. p. 475-480.

CPRM – Companhia de Pesquisa de Recursos Minerais 2008. Mapa Geológico do RioGrande do Sul. Escala 1:750.000. CPRM/MME. Available online inhttp://www.cprm.gov.br/

DOWA Engineering Co., Ltd. 2003. POSAM POrtable SpectrorAdiometer for Mineralidentification and MISO Ver.1.0 Mineral Identification Software User's Manual, 22 pp.Tokyo.

Embrapa - Empresa Brasileira de Pesquisa Agropécuária. 2009. A Embrapa nos biomasbrasileiros: Avanços no Manejo Sustentável dos Recursos Naturais. MAPA, Brasília.Available: http://www.embrapa.br/publicacoes/institucionais/laminas-biomas.pdf/view

Hoff, R.; Ducati, J. R.; Bergmann, M. 2009. Comparação de dados de modelo digital deelevação - MDE: ASTER e SRTM por processamento digital de imagem para identificaçãode terroir vitivinícola na Folha Encruzilhada do Sul, RS, Brasil. In: XIV SimpósioBrasileiro de Sensoriamento Remoto, 2009, Natal. Anais XIV Simpósio Brasileiro deSensoriamento Remoto. S. J. dos Campos : INPE, v.1. p.1-8. Available online inhttp://marte.dpi.inpe.br/col/dpi.inpe.br/sbsr@80/2008/11.18.02.00.46/doc/215-222.pdf

Hoff, R.; Ducati, J. R.; Flores, C. A.; Iglesias, C M F. 2007. Aspectos geológicos egeomorfológicos da identificação de critérios para estabelecimento de terroirs na MetadeSul (Rio Grande Do Sul, Brasil) pela aplicação de processamento digital de imagemASTER. In: V Congreso Uruguayo de Geología, 2007, Montevidéo. Resumenes VCongreso Uruguayo de Geología. Montevidéo : SUGEOLOGIA, 2007. v. 1. p. 120-120.

SRTM - Shuttle Radar Topography Mission. 2008. Available: http://www2.jpl.nasa.gov/srtm/

4 - 49



INFLUENCE OF BASALT ON THE TERROIR OF THE COLUMBIA VALLEY AMERICAN VITICULTURAL AREA

K. R. Pogue

Department of Geology, Whitman College 345 Boyer Ave., Walla Walla, Washington 99362 USA

[email protected] ABSTRACT

The Columbia Valley American Viticultural Area (AVA) of the Pacific Northwest, USA is the world's largest officially recognized viticultural area with basalt bedrock. However, most Columbia Valley vineyards are planted in soils derived from thick loess and glacial flood sediments, rather than the underlying bedrock. Recently, vineyard plantings have expanded into parts of the AVA where basalt and basalt weathering products, derived either naturally or through mechanical ripping, are a major soil component. Tests were conducted to determine how the addition of a basalt component to soils could affect the terroir of Columbia Valley vineyards. To test for the chemical influence of basalt, samples were obtained from soils representative of the range of basalt influence and analyzed for iron content. Increases of 77% to 233% in available iron were observed in vineyards with basalt component relative to vineyards planted in grass-covered loess. To measure the thermal influence of basalt, temperature data loggers were installed within soils and grape clusters in basalt-covered and grass-covered vineyards. Temperature loggers in the basalt-covered vineyard recorded an 18% increase in average soil temperature at a depth of 5 cm, a 13% increase in average soil temperature at a depth of 25 cm, and a 4% in average cluster temperature relative to those in the grass-covered vineyard. Cluster temperatures in the basalt-covered vineyard were generally higher than in the grass-covered vineyard from late morning through early evening, equilibrating rapidly near sunset. KEYWORDS basalt - terroir - soil - Columbia Valley INTRODUCTION

Worldwide, the percentage of vineyards planted in basalt-derived soils is relatively small. Notable viticultural areas with soils developed in weakly weathered to unweathered basalt include the Canary Islands, the Azores, and Sicily's Mt. Etna. Regions that host vineyards planted in older or more deeply weathered basaltic soils include western India, southern Australia, Oregon's Willamette Valley, south-central France, northern Italy, and Hungary's Badascony region. The world's largest government-designated viticultural region with basalt-dominated bedrock is the Columbia Valley AVA, which encompasses over 45,000 km2 of the states of Washington and Oregon (Gregutt, 2002). The basalt was erupted during the Miocene Epoch from volcanoes associated with the hot spot that now lies beneath Yellowstone National Park in Wyoming (Pogue, 2009). The Columbia Valley AVA presently contains over 2700 ha of vineyards that are located primarily on gentle slopes or on valley floors below 400 m in elevation. Almost all of the vineyards are planted in loess derived from the

deflation of sediments deposited by a series of catastrophic Pleistocene glacial outburst floods, known as the Missoula floods (Busacca, Meinert, 2003). At elevations below 330 m, the loess commonly overlies sand, silt, and gravel deposited by the Missoula floods, while above this elevation, the loess directly overlies basalt bedrock. Despite its age, the basalt is not deeply weathered due to the combined affects of the region's arid climate and the protective mantle of flood- and wind-deposited sediment. The most obvious effects of weathering are fracture networks filled with calcium carbonate and iron oxide-stained clays within the uppermost 1 m of the basalt. The soils in most Columbia Valley AVA vineyards were derived by the glacial and fluvial erosion of regions dominated by granitic and metasedimentary bedrock that lie north and east of the Columbia Valley AVA. They are therefore rich in quartz, muscovite mica, and potassium feldspar, minerals not present in the underlying basalt bedrock. Since the thickness of these soils generally exceeds the rooting depth of the vines, basalt has had, until recently, almost no influence on terroir.

Over the last 10 years, viticulture in the Columbia Valley AVA has rapidly expanded. Vineyards have recently been planted at higher elevations, on steeper slopes, and in rocky, alluvial soils. The soils in these vineyards are commonly much thinner than those of the traditional valley floor sites, and therefore vine roots are able to directly interact with basalt bedrock or basalt-derived alluvium or colluvium. In preparation for planting, the thin soils are often mechanically ripped to a depth of 0.5 to 1 m to increase water holding capacity and available rooting depth. The ripping process crushes the upper parts of the weakly weathered basalt bedrock and incorporates fractured basalt and basalt weathering products into the overlying sediments, significantly altering their mineralogy and chemistry. The introduction of basaltic minerals into soils derived from a granitic parent should increase the concentrations of elements that are present in higher concentrations in basalt, such as iron. Iron is an important nutrient for grapevines and unlike most elements, the concentration of iron in grapes and vineyard soils has been demonstrated to be directly related (Negre, Cordonnier, 1953). Iron concentrations in musts, which vary according to soil iron content, have been shown to affect the stability, clarity, and color of wines (Riganakos, Veltsistas, 2003).

The incorporation of fractured basalt by ripping also significantly alters both the texture and color of soils derived from fine-grained light-colored loess. In some recently planted Columbia Valley vineyards, ripping has produced soils with a very high ratio of rock to loess, and basalt is exposed over a significant percentage of the ground surface. The physical properties of these basalt-rich soils are very different from the loess-dominated soils that are typical of most Columbia Valley AVA vineyards. Unlike the highly erodible loess-dominated soils, the rocky coarse-textured basalt-rich soils require no cover crop. Relative to a vegetated ground surface, bare soil rich in basalt should absorb, store, and radiate more heat and conduct heat to deeper levels of the soil more effectively (Gladstones, 1994; White, 2003). Winegrowers have long recognized the thermal properties of basalt. In Germany's Forst region, it has even been imported to warm vineyard soils (Clarke, 2002).


To test the influence of basalt on the chemistry of Columbia Valley AVA soils, samples

were collected from diverse sites that typify the range of its involvement (Fig.1). Soil samples were collected from: 1) ripped alluvial soils rich in basalt cobbles, 2) ripped alluvial soils with scattered basalt cobbles, 3) steep hillsides with thin loess and basalt colluvium (unripped), 4) steep hillsides with thin loess and basalt colluvium (ripped), and 5) gently sloping topography

4 - 50



INFLUENCE OF BASALT ON THE TERROIR OF THE COLUMBIA VALLEY AMERICAN VITICULTURAL AREA

K. R. Pogue

Department of Geology, Whitman College 345 Boyer Ave., Walla Walla, Washington 99362 USA

[email protected] ABSTRACT

The Columbia Valley American Viticultural Area (AVA) of the Pacific Northwest, USA is the world's largest officially recognized viticultural area with basalt bedrock. However, most Columbia Valley vineyards are planted in soils derived from thick loess and glacial flood sediments, rather than the underlying bedrock. Recently, vineyard plantings have expanded into parts of the AVA where basalt and basalt weathering products, derived either naturally or through mechanical ripping, are a major soil component. Tests were conducted to determine how the addition of a basalt component to soils could affect the terroir of Columbia Valley vineyards. To test for the chemical influence of basalt, samples were obtained from soils representative of the range of basalt influence and analyzed for iron content. Increases of 77% to 233% in available iron were observed in vineyards with basalt component relative to vineyards planted in grass-covered loess. To measure the thermal influence of basalt, temperature data loggers were installed within soils and grape clusters in basalt-covered and grass-covered vineyards. Temperature loggers in the basalt-covered vineyard recorded an 18% increase in average soil temperature at a depth of 5 cm, a 13% increase in average soil temperature at a depth of 25 cm, and a 4% in average cluster temperature relative to those in the grass-covered vineyard. Cluster temperatures in the basalt-covered vineyard were generally higher than in the grass-covered vineyard from late morning through early evening, equilibrating rapidly near sunset. KEYWORDS basalt - terroir - soil - Columbia Valley INTRODUCTION

Worldwide, the percentage of vineyards planted in basalt-derived soils is relatively small. Notable viticultural areas with soils developed in weakly weathered to unweathered basalt include the Canary Islands, the Azores, and Sicily's Mt. Etna. Regions that host vineyards planted in older or more deeply weathered basaltic soils include western India, southern Australia, Oregon's Willamette Valley, south-central France, northern Italy, and Hungary's Badascony region. The world's largest government-designated viticultural region with basalt-dominated bedrock is the Columbia Valley AVA, which encompasses over 45,000 km2 of the states of Washington and Oregon (Gregutt, 2002). The basalt was erupted during the Miocene Epoch from volcanoes associated with the hot spot that now lies beneath Yellowstone National Park in Wyoming (Pogue, 2009). The Columbia Valley AVA presently contains over 2700 ha of vineyards that are located primarily on gentle slopes or on valley floors below 400 m in elevation. Almost all of the vineyards are planted in loess derived from the

deflation of sediments deposited by a series of catastrophic Pleistocene glacial outburst floods, known as the Missoula floods (Busacca, Meinert, 2003). At elevations below 330 m, the loess commonly overlies sand, silt, and gravel deposited by the Missoula floods, while above this elevation, the loess directly overlies basalt bedrock. Despite its age, the basalt is not deeply weathered due to the combined affects of the region's arid climate and the protective mantle of flood- and wind-deposited sediment. The most obvious effects of weathering are fracture networks filled with calcium carbonate and iron oxide-stained clays within the uppermost 1 m of the basalt. The soils in most Columbia Valley AVA vineyards were derived by the glacial and fluvial erosion of regions dominated by granitic and metasedimentary bedrock that lie north and east of the Columbia Valley AVA. They are therefore rich in quartz, muscovite mica, and potassium feldspar, minerals not present in the underlying basalt bedrock. Since the thickness of these soils generally exceeds the rooting depth of the vines, basalt has had, until recently, almost no influence on terroir.

Over the last 10 years, viticulture in the Columbia Valley AVA has rapidly expanded. Vineyards have recently been planted at higher elevations, on steeper slopes, and in rocky, alluvial soils. The soils in these vineyards are commonly much thinner than those of the traditional valley floor sites, and therefore vine roots are able to directly interact with basalt bedrock or basalt-derived alluvium or colluvium. In preparation for planting, the thin soils are often mechanically ripped to a depth of 0.5 to 1 m to increase water holding capacity and available rooting depth. The ripping process crushes the upper parts of the weakly weathered basalt bedrock and incorporates fractured basalt and basalt weathering products into the overlying sediments, significantly altering their mineralogy and chemistry. The introduction of basaltic minerals into soils derived from a granitic parent should increase the concentrations of elements that are present in higher concentrations in basalt, such as iron. Iron is an important nutrient for grapevines and unlike most elements, the concentration of iron in grapes and vineyard soils has been demonstrated to be directly related (Negre, Cordonnier, 1953). Iron concentrations in musts, which vary according to soil iron content, have been shown to affect the stability, clarity, and color of wines (Riganakos, Veltsistas, 2003).

The incorporation of fractured basalt by ripping also significantly alters both the texture and color of soils derived from fine-grained light-colored loess. In some recently planted Columbia Valley vineyards, ripping has produced soils with a very high ratio of rock to loess, and basalt is exposed over a significant percentage of the ground surface. The physical properties of these basalt-rich soils are very different from the loess-dominated soils that are typical of most Columbia Valley AVA vineyards. Unlike the highly erodible loess-dominated soils, the rocky coarse-textured basalt-rich soils require no cover crop. Relative to a vegetated ground surface, bare soil rich in basalt should absorb, store, and radiate more heat and conduct heat to deeper levels of the soil more effectively (Gladstones, 1994; White, 2003). Winegrowers have long recognized the thermal properties of basalt. In Germany's Forst region, it has even been imported to warm vineyard soils (Clarke, 2002).


To test the influence of basalt on the chemistry of Columbia Valley AVA soils, samples

were collected from diverse sites that typify the range of its involvement (Fig.1). Soil samples were collected from: 1) ripped alluvial soils rich in basalt cobbles, 2) ripped alluvial soils with scattered basalt cobbles, 3) steep hillsides with thin loess and basalt colluvium (unripped), 4) steep hillsides with thin loess and basalt colluvium (ripped), and 5) gently sloping topography

4 - 51



with thick loess. In addition, as a control, an artificial soil sample (sample #6) was created by crushing unweathered basalt. All samples were sieved to <1mm particle size. The availability of iron in all samples was determined by a commercial soil laboratory using diethylenetriaminepentaacetic acid (DTPA) as an extractant.

Figure 1 Diagrammatic cross-section showing location of samples relative to soil type and slope.

To measure the thermal effects of a basalt-covered ground surface, temperature data loggers

were inserted into the interiors of grape clusters in two vineyards located 2.5 km apart. The surface of one vineyard is covered almost entirely by basalt cobbles while the other by a combination of dry, grassy vegetation and brown, loess-based soil. Data loggers were placed in 4 clusters in each vineyard. Clusters were selected to be approximately 0.5 m above the ground surface and shaded by leaves from direct sunlight. Data loggers were also buried midway between two rows in each vineyard at depths of 5 cm and 25 cm. The ambient air temperature in each vineyard was recorded by a radiation-shielded temperature data logger positioned 1.5 m above the ground. Data were collected at various times during July and August of 2007. RESULTS AND DISCUSSION

Fig. 2 shows the concentration of available iron in each sample in parts per million. As

expected, sample #5 from the thick loess soils showed the lowest concentration (9 ppm). Relative to the other soils, these soils contain virtually no basaltic component. The highest concentration of iron (30 ppm) was measured in the alluvial soil with scattered basalt cobbles (sample #2). Being farther from the main stream channel, this soil is older and more deeply weathered than its cobble-rich counterpart (sample #1), which had a 37 percent lower iron concentration (19 ppm). The sample of unripped thin loess-based soil from a steep hillside (sample #3) contained 16 ppm iron, reflecting minor colluvial input from the basalt bedrock. The 44% increase in available iron in sample #4 (23 ppm) relative to sample #3 is likely related to the incorporation of weathered basalt by mechanical ripping. Sample #6, the

artificial soil created by crushing unweathered basalt, had an iron concentration of only 14 ppm, which emphasizes the critical role of weathering in the production of plant-available iron.

Figure 2 Available iron in each sample.

Fig. 3 shows a graph of the average temperatures of grape clusters in the basalt-covered and grass-covered vineyards and a graph of the difference in ambient air temperature in the two vineyards from 5 days in late July 2007. The grape cluster data loggers in the basalt-covered vineyard recorded higher temperatures between approximately 10:00 and 20:00 each day.

4 - 52



with thick loess. In addition, as a control, an artificial soil sample (sample #6) was created by crushing unweathered basalt. All samples were sieved to <1mm particle size. The availability of iron in all samples was determined by a commercial soil laboratory using diethylenetriaminepentaacetic acid (DTPA) as an extractant.

Figure 1 Diagrammatic cross-section showing location of samples relative to soil type and slope.

To measure the thermal effects of a basalt-covered ground surface, temperature data loggers

were inserted into the interiors of grape clusters in two vineyards located 2.5 km apart. The surface of one vineyard is covered almost entirely by basalt cobbles while the other by a combination of dry, grassy vegetation and brown, loess-based soil. Data loggers were placed in 4 clusters in each vineyard. Clusters were selected to be approximately 0.5 m above the ground surface and shaded by leaves from direct sunlight. Data loggers were also buried midway between two rows in each vineyard at depths of 5 cm and 25 cm. The ambient air temperature in each vineyard was recorded by a radiation-shielded temperature data logger positioned 1.5 m above the ground. Data were collected at various times during July and August of 2007. RESULTS AND DISCUSSION

Fig. 2 shows the concentration of available iron in each sample in parts per million. As

expected, sample #5 from the thick loess soils showed the lowest concentration (9 ppm). Relative to the other soils, these soils contain virtually no basaltic component. The highest concentration of iron (30 ppm) was measured in the alluvial soil with scattered basalt cobbles (sample #2). Being farther from the main stream channel, this soil is older and more deeply weathered than its cobble-rich counterpart (sample #1), which had a 37 percent lower iron concentration (19 ppm). The sample of unripped thin loess-based soil from a steep hillside (sample #3) contained 16 ppm iron, reflecting minor colluvial input from the basalt bedrock. The 44% increase in available iron in sample #4 (23 ppm) relative to sample #3 is likely related to the incorporation of weathered basalt by mechanical ripping. Sample #6, the

artificial soil created by crushing unweathered basalt, had an iron concentration of only 14 ppm, which emphasizes the critical role of weathering in the production of plant-available iron.

Figure 2 Available iron in each sample.

Fig. 3 shows a graph of the average temperatures of grape clusters in the basalt-covered and grass-covered vineyards and a graph of the difference in ambient air temperature in the two vineyards from 5 days in late July 2007. The grape cluster data loggers in the basalt-covered vineyard recorded higher temperatures between approximately 10:00 and 20:00 each day.

4 - 53



Figure 3 Grape cluster temperatures and difference in ambient air temperature in basalt- and grass-covered vineyards.

These time intervals are shaded on fig. 3. Although the ambient air temperature in the vineyards during these intervals generally differed by less than 1°C, the basalt-covered vineyard cluster temperatures were often 3° to 5°C higher than those in the grass-covered vineyard.

Fig. 4 compares the average ambient air, ground surface, subsurface, and cluster temperatures of the grass-covered and basalt-covered vineyards from 25 days in August of 2007. The higher ground surface and subsurface temperatures in the basalt-covered vineyard reflect the lower specific heat and higher thermal conductivity of basalt relative to grass and loess. Radiant heat supplied by the higher surface temperatures increased average cluster temperatures in the basalt-covered vineyard relative to the grass-covered vineyard. The relatively smaller increase in the average temperature of the basalt vineyard clusters reflects the high specific heat of water, which constitutes most of their mass. Variations in berry temperature have been shown to affect the production of phenolics, anthocyanins, and sugars (Bergqvist et al., 2001). The average ambient air temperature in the two vineyards was statistically the same, indicating that advection overwhelms the effects of the differing surface materials.

Figure 4 Comparison of average temperatures in basalt- and grass-covered vineyards.

CONCLUSIONS

The terroir of some Columbia Valley AVA vineyards is significantly influenced by the chemical and thermal properties of basalt. Vineyard soils in the Columbia Valley AVA that incorporate basalt bedrock or basalt alluvium show substantial increases in available iron relative to the more widely planted loess-based soils. The largest increases are observed in older alluvial soils and in mechanically ripped soils that incorporate weathered bedrock. Since the iron content of grapevines is directly related to the availability of iron in vineyard soil, increased iron should also be evident in the grapes and wines produced from basalt-rich soils.

Vineyards within the Columbia Valley AVA covered by fractured basalt bedrock or basalt-rich alluvium have higher average ground surface and subsurface temperatures than their grass-covered counterparts. From late morning to early evening, grape clusters in basalt-covered vineyards are heated to higher temperatures than clusters in grass covered vineyards. The extra heat is derived from infrared radiation from the sun-warmed dark-colored basalt, not from conduction from heated air. Cluster temperatures within the basalt-covered and grass-covered vineyards rapidly equilibrate near sunset. No evidence was observed of the oft-cited ability of surface stones to store heat and release it after sunset, at least not to the above ground part of the grapevines. Due to advection, vineyard surface material appears to have little affect on ambient air temperature. ACKNOWLEDGEMENTS

Christophe Baron of Cayuse Vineyards and Norm McKibben of Seven Hills Vineyards kindly provided permission to use their properties for this research. Students Greg Dering and Karl Lang provide valuable assistance in the field. Instrumentation and analyses were funded by Whitman College and the Keck Geology Consortium. BIBLIOGRAPHY Bergqvist J., Dokoozlian N., Ebisuda N., 2001. Sunlight exposure and temperature effects on

berry growth and composition of Cabernet Sauvignon and Grenache in the central San Joaquin valley of California. American Journal of Enology and Viticulture, 52, 1-7.

Busacca A., Meinert L., 2003. Wine and geology—The terroir of Washington State. In:

Western Cordillera and Adjacent Areas. Swanson, T.W., Ed. Geological Society of America Field Guide 4. p. 69–85.

Clarke O., 2002. Oz Clarke's New Wine Atlas: Wine and Wine Regions of the World.

Webster's International Publishers. London. Gladstones J., 1992. Viticulture and Environment. Winetitles. Underdale. Gregutt P., 2002. Washington Wines and Wineries, the Essential Guide. University of

California Press. Berkeley. Negre E. and Cordonnier R., 1953. Les Origines du Fer des Vins. Progres Agricole et

Viticole, 139: 160-164. Pogue K., 2009. Folds, floods, and fine wine: Geologic influences on the terroir of the Columbia

basin. In: Volcanoes to Vineyards: Geologic Field Trips through the Dynamic Landscape of the Pacific Northwest. O'Connor, J., Dorsey, R., and Madin, I., Eds. Geological Society of America Field Guide 15. p. 1-17.

White R., 2003. Soils for Fine Wines. Oxford University Press. Oxford.

4 - 54



Figure 3 Grape cluster temperatures and difference in ambient air temperature in basalt- and grass-covered vineyards.

These time intervals are shaded on fig. 3. Although the ambient air temperature in the vineyards during these intervals generally differed by less than 1°C, the basalt-covered vineyard cluster temperatures were often 3° to 5°C higher than those in the grass-covered vineyard.

Fig. 4 compares the average ambient air, ground surface, subsurface, and cluster temperatures of the grass-covered and basalt-covered vineyards from 25 days in August of 2007. The higher ground surface and subsurface temperatures in the basalt-covered vineyard reflect the lower specific heat and higher thermal conductivity of basalt relative to grass and loess. Radiant heat supplied by the higher surface temperatures increased average cluster temperatures in the basalt-covered vineyard relative to the grass-covered vineyard. The relatively smaller increase in the average temperature of the basalt vineyard clusters reflects the high specific heat of water, which constitutes most of their mass. Variations in berry temperature have been shown to affect the production of phenolics, anthocyanins, and sugars (Bergqvist et al., 2001). The average ambient air temperature in the two vineyards was statistically the same, indicating that advection overwhelms the effects of the differing surface materials.

Figure 4 Comparison of average temperatures in basalt- and grass-covered vineyards.

CONCLUSIONS

The terroir of some Columbia Valley AVA vineyards is significantly influenced by the chemical and thermal properties of basalt. Vineyard soils in the Columbia Valley AVA that incorporate basalt bedrock or basalt alluvium show substantial increases in available iron relative to the more widely planted loess-based soils. The largest increases are observed in older alluvial soils and in mechanically ripped soils that incorporate weathered bedrock. Since the iron content of grapevines is directly related to the availability of iron in vineyard soil, increased iron should also be evident in the grapes and wines produced from basalt-rich soils.

Vineyards within the Columbia Valley AVA covered by fractured basalt bedrock or basalt-rich alluvium have higher average ground surface and subsurface temperatures than their grass-covered counterparts. From late morning to early evening, grape clusters in basalt-covered vineyards are heated to higher temperatures than clusters in grass covered vineyards. The extra heat is derived from infrared radiation from the sun-warmed dark-colored basalt, not from conduction from heated air. Cluster temperatures within the basalt-covered and grass-covered vineyards rapidly equilibrate near sunset. No evidence was observed of the oft-cited ability of surface stones to store heat and release it after sunset, at least not to the above ground part of the grapevines. Due to advection, vineyard surface material appears to have little affect on ambient air temperature. ACKNOWLEDGEMENTS

Christophe Baron of Cayuse Vineyards and Norm McKibben of Seven Hills Vineyards kindly provided permission to use their properties for this research. Students Greg Dering and Karl Lang provide valuable assistance in the field. Instrumentation and analyses were funded by Whitman College and the Keck Geology Consortium. BIBLIOGRAPHY Bergqvist J., Dokoozlian N., Ebisuda N., 2001. Sunlight exposure and temperature effects on

berry growth and composition of Cabernet Sauvignon and Grenache in the central San Joaquin valley of California. American Journal of Enology and Viticulture, 52, 1-7.

Busacca A., Meinert L., 2003. Wine and geology—The terroir of Washington State. In:

Western Cordillera and Adjacent Areas. Swanson, T.W., Ed. Geological Society of America Field Guide 4. p. 69–85.

Clarke O., 2002. Oz Clarke's New Wine Atlas: Wine and Wine Regions of the World.

Webster's International Publishers. London. Gladstones J., 1992. Viticulture and Environment. Winetitles. Underdale. Gregutt P., 2002. Washington Wines and Wineries, the Essential Guide. University of

California Press. Berkeley. Negre E. and Cordonnier R., 1953. Les Origines du Fer des Vins. Progres Agricole et

Viticole, 139: 160-164. Pogue K., 2009. Folds, floods, and fine wine: Geologic influences on the terroir of the Columbia

basin. In: Volcanoes to Vineyards: Geologic Field Trips through the Dynamic Landscape of the Pacific Northwest. O'Connor, J., Dorsey, R., and Madin, I., Eds. Geological Society of America Field Guide 15. p. 1-17.

White R., 2003. Soils for Fine Wines. Oxford University Press. Oxford.

4 - 55



ARSENIC IN BERRIES AND ITS CORRELATION WITH NATURAL SOIL CONTENT: EXPERIENCE IN TRENTINO (ITALY)

D. Bertoldi(1,2), R. Larcher(1), G. Nicolini(1), M. Bertamini(1), G. Concheri(2)(1) IASMA – Fondazione E. Mach, via Mach 1, 38010 San Michele all’Adige (TN), Italy

[email protected](2) Università di Padove, Dip. Biotecnologie Agrarie, viale dell’Università, 16, 35020 Legnaro (PD), Italy

ABSTRACT Il lavoro presenta l’evoluzione dei contenuti di arsenico nelle uve durante lo sviluppo e la

maturazione, e la sua distribuzione nell’acino; verifica inoltre la relazione tra i contenuti di As nelle uve, nelle foglie e nei suoli caratterizzati da una dotazione differente e naturale di questo elemento.

Nella bacca l’arsenico cresce durante la stagione vegetativa e a maturazione è localizzato nella polpa (50%), nella buccia (40%) e in minima parte nei semi.

La correlazione tra i contenuti di As nelle bacche raccolte in 18 vigneti, nelle corrispondenti foglie e nei rispettivi suoli estratti con acetato di ammonio risulta statisticamente significativa.

KEYWORDSarsenico – arsenico biodisponibile – suolo – Vitis – acino – ICP-MS

ABSTRACT The work illustrates arsenic content in grapes during development and ripening and its

distribution in the berry, together with the relationship between As content in grape berries, leaves and soils where this element is naturally present in different amounts.

Arsenic increases in the berry during the growing season and is located in the pulp (50%), the skin (40%) and to a lesser extent in the seeds in ripe berries.

The correlation between the As content in berries collected in 18 vineyards and in the corresponding leaves and soils, extracted using ammonium acetate, is statistically significant.

KEYWORDSarsenic – bioavailable arsenic – soil – Vitis – grape berry – ICP-MS

INTRODUZIONE L’arsenico (As) è naturalmente presente nella crosta terrestre, diffuso in vari tipi di rocce e

soprattutto nei sedimenti argillosi e negli scisti. È un elemento calcofilo spesso associato a solfuri minerali soprattutto di Fe, Cu e Pb quali pirite, calcopirite e galena ed ha un comportamento geochimico simile al P (elemento antagonista). È presente nell’ambiente sia in forme inorganiche che organiche, interconvertibili tra loro (attraverso ad es. processi di metilazione) a seguito di processi biotici e abiotici.

Nei suoli non contaminati, i livelli di As variano tra <0.1 e 95 mg/kg con medie mondiali in genere inferiori a 10 mg/kg (Adriano, 2001; Kabata-Pendias, 2001). La solubilità di questo elemento - e conseguentemente la sua concentrazione nella soluzione del suolo - è influenzata principalmente dalle condizioni redox, dal pH, dalla presenza di sostanza organica e dall’attività biologica (Jones et al., 2000). Nella soluzione del suolo l’As è presente con stato di ossidazione (III) o (V). L’As(III) è più solubile, mobile e tossico

dell’As(V) che risulta essere comunque la specie fitodisponibile dominante nei suoli aerobici (Adriano, 2001).

All’As di origine naturale, geogenica, si può aggiungere quello di derivazione antropica. Tale elemento può essere infatti emesso nell’ambiente in seguito all’attività industriale e mineraria, alla combustione del carbone e dalle centrali geotermiche ed essere inoltre presente come tracce nei concimi fosfatici e in molti fitofarmaci e defoglianti. Conseguentemente, concentrazioni relativamente elevate possono talvolta essere ritrovate nel suolo e destare preoccupazione in riferimento ad un possibile assorbimento da parte delle piante con successivo trasferimento attraverso la catena alimentare fino all’uomo, con rischi di cancerogenicità. L’International Agency for Research on Cancer (IARC) ha infatti classificato l’As ed i suoi composti inorganci come “cancerogeni per l’uomo” (gruppo 1) e i suoi composti organici metabolizzati dall’uomo (acido monometilarsonico e acido dimetilarsinico) come “possibili cancerogeni per l’uomo” (gruppo 2B).

Benchè i minerali e i composti dell’As siano piuttosto solubili, tuttavia, la maggior parte dell’As nei suoli non risulta molto mobile né biodisponibile poichè generalmente fortemente associato a ossidi cristallini o amorfi di Fe e Al, nei terreni acidi, e al Ca in quelli basici (Wenzel et al., 2001; Kabata-Pendias, 2001).

In condizioni aerobiche il desorbimento di As(V) dal suolo aumenta all’aumentare del pH a causa dell’aumento delle cariche negative a livello dei colloidi del suolo (Smith et al., 1999; Adriano, 2001). Tuttavia, un calo del pH, per esempio a livello della rizosfera, può dissolvere gli ossidi e idrossidi di Fe e Al con la conseguente co-dissoluzione dell’As legato a questa frazione (Fitz, Wenzel, 2002). Al contrario, l’adsorbimento di As(III) aumenta con il pH.

Nonostante sia già stato osservato come l’As in genere si accumuli nelle radici e nelle foglie più vecchie (Kabata-Pendias, 2001), tuttavia informazioni più dettagliate sulla sua distribuzione in particolare nel caso della vite sono carenti.

Lo scopo del lavoro è stato quello di verificare l’esistenza di una correlazione tra i contenuti di As nei suoli e nelle uve al fine di valutare la pericolosità dell’utilizzo/consumo di uve prodotte in aree geologiche naturalmente ricche di questo elemento. Si è voluto inoltre aumentare le conoscenze circa l’accumulo e localizzazione dell’As nell’acino d’uva studiandone la variazione dei contenuti nella bacca durante lo sviluppo e la maturazione e la distribuzione nelle diverse porzioni della bacca stessa.

MATERIALI E METODI Evoluzione dei contenuti di As nella bacca.Per lo studio dell’evoluzione dei contenuti di As nella bacca intera durante lo sviluppo e la

maturazione, sono state considerate uve cv. Chardonnay (cl. 95, p.i. 3309), prelevate dall’allegagione (40 giorni prima dell’invaiatura) alla surmaturazione (60 gg dopo l’invaiatura) in 2 vigneti trentini localizzati a San Michele all’Adige (SM, 289 m s.l.m.) e Faedo-Maso Togn (MT, 723 m s.l.m.). Entrambi i vigneti insistevano su un suolo di tipo alcalino (pH 8.11 e 7.99 rispettivamente) benché con diversa dotazione di calcare totale (499 g/kg e 222 g/kg rispettivamente). Da ciascun vigneto sono stati raccolti a cadenza indicativamente settimanale, per un totale di 11-12 momenti, 4 campioni da 100 bacche ciascuno. Le bacche sono state lavate con acido nitrico 1%, risciacquate con acqua ultrapura, omogeneizzate, mineralizzate con acido nitrico concentrato in vaso chiuso (Bertoldi et al.,2009) e analizzate mediante uno spettrometro di massa con sorgente al plasma accoppiata induttivamente (ICP-MS) dotato di cella di collisione (gas He) per la rimozione di interferenti, utilizzando lo Sc come standard interno.

4 - 56



ARSENIC IN BERRIES AND ITS CORRELATION WITH NATURAL SOIL CONTENT: EXPERIENCE IN TRENTINO (ITALY)

D. Bertoldi(1,2), R. Larcher(1), G. Nicolini(1), M. Bertamini(1), G. Concheri(2)(1) IASMA – Fondazione E. Mach, via Mach 1, 38010 San Michele all’Adige (TN), Italy

[email protected](2) Università di Padove, Dip. Biotecnologie Agrarie, viale dell’Università, 16, 35020 Legnaro (PD), Italy

ABSTRACT Il lavoro presenta l’evoluzione dei contenuti di arsenico nelle uve durante lo sviluppo e la

maturazione, e la sua distribuzione nell’acino; verifica inoltre la relazione tra i contenuti di As nelle uve, nelle foglie e nei suoli caratterizzati da una dotazione differente e naturale di questo elemento.

Nella bacca l’arsenico cresce durante la stagione vegetativa e a maturazione è localizzato nella polpa (50%), nella buccia (40%) e in minima parte nei semi.

La correlazione tra i contenuti di As nelle bacche raccolte in 18 vigneti, nelle corrispondenti foglie e nei rispettivi suoli estratti con acetato di ammonio risulta statisticamente significativa.

KEYWORDSarsenico – arsenico biodisponibile – suolo – Vitis – acino – ICP-MS

ABSTRACT The work illustrates arsenic content in grapes during development and ripening and its

distribution in the berry, together with the relationship between As content in grape berries, leaves and soils where this element is naturally present in different amounts.

Arsenic increases in the berry during the growing season and is located in the pulp (50%), the skin (40%) and to a lesser extent in the seeds in ripe berries.

The correlation between the As content in berries collected in 18 vineyards and in the corresponding leaves and soils, extracted using ammonium acetate, is statistically significant.

KEYWORDSarsenic – bioavailable arsenic – soil – Vitis – grape berry – ICP-MS

INTRODUZIONE L’arsenico (As) è naturalmente presente nella crosta terrestre, diffuso in vari tipi di rocce e

soprattutto nei sedimenti argillosi e negli scisti. È un elemento calcofilo spesso associato a solfuri minerali soprattutto di Fe, Cu e Pb quali pirite, calcopirite e galena ed ha un comportamento geochimico simile al P (elemento antagonista). È presente nell’ambiente sia in forme inorganiche che organiche, interconvertibili tra loro (attraverso ad es. processi di metilazione) a seguito di processi biotici e abiotici.

Nei suoli non contaminati, i livelli di As variano tra <0.1 e 95 mg/kg con medie mondiali in genere inferiori a 10 mg/kg (Adriano, 2001; Kabata-Pendias, 2001). La solubilità di questo elemento - e conseguentemente la sua concentrazione nella soluzione del suolo - è influenzata principalmente dalle condizioni redox, dal pH, dalla presenza di sostanza organica e dall’attività biologica (Jones et al., 2000). Nella soluzione del suolo l’As è presente con stato di ossidazione (III) o (V). L’As(III) è più solubile, mobile e tossico

dell’As(V) che risulta essere comunque la specie fitodisponibile dominante nei suoli aerobici (Adriano, 2001).

All’As di origine naturale, geogenica, si può aggiungere quello di derivazione antropica. Tale elemento può essere infatti emesso nell’ambiente in seguito all’attività industriale e mineraria, alla combustione del carbone e dalle centrali geotermiche ed essere inoltre presente come tracce nei concimi fosfatici e in molti fitofarmaci e defoglianti. Conseguentemente, concentrazioni relativamente elevate possono talvolta essere ritrovate nel suolo e destare preoccupazione in riferimento ad un possibile assorbimento da parte delle piante con successivo trasferimento attraverso la catena alimentare fino all’uomo, con rischi di cancerogenicità. L’International Agency for Research on Cancer (IARC) ha infatti classificato l’As ed i suoi composti inorganci come “cancerogeni per l’uomo” (gruppo 1) e i suoi composti organici metabolizzati dall’uomo (acido monometilarsonico e acido dimetilarsinico) come “possibili cancerogeni per l’uomo” (gruppo 2B).

Benchè i minerali e i composti dell’As siano piuttosto solubili, tuttavia, la maggior parte dell’As nei suoli non risulta molto mobile né biodisponibile poichè generalmente fortemente associato a ossidi cristallini o amorfi di Fe e Al, nei terreni acidi, e al Ca in quelli basici (Wenzel et al., 2001; Kabata-Pendias, 2001).

In condizioni aerobiche il desorbimento di As(V) dal suolo aumenta all’aumentare del pH a causa dell’aumento delle cariche negative a livello dei colloidi del suolo (Smith et al., 1999; Adriano, 2001). Tuttavia, un calo del pH, per esempio a livello della rizosfera, può dissolvere gli ossidi e idrossidi di Fe e Al con la conseguente co-dissoluzione dell’As legato a questa frazione (Fitz, Wenzel, 2002). Al contrario, l’adsorbimento di As(III) aumenta con il pH.

Nonostante sia già stato osservato come l’As in genere si accumuli nelle radici e nelle foglie più vecchie (Kabata-Pendias, 2001), tuttavia informazioni più dettagliate sulla sua distribuzione in particolare nel caso della vite sono carenti.

Lo scopo del lavoro è stato quello di verificare l’esistenza di una correlazione tra i contenuti di As nei suoli e nelle uve al fine di valutare la pericolosità dell’utilizzo/consumo di uve prodotte in aree geologiche naturalmente ricche di questo elemento. Si è voluto inoltre aumentare le conoscenze circa l’accumulo e localizzazione dell’As nell’acino d’uva studiandone la variazione dei contenuti nella bacca durante lo sviluppo e la maturazione e la distribuzione nelle diverse porzioni della bacca stessa.

MATERIALI E METODI Evoluzione dei contenuti di As nella bacca.Per lo studio dell’evoluzione dei contenuti di As nella bacca intera durante lo sviluppo e la

maturazione, sono state considerate uve cv. Chardonnay (cl. 95, p.i. 3309), prelevate dall’allegagione (40 giorni prima dell’invaiatura) alla surmaturazione (60 gg dopo l’invaiatura) in 2 vigneti trentini localizzati a San Michele all’Adige (SM, 289 m s.l.m.) e Faedo-Maso Togn (MT, 723 m s.l.m.). Entrambi i vigneti insistevano su un suolo di tipo alcalino (pH 8.11 e 7.99 rispettivamente) benché con diversa dotazione di calcare totale (499 g/kg e 222 g/kg rispettivamente). Da ciascun vigneto sono stati raccolti a cadenza indicativamente settimanale, per un totale di 11-12 momenti, 4 campioni da 100 bacche ciascuno. Le bacche sono state lavate con acido nitrico 1%, risciacquate con acqua ultrapura, omogeneizzate, mineralizzate con acido nitrico concentrato in vaso chiuso (Bertoldi et al.,2009) e analizzate mediante uno spettrometro di massa con sorgente al plasma accoppiata induttivamente (ICP-MS) dotato di cella di collisione (gas He) per la rimozione di interferenti, utilizzando lo Sc come standard interno.

4 - 57



Distribuzione dell’As nell’acino.Per lo studio della distribuzione dell’As nelle diverse frazioni dell’acino, da ciascun

appezzamento sono stati raccolti settimanalmente ulteriori 4 campioni da 100 bacche. Tale campionamento è stato effettuato nell’intervallo di composizione delle uve tra i 20 e i 24° Brix (corrispondente ad un periodo di circa 4 settimane) in modo da corrispondere a diversi livelli di maturazione tecnologica. Queste bacche sono state lavate e divise in modo da ottenere separatamente campioni di bucce e semi. Le 2 frazioni sono state omogeneizzate, mineralizzate e analizzate mediante ICP-MS. Il contenuto di As nella polpa è calcolato per differenza a partire dal contenuto nella bacca intera a cui sono sottratti i contributi di semi e buccia.

Correlazione tra i contenuti di As di suoli, bacche e foglie.La correlazione tra i contenuti di As nelle bacche, nelle foglie e nei rispettivi suoli è stata

valutata su campioni raccolti in 11 diversi vigneti localizzati in provincia di Trento, caratterizzati da condizioni pedo-climatiche e geologiche differenti (pH del suolo variabile tra 5.6 e 8.1, calcare totale variabile tra 5 e 700 g/kg). In 7 di questi vigneti i campionamenti sono stati effettuati anche in un secondo anno.

Presso questi vigneti, in prossimità della vendemmia, sono stati quindi raccolti campioni di bacche (con le stesse modalità sopra descritte), foglie (quarta foglia dopo il secondo grappolo), e suoli (fino ad una profondità di 60 cm). Le foglie sono state lavate con acido citrico 0.2%, seccate a 40°C, macinate, mineralizzate con acido nitrico concentrato e acqua ossigenata in vaso chiuso e analizzate mediante ICP-MS. I suoli sono stati seccati a temperatura ambiente e setacciati a 2 mm. E’ stato quantificato sia l’As “pseudototale” dissolvibile in acqua regia dopo mineralizzazione in microonde (ISO 11466/1995), sia l’As estraibile in acetato di ammonio 1M a pH 7 (SSIR, 2004). Il termine “pseudototale” si riferisce alla non completa dissoluzione della frazione silicatica del suolo da parte dell’acqua regia (Page et al., 1982).

Elaborazioni statistiche L’elaborazione dei dati è stata effettuata mediante le procedure statistiche del pacchetto

software STATISTICA 8.0 (StatSoft, 2008).

RISULTATI E DISCUSSIONE In entrambi i vigneti, i contenuti di As (espressi per bacca) aumentano nell’acino durante lo

sviluppo e la maturazione (Fig. 1).

0

0.03

0.06

0.09

0.12

-45 -30 -15 0 15 30 45 60

gg dall'inizio invaiatura

µµ µµg/

100

bacc

he

vigneto SMvigneto MT

Fig. 1. Variazione dei contenuti (media ± err. std., N=4 per ogni punto) di As nella bacca durante la stagione nei 2 vigneti

A livello dell’acino, e considerando le medie geometriche, la maggior parte dell’As è localizzato nella polpa (50% rispetto al contenuto totale nella bacca) e nella buccia (40%), con contenuti percentuali non statisticamente differenti (test ANOVA, p<0.05, N=40) mentre il restante 10% è localizzato nei semi (Fig. 2).

polpa buccia semi0

20

40

60

80

100

%

Mediana 25%-75% Interv. Non-Outlier

Fig. 2. Distribuzione statistica del contenuto percentuale di As nelle diverse porzioni dell’acino (N=40)

Considerando tutti i campioni analizzati nei 2 anni, il contenuto di As nelle bacche varia tra <0.09 e 6.23 µg/kg peso secco (p.s.) mentre il contenuto nelle foglie varia tra 16.3 e 205.9 µg/kg p.s.. Tali valori sono decisamente o tendenzialmente inferiori a quelli riportati da Ko etal (2007), pari rispettivamente a 70 µg/kg p.s. per le bacche e 60-410 µg/kg p.s. per le foglie. Sono invece simili a quanto osservato da Fang et al (2010) in uva passa cinese, che riportano livelli sempre inferiori al loro limite di rilevamento (7 µg/kg).

Il contenuto “pseudototale” di As naturalmente presente nei corrispondenti suoli varia tra 5.67 e 76.7 mg/kg. In un caso il contenuto supera il valore di 50 mg/kg indicato come limite dal DM 471/99 per le aree ad uso commerciale/industriale. Si tratta di un vigneto posizionato in un’area geologicamente ricca di As per la presenza di pirite (FeS2) e calcopirite (CuFeS2)che contengono, come noto, significative quantità di altri elementi tra cui soprattutto As, Zn e Pb. Un ulteriore campione supera il valore di 20 mg/kg indicato come limite per le aree ad uso verde pubblico, privato e residenziale.

Il rapporto percentuale tra il contenuto estratto in acetato di ammonio e il contenuto “pseudototale” dissolvibile in acqua regia varia tra 0.13 e 2.55%, con valori tendenzialmente più bassi nei vigneti con suolo più acido e meno calcareo (dati non mostrati). Orescanin et al(2003) hanno riscontrato in suoli vitati croati una frazione di As scambiabile (estratta in acetato di ammonio) pari al 5-10% del contenuto totale, percentuali più che doppie rispetto al quelle da noi quantificate. Wenzel et al (2001) riportano invece contenuti percentuali di As inferiori al 3.8% del totale nella frazione “non-specifically sorbed” estratta con (NH4)2SO40.05M.

Il test di Pearson rileva l’esistenza di una correlazione significativa (r = 0.77; p<0.05) tra i contenuti di As nelle bacche e nelle foglie prelevate nello stesso vigneto e stagione (Fig. 3).

Una buona correlazione è stata osservata anche tra i contenuti rilevati nelle bacche e quelli del suolo estratto con acetato di ammonio 1 M (r = 0.76; p<0.05; Fig. 3), considerabili come i contenuti biodisponibili quindi più facilmente assorbibili dalla pianta. Minore e non statisticamente significativa risulta invece la correlazione (r = 0.42) tra i contenuti nelle bacche e quelli del suolo dissolvibili in acqua regia.

4 - 58



Distribuzione dell’As nell’acino.Per lo studio della distribuzione dell’As nelle diverse frazioni dell’acino, da ciascun

appezzamento sono stati raccolti settimanalmente ulteriori 4 campioni da 100 bacche. Tale campionamento è stato effettuato nell’intervallo di composizione delle uve tra i 20 e i 24° Brix (corrispondente ad un periodo di circa 4 settimane) in modo da corrispondere a diversi livelli di maturazione tecnologica. Queste bacche sono state lavate e divise in modo da ottenere separatamente campioni di bucce e semi. Le 2 frazioni sono state omogeneizzate, mineralizzate e analizzate mediante ICP-MS. Il contenuto di As nella polpa è calcolato per differenza a partire dal contenuto nella bacca intera a cui sono sottratti i contributi di semi e buccia.

Correlazione tra i contenuti di As di suoli, bacche e foglie.La correlazione tra i contenuti di As nelle bacche, nelle foglie e nei rispettivi suoli è stata

valutata su campioni raccolti in 11 diversi vigneti localizzati in provincia di Trento, caratterizzati da condizioni pedo-climatiche e geologiche differenti (pH del suolo variabile tra 5.6 e 8.1, calcare totale variabile tra 5 e 700 g/kg). In 7 di questi vigneti i campionamenti sono stati effettuati anche in un secondo anno.

Presso questi vigneti, in prossimità della vendemmia, sono stati quindi raccolti campioni di bacche (con le stesse modalità sopra descritte), foglie (quarta foglia dopo il secondo grappolo), e suoli (fino ad una profondità di 60 cm). Le foglie sono state lavate con acido citrico 0.2%, seccate a 40°C, macinate, mineralizzate con acido nitrico concentrato e acqua ossigenata in vaso chiuso e analizzate mediante ICP-MS. I suoli sono stati seccati a temperatura ambiente e setacciati a 2 mm. E’ stato quantificato sia l’As “pseudototale” dissolvibile in acqua regia dopo mineralizzazione in microonde (ISO 11466/1995), sia l’As estraibile in acetato di ammonio 1M a pH 7 (SSIR, 2004). Il termine “pseudototale” si riferisce alla non completa dissoluzione della frazione silicatica del suolo da parte dell’acqua regia (Page et al., 1982).

Elaborazioni statistiche L’elaborazione dei dati è stata effettuata mediante le procedure statistiche del pacchetto

software STATISTICA 8.0 (StatSoft, 2008).

RISULTATI E DISCUSSIONE In entrambi i vigneti, i contenuti di As (espressi per bacca) aumentano nell’acino durante lo

sviluppo e la maturazione (Fig. 1).

0

0.03

0.06

0.09

0.12

-45 -30 -15 0 15 30 45 60

gg dall'inizio invaiatura

µµ µµg/

100

bacc

he

vigneto SMvigneto MT

Fig. 1. Variazione dei contenuti (media ± err. std., N=4 per ogni punto) di As nella bacca durante la stagione nei 2 vigneti

A livello dell’acino, e considerando le medie geometriche, la maggior parte dell’As è localizzato nella polpa (50% rispetto al contenuto totale nella bacca) e nella buccia (40%), con contenuti percentuali non statisticamente differenti (test ANOVA, p<0.05, N=40) mentre il restante 10% è localizzato nei semi (Fig. 2).

polpa buccia semi0

20

40

60

80

100

%

Mediana 25%-75% Interv. Non-Outlier

Fig. 2. Distribuzione statistica del contenuto percentuale di As nelle diverse porzioni dell’acino (N=40)

Considerando tutti i campioni analizzati nei 2 anni, il contenuto di As nelle bacche varia tra <0.09 e 6.23 µg/kg peso secco (p.s.) mentre il contenuto nelle foglie varia tra 16.3 e 205.9 µg/kg p.s.. Tali valori sono decisamente o tendenzialmente inferiori a quelli riportati da Ko etal (2007), pari rispettivamente a 70 µg/kg p.s. per le bacche e 60-410 µg/kg p.s. per le foglie. Sono invece simili a quanto osservato da Fang et al (2010) in uva passa cinese, che riportano livelli sempre inferiori al loro limite di rilevamento (7 µg/kg).

Il contenuto “pseudototale” di As naturalmente presente nei corrispondenti suoli varia tra 5.67 e 76.7 mg/kg. In un caso il contenuto supera il valore di 50 mg/kg indicato come limite dal DM 471/99 per le aree ad uso commerciale/industriale. Si tratta di un vigneto posizionato in un’area geologicamente ricca di As per la presenza di pirite (FeS2) e calcopirite (CuFeS2)che contengono, come noto, significative quantità di altri elementi tra cui soprattutto As, Zn e Pb. Un ulteriore campione supera il valore di 20 mg/kg indicato come limite per le aree ad uso verde pubblico, privato e residenziale.

Il rapporto percentuale tra il contenuto estratto in acetato di ammonio e il contenuto “pseudototale” dissolvibile in acqua regia varia tra 0.13 e 2.55%, con valori tendenzialmente più bassi nei vigneti con suolo più acido e meno calcareo (dati non mostrati). Orescanin et al(2003) hanno riscontrato in suoli vitati croati una frazione di As scambiabile (estratta in acetato di ammonio) pari al 5-10% del contenuto totale, percentuali più che doppie rispetto al quelle da noi quantificate. Wenzel et al (2001) riportano invece contenuti percentuali di As inferiori al 3.8% del totale nella frazione “non-specifically sorbed” estratta con (NH4)2SO40.05M.

Il test di Pearson rileva l’esistenza di una correlazione significativa (r = 0.77; p<0.05) tra i contenuti di As nelle bacche e nelle foglie prelevate nello stesso vigneto e stagione (Fig. 3).

Una buona correlazione è stata osservata anche tra i contenuti rilevati nelle bacche e quelli del suolo estratto con acetato di ammonio 1 M (r = 0.76; p<0.05; Fig. 3), considerabili come i contenuti biodisponibili quindi più facilmente assorbibili dalla pianta. Minore e non statisticamente significativa risulta invece la correlazione (r = 0.42) tra i contenuti nelle bacche e quelli del suolo dissolvibili in acqua regia.

4 - 59



2 3 4 5 6 7 8 9 10 11 12

ln[suolo] o ln[foglia]

-4

-3

-2

-1

0

1

2

3

ln[b

acca

]

suolo (acqua regia) suolo (ammonio acetato) foglie

Fig. 3. Correlazione tra i contenuti (trasformati in logaritmo) di As nelle bacche, nelle foglie e nel suolo estratto con acqua regia o con acetato di ammonio 1M.

CONCLUSIONI Il presente lavoro ha messo in luce la dinamica dell’aumento dei contenuti di As nelle

bacche durante lo sviluppo e la maturazione, nonché la localizzazione largamente preferenziale di tale elemento a livello di buccia e polpa. Ha messo in evidenza inoltre l’esistenza di una buona correlazione dei contenuti di As nelle bacche mature con quelli nei suoli estratti con acetato di ammonio, oltre che con quelli presenti nelle foglie.

Viti cresciute sui suoli maggiormente dotati di As (es. particolari aree geologiche) presentano uve con contenuti più elevati di questo elemento. Tuttavia, nel piano sperimentale non sono state riscontrate quantità di As nelle uve tali da destare alcuna preoccupazione per la salute umana, essendo stati rilevati contenuti massimi pari a 6.23 µg/kg p.s. a fronte di una dose considerata tossica di 5-50 mg/giorno per gli adulti. I contenuti presenti in un kg di uva (peso secco) corrispondono invece al 25-50% della dose giornaliera di As richiesta per un uomo adulto mentre la quantità di As ingerita quotidianamente attraverso la dieta varia generalmente tra 0.04 e 1.4 mg (Pais, Jones, 1997).

RINGRAZIAMENTIIl presente lavoro è stato svolto nell’ambito di un dottorato di ricerca in Viticoltura,

Enologia e Marketing delle Imprese Vitivinicole (Università degli Studi di Padova). Gli autori ringraziano il personale tecnico dell’Unità Viticoltura della Fondazione Mach – Istituto Agrario San Michele all’Adige per l’aiuto nella fase di campionamento.

BIBLIOGRAFIAAdriano D.C., 2001. Arsenic. In: Trace elements in terrestrial environments.

Biogeochemistry, bioavailability, and risks of metals. Adriano D.C., Ed. II edizione. New York: Springer-Verlag. n: 220-261

Bertoldi D., Larcher R., Nicolini G., Bertamini M., Concheri G., 2009. Distribution of rare earth elements in Vitis vinifera L. ‘Chardonnay’ berries. Vitis, 48(1): 49-51.

Fang Y.L., Zhang A., Wang H., Li H., Zhang Z.W., Chen S.X., Luan L.Y., 2010. Health risk assessment of trace elements in Chinese raisins produced in Xinjiang province. Food Control, 21: 732-739.

Fitz W.J., Wenzel W.W., 2002. Arsenic trasformation in the soil-rhizosphere-plant system: fundamentals and potential application to phytoremediation. Journal of Biotechnology, 99: 259-278.

ISO 11466, 1995. Soil quality -Extraction of trace elements soluble in aqua regia. Jones C.A., Langer H.W., Anderson K., McDermott T.R., Inskeep W.P., 2000. Rates of

microbially mediated arsenate reduction and solubilization. Soil Science Society of America Journal, 64: 600-608.

Kabata-Pendias A., 2001. Elements of group V. In: Trace elements in soils and plants. Kabata-Pendias A., Ed. III edizione. Boca Raton: CRC Press. n: 225-232

Ko B.-G., Vogeler I., Bolan N. S., Clothier B., Green S., Kennedy J., 2007. Mobility of copper, chromium and arsenic from treated timber into grapevines. Science of the Total Environment, 388: 35-42.

Orescanin V., Katunar A., Kutle A., Valkovic V., 2003. Heavy metal in soil, grape, and wine. Journal of Trace and Microprobe Techniques, 21(1): 171-180.

Page A.L., Miller R.H., Kenney D.R., 1982. Methods of Soil Analysis. Part 2. Chemical and Microbiological Properties. 2° edizione. Madison: American Society of Agronomy, Soil Science Society of America.

Pais I., Jones J.B., 1997. Trace elements. In :The Handbook of trace elements. Pais I., Jones J.B, Eds. Boca Raton: St. Lucie Press. n: 86

Smith E., Naidu R., Alston A.M., 1999. Chemistry of arsenic in soils: I. Sorption of arsenate and arsenite by four Australian soils. Journal of Environmental Quality, 28: 1719-1726.

SSIR (2004). Soil Survey Laboratory Methods Manual, Soil Survey Investigation Report n° 42, metodo 5A8.

StatSoft Inc., 2008. STATISTICA (data analysis software system), version 8.0. Tulsa, OK, 74104, USA

Wenzel W.W., Kirchbaumer N., Prohaska T., Stingeder G., Lombi E., Adriano D.C., 2001. Arsenic fractionation in soils using an improved sequential extraction procedure. Analytica Chimica Acta, 436: 309-323.

4 - 60



2 3 4 5 6 7 8 9 10 11 12

ln[suolo] o ln[foglia]

-4

-3

-2

-1

0

1

2

3

ln[b

acca

]

suolo (acqua regia) suolo (ammonio acetato) foglie

Fig. 3. Correlazione tra i contenuti (trasformati in logaritmo) di As nelle bacche, nelle foglie e nel suolo estratto con acqua regia o con acetato di ammonio 1M.

CONCLUSIONI Il presente lavoro ha messo in luce la dinamica dell’aumento dei contenuti di As nelle

bacche durante lo sviluppo e la maturazione, nonché la localizzazione largamente preferenziale di tale elemento a livello di buccia e polpa. Ha messo in evidenza inoltre l’esistenza di una buona correlazione dei contenuti di As nelle bacche mature con quelli nei suoli estratti con acetato di ammonio, oltre che con quelli presenti nelle foglie.

Viti cresciute sui suoli maggiormente dotati di As (es. particolari aree geologiche) presentano uve con contenuti più elevati di questo elemento. Tuttavia, nel piano sperimentale non sono state riscontrate quantità di As nelle uve tali da destare alcuna preoccupazione per la salute umana, essendo stati rilevati contenuti massimi pari a 6.23 µg/kg p.s. a fronte di una dose considerata tossica di 5-50 mg/giorno per gli adulti. I contenuti presenti in un kg di uva (peso secco) corrispondono invece al 25-50% della dose giornaliera di As richiesta per un uomo adulto mentre la quantità di As ingerita quotidianamente attraverso la dieta varia generalmente tra 0.04 e 1.4 mg (Pais, Jones, 1997).

RINGRAZIAMENTIIl presente lavoro è stato svolto nell’ambito di un dottorato di ricerca in Viticoltura,

Enologia e Marketing delle Imprese Vitivinicole (Università degli Studi di Padova). Gli autori ringraziano il personale tecnico dell’Unità Viticoltura della Fondazione Mach – Istituto Agrario San Michele all’Adige per l’aiuto nella fase di campionamento.

BIBLIOGRAFIAAdriano D.C., 2001. Arsenic. In: Trace elements in terrestrial environments.

Biogeochemistry, bioavailability, and risks of metals. Adriano D.C., Ed. II edizione. New York: Springer-Verlag. n: 220-261

Bertoldi D., Larcher R., Nicolini G., Bertamini M., Concheri G., 2009. Distribution of rare earth elements in Vitis vinifera L. ‘Chardonnay’ berries. Vitis, 48(1): 49-51.

Fang Y.L., Zhang A., Wang H., Li H., Zhang Z.W., Chen S.X., Luan L.Y., 2010. Health risk assessment of trace elements in Chinese raisins produced in Xinjiang province. Food Control, 21: 732-739.

Fitz W.J., Wenzel W.W., 2002. Arsenic trasformation in the soil-rhizosphere-plant system: fundamentals and potential application to phytoremediation. Journal of Biotechnology, 99: 259-278.

ISO 11466, 1995. Soil quality -Extraction of trace elements soluble in aqua regia. Jones C.A., Langer H.W., Anderson K., McDermott T.R., Inskeep W.P., 2000. Rates of

microbially mediated arsenate reduction and solubilization. Soil Science Society of America Journal, 64: 600-608.

Kabata-Pendias A., 2001. Elements of group V. In: Trace elements in soils and plants. Kabata-Pendias A., Ed. III edizione. Boca Raton: CRC Press. n: 225-232

Ko B.-G., Vogeler I., Bolan N. S., Clothier B., Green S., Kennedy J., 2007. Mobility of copper, chromium and arsenic from treated timber into grapevines. Science of the Total Environment, 388: 35-42.

Orescanin V., Katunar A., Kutle A., Valkovic V., 2003. Heavy metal in soil, grape, and wine. Journal of Trace and Microprobe Techniques, 21(1): 171-180.

Page A.L., Miller R.H., Kenney D.R., 1982. Methods of Soil Analysis. Part 2. Chemical and Microbiological Properties. 2° edizione. Madison: American Society of Agronomy, Soil Science Society of America.

Pais I., Jones J.B., 1997. Trace elements. In :The Handbook of trace elements. Pais I., Jones J.B, Eds. Boca Raton: St. Lucie Press. n: 86

Smith E., Naidu R., Alston A.M., 1999. Chemistry of arsenic in soils: I. Sorption of arsenate and arsenite by four Australian soils. Journal of Environmental Quality, 28: 1719-1726.

SSIR (2004). Soil Survey Laboratory Methods Manual, Soil Survey Investigation Report n° 42, metodo 5A8.

StatSoft Inc., 2008. STATISTICA (data analysis software system), version 8.0. Tulsa, OK, 74104, USA

Wenzel W.W., Kirchbaumer N., Prohaska T., Stingeder G., Lombi E., Adriano D.C., 2001. Arsenic fractionation in soils using an improved sequential extraction procedure. Analytica Chimica Acta, 436: 309-323.

4 - 61



EVALUATION OF TWO TRANSMITTANCE METERS IN ESTIMATING CHLOROPHYLL AND NITROGEN

CONCENTRATIONS IN GRAPEVINE CULTIVARS

D. Taskos (1), K. Karakioulakis(2), N. Theodorou(2), J.T. Tsialtas(3), E. Zioziou(2),N. Nikolaou(2), S. Koundouras(2)

(1) Boutari S.A., Goumenissa Winery, 613 00 Goumenissa, Greece, [email protected] (2) Laboratory of Viticulture, Aristotle University of Thessaloniki, 541 24, Thessaloniki, Greece,

[email protected] (3) NAGREF, Cotton and Industrial Plants Institute, 574 00 Sindos, Greece, [email protected]

ABSTRACT Two transmittance-based chlorophyll meters (SPAD-502 and CCM-200) were evaluated in

estimating chlorophyll (Chl) and nitrogen (N) levels in grapevine leaves. The study was conducted in a fertilization experiment [0 (N0), 60 (N1) and 120 (N2) kg N/ha] during the summer 2009, in two commercial vineyards located in Northern Greece and planted with cvs Cabernet-Sauvignon and Xinomavro (Vitis vinifera L.). When data were pooled over cultivars and samplings, leaves of N2 vines had the highest N and Chl content, as well as SPAD and CCM readings, followed by the respective values of N1. However, neither of the devices could detect the seasonal decline in leaf N and Chl content. Significant relationships between extracted Chl and measured leaf N were found in both cultivars. A strong linear function related SPAD and CCM readings in both cultivars. Total Chl and N were strongly correlated with SPAD and CCM readings in Cabernet Sauvignon (p<0.001) while relationships were poor for SPAD and not significant for CCM in Xinomavro. The results suggest that non-destructive chlorophyll estimations by transmittance-based meters are not applicable in all situations without specific calibrations necessary to improve their utility and accuracy over grapevine cultivars.

KEYWORD SPAD-502 – CCM-200 – chlorophyll – nitrogen – grapevine – N fertilization

INTRODUCTIONNitrogen (N) is the most important nutrient in grapevine, as it participates in many

physiological processes and has the potential to manipulate vine growth and productivity with significant implications for grape and wine composition (Bell and Henschke, 2005). Monitoring of vine N status in the field can be important in determining N fertilizer amount and time of application. Since chlorophyll (Chl) is a nitrogenous pigment, leaf Chl content provides an indirect estimation of N plant status (Steele et al., 2008). However, conventional extraction of leaf Chl with various organic solvents is laborious, time consuming and destructive, thus not adapted for N fertilization scheduling. Recently, non-destructive leaf "greenness" measurements have been advocated for rapid determination of leaf Chl and N status, mainly in annual crops (Filella et al., 1995; Bullock and Anderson, 1998) while fewer studies have been conducted on woody species as grapevine (Fanizza et al., 1991).

The aim of the present study was to evaluate the utility of two handheld transmittance-based Chl content meters in estimating Chl and N levels in intact leaves of two grapevine cultivars, in an N fertilization experiment.

MATERIALS AND METHODS The study was conducted in two commercial vineyard blocks located in Goumenissa

(Northern Greece) in the summer of 2009, planted with cvs Cabernet-Sauvignon and Xinomavro (Vitis vinifera L.) respectively and grafted onto 1103P. Vines were spaced 2.2 1.3 m and trained on a spur-pruned bilateral cordon. Nitrogen in the form of NH4NO3 and corresponding to three rates [0 (N0), 60 (N1) and 120 (N2) kg/ha of N] was applied at budburst. The experimental design was that of completely randomised blocks with three replications. Individual plots consisted of 6 vines distributed on two adjacent rows, and were separated by at least 6 border vines within a row.

Two handheld chlorophyll meters were evaluated: soil-plant analysis development meter (SPAD-502, Minolta Co., Osaka, Japan) and chlorophyll content meter (CCM-200, Opti-Sciences, Tyngsboro, USA). Measurements with the SPAD and CCM devices were conducted on three exterior, fully expanded leaves per plot, located on the basal shoot nodes, on four occasions during the growing season [berry set (d1), bunch closure (d2), veraison (d3) and harvest (d4)]. For each leaf, three readings on separate lobs were averaged to represent one observation. Immediately following Chl meter readings, leaves were cut, sealed in plastic bags and transported to the laboratory in a cooler for Chl and N determination. For each of the leaves sampled, three 1 cm2 disks were cut from the same leaf areas used for Chl meter readings, weighed and extracted for 3 h in 80% ethanol solution, at 78°C in a J.P. Selecta Precisterm bath (Barcelona, Spain). Chl a and Chl b concentration of the aliquot was estimated according to the equations proposed by Arnon (1949) after measurement of the optical density at 645 and 663 nm using a 6305 UV/VIS mini spectrophotometer (Jenway Ltd, Essex, UK), and results were expressed on a fresh weight basis. The remaining leaf tissues were dried at 70°C and used for total leaf nitrogen measurement (% dry weight) by an automated combustion elemental analyzer (PDZ Europa, Cheshire, UK).

Data were subjected to analysis of variance and correlation analysis using SPSS software (version 14.0, SPSS Inc., IL, USA). Only the mean of the three measurements per plot was used in data analysis. Comparison of means was performed using Duncan’s multiple range test at p<0.05.

RESULTS AND DISCUSSIONLeaf N concentration was reduced significantly with the progress of growing season in both

varieties, especially until veraison (d3). N fertilization significantly increased leaf N concentration with higher levels in N2 vines for both cultivars (Fig. 1). When data were pooled over samplings, N concentration was 1.96 % in N0, 2.09 % in N1 and 2.38 % in N2 in Cabernet Sauvignon (p<0.001) and 1.94 %, 2.08 % and 2.21 % respectively, in Xinomavro (p<0.001). ANOVA did not detect any sampling fertilization interaction in any of the varieties studied. Direct comparisons among cultivars could not be made, as cultivars were located in separate blocs.

4 - 62



EVALUATION OF TWO TRANSMITTANCE METERS IN ESTIMATING CHLOROPHYLL AND NITROGEN

CONCENTRATIONS IN GRAPEVINE CULTIVARS

D. Taskos (1), K. Karakioulakis(2), N. Theodorou(2), J.T. Tsialtas(3), E. Zioziou(2),N. Nikolaou(2), S. Koundouras(2)

(1) Boutari S.A., Goumenissa Winery, 613 00 Goumenissa, Greece, [email protected] (2) Laboratory of Viticulture, Aristotle University of Thessaloniki, 541 24, Thessaloniki, Greece,

[email protected] (3) NAGREF, Cotton and Industrial Plants Institute, 574 00 Sindos, Greece, [email protected]

ABSTRACT Two transmittance-based chlorophyll meters (SPAD-502 and CCM-200) were evaluated in

estimating chlorophyll (Chl) and nitrogen (N) levels in grapevine leaves. The study was conducted in a fertilization experiment [0 (N0), 60 (N1) and 120 (N2) kg N/ha] during the summer 2009, in two commercial vineyards located in Northern Greece and planted with cvs Cabernet-Sauvignon and Xinomavro (Vitis vinifera L.). When data were pooled over cultivars and samplings, leaves of N2 vines had the highest N and Chl content, as well as SPAD and CCM readings, followed by the respective values of N1. However, neither of the devices could detect the seasonal decline in leaf N and Chl content. Significant relationships between extracted Chl and measured leaf N were found in both cultivars. A strong linear function related SPAD and CCM readings in both cultivars. Total Chl and N were strongly correlated with SPAD and CCM readings in Cabernet Sauvignon (p<0.001) while relationships were poor for SPAD and not significant for CCM in Xinomavro. The results suggest that non-destructive chlorophyll estimations by transmittance-based meters are not applicable in all situations without specific calibrations necessary to improve their utility and accuracy over grapevine cultivars.

KEYWORD SPAD-502 – CCM-200 – chlorophyll – nitrogen – grapevine – N fertilization

INTRODUCTIONNitrogen (N) is the most important nutrient in grapevine, as it participates in many

physiological processes and has the potential to manipulate vine growth and productivity with significant implications for grape and wine composition (Bell and Henschke, 2005). Monitoring of vine N status in the field can be important in determining N fertilizer amount and time of application. Since chlorophyll (Chl) is a nitrogenous pigment, leaf Chl content provides an indirect estimation of N plant status (Steele et al., 2008). However, conventional extraction of leaf Chl with various organic solvents is laborious, time consuming and destructive, thus not adapted for N fertilization scheduling. Recently, non-destructive leaf "greenness" measurements have been advocated for rapid determination of leaf Chl and N status, mainly in annual crops (Filella et al., 1995; Bullock and Anderson, 1998) while fewer studies have been conducted on woody species as grapevine (Fanizza et al., 1991).

The aim of the present study was to evaluate the utility of two handheld transmittance-based Chl content meters in estimating Chl and N levels in intact leaves of two grapevine cultivars, in an N fertilization experiment.

MATERIALS AND METHODS The study was conducted in two commercial vineyard blocks located in Goumenissa

(Northern Greece) in the summer of 2009, planted with cvs Cabernet-Sauvignon and Xinomavro (Vitis vinifera L.) respectively and grafted onto 1103P. Vines were spaced 2.2 1.3 m and trained on a spur-pruned bilateral cordon. Nitrogen in the form of NH4NO3 and corresponding to three rates [0 (N0), 60 (N1) and 120 (N2) kg/ha of N] was applied at budburst. The experimental design was that of completely randomised blocks with three replications. Individual plots consisted of 6 vines distributed on two adjacent rows, and were separated by at least 6 border vines within a row.

Two handheld chlorophyll meters were evaluated: soil-plant analysis development meter (SPAD-502, Minolta Co., Osaka, Japan) and chlorophyll content meter (CCM-200, Opti-Sciences, Tyngsboro, USA). Measurements with the SPAD and CCM devices were conducted on three exterior, fully expanded leaves per plot, located on the basal shoot nodes, on four occasions during the growing season [berry set (d1), bunch closure (d2), veraison (d3) and harvest (d4)]. For each leaf, three readings on separate lobs were averaged to represent one observation. Immediately following Chl meter readings, leaves were cut, sealed in plastic bags and transported to the laboratory in a cooler for Chl and N determination. For each of the leaves sampled, three 1 cm2 disks were cut from the same leaf areas used for Chl meter readings, weighed and extracted for 3 h in 80% ethanol solution, at 78°C in a J.P. Selecta Precisterm bath (Barcelona, Spain). Chl a and Chl b concentration of the aliquot was estimated according to the equations proposed by Arnon (1949) after measurement of the optical density at 645 and 663 nm using a 6305 UV/VIS mini spectrophotometer (Jenway Ltd, Essex, UK), and results were expressed on a fresh weight basis. The remaining leaf tissues were dried at 70°C and used for total leaf nitrogen measurement (% dry weight) by an automated combustion elemental analyzer (PDZ Europa, Cheshire, UK).

Data were subjected to analysis of variance and correlation analysis using SPSS software (version 14.0, SPSS Inc., IL, USA). Only the mean of the three measurements per plot was used in data analysis. Comparison of means was performed using Duncan’s multiple range test at p<0.05.

RESULTS AND DISCUSSIONLeaf N concentration was reduced significantly with the progress of growing season in both

varieties, especially until veraison (d3). N fertilization significantly increased leaf N concentration with higher levels in N2 vines for both cultivars (Fig. 1). When data were pooled over samplings, N concentration was 1.96 % in N0, 2.09 % in N1 and 2.38 % in N2 in Cabernet Sauvignon (p<0.001) and 1.94 %, 2.08 % and 2.21 % respectively, in Xinomavro (p<0.001). ANOVA did not detect any sampling fertilization interaction in any of the varieties studied. Direct comparisons among cultivars could not be made, as cultivars were located in separate blocs.

4 - 63



Figure 1. Seasonal variation of leaf N concentration (% dry weight) in Cabernet Sauvignon and Xinomavro;

N0, N1 and N2 correspond to 0, 60 and 120 kg N/ha, respectively. Total Chl (a+b) followed a similar seasonal pattern with leaf N, decreasing values from d1

to d4 in all N treatments (Fig. 2). Chl a was more intensively degraded than Chl b (Netto etal., 2005) before veraison (d3), especially in Cabernet Sauvignon, whereas Chl a and Chl b followed an opposite pattern after veraison, with increasing values for Chl a (Fig. 2). These results largely explain the late season increase in the Chl a/b ratio in both varieties. According to Kitajima and Hogan (2003), increase of the Chl a/b ratio is an indication of plant acclimation to N limitation, conditions that occurred during the late stages of the growing period in this study.

N application significantly increased Chl content in leaves of both varieties, with highest levels in N2 for both cultivars (4.40 and 3.83 mg/g in Cabernet Sauvignon and Xinomavro respectively, pooled data across samplings) while significant differences between N0 and N1 were only observed in Xinomavro (3.11 and 3.43 mg/g respectively). Similar results were obtained when Chl content was expressed per leaf area (data not shown).

A regression using data from the four sampling times (n=72) showed a strong positive correlation between leaf N and extracted Chl in both varieties (p<0.001; Tab. 1 and Tab. 2) suggesting a direct response of Chl synthesis to N levels in leaves (Syvertsen, 1987), with higher coefficients for Chl a compared to Chl b. All Chl traits were positively correlated with each other with few exceptions (Tab. 1 and Tab. 2). However, variation of total Chl was highly related to changes in Chl a, in both varieties.

Leaves of N2 vines had constantly the highest SPAD and CCM readings followed by the respective readings of N1 in both cultivars (Fig. 3). However, Chl meter readings from both devices remained relatively stable during the growth period, with no significant difference between samplings.

SPAD and CCM readings were strongly correlated with each other in both varieties (Tab. 1 and Tab. 2). Although previous studies in grapevines have reported quadratic relationships between SPAD readings and leaf traits (Steele et al., 2008), in our study, best-fitted curves between both Chl meter readings and leaf Chl or N were linear (Fanizza et al., 1991). This result is probably due to the narrow range of Chl variation in the conditions of this experiment (Jifon et al., 2005).

Figure 2. Seasonal variation of leaf chlorophyll [Chl a, Chl b, Chl (a+b)] concentration and Chl a/b ratio in

Cabernet Sauvignon and Xinomavro; N0, N1 and N2 correspond to 0, 60 and 120 kg N/ha, respectively.

4 - 64



Figure 1. Seasonal variation of leaf N concentration (% dry weight) in Cabernet Sauvignon and Xinomavro;

N0, N1 and N2 correspond to 0, 60 and 120 kg N/ha, respectively. Total Chl (a+b) followed a similar seasonal pattern with leaf N, decreasing values from d1

to d4 in all N treatments (Fig. 2). Chl a was more intensively degraded than Chl b (Netto etal., 2005) before veraison (d3), especially in Cabernet Sauvignon, whereas Chl a and Chl b followed an opposite pattern after veraison, with increasing values for Chl a (Fig. 2). These results largely explain the late season increase in the Chl a/b ratio in both varieties. According to Kitajima and Hogan (2003), increase of the Chl a/b ratio is an indication of plant acclimation to N limitation, conditions that occurred during the late stages of the growing period in this study.

N application significantly increased Chl content in leaves of both varieties, with highest levels in N2 for both cultivars (4.40 and 3.83 mg/g in Cabernet Sauvignon and Xinomavro respectively, pooled data across samplings) while significant differences between N0 and N1 were only observed in Xinomavro (3.11 and 3.43 mg/g respectively). Similar results were obtained when Chl content was expressed per leaf area (data not shown).

A regression using data from the four sampling times (n=72) showed a strong positive correlation between leaf N and extracted Chl in both varieties (p<0.001; Tab. 1 and Tab. 2) suggesting a direct response of Chl synthesis to N levels in leaves (Syvertsen, 1987), with higher coefficients for Chl a compared to Chl b. All Chl traits were positively correlated with each other with few exceptions (Tab. 1 and Tab. 2). However, variation of total Chl was highly related to changes in Chl a, in both varieties.

Leaves of N2 vines had constantly the highest SPAD and CCM readings followed by the respective readings of N1 in both cultivars (Fig. 3). However, Chl meter readings from both devices remained relatively stable during the growth period, with no significant difference between samplings.

SPAD and CCM readings were strongly correlated with each other in both varieties (Tab. 1 and Tab. 2). Although previous studies in grapevines have reported quadratic relationships between SPAD readings and leaf traits (Steele et al., 2008), in our study, best-fitted curves between both Chl meter readings and leaf Chl or N were linear (Fanizza et al., 1991). This result is probably due to the narrow range of Chl variation in the conditions of this experiment (Jifon et al., 2005).

Figure 2. Seasonal variation of leaf chlorophyll [Chl a, Chl b, Chl (a+b)] concentration and Chl a/b ratio in

Cabernet Sauvignon and Xinomavro; N0, N1 and N2 correspond to 0, 60 and 120 kg N/ha, respectively.

4 - 65



Figure 3. Seasonal variation of SPAD and CCM readings in Cabernet Sauvignon and Xinomavro; N0, N1 and

N2 correspond to 0, 60 and 120 kg Ν/ha, respectively. In Cabernet Sauvignon, SPAD and CCM readings were significantly correlated with leaf

Chl traits and N concentration, although correlation coefficients were low (Tab. 1). Moreover, Chl meter readings were better correlated with Chl a than Chl b, as previously reported in other plants (Madeira et al., 2003). In Xinomavro, relationships were generally poor for SPAD and not significant for CCM (Tab. 2), possibly due to the thicker leaves of this variety (Jiffon et al., 2005).

Table 1. Correlation coefficients and significance level for the leaf traits determined in Cabernet Sauvignon.

*, **, ***: significant coefficients at p<0.05, p<0.01, or p<0.001, respectively, ns: not significant (n=72); Chl a, Chl b, Chl (a+b): (mg g-1 fresh weight); N (% dry weight).

Chl a Chl b Chl a/b N SPAD CCM Chl (a+b) 0.878*** 0.740*** 0.543*** 0.878*** 0.490*** 0.461*** Chl a 0.329** 0.868*** 0.886*** 0.484*** 0.437*** Chl b ns 0.488*** 0.286* 0.295* Chl a/b 0.644*** 0.371*** 0.329** N 0.464*** 0.419*** SPAD 0.952***

Table 2. Correlation coefficients and significance level for the leaf traits determined in Xinomavro. *, **, ***: significant coefficients at p<0.05, p<0.01, or p<0.001, respectively, ns: not significant (n=72);

Chl a, Chl b, Chl (a+b): (mg g-1 fresh weight); N (% dry weight). Chl a Chl b Chl a/b N SPAD CCM Chl (a+b) 0.907*** 0.893*** 0.507*** 0.824*** 0.348** ns Chl a 0.620*** 0.811*** 0.825*** 0.294* ns Chl b ns 0.659*** 0.305** ns Chl a/b 0.545*** ns ns N 0.345** 0.246* SPAD 0.903***

CONCLUSIONS Leaf Chl and N concentration were linearly related to SPAD and CCM readings in

grapevine. However, correlation coefficients were stronger in Cabernet Sauvignon than Xinomavro, especially for the CCM device. Moreover, none of the devices was able to detect the seasonal N and Chl pattern. The results suggest that non-destructive chlorophyll estimations by transmittance-based meters are not applicable in all situations for accurate N status monitoring during vine growth cycle, and that specific calibrations are recommendable to improve their utility and accuracy across grapevine cultivars. However, both devices accurately distinguished N application levels and thus can serve as an indicator of seasonal N nutritional status.

BIBLIOGRAPHY Arnon D.I., 1949. Copper enzymes in isolated chlorophlasts and polyphenol oxidase in Beta

vulgaris. Plant Phys., 24: 1-15. Bell S.-J. and Henschke P.A., 2005. Implications of nitrogen nutrition for grapes,

fermentation and wine. Aust. J. Grape Wine Res., 11: 242-295. Bullock D.G. and Anderson D.S., 1998. Evaluation of the Minolta SPAD-502 chlorophyll

meter for nitrogen management in corn. J. Plant Nutr., 21: 741-755. Fanizza G., Della Gatta C. and Bagnulo C., 1991. A non-destructive determination of leaf

chlorophyll in Vitis vinifera. Ann. appl. Biol., 119: 203-205. Filella I., Serrano I., Serra J. and Penuelas J., 1995. Evaluating wheat nitrogen status with

canopy reflectances indices and discriminant analysis. Crop. Sci., 35: 1400-1405. Jifon J.L., Syvertsen J.P. and Whaley E., 2005. Growth environment and leaf anatomy affect

non-destructive estimates of chlorophyll and nitrogen in Citrus sp. leaves. J. Amer. Soc. Hort. Sci., 130: 152-158.

Kitajima K. and Hogan K.P., 2003. Increases of chlorophyll a/b ratios during acclimation of tropical woody seedlings to nitrogen limitation and high light. Plant Cell Environ. 26: 857-865.

Madeira A.C., Ferreira A., de Varennes A. and Vieira A.I., 2003. SPAD meter versus Tristimulus colorimeter to estimate chlorophyll content and leaf color in sweet pepper. Commun. Soil Sci. Plant Anal., 34: 2461-2470.

Netto A.T., Campostrini E., Goncalves de Oiveira J., Bressan-Smith R.E., 2004. Photosynthetic pigments, nitrogen, chlorophyll a fluorescence and SPAD-502 readings in coffee leaves. Scientia Hort., 104: 199-209.

Steel M.R., Gitelson A.A. and Rundquist D.C., 2008. A comparison of two techniques for nondestructive measurement of chlorophyll content in grapevine leaves. Agron. J., 100: 779-782.

4 - 66



Figure 3. Seasonal variation of SPAD and CCM readings in Cabernet Sauvignon and Xinomavro; N0, N1 and

N2 correspond to 0, 60 and 120 kg Ν/ha, respectively. In Cabernet Sauvignon, SPAD and CCM readings were significantly correlated with leaf

Chl traits and N concentration, although correlation coefficients were low (Tab. 1). Moreover, Chl meter readings were better correlated with Chl a than Chl b, as previously reported in other plants (Madeira et al., 2003). In Xinomavro, relationships were generally poor for SPAD and not significant for CCM (Tab. 2), possibly due to the thicker leaves of this variety (Jiffon et al., 2005).

Table 1. Correlation coefficients and significance level for the leaf traits determined in Cabernet Sauvignon.

*, **, ***: significant coefficients at p<0.05, p<0.01, or p<0.001, respectively, ns: not significant (n=72); Chl a, Chl b, Chl (a+b): (mg g-1 fresh weight); N (% dry weight).

Chl a Chl b Chl a/b N SPAD CCM Chl (a+b) 0.878*** 0.740*** 0.543*** 0.878*** 0.490*** 0.461*** Chl a 0.329** 0.868*** 0.886*** 0.484*** 0.437*** Chl b ns 0.488*** 0.286* 0.295* Chl a/b 0.644*** 0.371*** 0.329** N 0.464*** 0.419*** SPAD 0.952***

Table 2. Correlation coefficients and significance level for the leaf traits determined in Xinomavro. *, **, ***: significant coefficients at p<0.05, p<0.01, or p<0.001, respectively, ns: not significant (n=72);

Chl a, Chl b, Chl (a+b): (mg g-1 fresh weight); N (% dry weight). Chl a Chl b Chl a/b N SPAD CCM Chl (a+b) 0.907*** 0.893*** 0.507*** 0.824*** 0.348** ns Chl a 0.620*** 0.811*** 0.825*** 0.294* ns Chl b ns 0.659*** 0.305** ns Chl a/b 0.545*** ns ns N 0.345** 0.246* SPAD 0.903***

CONCLUSIONS Leaf Chl and N concentration were linearly related to SPAD and CCM readings in

grapevine. However, correlation coefficients were stronger in Cabernet Sauvignon than Xinomavro, especially for the CCM device. Moreover, none of the devices was able to detect the seasonal N and Chl pattern. The results suggest that non-destructive chlorophyll estimations by transmittance-based meters are not applicable in all situations for accurate N status monitoring during vine growth cycle, and that specific calibrations are recommendable to improve their utility and accuracy across grapevine cultivars. However, both devices accurately distinguished N application levels and thus can serve as an indicator of seasonal N nutritional status.

BIBLIOGRAPHY Arnon D.I., 1949. Copper enzymes in isolated chlorophlasts and polyphenol oxidase in Beta

vulgaris. Plant Phys., 24: 1-15. Bell S.-J. and Henschke P.A., 2005. Implications of nitrogen nutrition for grapes,

fermentation and wine. Aust. J. Grape Wine Res., 11: 242-295. Bullock D.G. and Anderson D.S., 1998. Evaluation of the Minolta SPAD-502 chlorophyll

meter for nitrogen management in corn. J. Plant Nutr., 21: 741-755. Fanizza G., Della Gatta C. and Bagnulo C., 1991. A non-destructive determination of leaf

chlorophyll in Vitis vinifera. Ann. appl. Biol., 119: 203-205. Filella I., Serrano I., Serra J. and Penuelas J., 1995. Evaluating wheat nitrogen status with

canopy reflectances indices and discriminant analysis. Crop. Sci., 35: 1400-1405. Jifon J.L., Syvertsen J.P. and Whaley E., 2005. Growth environment and leaf anatomy affect

non-destructive estimates of chlorophyll and nitrogen in Citrus sp. leaves. J. Amer. Soc. Hort. Sci., 130: 152-158.

Kitajima K. and Hogan K.P., 2003. Increases of chlorophyll a/b ratios during acclimation of tropical woody seedlings to nitrogen limitation and high light. Plant Cell Environ. 26: 857-865.

Madeira A.C., Ferreira A., de Varennes A. and Vieira A.I., 2003. SPAD meter versus Tristimulus colorimeter to estimate chlorophyll content and leaf color in sweet pepper. Commun. Soil Sci. Plant Anal., 34: 2461-2470.

Netto A.T., Campostrini E., Goncalves de Oiveira J., Bressan-Smith R.E., 2004. Photosynthetic pigments, nitrogen, chlorophyll a fluorescence and SPAD-502 readings in coffee leaves. Scientia Hort., 104: 199-209.

Steel M.R., Gitelson A.A. and Rundquist D.C., 2008. A comparison of two techniques for nondestructive measurement of chlorophyll content in grapevine leaves. Agron. J., 100: 779-782.

4 - 67



ASSESSMENT OF THE OPTIMAL NUMBER OF OBSERVATIONS IN THE STUDY OF VINEYARD SOIL (RIGOSOL)

Djordjević, A.1, *Životić, Lj. 1, Sivčev, B.1, Pajić, V.1, Ranković-Vasić, Z.1, Radovanović,

D1. 1University of Belgrade, Faculty of Agriculture, Nemanjina 6, Belgrade, Zemun, Republic of Serbia,

*[email protected]

ABSTRACT A study of soil pH on the experimental field resulted in a high variability of pH on a very

small scale. This kind of heterogenity in soil pH have effects on growth of two grapevine varieties on rootstock Kober 5BB: Riesling and Pinot Noir A number of 104 soil samples were taken from an area of 1.43 ha from two depths. A goal of this experiment was to find the optimum number of samples for pH studies, and to implement the obtained results in further investigation on experimental fields. Therefore, in this paper we compared diferent deterministic interpolation techniques: inverse distance weight, splines and local polynomial interpolation, on the results of soil pH. Root mean square error (RMSE) statistitics obtained after cross validation procedure was used for the choice of appropriate exponent value for IDW, spline and local interpolation. The obtained interpolation parameters were used for mapping the field and the most accurate technique was IDW, which was further used in creation of pH maps with lower number of samples: 54, 34, 29, 24, 19 and only 14 pH samples. Maps were classified and compared by means of percentage difference in area among classes of pH in respect to classes obtained after maximum sampling. The results indicated that the criteria of 15% of change in pH area over classes could be satisfied with only on third of the samples. An obtained results will be used for further sampling of the whole experimental area.

KEYWORD vineyard, soil, pH, interpolation, IDW, RBF, LP

INTRODUCTION The feasibility of site-specific management relies on the understanding of temporal and

spatial components of variability. Spatial variability of soil properties may vary due to natural variations or imposed sources of variability. The variability of soil properties could be vertically and horizontally along a field. The structure of variability in soil properties showed differences according to sampling spacing, soil properties, and method used in the study (Trangmar et al., 1985). Geographical Information Systems (GIS) have potential for handling information on variable soil conditions at all scales (Lark and Bolam, 1997) and the standard use of GIS implies the manipulation of data layer and generation of secondary data.

The accuracy of both geostatistical and deterministic interpolation methods was analysed in several studies (Laslet et al., 1987; Weber and Englund, 1992; Gotway et al., 1996; Kravchenko and Bullock, 1999; Li et al., 2007). Laslett et al. (1987) found splines to be better than IDW and Kriging in the analysis of soil pH, while other authors have found kriging better then IDW in the analysis of some other soil properties (Gotway et al., 1996). Weber and Englund (1992) have found IDW producing better results than kriging. Many conflincting reports were found taking into consideration the use of basic statistics to

predetermine both interpolation method and their parameters, and they are reported in Kravchenko and Bullock (1999), Weber and Englund (1992) and Gotway et al. (1996).

Taking into consideration the variability of the results of previous studies the objectives of this study were to: a) assses the accuracy of well known deterministic interpolation techniques, IDW, splines and local interpolation, in mapping soil topsoil and subsoil pH through the manipulation of various parameters attributable to each technique, b) choose the appropriate technique for the examination of classes of topsoil and subsoil pH on the lower sampling density datasets obtained by eliminating samplng points from the dense grid, c) predict the optimal number of samples in mapping the soil pH on a basis of a produced maps.

MATERIALS AND METHODS The study area is vineyard at Experimental Station Faculty of Agriculture “Radmilovac”,

which is located 8 km south-east from Belgrade on hilly terrain, at an altitude of 150 m. Total area of experimental field is 1.43 ha. Experimental field has rectangular shape, with a length of 140 m, and width of 102 m. According to soil taxonomy soils are defined as Rigosols. The region is characterized with temperate continental climate; air temperature for the period 1961-2001 increase for 1°C and mean annual is 11.8°C, which is in accordance with the forecast change (Vuković et al., 2009). During vegetation period precipitation average 401.7 mm and belong to “sub-humid category”. Kober 5BB is ideal rootstock drought common and tolerance to lime-based soils. An experimental field is divided in two parts; the upper part of the field is under cv. Pinot Noir, and the downer part of the field is under cv. Riesling. Total number of 104 samples for the topsoil (0-30 cm) and subsoil (30-60 cm) were taken. The sampling space within a row was 10 m, while inter-row space was 15 m. Each sampling point and borders of the field were recorded and geo-referenced by using Trimble global positioning system. Soil pH in 1 N KCl was measured with glass electrode by using pH meter method.

Soils with a pH less than 4.5 in KCl in the topsoil could be considered toxic to most crops. The effect of pH of crops is presented by soil pH classes in Tab. 1 (Cerling, 1990). Three deterministic interpolative techniques characterized with their simplicity and easy handling are used in this study.

Tab. 1 Classes of soil pH in KCl Group Intensity of acidity pH in KCl

1 Very extreme <4.5 2 Moderately acid 4.5-5.0 3 Acid 5.0-5.5 4 Acid to neutral 5.5-6.0 5 Neutral >6.0

Inverse distance weight (IDW) is deterministic interpolation technique which implements

directly the assumption that the things that are close to one another are more alike than those that are farther apart, therefore, within this interpolator a value at an un sampled location is a weighted average of known data points within surrounding neighborhood (Shepard, 1968). Measured values closer to the prediction location will have more impact than those farther away. Therefore, an assumption is that a local influence of each measured point diminishes with distance. It is an exact interpolator. The area calculated by using IDW depends on the selection of neighborhood strategy and power parameter.

Radial basis functions (RBF) methods are a series of exact interpolation techniques. It consists of five different functions each resulting in different interpolation surface. Although

4 - 68



ASSESSMENT OF THE OPTIMAL NUMBER OF OBSERVATIONS IN THE STUDY OF VINEYARD SOIL (RIGOSOL)

Djordjević, A.1, *Životić, Lj. 1, Sivčev, B.1, Pajić, V.1, Ranković-Vasić, Z.1, Radovanović,

D1. 1University of Belgrade, Faculty of Agriculture, Nemanjina 6, Belgrade, Zemun, Republic of Serbia,

*[email protected]

ABSTRACT A study of soil pH on the experimental field resulted in a high variability of pH on a very

small scale. This kind of heterogenity in soil pH have effects on growth of two grapevine varieties on rootstock Kober 5BB: Riesling and Pinot Noir A number of 104 soil samples were taken from an area of 1.43 ha from two depths. A goal of this experiment was to find the optimum number of samples for pH studies, and to implement the obtained results in further investigation on experimental fields. Therefore, in this paper we compared diferent deterministic interpolation techniques: inverse distance weight, splines and local polynomial interpolation, on the results of soil pH. Root mean square error (RMSE) statistitics obtained after cross validation procedure was used for the choice of appropriate exponent value for IDW, spline and local interpolation. The obtained interpolation parameters were used for mapping the field and the most accurate technique was IDW, which was further used in creation of pH maps with lower number of samples: 54, 34, 29, 24, 19 and only 14 pH samples. Maps were classified and compared by means of percentage difference in area among classes of pH in respect to classes obtained after maximum sampling. The results indicated that the criteria of 15% of change in pH area over classes could be satisfied with only on third of the samples. An obtained results will be used for further sampling of the whole experimental area.

KEYWORD vineyard, soil, pH, interpolation, IDW, RBF, LP

INTRODUCTION The feasibility of site-specific management relies on the understanding of temporal and

spatial components of variability. Spatial variability of soil properties may vary due to natural variations or imposed sources of variability. The variability of soil properties could be vertically and horizontally along a field. The structure of variability in soil properties showed differences according to sampling spacing, soil properties, and method used in the study (Trangmar et al., 1985). Geographical Information Systems (GIS) have potential for handling information on variable soil conditions at all scales (Lark and Bolam, 1997) and the standard use of GIS implies the manipulation of data layer and generation of secondary data.

The accuracy of both geostatistical and deterministic interpolation methods was analysed in several studies (Laslet et al., 1987; Weber and Englund, 1992; Gotway et al., 1996; Kravchenko and Bullock, 1999; Li et al., 2007). Laslett et al. (1987) found splines to be better than IDW and Kriging in the analysis of soil pH, while other authors have found kriging better then IDW in the analysis of some other soil properties (Gotway et al., 1996). Weber and Englund (1992) have found IDW producing better results than kriging. Many conflincting reports were found taking into consideration the use of basic statistics to

predetermine both interpolation method and their parameters, and they are reported in Kravchenko and Bullock (1999), Weber and Englund (1992) and Gotway et al. (1996).

Taking into consideration the variability of the results of previous studies the objectives of this study were to: a) assses the accuracy of well known deterministic interpolation techniques, IDW, splines and local interpolation, in mapping soil topsoil and subsoil pH through the manipulation of various parameters attributable to each technique, b) choose the appropriate technique for the examination of classes of topsoil and subsoil pH on the lower sampling density datasets obtained by eliminating samplng points from the dense grid, c) predict the optimal number of samples in mapping the soil pH on a basis of a produced maps.

MATERIALS AND METHODS The study area is vineyard at Experimental Station Faculty of Agriculture “Radmilovac”,

which is located 8 km south-east from Belgrade on hilly terrain, at an altitude of 150 m. Total area of experimental field is 1.43 ha. Experimental field has rectangular shape, with a length of 140 m, and width of 102 m. According to soil taxonomy soils are defined as Rigosols. The region is characterized with temperate continental climate; air temperature for the period 1961-2001 increase for 1°C and mean annual is 11.8°C, which is in accordance with the forecast change (Vuković et al., 2009). During vegetation period precipitation average 401.7 mm and belong to “sub-humid category”. Kober 5BB is ideal rootstock drought common and tolerance to lime-based soils. An experimental field is divided in two parts; the upper part of the field is under cv. Pinot Noir, and the downer part of the field is under cv. Riesling. Total number of 104 samples for the topsoil (0-30 cm) and subsoil (30-60 cm) were taken. The sampling space within a row was 10 m, while inter-row space was 15 m. Each sampling point and borders of the field were recorded and geo-referenced by using Trimble global positioning system. Soil pH in 1 N KCl was measured with glass electrode by using pH meter method.

Soils with a pH less than 4.5 in KCl in the topsoil could be considered toxic to most crops. The effect of pH of crops is presented by soil pH classes in Tab. 1 (Cerling, 1990). Three deterministic interpolative techniques characterized with their simplicity and easy handling are used in this study.

Tab. 1 Classes of soil pH in KCl Group Intensity of acidity pH in KCl

1 Very extreme <4.5 2 Moderately acid 4.5-5.0 3 Acid 5.0-5.5 4 Acid to neutral 5.5-6.0 5 Neutral >6.0

Inverse distance weight (IDW) is deterministic interpolation technique which implements

directly the assumption that the things that are close to one another are more alike than those that are farther apart, therefore, within this interpolator a value at an un sampled location is a weighted average of known data points within surrounding neighborhood (Shepard, 1968). Measured values closer to the prediction location will have more impact than those farther away. Therefore, an assumption is that a local influence of each measured point diminishes with distance. It is an exact interpolator. The area calculated by using IDW depends on the selection of neighborhood strategy and power parameter.

Radial basis functions (RBF) methods are a series of exact interpolation techniques. It consists of five different functions each resulting in different interpolation surface. Although

4 - 69



being exact interpolators these methods could predict values higher and lower than measured maximum and minimum values. RBFs are formed in geostatistics around each data location. Splines consist of polynomials, which describe pieces of a line or surface, and they are fitted together so that they join smoothly (Webster and Oliver, 2001).

Local Polynomial interpolation (Schaum, 2008) is an inexact interpolator that predicts a value that is different from the measured value. It presents a kind of combination of global polynomial methods and moving average procedure. Local polynomial interpolation fits many polynomials, each within specified overlapping neighborhoods. This kind of interpolation provides surfaces that accounts for more local variation. Instead of using all data, like in global polynomial interpolation, it uses data within localized windows.

Statistical analysis and spatial predictions in this paper were conducted by using Geostatistical Wizard implemented in ArcGIS 9.2 version.

RESULTS AND DISCUSSION Data description is achieved through basic summary statistics, including means, medians,

variances and skewness. A statistical summary of the pH in 1N KCl on two depths is presented in Tab. 2.

Tab. 2 Summary statistics for topsoil (0-30 cm) and subsoil (30-60 cm) pH in KCl

Soil pH N Min Max Range Mean Median Var CV (%) Skewness Kurtosis

Topsoil (0-30 cm) 104 3.73 7.93 4.2 5.31 4.86 1.48 23.1 0.479 1.775

Subsoil (30-60 cm) 104 3.55 7.36 3.81 5.05 4.57 1.38 23.3 0.707 2.07

Cross-validation is commonly used to validate the accuracy of interpolation (Voltz and

Webster, 1990). It is achieved by eliminating information, generally one observation at a time, estimating the value at that location with the remaining data and then computing the difference between the actual and estimated value for each data location (Davis, 1987). For the comparison of different interpolation techniques, we examined the difference between the measured and predicted data by using the mean error and the root mean squared error (Robinson and Metternicht, 2006). The best cross-validation parameters for IDW, splines and local interpolation are shown in Tab. 3.

Tab. 3 Parameters returning the lowest RMSE for IDW, Splines and Local Polynomial

Interpolation for top soil (0-30 cm) and subsoil (30-60 cm) Soil pH Power Neighboor ME RMSE

Topsoil (0-30 cm) IDW 4 10 -0.005068 0.3461

Splines 1 10 -0.01349 0.3846 Local 2 10 0.01377 0.3625

Subsoil (30-60 cm) IDW 4 10 -0.0247 0.4123

Splines 1 10 -0.02135 0.431 Local 2 10 0.01069 0.4802

In all IDW tests, the best weighting parameter was found to be four. This suggests that the

weights diminish rapidly from the sample point over the chosen radius. In all cases for RBFs, the best exponent value was found to be one (completely regularized spline) suggesting that

lower order polynomials were sufficient at representing the variation on the field. The same neighborhood variation was used also for local interpolation.

The interpolated maps of soil pH with the lowest RMSE from the cross-validation process are presented in Fig. 1 and Fig. 2 for each method.

Fig. 1 Interpolated maps of pH in KCl for the depth 0-30 cm, created on a basis of the lowest

RMSE obtained after cross validation by deterministic interpolative techniques.

Fig. 2 Interpolated maps of pH in KCl transformed with square root transformation, for the depth 30-60 cm, created on a basis of the lowest RMSE obtained after cross validation by

deterministic interpolative techniques.

Obtained results in Tab. 3 suggest IDW as the most accurate technique among three techniques used for both topsoil and subsoil pH analysis.

In further analysis, our goal was to obtain the maps of soil pH with lower number of sampling points by use of this technique. In the Tab. 4 and Tab. 5 are presented the areas for each pH class obtained after interpolation with diminished number of sampling points. Maps were produced with 54, 34, 29, 24, 19 and 14 sampling points. The simple comparison of the area of pH classes obtained with lower number of points with those pH classes obtained with maximum observation points is given as a percentage difference.

Tab. 4 The area of different pH classes in topsoil (0-30 cm) obtained with IDW for lower number of sampling points used, also expressed as a percentage difference

pH class

A104 (m2)

A54 (m2) % A34

(m2) % A29 (m2) % A24

(m2) % A19 (m2) % A14

(m2) %

<4.5 5783.2 5924.3 2.4 5617.3 2.9 5893.9 1.9 5431.3 6.1 5514.4 4.6 5161.0 10.8 4.5-5 2151.1 2350.7 9.3 2540.4 18.1 2331.8 8.4 2633.2 22.4 3321.8 54.4 4470.3 107.8 5-5.5 1157.4 998.5 13.7 936.5 19.1 985.5 14.9 1330.0 14.9 1357.4 17.3 733.0 36.7 5.5-6 943.8 1165.6 23.5 772.6 18.1 814.6 13.7 597.2 36.7 513.9 45.6 457.5 51.5

>6 4230.7 3827.1 9.5 4399.2 4.0 4240.2 0.2 4274.4 1.0 3558.6 15.9 3444.3 18.6

RBF LPIDW

<4.5

4.5-5.0

5.0-5.5

5.5-6.0

>6.0

pH in KCl

IDW RBF LP

<4.5

4.5-5.0

5.0-5.5

5.5-6.0

>6.0

pH in KCl

4 - 70



being exact interpolators these methods could predict values higher and lower than measured maximum and minimum values. RBFs are formed in geostatistics around each data location. Splines consist of polynomials, which describe pieces of a line or surface, and they are fitted together so that they join smoothly (Webster and Oliver, 2001).

Local Polynomial interpolation (Schaum, 2008) is an inexact interpolator that predicts a value that is different from the measured value. It presents a kind of combination of global polynomial methods and moving average procedure. Local polynomial interpolation fits many polynomials, each within specified overlapping neighborhoods. This kind of interpolation provides surfaces that accounts for more local variation. Instead of using all data, like in global polynomial interpolation, it uses data within localized windows.

Statistical analysis and spatial predictions in this paper were conducted by using Geostatistical Wizard implemented in ArcGIS 9.2 version.

RESULTS AND DISCUSSION Data description is achieved through basic summary statistics, including means, medians,

variances and skewness. A statistical summary of the pH in 1N KCl on two depths is presented in Tab. 2.

Tab. 2 Summary statistics for topsoil (0-30 cm) and subsoil (30-60 cm) pH in KCl

Soil pH N Min Max Range Mean Median Var CV (%) Skewness Kurtosis

Topsoil (0-30 cm) 104 3.73 7.93 4.2 5.31 4.86 1.48 23.1 0.479 1.775

Subsoil (30-60 cm) 104 3.55 7.36 3.81 5.05 4.57 1.38 23.3 0.707 2.07

Cross-validation is commonly used to validate the accuracy of interpolation (Voltz and

Webster, 1990). It is achieved by eliminating information, generally one observation at a time, estimating the value at that location with the remaining data and then computing the difference between the actual and estimated value for each data location (Davis, 1987). For the comparison of different interpolation techniques, we examined the difference between the measured and predicted data by using the mean error and the root mean squared error (Robinson and Metternicht, 2006). The best cross-validation parameters for IDW, splines and local interpolation are shown in Tab. 3.

Tab. 3 Parameters returning the lowest RMSE for IDW, Splines and Local Polynomial

Interpolation for top soil (0-30 cm) and subsoil (30-60 cm) Soil pH Power Neighboor ME RMSE

Topsoil (0-30 cm) IDW 4 10 -0.005068 0.3461

Splines 1 10 -0.01349 0.3846 Local 2 10 0.01377 0.3625

Subsoil (30-60 cm) IDW 4 10 -0.0247 0.4123

Splines 1 10 -0.02135 0.431 Local 2 10 0.01069 0.4802

In all IDW tests, the best weighting parameter was found to be four. This suggests that the

weights diminish rapidly from the sample point over the chosen radius. In all cases for RBFs, the best exponent value was found to be one (completely regularized spline) suggesting that

lower order polynomials were sufficient at representing the variation on the field. The same neighborhood variation was used also for local interpolation.

The interpolated maps of soil pH with the lowest RMSE from the cross-validation process are presented in Fig. 1 and Fig. 2 for each method.

Fig. 1 Interpolated maps of pH in KCl for the depth 0-30 cm, created on a basis of the lowest

RMSE obtained after cross validation by deterministic interpolative techniques.

Fig. 2 Interpolated maps of pH in KCl transformed with square root transformation, for the depth 30-60 cm, created on a basis of the lowest RMSE obtained after cross validation by

deterministic interpolative techniques.

Obtained results in Tab. 3 suggest IDW as the most accurate technique among three techniques used for both topsoil and subsoil pH analysis.

In further analysis, our goal was to obtain the maps of soil pH with lower number of sampling points by use of this technique. In the Tab. 4 and Tab. 5 are presented the areas for each pH class obtained after interpolation with diminished number of sampling points. Maps were produced with 54, 34, 29, 24, 19 and 14 sampling points. The simple comparison of the area of pH classes obtained with lower number of points with those pH classes obtained with maximum observation points is given as a percentage difference.

Tab. 4 The area of different pH classes in topsoil (0-30 cm) obtained with IDW for lower number of sampling points used, also expressed as a percentage difference

pH class

A104 (m2)

A54 (m2) % A34

(m2) % A29 (m2) % A24

(m2) % A19 (m2) % A14

(m2) %

<4.5 5783.2 5924.3 2.4 5617.3 2.9 5893.9 1.9 5431.3 6.1 5514.4 4.6 5161.0 10.8 4.5-5 2151.1 2350.7 9.3 2540.4 18.1 2331.8 8.4 2633.2 22.4 3321.8 54.4 4470.3 107.8 5-5.5 1157.4 998.5 13.7 936.5 19.1 985.5 14.9 1330.0 14.9 1357.4 17.3 733.0 36.7 5.5-6 943.8 1165.6 23.5 772.6 18.1 814.6 13.7 597.2 36.7 513.9 45.6 457.5 51.5

>6 4230.7 3827.1 9.5 4399.2 4.0 4240.2 0.2 4274.4 1.0 3558.6 15.9 3444.3 18.6

RBF LPIDW

<4.5

4.5-5.0

5.0-5.5

5.5-6.0

>6.0

pH in KCl

IDW RBF LP

<4.5

4.5-5.0

5.0-5.5

5.5-6.0

>6.0

pH in KCl

4 - 71



This kind of analysis was found to be not very reliable due to the fact that the obtained results expressed as a percentage of difference among classes depends on the number of sampling points included in each class, and if the number of sampling points is very low, the borders among classes will be narrower. Nevertheless, for the two major and grapevine growth and development limiting classes in our experimental field, namely, pH <4.5 and pH >6, which sum 70% of total area of topsoil, the percentage of difference seems to be quite reasonable, being lower than 10% in 54, 34, 29 and 24 sampling points maps. The map produced with 29 points is if relatively expressed more accurate than those produced with 34 and 54 points. The maps produced with 14 and 19 sampling points cannot be considered reliable.

Regarding subsoil results (Tab. 5), two major classes (pH <4.5 and pH >6), take 71% of total area, and the same behavior was obtained for 54, 34, 29 and 24 samples, meaning that 24 samples are adequate to represent these classes. The presentation of other classes is acceptable with 34 sampling points, meaning almost 1/3 of maximum. Tab. 5 The area of different pH classes (in m2) in subsoil (30-60 cm) obtained with IDW for

lower number of sampling points used, also expressed as a percentage difference

pH class

B104 (m2)

B54 (m2) % B34

(m2) % B29 (m2) % B24

(m2) % B19 (m2) % B14

(m2) %

<4.5 7157.1 7055.6 1.4 7081.7 1.1 7088.2 1.0 6623.3 7.5 7102.1 0.8 7612.4 6.4 4.5-5 1635.5 1912.1 16.9 1604.5 1.9 1735.7 6.1 2107.3 28.8 2220.5 35.8 2596.6 58.8 5-5.5 1354.7 1318.6 2.7 1185.4 12.5 1320.7 2.5 1380.3 1.9 1066.4 21.3 478.3 64.7 5.5-6 1110.5 1105.9 0.4 1215.3 9.4 1460.6 31.5 1199.1 8.0 504.2 54.6 450.6 59.4

>6 3008.3 2873.9 4.5 3179.1 5.7 2660.9 11.5 2956.0 1.7 3372.9 12.1 3128.2 4.0

CONCLUSIONS In this paper, we tried to express the estimation accuracy of interpolative methods as a

relative difference in the area of pH classes obtained with lower number of points to those obtained with maximum number of sampling points. A decline in number of sampling points that we used to produce maps of soil pH indicated that adequate maps could be created with 29 samples in topsoil depth and with 34 samples in subsoil depth. The higher number of sampling points used for map creation does not seem to be an accuracy advantage if expressed in relative units.

Studies that involve soil sampling following a regular grid are limited to some hectares, and a similar approach used for larger areas is time consuming and costly. A balance must be found among the scientific objective, human and monetary resources, and time. The results of this study are very important because the optimal number of samples which is obtained to be one third of maximum for both soil depths should be a guide for further analysis on the field of ES “Radmilovac” which consists of more tenths of hectares of all type of cultivation, including 12 hectares of grapevine that should be investigated in future.

ACKNOWLEDGMENTS This research was part of project: «Organic production of grape and wine and all grapevine

products» and is partially sponsored by the Republic of Serbia, Ministry of Science, Technologies and Development under grant no TR-20093.

BIBLIOGRAPHY

Cerling, V.V. (1990). Diagnostika patini seljsko-hozalistcennvih kultur. Moskva-Leningrad: Agropromizdat.

Davis, B.M. (1987). Uses and abuses of cross-validation in geostatistics. Math. Geol. 19, 241–248.

Gotway, C.A., Ferguson, R.B., Hergert,G.W. and Peterson, T.A. (1996). Comparison of kriging and inverse-distance methods for mapping soil parameters. Am. J. Soil Sci. 60, 1237–1247.

Kravchenko, A.N. and Bullock, D.G. (1999). A comparative study of interpolation methods for mapping soil properties. J. Agronomy 91, 393–400.

Lark, R.M., and Bolam, H.C. (1997). Uncertainty in prediction and interpretation of spatially variable data on soils. Geoderma 77: 263-282.

Laslett, G.M., McBratney, A.B., Pahl, P.J. and Hutchinson, M.F. (1987). Comparison of several spatial prediction methods for soil pH. J. Soil Sci. 38, 325–341.

Li Yan, Shi Zhou, Wu Ci-fang, Li Hong-yi, and Li Feng (2007). Improved Prediction and Reduction of Sampling Density for Soil Salinity by Different Geostatistical Methods Agricultural Sciences in China, Vol. 6 (7): 832-841.

Robinson, T.P. and Metternicht, G. (2006). Testing the performance of spatial interpolation techniques for mapping soil properties. Computers and Electronics in Agriculture 50: 97–108.

Shepard, D. (1968). A two-dimensional interpolation function for irregularly-spaced data, Proc. 23rd National Conference ACM, ACM, 517-524.

Schaum, A. (2008). Principles of local polynomial interpolation, aipr, pp.1-6, 37th IEEE Applied Imagery Pattern Recognition Workshop.

Trangmar, B.B., Yost,R.S., & Uehara, G. (1985). Application of geostatistics to spatial studies of soil properties. Advancement in Agronomy, 38, 45–94.

Voltz, M., Webster, R. (1990). A comparison of kriging, cubic splines and classification for predicting soil properties from sample information. J.Soil Sci. 41, 473–490.

Vuković, А., Đurđević V., Petrović N., Sivčev B., Ranković-Vasić Z. (2009). Simulation of climate changes for Europe with special analysis for important vineyard areas of Serbia, Proceeding 32nd World Congress of Vine and Wine, pp. 47-48. www.oiv2009.hr

Weber, D. and Englund, E. (1992). Evaluation and comparison of spatial interpolators. Math. Geol. 24, 381–391.

Webster, R. and Oliver, M.A. (2001). Geostatistics for Environmental Scientists. John Wiley and Sons, Brisbane, Australia.

4 - 72



This kind of analysis was found to be not very reliable due to the fact that the obtained results expressed as a percentage of difference among classes depends on the number of sampling points included in each class, and if the number of sampling points is very low, the borders among classes will be narrower. Nevertheless, for the two major and grapevine growth and development limiting classes in our experimental field, namely, pH <4.5 and pH >6, which sum 70% of total area of topsoil, the percentage of difference seems to be quite reasonable, being lower than 10% in 54, 34, 29 and 24 sampling points maps. The map produced with 29 points is if relatively expressed more accurate than those produced with 34 and 54 points. The maps produced with 14 and 19 sampling points cannot be considered reliable.

Regarding subsoil results (Tab. 5), two major classes (pH <4.5 and pH >6), take 71% of total area, and the same behavior was obtained for 54, 34, 29 and 24 samples, meaning that 24 samples are adequate to represent these classes. The presentation of other classes is acceptable with 34 sampling points, meaning almost 1/3 of maximum. Tab. 5 The area of different pH classes (in m2) in subsoil (30-60 cm) obtained with IDW for

lower number of sampling points used, also expressed as a percentage difference

pH class

B104 (m2)

B54 (m2) % B34

(m2) % B29 (m2) % B24

(m2) % B19 (m2) % B14

(m2) %

<4.5 7157.1 7055.6 1.4 7081.7 1.1 7088.2 1.0 6623.3 7.5 7102.1 0.8 7612.4 6.4 4.5-5 1635.5 1912.1 16.9 1604.5 1.9 1735.7 6.1 2107.3 28.8 2220.5 35.8 2596.6 58.8 5-5.5 1354.7 1318.6 2.7 1185.4 12.5 1320.7 2.5 1380.3 1.9 1066.4 21.3 478.3 64.7 5.5-6 1110.5 1105.9 0.4 1215.3 9.4 1460.6 31.5 1199.1 8.0 504.2 54.6 450.6 59.4

>6 3008.3 2873.9 4.5 3179.1 5.7 2660.9 11.5 2956.0 1.7 3372.9 12.1 3128.2 4.0

CONCLUSIONS In this paper, we tried to express the estimation accuracy of interpolative methods as a

relative difference in the area of pH classes obtained with lower number of points to those obtained with maximum number of sampling points. A decline in number of sampling points that we used to produce maps of soil pH indicated that adequate maps could be created with 29 samples in topsoil depth and with 34 samples in subsoil depth. The higher number of sampling points used for map creation does not seem to be an accuracy advantage if expressed in relative units.

Studies that involve soil sampling following a regular grid are limited to some hectares, and a similar approach used for larger areas is time consuming and costly. A balance must be found among the scientific objective, human and monetary resources, and time. The results of this study are very important because the optimal number of samples which is obtained to be one third of maximum for both soil depths should be a guide for further analysis on the field of ES “Radmilovac” which consists of more tenths of hectares of all type of cultivation, including 12 hectares of grapevine that should be investigated in future.

ACKNOWLEDGMENTS This research was part of project: «Organic production of grape and wine and all grapevine

products» and is partially sponsored by the Republic of Serbia, Ministry of Science, Technologies and Development under grant no TR-20093.

BIBLIOGRAPHY

Cerling, V.V. (1990). Diagnostika patini seljsko-hozalistcennvih kultur. Moskva-Leningrad: Agropromizdat.

Davis, B.M. (1987). Uses and abuses of cross-validation in geostatistics. Math. Geol. 19, 241–248.

Gotway, C.A., Ferguson, R.B., Hergert,G.W. and Peterson, T.A. (1996). Comparison of kriging and inverse-distance methods for mapping soil parameters. Am. J. Soil Sci. 60, 1237–1247.

Kravchenko, A.N. and Bullock, D.G. (1999). A comparative study of interpolation methods for mapping soil properties. J. Agronomy 91, 393–400.

Lark, R.M., and Bolam, H.C. (1997). Uncertainty in prediction and interpretation of spatially variable data on soils. Geoderma 77: 263-282.

Laslett, G.M., McBratney, A.B., Pahl, P.J. and Hutchinson, M.F. (1987). Comparison of several spatial prediction methods for soil pH. J. Soil Sci. 38, 325–341.

Li Yan, Shi Zhou, Wu Ci-fang, Li Hong-yi, and Li Feng (2007). Improved Prediction and Reduction of Sampling Density for Soil Salinity by Different Geostatistical Methods Agricultural Sciences in China, Vol. 6 (7): 832-841.

Robinson, T.P. and Metternicht, G. (2006). Testing the performance of spatial interpolation techniques for mapping soil properties. Computers and Electronics in Agriculture 50: 97–108.

Shepard, D. (1968). A two-dimensional interpolation function for irregularly-spaced data, Proc. 23rd National Conference ACM, ACM, 517-524.

Schaum, A. (2008). Principles of local polynomial interpolation, aipr, pp.1-6, 37th IEEE Applied Imagery Pattern Recognition Workshop.

Trangmar, B.B., Yost,R.S., & Uehara, G. (1985). Application of geostatistics to spatial studies of soil properties. Advancement in Agronomy, 38, 45–94.

Voltz, M., Webster, R. (1990). A comparison of kriging, cubic splines and classification for predicting soil properties from sample information. J.Soil Sci. 41, 473–490.

Vuković, А., Đurđević V., Petrović N., Sivčev B., Ranković-Vasić Z. (2009). Simulation of climate changes for Europe with special analysis for important vineyard areas of Serbia, Proceeding 32nd World Congress of Vine and Wine, pp. 47-48. www.oiv2009.hr

Weber, D. and Englund, E. (1992). Evaluation and comparison of spatial interpolators. Math. Geol. 24, 381–391.

Webster, R. and Oliver, M.A. (2001). Geostatistics for Environmental Scientists. John Wiley and Sons, Brisbane, Australia.

4 - 73



RISPOSTE ENOLOGICHE DEL NERO D’AVOLA SU SUOLI A DIVERSO GRADO DI SALINITA’

Antonio Sparacio1, Giuseppe Genna1, Leo Prinzivalli1, Salvatore Sparla1, Vincenzo Melia1, Salvatore Raimondi2, Antonella Verzera3

(1) Istituto Regionale della Vite e del Vino – Via Libertà 66, Palermo – Italia [email protected](2) DAAT – Università degli Studi di Palermo- Viale delle Scienze, Palermo – Italia [email protected] (3) DCOB - Università degli Studi di Messina – Salita Sperone 31, Messina – Italia [email protected]

RIASSUNTO Vengono riportati i risultati enologici di uno studio condotto sul Nero d’Avola in un tipico

ambiente viticolo siciliano, in cui insistono suoli che presentano un diverso grado di salinità. La salinità di un suolo è il tenore in sali solubili presenti in un terreno. I Sali sono indispensabili per la vita delle piante, ma se la loro quantità è elevata può pregiudicarne la vita. Un suolo si definisce salino quando il valore della conduttività elettrica dell’estratto acquoso a saturazione è pari o superiore a 4. La conduttività elettrica (ECe) è direttamente proporzionale al contenuto di sali solubili. In Sicilia i suoli “affetti” da salinità occupano un’area di 600.000 ettari, concentrati principalmente nella Sicilia centro meridionale ed in parte in quella occidentale. La prova sperimentale si è svolta in un’azienda viticola ubicata nel comune di Santa Margherita Belice (AG) a 280 m. slm, in un vigneto di Nero d’Avola, allevato a controspalliera. La caratteristica di questo vigneto è quella avere lungo i filari, che dall’alto vanno verso il basso, un diverso tenore di contenuto salino tanto che è stato possibile impostare tre differenti tesi. Alla vendemmia le uve delle singole tesi sono state vinificate, presso la cantina sperimentale dell’IRVV, adottando un identico protocollo di trasformazione per non interferire sulla qualità finale dei prodotti. Per verificare eventuali differenze nei vini delle diverse tesi, sono stati determinati i parametri analitici più importanti, tra cui i polifenoli, gli antociani, i flavonoidi, la componente minerale, ecc. Sono state effettuate, inoltre, le analisi strumentali qualitative e quantitative dei composti volatili responsabili della componente aromatica.

PAROLE CHIAVE Nero d’Avola – Sicilia - suoli salini - salinità

ABSTRACT We show the results of a study on Nero d'Avola in a typical Sicilian environment, with soil

at different salinity. The salinity of soil is its content of soluble salts. The salts are essential for plant life, but high quantity can affect negatively. A soil is defined saline as the value of electrical conductivity of the aqueous extract at saturation is equal to or greater than 4. Electrical conductivity (ECe) is directly proportional to the content of soluble salts. In Sicily, the land "affected" by salinity have an area of 600,000 hectares, concentrated mainly in central southern Sicily and partly in the west. The experimental test was conducted in the municipality of Santa Margherita Belice (AG) at 280 m. asl, in a vineyard of Nero d'Avola, trained in espalier. The characteristic of this vineyard is to have along the rows which concentration of salt content changes so that it was possible to set three different thesis. At harvest the grapes of each thesis were fermented in the experimental winery of IRVV by identical protocol processing for not interfering on the quality of final products. To verify possible differences in the wines of various thesis, the most important analytical parameters have been determined, including polyphenols, anthocyanins, flavonoids, the mineral

component, etc. We realize also instrumental qualitative and quantitative analysis of volatile compounds responsible for flavor component.

KEYWORD Nero d’Avola – Sicily - salinity

INTRODUZIONE La salinità di un suolo è il tenore in sali solubili presenti in un terreno (vengono definiti sali

solubili tutti i composti chimici caratterizzati da solubilità più elevata di quella del gesso). I sali sono indispensabili per la vita delle piante, ma se la loro quantità nel terreno è elevata, può pregiudicarne la vita. Un suolo si definisce salino quando il valore della conduttività elettrica dell’estratto acquoso a saturazione è pari o superiore a 4. La conduttività elettrica (ECe) è direttamente proporzionale al contenuto di sali solubili. L’eccesso di salinità può provocare essiccamento fisiologico dei vegetali, aumento della resistenza idraulica delle radici e delle foglie, alterazione del contenuto di ormoni, danneggiamento diretto del processo di fotosintesi, ecc. In Sicilia i suoli affetti da salinità occupano un’area di 600.000 ettari, concentrati principalmente nella Sicilia centro meridionale ed in parte in quella occidentale

(figura 1). La bibliografia scientifica relativa al comportamento della vite sui suoli salini è abbastanza carente, pertanto si è pensato di effettuare uno studio per verificare l’influenza della salinità sulla qualità delle produzioni vitivinicole, con lo scopo anche di valorizzare una produzione di vini specifici legati a particolari ambienti pedo-climatici.

Figura 1 - I suoli salini in Sicilia

MATERIALI E METODI Le prove sperimentali si sono svolte nel biennio 2007-2008 in un’azienda viticola ubicata

nel comune di Santa Margherita Belice (AG) a 280 m. slm; è stato scelto un vigneto di Nero d’Avola, allevato a controspalliera con potatura a cordone speronato, in leggera pendenza con esposizione a sud-est. La caratteristica di questo vigneto è quella avere lungo i filari, che dall’alto vanno verso il basso, un diverso tenore di contenuto salino (grafico 1) tanto che è stato possibile impostare tre differenti tesi: Tesi 1: contenuto salino trascurabile (“test” - valore medio dei primi 105 cm. ECe 0,7 dS m-1) Tesi 2: contenuto salino medio (“mediamente salino” - ECe dei primi 55 cm. 1,2 dS m-1, da 55 a 105 cm. 2,1 dS m-1) Tesi 3: contenuto salino forte (“salino” - ECe dei primi 55 cm. 1,0 dS m-1, da 55 a 105 cm. 7,6 dS m-1) Alla vendemmia le uve delle singole tesi sono state raccolte in cassette e trasferite presso la cantina sperimentale dell’IRVV per la vinificazione. Per non interferire sulla qualità finale dei prodotti, è stato adottato un identico protocollo di trasformazione delle uve per le tre tesi che comprende: raccolta manuale delle uve in cassette, pigiadiraspatura, aggiunta di 5 g/hl di SO2, inoculo di lieviti selezionati, fermentazione a temperatura controllata (+28 °C.) con tre follature al giorno, svinatura e pressatura delle vinacce, travasi (almeno 2) ed imbottigliamento. Per verificare eventuali differenze nei vini delle diverse tesi, sono stati

4 - 74



RISPOSTE ENOLOGICHE DEL NERO D’AVOLA SU SUOLI A DIVERSO GRADO DI SALINITA’

Antonio Sparacio1, Giuseppe Genna1, Leo Prinzivalli1, Salvatore Sparla1, Vincenzo Melia1, Salvatore Raimondi2, Antonella Verzera3

(1) Istituto Regionale della Vite e del Vino – Via Libertà 66, Palermo – Italia [email protected](2) DAAT – Università degli Studi di Palermo- Viale delle Scienze, Palermo – Italia [email protected] (3) DCOB - Università degli Studi di Messina – Salita Sperone 31, Messina – Italia [email protected]

RIASSUNTO Vengono riportati i risultati enologici di uno studio condotto sul Nero d’Avola in un tipico

ambiente viticolo siciliano, in cui insistono suoli che presentano un diverso grado di salinità. La salinità di un suolo è il tenore in sali solubili presenti in un terreno. I Sali sono indispensabili per la vita delle piante, ma se la loro quantità è elevata può pregiudicarne la vita. Un suolo si definisce salino quando il valore della conduttività elettrica dell’estratto acquoso a saturazione è pari o superiore a 4. La conduttività elettrica (ECe) è direttamente proporzionale al contenuto di sali solubili. In Sicilia i suoli “affetti” da salinità occupano un’area di 600.000 ettari, concentrati principalmente nella Sicilia centro meridionale ed in parte in quella occidentale. La prova sperimentale si è svolta in un’azienda viticola ubicata nel comune di Santa Margherita Belice (AG) a 280 m. slm, in un vigneto di Nero d’Avola, allevato a controspalliera. La caratteristica di questo vigneto è quella avere lungo i filari, che dall’alto vanno verso il basso, un diverso tenore di contenuto salino tanto che è stato possibile impostare tre differenti tesi. Alla vendemmia le uve delle singole tesi sono state vinificate, presso la cantina sperimentale dell’IRVV, adottando un identico protocollo di trasformazione per non interferire sulla qualità finale dei prodotti. Per verificare eventuali differenze nei vini delle diverse tesi, sono stati determinati i parametri analitici più importanti, tra cui i polifenoli, gli antociani, i flavonoidi, la componente minerale, ecc. Sono state effettuate, inoltre, le analisi strumentali qualitative e quantitative dei composti volatili responsabili della componente aromatica.

PAROLE CHIAVE Nero d’Avola – Sicilia - suoli salini - salinità

ABSTRACT We show the results of a study on Nero d'Avola in a typical Sicilian environment, with soil

at different salinity. The salinity of soil is its content of soluble salts. The salts are essential for plant life, but high quantity can affect negatively. A soil is defined saline as the value of electrical conductivity of the aqueous extract at saturation is equal to or greater than 4. Electrical conductivity (ECe) is directly proportional to the content of soluble salts. In Sicily, the land "affected" by salinity have an area of 600,000 hectares, concentrated mainly in central southern Sicily and partly in the west. The experimental test was conducted in the municipality of Santa Margherita Belice (AG) at 280 m. asl, in a vineyard of Nero d'Avola, trained in espalier. The characteristic of this vineyard is to have along the rows which concentration of salt content changes so that it was possible to set three different thesis. At harvest the grapes of each thesis were fermented in the experimental winery of IRVV by identical protocol processing for not interfering on the quality of final products. To verify possible differences in the wines of various thesis, the most important analytical parameters have been determined, including polyphenols, anthocyanins, flavonoids, the mineral

component, etc. We realize also instrumental qualitative and quantitative analysis of volatile compounds responsible for flavor component.

KEYWORD Nero d’Avola – Sicily - salinity

INTRODUZIONE La salinità di un suolo è il tenore in sali solubili presenti in un terreno (vengono definiti sali

solubili tutti i composti chimici caratterizzati da solubilità più elevata di quella del gesso). I sali sono indispensabili per la vita delle piante, ma se la loro quantità nel terreno è elevata, può pregiudicarne la vita. Un suolo si definisce salino quando il valore della conduttività elettrica dell’estratto acquoso a saturazione è pari o superiore a 4. La conduttività elettrica (ECe) è direttamente proporzionale al contenuto di sali solubili. L’eccesso di salinità può provocare essiccamento fisiologico dei vegetali, aumento della resistenza idraulica delle radici e delle foglie, alterazione del contenuto di ormoni, danneggiamento diretto del processo di fotosintesi, ecc. In Sicilia i suoli affetti da salinità occupano un’area di 600.000 ettari, concentrati principalmente nella Sicilia centro meridionale ed in parte in quella occidentale

(figura 1). La bibliografia scientifica relativa al comportamento della vite sui suoli salini è abbastanza carente, pertanto si è pensato di effettuare uno studio per verificare l’influenza della salinità sulla qualità delle produzioni vitivinicole, con lo scopo anche di valorizzare una produzione di vini specifici legati a particolari ambienti pedo-climatici.

Figura 1 - I suoli salini in Sicilia

MATERIALI E METODI Le prove sperimentali si sono svolte nel biennio 2007-2008 in un’azienda viticola ubicata

nel comune di Santa Margherita Belice (AG) a 280 m. slm; è stato scelto un vigneto di Nero d’Avola, allevato a controspalliera con potatura a cordone speronato, in leggera pendenza con esposizione a sud-est. La caratteristica di questo vigneto è quella avere lungo i filari, che dall’alto vanno verso il basso, un diverso tenore di contenuto salino (grafico 1) tanto che è stato possibile impostare tre differenti tesi: Tesi 1: contenuto salino trascurabile (“test” - valore medio dei primi 105 cm. ECe 0,7 dS m-1) Tesi 2: contenuto salino medio (“mediamente salino” - ECe dei primi 55 cm. 1,2 dS m-1, da 55 a 105 cm. 2,1 dS m-1) Tesi 3: contenuto salino forte (“salino” - ECe dei primi 55 cm. 1,0 dS m-1, da 55 a 105 cm. 7,6 dS m-1) Alla vendemmia le uve delle singole tesi sono state raccolte in cassette e trasferite presso la cantina sperimentale dell’IRVV per la vinificazione. Per non interferire sulla qualità finale dei prodotti, è stato adottato un identico protocollo di trasformazione delle uve per le tre tesi che comprende: raccolta manuale delle uve in cassette, pigiadiraspatura, aggiunta di 5 g/hl di SO2, inoculo di lieviti selezionati, fermentazione a temperatura controllata (+28 °C.) con tre follature al giorno, svinatura e pressatura delle vinacce, travasi (almeno 2) ed imbottigliamento. Per verificare eventuali differenze nei vini delle diverse tesi, sono stati

4 - 75



determinati, adottando le metodiche ufficiali di analisi, i parametri analitici più importanti, tra cui anche i polifenoli, gli antociani, i flavonoidi, la componente minerale, ecc. Sui vini sono state effettuate, inoltre, le analisi strumentali qualitative e quantitative dei composti volatili responsabili dell’aroma.

RISULTATI E DISCUSSIONE I parametri analitici dei mosti (tabella 1) delle singole tesi non mostrano accentuate

differenze nella gradazione zuccherina e nell’acidità titolabile, delle differenze più significative, invece, ci sono nel pH. Nei vini delle diverse tesi esistono differenze abbastanza significative dei dati relativi ai polifenoli, agli antociani, ai flavonoidi ed all’intensità colorante; quest’ultimi parametri risultano più alti nella tesi con contenuto salino forte. Il contenuto salino dei suoli ha influenzato in maniera decisiva la componente minerale dei vini (grafico2.). In particolare tra i diversi parametri quello che ha subito una certa variazione fra le diverse tesi è il dato relativo ai solfati, con valori più bassi nel test e valori più elevati nel vino della tesi “salino”; tale andamento è simile nelle due annate di osservazione anche se con valori differenti. Le analisi sulla componente volatile dei vini sono state effettuate sui campioni di entrambe le vendemmie (tabella 2). Nei vini del 2007 i campioni delle tesi “mediamente salino” e “salino” presentano valori simili degli esteri e comunque più elevati rispetto al test. La tesi “salino” presenta una quantità superiore (dati non riportati) di succinato di dietile (delicato odore di frutta), mentre il campione “mediamente salino” si differenzia per un quantitativo più elevato (dati non riportati) di esanoato di etile (fruttato di mela). Anche gli alcoli presentano valori più alti nelle tesi “salino” e “mediamente salino”. La compente terpenica è più elevata nel campione della seconda tesi e più basso nel test; in questa classe sono stati rilevati (dati non riportati) il limonene (solo nel “salino”), il terpinolene (solo nel “mediamente salino”), il ß-linalolo (costante nei tre campioni), mentre il (Z)-nerolidolo (delicatamente erbaceo, floreale) presenta quantità che diminuiscono dal “salino” al “test”. Anche nei campioni del 2008 è possibile riscontrare una notevole differenza nei contenuti di esteri, alcoli, acidi e terpeni delle tesi “salino” e “mediamente salino”, in genere più elevati rispetto al test. Da un punto di vista aromatico, e per le due annate di osservazione, si può senza dubbio affermare che i vini delle tesi “mediamente salino” e “salino” sono quelli più apprezzati, mentre quelli relativi al testimone risultano di aroma complessivamente meno intenso ed armonico. CONCLUSIONI

Alla luce dei dati rilevati sui vini frutto di questa sperimentazione biennale, è possibile tracciare un quadro abbastanza preciso sui risultati ottenuti. Appare chiaro come la salinità, che caratterizza in modo particolare i suoli di quest’areale viticolo siciliano, influenzi decisamente il comportamento e le risposte enologiche del Nero d’Avola. Infatti i vini che si ottengono si caratterizzano per avere dei parametri analitici “migliori” rispetto a quelli ottenuti nel terreno in cui il livello di salinità è basso; anche il profilo aromatico di questi vini, come si evince dai dati strumentali sulla componente volatile, è senza dubbio più complesso, più intenso ed armonico. La componente minerale dei vini, più elevata nei campioni ottenuti da suoli con media ed alta salinità, può influenzare positivamente il loro quadro organolettico; infatti le analisi sensoriali effettuate sui campioni ottenuti nella vendemmia 2007, i cui dati sono riportati in un precedente lavoro (Sparacio et al., 2009), evidenziano un particolare gradimento della componente gustativa da parte del gruppo di assaggiatori. Considerato che su questi suoli il Nero d’Avola ha dato dei risultati incoraggianti, soprattutto per quel che riguarda le caratteristiche dei vini che si possono produrre, resta da verificare fino a quali livelli di salinità è possibile riuscire a produrre a determinati livelli qualitativi

0

2

4

6

8

10

10 25 50 75 100

profondità (cm)

ECe

dS m

-1

tesi 1

tesi 2

tesi 3

senza interferire negativamente sulla vita delle piante. In questo senso risulta di particolare importanza l’ausilio dell’irrigazione, da effettuare con acque “dolci” per lisciviare i sali che in estate, per effetto di elevati livelli di evapotraspirazione, tendono ad aumentare la loro concentrazione nel terreno.

RINGRAZIAMENTI Si ringrazia la Dr.ssa Paola Catanzaro del laboratorio centrale dell’IRVV per le analisi

relative alla componente minerale dei vini.

BIBLIOGRAFIA - Chapman V., (1966). Salinity and Acidity. H. Boyko (ed.), Junk Publ. The Hague, Netherlands, 23-42. - Fregoni M., (1998). Viticoltura di qualità. Edizioni L’Informatore Agrario - Moolman J.H., (1983). The effect of irrigation practices in the Bree River valley on the salt content of a small river. Irr. Sci., 4.- Sparacio A. et al., (2009). Suoli salini e qualità del Nero d’Avola. In Atti Enoforum, Piacenza, Sive. - Verzera A. et al. The influence of the soil salinity on the sensory characteristic and volatile aroma compounds of the “Nero d’Avola” wine. In corso di pubblicazione

Grafico 1 - Andamento della salinità nel suolo

Tabella 1 -Alcuni parametri analitici delle diverse tesi Mosto Vino

Babo

Acid. Tit. (g/l)

pH Alcool %Acid. Tit. (g/l)

Acido Tartarico

(g/l)

Polifenoli totali (mg/l)

Antociani (mg/l)

Flavonoidi (mg/l)

Intensità colorante Tonalità

Test 17,0 7,0 3,13 12,0 6 3,19 1336 270 852 7,84 0,53 Med. Salino 17,5 7,1 3,03 12,5 6,7 3,61 1491 295 960 9,34 0,44

2007

Salino 17,6 7,2 3,29 12,5 6,4 4,16 1659 329 1165 9,56 0,43 Test 16,8 7,7 3,32 11,6 5,3 3,60 1342 248 1130 7,49 0,48 Med. Salino 17,3 6,0 3,31 11,7 6,2 3,80 1309 261 1153 7,64 0,42

2008

Salino 17,5 6,6 3,41 12,1 5,1 2,90 1622 281 1378 8,81 0,41

4 - 76



determinati, adottando le metodiche ufficiali di analisi, i parametri analitici più importanti, tra cui anche i polifenoli, gli antociani, i flavonoidi, la componente minerale, ecc. Sui vini sono state effettuate, inoltre, le analisi strumentali qualitative e quantitative dei composti volatili responsabili dell’aroma.

RISULTATI E DISCUSSIONE I parametri analitici dei mosti (tabella 1) delle singole tesi non mostrano accentuate

differenze nella gradazione zuccherina e nell’acidità titolabile, delle differenze più significative, invece, ci sono nel pH. Nei vini delle diverse tesi esistono differenze abbastanza significative dei dati relativi ai polifenoli, agli antociani, ai flavonoidi ed all’intensità colorante; quest’ultimi parametri risultano più alti nella tesi con contenuto salino forte. Il contenuto salino dei suoli ha influenzato in maniera decisiva la componente minerale dei vini (grafico2.). In particolare tra i diversi parametri quello che ha subito una certa variazione fra le diverse tesi è il dato relativo ai solfati, con valori più bassi nel test e valori più elevati nel vino della tesi “salino”; tale andamento è simile nelle due annate di osservazione anche se con valori differenti. Le analisi sulla componente volatile dei vini sono state effettuate sui campioni di entrambe le vendemmie (tabella 2). Nei vini del 2007 i campioni delle tesi “mediamente salino” e “salino” presentano valori simili degli esteri e comunque più elevati rispetto al test. La tesi “salino” presenta una quantità superiore (dati non riportati) di succinato di dietile (delicato odore di frutta), mentre il campione “mediamente salino” si differenzia per un quantitativo più elevato (dati non riportati) di esanoato di etile (fruttato di mela). Anche gli alcoli presentano valori più alti nelle tesi “salino” e “mediamente salino”. La compente terpenica è più elevata nel campione della seconda tesi e più basso nel test; in questa classe sono stati rilevati (dati non riportati) il limonene (solo nel “salino”), il terpinolene (solo nel “mediamente salino”), il ß-linalolo (costante nei tre campioni), mentre il (Z)-nerolidolo (delicatamente erbaceo, floreale) presenta quantità che diminuiscono dal “salino” al “test”. Anche nei campioni del 2008 è possibile riscontrare una notevole differenza nei contenuti di esteri, alcoli, acidi e terpeni delle tesi “salino” e “mediamente salino”, in genere più elevati rispetto al test. Da un punto di vista aromatico, e per le due annate di osservazione, si può senza dubbio affermare che i vini delle tesi “mediamente salino” e “salino” sono quelli più apprezzati, mentre quelli relativi al testimone risultano di aroma complessivamente meno intenso ed armonico. CONCLUSIONI

Alla luce dei dati rilevati sui vini frutto di questa sperimentazione biennale, è possibile tracciare un quadro abbastanza preciso sui risultati ottenuti. Appare chiaro come la salinità, che caratterizza in modo particolare i suoli di quest’areale viticolo siciliano, influenzi decisamente il comportamento e le risposte enologiche del Nero d’Avola. Infatti i vini che si ottengono si caratterizzano per avere dei parametri analitici “migliori” rispetto a quelli ottenuti nel terreno in cui il livello di salinità è basso; anche il profilo aromatico di questi vini, come si evince dai dati strumentali sulla componente volatile, è senza dubbio più complesso, più intenso ed armonico. La componente minerale dei vini, più elevata nei campioni ottenuti da suoli con media ed alta salinità, può influenzare positivamente il loro quadro organolettico; infatti le analisi sensoriali effettuate sui campioni ottenuti nella vendemmia 2007, i cui dati sono riportati in un precedente lavoro (Sparacio et al., 2009), evidenziano un particolare gradimento della componente gustativa da parte del gruppo di assaggiatori. Considerato che su questi suoli il Nero d’Avola ha dato dei risultati incoraggianti, soprattutto per quel che riguarda le caratteristiche dei vini che si possono produrre, resta da verificare fino a quali livelli di salinità è possibile riuscire a produrre a determinati livelli qualitativi

0

2

4

6

8

10

10 25 50 75 100

profondità (cm)

ECe

dS m

-1

tesi 1

tesi 2

tesi 3

senza interferire negativamente sulla vita delle piante. In questo senso risulta di particolare importanza l’ausilio dell’irrigazione, da effettuare con acque “dolci” per lisciviare i sali che in estate, per effetto di elevati livelli di evapotraspirazione, tendono ad aumentare la loro concentrazione nel terreno.

RINGRAZIAMENTI Si ringrazia la Dr.ssa Paola Catanzaro del laboratorio centrale dell’IRVV per le analisi

relative alla componente minerale dei vini.

BIBLIOGRAFIA - Chapman V., (1966). Salinity and Acidity. H. Boyko (ed.), Junk Publ. The Hague, Netherlands, 23-42. - Fregoni M., (1998). Viticoltura di qualità. Edizioni L’Informatore Agrario - Moolman J.H., (1983). The effect of irrigation practices in the Bree River valley on the salt content of a small river. Irr. Sci., 4.- Sparacio A. et al., (2009). Suoli salini e qualità del Nero d’Avola. In Atti Enoforum, Piacenza, Sive. - Verzera A. et al. The influence of the soil salinity on the sensory characteristic and volatile aroma compounds of the “Nero d’Avola” wine. In corso di pubblicazione

Grafico 1 - Andamento della salinità nel suolo

Tabella 1 -Alcuni parametri analitici delle diverse tesi Mosto Vino

Babo

Acid. Tit. (g/l)

pH Alcool %Acid. Tit. (g/l)

Acido Tartarico

(g/l)

Polifenoli totali (mg/l)

Antociani (mg/l)

Flavonoidi (mg/l)

Intensità colorante Tonalità

Test 17,0 7,0 3,13 12,0 6 3,19 1336 270 852 7,84 0,53 Med. Salino 17,5 7,1 3,03 12,5 6,7 3,61 1491 295 960 9,34 0,44

2007

Salino 17,6 7,2 3,29 12,5 6,4 4,16 1659 329 1165 9,56 0,43 Test 16,8 7,7 3,32 11,6 5,3 3,60 1342 248 1130 7,49 0,48 Med. Salino 17,3 6,0 3,31 11,7 6,2 3,80 1309 261 1153 7,64 0,42

2008

Salino 17,5 6,6 3,41 12,1 5,1 2,90 1622 281 1378 8,81 0,41

4 - 77



Grafico 2 – Componente minerale dei vini

Tabella 2 - Frazione volatile dei vini delle diverse tesi (mg/L-1) Esteri Alcoli Acidi Terpeni

Test 970,57 129,55 5,93 0,67 Mediamente salino

1237,72 170,96 4,44 4,72 2007

Salino 1295,67 184,96 4,85 2,31 Test 1213,35 242,62 10,08 4,51 Mediamente salino

1878,55 301,82 27,63 8,77 2008

Salino 2082,91 291,81 18,01 8,36

635648

704

403426

469

35 32 29 27 33 36

135140

136129120

142

12 13 10 10 9 14

65 71 72 79 75 64

0

100

200

300

400

500

600

700

800

mg/

l

SOLFATI CLORURI Mg Na Ca

Test-2007

Med. Salino-2007

Salino-2007

Test-2008

Med. Salino-2008

Salino-2008

PROPOSAL OF ZONIFICATION AND CHARACTERIZATION OF TERROIRS IN THE YALDE-NAJERILLA-URUÑUELA VINE

GROWING AREA (DOC RIOJA, SPAIN), BASED ON THE SOIL INFLUENCE.

E. García-Escudero1, J. Mª. Martínez1, E. P. Pérez1, R. López1 and I. Martín11Servicio de Investigación y Desarrollo Tecnológico Agroalimentario (SIDTA-CIDA)-ICVV.

Ctra. Logroño-Mendavia NA-134 Km. 90. 26071 Logroño, La Rioja. (Spain). e-mail: [email protected], Tfno:+36-941291833

ABSTRACT Natural Terroir Units (NTU) are being delimited in vine growing area DOCa Rioja, in

collaboration with Uruñuela Cooperative, to characterized specific and singular Tempranillo (Vitis vinifera, L.) wines. NTU selection is based on detailed cartography (1:20.000), managed by the Soil Information System of La Rioja (SISR), and in the analysis of pedologic, climatic, lithologic, and relief features of Najerilla Valley.

The five NTU, placed on river and torrential platforms with similar lithology of original materials, have been selected with series of soils belong to the Alfisol, Inceptisol and Mollisol orders. The main purpose of this project is to measure the influence produced by soil properties of each series of soil (effective depth, water reserve, clay and carbonates percentage, potassium and magnesium) in musts and wines of this vine growing area.

KEYWORDSTerroir – soil – Tempranillo – grapevine - wine

INTRODUCTIONTerroir can be defined as an interactive ecosystem, in a given place, including climate, soil,

and the vine (rootstock and cultivar), (Seguin G.,1988; Van Leeuwen C. et al., 2006). The effect of climate was greatest on most parameters, followed by soil and cultivar. (Van Leeuwen et al., 2004).

As part of the characterization of wine terroirs, a proposal for the establishment of possible relationships between natural factors, especially soil, and the physico-chemical and organoleptic characteristics of wines made with Tempranillo (Vitis vinifera, L.) in the Uruñuela environment (La Rioja, Spain) is put forward. This municipality, with a vineyard area of 1,200 ha, is located in the lower Najerilla riverine, whose soils are gravels, cobbles and stones materials originating in parent silty-sand and sand matrix. The main pedogenetic processes relate partial translocation of carbonates and clay illuviation. The surface was modeled on the Neogene geological materials of Najera formation. Those original materials are covered with Quaternary deposits from the river (terrace) and torrential (alluvial fan and glacis) modeling.

The primary landforms are slopes and platforms, with altitudes ranging from 440 to 583 m.s.l., as a result of an intense water modeling.

The area is characterized by a dry Mediterranean climate, with semi-arid tendency, and a strong daily, monthly, seasonal and annual thermal oscillation. As a unit climate can be classified as temperate mesomediterranean (Papadakis). The average values of annual precipitation and temperature are 436 mm and 13.2 ºC respectively.

4 - 78



Grafico 2 – Componente minerale dei vini

Tabella 2 - Frazione volatile dei vini delle diverse tesi (mg/L-1) Esteri Alcoli Acidi Terpeni

Test 970,57 129,55 5,93 0,67 Mediamente salino

1237,72 170,96 4,44 4,72 2007

Salino 1295,67 184,96 4,85 2,31 Test 1213,35 242,62 10,08 4,51 Mediamente salino

1878,55 301,82 27,63 8,77 2008

Salino 2082,91 291,81 18,01 8,36

635648

704

403426

469

35 32 29 27 33 36

135140

136129120

142

12 13 10 10 9 14

65 71 72 79 75 64

0

100

200

300

400

500

600

700

800

mg/

l

SOLFATI CLORURI Mg Na Ca

Test-2007

Med. Salino-2007

Salino-2007

Test-2008

Med. Salino-2008

Salino-2008

PROPOSAL OF ZONIFICATION AND CHARACTERIZATION OF TERROIRS IN THE YALDE-NAJERILLA-URUÑUELA VINE

GROWING AREA (DOC RIOJA, SPAIN), BASED ON THE SOIL INFLUENCE.

E. García-Escudero1, J. Mª. Martínez1, E. P. Pérez1, R. López1 and I. Martín11Servicio de Investigación y Desarrollo Tecnológico Agroalimentario (SIDTA-CIDA)-ICVV.

Ctra. Logroño-Mendavia NA-134 Km. 90. 26071 Logroño, La Rioja. (Spain). e-mail: [email protected], Tfno:+36-941291833

ABSTRACT Natural Terroir Units (NTU) are being delimited in vine growing area DOCa Rioja, in

collaboration with Uruñuela Cooperative, to characterized specific and singular Tempranillo (Vitis vinifera, L.) wines. NTU selection is based on detailed cartography (1:20.000), managed by the Soil Information System of La Rioja (SISR), and in the analysis of pedologic, climatic, lithologic, and relief features of Najerilla Valley.

The five NTU, placed on river and torrential platforms with similar lithology of original materials, have been selected with series of soils belong to the Alfisol, Inceptisol and Mollisol orders. The main purpose of this project is to measure the influence produced by soil properties of each series of soil (effective depth, water reserve, clay and carbonates percentage, potassium and magnesium) in musts and wines of this vine growing area.

KEYWORDSTerroir – soil – Tempranillo – grapevine - wine

INTRODUCTIONTerroir can be defined as an interactive ecosystem, in a given place, including climate, soil,

and the vine (rootstock and cultivar), (Seguin G.,1988; Van Leeuwen C. et al., 2006). The effect of climate was greatest on most parameters, followed by soil and cultivar. (Van Leeuwen et al., 2004).

As part of the characterization of wine terroirs, a proposal for the establishment of possible relationships between natural factors, especially soil, and the physico-chemical and organoleptic characteristics of wines made with Tempranillo (Vitis vinifera, L.) in the Uruñuela environment (La Rioja, Spain) is put forward. This municipality, with a vineyard area of 1,200 ha, is located in the lower Najerilla riverine, whose soils are gravels, cobbles and stones materials originating in parent silty-sand and sand matrix. The main pedogenetic processes relate partial translocation of carbonates and clay illuviation. The surface was modeled on the Neogene geological materials of Najera formation. Those original materials are covered with Quaternary deposits from the river (terrace) and torrential (alluvial fan and glacis) modeling.

The primary landforms are slopes and platforms, with altitudes ranging from 440 to 583 m.s.l., as a result of an intense water modeling.

The area is characterized by a dry Mediterranean climate, with semi-arid tendency, and a strong daily, monthly, seasonal and annual thermal oscillation. As a unit climate can be classified as temperate mesomediterranean (Papadakis). The average values of annual precipitation and temperature are 436 mm and 13.2 ºC respectively.

4 - 79



Different areas, defined as Natural Terroir Units (NTU) (Carey et al., 2008), were delimited. The selection of NTUs was based on detailed mapping (1:20,000) which manages the Soil Information System of La Rioja (SISR), defining for the area 29 series and 23 map units (Soil Surface Staff, 2006). The climatic, soil, lithology and relief of the lower Najerilla riverine are also taken into account.

MATERIALS AND METHODS Based on their distribution and importance (Fig. 1), we considered five NTUs. In each of the

NTUs three representative vineyards were chosen (Fig. 2 and Tab. 1). To reduce variability attributable to variety for farm management factors, the study was focused on the Tempranillo variety, grafted on Richter 110 rootstock. The vineyards were in full production, with an age ranging from 10 to 25 years, and an average planting density 3,000 plants/ ha, with a deviation of ± 200. According to the characteristics of each NTU, the vineyard was trellised in VSP (“espalier”) (Royat double cordon system) or bush vines (“gobelet”).

Fig. 1. Soil mapping (1:20.000) of Uruñuela (DOCa Rioja-Spain).

In each plot, a "sampling unit" was bounded in which all surveys and measures were conducted. Monitored parameters were: • Vineyard nutritional status. Mineral composition (macro and micro-elements) of leaf

blades and petioles sampled at veraison. (García-Escudero E. et al., 2006). • Yield components, vigour indicators and vegetative-productive plant balance. At

harvest, grape yield, cluster number, average berry and cluster weight per vine were measured. Wood pruning per vine, shoot weight and Ravaz Index were assessed at post harvest.

• Ripening process. Winemaking. To determine the optimal date of harvest, sugar concentration (probable degree), acidity and colour components were monitored in musts during maturation on a weekly basis. For each plot a sample or 150 kg was harvested and then elaborated in 100 l. stainless steel vats. The alcoholic fermentation was conducted with active dry yeasts, at room temperature. Malolactic fermentation was induced by commercial lactic bacteria inoculation. Thus, fifteen wines were elaborated, from the experimental vineyards (three plots per NTU), in which pH, titratable acidity, tartaric acid, potassium,

alcoholic degree, colour intensity, tonality, total phenolic index and anthocyanins were determined. Wines were blindly tasted.

• Weather data. Climatic data were measured in the close vicinity with an automatic weather station (coordinates (zone 30) UTM X: 523,622, UTM Y: 4,701,100 and altitude: 465 m) management by Agroclimatic Information Service of La Rioja (SIAR).

Fig. 2. Map and topographic section across of selectioned plots in the delimited NTUs

Tab.1. Soil characteristics and representative soil profiles of Natural Terroir Units (NTUs). NTU 1 NTU 2 NTU 3 NTU 5 NTU 6

Classification (Soil Taxonomy, 2006)

Calcic Haploxeralfs, loamy-skeletal, mixed, mesic

Fluventic Haploxerepts,

loamy, mixed, mesic

Calcic Palexeralfs, clayey, mixed,

mesic

Typic Calcixerepts, loamy-skeletal,

carbonatic, mesic

Petrocalcic Palexerolls, loamy-skeletal, carbonatic,

mesic

Landscape unit Terrace II Najerilla river

Alluvial depth Yalde river

Terrace III Najerilla river Glacis IV Terrace IV Najerilla

river

Parent materials Gravels and cobbles in sand matrix

Sands, silts and clays

Gravels and cobbles in sand matrix

Gravels and cobbles in sandy loam matrix


Rock fragments (%) 40-80 5-10 15-40 15-40 40-80 Effective depth (cm) 110 150 90 115 50

Clay (%) 17-30 24-25 45-55 15 20 Carbonates (%) 3.5-5 3-7.5 40-50 45-60 55

Water availability Low Moderate-high Moderate-high Moderate-low Low

Representative soil profile

4 - 80



Different areas, defined as Natural Terroir Units (NTU) (Carey et al., 2008), were delimited. The selection of NTUs was based on detailed mapping (1:20,000) which manages the Soil Information System of La Rioja (SISR), defining for the area 29 series and 23 map units (Soil Surface Staff, 2006). The climatic, soil, lithology and relief of the lower Najerilla riverine are also taken into account.

MATERIALS AND METHODS Based on their distribution and importance (Fig. 1), we considered five NTUs. In each of the

NTUs three representative vineyards were chosen (Fig. 2 and Tab. 1). To reduce variability attributable to variety for farm management factors, the study was focused on the Tempranillo variety, grafted on Richter 110 rootstock. The vineyards were in full production, with an age ranging from 10 to 25 years, and an average planting density 3,000 plants/ ha, with a deviation of ± 200. According to the characteristics of each NTU, the vineyard was trellised in VSP (“espalier”) (Royat double cordon system) or bush vines (“gobelet”).

Fig. 1. Soil mapping (1:20.000) of Uruñuela (DOCa Rioja-Spain).

In each plot, a "sampling unit" was bounded in which all surveys and measures were conducted. Monitored parameters were: • Vineyard nutritional status. Mineral composition (macro and micro-elements) of leaf

blades and petioles sampled at veraison. (García-Escudero E. et al., 2006). • Yield components, vigour indicators and vegetative-productive plant balance. At

harvest, grape yield, cluster number, average berry and cluster weight per vine were measured. Wood pruning per vine, shoot weight and Ravaz Index were assessed at post harvest.

• Ripening process. Winemaking. To determine the optimal date of harvest, sugar concentration (probable degree), acidity and colour components were monitored in musts during maturation on a weekly basis. For each plot a sample or 150 kg was harvested and then elaborated in 100 l. stainless steel vats. The alcoholic fermentation was conducted with active dry yeasts, at room temperature. Malolactic fermentation was induced by commercial lactic bacteria inoculation. Thus, fifteen wines were elaborated, from the experimental vineyards (three plots per NTU), in which pH, titratable acidity, tartaric acid, potassium,

alcoholic degree, colour intensity, tonality, total phenolic index and anthocyanins were determined. Wines were blindly tasted.

• Weather data. Climatic data were measured in the close vicinity with an automatic weather station (coordinates (zone 30) UTM X: 523,622, UTM Y: 4,701,100 and altitude: 465 m) management by Agroclimatic Information Service of La Rioja (SIAR).

Fig. 2. Map and topographic section across of selectioned plots in the delimited NTUs

Tab.1. Soil characteristics and representative soil profiles of Natural Terroir Units (NTUs). NTU 1 NTU 2 NTU 3 NTU 5 NTU 6

Classification (Soil Taxonomy, 2006)

Calcic Haploxeralfs, loamy-skeletal, mixed, mesic

Fluventic Haploxerepts,

loamy, mixed, mesic

Calcic Palexeralfs, clayey, mixed,

mesic

Typic Calcixerepts, loamy-skeletal,

carbonatic, mesic

Petrocalcic Palexerolls, loamy-skeletal, carbonatic,

mesic

Landscape unit Terrace II Najerilla river

Alluvial depth Yalde river

Terrace III Najerilla river Glacis IV Terrace IV Najerilla

river

Parent materials Gravels and cobbles in sand matrix

Sands, silts and clays


Gravels and cobbles in sandy loam matrix


Rock fragments (%) 40-80 5-10 15-40 15-40 40-80 Effective depth (cm) 110 150 90 115 50

Clay (%) 17-30 24-25 45-55 15 20 Carbonates (%) 3.5-5 3-7.5 40-50 45-60 55

Water availability Low Moderate-high Moderate-high Moderate-low Low

Representative soil profile

4 - 81



• Statistical analysis. Treatment effects on measured variables were tested using ANOVA (univariate linear model), and comparisons among treatment means were made using the Tukey´s test calculated at p < 0.05.

RESULTS AND DISCUSSION We present 2009 campaign results respect to mineral composition of leaves (Tab. 2), yield

components (Tab. 3), physical-chemical parameters of wines produced (Tab. 4 and Tab. 5). In a preliminary assessment, the results show clear differences for the NTU 2. This NTU, corresponds to a Yalde river flood plain, with high soil fertility and elevate effective depth, presented the highest values of yield and vigour. Also, the physical-chemical analysis of wines in the NTU 2 reached the lowest alcoholic degree and colour parameters (anthocyanins, colour intensity and total polyphenols). Components of the acidity in NTU 2 wine showed the lowest tartaric acid and the highest potassium concentrations, which contributed to the increase of wine pH. In NTU 2 must, potassium highest levels were observed, this was in accordance with the potassium concentration in both leaf blade and petiole. In the blind tasting panel the NTU 2 wine had the lowest score (data not shown).

Tab. 2. Principal mineral composition of leaf blades and petioles (% dry matter), at veraison. Uruñuela 2009.

Different letters in same column indicate significant differences (p<0.05) in leaf analysis data between NTUs, applying Tukey´s test.

Tab. 3. Yield components. Vigour parameters. Uruñuela 2009.

Different letters in same column indicate significant differences (p<0.05) in yield components data between NTUs, applying Tukey´s test

Tab. 4. Analytic composition of wines. Alcoholic degree and colour parameters. Uruñuela 2009.

Different letters in same column indicate significant differences (p<0.05) in wine analysis data between NTUs, applying Tukey´s test.

Leaf tissue NTU N P K Ca Mg

1 2.15 ab 0.15 0.66 b 2.82 0.31 2 2.47 a 0.19 1.02 a 3.23 0.28 3 2.08 ab 0.16 0.68 b 3.11 0.38 5 2.18 ab 0.21 0.56 b 2.96 0.33

Leaf blade

6 2.04 b 0.15 0.60 b 3.24 0.32 1 0.46 0.16 0.65 b 1.66 0.66 2 0.52 0.22 1.90 a 1.89 0.53 3 0.43 0.14 0.65 b 1.64 0.76 5 0.46 0.25 0.57 b 1.97 0.74

Petiole

6 0.42 0.20 0.64 b 2.27 0.77

NTU Weightshoot (g)

Weightbunch (kg)

Weight of 100 berries (g)

Grape yield (kg/vine)

Weight of pruning wood (WPW) (kg wood/vine)

Ravaz index

1 73.1 0.367 235.6 4.1 0.79 ab 4.61 2 104.9 0.486 257.6 6.05 1.11 a 5.43 3 85.3 0.414 246.1 5.25 0.95 ab 5.65 5 81.8 0.374 235.3 4.91 0.78 ab 6.31 6 59.9 0.340 216.3 4.93 0.60 b 8.61

NTU Alcoholic degree

Colour intensity Tonality Total phenolic

index (280 nm) Anthocyanins

(mg/l)1 12.7 8.30 ab 0.600 ab 53.9 717.0 2 12.6 6.01 b 0.672 a 45.0 656.0 3 12.9 8.02 ab 0.567 ab 52.0 737.9 5 12.5 11.5 a 0.502 b 58.0 890.2 6 13.1 10.9 a 0.518 b 58.8 786.7

Tab. 5. Analytic composition of wines. Acidity parameters.Uruñuela 2009.


CONCLUSIONS In Tempranillo (Vitis vinifera L.) vineyards of Uruñuela municipality (La Rioja region), we

can define five Natural Terroir Units (NTUs), based on detailed cartography and in the pedologic, climatic, lithologic, and relief characteristics, with a different soil type in each NTU. This terroir survey, carried out during 2009 season, with three homogeneous and representative vineyard plots of each soil type, revealed some different between NTUs in the vine and wine parameters monitored. However, more experiments seasons are require to can discuss the terroir effect on vine development and wine typicity in this area.

ACKNOWLEDGMENTS This survey is based on research (P.R.12.09) that was financially supported by Consejería

de Agricultura, Ganadería y Desarrollo Rural, La Rioja Government. We thank Bodegas Patrocinio, Uruñuela, for providing the experimental vineyards and

collaborate in the study.

BIBLIOGRAPHYCarey V., Saayman D., Archer E., Barbeau G. and Wallace M., 2008. Viticultural terroirs in

Stellenbosch, South Africa. I. The identification of natural terroir units. JournalInternational Science Vigne Vin, n 4: 169-183.

García-Escudero E., Lorenzo I., Romero I., García C., Villar M.T., López D., Ibáñez S., Martín I., 2006. Niveles de referencia en base a calidad para el diagnóstico foliar en el ámbito de la D.O.Ca. Rioja. Nutrición mineral: aspectos fisiológicos, agronómicos y ambientales. XI Simposio ibérico sobre nutrición mineral de las plantas, n 2: 335-342.

Seguin G., 1988. Ecosystems of the great red wines produced in the maritime climate of Bordeaus. In: Proceedings of the Symposium on Maritime Climate Winegrowing. L. Fuller-Perrine, Eds. Edition. Department of Horticultural Sciences, Cornell University, Geneva, N.Y., n: 36-53.

Soil Survey Staff Natural Resources Conservation Service, U.S. Department of Agriculture, 2006. Keys to Soil Taxonomy. Tenth Edition. Blacksburg, Virginia: Pocahontas Press, Inc.

Van Leeuwen C., Friant P., Choné X., Tregoat O., Koundouras S., Dubourdieu D., 2004. Influence of climate, soil, and cultivar on terroir. American Journal of Enology and Viticulture, n 55.3: 207-217.

Van Leeuwen C., Seguin G., 2006. The concept of terroir in viticulture. Journal Wine Research, n 17.1: 1-10.

NTU pH Titratable acidity (g/l tartaric acid)

Tartaric acid (g/l)

Potassium (mg/l)

1 3.72 ab 5.28 2.11 ab 1244.8 ab2 3.90 a 4.71 1.74 b 1427.9 a3 3.67 ab 5.55 2.31 ab 1216.8 ab5 3.57 b 5.55 2.78 a 1098.2 b6 3.68 ab 5.32 2.60 ab 1226.8 ab

4 - 82



• Statistical analysis. Treatment effects on measured variables were tested using ANOVA (univariate linear model), and comparisons among treatment means were made using the Tukey´s test calculated at p < 0.05.

RESULTS AND DISCUSSION We present 2009 campaign results respect to mineral composition of leaves (Tab. 2), yield

components (Tab. 3), physical-chemical parameters of wines produced (Tab. 4 and Tab. 5). In a preliminary assessment, the results show clear differences for the NTU 2. This NTU, corresponds to a Yalde river flood plain, with high soil fertility and elevate effective depth, presented the highest values of yield and vigour. Also, the physical-chemical analysis of wines in the NTU 2 reached the lowest alcoholic degree and colour parameters (anthocyanins, colour intensity and total polyphenols). Components of the acidity in NTU 2 wine showed the lowest tartaric acid and the highest potassium concentrations, which contributed to the increase of wine pH. In NTU 2 must, potassium highest levels were observed, this was in accordance with the potassium concentration in both leaf blade and petiole. In the blind tasting panel the NTU 2 wine had the lowest score (data not shown).

Tab. 2. Principal mineral composition of leaf blades and petioles (% dry matter), at veraison. Uruñuela 2009.

Different letters in same column indicate significant differences (p<0.05) in leaf analysis data between NTUs, applying Tukey´s test.

Tab. 3. Yield components. Vigour parameters. Uruñuela 2009.

Different letters in same column indicate significant differences (p<0.05) in yield components data between NTUs, applying Tukey´s test

Tab. 4. Analytic composition of wines. Alcoholic degree and colour parameters. Uruñuela 2009.


Leaf tissue NTU N P K Ca Mg

1 2.15 ab 0.15 0.66 b 2.82 0.31 2 2.47 a 0.19 1.02 a 3.23 0.28 3 2.08 ab 0.16 0.68 b 3.11 0.38 5 2.18 ab 0.21 0.56 b 2.96 0.33

Leaf blade

6 2.04 b 0.15 0.60 b 3.24 0.32 1 0.46 0.16 0.65 b 1.66 0.66 2 0.52 0.22 1.90 a 1.89 0.53 3 0.43 0.14 0.65 b 1.64 0.76 5 0.46 0.25 0.57 b 1.97 0.74

Petiole

6 0.42 0.20 0.64 b 2.27 0.77

NTU Weightshoot (g)

Weightbunch (kg)

Weight of 100 berries (g)

Grape yield (kg/vine)

Weight of pruning wood (WPW) (kg wood/vine)

Ravaz index

1 73.1 0.367 235.6 4.1 0.79 ab 4.61 2 104.9 0.486 257.6 6.05 1.11 a 5.43 3 85.3 0.414 246.1 5.25 0.95 ab 5.65 5 81.8 0.374 235.3 4.91 0.78 ab 6.31 6 59.9 0.340 216.3 4.93 0.60 b 8.61

NTU Alcoholic degree

Colour intensity Tonality Total phenolic

index (280 nm) Anthocyanins

(mg/l)1 12.7 8.30 ab 0.600 ab 53.9 717.0 2 12.6 6.01 b 0.672 a 45.0 656.0 3 12.9 8.02 ab 0.567 ab 52.0 737.9 5 12.5 11.5 a 0.502 b 58.0 890.2 6 13.1 10.9 a 0.518 b 58.8 786.7

Tab. 5. Analytic composition of wines. Acidity parameters.Uruñuela 2009.


CONCLUSIONS In Tempranillo (Vitis vinifera L.) vineyards of Uruñuela municipality (La Rioja region), we

can define five Natural Terroir Units (NTUs), based on detailed cartography and in the pedologic, climatic, lithologic, and relief characteristics, with a different soil type in each NTU. This terroir survey, carried out during 2009 season, with three homogeneous and representative vineyard plots of each soil type, revealed some different between NTUs in the vine and wine parameters monitored. However, more experiments seasons are require to can discuss the terroir effect on vine development and wine typicity in this area.

ACKNOWLEDGMENTS This survey is based on research (P.R.12.09) that was financially supported by Consejería

de Agricultura, Ganadería y Desarrollo Rural, La Rioja Government. We thank Bodegas Patrocinio, Uruñuela, for providing the experimental vineyards and

collaborate in the study.

BIBLIOGRAPHYCarey V., Saayman D., Archer E., Barbeau G. and Wallace M., 2008. Viticultural terroirs in

Stellenbosch, South Africa. I. The identification of natural terroir units. JournalInternational Science Vigne Vin, n 4: 169-183.

García-Escudero E., Lorenzo I., Romero I., García C., Villar M.T., López D., Ibáñez S., Martín I., 2006. Niveles de referencia en base a calidad para el diagnóstico foliar en el ámbito de la D.O.Ca. Rioja. Nutrición mineral: aspectos fisiológicos, agronómicos y ambientales. XI Simposio ibérico sobre nutrición mineral de las plantas, n 2: 335-342.

Seguin G., 1988. Ecosystems of the great red wines produced in the maritime climate of Bordeaus. In: Proceedings of the Symposium on Maritime Climate Winegrowing. L. Fuller-Perrine, Eds. Edition. Department of Horticultural Sciences, Cornell University, Geneva, N.Y., n: 36-53.

Soil Survey Staff Natural Resources Conservation Service, U.S. Department of Agriculture, 2006. Keys to Soil Taxonomy. Tenth Edition. Blacksburg, Virginia: Pocahontas Press, Inc.

Van Leeuwen C., Friant P., Choné X., Tregoat O., Koundouras S., Dubourdieu D., 2004. Influence of climate, soil, and cultivar on terroir. American Journal of Enology and Viticulture, n 55.3: 207-217.

Van Leeuwen C., Seguin G., 2006. The concept of terroir in viticulture. Journal Wine Research, n 17.1: 1-10.

NTU pH Titratable acidity (g/l tartaric acid)

Tartaric acid (g/l)

Potassium (mg/l)

1 3.72 ab 5.28 2.11 ab 1244.8 ab2 3.90 a 4.71 1.74 b 1427.9 a3 3.67 ab 5.55 2.31 ab 1216.8 ab5 3.57 b 5.55 2.78 a 1098.2 b6 3.68 ab 5.32 2.60 ab 1226.8 ab

4 - 83



VALIDATION OF THE VITICULTURE ZONING METHODOLOGY APPLIED TO DETERMINE THE HOMOGENOUS SOIL UNITS

PRESENT ON D.O. RIBERA DEL DUERO REGION

*González-SanJosé ML(1), Gómez-Miguel V(2), Rivero-Pérez MD(1), Mihnea M(1) y Velasco-López T(1)

(1)Department of Biotechnology and Food Science. University of Burgos. Plaza Misael Bañuelos s/n, 09001 Burgos, Spain.

[email protected]. * Author to whom correspondence should be sent. (2)Dpto Edafología. Escuela Técnica Superior de Ingenieros Agrónomos, Universidad Politécnica de Madrid,

28040 Madrid, Spain. [email protected]

ABSTRACTThe methodology to viticulture zoning developed and proposed by Gómez-Miguel and

Sotés (1992) has been studied in order to validate it. This was the main aim of this work, which shows only partial results because data from more vintages will be collected during the next vintages.

The proposed validation is based on the comparison of quality levels of the viticulture products (grapes) grown in different Homogeneous Soil Units (HSU) but classified as the same level of quality. HSUs classified as optimum in Ribera del Duero Denomination of Origin (D.O.) region were chosen for this validation study. The three more important Optimum Units were selected. They represented around of 50% of the global surface of vineyards on the Ribera del Duero viticulture D.O. zone. Five different vineyards in each Unit were chosen. Trying to select the most similar vineyards to reduce variability factors, other selection criteria applied were grape variety, clone, rootstocks, age, training systems and cultural practices.

Three grape samples were collected around of each selected vineyards at the “Technological maturity” stage of the grapes. Different oenological quality parameters were analysed on the collected grapes. After the statistical treatment of the whole analytical data, obtained from grapes collected during two consecutive vintages, some significant results can be pointed out. Among them, it is interesting to note that, in general, variability due to vintage was stronger than that due to the HSU. In a similar way, variability due to vineyards was also significant, and in general, it was bigger than variability due to Units. Furthermore, the whole data showed similar levels of quality after comparing grapes from each HSU studied.

These results seem to validate the proposed methodology. That is, the methodology is valid to determine HSU which can produce grape of a similar quality, and then it can be applied to the correct or appropriate use of the agriculture medium.

KEYWORDViticulture zoning methodology – validation – grape – quality

INTRODUCTIONNowadays, it is undisputable the relationships among soil, climate, landscape and other

factors of the agriculture medium, with the characteristics of the wine grapes as composition, colour, astringency, and so on. In fact, a lot researcher groups all over the world have been

studying these influences and relationships for decades. To sum up all the precedent studies, now it is totally accepted that the interaction “terroir”-vine-viticulture is the base to obtain quality grapes from which make quality wine. Furthermore, this multiple interaction is the base to obtain wines with personality and with particularly expression of the medium in which grapes grown.

During the last 90’s, Gómez-Miguel and Sotés developed and proposed a methodology to viticulture zoning (Sotés y Gómes-Miguel, 1992; Gómez-Miguel and Sotés, 2003). This methodology has been applied in the most significant and important Spanish Viticultural and Oenological Denomination of Origin (D.O.) regions, such as Ribera del Duero, Rioja, Toro, among others. Cited authors indicated this methodology is useful to determine Homogeneous Soil Units (HSU) even if these co-exit in the same vineyard. So, cited authors said that this methodology is an appropriate method to the correct ordering and exploitation of the medium according to its viticulture and oenological use.

The validation of this methodology has not been carried out completely yet, and this is the main aim of this works. The proposed validation is based on the comparison of quality levels of the viticulture products (grapes) grown in different HSU classified as the same level of quality. The study is being carried out in vineyards of Ribera del Duero D.O. region. HSUs, classified as optimum, were chosen for this validation study. The three more important Optimum Units were selected. They represented around of 50% of the global surface of vineyards in the Ribera del Duero D.O. region.

MATERIALS AND METHODS The three more important and extensive HSUs selected for this study were defined from

the zoning study of Ribera del Duero Denomination of Origin carried out by Sotés and Gómez-Miguel (1992). They were the Units 6, 11 and 14. For more information about this Units consult (González-Sanjosé et al., 2008). As it is described in the previous cited work, five different vineyards in each Unit were chosen. Trying to select the most similar vineyards to reduce variability factors, other selection criteria applied were grape variety, clone, rootstocks, age, training systems and other cultural practices

Three grape samples were collected, around each selected vineyard, at the “Technological maturity” stage of the grapes, which are correlated with adequate levels of sugar, acidity, phenolic content (nowadays named phenolic maturity), so that good sanitary stages and even with good levels of aroma precursor compounds (González-Sanjosé et al. 1991). The harvesting periods were around the middle of October, and in general time between the first and the last sampling was around two weeks. Depending on the vintage, time sampling difference was large, 16 days in the first vintages, or short, tree days in the second one.

Three lots of 25 Kg of grapes were picked up from each selected vineyard. One cluster by vine was taken, and sampled vines were randomly chosen with a Z distribution around the vineyard. Grapes were transported in plastic boxes to the laboratory as soon as possible after their collection. From each sampled lot, groups of single grapes were obtained separating manually two o three grapes of each cluster. Then, three groups of 100 single grapes were randomly formed and they were used to analyse the composition of the grapes. Grapes were manually peeled, skins were used to evaluate phenolic composition, pulps were pressed to obtain the respective must where parameters related to sugars and acidity were measured. Titrable acidity (TA) expressed as g/L of tartaric acid, pH, conductivity (Cnd), K and Cl measurements so as reducing sugars (RS) were determined according to OIV methods (1990). ºBrix was evaluated by direct measured on refractometer. Malic Acid content was evaluated by enzymatic methodology. Phenolic extracts were obtained by maceration of skins with

4 - 84



VALIDATION OF THE VITICULTURE ZONING METHODOLOGY APPLIED TO DETERMINE THE HOMOGENOUS SOIL UNITS

PRESENT ON D.O. RIBERA DEL DUERO REGION

*González-SanJosé ML(1), Gómez-Miguel V(2), Rivero-Pérez MD(1), Mihnea M(1) y Velasco-López T(1)

(1)Department of Biotechnology and Food Science. University of Burgos. Plaza Misael Bañuelos s/n, 09001 Burgos, Spain.

[email protected]. * Author to whom correspondence should be sent. (2)Dpto Edafología. Escuela Técnica Superior de Ingenieros Agrónomos, Universidad Politécnica de Madrid,

28040 Madrid, Spain. [email protected]

ABSTRACTThe methodology to viticulture zoning developed and proposed by Gómez-Miguel and

Sotés (1992) has been studied in order to validate it. This was the main aim of this work, which shows only partial results because data from more vintages will be collected during the next vintages.

The proposed validation is based on the comparison of quality levels of the viticulture products (grapes) grown in different Homogeneous Soil Units (HSU) but classified as the same level of quality. HSUs classified as optimum in Ribera del Duero Denomination of Origin (D.O.) region were chosen for this validation study. The three more important Optimum Units were selected. They represented around of 50% of the global surface of vineyards on the Ribera del Duero viticulture D.O. zone. Five different vineyards in each Unit were chosen. Trying to select the most similar vineyards to reduce variability factors, other selection criteria applied were grape variety, clone, rootstocks, age, training systems and cultural practices.

Three grape samples were collected around of each selected vineyards at the “Technological maturity” stage of the grapes. Different oenological quality parameters were analysed on the collected grapes. After the statistical treatment of the whole analytical data, obtained from grapes collected during two consecutive vintages, some significant results can be pointed out. Among them, it is interesting to note that, in general, variability due to vintage was stronger than that due to the HSU. In a similar way, variability due to vineyards was also significant, and in general, it was bigger than variability due to Units. Furthermore, the whole data showed similar levels of quality after comparing grapes from each HSU studied.

These results seem to validate the proposed methodology. That is, the methodology is valid to determine HSU which can produce grape of a similar quality, and then it can be applied to the correct or appropriate use of the agriculture medium.

KEYWORDViticulture zoning methodology – validation – grape – quality

INTRODUCTIONNowadays, it is undisputable the relationships among soil, climate, landscape and other

factors of the agriculture medium, with the characteristics of the wine grapes as composition, colour, astringency, and so on. In fact, a lot researcher groups all over the world have been

studying these influences and relationships for decades. To sum up all the precedent studies, now it is totally accepted that the interaction “terroir”-vine-viticulture is the base to obtain quality grapes from which make quality wine. Furthermore, this multiple interaction is the base to obtain wines with personality and with particularly expression of the medium in which grapes grown.

During the last 90’s, Gómez-Miguel and Sotés developed and proposed a methodology to viticulture zoning (Sotés y Gómes-Miguel, 1992; Gómez-Miguel and Sotés, 2003). This methodology has been applied in the most significant and important Spanish Viticultural and Oenological Denomination of Origin (D.O.) regions, such as Ribera del Duero, Rioja, Toro, among others. Cited authors indicated this methodology is useful to determine Homogeneous Soil Units (HSU) even if these co-exit in the same vineyard. So, cited authors said that this methodology is an appropriate method to the correct ordering and exploitation of the medium according to its viticulture and oenological use.

The validation of this methodology has not been carried out completely yet, and this is the main aim of this works. The proposed validation is based on the comparison of quality levels of the viticulture products (grapes) grown in different HSU classified as the same level of quality. The study is being carried out in vineyards of Ribera del Duero D.O. region. HSUs, classified as optimum, were chosen for this validation study. The three more important Optimum Units were selected. They represented around of 50% of the global surface of vineyards in the Ribera del Duero D.O. region.

MATERIALS AND METHODS The three more important and extensive HSUs selected for this study were defined from

the zoning study of Ribera del Duero Denomination of Origin carried out by Sotés and Gómez-Miguel (1992). They were the Units 6, 11 and 14. For more information about this Units consult (González-Sanjosé et al., 2008). As it is described in the previous cited work, five different vineyards in each Unit were chosen. Trying to select the most similar vineyards to reduce variability factors, other selection criteria applied were grape variety, clone, rootstocks, age, training systems and other cultural practices

Three grape samples were collected, around each selected vineyard, at the “Technological maturity” stage of the grapes, which are correlated with adequate levels of sugar, acidity, phenolic content (nowadays named phenolic maturity), so that good sanitary stages and even with good levels of aroma precursor compounds (González-Sanjosé et al. 1991). The harvesting periods were around the middle of October, and in general time between the first and the last sampling was around two weeks. Depending on the vintage, time sampling difference was large, 16 days in the first vintages, or short, tree days in the second one.

Three lots of 25 Kg of grapes were picked up from each selected vineyard. One cluster by vine was taken, and sampled vines were randomly chosen with a Z distribution around the vineyard. Grapes were transported in plastic boxes to the laboratory as soon as possible after their collection. From each sampled lot, groups of single grapes were obtained separating manually two o three grapes of each cluster. Then, three groups of 100 single grapes were randomly formed and they were used to analyse the composition of the grapes. Grapes were manually peeled, skins were used to evaluate phenolic composition, pulps were pressed to obtain the respective must where parameters related to sugars and acidity were measured. Titrable acidity (TA) expressed as g/L of tartaric acid, pH, conductivity (Cnd), K and Cl measurements so as reducing sugars (RS) were determined according to OIV methods (1990). ºBrix was evaluated by direct measured on refractometer. Malic Acid content was evaluated by enzymatic methodology. Phenolic extracts were obtained by maceration of skins with

4 - 85



methanol acidulate with formic acid, according to the procedure described by Izcara and González-Sanjosé, (2001). Global phenolic content (total polyphenols, TP, expressed as mg/L of gallic acid), so as some phenolic families (total anthocyanins, ACYpH, expressed as mg/L of malvidin-3-glucoside, and catechins (CAT) quantified as mg/L of D-catechin) were evaluated by classical spectrophotometric methodologies, all of them described in García-Barcelo (1990). Furthermore, total phenolic-tartaric-ester contents (E-TH2) were analysed according to Mazza et al., (1999) and Total Flavonol levels (FLA Neu) by Neu reactions. The Glories colour parameters, colour intensity (CIntensity), tonality (To) and percentage of red, yellow (Ye) and Blue were measured on methanol extract of the skins.

The analysis of the variance (ANOVA) and the Least Significant Difference test (LSD) were used to detect differences and to establish which data could be considered statistically different. A significance level of α = 0.05 was used. Multivariante analysis were also applied, Factorial Analysis were applied to the global analysis of the data. All statistical analyses were carried out using the statistics package Statgraphics Plus 4.0 (1999, Manugistics Inc.).

RESULTS AND DISCUSSION Firstly to comment the obtained results, authors want to note that this paper present partial

results (from two vintages) which will be completed and corroborated, with data from new vintages. Secondly, it is important to comment that the climatology on the viticultural region under study was very unfortunate in different aspect during the year under study. So, the meteorology of the first year was very adverse, affecting notably to the development of clusters and grapes so as the ripening process. Important Spring frost, strong hails during May, so as a warm summer, not hot enough, caused that some of the selected vineyards did not show the best conditions to produce adequate quality grapes. According to these comments, even if grapes were harvested from the 15 vineyards under study only data from 11 of them will be showed and commented in this work. Only data from grapes with an adequate level of technological quality will be considered. During the second vintage meteorology was loss adverse, however a strong hail storm during spring and a frost at the beginning of the autumn, damaged largely two of the vineyards selected for this study. For that reason, at in the first year of study, even if samples were collected from the 15 selected vineyards, only data from 13 of them have been included in this study.

The diverse factors of variability on the composition of the grapes are well known. Some of them are: the intra vineyard variability, due to own metabolism of each vine and cluster; the inter variability due to vineyards even if these are close, due to soil units, cultural practice and so on and the inter variability due to the vintages, especially associate to climatic conditions. The experimental design applied in this study try to consider all of them. So, the three lots collected around each selected vineyard try to collect the intra-vineyard-variability and the five vineyards selected form each HSU try to collect the inter-vineyard-variability. Obviously the extension of this study to different vinatge tries to collect also the inter-annual-variability. The two last types of variability are showed in figure 1, which showed, as example, the results from two of the parameters studied. Similar results were observed in the other studied oenological characteristics.

The results showed large variability due to vineyards, but this was also dependent on the year or vintage. So, the inter-vineyards variability can be very important (large vertical lines) or insignificant (short vertical lines).

Global results also showed a clear effect of the year, as it can be observed in the figure 2. Factorial Component Analysis showed how data of the grapes from each vintage were well separated on the left and the right of the figure, respectively. The multivariante analysis also

showed the intra-vineyards variability of the vineyards, which is showed by the dispersion of the points. Very close points correspond with grape-lots from the vineyards with a small intra-variability and disperse points correspond with vineyards with a large intra-variability.

Figure 1. Scatter-plots of Variance Components Analysis. Solid horizontal lines are drawn at the means of the data for each factor level (HTU). Points are drawn at the average values for each vineyard of each HTU, vertical lines indicate the difference among means of each vineyard and the means of its respective HTU.

Figure 2. Distribution, on the plane defined by the two main Factor Components, of the grapes from each lot analysed on the two studied vintages. Each point is the score of each lot after a multi-data analyse. Point on the left/right: grapes from 1st and 2nd vintage respectively.

Firs t Compone nt

Se

con

d C

om

po

ne

nt

43210-1-2-3-4-5

5

4

3

2

1

0

-1

-2

-3

-4

UH T6

1114

2 0 0 8

U H T

Intens

idad

6 1 1 1 49

1 1

1 3

1 5

1 7

1 9

2 1

2 0 0 8

U H T

Acide

z total

6 1 1 1 43 ,7

5 ,7

7 ,7

9 ,7

1 1 ,7

2 0 0 7

U H T

Acide

z total

6 1 1 1 44 ,4

5 ,4

6 ,4

7 ,4

8 ,4

9 ,4

1 s t v in ta g e 2 n d v in ta g e

Tritu

labl

eA

cidi

ty

2 0 0 7

U H T

Intens

idad

6 1 1 1 40 ,9

1 ,2

1 ,5

1 ,8

2 ,1

2 ,4

Col

ourI

nten

sity

H S U H S U

2 0 0 8

U H T

Intens

idad

6 1 1 1 49

1 1

1 3

1 5

1 7

1 9

2 1

2 0 0 8

U H T

Acide

z total

6 1 1 1 43 ,7

5 ,7

7 ,7

9 ,7

1 1 ,7

2 0 0 7

U H T

Acide

z total

6 1 1 1 44 ,4

5 ,4

6 ,4

7 ,4

8 ,4

9 ,4


Tritu

labl

eA

cidi

ty

2 0 0 7

U H T

Intens

idad

6 1 1 1 40 ,9

1 ,2

1 ,5

1 ,8

2 ,1

2 ,4

Col

ourI

nten

sity

H S U H S U

4 - 86



methanol acidulate with formic acid, according to the procedure described by Izcara and González-Sanjosé, (2001). Global phenolic content (total polyphenols, TP, expressed as mg/L of gallic acid), so as some phenolic families (total anthocyanins, ACYpH, expressed as mg/L of malvidin-3-glucoside, and catechins (CAT) quantified as mg/L of D-catechin) were evaluated by classical spectrophotometric methodologies, all of them described in García-Barcelo (1990). Furthermore, total phenolic-tartaric-ester contents (E-TH2) were analysed according to Mazza et al., (1999) and Total Flavonol levels (FLA Neu) by Neu reactions. The Glories colour parameters, colour intensity (CIntensity), tonality (To) and percentage of red, yellow (Ye) and Blue were measured on methanol extract of the skins.

The analysis of the variance (ANOVA) and the Least Significant Difference test (LSD) were used to detect differences and to establish which data could be considered statistically different. A significance level of α = 0.05 was used. Multivariante analysis were also applied, Factorial Analysis were applied to the global analysis of the data. All statistical analyses were carried out using the statistics package Statgraphics Plus 4.0 (1999, Manugistics Inc.).

RESULTS AND DISCUSSION Firstly to comment the obtained results, authors want to note that this paper present partial

results (from two vintages) which will be completed and corroborated, with data from new vintages. Secondly, it is important to comment that the climatology on the viticultural region under study was very unfortunate in different aspect during the year under study. So, the meteorology of the first year was very adverse, affecting notably to the development of clusters and grapes so as the ripening process. Important Spring frost, strong hails during May, so as a warm summer, not hot enough, caused that some of the selected vineyards did not show the best conditions to produce adequate quality grapes. According to these comments, even if grapes were harvested from the 15 vineyards under study only data from 11 of them will be showed and commented in this work. Only data from grapes with an adequate level of technological quality will be considered. During the second vintage meteorology was loss adverse, however a strong hail storm during spring and a frost at the beginning of the autumn, damaged largely two of the vineyards selected for this study. For that reason, at in the first year of study, even if samples were collected from the 15 selected vineyards, only data from 13 of them have been included in this study.

The diverse factors of variability on the composition of the grapes are well known. Some of them are: the intra vineyard variability, due to own metabolism of each vine and cluster; the inter variability due to vineyards even if these are close, due to soil units, cultural practice and so on and the inter variability due to the vintages, especially associate to climatic conditions. The experimental design applied in this study try to consider all of them. So, the three lots collected around each selected vineyard try to collect the intra-vineyard-variability and the five vineyards selected form each HSU try to collect the inter-vineyard-variability. Obviously the extension of this study to different vinatge tries to collect also the inter-annual-variability. The two last types of variability are showed in figure 1, which showed, as example, the results from two of the parameters studied. Similar results were observed in the other studied oenological characteristics.

The results showed large variability due to vineyards, but this was also dependent on the year or vintage. So, the inter-vineyards variability can be very important (large vertical lines) or insignificant (short vertical lines).

Global results also showed a clear effect of the year, as it can be observed in the figure 2. Factorial Component Analysis showed how data of the grapes from each vintage were well separated on the left and the right of the figure, respectively. The multivariante analysis also

showed the intra-vineyards variability of the vineyards, which is showed by the dispersion of the points. Very close points correspond with grape-lots from the vineyards with a small intra-variability and disperse points correspond with vineyards with a large intra-variability.

Figure 1. Scatter-plots of Variance Components Analysis. Solid horizontal lines are drawn at the means of the data for each factor level (HTU). Points are drawn at the average values for each vineyard of each HTU, vertical lines indicate the difference among means of each vineyard and the means of its respective HTU.

Figure 2. Distribution, on the plane defined by the two main Factor Components, of the grapes from each lot analysed on the two studied vintages. Each point is the score of each lot after a multi-data analyse. Point on the left/right: grapes from 1st and 2nd vintage respectively.

Firs t Compone nt

Se

con

d C

om

po

ne

nt

43210-1-2-3-4-5

5

4

3

2

1

0

-1

-2

-3

-4

UH T6

1114

2 0 0 8

U H T

Intens

idad

6 1 1 1 49

1 1

1 3

1 5

1 7

1 9

2 1

2 0 0 8

U H T

Acide

z total

6 1 1 1 43 ,7

5 ,7

7 ,7

9 ,7

1 1 ,7

2 0 0 7

U H T

Acide

z total

6 1 1 1 44 ,4

5 ,4

6 ,4

7 ,4

8 ,4

9 ,4


Tritu

labl

eA

cidi

ty

2 0 0 7

U H T

Intens

idad

6 1 1 1 40 ,9

1 ,2

1 ,5

1 ,8

2 ,1

2 ,4

Col

ourI

nten

sity

H S U H S U

2 0 0 8

U H T

Intens

idad

6 1 1 1 49

1 1

1 3

1 5

1 7

1 9

2 1

2 0 0 8

U H T

Acide

z total

6 1 1 1 43 ,7

5 ,7

7 ,7

9 ,7

1 1 ,7

2 0 0 7

U H T

Acide

z total

6 1 1 1 44 ,4

5 ,4

6 ,4

7 ,4

8 ,4

9 ,4


Tritu

labl

eA

cidi

ty

2 0 0 7

U H T

Intens

idad

6 1 1 1 40 ,9

1 ,2

1 ,5

1 ,8

2 ,1

2 ,4

Col

ourI

nten

sity

H S U H S U

4 - 87



Vintage 1rst 2nd UHT 6 11 14 6 11 14

SR (g/L) 228,8 ± 7 a 226,9 ±17,9 a 226,7 ± 6,6 a 216,3 ± 13,7 a 220,7 ± 11,3 a 225,3 ± 19,9 a

º Brix 22,2 ± 1,1 b 22,2 ± 1,9 b 24,4 ± 0,4 a 22,4 ± 1,1 a 23,2 ± 1,2 a 22,9 ± 1 a

Malic (g/L) 4,4 ± 1 a 3,7 ± 1 ab 2,9 ± 0,8 b 5,5 ± 1,3 a 3,9 ± 0,5 b 4,2 ± 0,8 b

pH 3,29 ± 0,19 b 3,3 ± 0,15 b 3,5 ± 0,06 a 3,57 ± 0,26 a 3,5 ± 0,1 a 3,64 ± 0,27 a

TA (g/L) 7,18 ± 1,4 a 6,96 ± 1,39 a 5,01 ± 0,5 b 7,01 ± 2,03 a 6,16 ± 0,41 ab 5,18 ± 1,33 b

Cnd (mS) 2,22 ± 0,24 a 2,08 ± 0,22 a 2,13 ± 0,15 a 2,44 ± 0,18 a 2,22 ± 0,28 b 2,37 ± 0,25 ab

K (mg/L) 1353 ± 178 ab 1229 ± 67 b 1386 ± 225 a 1642 ± 176 a 1452 ± 84 b 1621 ± 281 a

Cl (mg/L) 18,5 ± 4,7 a 17,2 ± 5,9 a 19,2 ± 4,3 a 40,5 ± 17,5 a 36,5 ± 12,9 a 39 ± 18,5 a

TP (mg/L) 1678 ± 217 b 1734 ± 266 b 2104 ± 227 a 1857 ± 461 a 1760 ± 290 a 1863 ± 237 a

CAT (mg/L) 316 ± 66 c 383 ± 54 b 496 ± 110 a 516 ± 69 a 440 ± 163 a 495 ± 98 aACY pH (mg/L) 1100 ± 85 a 1076 ± 179 a 1100, ± 57 a 964 ± 88 a 877 ± 205 a 942 ± 123 a

FLA N (mg/L) 74,8 ± 9,6 b 72,2 ± 11,5 b 83,9 ± 6,6 a 104,7 ± 7,3 b 105,5 ± 21,1 b 123 ± 35,4 a

E-TH2 (mg/L) 14,3 ± 1,66 a 14,4 ± 2,5 a 13,9 ± 0,6 a 13,8 ± 1,3 b 13 ± 2,5 b 16,4 ± 3,4 aCIntensity

(1mm) 1,38 ± 0,20 b 1,43 ± 0,34 b 1,89 ± 0,24 a 2,18 ± 0,3 b 2,41 ± 0,32 a 2,39 ± 0,24 ab

Tonality 0,47 ± 0,06 a 0,45 ± 0,07 a 0,43 ± 0,02 a 0,22 ± 0,02 a 0,23 ± 0,03 a 0,22 ± 0,02 a

% Yellow 29,4 ± 2,6 a 28,7 ± 2,95 a 28,3 ± 1 a 17,7 ± 0,9 a 17,6 ± 1,67 a 17,6 ± 1,1 a

% Red 63,1 ± 3,3 a 64,9 ± 4,2 a 65,5 ± 1,2 a 79,7 ± 2,2 a 76,4 ± 4,2 b 78,8 ± 3,5 ab

% Blue 7,5 ± 1,87 a 5,9 ± 1,14 b 6,2 ± 0,6 b 2,6 ± 1,56 b 4,8 ± 1,81 a 3,6 ± 2,5 ab

Furthermore, multivariante analysis showed that the grapes from the different studied units were very similar in composition among them, although this similitude depends also on the vintage. So, data from second vintage are globally more aggregated than those from the first one and no intra grouping were glimpsed. However, data from the first vintage allow glimpsing a slight aggregation of the data by units. This fact is also observed by invariant analysis of the studies variables (table 1) which showed that data from the first vintage showed more statistical significant differences among HSU than data from the second one.

Table 1. Mean values and deviation of each indicated parameter and HSU obtained at each studied vintages. Letters indicate significant differences among values for each vintage. LSD (Fisher's least significant difference) method to α = 0,05 was applied.

The similitude among grapes of the three studied units will be also clearly observed in the figure 3, which showed the global data summarized on the graphical representation of the average data of grapes from each HSU studied. The general composition profiles showed in this figure were very similar, that means that, in general, grapes showed similar oenological characteristics independent of the soil units in which they were cultivated.

CONCLUSIONS These results seem to validate the proposed methodology. That is to say, the methodology is

valid to determine HSU which can produce grapes of the similar quality, and then it can be applied to the correct or appropriate use of the agriculture medium.

Figure 3. Global oenological characteristic profile of the grapes of each studied HSU. Points showed mean values of grapes by units independent on vineyard and vintage. n= 27, 24 and 21 to HSU 6, 11 and 14, respectively. .

ACKNOWLEDGMENTS The authors want to thank all the wineries and vineyards’ owners who give them the

possibility to take samples from the selected vineyards. Furthermore, they thank the financial support from the Autonomous Government of Castilla y León, Spain, through the Research Proyect of reference BU025A06.

BIBLIOGRAPHYGarcía-Barceló J. 1990. Técnicas analíticas para vinos. GAB Ed, Barcelona, Spain. Gómez-Miguel V. and Sotés V. 2003. Zonificación del “terroir” en España. In: “Terroir,

Zonazione, Viticoltura: trattato internazionale”. Ed. Fregoni M., Schuster D. and Paoletti A. Phytoline, Rivoli Veronese, Italy. pp 187-226

González-Sanjosé M.L., Barrón L.J.R., Junquera B., Robredo L.M. 1991. Application of principal component analysis to ripening indices for wine grapes. Journal of Food Composition and Analysis, 4, 245-255.

González-Sanjosé M.L., Rivero M.D., Bleoju M. and Gómez- Miguel, V. 2008. In: VIIe Congrès International des Terroirs Viticoles 2008 Acta. ACW Ed. Nyon, Suisse. Vol.2, 641-7.

Izcara E. and González-Sanjosé M.L. 2001. Análisis de métodos rápidos de extracción para seguir la maduración fenólica de la uva. Enólogos, 14, 14-18.

OIV. 1990. Recueil des Méthodes Internationales d’Analyse des vins et des moûts. OfficeInternational de la Vigne et du Vin, París.

Mazza G., Fukumoto L., Delaquis P., Girard B. and Ewert B. 1999. Anthocyanins, phenolics and color of Cabernet Franc, Merlot, and Pinot Noir wines from British Columbia. Journal of Agriculture and Food Chemistry, 47, 4009-4017.

Sotés V. and Gómez-Miguel V. 1992. Delimitación de zonas vitícolas en la Denominación de Origen Ribera de Duero. Informes Técnicos ETSIA. UPM, Madrid, Spain.

0

8

16

24

32

TA

Malic

RS/10

º Brix

pH

K/100

Cl

Conductiv ity

TP (g/L)

CAT/100

Acy pH (g/L)

FLA Neu/10

E-TH2

CIntensity

Tonality

% Ye

% Red/10

% Blue

HSU 6 HSU 11 HSU 14

4 - 88



Vintage 1rst 2nd UHT 6 11 14 6 11 14

SR (g/L) 228,8 ± 7 a 226,9 ±17,9 a 226,7 ± 6,6 a 216,3 ± 13,7 a 220,7 ± 11,3 a 225,3 ± 19,9 a

º Brix 22,2 ± 1,1 b 22,2 ± 1,9 b 24,4 ± 0,4 a 22,4 ± 1,1 a 23,2 ± 1,2 a 22,9 ± 1 a

Malic (g/L) 4,4 ± 1 a 3,7 ± 1 ab 2,9 ± 0,8 b 5,5 ± 1,3 a 3,9 ± 0,5 b 4,2 ± 0,8 b

pH 3,29 ± 0,19 b 3,3 ± 0,15 b 3,5 ± 0,06 a 3,57 ± 0,26 a 3,5 ± 0,1 a 3,64 ± 0,27 a

TA (g/L) 7,18 ± 1,4 a 6,96 ± 1,39 a 5,01 ± 0,5 b 7,01 ± 2,03 a 6,16 ± 0,41 ab 5,18 ± 1,33 b

Cnd (mS) 2,22 ± 0,24 a 2,08 ± 0,22 a 2,13 ± 0,15 a 2,44 ± 0,18 a 2,22 ± 0,28 b 2,37 ± 0,25 ab

K (mg/L) 1353 ± 178 ab 1229 ± 67 b 1386 ± 225 a 1642 ± 176 a 1452 ± 84 b 1621 ± 281 a

Cl (mg/L) 18,5 ± 4,7 a 17,2 ± 5,9 a 19,2 ± 4,3 a 40,5 ± 17,5 a 36,5 ± 12,9 a 39 ± 18,5 a

TP (mg/L) 1678 ± 217 b 1734 ± 266 b 2104 ± 227 a 1857 ± 461 a 1760 ± 290 a 1863 ± 237 a

CAT (mg/L) 316 ± 66 c 383 ± 54 b 496 ± 110 a 516 ± 69 a 440 ± 163 a 495 ± 98 aACY pH (mg/L) 1100 ± 85 a 1076 ± 179 a 1100, ± 57 a 964 ± 88 a 877 ± 205 a 942 ± 123 a

FLA N (mg/L) 74,8 ± 9,6 b 72,2 ± 11,5 b 83,9 ± 6,6 a 104,7 ± 7,3 b 105,5 ± 21,1 b 123 ± 35,4 a

E-TH2 (mg/L) 14,3 ± 1,66 a 14,4 ± 2,5 a 13,9 ± 0,6 a 13,8 ± 1,3 b 13 ± 2,5 b 16,4 ± 3,4 aCIntensity

(1mm) 1,38 ± 0,20 b 1,43 ± 0,34 b 1,89 ± 0,24 a 2,18 ± 0,3 b 2,41 ± 0,32 a 2,39 ± 0,24 ab

Tonality 0,47 ± 0,06 a 0,45 ± 0,07 a 0,43 ± 0,02 a 0,22 ± 0,02 a 0,23 ± 0,03 a 0,22 ± 0,02 a

% Yellow 29,4 ± 2,6 a 28,7 ± 2,95 a 28,3 ± 1 a 17,7 ± 0,9 a 17,6 ± 1,67 a 17,6 ± 1,1 a

% Red 63,1 ± 3,3 a 64,9 ± 4,2 a 65,5 ± 1,2 a 79,7 ± 2,2 a 76,4 ± 4,2 b 78,8 ± 3,5 ab

% Blue 7,5 ± 1,87 a 5,9 ± 1,14 b 6,2 ± 0,6 b 2,6 ± 1,56 b 4,8 ± 1,81 a 3,6 ± 2,5 ab

Furthermore, multivariante analysis showed that the grapes from the different studied units were very similar in composition among them, although this similitude depends also on the vintage. So, data from second vintage are globally more aggregated than those from the first one and no intra grouping were glimpsed. However, data from the first vintage allow glimpsing a slight aggregation of the data by units. This fact is also observed by invariant analysis of the studies variables (table 1) which showed that data from the first vintage showed more statistical significant differences among HSU than data from the second one.

Table 1. Mean values and deviation of each indicated parameter and HSU obtained at each studied vintages. Letters indicate significant differences among values for each vintage. LSD (Fisher's least significant difference) method to α = 0,05 was applied.

The similitude among grapes of the three studied units will be also clearly observed in the figure 3, which showed the global data summarized on the graphical representation of the average data of grapes from each HSU studied. The general composition profiles showed in this figure were very similar, that means that, in general, grapes showed similar oenological characteristics independent of the soil units in which they were cultivated.

CONCLUSIONS These results seem to validate the proposed methodology. That is to say, the methodology is

valid to determine HSU which can produce grapes of the similar quality, and then it can be applied to the correct or appropriate use of the agriculture medium.

Figure 3. Global oenological characteristic profile of the grapes of each studied HSU. Points showed mean values of grapes by units independent on vineyard and vintage. n= 27, 24 and 21 to HSU 6, 11 and 14, respectively. .

ACKNOWLEDGMENTS The authors want to thank all the wineries and vineyards’ owners who give them the

possibility to take samples from the selected vineyards. Furthermore, they thank the financial support from the Autonomous Government of Castilla y León, Spain, through the Research Proyect of reference BU025A06.

BIBLIOGRAPHYGarcía-Barceló J. 1990. Técnicas analíticas para vinos. GAB Ed, Barcelona, Spain. Gómez-Miguel V. and Sotés V. 2003. Zonificación del “terroir” en España. In: “Terroir,

Zonazione, Viticoltura: trattato internazionale”. Ed. Fregoni M., Schuster D. and Paoletti A. Phytoline, Rivoli Veronese, Italy. pp 187-226

González-Sanjosé M.L., Barrón L.J.R., Junquera B., Robredo L.M. 1991. Application of principal component analysis to ripening indices for wine grapes. Journal of Food Composition and Analysis, 4, 245-255.

González-Sanjosé M.L., Rivero M.D., Bleoju M. and Gómez- Miguel, V. 2008. In: VIIe Congrès International des Terroirs Viticoles 2008 Acta. ACW Ed. Nyon, Suisse. Vol.2, 641-7.

Izcara E. and González-Sanjosé M.L. 2001. Análisis de métodos rápidos de extracción para seguir la maduración fenólica de la uva. Enólogos, 14, 14-18.

OIV. 1990. Recueil des Méthodes Internationales d’Analyse des vins et des moûts. OfficeInternational de la Vigne et du Vin, París.

Mazza G., Fukumoto L., Delaquis P., Girard B. and Ewert B. 1999. Anthocyanins, phenolics and color of Cabernet Franc, Merlot, and Pinot Noir wines from British Columbia. Journal of Agriculture and Food Chemistry, 47, 4009-4017.

Sotés V. and Gómez-Miguel V. 1992. Delimitación de zonas vitícolas en la Denominación de Origen Ribera de Duero. Informes Técnicos ETSIA. UPM, Madrid, Spain.

0

8

16

24

32

TA

Malic

RS/10

º Brix

pH

K/100

Cl

Conductiv ity

TP (g/L)

CAT/100

Acy pH (g/L)

FLA Neu/10

E-TH2

CIntensity

Tonality

% Ye

% Red/10

% Blue

HSU 6 HSU 11 HSU 14

4 - 89



TYPICITE ET TERROIR : IMPORTANCE RELATIVE DU TYPE DE

SOL ET DU NIVEAU DE MATURITE DANS LA TYPOLOGIE

SENSORIELLE DU VIN.

TYPICALITY AND TERROIR : COMPARISON OF SOIL TYPE AND

HARVEST DATE EFFECTS ON THE SENSORIAL STYLE OF WINE.

Champenois Réjane (1)

, Cadot Yves (1)

, Caille Soline

(2), Samson Alain

(3) and Cheynier

Véronique (2)

(1)

INRA, UE 1117, UMT Vinitera,

F-49070 Beaucouzé, France

[email protected]

(2) INRA, UMR1083 Sciences pour l'Œnologie,

F-34060 Montpellier, France

(3) INRA, UE999 Pech-Rouge,

F-11430 Gruissan, France.

RESUME

Le lien fonctionnel entre typicité et terroir a été étudié en prenant en compte deux

dimensions importantes : le type de sol et la date de vendanges. Ces deux facteurs sont, à des

degrés divers, considérés comme facteurs explicatifs de l’effet terroir. Trois parcelles de

Cabernet franc, sur trois types de sols différents et revendiquant des AOP variées (“Anjou

Villages”, “Anjou Rouge” et “Saumur Champigny”) ont été vinifiées, en triplicata, à deux

dates espacées de 14 jours. Les vins, vinifiés selon un protocole standardisé, ont été analysés

sensoriellement par un jury de professionnels (question de typicité : Anjou Rouge vs Anjou

Villages) et par un jury expert (profil conventionnel). Pour évaluer la notion de maturité

phénolique (teneur et aptitude à l’extraction), les composés phénoliques ont été analysés à la

vendange, au décuvage, mais également au moment de l’analyse sensorielle.

Les résultats montrent que si le type de sol a permis de discriminer les profils sensoriels des

vins, son effet sur la typicité a été faible. La date de vendanges, au contraire, a permis de

discriminer les profils sensoriels mais également les notes de typicité. Concernant les

composés phénoliques, si la teneur et la composition en anthocyanes était dépendante de la

date de vendanges, elle n’a pas été explicative de la typicité, sauf quand les anthocyanes

totales ont été mesurées lors de l’analyse sensorielle (effet couleur). La quantité de tanins

condensés n’est pas apparue dépendante des parcelles mais de la date de vendanges. La

qualité des tanins contenus dans le vin au décuvage s’est révélée différente selon la date de

vendanges et explicative de la typicité. Enfin, la couleur des vins, liée à leur composition en

composés phénoliques, a influencé la perception de la typicité.

Cette étude illustre l’importance de certaines pratiques dans l’effet terroir, le type de sol

ayant un effet direct beaucoup moins important que ne laissent supposer les résultats

d’enquêtes auprès des producteurs.

MOTS CLES

Terroir – Typicité - Tanins condensés – Anthocyanes – Cabernet franc - Vitis vinifera

ABSTRACT

Harvest date is a critical point to the winemaker, in order to produce wine with a distinctive

style. In particular the relation between ripening stage and extractability of flavonoids must be

highlighted.

The extractability of flavonoids (flavan-3-ols, anthocyanins) from grapes was monitored at

two stages of maturity (veraison + 30 days, veraison + 44 days). Berries were obtained from

three plots with different types of soil in term of water status, from 3 AOC (Anjou-Villages-

Brissac, Anjou and Saumur Champigny) and were elaborated in triplicate. Flavonoids were

analysed before and after winemaking, by RP-LC-DAD, after fractionation and thiolysis for

the proanthocyanidins. Sensory analysis was performed eight month after harvest, by a

sensory expert panel (Quantitative descriptive analysis) and by wine experts, (assessment of

the typicality). Wine experts were producers, winemakers, and oenologists from the area.

The results showed that the type of soil allowed to discriminate the wines according to

some sensory attributes, but its effect on the typicality was weak. On the contrary, the date of

grape harvest, allowed discriminating the wine according to their sensory profiles and also to

their typicality scores. Concerning the flavonoids, if the content and the composition in

anthocyanins were dependent on the date of grape harvest, it was not connected to the

typicality, except when anthocyanins were analyzed just before sensory analysis. The quantity

of condensed tannins was not dependent on plots but on harvest date. The quality of tannins

contained in the wine at devatting was different according to hatvest date. Moreover, quantity

and quality of condensed tanins were highly correlated to the typicality scores. Finally, if the

anthocyanin contents of wines were correlated with typicality, the composition in the final

wine were not predicted by composition at devatting. The influence of anthocyanins seemed

to be due to perception of the color of wines in the typicality judgment.

This study illustrated the importance of harvest and vatting practices in the terroir effect,

with a soil effect less important as often admitted.

KEYWORDS

Terroir – Typicality – Condensed tanins – Anthocyanins - Cabernet franc - Vitis vinifera

INTRODUCTION

La date de récolte est un moment important dans le cycle de production : elle synthétise un

itinéraire viticole et suppose dans ses choix des anticipations quant au futur itinéraire

œnologique. Pour comprendre les déterminants des choix de dates de récolte, des critères

technologiques sont aisément identifiables. Ces critères technologiques sont plus ou moins

bien connus et/ou maîtrisés. Parmi les déterminants technologiques, la « maturité phénolique

» est un critère essentiel d’évaluation de la maturité. La maturité phénolique prend en compte

la teneur globale en polyphénols, mais aussi leur structure et leur aptitude à l’extraction

(Glories, 1998). Ainsi, la maturité phénolique peut être définie comme le niveau de maturité

permettant l’obtention simultanée d’un potentiel important et d’une bonne capacité de

diffusion dans le vin. Néanmoins, cette notion reste encore peu précise, car les évolutions au

cours de la maturation ne sont pas clairement établies, et les propriétés des tanins condensés et

les déterminants de leur extractibilité sont peu connus (Cadot, et al., 2006; Fournand et al.,

2006). De nombreux auteurs montrent l’importance de ces composés sur la qualité des

produits (visuelles, gustatives et somesthésiques), en particulier dans le cas des vins rouges

(Brossaud et al., 2001; Vidal et al., 2004). La synthèse de ces composés est fortement

dépendante de facteurs biotiques et abiotiques (Winkel-Shirley, 2002).

4 - 90



TYPICITE ET TERROIR : IMPORTANCE RELATIVE DU TYPE DE

SOL ET DU NIVEAU DE MATURITE DANS LA TYPOLOGIE

SENSORIELLE DU VIN.

TYPICALITY AND TERROIR : COMPARISON OF SOIL TYPE AND

HARVEST DATE EFFECTS ON THE SENSORIAL STYLE OF WINE.

Champenois Réjane (1)

, Cadot Yves (1)

, Caille Soline

(2), Samson Alain

(3) and Cheynier

Véronique (2)

(1)

INRA, UE 1117, UMT Vinitera,

F-49070 Beaucouzé, France

[email protected]

(2) INRA, UMR1083 Sciences pour l'Œnologie,

F-34060 Montpellier, France

(3) INRA, UE999 Pech-Rouge,

F-11430 Gruissan, France.

RESUME

Le lien fonctionnel entre typicité et terroir a été étudié en prenant en compte deux

dimensions importantes : le type de sol et la date de vendanges. Ces deux facteurs sont, à des

degrés divers, considérés comme facteurs explicatifs de l’effet terroir. Trois parcelles de

Cabernet franc, sur trois types de sols différents et revendiquant des AOP variées (“Anjou

Villages”, “Anjou Rouge” et “Saumur Champigny”) ont été vinifiées, en triplicata, à deux

dates espacées de 14 jours. Les vins, vinifiés selon un protocole standardisé, ont été analysés

sensoriellement par un jury de professionnels (question de typicité : Anjou Rouge vs Anjou

Villages) et par un jury expert (profil conventionnel). Pour évaluer la notion de maturité

phénolique (teneur et aptitude à l’extraction), les composés phénoliques ont été analysés à la

vendange, au décuvage, mais également au moment de l’analyse sensorielle.

Les résultats montrent que si le type de sol a permis de discriminer les profils sensoriels des

vins, son effet sur la typicité a été faible. La date de vendanges, au contraire, a permis de

discriminer les profils sensoriels mais également les notes de typicité. Concernant les

composés phénoliques, si la teneur et la composition en anthocyanes était dépendante de la

date de vendanges, elle n’a pas été explicative de la typicité, sauf quand les anthocyanes

totales ont été mesurées lors de l’analyse sensorielle (effet couleur). La quantité de tanins

condensés n’est pas apparue dépendante des parcelles mais de la date de vendanges. La

qualité des tanins contenus dans le vin au décuvage s’est révélée différente selon la date de

vendanges et explicative de la typicité. Enfin, la couleur des vins, liée à leur composition en

composés phénoliques, a influencé la perception de la typicité.

Cette étude illustre l’importance de certaines pratiques dans l’effet terroir, le type de sol

ayant un effet direct beaucoup moins important que ne laissent supposer les résultats

d’enquêtes auprès des producteurs.

MOTS CLES

Terroir – Typicité - Tanins condensés – Anthocyanes – Cabernet franc - Vitis vinifera

ABSTRACT

Harvest date is a critical point to the winemaker, in order to produce wine with a distinctive

style. In particular the relation between ripening stage and extractability of flavonoids must be

highlighted.

The extractability of flavonoids (flavan-3-ols, anthocyanins) from grapes was monitored at

two stages of maturity (veraison + 30 days, veraison + 44 days). Berries were obtained from

three plots with different types of soil in term of water status, from 3 AOC (Anjou-Villages-

Brissac, Anjou and Saumur Champigny) and were elaborated in triplicate. Flavonoids were

analysed before and after winemaking, by RP-LC-DAD, after fractionation and thiolysis for

the proanthocyanidins. Sensory analysis was performed eight month after harvest, by a

sensory expert panel (Quantitative descriptive analysis) and by wine experts, (assessment of

the typicality). Wine experts were producers, winemakers, and oenologists from the area.

The results showed that the type of soil allowed to discriminate the wines according to

some sensory attributes, but its effect on the typicality was weak. On the contrary, the date of

grape harvest, allowed discriminating the wine according to their sensory profiles and also to

their typicality scores. Concerning the flavonoids, if the content and the composition in

anthocyanins were dependent on the date of grape harvest, it was not connected to the

typicality, except when anthocyanins were analyzed just before sensory analysis. The quantity

of condensed tannins was not dependent on plots but on harvest date. The quality of tannins

contained in the wine at devatting was different according to hatvest date. Moreover, quantity

and quality of condensed tanins were highly correlated to the typicality scores. Finally, if the

anthocyanin contents of wines were correlated with typicality, the composition in the final

wine were not predicted by composition at devatting. The influence of anthocyanins seemed

to be due to perception of the color of wines in the typicality judgment.

This study illustrated the importance of harvest and vatting practices in the terroir effect,

with a soil effect less important as often admitted.

KEYWORDS

Terroir – Typicality – Condensed tanins – Anthocyanins - Cabernet franc - Vitis vinifera

INTRODUCTION

La date de récolte est un moment important dans le cycle de production : elle synthétise un

itinéraire viticole et suppose dans ses choix des anticipations quant au futur itinéraire

œnologique. Pour comprendre les déterminants des choix de dates de récolte, des critères

technologiques sont aisément identifiables. Ces critères technologiques sont plus ou moins

bien connus et/ou maîtrisés. Parmi les déterminants technologiques, la « maturité phénolique

» est un critère essentiel d’évaluation de la maturité. La maturité phénolique prend en compte

la teneur globale en polyphénols, mais aussi leur structure et leur aptitude à l’extraction

(Glories, 1998). Ainsi, la maturité phénolique peut être définie comme le niveau de maturité

permettant l’obtention simultanée d’un potentiel important et d’une bonne capacité de

diffusion dans le vin. Néanmoins, cette notion reste encore peu précise, car les évolutions au

cours de la maturation ne sont pas clairement établies, et les propriétés des tanins condensés et

les déterminants de leur extractibilité sont peu connus (Cadot, et al., 2006; Fournand et al.,

2006). De nombreux auteurs montrent l’importance de ces composés sur la qualité des

produits (visuelles, gustatives et somesthésiques), en particulier dans le cas des vins rouges

(Brossaud et al., 2001; Vidal et al., 2004). La synthèse de ces composés est fortement

dépendante de facteurs biotiques et abiotiques (Winkel-Shirley, 2002).

4 - 91



Tableau 1. Relation entre scores de typicité et parcelles

/ dates de vendange. AR : « Anjou Rouge ». AV : « Anjou-Villages ». ANOVA ; tests Newman-Keuls

Typicality "AR" Typicality "AV"Category LS means Groups Category LS means GroupsCYR 5.10 A LEB 3.53 ALEB 5.01 A CYR 3.25 ABMO 4.83 A BMO 2.84 AV44 5.37 A V44 4.19 AV30 4.59 B V30 2.22 B

Le sol est généralement mis en avant dans le système terroir. Ce système confère au vin des

caractéristiques particulières, la typicité. La typicité d’un produit peut être caractérisée par des

propriétés d’appartenance à un type, mais également des propriétés de distinction (Casabianca

et al., 2005). La typicité liée au terroir est associée à une origine géographique délimitée, et

présente des caractéristiques repérables et revendiquées (Cadot, 2006).

Nous faisons l’hypothèse que si le sol peut influencer la qualité du produit, ce sont certains

actes techniques saillants qui expliquent le mieux la qualité finale du vin. Dans le cas de cette

étude, nous avons évalué l’influence des actes techniques associés à la notion de « maturité

phénolique ».

MATERIEL ET METHODES

Le dispositif expérimental était constitué de 3 parcelles sur lesquelles le producteur a

conduit les vignes dans l’optique de produire des vins revendiquant les AOC suivantes :

« Saumur Champigny » (CYR), « Anjou » (BMO) et « Anjou Villages-Brissac » (LEB). Les

récoltes ont été réalisées à deux dates espacées de 14 jours : véraison + 35 jours (V30) et

véraison + 49 jours (V44) (29 septembre 2008 et 13 octobre 2008). Les vinifications ont été

réalisées en triple. La durée de cuvaison était de 9 jours, correspondant à une valeur commune

dans la région. 200 baies (encuvage - décuvage) ont été analysées selon le protocole décrit par

Roggero pour les anthocyanes (Roggero et al., 1992) et par Cadot pour les tanins condensés

(Cadot et al., 2006). Le rendement de thiolyse a été évalué en calculant de rapport entre

l’intégration de l’extrait après fractionement avant thiolyse et après thiolyse, afin de rendre

compte d’une possible évolution du rendement de thiolyse en fonction du niveau de maturité

(Kennedy & Jones, 2001). Au moment de l’analyse sensorielle (juin 2009), les vins ont

également été analysés, selon des méthodes classiques. Sur les vins finis, deux analyses

sensorielles ont été réalisées : (i) un profil conventionnel par un jury expert, méthode dérivée

de l’analyse quantitative descriptive (QDA), (Stone, 1974), (ii) deux questions de typicité, par

un jury de professionnels, composé de vignerons, œnologues et techniciens de la zone

d’étude, selon une méthode dérivée de Ballester (Ballester et al., 2005). Le dispositif

expérimental concernant les analyses sensorielles a été décrit précédemment par Cadot (Cadot

et al., 2010). Treize descripteurs ont été générés par le jury expert (2 visuels, 8 olfactifs, 3

gustatifs et somesthésiques). Les traitements statistiques ont été réalisés avec le logiciel

XLSTAT® 2009 (ADDINSOFT, France).

RESULTATS – DISCUSSION

L’origine des vins n’a pas eu d’incidence sur les notes de typicité, au contraire des dates de

vendanges : les récoltes les plus tardives ont été jugées plus typiques pour les deux tests.

(Tableau 1). Il est clair que le niveau de maturité plus avancé a eu une influence sur la qualité

des vins (qualité au sens de composition spécifique). Cet effet été plus net pour la typicité «

Anjou-Villages ».En effet, cette Appellation d’Origine Contrôlée (AOC) est considérée

comme une AOC « premium ». Néanmoins plusieurs descripteurs du profil conventionnel ont

été significativement reliés à l’origine des vins (Tableau 2). LEB a été noté significativement

(+) foncé, (+) fruits cuits, (+) alcool, et (+)

astringent. BMO a été noté significativement

(+) cassis et (+) astringent. Enfin, CYR a été

noté significativement (–) intense, (+) soufre

et (+) acide. Concernant les dates de

vendanges, V30 a été noté (+) clair, (+)

rouge orangé, (+) herbacé et (+) humus, ce

Tableau 2. Profil sensoriel. Résultats de l’ANOVA. Les

probabilités inférieures à 0.05 sont notées en gras.

Attributes Model Judge Origin Date Origin*Date

Colour intensity <0.0001 <0.0001 <0.0001 <0.0001 0.001

Shade <0.0001 <0.0001 0.767 <0.0001 0.131

Herbaceous <0.0001 <0.0001 0.063 0.010 0.424

Blackcurrant <0.0001 <0.0001 0.0001 0.492 0.172

Other red fruits <0.0001 <0.0001 0.571 0.114 0.266

Cooked fruits <0.0001 <0.0001 0.002 0.0004 0.633

Spice <0.0001 <0.0001 0.142 0.034 0.542

Humus <0.0001 <0.0001 0.465 <0.0001 0.360

Alcohol <0.0001 <0.0001 0.024 0.022 0.787

Sulfur <0.0001 <0.0001 <0.0001 0.174 0.001

Acid <0.0001 <0.0001 0.016 0.636 0.978

Bitter <0.0001 <0.0001 0.337 0.036 0.368

Astringent <0.0001 <0.0001 <0.0001 0.659 0.392

qui est en accord avec le niveau de maturité plus faible. V44 a été noté (+) foncé, (+) violet,

(+) fruits cuits, (+) épices, (+) alcool et (+) amer. Les analyses biochimiques ont confirmé ces

résultats concernant l’effet date de vendanges. En effet, les vins des dates V44 étaient

significativement plus alcoolisés, plus

riches en anthocyanes, à nuance bleue

plus marquée (A620) et moins acides

(Tableau 3). La figure 2 illustre le lien

entre la teneur en tanins condensés et (i)

la typicité « Anjou Villages » d’une

part, (ii), la date de récolte et l’origine

des raisins d’autre part. La teneur en

tanins condensés a eu des conséquences

sur la typicité, et les différences entre

teneurs ne s’expliquent pas par

l’origine, mais par les dates de récolte.

Une teneur plus élevée en tanins

condensés au décuvage, expliquée par

une date de récolte plus tardive a eu comme conséquence une note de typicité plus élevée. De

la même façon, la figure 3 illustre le lien entre le degré moyen de polymérisation (mDP) des

tanins condensés des tanins extraits (décuvage) et (i) la typicité « Anjou Villages » d’une part,

(ii), la date de récolte et l’origine des raisins d’autre part. Un mDP des vins au décuvage, plus

élevé, a eu des conséquences positives sur la note de typicité. Les différences entre mDP ne

s’expliquent pas par l’origine des parcelles, mais par les dates de récolte : les mDP au

décuvage ont été significativement plus élevés

pour les vendanges tardives. Le tableau 3

résume l’ensemble des relations entre la

composition biochimique des vins, la note de

typicité, la date de récolte et l’origine des

raisins. Le rendement de thiolyse a été

significativement plus faible lors des

vendanges plus tardives, confirmant

l’hypothèse émise par Kennedy, expliquant

que l’évolution des tanins condensés lors de la

maturation pourrait être en partie masquée par

une diminution du rendement de

dépolymérisation (Kennedy & Jones, 2001).

Globalement, la composition en tanins

condensés lors du décuvage (quantité et

composition spécifique) a été différente selon

la date de récolte, et elle a affecté la note de

typicité. L’origine des parcelles n’a pas été

mise en évidence, sauf pour la teneur en

catéchine des unités terminales. Lors d’études

précédentes sur Cabernet, aucune évolution

nette de la composition des raisins durant la

maturation n’a été mise en évidence

(Harbertson et al., 2002; Kennedy et al., 2002,

Fournand et al., 2006). Dans cette étude, la

Figure 1. Teneurs en tanins condensés (vin au décuvage).

Relation avec le score de typicité, la parcelle et la date de

vendanges. Box plots. AV+ : notes typicité sup. ; AV= :

notes typicité moy. ; AV – : notes typicité inf.

Figure 2. Degré de polymérisation des tanins condensés (vin

au décuvage). Relation avec le score de typicité, la parcelle

et la date de vendanges. Box plots. AV+ : notes typicité sup. ; AV= : notes typicité moy. ; AV – : notes typicité inf.

4 - 92



Tableau 1. Relation entre scores de typicité et parcelles

/ dates de vendange. AR : « Anjou Rouge ». AV : « Anjou-Villages ». ANOVA ; tests Newman-Keuls

Typicality "AR" Typicality "AV"Category LS means Groups Category LS means GroupsCYR 5.10 A LEB 3.53 ALEB 5.01 A CYR 3.25 ABMO 4.83 A BMO 2.84 AV44 5.37 A V44 4.19 AV30 4.59 B V30 2.22 B

Le sol est généralement mis en avant dans le système terroir. Ce système confère au vin des

caractéristiques particulières, la typicité. La typicité d’un produit peut être caractérisée par des

propriétés d’appartenance à un type, mais également des propriétés de distinction (Casabianca

et al., 2005). La typicité liée au terroir est associée à une origine géographique délimitée, et

présente des caractéristiques repérables et revendiquées (Cadot, 2006).

Nous faisons l’hypothèse que si le sol peut influencer la qualité du produit, ce sont certains

actes techniques saillants qui expliquent le mieux la qualité finale du vin. Dans le cas de cette

étude, nous avons évalué l’influence des actes techniques associés à la notion de « maturité

phénolique ».

MATERIEL ET METHODES

Le dispositif expérimental était constitué de 3 parcelles sur lesquelles le producteur a

conduit les vignes dans l’optique de produire des vins revendiquant les AOC suivantes :

« Saumur Champigny » (CYR), « Anjou » (BMO) et « Anjou Villages-Brissac » (LEB). Les

récoltes ont été réalisées à deux dates espacées de 14 jours : véraison + 35 jours (V30) et

véraison + 49 jours (V44) (29 septembre 2008 et 13 octobre 2008). Les vinifications ont été

réalisées en triple. La durée de cuvaison était de 9 jours, correspondant à une valeur commune

dans la région. 200 baies (encuvage - décuvage) ont été analysées selon le protocole décrit par

Roggero pour les anthocyanes (Roggero et al., 1992) et par Cadot pour les tanins condensés

(Cadot et al., 2006). Le rendement de thiolyse a été évalué en calculant de rapport entre

l’intégration de l’extrait après fractionement avant thiolyse et après thiolyse, afin de rendre

compte d’une possible évolution du rendement de thiolyse en fonction du niveau de maturité

(Kennedy & Jones, 2001). Au moment de l’analyse sensorielle (juin 2009), les vins ont

également été analysés, selon des méthodes classiques. Sur les vins finis, deux analyses

sensorielles ont été réalisées : (i) un profil conventionnel par un jury expert, méthode dérivée

de l’analyse quantitative descriptive (QDA), (Stone, 1974), (ii) deux questions de typicité, par

un jury de professionnels, composé de vignerons, œnologues et techniciens de la zone

d’étude, selon une méthode dérivée de Ballester (Ballester et al., 2005). Le dispositif

expérimental concernant les analyses sensorielles a été décrit précédemment par Cadot (Cadot

et al., 2010). Treize descripteurs ont été générés par le jury expert (2 visuels, 8 olfactifs, 3

gustatifs et somesthésiques). Les traitements statistiques ont été réalisés avec le logiciel

XLSTAT® 2009 (ADDINSOFT, France).

RESULTATS – DISCUSSION

L’origine des vins n’a pas eu d’incidence sur les notes de typicité, au contraire des dates de

vendanges : les récoltes les plus tardives ont été jugées plus typiques pour les deux tests.

(Tableau 1). Il est clair que le niveau de maturité plus avancé a eu une influence sur la qualité

des vins (qualité au sens de composition spécifique). Cet effet été plus net pour la typicité «

Anjou-Villages ».En effet, cette Appellation d’Origine Contrôlée (AOC) est considérée

comme une AOC « premium ». Néanmoins plusieurs descripteurs du profil conventionnel ont

été significativement reliés à l’origine des vins (Tableau 2). LEB a été noté significativement

(+) foncé, (+) fruits cuits, (+) alcool, et (+)

astringent. BMO a été noté significativement

(+) cassis et (+) astringent. Enfin, CYR a été

noté significativement (–) intense, (+) soufre

et (+) acide. Concernant les dates de

vendanges, V30 a été noté (+) clair, (+)

rouge orangé, (+) herbacé et (+) humus, ce

Tableau 2. Profil sensoriel. Résultats de l’ANOVA. Les

probabilités inférieures à 0.05 sont notées en gras.

Attributes Model Judge Origin Date Origin*Date

Colour intensity <0.0001 <0.0001 <0.0001 <0.0001 0.001

Shade <0.0001 <0.0001 0.767 <0.0001 0.131

Herbaceous <0.0001 <0.0001 0.063 0.010 0.424

Blackcurrant <0.0001 <0.0001 0.0001 0.492 0.172

Other red fruits <0.0001 <0.0001 0.571 0.114 0.266

Cooked fruits <0.0001 <0.0001 0.002 0.0004 0.633

Spice <0.0001 <0.0001 0.142 0.034 0.542

Humus <0.0001 <0.0001 0.465 <0.0001 0.360

Alcohol <0.0001 <0.0001 0.024 0.022 0.787

Sulfur <0.0001 <0.0001 <0.0001 0.174 0.001

Acid <0.0001 <0.0001 0.016 0.636 0.978

Bitter <0.0001 <0.0001 0.337 0.036 0.368

Astringent <0.0001 <0.0001 <0.0001 0.659 0.392

qui est en accord avec le niveau de maturité plus faible. V44 a été noté (+) foncé, (+) violet,

(+) fruits cuits, (+) épices, (+) alcool et (+) amer. Les analyses biochimiques ont confirmé ces

résultats concernant l’effet date de vendanges. En effet, les vins des dates V44 étaient

significativement plus alcoolisés, plus

riches en anthocyanes, à nuance bleue

plus marquée (A620) et moins acides

(Tableau 3). La figure 2 illustre le lien

entre la teneur en tanins condensés et (i)

la typicité « Anjou Villages » d’une

part, (ii), la date de récolte et l’origine

des raisins d’autre part. La teneur en

tanins condensés a eu des conséquences

sur la typicité, et les différences entre

teneurs ne s’expliquent pas par

l’origine, mais par les dates de récolte.

Une teneur plus élevée en tanins

condensés au décuvage, expliquée par

une date de récolte plus tardive a eu comme conséquence une note de typicité plus élevée. De

la même façon, la figure 3 illustre le lien entre le degré moyen de polymérisation (mDP) des

tanins condensés des tanins extraits (décuvage) et (i) la typicité « Anjou Villages » d’une part,

(ii), la date de récolte et l’origine des raisins d’autre part. Un mDP des vins au décuvage, plus

élevé, a eu des conséquences positives sur la note de typicité. Les différences entre mDP ne

s’expliquent pas par l’origine des parcelles, mais par les dates de récolte : les mDP au

décuvage ont été significativement plus élevés

pour les vendanges tardives. Le tableau 3

résume l’ensemble des relations entre la

composition biochimique des vins, la note de

typicité, la date de récolte et l’origine des

raisins. Le rendement de thiolyse a été

significativement plus faible lors des

vendanges plus tardives, confirmant

l’hypothèse émise par Kennedy, expliquant

que l’évolution des tanins condensés lors de la

maturation pourrait être en partie masquée par

une diminution du rendement de

dépolymérisation (Kennedy & Jones, 2001).

Globalement, la composition en tanins

condensés lors du décuvage (quantité et

composition spécifique) a été différente selon

la date de récolte, et elle a affecté la note de

typicité. L’origine des parcelles n’a pas été

mise en évidence, sauf pour la teneur en

catéchine des unités terminales. Lors d’études

précédentes sur Cabernet, aucune évolution

nette de la composition des raisins durant la

maturation n’a été mise en évidence

(Harbertson et al., 2002; Kennedy et al., 2002,

Fournand et al., 2006). Dans cette étude, la

Figure 1. Teneurs en tanins condensés (vin au décuvage).

Relation avec le score de typicité, la parcelle et la date de

vendanges. Box plots. AV+ : notes typicité sup. ; AV= :

notes typicité moy. ; AV – : notes typicité inf.

Figure 2. Degré de polymérisation des tanins condensés (vin

au décuvage). Relation avec le score de typicité, la parcelle

et la date de vendanges. Box plots. AV+ : notes typicité sup. ; AV= : notes typicité moy. ; AV – : notes typicité inf.

4 - 93



composition des raisins a affecté leur

extractibilité, modifiant ainsi la composition du

vin au décuvage. L’évolution de la composition

des tanins condensés des baies semble mal

mesurée par les méthodes employées. Mais elle

est révélée par (i) une évolution du rendement

de thiolyse (ce qui suggère une modification de

la composition des baies), (ii) par une

composition différente après décuvage (ce qui

suggère une modification de leur extractibilité).

Ainsi, la notion de « maturité phénolique »

semble bien démontrée. Les variations de

composition en tanins condensés en fonction de

l’origine des baies, mises en évidence par des

études précédentes pouvaient s’expliquer par

les différences de niveau de maturité des baies

des zones récoltées et du matériel végétal

différent : porte greffe, clone et âge de la vigne,

différents en fonction des zones, (Brossaud, et

al., 1999). Dans le cas de cette étude, l’origine

des baies a eu un effet limité, voire nul.

CONCLUSION

Les résultats montrent que l’origine des vins

a permis de discriminer les profils sensoriels

des vins, mais que son effet sur la typicité a été faible. Au contraire, la date de vendanges a

permis de discriminer les profils sensoriels mais également les notes de typicité. Concernant

les composés phénoliques, la teneur et la composition en anthocyanes n’ont pas été

explicatives de la typicité, sauf quand les anthocyanes totales ont été mesurées lors de

l’analyse sensorielle (effet couleur). La quantité de tanins condensés n’est pas apparue

dépendante des parcelles mais de la date de vendanges. La « qualité » des tanins contenus

dans le vin au décuvage s’est révélée différente selon la date de vendanges et explicative de la

typicité. Enfin, la couleur des vins, liée à leur composition en composés phénoliques, a

influencé la perception de la typicité.

Cette étude illustre l’importance des pratiques liées à la notion de « maturité phénolique »

sur la typicité des vins. L’origine des parcelles a eu un effet beaucoup moins important que ne

laissent supposer l’idée, communément admise, de la prévalence du sol dans le système des

AOC.

REMERCIEMENTS

Les auteurs tiennent à remercier les vignerons propriétaires des parcelles expérimentales,

Marie-Hélène Bouvet et Anne Mège pour la prise en charge des prélèvements et des analyses

et Michel Cosneau pour la vinification et l’élevage des vins. Nous remercions

particulièrement Mr. Erik Picou pour son aide dans la réalisation des profils sensoriels ainsi

que tous les dégustateurs. Ces travaux ont été conduits par l’INRA, avec le soutien du Conseil

Régional des Pays de la Loire, de Viniflhor et d’InterLoire

Tableau 3. Relations entre composition des vins et

scores de typicité / dates de récolte. ANOVAs.

Seules les probabilités <0.05 sont reportées.

Biochem. Typicality Date Plot

Harvest date 0.020 N/A N/A

Plot N/A N/A

Total anthocyanins

Delphinidin 0.001 0.018

Cyanidin

Petunidin

Peonidin 0.001

Malvidin 0.005

Acetyled 0.012

Coumaroyled 0.0001 < 0.001 0.026

Condensed tanins 0.050 < 0.001

Thiolysis yield 0.002

Catechin 0.040

Epicatechin 0.022 < 0.0001

Epicatechin-3-gal. 0.031 0.006

Epigallocatechin 0.083 < 0.0001

mDP 0.016 < 0.0001

% galloylation

% prodelphinidin 0.005 < 0.0001

Total acidity 0.018 0.004

Alcohol 0.006 < 0.0001

Dry extract

Total phenolic index

Total anthocyanins 0.006 < 0.0001

A420 0.024

A520

A620 0.030 0.005

Wine at

sensory

analysis

Wine at

devatting

Pr (ANOVA)

BIBLIOGRAPHIE

Ballester, J., Dacremont, C., Le Fur, Y., & Etievant, P. (2005). The role of olfaction in the

elaboration and use of the Chardonnay wine concept. Food Quality and Preference, 16(4),

351-359.

Brossaud, F., Cheynier, V., Asselin, C., & Moutounet, M. (1999). Flavonoid compositional

differences of grapes among site test plantings of Cabernet franc. American Journal of Enology

and Viticulture, 50(3), 277-284.

Brossaud, F., Cheynier, V., & Noble, A. C. (2001). Bitterness and astringency of grape and wine

polyphenols. Australian Journal of Grape and Wine Research, 7(1), 33-39.

Cadot, Y. (2006). Le lien du vin au terroir : complexité du concept de typicité. Revue des

Oenologues, 118, 9-11.

Cadot, Y., Caille, S., Samson, A., Barbeau, G., & Cheynier, V. (2010). Sensory dimension of

wine typicality related to a terroir by Quantitative Descriptive Analysis, Just About Right

analysis and typicality assessment. Analytica Chimica Acta, 660(1-2), 53-62.

Cadot, Y., Miñana-Castelló, M. T., & Chevalier, M. (2006). Flavan-3-ol compositional changes in

grape berries (Vitis vinifera L. cv Cabernet Franc) before veraison, using two complementary

analytical approaches, HPLC reversed phase and histochemistry. Analytica Chimica Acta,

563(1-2), 65-75.

Casabianca, F., Sylvander, B., Noel, Y., Beranger, C., Coulon, J. B., & Roncin, F. (2005). Terroir

et typicité : deux concepts clés des appellations d'origine contrôlée, essai de définitions

scientifiques et opérationnelles. In I.-. INAO, Colloque International de restitution des travaux

de recherches sur les indications et appellations d'origine géographiques. Paris (France).

Fournand, D., Vicens, A., Sidhoum, L., Souquet, J. M., Moutounet, M., & Cheynier, V. (2006).

Accumulation and Extractability of Grape Skin Tannins and Anthocyanins at Different

Advanced Physiological Stages. Journal of Agricultural and Food Chemistry, 54(19), 7331-

7338.

Glories, Y. (1998). Les composés phénoliques. La notion de maturation phénolique. In P.

Ribereau-Gayon, Traité d'œnologie. Volume II. Chimie du vin stabilisation et traitements.

Paris: Dunod.

Harbertson, J. F., Kennedy, J. A., & Adams, D. O. (2002). Tannin in skins and seeds of Cabernet

Sauvignon, Syrah, and Pinot noir berries during ripening. American Journal of Enology and

Viticulture, 53(1), 54-59.

Kennedy, J. A., & Jones, G. P. (2001). Analysis of proanthocyanidin cleavage products following

acid-catalysis in the presence of excess phloroglucinol. Journal of Agricultural and Food

Chemistry, 49(4), 1740-1746.

Kennedy, J. A., Matthews, M. A., & Waterhouse, A. L. (2002). Effect of maturity and vine water

status on grape skin and wine flavonoids. American Journal of Enology and Viticulture, 53(4),

268-274.

Roggero, J. P., Archier, P., & Coen, S. (1992). Etude par CLHP des compositions phenolique et

anthocyanique d'un moût de raisin en fermentation. Sciences Des Aliments, 12, 37-46.

Stone, H. (1974). Sensory evaluation by quantitative descriptive analysis. Food Technology, 24-

28.

Vidal, S., Francis, L., Noble, A., Kwiatkowski, M., Cheynier, V., & Waters, E. (2004). Taste and

mouth-feel properties of different types of tannin-like polyphenolic compounds and

anthocyanins in wine. Analytica Chimica Acta, 513(1), 57-65.

Winkel-Shirley, B. (2002). Biosynthesis of flavonoids and effects of stress. Current Opinion in

Plant Biology, 5(3), 218-223.

4 - 94



composition des raisins a affecté leur

extractibilité, modifiant ainsi la composition du

vin au décuvage. L’évolution de la composition

des tanins condensés des baies semble mal

mesurée par les méthodes employées. Mais elle

est révélée par (i) une évolution du rendement

de thiolyse (ce qui suggère une modification de

la composition des baies), (ii) par une

composition différente après décuvage (ce qui

suggère une modification de leur extractibilité).

Ainsi, la notion de « maturité phénolique »

semble bien démontrée. Les variations de

composition en tanins condensés en fonction de

l’origine des baies, mises en évidence par des

études précédentes pouvaient s’expliquer par

les différences de niveau de maturité des baies

des zones récoltées et du matériel végétal

différent : porte greffe, clone et âge de la vigne,

différents en fonction des zones, (Brossaud, et

al., 1999). Dans le cas de cette étude, l’origine

des baies a eu un effet limité, voire nul.

CONCLUSION

Les résultats montrent que l’origine des vins

a permis de discriminer les profils sensoriels

des vins, mais que son effet sur la typicité a été faible. Au contraire, la date de vendanges a

permis de discriminer les profils sensoriels mais également les notes de typicité. Concernant

les composés phénoliques, la teneur et la composition en anthocyanes n’ont pas été

explicatives de la typicité, sauf quand les anthocyanes totales ont été mesurées lors de

l’analyse sensorielle (effet couleur). La quantité de tanins condensés n’est pas apparue

dépendante des parcelles mais de la date de vendanges. La « qualité » des tanins contenus

dans le vin au décuvage s’est révélée différente selon la date de vendanges et explicative de la

typicité. Enfin, la couleur des vins, liée à leur composition en composés phénoliques, a

influencé la perception de la typicité.

Cette étude illustre l’importance des pratiques liées à la notion de « maturité phénolique »

sur la typicité des vins. L’origine des parcelles a eu un effet beaucoup moins important que ne

laissent supposer l’idée, communément admise, de la prévalence du sol dans le système des

AOC.

REMERCIEMENTS

Les auteurs tiennent à remercier les vignerons propriétaires des parcelles expérimentales,

Marie-Hélène Bouvet et Anne Mège pour la prise en charge des prélèvements et des analyses

et Michel Cosneau pour la vinification et l’élevage des vins. Nous remercions

particulièrement Mr. Erik Picou pour son aide dans la réalisation des profils sensoriels ainsi

que tous les dégustateurs. Ces travaux ont été conduits par l’INRA, avec le soutien du Conseil

Régional des Pays de la Loire, de Viniflhor et d’InterLoire

Tableau 3. Relations entre composition des vins et

scores de typicité / dates de récolte. ANOVAs.

Seules les probabilités <0.05 sont reportées.

Biochem. Typicality Date Plot

Harvest date 0.020 N/A N/A

Plot N/A N/A

Total anthocyanins

Delphinidin 0.001 0.018

Cyanidin

Petunidin

Peonidin 0.001

Malvidin 0.005

Acetyled 0.012

Coumaroyled 0.0001 < 0.001 0.026

Condensed tanins 0.050 < 0.001

Thiolysis yield 0.002

Catechin 0.040

Epicatechin 0.022 < 0.0001

Epicatechin-3-gal. 0.031 0.006

Epigallocatechin 0.083 < 0.0001

mDP 0.016 < 0.0001

% galloylation

% prodelphinidin 0.005 < 0.0001

Total acidity 0.018 0.004

Alcohol 0.006 < 0.0001

Dry extract

Total phenolic index

Total anthocyanins 0.006 < 0.0001

A420 0.024

A520

A620 0.030 0.005

Wine at

sensory

analysis

Wine at

devatting

Pr (ANOVA)

BIBLIOGRAPHIE

Ballester, J., Dacremont, C., Le Fur, Y., & Etievant, P. (2005). The role of olfaction in the

elaboration and use of the Chardonnay wine concept. Food Quality and Preference, 16(4),

351-359.

Brossaud, F., Cheynier, V., Asselin, C., & Moutounet, M. (1999). Flavonoid compositional

differences of grapes among site test plantings of Cabernet franc. American Journal of Enology

and Viticulture, 50(3), 277-284.

Brossaud, F., Cheynier, V., & Noble, A. C. (2001). Bitterness and astringency of grape and wine

polyphenols. Australian Journal of Grape and Wine Research, 7(1), 33-39.

Cadot, Y. (2006). Le lien du vin au terroir : complexité du concept de typicité. Revue des

Oenologues, 118, 9-11.

Cadot, Y., Caille, S., Samson, A., Barbeau, G., & Cheynier, V. (2010). Sensory dimension of

wine typicality related to a terroir by Quantitative Descriptive Analysis, Just About Right

analysis and typicality assessment. Analytica Chimica Acta, 660(1-2), 53-62.

Cadot, Y., Miñana-Castelló, M. T., & Chevalier, M. (2006). Flavan-3-ol compositional changes in

grape berries (Vitis vinifera L. cv Cabernet Franc) before veraison, using two complementary

analytical approaches, HPLC reversed phase and histochemistry. Analytica Chimica Acta,

563(1-2), 65-75.

Casabianca, F., Sylvander, B., Noel, Y., Beranger, C., Coulon, J. B., & Roncin, F. (2005). Terroir

et typicité : deux concepts clés des appellations d'origine contrôlée, essai de définitions

scientifiques et opérationnelles. In I.-. INAO, Colloque International de restitution des travaux

de recherches sur les indications et appellations d'origine géographiques. Paris (France).

Fournand, D., Vicens, A., Sidhoum, L., Souquet, J. M., Moutounet, M., & Cheynier, V. (2006).

Accumulation and Extractability of Grape Skin Tannins and Anthocyanins at Different

Advanced Physiological Stages. Journal of Agricultural and Food Chemistry, 54(19), 7331-

7338.

Glories, Y. (1998). Les composés phénoliques. La notion de maturation phénolique. In P.

Ribereau-Gayon, Traité d'œnologie. Volume II. Chimie du vin stabilisation et traitements.

Paris: Dunod.

Harbertson, J. F., Kennedy, J. A., & Adams, D. O. (2002). Tannin in skins and seeds of Cabernet

Sauvignon, Syrah, and Pinot noir berries during ripening. American Journal of Enology and

Viticulture, 53(1), 54-59.

Kennedy, J. A., & Jones, G. P. (2001). Analysis of proanthocyanidin cleavage products following

acid-catalysis in the presence of excess phloroglucinol. Journal of Agricultural and Food

Chemistry, 49(4), 1740-1746.

Kennedy, J. A., Matthews, M. A., & Waterhouse, A. L. (2002). Effect of maturity and vine water

status on grape skin and wine flavonoids. American Journal of Enology and Viticulture, 53(4),

268-274.

Roggero, J. P., Archier, P., & Coen, S. (1992). Etude par CLHP des compositions phenolique et

anthocyanique d'un moût de raisin en fermentation. Sciences Des Aliments, 12, 37-46.

Stone, H. (1974). Sensory evaluation by quantitative descriptive analysis. Food Technology, 24-

28.

Vidal, S., Francis, L., Noble, A., Kwiatkowski, M., Cheynier, V., & Waters, E. (2004). Taste and

mouth-feel properties of different types of tannin-like polyphenolic compounds and

anthocyanins in wine. Analytica Chimica Acta, 513(1), 57-65.

Winkel-Shirley, B. (2002). Biosynthesis of flavonoids and effects of stress. Current Opinion in

Plant Biology, 5(3), 218-223.

4 - 95



POTENTIEL DES SOLS VITICOLES ET QUALITÉ DES VINS

BROUSSET JMa, PICQUE Da, GUERIN Lb, GOULET Eb,c, PERROT Na

aUMR 782, GMPA, INRA, AgroParisTech, F-78850 Thiverval-Grignon bIFV du Val de Loire, 42, rue G. Morel, F-49071 Beaucouzé / 46, Av. G. Eiffel, F-37095 Tours cedex 2

cInterLoire, 12, rue E. Pallu – F-37000 TOURS

RÉSUMÉ La qualité des vins dépend de différents facteurs et procédés, notamment de la nature des

terrains viticoles. Dans ce travail, nous avons cherché à établir les liens entre descripteurs pédologiques des parcelles et descripteurs sensoriels des vins. Sur la base de Classifications Ascendantes Hiérarchiques (CAH) et d’Analyses en Composante Principale (ACP), il a été possible d’établir des liens entre la nature des parcelles (sableuse, argileuse, sablo-graveuleuse) et certains descripteurs sensoriels des vins (chaleur, astringence, fruit noir) et plus globalement avec le type de vins élaborés.

Mots-clés : pédologie, type de vin, CAH, ACP.

ABSTRACT Wine quality depends on various factors and processes, including type of soil. In this study,

we sought to establish links between pedological data and sensory attributes of wines. Based on Hierarchical Ascendant Classification (HAC) and Principal Component Analysis (PCA), it was possible to establish links between the nature of the parcels (sandy, clayey, gravelly-sand) and some wine sensory descriptors (heat, astringency, black fruit) and more generally with the type of wines.

Keywords : pedology, Wine type, HAC, PCA.

1. INTRODUCTION L’élaboration de vins de qualité dépend notamment de la qualité des baies de raisin

récoltées, c'est-à-dire de leur degré de maturité, et de la conduite des fermentations. C’est la première étape du processus qui nous intéresse ici. L’évolution des raisins dépend de nombreux facteurs dont les principaux sont liés au climat et au sol (Garcia de Cortazar Atauri, 2006). Cependant, tandis que le climat est lié à un millésime, les données pédologiques sont peu variables dans le temps. D’après les travaux de Morlat (2001), les sols peuvent être regroupés selon certains paramètres (texture, profondeur, calcaire actif, …). L’ensemble de ces paramètres influence fortement le fonctionnement de la plante et donc la composition des baies. En se basant sur l’hypothèse de l’existence d’un potentiel qualitatif d’un sol, nous proposons d’essayer d’établir une relation entre type de sol et type de vin. Pour cela, des classifications des parcelles (données pédologiques) et des vins qui en sont issus (données sensorielles) ont été réalisées par des méthodes de Classification Ascendante Hiérarchique (CAH) et d’Analyse en Composante Principale (ACP). Par la suite, une identification des liens est entreprise entre d’une part les données pédologiques et les données sensorielles des vins et d’autre part entre classification des parcelles et classification des vins.

Le couplage des données pédologiques et des données d’analyse sensorielle des vins permettra d’obtenir une classification qui servira de base pour l’étude de la maturation des baies.

2. MATÉRIEL ET MÉTHODES

2.1 Le réseau de parcelles Cette étude s’appuie sur les données pédologiques collectées par la Cellule Terroir Viticole

d’Angers. Le réseau de parcelles, utilisé par l’Institut Français de la Vigne et du Vin (IFV) depuis une dizaine d’années, est situé en moyenne vallée de la Loire et le cépage est le Cabernet franc. Les 26 parcelles sont situées sur les 5 appellations « rouges » de la région (tableau 1). Les variables pédologiques utilisées pour cette classification sont issues de la bibliographie (Morlat, 2001) et d’un recueil d’expertise réalisé auprès d’experts de la filière (vignerons et communauté scientifique). Les éléments de texture et la quantité de calcaire ont été pris en compte sur 3 horizons (en surface, en profondeur et roche mère). Le pH et la profondeur du sol exploitable par les racines ont été également utilisés (tableau 2).

Tableau 1. Présentation des parcelles du réseau Cabernet franc de la moyenne vallée de la Loire

Région AOC Parcelles Bourgueil 201, 202, 203, 204, 205 St Nicolas de Bourgueil 206, 207, 208, 219 Touraine Chinon 210, 211, 212, 213, 218 Anjou PAP, LEB, BMO, CHAL, MAT, RAH, LEC, CHAU, MAR AnjouSaumur CYR, SOU, MB

Tableau 2. Variables pédologiques caractérisant les sols viticoles

AR

Gs

(%)

AR

Gp

(%)

RG

rm

(%)

Ss (%

)

Sp (%

)

Srm

(%)

Ls (

%)

Lp

(%)

Lrm

(%)

EG

s (%

)

EG

p (%

)

EG

rm

(%) pH

PRO

F (c

m)

CA

s(g

/kg)

CA

p(g

/kg)

CA

rm(g

/kg)

Des

crip

teur

s qua

ntita

tifs

Arg

ile (s

urfa

ce)

Arg

ile (p

rofo

ndeu

r)

Arg

ile (r

oche

mèr

e)

Sabl

e G

ross

ier

(sur

face

)

Sabl

e G

ross

ier

(pro

fond

eur)

Sabl

e G

ross

ier (

roch

e m

ère)

Lim

on (s

urfa

ce)

Lim

on (p

rofo

ndeu

r)

Lim

on (r

oche

mèr

e)

Elém

ents

gro

ssie

rs

(sur

face

)

Elém

ents

gro

ssie

rs

(pro

fond

eur)

Elém

ents

gro

ssie

rs

(roc

he m

ère)

pH

Prof

onde

ur d

u so

l ex

ploi

tabl

e pa

r les

raci

nes

Cal

caire

Act

if (s

urfa

ce)

Cal

caire

Act

if (p

rofo

ndeu

r)

Cal

caire

Act

if (r

oche

m

ère)

2.2. Analyse des vins Sur l’ensemble de ces parcelles, les raisins sont vinifiés de manière identique afin

d’introduire le minimum de biais sur le vin fini. Après l’élevage, les vins sont dégustés par un jury professionnel composé de 15 viticulteurs de la région. Plusieurs variables sont ainsi évaluées (chaleur, acidité, notes aromatiques, astringence et amertume).

2.3. Traitement des données Le principe de la CAH (Lebart et al., 1995) consiste à fusionner à chaque étape les deux

clusters les plus proches au sens de la distance choisie. Le processus de classification s’arrête quand les deux clusters restant fusionnent dans l’unique cluster qui contient toutes les observations. Une hiérarchie de partitions se présente donc sous la forme de dendrogramme.

L’Analyse en Composante Principale (ACP) est une méthode de représentation graphique en 2 dimensions d’un ensemble d’observations caractérisées par plus de deux paramètres. Le

4 - 96



POTENTIEL DES SOLS VITICOLES ET QUALITÉ DES VINS

BROUSSET JMa, PICQUE Da, GUERIN Lb, GOULET Eb,c, PERROT Na

aUMR 782, GMPA, INRA, AgroParisTech, F-78850 Thiverval-Grignon bIFV du Val de Loire, 42, rue G. Morel, F-49071 Beaucouzé / 46, Av. G. Eiffel, F-37095 Tours cedex 2

cInterLoire, 12, rue E. Pallu – F-37000 TOURS

RÉSUMÉ La qualité des vins dépend de différents facteurs et procédés, notamment de la nature des

terrains viticoles. Dans ce travail, nous avons cherché à établir les liens entre descripteurs pédologiques des parcelles et descripteurs sensoriels des vins. Sur la base de Classifications Ascendantes Hiérarchiques (CAH) et d’Analyses en Composante Principale (ACP), il a été possible d’établir des liens entre la nature des parcelles (sableuse, argileuse, sablo-graveuleuse) et certains descripteurs sensoriels des vins (chaleur, astringence, fruit noir) et plus globalement avec le type de vins élaborés.

Mots-clés : pédologie, type de vin, CAH, ACP.

ABSTRACT Wine quality depends on various factors and processes, including type of soil. In this study,

we sought to establish links between pedological data and sensory attributes of wines. Based on Hierarchical Ascendant Classification (HAC) and Principal Component Analysis (PCA), it was possible to establish links between the nature of the parcels (sandy, clayey, gravelly-sand) and some wine sensory descriptors (heat, astringency, black fruit) and more generally with the type of wines.

Keywords : pedology, Wine type, HAC, PCA.

1. INTRODUCTION L’élaboration de vins de qualité dépend notamment de la qualité des baies de raisin

récoltées, c'est-à-dire de leur degré de maturité, et de la conduite des fermentations. C’est la première étape du processus qui nous intéresse ici. L’évolution des raisins dépend de nombreux facteurs dont les principaux sont liés au climat et au sol (Garcia de Cortazar Atauri, 2006). Cependant, tandis que le climat est lié à un millésime, les données pédologiques sont peu variables dans le temps. D’après les travaux de Morlat (2001), les sols peuvent être regroupés selon certains paramètres (texture, profondeur, calcaire actif, …). L’ensemble de ces paramètres influence fortement le fonctionnement de la plante et donc la composition des baies. En se basant sur l’hypothèse de l’existence d’un potentiel qualitatif d’un sol, nous proposons d’essayer d’établir une relation entre type de sol et type de vin. Pour cela, des classifications des parcelles (données pédologiques) et des vins qui en sont issus (données sensorielles) ont été réalisées par des méthodes de Classification Ascendante Hiérarchique (CAH) et d’Analyse en Composante Principale (ACP). Par la suite, une identification des liens est entreprise entre d’une part les données pédologiques et les données sensorielles des vins et d’autre part entre classification des parcelles et classification des vins.

Le couplage des données pédologiques et des données d’analyse sensorielle des vins permettra d’obtenir une classification qui servira de base pour l’étude de la maturation des baies.

2. MATÉRIEL ET MÉTHODES

2.1 Le réseau de parcelles Cette étude s’appuie sur les données pédologiques collectées par la Cellule Terroir Viticole

d’Angers. Le réseau de parcelles, utilisé par l’Institut Français de la Vigne et du Vin (IFV) depuis une dizaine d’années, est situé en moyenne vallée de la Loire et le cépage est le Cabernet franc. Les 26 parcelles sont situées sur les 5 appellations « rouges » de la région (tableau 1). Les variables pédologiques utilisées pour cette classification sont issues de la bibliographie (Morlat, 2001) et d’un recueil d’expertise réalisé auprès d’experts de la filière (vignerons et communauté scientifique). Les éléments de texture et la quantité de calcaire ont été pris en compte sur 3 horizons (en surface, en profondeur et roche mère). Le pH et la profondeur du sol exploitable par les racines ont été également utilisés (tableau 2).

Tableau 1. Présentation des parcelles du réseau Cabernet franc de la moyenne vallée de la Loire

Région AOC Parcelles Bourgueil 201, 202, 203, 204, 205 St Nicolas de Bourgueil 206, 207, 208, 219 Touraine Chinon 210, 211, 212, 213, 218 Anjou PAP, LEB, BMO, CHAL, MAT, RAH, LEC, CHAU, MAR AnjouSaumur CYR, SOU, MB

Tableau 2. Variables pédologiques caractérisant les sols viticoles

AR

Gs

(%)

AR

Gp

(%)

RG

rm

(%)

Ss (%

)

Sp (%

)

Srm

(%)

Ls (

%)

Lp

(%)

Lrm

(%)

EG

s (%

)

EG

p (%

)

EG

rm

(%) pH

PRO

F (c

m)

CA

s(g

/kg)

CA

p(g

/kg)

CA

rm(g

/kg)

Des

crip

teur

s qua

ntita

tifs

Arg

ile (s

urfa

ce)

Arg

ile (p

rofo

ndeu

r)

Arg

ile (r

oche

mèr

e)

Sabl

e G

ross

ier

(sur

face

)

Sabl

e G

ross

ier

(pro

fond

eur)

Sabl

e G

ross

ier (

roch

e m

ère)

Lim

on (s

urfa

ce)

Lim

on (p

rofo

ndeu

r)

Lim

on (r

oche

mèr

e)

Elém

ents

gro

ssie

rs

(sur

face

)

Elém

ents

gro

ssie

rs

(pro

fond

eur)

Elém

ents

gro

ssie

rs

(roc

he m

ère)

pH

Prof

onde

ur d

u so

l ex

ploi

tabl

e pa

r les

raci

nes

Cal

caire

Act

if (s

urfa

ce)

Cal

caire

Act

if (p

rofo

ndeu

r)

Cal

caire

Act

if (r

oche

m

ère)

2.2. Analyse des vins Sur l’ensemble de ces parcelles, les raisins sont vinifiés de manière identique afin

d’introduire le minimum de biais sur le vin fini. Après l’élevage, les vins sont dégustés par un jury professionnel composé de 15 viticulteurs de la région. Plusieurs variables sont ainsi évaluées (chaleur, acidité, notes aromatiques, astringence et amertume).

2.3. Traitement des données Le principe de la CAH (Lebart et al., 1995) consiste à fusionner à chaque étape les deux

clusters les plus proches au sens de la distance choisie. Le processus de classification s’arrête quand les deux clusters restant fusionnent dans l’unique cluster qui contient toutes les observations. Une hiérarchie de partitions se présente donc sous la forme de dendrogramme.

L’Analyse en Composante Principale (ACP) est une méthode de représentation graphique en 2 dimensions d’un ensemble d’observations caractérisées par plus de deux paramètres. Le

4 - 97



calcul des composantes principales est effectué à partir de la matrice des corrélations. Le but est d’obtenir une représentation la plus fidèle possible de l’ensemble des observations. La représentativité des variables et des observations est jugée par la valeur du cosinus carré. Plus cette valeur est proche de 1, meilleure est la représentation dans le plan.

3. RÉSULTATS ET DISCUSSION

3.1. Classification des parcelles par CAH et ACP Le dendrogramme obtenu par la CAH (figure 1) montre une séparation claire en 4 groupes.

La parcelle CHAU (S4) se démarque et semble être un cas particulier compte tenu de la faible profondeur du sol exploitable par les racines. Les deux groupes suivants de parcelles se différencient par une appartenance à des terrains à dominante sableuse (S1/S2) ou à dominante argileuse (S3). Dans le premier cas, on trouve essentiellement des parcelles de la région de Touraine. Dans le second, les parcelles d’Anjou et du Saumurois sont les plus représentées. Deux sous groupes peuvent être définis au sein du groupe des parcelles sableuses par la quantité d’éléments grossiers présent dans le sol. On distingue ainsi des parcelles sableuses (S2) et des parcelles sablo graveleuses (S1).

Dendrogramme

MB

CH

AL

211

MA

TP

AP

CY

R21

320

3M

AR

LEB

CH

AU

SO

U21

020

620

2B

MO

212

208

204

218

201

207

205

LEC

219

RA

H

0

5000

10000

15000

20000

25000

30000

35000

Dis

sim

ilarit

é

Biplot (axes F1 et F2 : 57,12 %)

MB

CYR

SOU

MAR

CHAU

LEC

RAH

MAT

CHALBMO

LEB

PAP219

218 213

212

211

210

208

207

206

205

204

203

202201

CA Rm

CApCAs

pHProfondeur totale sol

éléments grossiers roche

mère

éléments grossiers

profondeur

éléments grossiers surface

%Lrm

%Lp

%Ls

%Srm%Sp

%Ss

%ARGrm

%ARGp%ARGs

-10

-5

0

5

10

15

20

-25 -20 -15 -10 -5 0 5 10 15 20

F1 (33,04 %)

F2 (2

4,08

%)

Figure 1 : Arbre de classification des parcelles en fonction des données pédologiques

Figure 2. Représentation des variables et des parcelles sur F1/F2 de l’ACP

Le plan factoriel F1/F2 de l’ACP (figure 2) réalisée sur les mêmes données pédologiques explique 57% de la variance. Sur la composante F1, une forte corrélation est observée entre les variables % Argile et % Limon (tout horizon de sol confondu) en nette opposition avec les variables % Sable. La deuxième composante met en évidence une influence de la profondeur du sol (PROF), du pH et des variables % Sable en opposition avec les Eléments Grossiers (% EG tout horizon de sol confondu). La projection des individus sur le plan factoriel de l’ACP nous permet également de caractériser différents groupes. Les groupes définis à partir de la CAH peuvent également être définis dans ce plan.

Tableau 3. Récapitulatif des résultats obtenus par CAH et ACP sur les données pédologiques

Groupes Parcelles Caractérisation S1 RAH, LEC, 205, 207, 219 Texture sablo-graveleuse S2 SOU, BMO, 201, 202, 204, 206, 208, 210, 212, 218 Texture sableuse S3 LEB, MAR, CYR, 203, 213, PAP, CHAL, MAT, MB, 211 Texture argileuse S4 CHAU Texture sablo-graveleuse + profondeur

S1 S3S4S2

S4

S1

S2

S3

Ainsi, la combinaison des résultats de ces deux analyses nous permet de déterminer 3 groupes et un cas particulier (tableau 3) et de les caractériser en fonction des variables pédologiques. Une classification similaire a été réalisée sans la parcelle CHAU afin d’en atténuer l’effet par l’intermédiaire de la variable « profondeur du sol », mais la classification reste inchangée.

3.2 Classification des vins par CAH La figure 3 présente le dendrogramme obtenu par la méthode de classification réalisée sur

les variables de dégustation des vins. Dendrogramme

CH

AL

RA

HC

HA

ULE

BC

YR

LEC

MA

RB

MO

PA

PM

AT

203

204

202

212

211

201

210

213

SO

U21

920

520

620

720

821

8

0

5

10

15

20

25

30

Dis

sim

ilarit

é

Figure 3. Arbre de classification des vin identifiés par leur nom de parcelle en fonction des données sensorielles

Trois groupes ont été identifiés par cette méthode (tableau 4) et permettent de définir 3 classes de vins ayant leurs caractéristiques propres. Les groupes V1 et V2 sont caractérisés par des parcelles de Touraine (SOU étant une exception) tandis que le groupe V3 est caractérisé par des parcelles d’Anjou.

Tableau 4. Groupes et caractéristiques des vins identifiés par leurs noms de parcelle sur la base les données sensorielles

Groupes Vins issus des parcelles Caractéristiques V1 (vert) 213, SOU, 219, 205, 206, 207, 208, 218 Peu d’alcool, arômes de début de maturité

V2 (jaune) 203, 204, 202, 212, 211, 201, 210 Amertume, arômes plus évolués

V3 (rouge) CHAL, RAH, CHAU, LEB, CYR, LEC, MAR, BMO, PAP, MAT Diversité des arômes, structure plus appréciée

Le groupe V1 présente des caractéristiques de vins légers, destinés à être bu rapidement (taux d’alcool plus faible, arômes de début de maturation, peu de tanins). A l’inverse, le groupe V3 présente des caractéristiques de vins plus structurés, pouvant être conservés quelques années (chaleur, tanins, arômes plus complexes). Le groupe V2 est un intermédiaire, avec une structure plus appréciée des dégustateurs que celle du groupe V1 mais une présence de l’amertume encore prononcée (signe d’une maturité phénolique plus faible que celle du groupe V3).

3.3. Liens entre caractéristiques pédologiques des parcelles et caractéristiques sensorielles des vins

Les variables issues de l’analyse sensorielle des vins ont été intégrées par groupe comme variables supplémentaires dans l’ACP réalisée sur les variables pédologiques. Elles ont été divisées en 3 groupes de manière à ce que les groupes correspondent aux différentes maturités des baies. La maturité technologique est caractérisée par le sucre et l’acidité, la maturité phénolique par l’astringence et l’amertume et la maturité aromatique par les notes

V3 V2 V1

4 - 98



calcul des composantes principales est effectué à partir de la matrice des corrélations. Le but est d’obtenir une représentation la plus fidèle possible de l’ensemble des observations. La représentativité des variables et des observations est jugée par la valeur du cosinus carré. Plus cette valeur est proche de 1, meilleure est la représentation dans le plan.

3. RÉSULTATS ET DISCUSSION

3.1. Classification des parcelles par CAH et ACP Le dendrogramme obtenu par la CAH (figure 1) montre une séparation claire en 4 groupes.

La parcelle CHAU (S4) se démarque et semble être un cas particulier compte tenu de la faible profondeur du sol exploitable par les racines. Les deux groupes suivants de parcelles se différencient par une appartenance à des terrains à dominante sableuse (S1/S2) ou à dominante argileuse (S3). Dans le premier cas, on trouve essentiellement des parcelles de la région de Touraine. Dans le second, les parcelles d’Anjou et du Saumurois sont les plus représentées. Deux sous groupes peuvent être définis au sein du groupe des parcelles sableuses par la quantité d’éléments grossiers présent dans le sol. On distingue ainsi des parcelles sableuses (S2) et des parcelles sablo graveleuses (S1).

Dendrogramme

MB

CH

AL

211

MA

TP

AP

CY

R21

320

3M

AR

LEB

CH

AU

SO

U21

020

620

2B

MO

212

208

204

218

201

207

205

LEC

219

RA

H

0

5000

10000

15000

20000

25000

30000

35000

Dis

sim

ilarit

é

Biplot (axes F1 et F2 : 57,12 %)

MB

CYR

SOU

MAR

CHAU

LEC

RAH

MAT

CHALBMO

LEB

PAP219

218 213

212

211

210

208

207

206

205

204

203

202201

CA Rm

CApCAs



mère


profondeur


%Lrm

%Lp

%Ls

%Srm%Sp

%Ss

%ARGrm

%ARGp%ARGs

-10

-5

0

5

10

15

20

-25 -20 -15 -10 -5 0 5 10 15 20

F1 (33,04 %)

F2 (2

4,08

%)

Figure 1 : Arbre de classification des parcelles en fonction des données pédologiques

Figure 2. Représentation des variables et des parcelles sur F1/F2 de l’ACP

Le plan factoriel F1/F2 de l’ACP (figure 2) réalisée sur les mêmes données pédologiques explique 57% de la variance. Sur la composante F1, une forte corrélation est observée entre les variables % Argile et % Limon (tout horizon de sol confondu) en nette opposition avec les variables % Sable. La deuxième composante met en évidence une influence de la profondeur du sol (PROF), du pH et des variables % Sable en opposition avec les Eléments Grossiers (% EG tout horizon de sol confondu). La projection des individus sur le plan factoriel de l’ACP nous permet également de caractériser différents groupes. Les groupes définis à partir de la CAH peuvent également être définis dans ce plan.

Tableau 3. Récapitulatif des résultats obtenus par CAH et ACP sur les données pédologiques

Groupes Parcelles Caractérisation S1 RAH, LEC, 205, 207, 219 Texture sablo-graveleuse S2 SOU, BMO, 201, 202, 204, 206, 208, 210, 212, 218 Texture sableuse S3 LEB, MAR, CYR, 203, 213, PAP, CHAL, MAT, MB, 211 Texture argileuse S4 CHAU Texture sablo-graveleuse + profondeur

S1 S3S4S2

S4

S1

S2

S3

Ainsi, la combinaison des résultats de ces deux analyses nous permet de déterminer 3 groupes et un cas particulier (tableau 3) et de les caractériser en fonction des variables pédologiques. Une classification similaire a été réalisée sans la parcelle CHAU afin d’en atténuer l’effet par l’intermédiaire de la variable « profondeur du sol », mais la classification reste inchangée.

3.2 Classification des vins par CAH La figure 3 présente le dendrogramme obtenu par la méthode de classification réalisée sur

les variables de dégustation des vins. Dendrogramme

CH

AL

RA

HC

HA

ULE

BC

YR

LEC

MA

RB

MO

PA

PM

AT

203

204

202

212

211

201

210

213

SO

U21

920

520

620

720

821

8

0

5

10

15

20

25

30

Dis

sim

ilarit

é

Figure 3. Arbre de classification des vin identifiés par leur nom de parcelle en fonction des données sensorielles

Trois groupes ont été identifiés par cette méthode (tableau 4) et permettent de définir 3 classes de vins ayant leurs caractéristiques propres. Les groupes V1 et V2 sont caractérisés par des parcelles de Touraine (SOU étant une exception) tandis que le groupe V3 est caractérisé par des parcelles d’Anjou.

Tableau 4. Groupes et caractéristiques des vins identifiés par leurs noms de parcelle sur la base les données sensorielles

Groupes Vins issus des parcelles Caractéristiques V1 (vert) 213, SOU, 219, 205, 206, 207, 208, 218 Peu d’alcool, arômes de début de maturité

V2 (jaune) 203, 204, 202, 212, 211, 201, 210 Amertume, arômes plus évolués

V3 (rouge) CHAL, RAH, CHAU, LEB, CYR, LEC, MAR, BMO, PAP, MAT Diversité des arômes, structure plus appréciée

Le groupe V1 présente des caractéristiques de vins légers, destinés à être bu rapidement (taux d’alcool plus faible, arômes de début de maturation, peu de tanins). A l’inverse, le groupe V3 présente des caractéristiques de vins plus structurés, pouvant être conservés quelques années (chaleur, tanins, arômes plus complexes). Le groupe V2 est un intermédiaire, avec une structure plus appréciée des dégustateurs que celle du groupe V1 mais une présence de l’amertume encore prononcée (signe d’une maturité phénolique plus faible que celle du groupe V3).

3.3. Liens entre caractéristiques pédologiques des parcelles et caractéristiques sensorielles des vins

Les variables issues de l’analyse sensorielle des vins ont été intégrées par groupe comme variables supplémentaires dans l’ACP réalisée sur les variables pédologiques. Elles ont été divisées en 3 groupes de manière à ce que les groupes correspondent aux différentes maturités des baies. La maturité technologique est caractérisée par le sucre et l’acidité, la maturité phénolique par l’astringence et l’amertume et la maturité aromatique par les notes

V3 V2 V1

4 - 99



aromatiques (végétal, fruit rouge, fruit noir, animal et empyreumatique) Les figures 4, 5 et 6 présentent les coordonnées des variables (pédologiques et sensorielles) dans le plan factoriel F1/F2 de l’ACP.

Variables (axes F1 et F2 : 57,12 %)

CA RmCAp

CAs



mèreéléments grossiers

profondeur


%Lrm

%Lp

%Ls

%Srm

%Sp

%Ss

%ARGrm

%ARGp%ARGs

Acidité

Chaleur - alcool

-1

-0,75

-0,5

-0,25

0

0,25

0,5

0,75

1

-1 -0,75 -0,5 -0,25 0 0,25 0,5 0,75 1

F1 (33,04 %)

F2 (2

4,08

%)

variables actives Variables supplémentaires


CA RmCAp

CAs




profondeur


%Lrm

%Lp

%Ls

%Srm

%Sp

%Ss

%ARGrm

%ARGp%ARGs

Amertume

Astringence

-1

-0,75

-0,5

-0,25

0

0,25

0,5

0,75

1

-1 -0,75 -0,5 -0,25 0 0,25 0,5 0,75 1

F1 (33,04 %)

F2 (2

4,08

%)



%ARGs

%ARGp

%ARGrm

%Ss

%Sp

%Srm

%Ls

%Lp

%Lrm



profondeur


mère

Profondeur totale sol

pH

CAsCAp

CA Rm

Fruits rouges

Fruits noirs

Note végétale

Note animale

Note empyreumatique

-1

-0,75

-0,5

-0,25

0

0,25

0,5

0,75

1

-1 -0,75 -0,5 -0,25 0 0,25 0,5 0,75 1

F1 (33,04 %)

F2 (2

4,08

%)


Figure 4, 5 et 6. Représentation des variables et des variables supplémentaires dans F1/F2

L’alcool est situé dans la partie supérieure droite du plan. Cette variable est anti-corrélée à une profondeur de sol importante et à la présence de sable. Dans le plan F1/F2, la variable acidité est mal représentée (cosinus carré proche de 0) ; son interprétation est donc impossible. L’astringence se situe à l’opposé des variables sable et profondeur du sol tandis que la variable amertume se situe à l’opposé des variables argile et limon. Les variables aromatiques se répartissent de manière logique avec une évolution selon l’axe 2 liée à la maturation des baies (végétal puis fruit rouge et enfin fruit noir). La note aromatique « fruit noir », représentative des vins issus de baies de maturité plus aboutie, est anti-corrélée avec la présence de sable et une profondeur du sol importante. Les notes empyreumatiques et animales, situées dans la partie supérieure droite du plan, sont également anti-corrélées avec la présence de sable et la profondeur du sol. Il semblerait que le sable et la profondeur du sol soient des facteurs limitants de la maturité et de la diversité des arômes.

D’une manière plus générale, la tendance qui ressort de ces 3 figures semble être la capacité plus prononcée des sols argileux à produire des vins plus structurés (chaleur, astringence, fruit noir), issus de baies de maturité plus poussée. Les sols sableux seraient plus propices à l’élaboration de vins légers.

La figure 7 est la représentation graphique des parcelles dans le plan factoriel F1/F2 réalisée sur les données pédologiques. Chacune des parcelles est représentée par la couleur correspondante aux groupes de vins issus de la classification sur les données sensorielles répertoriées dans le tableau 4. Une séparation est observée selon l’axe 1. Le groupe de vin V1 (vert) se situe principalement sur la gauche du plan factoriel, caractérisé par les variables % Sable et % d’Eléments Grossiers (parcelles sableuses et sablo graveleuses). Le groupe V3 (rouge) correspond essentiellement à des parcelles à tendance argileuse. Enfin, les vins du groupe V2 (jaune) se répartissent de manière hétérogène dans le plan factoriel. Ces observations confirment les tendances observées dans les figures 4, 5 et 6. Les vins plus structurés (V3) sont issus des parcelles les plus argileuses alors que les vins plus légers (V1) sont produits sur les parcelles à tendance sableuse.

Cependant, les pourcentages d’explication des plans ainsi que la représentation de certaines variables supplémentaires sont relativement faibles. Les résultats doivent être interprétés avec prudence ; nous ne pouvons parler que de tendances.

Observations (axes F1 et F2 : 57,12 %)

MBCYR

SOU

MAR

CHAU

LEC

RAH

MAT

CHALBMO

LEB

PAP219

218 213212

211

210208

207

206

205

204

203

202

201

-8

-6

-4

-2

0

2

4

6

8

-6 -4 -2 0 2 4 6

F1 (33,04 %)

F2 (2

4,08

%)

Figure 7. Positionnement des parcelles (groupe S1 à S4) dans le plan F1/F2 de l’ACP réalisée sur les données pédologiques. Les couleurs correspondent aux types de vins issus des parcelles.

4. CONCLUSION Nous avons pu identifier 4 groupes de parcelles dont la principale opposition se fait au

niveau de la texture (sableuse ou argileuse). On note également une importance du taux d’éléments grossiers en ce qui concerne les parcelles sableuses. En mettant en relation cette classification avec les données d’analyse sensorielle des vins, nous avons également montré un lien entre type de sol et type de vin. Les parcelles à dominante sableuses sont propices à l’élaboration de vins légers, nécessitant une maturation des baies moins poussée, tandis que les parcelles argileuses permettent d’envisager des vins plus structurés destinés à être conservés quelques temps.

Cependant, la significativité de ce lien est moyenne. La qualité des vin ne dépend pas uniquement des facteurs pédologiques. Le système est plus complexe à décrire et certaines variables (météorologie, techniques culturales) manquent pour mieux expliquer la qualité des vins.

Références Coombe, B.G., McCarthy, M.G., 2000, Dynamics of grape berry growth and physiology of

ripening. Aust. J. Grape Wine Res. 6:131-135. Garcia de Cortazar Atauri, I., 2006, Adaptation du modèle STICS à la vigne (Vitis vinifera

L.). Utilisation dans le cadre d’une étude d’impacte du changement climatique à l’échelle de la France. Thèse, Ecole Nationale Supérieure Agronomique de Montpellier.

Lebart, L., Morineau, A., Piron, M., 1995, Statistique exploratoire multidimensionnelle. Dunod, Paris.

Morlat, R., 2001, Le terroir viticole : Contribution à l'étude de sa caractérisation et de son influence sur les vins. Applications aux vignobles rouges de Moyenne Vallée de la Loire. Thèse de doctorat d'Etat de l’Université de Bordeaux.

SABLE ARGILE

S4

S1

S2

S3

4 - 100



aromatiques (végétal, fruit rouge, fruit noir, animal et empyreumatique) Les figures 4, 5 et 6 présentent les coordonnées des variables (pédologiques et sensorielles) dans le plan factoriel F1/F2 de l’ACP.


CA RmCAp

CAs




profondeur


%Lrm

%Lp

%Ls

%Srm

%Sp

%Ss

%ARGrm

%ARGp%ARGs

Acidité

Chaleur - alcool

-1

-0,75

-0,5

-0,25

0

0,25

0,5

0,75

1

-1 -0,75 -0,5 -0,25 0 0,25 0,5 0,75 1

F1 (33,04 %)

F2 (2

4,08

%)



CA RmCAp

CAs




profondeur


%Lrm

%Lp

%Ls

%Srm

%Sp

%Ss

%ARGrm

%ARGp%ARGs

Amertume

Astringence

-1

-0,75

-0,5

-0,25

0

0,25

0,5

0,75

1

-1 -0,75 -0,5 -0,25 0 0,25 0,5 0,75 1

F1 (33,04 %)

F2 (2

4,08

%)



%ARGs

%ARGp

%ARGrm

%Ss

%Sp

%Srm

%Ls

%Lp

%Lrm



profondeur


mère

Profondeur totale sol

pH

CAsCAp

CA Rm

Fruits rouges

Fruits noirs

Note végétale

Note animale

Note empyreumatique

-1

-0,75

-0,5

-0,25

0

0,25

0,5

0,75

1

-1 -0,75 -0,5 -0,25 0 0,25 0,5 0,75 1

F1 (33,04 %)

F2 (2

4,08

%)


Figure 4, 5 et 6. Représentation des variables et des variables supplémentaires dans F1/F2

L’alcool est situé dans la partie supérieure droite du plan. Cette variable est anti-corrélée à une profondeur de sol importante et à la présence de sable. Dans le plan F1/F2, la variable acidité est mal représentée (cosinus carré proche de 0) ; son interprétation est donc impossible. L’astringence se situe à l’opposé des variables sable et profondeur du sol tandis que la variable amertume se situe à l’opposé des variables argile et limon. Les variables aromatiques se répartissent de manière logique avec une évolution selon l’axe 2 liée à la maturation des baies (végétal puis fruit rouge et enfin fruit noir). La note aromatique « fruit noir », représentative des vins issus de baies de maturité plus aboutie, est anti-corrélée avec la présence de sable et une profondeur du sol importante. Les notes empyreumatiques et animales, situées dans la partie supérieure droite du plan, sont également anti-corrélées avec la présence de sable et la profondeur du sol. Il semblerait que le sable et la profondeur du sol soient des facteurs limitants de la maturité et de la diversité des arômes.

D’une manière plus générale, la tendance qui ressort de ces 3 figures semble être la capacité plus prononcée des sols argileux à produire des vins plus structurés (chaleur, astringence, fruit noir), issus de baies de maturité plus poussée. Les sols sableux seraient plus propices à l’élaboration de vins légers.

La figure 7 est la représentation graphique des parcelles dans le plan factoriel F1/F2 réalisée sur les données pédologiques. Chacune des parcelles est représentée par la couleur correspondante aux groupes de vins issus de la classification sur les données sensorielles répertoriées dans le tableau 4. Une séparation est observée selon l’axe 1. Le groupe de vin V1 (vert) se situe principalement sur la gauche du plan factoriel, caractérisé par les variables % Sable et % d’Eléments Grossiers (parcelles sableuses et sablo graveleuses). Le groupe V3 (rouge) correspond essentiellement à des parcelles à tendance argileuse. Enfin, les vins du groupe V2 (jaune) se répartissent de manière hétérogène dans le plan factoriel. Ces observations confirment les tendances observées dans les figures 4, 5 et 6. Les vins plus structurés (V3) sont issus des parcelles les plus argileuses alors que les vins plus légers (V1) sont produits sur les parcelles à tendance sableuse.

Cependant, les pourcentages d’explication des plans ainsi que la représentation de certaines variables supplémentaires sont relativement faibles. Les résultats doivent être interprétés avec prudence ; nous ne pouvons parler que de tendances.

Observations (axes F1 et F2 : 57,12 %)

MBCYR

SOU

MAR

CHAU

LEC

RAH

MAT

CHALBMO

LEB

PAP219

218 213212

211

210208

207

206

205

204

203

202

201

-8

-6

-4

-2

0

2

4

6

8

-6 -4 -2 0 2 4 6

F1 (33,04 %)

F2 (2

4,08

%)

Figure 7. Positionnement des parcelles (groupe S1 à S4) dans le plan F1/F2 de l’ACP réalisée sur les données pédologiques. Les couleurs correspondent aux types de vins issus des parcelles.

4. CONCLUSION Nous avons pu identifier 4 groupes de parcelles dont la principale opposition se fait au

niveau de la texture (sableuse ou argileuse). On note également une importance du taux d’éléments grossiers en ce qui concerne les parcelles sableuses. En mettant en relation cette classification avec les données d’analyse sensorielle des vins, nous avons également montré un lien entre type de sol et type de vin. Les parcelles à dominante sableuses sont propices à l’élaboration de vins légers, nécessitant une maturation des baies moins poussée, tandis que les parcelles argileuses permettent d’envisager des vins plus structurés destinés à être conservés quelques temps.

Cependant, la significativité de ce lien est moyenne. La qualité des vin ne dépend pas uniquement des facteurs pédologiques. Le système est plus complexe à décrire et certaines variables (météorologie, techniques culturales) manquent pour mieux expliquer la qualité des vins.

Références Coombe, B.G., McCarthy, M.G., 2000, Dynamics of grape berry growth and physiology of

ripening. Aust. J. Grape Wine Res. 6:131-135. Garcia de Cortazar Atauri, I., 2006, Adaptation du modèle STICS à la vigne (Vitis vinifera

L.). Utilisation dans le cadre d’une étude d’impacte du changement climatique à l’échelle de la France. Thèse, Ecole Nationale Supérieure Agronomique de Montpellier.

Lebart, L., Morineau, A., Piron, M., 1995, Statistique exploratoire multidimensionnelle. Dunod, Paris.

Morlat, R., 2001, Le terroir viticole : Contribution à l'étude de sa caractérisation et de son influence sur les vins. Applications aux vignobles rouges de Moyenne Vallée de la Loire. Thèse de doctorat d'Etat de l’Université de Bordeaux.

SABLE ARGILE

S4

S1

S2

S3

4 - 101



INFLUENZA DI ALCUNI ASPETTI AMBIENTALI SUL CONTENUTO DISTILBENI NEL VINO NELL’AREA DELLA DOC “SANGIOVESE DI

ROMAGNA” (ITALIA)

Matteo GATTI(1,2), Silvia CIVARDI(2), Maurizio ZAMBONI(2), Luigi BAVARESCO(2),Federico FERRARI(3), Stefano RAIMONDI(4), Marco SIMONI(5), Driss ELOTHMANI(1),

Frédérique JOURJON(1)

(1) Laboratoire GRAPPE, Ecole Supérieure d’Agriculture55 rue Rabelais, B.P. 30748, 49007 Angers Cedex 01, France

[email protected]; [email protected](2)Istituto di Frutti-Viticoltura, Università Cattolica del Sacro Cuore

Via E. Parmense 84, 29199 Piacenza, [email protected]; [email protected]; [email protected]; [email protected]

(3) Istituto di Chimica Agraria e Ambientale, Università Cattolica del Sacro CuoreVia E. Parmense 84, 29199 Piacenza, Italia

[email protected](4)I.TER Soc. coop.

Via Brugnoli 11, 40122 Bologna, [email protected]

(5) ASTRA Innovazione e Sviluppo s.r.l.Via Tebano 45, 48018 Faenza (RA), Italia

[email protected]

RIASSUNTONell’ambito della zonazione della Doc “Sangiovese di Romagna” sono stati descritti 25 siti

sperimentali, aventi diversa origine geologica, in cui è stato individuato un vigneto omogeneo perla determinazione dei principali parametri viticoli ed enologici. In seguito è stato analizzato ilcontenuto di stilbeni nei vini al fine di approfondirne il legame con le caratteristichegeopedologiche. Lo studio descrive la relazione positiva tra l’altitudine e il contenuto di trans-piceide nelle province di Forlì e Ravenna e di trans-resveratrolo a Ravenna. I suoli con maggiorecalcare attivo hanno fornito vini più ricchi in stilbeni.

PAROLE CHIAVEFormazione geologica – Calcare attivo - Stilbeni – Sangiovese

ABSTRACTThe “Sangiovese di Romagna” zoning characterized 25 sites, based on a different geological

origin. For each site, a representative commercial vineyard was chosen and the main viticulturaland oenological parameters were recorded. The wine stilbene content was analyzed to investigatethe effect of the geological origin and the soil composition. Positive relations between siteelevation and trans-resveratrol and site elevation and trans-piceid were observed in the Ravennaand, Forlì and Ravenna area, respectively. The higher the active lime in the soils the richer thestilbenes in the wines.

KEYWORDGeology – Active lime – Stilbenes – Sangiovese

INTRODUZIONEGli stilbeni sono composti fenolici a basso peso molecolare presenti in numerose famiglie di

piante, tra cui le Vitaceae, aventi carattere costitutivo oppure inducibile dall’esposizione a fattoridi stress. Da lungo tempo è assodato che il trans-resveratrolo, il primo ad essere stato scoperto(Langcake, Price, 1977) e quindi il più noto, svolge un importante ruolo antifungino(fitoalessina), ma recenti sperimentazioni lo hanno indicato come importante antiossidantenaturale (Sienmann, Creasy, 1992; Renaud, de Lorgeril, 1992), nonché come possibile fattore dilongevità della pianta, poiché implicato nel metabolismo delle sirtuine della vite (Busconi et al.,2009). Gli stilbeni rivestono pertanto un’importanza notevole sia sul piano della vitalità delvigneto che dal punto di vista salutistico e nutraceutico rappresentando un parametro qualitativodell’uva e dei vini che da essa ne derivano. Nella vite, la sintesi di dette molecole è legataall’attivazione della Stilbene sintasi (Rupprich, Kindl, 1978) da parte di elicitori biotici, tra cui leprincipali crittogame (Jeandet et al., 1995; Bavaresco et al., 1997, 2008a), ma anche abioticilegati alle condizioni ambientali (Bavaresco et al. 2005, 2007; de Andrés-de Prado et al., 2007) ecolturali (Bavaresco et al., 2008b). Poiché le potenzialità viticole di un ambiente dipendono dalleinterazioni che legano il vitigno a specifiche condizioni climatiche, geopedologiche,paesaggistiche e colturali così come indicato dalla definizione di terroir, recentemente adottatadall’Oiv, si ritiene opportuno inserire nell’ambito della zonazione viticola per l’individuazionedelle terre più idonee alla coltura della vite, lo studio di alcuni metaboliti secondari al fine diindividuarne il legame con il territorio, indipendentemente dall’intensità della pressione fungina(Adrian et al., 2000).

Il presente lavoro vuole approfondire l’eventuale interazione esistente tra l’origine geologicadel suolo e il tenore di stilbeni del vino, finora sconosciuta, nonché verificare in pieno campol’effetto del calcare attivo e di alcune componenti paesaggistiche.

MATERIALI E METODINell’ambito di un progetto di zonazione viticola della Collina Romagnola, coordinato dal

C.R.P.V. e riferito, nello specifico, all’area di produzione del “Sangiovese di Romagna”, sonostati individuati 25 siti omogenei nelle colline di Faenza (otto), Forlì-Cesena (dieci) e Rimini(sette). La scelta dei vigneti ha preso in considerazione le principali unità geologichedell’Appennino romagnolo formatesi tra la fine del Terziario e l’inizio del Quaternario ovvero, laFormazione Marnoso Arenacea (FMA), la Formazione delle Argille Azzurre (FAA) e la piùrecente Formazione del Margine Appenninico (AES). I vigneti, impiantati a Sangiovese biotiporomagnolo spesso derivato da materiale standard e innestato su ibridi Berlandieri per Riparia,erano allevati a cordone speronato con una densità d’impianto di circa 3500 ceppi/ha e distanzamedia tra i filari pari a 3 m. I siti osservati erano collocati a un’altitudine variabile tra i 45 e i 280m s.l.m. con pendenza differente a seconda della relativa unità geologica. Lo studio dell’effettodella formazione geologica è stato epurato dell’interazione con l’altitudine individuando due zonealtimetriche (< 100 m e 100-160 m) entro le quali sono stati considerati tre siti per ciascuna unitàgeologica; solo per FMA è stata considerata unicamente l’altimetria superiore. Annualmente sonostate rilevate le principali caratteristiche vegeto-produttive delle viti e, alla vendemmia, è stataprelevata una quota di uva pari a circa 80 kg, poi micro-vinificata. Sui vini ottenuti è stataeseguita la determinazione dei principali parametri della qualità e, nel maggio 2009, dopocentrifugazione di 5 minuti a 3500 rpm, è stato dosato il contenuto degli stilbeni nei vini dellavendemmia 2008 per iniezione diretta in HPLC (Agilent HP 1100 – Waldbronn, Germania)

4 - 102



INFLUENZA DI ALCUNI ASPETTI AMBIENTALI SUL CONTENUTO DISTILBENI NEL VINO NELL’AREA DELLA DOC “SANGIOVESE DI

ROMAGNA” (ITALIA)

Matteo GATTI(1,2), Silvia CIVARDI(2), Maurizio ZAMBONI(2), Luigi BAVARESCO(2),Federico FERRARI(3), Stefano RAIMONDI(4), Marco SIMONI(5), Driss ELOTHMANI(1),

Frédérique JOURJON(1)

(1) Laboratoire GRAPPE, Ecole Supérieure d’Agriculture55 rue Rabelais, B.P. 30748, 49007 Angers Cedex 01, France

[email protected]; [email protected](2)Istituto di Frutti-Viticoltura, Università Cattolica del Sacro Cuore

Via E. Parmense 84, 29199 Piacenza, [email protected]; [email protected]; [email protected]; [email protected]

(3) Istituto di Chimica Agraria e Ambientale, Università Cattolica del Sacro CuoreVia E. Parmense 84, 29199 Piacenza, Italia

[email protected](4)I.TER Soc. coop.

Via Brugnoli 11, 40122 Bologna, [email protected]

(5) ASTRA Innovazione e Sviluppo s.r.l.Via Tebano 45, 48018 Faenza (RA), Italia

[email protected]

RIASSUNTONell’ambito della zonazione della Doc “Sangiovese di Romagna” sono stati descritti 25 siti

sperimentali, aventi diversa origine geologica, in cui è stato individuato un vigneto omogeneo perla determinazione dei principali parametri viticoli ed enologici. In seguito è stato analizzato ilcontenuto di stilbeni nei vini al fine di approfondirne il legame con le caratteristichegeopedologiche. Lo studio descrive la relazione positiva tra l’altitudine e il contenuto di trans-piceide nelle province di Forlì e Ravenna e di trans-resveratrolo a Ravenna. I suoli con maggiorecalcare attivo hanno fornito vini più ricchi in stilbeni.

PAROLE CHIAVEFormazione geologica – Calcare attivo - Stilbeni – Sangiovese

ABSTRACTThe “Sangiovese di Romagna” zoning characterized 25 sites, based on a different geological

origin. For each site, a representative commercial vineyard was chosen and the main viticulturaland oenological parameters were recorded. The wine stilbene content was analyzed to investigatethe effect of the geological origin and the soil composition. Positive relations between siteelevation and trans-resveratrol and site elevation and trans-piceid were observed in the Ravennaand, Forlì and Ravenna area, respectively. The higher the active lime in the soils the richer thestilbenes in the wines.

KEYWORDGeology – Active lime – Stilbenes – Sangiovese

INTRODUZIONEGli stilbeni sono composti fenolici a basso peso molecolare presenti in numerose famiglie di

piante, tra cui le Vitaceae, aventi carattere costitutivo oppure inducibile dall’esposizione a fattoridi stress. Da lungo tempo è assodato che il trans-resveratrolo, il primo ad essere stato scoperto(Langcake, Price, 1977) e quindi il più noto, svolge un importante ruolo antifungino(fitoalessina), ma recenti sperimentazioni lo hanno indicato come importante antiossidantenaturale (Sienmann, Creasy, 1992; Renaud, de Lorgeril, 1992), nonché come possibile fattore dilongevità della pianta, poiché implicato nel metabolismo delle sirtuine della vite (Busconi et al.,2009). Gli stilbeni rivestono pertanto un’importanza notevole sia sul piano della vitalità delvigneto che dal punto di vista salutistico e nutraceutico rappresentando un parametro qualitativodell’uva e dei vini che da essa ne derivano. Nella vite, la sintesi di dette molecole è legataall’attivazione della Stilbene sintasi (Rupprich, Kindl, 1978) da parte di elicitori biotici, tra cui leprincipali crittogame (Jeandet et al., 1995; Bavaresco et al., 1997, 2008a), ma anche abioticilegati alle condizioni ambientali (Bavaresco et al. 2005, 2007; de Andrés-de Prado et al., 2007) ecolturali (Bavaresco et al., 2008b). Poiché le potenzialità viticole di un ambiente dipendono dalleinterazioni che legano il vitigno a specifiche condizioni climatiche, geopedologiche,paesaggistiche e colturali così come indicato dalla definizione di terroir, recentemente adottatadall’Oiv, si ritiene opportuno inserire nell’ambito della zonazione viticola per l’individuazionedelle terre più idonee alla coltura della vite, lo studio di alcuni metaboliti secondari al fine diindividuarne il legame con il territorio, indipendentemente dall’intensità della pressione fungina(Adrian et al., 2000).

Il presente lavoro vuole approfondire l’eventuale interazione esistente tra l’origine geologicadel suolo e il tenore di stilbeni del vino, finora sconosciuta, nonché verificare in pieno campol’effetto del calcare attivo e di alcune componenti paesaggistiche.

MATERIALI E METODINell’ambito di un progetto di zonazione viticola della Collina Romagnola, coordinato dal

C.R.P.V. e riferito, nello specifico, all’area di produzione del “Sangiovese di Romagna”, sonostati individuati 25 siti omogenei nelle colline di Faenza (otto), Forlì-Cesena (dieci) e Rimini(sette). La scelta dei vigneti ha preso in considerazione le principali unità geologichedell’Appennino romagnolo formatesi tra la fine del Terziario e l’inizio del Quaternario ovvero, laFormazione Marnoso Arenacea (FMA), la Formazione delle Argille Azzurre (FAA) e la piùrecente Formazione del Margine Appenninico (AES). I vigneti, impiantati a Sangiovese biotiporomagnolo spesso derivato da materiale standard e innestato su ibridi Berlandieri per Riparia,erano allevati a cordone speronato con una densità d’impianto di circa 3500 ceppi/ha e distanzamedia tra i filari pari a 3 m. I siti osservati erano collocati a un’altitudine variabile tra i 45 e i 280m s.l.m. con pendenza differente a seconda della relativa unità geologica. Lo studio dell’effettodella formazione geologica è stato epurato dell’interazione con l’altitudine individuando due zonealtimetriche (< 100 m e 100-160 m) entro le quali sono stati considerati tre siti per ciascuna unitàgeologica; solo per FMA è stata considerata unicamente l’altimetria superiore. Annualmente sonostate rilevate le principali caratteristiche vegeto-produttive delle viti e, alla vendemmia, è stataprelevata una quota di uva pari a circa 80 kg, poi micro-vinificata. Sui vini ottenuti è stataeseguita la determinazione dei principali parametri della qualità e, nel maggio 2009, dopocentrifugazione di 5 minuti a 3500 rpm, è stato dosato il contenuto degli stilbeni nei vini dellavendemmia 2008 per iniezione diretta in HPLC (Agilent HP 1100 – Waldbronn, Germania)

4 - 103



abbinata a un detector a diodi (DAD) settato a 306 e 325 nm utilizzando una colonna C18Supelco Supelcosil.

I dati raccolti sono stati elaborati mediante l’analisi della varianza (ANOVA) a un criterio diclassificazione considerando come fonti di variazione la Formazione geologica e il Calcareattivo; le medie sono state comparate con il test SNK (p<0.05). La relazione esistente traaltitudine e tenore di stilbeni del vino è stata studiata mediante la regressione lineare.

RISULTATI E DISCUSSIONENei vini Sangiovese 2008 analizzati è stata ricercata la presenza di numerosi stilbeni

riscontrando concentrazioni interessanti di trans-piceide e trans-resveratrolo. Non sono statirilevati il pallidolo, il piceatannolo, δ- ed ε-viniferina nonchè la forma cis- di piceide eresveratrolo. Il trans-piceide è variato da concentrazioni pressoché nulle in alcuni vigneti delmargine appenninico a livelli considerevoli superiori a 6 mg/L di alcuni siti delle Argille azzurredel riminese. Le concentrazioni di trans-resveratrolo, seppur inferiori rispetto al precedentecomposto, sono oscillate tra 0.3 e 4.5 mg/L.

Il ruolo dell’altitudine sul tenore di trans-piceide e trans-resveratrolo è stato ricercato mediantela definizione di correlazioni lineari per le diverse province osservando coefficienti dicorrelazione altamente significativi (p<0.01) nel caso del trans-piceide a Forlì-Cesena e aRavenna e del trans-resveratrolo a Ravenna. Il risultato, parzialmente atteso, concorda conquanto osservato nel corso di precedenti sperimentazioni (Bavaresco et al., 2007) ma assume,nelle condizioni di coltura descritte, un andamento lineare senza raggiungere un plateau (Figure1, 2 e 3).

Figura 1 : Effetto dell’altitudine sul contenuto ditrans-piceide nei vini Sangiovese della provincia di

Forlì-Cesena


Ravenna

Tabella 1 : Contenuto di trans-resveratrolo e di trans-piceide nei vini Sangiovese in funzione del calcare attivodel suolo

Calcare attivo trans-Resveratrolo (mg/L) trans-Piceide (mg/L)< 2 % 1.10 a 1.07 a

2-5 % 1.44 ab 1.30 a

> 5 % 1.90 b 3.51 bA lettere diverse corrispondono medie significativamente differenti per p ≤ 0,05 al test S-N-K

Figura 3 : Effetto dell’altitudine sul contenuto di trans-resveratrolo nei vini Sangiovese del Ravennate

L’effetto del calcare attivo, contenuto nei primi strati del profilo, sul contenuto di stilbeni delvino è stato studiato suddividendo i siti in tre classi (Tab. 1). Il trans-resveratrolo è variato da unminimo di 1.10 mg/L a un massimo di 1.97 mg/L; i vini dei siti con calcare attivo inferiore al 2%erano significativamente meno ricchi in trans-resveratrolo rispetto a quelli più calcarei conalmeno il 5% di calcare attivo. Considerando il trans-piceide che è oscillato tra 1.07 mg/L e 3.51mg/L, i siti con calcare attivo superiore al 5% hanno mostrato un contenuto superiore a tutti glialtri in misura più che significativa. Dalle prime osservazioni ci pare corretto ritenere che inquesti ambienti sia necessaria una “buona” dotazione di calcare attivo per registrare un sensibileincremento degli stilbeni nel vino e in particolare nel caso del trans-piceide per il quale si ritienenecessaria una dotazione del suolo “molto buona”. I risultati ottenuti hanno inoltre confermato,sulla base della maggior variabilità dei 25 vigneti, quanto precedentemente osservato su vitiallevate in vaso (Bavaresco et al., 2005; 2008a).

Sulla base dei riscontri precedentemente descritti diventa interessante analizzare l’eventualeeffetto che la formazione geologica del sito di coltivazione, esercita sulla sintesi degli stilbeni.Infatti, l’Appennino romagnolo si compone di rocce sedimentarie di prevalente origine marina,spesso diversificabili in funzione del tenore in calcare attivo dei suoli, nonché per componentipaesaggistiche rilevanti al fine della viticoltura di qualità. I suoli del margine appenninico, darocce risalenti al pleistocene (AES), sono moderatamente ondulati, profondi, a tessitura fine,soggetti a ristagno idrico e privi di calcare. Considerando la formazione pliocenica delle ArgilleAzzurre emergono notevoli differenze rispetto ad AES, legate all’inclinazione dei pendii, allaminore profondità del suolo, alla tessitura media e alla dotazione calcarea più consistente. LaFormazione Marnoso Arenacea (FMA) ha suoli a tessitura molto variabile riguardo alla tipologiadel substrato, mentre la dotazione calcarea è generalmente elevata.

Complessivamente si può affermare che la formazione geologica abbia influenzato la sintesidegli stilbeni poi riscontrati nel vino. I siti della Formazione delle Argille Azzurre hannopresentato una concentrazione di trans-piceide di 2.86 mg/L, significativamente superiorerispetto a quella del Margine appenninico (0.94 mg/L); la Formazione Marnoso Arenacea si èinvece collocata su valori intermedi (Fig. 4). I vini afferenti alle unità geologiche FAA e FMAhanno mostrato un tenore di trans-resveratrolo significativamente superiore (circa 1.6 mg/L)rispetto a quelli del margine appenninico (0.99 mg/L).

4 - 104



abbinata a un detector a diodi (DAD) settato a 306 e 325 nm utilizzando una colonna C18Supelco Supelcosil.

I dati raccolti sono stati elaborati mediante l’analisi della varianza (ANOVA) a un criterio diclassificazione considerando come fonti di variazione la Formazione geologica e il Calcareattivo; le medie sono state comparate con il test SNK (p<0.05). La relazione esistente traaltitudine e tenore di stilbeni del vino è stata studiata mediante la regressione lineare.

RISULTATI E DISCUSSIONENei vini Sangiovese 2008 analizzati è stata ricercata la presenza di numerosi stilbeni

riscontrando concentrazioni interessanti di trans-piceide e trans-resveratrolo. Non sono statirilevati il pallidolo, il piceatannolo, δ- ed ε-viniferina nonchè la forma cis- di piceide eresveratrolo. Il trans-piceide è variato da concentrazioni pressoché nulle in alcuni vigneti delmargine appenninico a livelli considerevoli superiori a 6 mg/L di alcuni siti delle Argille azzurredel riminese. Le concentrazioni di trans-resveratrolo, seppur inferiori rispetto al precedentecomposto, sono oscillate tra 0.3 e 4.5 mg/L.

Il ruolo dell’altitudine sul tenore di trans-piceide e trans-resveratrolo è stato ricercato mediantela definizione di correlazioni lineari per le diverse province osservando coefficienti dicorrelazione altamente significativi (p<0.01) nel caso del trans-piceide a Forlì-Cesena e aRavenna e del trans-resveratrolo a Ravenna. Il risultato, parzialmente atteso, concorda conquanto osservato nel corso di precedenti sperimentazioni (Bavaresco et al., 2007) ma assume,nelle condizioni di coltura descritte, un andamento lineare senza raggiungere un plateau (Figure1, 2 e 3).


Forlì-Cesena


Ravenna

Tabella 1 : Contenuto di trans-resveratrolo e di trans-piceide nei vini Sangiovese in funzione del calcare attivodel suolo

Calcare attivo trans-Resveratrolo (mg/L) trans-Piceide (mg/L)< 2 % 1.10 a 1.07 a

2-5 % 1.44 ab 1.30 a

> 5 % 1.90 b 3.51 bA lettere diverse corrispondono medie significativamente differenti per p ≤ 0,05 al test S-N-K

Figura 3 : Effetto dell’altitudine sul contenuto di trans-resveratrolo nei vini Sangiovese del Ravennate

L’effetto del calcare attivo, contenuto nei primi strati del profilo, sul contenuto di stilbeni delvino è stato studiato suddividendo i siti in tre classi (Tab. 1). Il trans-resveratrolo è variato da unminimo di 1.10 mg/L a un massimo di 1.97 mg/L; i vini dei siti con calcare attivo inferiore al 2%erano significativamente meno ricchi in trans-resveratrolo rispetto a quelli più calcarei conalmeno il 5% di calcare attivo. Considerando il trans-piceide che è oscillato tra 1.07 mg/L e 3.51mg/L, i siti con calcare attivo superiore al 5% hanno mostrato un contenuto superiore a tutti glialtri in misura più che significativa. Dalle prime osservazioni ci pare corretto ritenere che inquesti ambienti sia necessaria una “buona” dotazione di calcare attivo per registrare un sensibileincremento degli stilbeni nel vino e in particolare nel caso del trans-piceide per il quale si ritienenecessaria una dotazione del suolo “molto buona”. I risultati ottenuti hanno inoltre confermato,sulla base della maggior variabilità dei 25 vigneti, quanto precedentemente osservato su vitiallevate in vaso (Bavaresco et al., 2005; 2008a).

Sulla base dei riscontri precedentemente descritti diventa interessante analizzare l’eventualeeffetto che la formazione geologica del sito di coltivazione, esercita sulla sintesi degli stilbeni.Infatti, l’Appennino romagnolo si compone di rocce sedimentarie di prevalente origine marina,spesso diversificabili in funzione del tenore in calcare attivo dei suoli, nonché per componentipaesaggistiche rilevanti al fine della viticoltura di qualità. I suoli del margine appenninico, darocce risalenti al pleistocene (AES), sono moderatamente ondulati, profondi, a tessitura fine,soggetti a ristagno idrico e privi di calcare. Considerando la formazione pliocenica delle ArgilleAzzurre emergono notevoli differenze rispetto ad AES, legate all’inclinazione dei pendii, allaminore profondità del suolo, alla tessitura media e alla dotazione calcarea più consistente. LaFormazione Marnoso Arenacea (FMA) ha suoli a tessitura molto variabile riguardo alla tipologiadel substrato, mentre la dotazione calcarea è generalmente elevata.

Complessivamente si può affermare che la formazione geologica abbia influenzato la sintesidegli stilbeni poi riscontrati nel vino. I siti della Formazione delle Argille Azzurre hannopresentato una concentrazione di trans-piceide di 2.86 mg/L, significativamente superiorerispetto a quella del Margine appenninico (0.94 mg/L); la Formazione Marnoso Arenacea si èinvece collocata su valori intermedi (Fig. 4). I vini afferenti alle unità geologiche FAA e FMAhanno mostrato un tenore di trans-resveratrolo significativamente superiore (circa 1.6 mg/L)rispetto a quelli del margine appenninico (0.99 mg/L).

4 - 105



b

b

a

0

0.5

1

1.5

2

2.5

3

3.5

AES FMA FAA

tran

s-P

icei

de(m

g/L)

Figura 4: Contenuto di trans-piceide nei vini Sangiovese di Romagna in funzione dell’origine geologica

bb

a

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

AES FMA FAA

tran

s-R

esve

ratr

olo

(mg/

L)

Figura 5: Contenuto di trans-resveratrolo nei vini Sangiovese di Romagna in funzione dell’origine geologica

I siti delle Argille Azzurre hanno presentato una concentrazione di trans-piceide di 2.86 mg/L,significativamente superiore rispetto a quella del Margine appenninico (0.94 mg/L); laFormazione Marnoso-arenacea si è invece collocata su valori intermedi. I vini afferenti alle unitàgeologiche FAA e FMA hanno mostrato un tenore di trans-resveratrolo significativamentesuperiore (circa 1.6 mg/L) rispetto a quelli del margine appenninico (0.99 mg/L) (Fig. 5). Seppurcon riferimento alla sola vendemmia 2008, è ragionevole osservare come i livelli più scarsi ditrans-resveratrolo e di trans-piceide siano collegati a siti del margine appenninico ben distintidalle altre formazioni geologiche in particolare per il modesto contenuto in calcare degli stratisuperficiali. Le dinamiche dell’acqua nel suolo e in particolare le conoscenze degli equilibri delpotassio legati alla componente argillosa e alla capacità di scambio cationico potrebbero spiegareulteriormente tale evidenza (Bavaresco et al., 2006).

CONCLUSIONIIl lavoro verifica l’esistenza di una relazione positiva esistente tra l’altitudine del sito di

coltivazione e la presenza di stilbeni nei vini “Sangiovese di Romagna”. L’importanza del calcareattivo nell’induzione della sintesi degli stilbeni e in particolare del trans-resveratrolo è stataconfermata. Sulla base di un’esperienza di pieno campo condotta in 25 siti sperimentali, è statoindividuato il 5% di calcare attivo come soglia minima necessaria per incrementare la

concentrazione di stilbeni nei vini. La formazione geologica ha presumibilmente influenzato lasintesi degli stilbeni nella vite poiché nei vini prodotti sulla Formazione delle Argille Azzurresono sempre state osservate concentrazioni superiori di trans-piceide e trans-resveratrolo rispettoa quelli del Margine appenninico.

I risultati presentati necessitano comunque di ulteriori conferme e rappresentano un contributopreliminare allo studio delle interazioni intercorrenti tra il terroir e la sintesi degli stilbeni.

BIBLIOGRAFIAAdrian M., Jeandet P., Breuil A.C., Levite A. D., Debord S., Bessis, R., 2000. Assay of

resveratrol and derivative stilbenes in wines by direct injection High Performance LiquidChromatography. Am. J. Enol. Vitic., 51: 37-41.

Bavaresco, L., Petegolli D., Cantù E., Fregoni M., Chiusa G., Trevisan M.,1997. Elicitation andaccumulation of stilbene phytoalexins in grapevine berries infected by B. cinerea. Vitis, 36: 77-83.

Bavaresco L., Civardi S., Pezzutto S., Vezzulli S., Ferrari F., 2005. Grape production,technological parameters, and stilbenic compounds as affected by lime-induced chlorosis. Vitis,44: 63-65.

Bavaresco L., Civardi S., Pezzutto S., Ferrari F., 2006. Effetto della concimazione potassica sullanutrizione minerale, produzione, qualità e stilbeni del vitigno Cabernet Sauvignon. ItalusHortus 13 (3): 85-89.

Bavaresco L., Pezzutto S., Gatti M., Mattivi F., 2007. Role of the variety and someenvironmental factors on grape stilbenes. Vitis, 46: 57-61.

Bavaresco L., Vezzulli S., Civardi S., Gatti M., Battilani P., Pietri A., Ferrari F., 2008a. Effect oflime-induced chlorosis on ochratoxin-A and stilbenic phytoalexins in grapevine (V. vinifera L.)berries infected by Aspergillus carbonarius. J. Agric. Food Chem., 56: 2085-2089.

Bavaresco L., Gatti M., Pezzutto S., Fregoni M., Mattivi F., 2008b. Effect of leaf removal ongrape yield, quality, and stilbenes. Am. J. Enol. Vitic., 59: 292-298.

Busconi M., Reggi S., Fogher C., Bavaresco L., 2009. Evidence of a sirtuin gene family ingrapevine (Vitis vinifera L.). Plant. Physiol. Biochem., 47: 650-652.

De Andrés-de Prado, R, Yuste-Rojas, M., Sort, X., Andrés-Lacueva, C., Torres, M., Lamuela-Raventós, R.M., 2007. Effect of soil type on wine produced from Vitis vinifera L. cv. Grenachein commercial vineyards. J. Agric. Food Chem., 55, 779-786.

Jeandet P., Bessis R., Sbaghi M., Meunier P., Trollat P., 1995. Resveratrol content of wines ofdifferent ages: relationships with fungal disease pressure in the vineyard. Am. J. Enol. Vitic.,46: 1-3.

Langcake P., Pryce R.J., 1977. The production of resveratrol and the viniferins by grapevines inresponse to ultraviolet irradiation. Phytochemistry, 16: 1193-1196.

Renaud S., De Lorgeril M., 1992. Wine, alcohol, platelets, and the Franch paradox for coronaryhearth disease. The Lancet, 339: 1523-1526.

Rupprich N., Kindl H., 1978. Stilbene synthases and stilbene carboxylates synthases, I.Enzymaticsynthesis of 3,5,4’ –trihydroxystilbene from p-coumaroil coenzyme A and malonylcoenzyme A. Hoppe-Seyler’s Z. Physiol. Chem., 359: 165-172.

Sienmann E.H., Creasy L.L., 1992. Concentration of phytoalexin resveratrol in wine. Am. J.Enol. Vitic., 43: 49-52.

4 - 106



b

b

a

0

0.5

1

1.5

2

2.5

3

3.5

AES FMA FAA

tran

s-P

icei

de(m

g/L)

Figura 4: Contenuto di trans-piceide nei vini Sangiovese di Romagna in funzione dell’origine geologica

bb

a

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

AES FMA FAA

tran

s-R

esve

ratr

olo

(mg/

L)

Figura 5: Contenuto di trans-resveratrolo nei vini Sangiovese di Romagna in funzione dell’origine geologica

I siti delle Argille Azzurre hanno presentato una concentrazione di trans-piceide di 2.86 mg/L,significativamente superiore rispetto a quella del Margine appenninico (0.94 mg/L); laFormazione Marnoso-arenacea si è invece collocata su valori intermedi. I vini afferenti alle unitàgeologiche FAA e FMA hanno mostrato un tenore di trans-resveratrolo significativamentesuperiore (circa 1.6 mg/L) rispetto a quelli del margine appenninico (0.99 mg/L) (Fig. 5). Seppurcon riferimento alla sola vendemmia 2008, è ragionevole osservare come i livelli più scarsi ditrans-resveratrolo e di trans-piceide siano collegati a siti del margine appenninico ben distintidalle altre formazioni geologiche in particolare per il modesto contenuto in calcare degli stratisuperficiali. Le dinamiche dell’acqua nel suolo e in particolare le conoscenze degli equilibri delpotassio legati alla componente argillosa e alla capacità di scambio cationico potrebbero spiegareulteriormente tale evidenza (Bavaresco et al., 2006).

CONCLUSIONIIl lavoro verifica l’esistenza di una relazione positiva esistente tra l’altitudine del sito di

coltivazione e la presenza di stilbeni nei vini “Sangiovese di Romagna”. L’importanza del calcareattivo nell’induzione della sintesi degli stilbeni e in particolare del trans-resveratrolo è stataconfermata. Sulla base di un’esperienza di pieno campo condotta in 25 siti sperimentali, è statoindividuato il 5% di calcare attivo come soglia minima necessaria per incrementare la

concentrazione di stilbeni nei vini. La formazione geologica ha presumibilmente influenzato lasintesi degli stilbeni nella vite poiché nei vini prodotti sulla Formazione delle Argille Azzurresono sempre state osservate concentrazioni superiori di trans-piceide e trans-resveratrolo rispettoa quelli del Margine appenninico.

I risultati presentati necessitano comunque di ulteriori conferme e rappresentano un contributopreliminare allo studio delle interazioni intercorrenti tra il terroir e la sintesi degli stilbeni.

BIBLIOGRAFIAAdrian M., Jeandet P., Breuil A.C., Levite A. D., Debord S., Bessis, R., 2000. Assay of

resveratrol and derivative stilbenes in wines by direct injection High Performance LiquidChromatography. Am. J. Enol. Vitic., 51: 37-41.

Bavaresco, L., Petegolli D., Cantù E., Fregoni M., Chiusa G., Trevisan M.,1997. Elicitation andaccumulation of stilbene phytoalexins in grapevine berries infected by B. cinerea. Vitis, 36: 77-83.

Bavaresco L., Civardi S., Pezzutto S., Vezzulli S., Ferrari F., 2005. Grape production,technological parameters, and stilbenic compounds as affected by lime-induced chlorosis. Vitis,44: 63-65.

Bavaresco L., Civardi S., Pezzutto S., Ferrari F., 2006. Effetto della concimazione potassica sullanutrizione minerale, produzione, qualità e stilbeni del vitigno Cabernet Sauvignon. ItalusHortus 13 (3): 85-89.

Bavaresco L., Pezzutto S., Gatti M., Mattivi F., 2007. Role of the variety and someenvironmental factors on grape stilbenes. Vitis, 46: 57-61.

Bavaresco L., Vezzulli S., Civardi S., Gatti M., Battilani P., Pietri A., Ferrari F., 2008a. Effect oflime-induced chlorosis on ochratoxin-A and stilbenic phytoalexins in grapevine (V. vinifera L.)berries infected by Aspergillus carbonarius. J. Agric. Food Chem., 56: 2085-2089.

Bavaresco L., Gatti M., Pezzutto S., Fregoni M., Mattivi F., 2008b. Effect of leaf removal ongrape yield, quality, and stilbenes. Am. J. Enol. Vitic., 59: 292-298.

Busconi M., Reggi S., Fogher C., Bavaresco L., 2009. Evidence of a sirtuin gene family ingrapevine (Vitis vinifera L.). Plant. Physiol. Biochem., 47: 650-652.

De Andrés-de Prado, R, Yuste-Rojas, M., Sort, X., Andrés-Lacueva, C., Torres, M., Lamuela-Raventós, R.M., 2007. Effect of soil type on wine produced from Vitis vinifera L. cv. Grenachein commercial vineyards. J. Agric. Food Chem., 55, 779-786.

Jeandet P., Bessis R., Sbaghi M., Meunier P., Trollat P., 1995. Resveratrol content of wines ofdifferent ages: relationships with fungal disease pressure in the vineyard. Am. J. Enol. Vitic.,46: 1-3.

Langcake P., Pryce R.J., 1977. The production of resveratrol and the viniferins by grapevines inresponse to ultraviolet irradiation. Phytochemistry, 16: 1193-1196.

Renaud S., De Lorgeril M., 1992. Wine, alcohol, platelets, and the Franch paradox for coronaryhearth disease. The Lancet, 339: 1523-1526.

Rupprich N., Kindl H., 1978. Stilbene synthases and stilbene carboxylates synthases, I.Enzymaticsynthesis of 3,5,4’ –trihydroxystilbene from p-coumaroil coenzyme A and malonylcoenzyme A. Hoppe-Seyler’s Z. Physiol. Chem., 359: 165-172.

Sienmann E.H., Creasy L.L., 1992. Concentration of phytoalexin resveratrol in wine. Am. J.Enol. Vitic., 43: 49-52.

4 - 107



UVE E VINI IN VULCANITI BASICHE ANOROGENICHE DEI

LESSINI MERIDIONALI, IMPRONTA PETROCHIMICA E

ASSIMILAZIONE DI METALLI PESANTI.

D. G. Ferioli 1, P. Bartolomei

2, M. Esposito

1, E. Marrocchino

3, L. Sansone

4, M. Borgo

4,

N. Belfiore 4, D. Tomasi

4, R. Tassinari

3, C. Vaccaro

3, M. Niero

4, P. Biondini

5

1 U-SERIES, Via Ferrarese, 131, 40128 Bologna - E-mail [email protected]

2 ENEA, via dei Colli, 16; 40136 Bologna

3 Dipartimento di Scienze della Terra, Università di Ferrara, Via Saragat 1, 44100 Ferrara.

4 CRA-Centro di Ricerca per la Viticoltura, Viale XXVIII Aprile, 26 31015 Conegliano (TV).

5 Delegazione Pontificia per il Santuario della Santa Casa di Loreto, Piazza della Madonna, 1 60025 Loreto

(AN)

RIASSUNTO

Nel 2009 sono stati prelevati e analizzati mediante XRF (X-ray fluorescence) campioni di

suolo, in vigneti sperimentali siti nelle province di Vicenza e di Ancona. Sono stati inoltre

determinati in 2 campioni di mosto e 2 di vino delle varietà Verdicchio e Refosco dal

peduncolo rosso, ed in 2 di uva Refosco dal peduncolo rosso, gli elementi in traccia mediante

ICP-MS (Inductively coupled plasma-mass spectrometry). Lo studio ha consentito di definire

la qualità di uva e vini, i rapporti caratteristici per ogni tipologia di suolo, e caratterizzare

l’impronta geochimica in un ampio areale in cui le modeste differenze climatiche non

influiscono sulle dinamiche di assimilazione. Sono state definite le relazioni fra matrice suolo

e vino attraverso il confronto fra le concentrazioni dei metalli analizzati nelle varie matrici e

varietà di uva.

PAROLE CHIAVE Uva – vino – suolo – impronta geochimica

ABSTRACT In 2009, 18 samples of soils, coming from experimental vineyards in Vicenza and Ancona,

were collected and analysed using XRF technique, to characterize major and minor element

concentration. Moreover, 2 samples of must, 2 samples of wine (one of each varieties

Verdicchio and Refosco dal peduncolo rosso) and 2 samples of grapes Refosco dal peduncolo

rosso, were investigated using ICP-MS (Inductively coupled plasma-mass spectrometry)

technique in order to define their trace elements concentrations. The aim of this study has

been not only to characterize the quality of the grapes and the wines, but also to define the

typical ratios between these two variable for each soils, and to outline geochemical

fingerprints in a wide area where climatic differences have limited influence on the

assimilation processes. The comparison of heavy metals concentrations between the several

matrix and the varieties of grapes allow to define the relationship between soil matrix and

wine.

KEYWORD Grape – wine – soil – geochemical fingerprints

INTRODUZIONE

La risposta delle differenti varietà di uva all’ambiente geolitologico e microclimatico può

fornire preziose informazioni per definire l’impronta geochimica dei prodotti alimentari e

caratterizzare i contenuti dei macro e micronutrienti essenziali per l’alimentazione umana nei

prodotti alimentari. Questi dati caratteristici dell’ambiente geolitologico e microclimatico

consentono di procedere all’identificazione delle aree di origine e quindi alla certificazione

dei prodotti. La conoscenza delle concentrazioni dei macro e micronutrienti inorganici nei

prodotti alimentari è indispensabile per la tutela dei consumatori e per la valorizzazione dei

prodotti alimentari e risponde alla crescente richiesta di prodotti certificati sulla base

dell’origine geografica. Le attuali etichettature, importante passo avanti verso la tracciabilità

dei prodotti, informano i consumatori dell’origine e delle procedure agronomiche, rendendoli

consapevoli della storia degli alimenti. Una più completa descrizione delle caratteristiche del

prodotto si potrebbe ottenere integrando l’etichetta con l’indicazione del contenuto in macro e

micronutrienti inorganici. Questa strategia ridurrebbe i rischi di immissione nel mercato di

prodotti non locali ottenuti con pratiche agronomiche intensive e su suoli qualitativamente

non idonei (sia per processi di impoverimento dei nutrienti minerali sia per eccesso di metalli

inquinanti). Per attuare questo sistema di difesa e tutela dei prodotti autoctoni, occorre

conoscere le caratteristiche geochimiche dei siti produttivi, e quindi fornire l’impronta digitale

del prodotto certificato e accertare mescolanze e/o sostituzioni con prodotti di altra

provenienza. La verifica della congruità della provenienza geografica dichiarata in etichetta,

potrebbe avvenire tramite il confronto con la distribuzione dei macro e micronutrienti

inorganici tipici dell’area di produzione. L’analisi geochimica conoscitiva risulta quindi

indispensabile per una corretta valutazione della provenienza dei prodotti alimentari e la

verifica della qualità dichiarata in etichetta e dell’assenza di manipolazioni. Si propone con

questo lavoro lo sviluppo di una metodologia analitica per la tracciabilità e la realizzazione di

una banca dati nazionale sulla qualità dei prodotti e sui range di composizione tipici e

caratteristici delle aree di provenienza. Il presente lavoro è stato condotto in un’area

caratterizzata da vigneti che producono uve di elevata qualità, impiantati su suoli

particolarmente ricchi in metalli di transizione (Cr, Ni, Co, V, di elevato valore nutrizionale in

basse concentrazioni ma potenzialmente tossico-nocivi se in elevate concentrazioni) al fine di

mostrare la capacità delle cultivar nel selezionare gli elementi chimici e quindi fornire le

corrette dosi dei macro e micronutrienti essenziali anche in presenza di anomali arricchimenti

e disponibilità nei suoli. La definizione dell’importanza nutrizionale di questi elementi nelle

uve e nei vini ha stimolato il presente studio condotto in una delle aree vulcaniche basiche

famose per gli antichi e pregiati vitigni.

MATERIALI E METODI

Inquadramento geologico

I due campi sperimentali di Verdicchio e Refosco, oggetto della presente sperimentazione,

sono localizzati, rispettivamente, nei comuni di Gambellara e Mason in provincia di Vicenza.

Sono costituiti da suoli vulcanici basici riferibili al magmatismo anorogenico terziario, che ha

interessato la porzione meridionale del Sudalpino Veneto. Nei Lessini affiorano prodotti

vulcanici basici subaerei e vulcanoclastici sottomarini, questi ultimi nel settore più

meridionale dell’area (ove insistono i campi sperimentali del CRA-VIT), caratterizzati dalle

tipiche strutture ialoclastiche di ambiente sottomarino di mare poco profondo. Le vulcaniti

affiorano nella zona centrale delle Prealpi Venete in corrispondenza ed allineati lungo il

prolungamento della cosiddetta “Flessura pedemontana” (Caputo e Bosellini 1994),

importante struttura tettonica che si estende dalla Linea Schio-Vicenza, a ovest, all’accidente

trasversale Fadalto-Vittorio Veneto, a est, per una distanza di circa 80 km. Si tratta a grande

4 - 108



UVE E VINI IN VULCANITI BASICHE ANOROGENICHE DEI

LESSINI MERIDIONALI, IMPRONTA PETROCHIMICA E

ASSIMILAZIONE DI METALLI PESANTI.

D. G. Ferioli 1, P. Bartolomei

2, M. Esposito

1, E. Marrocchino

3, L. Sansone

4, M. Borgo

4,

N. Belfiore 4, D. Tomasi

4, R. Tassinari

3, C. Vaccaro

3, M. Niero

4, P. Biondini

5

1 U-SERIES, Via Ferrarese, 131, 40128 Bologna - E-mail [email protected]

2 ENEA, via dei Colli, 16; 40136 Bologna

3 Dipartimento di Scienze della Terra, Università di Ferrara, Via Saragat 1, 44100 Ferrara.

4 CRA-Centro di Ricerca per la Viticoltura, Viale XXVIII Aprile, 26 31015 Conegliano (TV).

5 Delegazione Pontificia per il Santuario della Santa Casa di Loreto, Piazza della Madonna, 1 60025 Loreto

(AN)

RIASSUNTO

Nel 2009 sono stati prelevati e analizzati mediante XRF (X-ray fluorescence) campioni di

suolo, in vigneti sperimentali siti nelle province di Vicenza e di Ancona. Sono stati inoltre

determinati in 2 campioni di mosto e 2 di vino delle varietà Verdicchio e Refosco dal

peduncolo rosso, ed in 2 di uva Refosco dal peduncolo rosso, gli elementi in traccia mediante

ICP-MS (Inductively coupled plasma-mass spectrometry). Lo studio ha consentito di definire

la qualità di uva e vini, i rapporti caratteristici per ogni tipologia di suolo, e caratterizzare

l’impronta geochimica in un ampio areale in cui le modeste differenze climatiche non

influiscono sulle dinamiche di assimilazione. Sono state definite le relazioni fra matrice suolo

e vino attraverso il confronto fra le concentrazioni dei metalli analizzati nelle varie matrici e

varietà di uva.

PAROLE CHIAVE Uva – vino – suolo – impronta geochimica

ABSTRACT In 2009, 18 samples of soils, coming from experimental vineyards in Vicenza and Ancona,

were collected and analysed using XRF technique, to characterize major and minor element

concentration. Moreover, 2 samples of must, 2 samples of wine (one of each varieties

Verdicchio and Refosco dal peduncolo rosso) and 2 samples of grapes Refosco dal peduncolo

rosso, were investigated using ICP-MS (Inductively coupled plasma-mass spectrometry)

technique in order to define their trace elements concentrations. The aim of this study has

been not only to characterize the quality of the grapes and the wines, but also to define the

typical ratios between these two variable for each soils, and to outline geochemical

fingerprints in a wide area where climatic differences have limited influence on the

assimilation processes. The comparison of heavy metals concentrations between the several

matrix and the varieties of grapes allow to define the relationship between soil matrix and

wine.

KEYWORD Grape – wine – soil – geochemical fingerprints

INTRODUZIONE

La risposta delle differenti varietà di uva all’ambiente geolitologico e microclimatico può

fornire preziose informazioni per definire l’impronta geochimica dei prodotti alimentari e

caratterizzare i contenuti dei macro e micronutrienti essenziali per l’alimentazione umana nei

prodotti alimentari. Questi dati caratteristici dell’ambiente geolitologico e microclimatico

consentono di procedere all’identificazione delle aree di origine e quindi alla certificazione

dei prodotti. La conoscenza delle concentrazioni dei macro e micronutrienti inorganici nei

prodotti alimentari è indispensabile per la tutela dei consumatori e per la valorizzazione dei

prodotti alimentari e risponde alla crescente richiesta di prodotti certificati sulla base

dell’origine geografica. Le attuali etichettature, importante passo avanti verso la tracciabilità

dei prodotti, informano i consumatori dell’origine e delle procedure agronomiche, rendendoli

consapevoli della storia degli alimenti. Una più completa descrizione delle caratteristiche del

prodotto si potrebbe ottenere integrando l’etichetta con l’indicazione del contenuto in macro e

micronutrienti inorganici. Questa strategia ridurrebbe i rischi di immissione nel mercato di

prodotti non locali ottenuti con pratiche agronomiche intensive e su suoli qualitativamente

non idonei (sia per processi di impoverimento dei nutrienti minerali sia per eccesso di metalli

inquinanti). Per attuare questo sistema di difesa e tutela dei prodotti autoctoni, occorre

conoscere le caratteristiche geochimiche dei siti produttivi, e quindi fornire l’impronta digitale

del prodotto certificato e accertare mescolanze e/o sostituzioni con prodotti di altra

provenienza. La verifica della congruità della provenienza geografica dichiarata in etichetta,

potrebbe avvenire tramite il confronto con la distribuzione dei macro e micronutrienti

inorganici tipici dell’area di produzione. L’analisi geochimica conoscitiva risulta quindi

indispensabile per una corretta valutazione della provenienza dei prodotti alimentari e la

verifica della qualità dichiarata in etichetta e dell’assenza di manipolazioni. Si propone con

questo lavoro lo sviluppo di una metodologia analitica per la tracciabilità e la realizzazione di

una banca dati nazionale sulla qualità dei prodotti e sui range di composizione tipici e

caratteristici delle aree di provenienza. Il presente lavoro è stato condotto in un’area

caratterizzata da vigneti che producono uve di elevata qualità, impiantati su suoli

particolarmente ricchi in metalli di transizione (Cr, Ni, Co, V, di elevato valore nutrizionale in

basse concentrazioni ma potenzialmente tossico-nocivi se in elevate concentrazioni) al fine di

mostrare la capacità delle cultivar nel selezionare gli elementi chimici e quindi fornire le

corrette dosi dei macro e micronutrienti essenziali anche in presenza di anomali arricchimenti

e disponibilità nei suoli. La definizione dell’importanza nutrizionale di questi elementi nelle

uve e nei vini ha stimolato il presente studio condotto in una delle aree vulcaniche basiche

famose per gli antichi e pregiati vitigni.

MATERIALI E METODI

Inquadramento geologico

I due campi sperimentali di Verdicchio e Refosco, oggetto della presente sperimentazione,

sono localizzati, rispettivamente, nei comuni di Gambellara e Mason in provincia di Vicenza.

Sono costituiti da suoli vulcanici basici riferibili al magmatismo anorogenico terziario, che ha

interessato la porzione meridionale del Sudalpino Veneto. Nei Lessini affiorano prodotti

vulcanici basici subaerei e vulcanoclastici sottomarini, questi ultimi nel settore più

meridionale dell’area (ove insistono i campi sperimentali del CRA-VIT), caratterizzati dalle

tipiche strutture ialoclastiche di ambiente sottomarino di mare poco profondo. Le vulcaniti

affiorano nella zona centrale delle Prealpi Venete in corrispondenza ed allineati lungo il

prolungamento della cosiddetta “Flessura pedemontana” (Caputo e Bosellini 1994),

importante struttura tettonica che si estende dalla Linea Schio-Vicenza, a ovest, all’accidente

trasversale Fadalto-Vittorio Veneto, a est, per una distanza di circa 80 km. Si tratta a grande

4 - 109



scala di due pieghe parallele, una anticlinale, a nord, ed una sinclinale, a sud, con asse in

direzione ENE-WSW (Caputo e Bosellini 1994) che collega l’altopiano di Asiago con la

pianura alluvionale della fossa tettonica di Schio, struttura lungo la quale si sono realizzate

importanti discontinuità tettoniche che hanno consentito la risalita di magmi basici poco

differenziati che rappresentano oggi i depositi ialoclastici che si appoggiano sulla struttura a

pieghe.

L’evoluzione tettonica tardo Eocenica - Oligocenica, ha portato allo sviluppo di sistemi di

faglie regionali con andamento NNE che hanno consentito la risalita di basalti alcalini. Nelle

aree più settentrionali le colate sono sub–aeree ed i magmi alcalini ad affinità sodica, mentre

nelle aree più meridionali le colate sono di ambiente sottomarino ed affiorano rocce

appartenenti alle tre serie: magmi alcalini sodici, magmi transizionali e magmi alcalino

potassici. I suoli dei vigneti sono il prodotto delle trasformazioni pedogenetiche di depositi

ialoclastici e brecciole a pillows lava, derivate da eruzioni in ambiente sottomarino. La natura

ialoclastica ha favorito i processi pedogenetici e quindi lo sviluppo di suoli idonei alla

coltivazione della vite. I campi sperimentali, come mostrano i dati petrochimici sui suoli, non

derivano da un’unica colata, ma da manifestazioni vulcaniche differenti, infatti i campi più

settentrionali sono impostati su colate basaltiche alcalino-sodiche in cui la composizione dei

suoli è omogenea, mentre il sito sperimentale più meridionale, ubicato nel settore

geograficamente attribuibile alla transizione con l’area Berica, è impostato su una serie di

colate basaltiche sottomarine in cui si ha una significativa variazione del rapporto Na2O/ K2O.

Anche i suoli dei vigneti dell’area di Gambellara sono derivati da varie colate basaltiche di

composizione sia alcalina sodica che potassica, questo comporta una variazione del rapporto

Na2O/K2O nei suoli che si riflette anche sulle uve. Le significative differenze nei rapporti fra

gli alcali non possono essere dovute a weathering in quanto i terreni, che distano uno

dall’altro meno di 3 km, hanno stessa età, morfologia, condizioni climatiche, esposizione e

strutture vulcaniche (entrambi i siti sono impostati su colate sottomarine ialoclastiche).

Essendo gli elementi alcalini estremamente mobili, il mantenimento delle originarie

differenze composizionali consente di affermare che i processi alterativi non hanno potuto

differenziare in maniera significativa le caratteristiche geochimiche di questi suoli, che sono

stati interessati dalla stessa storia.

Misure in spettrometria γ hanno consentito di escludere concentrazioni significative di

elementi radioattivi e quindi il rispetto dei requisiti previsti dalla radioprotezione nonostante

nel 1986 l’Italia settentrionale, ed in particolar modo le regioni Friuli Venezia Giulia e

Veneto, sia stata interessata dalla ricaduta radioattiva conseguente all’incidente di Chernobyl.

Campionatura

Campioni di suolo, in numero variabile da tre a cinque, in funzione dell’estensione del

vigneto, sono stati prelevati a tre profondità (0-10; 10-30; 30-60cm), nella zona del sottofila,

in entrambe le province di Vicenza e Ancona (vigneti sperimentali della Delegazione

Pontificia per il Santuario della Santa Casa di Loreto – Ancona). Sono state considerate due

varietà, il Verdicchio ed il Refosco p.r. e per ciascuna di esse è stato analizzato un campione

di vino del 2007 vinificato dalla medesima azienda. Per meglio comprendere il processo di

trasferimento dei metalli dal suolo, alla pianta ed al vino, durante la vendemmia del 2009, per

la varietà Refosco, sono stati indagati anche i prodotti delle varie fasi di vinificazione (uva,

mosto e vino). Per evidenziare i contributi dovuti al succo e alle bucce, si sono separati, per

centrifugazione, il residuo solido (buccia e semi) ed il succo (polpa) da sottoporre ad analisi

(Ferioli 2010) .

Metodologie analitiche

I “metalli pesanti” e le terre rare presenti in vini, succhi, bucce e mosti sono stati determinati

mediante analisi in ICP-MS (Inductively coupled plasma-mass spectrometry) con lo

spettrometro Serie X della Thermo Electron Corporation, dotato di dispositivo a cella di

collisione/reazione CCTED

per la riduzione/eliminazione delle principali interferenze

poliatomiche ed isobariche, in dotazione al Dipartimento di Scienze della Terra

dell’Università di Ferrara.

Ogni campione è stato sottoposto a due cicli di analisi; un primo per la determinazione di Li,

Be, B, Na, Al, K, Rb, Ca, Sr, Ba, Mg, Mn, Fe, V, Cr, Co, Ni, Cu, Zn, Ga, As, Se, Mo, Ag, Cd,

Sb, Te, Hg, Tl, Pb, Bi, U ed un secondo per la determinazione delle concentrazioni di Rb, Sr,

Y, Zr, Nb, La, Ce, Pr, Nd, Sm Eu, Gd, Tb, Dy, Ho Er, Tm, Yb, Lu, Hf, Ta, Th, U.

La determinazione degli elementi maggiori e in traccia nei suoli è stata eseguita mediante

XRF (X-ray fluorescence), (spettrometro a dispersione di lunghezza d’onda ARL Advant’XP

in dotazione ai laboratori del Dipartimento di Scienze della Terra dell’Università di Ferrara).

La composizione chimica dei suoli è espressa in percentuale in peso dei seguenti ossidi

(SiO2, TiO2, Al2O3, Fe2O3, MnO, MgO, CaO, Na2O, K2O, P2O5 ) ed in ppm dei seguenti

elementi in traccia: Ba, Ce, Co, Cr, La, Nb, Ni, Pb, Rb, Sr, Th, V,Y, Zn, Zr, Cu, Ga, Nd, S,

Sc. Il contenuto totale in fasi volatili nei suoli é stato determinato come perdita in peso dopo

calcinazione in muffola a 1000° Celsius (Loss On Ignition). Infine nei suoli, nelle uve, mosti e

vini sono stati misurati i radionuclidi naturali (Ra-226 e K-40) ed artificiali (Cs-137)

attraverso spettrometria gamma e condotte secondo la norma UNI 10797 - 1999. Le

concentrazioni di K-40 e Cs-137 sono misurabili direttamente, il Ra-226 è stato calcolato

tramite i figli a vita breve (Pb-214 e Bi-214) mentre il Th-232 è stato calcolato tramite il Ra-

228, a sua volta calcolato tramite Ac-228. Le misure sono state effettuate con rivelatori al

Germanio Iperpuro (HPGe), calibrati in energia e in efficienza con standards radioattivi forniti

da IAEA (International Atomic Energy Agency), NIST (National Institute of Standards and

Technology) ENEA (Ente per le Nuove tecnologie, l’Energia e l’Ambiente). Le sorgenti

radioattive utilizzate sono IAEA-RGU-1, IAEA RGTh-1, IAEA RGK-1, IAEA 300, IAEA

375, IAEA 315 ed IAEA SOIL 6, NIST SRM 4350B ed ENEA MRS 1057.

RISULTATI E DISCUSSIONE

Dalla classificazione petrologica si deduce che i suoli di Gambellara derivano da ialoclastiti

ed hanno composizione alcali basaltica di serie sodica mentre quelli di Mason da colate di

serie transizionale debolmente ricche in Potassio. I suoli veneti sono stati confrontati coi suoli

derivati da rocce sedimentarie di omologhi vigneti sperimentali presenti nelle Marche.

L’elaborazione dei dati relativi a Co, Cr, Ni, Pb, V, Zn, Cu normalizzati ai limiti di legge

(D.Lgs. 152/2006) ha evidenziato il superamento dei parametri per Co, Cr, Ni e V “Fig. 1”.

4 - 110



scala di due pieghe parallele, una anticlinale, a nord, ed una sinclinale, a sud, con asse in

direzione ENE-WSW (Caputo e Bosellini 1994) che collega l’altopiano di Asiago con la

pianura alluvionale della fossa tettonica di Schio, struttura lungo la quale si sono realizzate

importanti discontinuità tettoniche che hanno consentito la risalita di magmi basici poco

differenziati che rappresentano oggi i depositi ialoclastici che si appoggiano sulla struttura a

pieghe.

L’evoluzione tettonica tardo Eocenica - Oligocenica, ha portato allo sviluppo di sistemi di

faglie regionali con andamento NNE che hanno consentito la risalita di basalti alcalini. Nelle

aree più settentrionali le colate sono sub–aeree ed i magmi alcalini ad affinità sodica, mentre

nelle aree più meridionali le colate sono di ambiente sottomarino ed affiorano rocce

appartenenti alle tre serie: magmi alcalini sodici, magmi transizionali e magmi alcalino

potassici. I suoli dei vigneti sono il prodotto delle trasformazioni pedogenetiche di depositi

ialoclastici e brecciole a pillows lava, derivate da eruzioni in ambiente sottomarino. La natura

ialoclastica ha favorito i processi pedogenetici e quindi lo sviluppo di suoli idonei alla

coltivazione della vite. I campi sperimentali, come mostrano i dati petrochimici sui suoli, non

derivano da un’unica colata, ma da manifestazioni vulcaniche differenti, infatti i campi più

settentrionali sono impostati su colate basaltiche alcalino-sodiche in cui la composizione dei

suoli è omogenea, mentre il sito sperimentale più meridionale, ubicato nel settore

geograficamente attribuibile alla transizione con l’area Berica, è impostato su una serie di

colate basaltiche sottomarine in cui si ha una significativa variazione del rapporto Na2O/ K2O.

Anche i suoli dei vigneti dell’area di Gambellara sono derivati da varie colate basaltiche di

composizione sia alcalina sodica che potassica, questo comporta una variazione del rapporto

Na2O/K2O nei suoli che si riflette anche sulle uve. Le significative differenze nei rapporti fra

gli alcali non possono essere dovute a weathering in quanto i terreni, che distano uno

dall’altro meno di 3 km, hanno stessa età, morfologia, condizioni climatiche, esposizione e

strutture vulcaniche (entrambi i siti sono impostati su colate sottomarine ialoclastiche).

Essendo gli elementi alcalini estremamente mobili, il mantenimento delle originarie

differenze composizionali consente di affermare che i processi alterativi non hanno potuto

differenziare in maniera significativa le caratteristiche geochimiche di questi suoli, che sono

stati interessati dalla stessa storia.

Misure in spettrometria γ hanno consentito di escludere concentrazioni significative di

elementi radioattivi e quindi il rispetto dei requisiti previsti dalla radioprotezione nonostante

nel 1986 l’Italia settentrionale, ed in particolar modo le regioni Friuli Venezia Giulia e

Veneto, sia stata interessata dalla ricaduta radioattiva conseguente all’incidente di Chernobyl.

Campionatura

Campioni di suolo, in numero variabile da tre a cinque, in funzione dell’estensione del

vigneto, sono stati prelevati a tre profondità (0-10; 10-30; 30-60cm), nella zona del sottofila,

in entrambe le province di Vicenza e Ancona (vigneti sperimentali della Delegazione

Pontificia per il Santuario della Santa Casa di Loreto – Ancona). Sono state considerate due

varietà, il Verdicchio ed il Refosco p.r. e per ciascuna di esse è stato analizzato un campione

di vino del 2007 vinificato dalla medesima azienda. Per meglio comprendere il processo di

trasferimento dei metalli dal suolo, alla pianta ed al vino, durante la vendemmia del 2009, per

la varietà Refosco, sono stati indagati anche i prodotti delle varie fasi di vinificazione (uva,

mosto e vino). Per evidenziare i contributi dovuti al succo e alle bucce, si sono separati, per

centrifugazione, il residuo solido (buccia e semi) ed il succo (polpa) da sottoporre ad analisi

(Ferioli 2010) .

Metodologie analitiche

I “metalli pesanti” e le terre rare presenti in vini, succhi, bucce e mosti sono stati determinati

mediante analisi in ICP-MS (Inductively coupled plasma-mass spectrometry) con lo

spettrometro Serie X della Thermo Electron Corporation, dotato di dispositivo a cella di

collisione/reazione CCTED

per la riduzione/eliminazione delle principali interferenze

poliatomiche ed isobariche, in dotazione al Dipartimento di Scienze della Terra

dell’Università di Ferrara.

Ogni campione è stato sottoposto a due cicli di analisi; un primo per la determinazione di Li,

Be, B, Na, Al, K, Rb, Ca, Sr, Ba, Mg, Mn, Fe, V, Cr, Co, Ni, Cu, Zn, Ga, As, Se, Mo, Ag, Cd,

Sb, Te, Hg, Tl, Pb, Bi, U ed un secondo per la determinazione delle concentrazioni di Rb, Sr,

Y, Zr, Nb, La, Ce, Pr, Nd, Sm Eu, Gd, Tb, Dy, Ho Er, Tm, Yb, Lu, Hf, Ta, Th, U.

La determinazione degli elementi maggiori e in traccia nei suoli è stata eseguita mediante

XRF (X-ray fluorescence), (spettrometro a dispersione di lunghezza d’onda ARL Advant’XP

in dotazione ai laboratori del Dipartimento di Scienze della Terra dell’Università di Ferrara).

La composizione chimica dei suoli è espressa in percentuale in peso dei seguenti ossidi

(SiO2, TiO2, Al2O3, Fe2O3, MnO, MgO, CaO, Na2O, K2O, P2O5 ) ed in ppm dei seguenti

elementi in traccia: Ba, Ce, Co, Cr, La, Nb, Ni, Pb, Rb, Sr, Th, V,Y, Zn, Zr, Cu, Ga, Nd, S,

Sc. Il contenuto totale in fasi volatili nei suoli é stato determinato come perdita in peso dopo

calcinazione in muffola a 1000° Celsius (Loss On Ignition). Infine nei suoli, nelle uve, mosti e

vini sono stati misurati i radionuclidi naturali (Ra-226 e K-40) ed artificiali (Cs-137)

attraverso spettrometria gamma e condotte secondo la norma UNI 10797 - 1999. Le

concentrazioni di K-40 e Cs-137 sono misurabili direttamente, il Ra-226 è stato calcolato

tramite i figli a vita breve (Pb-214 e Bi-214) mentre il Th-232 è stato calcolato tramite il Ra-

228, a sua volta calcolato tramite Ac-228. Le misure sono state effettuate con rivelatori al

Germanio Iperpuro (HPGe), calibrati in energia e in efficienza con standards radioattivi forniti

da IAEA (International Atomic Energy Agency), NIST (National Institute of Standards and

Technology) ENEA (Ente per le Nuove tecnologie, l’Energia e l’Ambiente). Le sorgenti

radioattive utilizzate sono IAEA-RGU-1, IAEA RGTh-1, IAEA RGK-1, IAEA 300, IAEA

375, IAEA 315 ed IAEA SOIL 6, NIST SRM 4350B ed ENEA MRS 1057.

RISULTATI E DISCUSSIONE

Dalla classificazione petrologica si deduce che i suoli di Gambellara derivano da ialoclastiti

ed hanno composizione alcali basaltica di serie sodica mentre quelli di Mason da colate di

serie transizionale debolmente ricche in Potassio. I suoli veneti sono stati confrontati coi suoli

derivati da rocce sedimentarie di omologhi vigneti sperimentali presenti nelle Marche.

L’elaborazione dei dati relativi a Co, Cr, Ni, Pb, V, Zn, Cu normalizzati ai limiti di legge

(D.Lgs. 152/2006) ha evidenziato il superamento dei parametri per Co, Cr, Ni e V “Fig. 1”.

4 - 111



Fig. 1 Concentrazioni in ppm dei metalli nei suoli normalizzati ai valori limiti D.Lgs.

152/2006 per uso verde pubblico, privato e residenziale.

Questi superamenti sono da imputare a cause naturali essendo le elevate concentrazioni in

elementi di transizione caratteristiche dei suoli di origine vulcanica basica. Tale anomalia non

viene trasferita alle uve e ai vini dei Lessini che hanno infatti concentrazioni

significativamente inferiori ai limiti previsti dal Decreto Ministeriale 29 dicembre 1986

“Caratteristiche e limiti di alcune sostanze contenute nei vini”, dal regolamento europeo

2006/1881 CE, che definisce i tenori massimi di alcuni contaminanti nei prodotti alimentari e

dalle linee guida dettate dal O.I.V. (Office International de la Vigne et du Vin). A parità di

concentrazione nei suoli veneti inoltre il Refosco assimila i metalli con un ordine di grandezza

superiore (scala logaritmica) rispetto al Verdicchio.

Le concentrazioni degli elementi di interesse ambientale nei suoli dei vigneti a Verdicchio e

a Refosco dei Lessini sono state inoltre confrontate con quelli di Loreto (AN - Marche)

coltivati anch’essi a Verdicchio e Refosco p. r. Nelle Marche, uve e vini di Refosco e

Verdicchio mostrano basse concentrazioni di questi elementi essenziali “Fig. 2”.

(a)

(b)

Fig.2. Confronto dei trends di concentrazione dei metalli (a) e di quelli di interesse

ambientale (b) nel Refosco dei vigneti del Veneto e Marche.

Il rapporto fra i vari metalli nei suoli e nelle cultivars consente di discriminare i contributi

dovuti al suolo e i rapporti legati alle capacità di assimilazione delle piante. Ad esempio le

concentrazioni di Magnesio e Stronzio nelle uve e nei vini per la stessa cultivar ottenute nel

Veneto e nelle Marche sono confrontabili, nonostante le forti differenze riscontrate fra i suoli

vulcanici dei Lessini e sedimentari carbonatici delle Marche per cui essi sono caratteristici e

non sono significativi ai fini della tracciabilità geografica “Figg. 2, 3”.

(a)

(b)

Fig.3. Confronto dei trends di concentrazione dei metalli(a) e di quelli di interesse

ambientale (b) nel Verdicchio dei vigneti del Veneto e Marche.

Le due varietà sono state caratterizzate mediante i coefficienti di assimilazione K espressi

come rapporto fra concentrazione dell’elemento Cipianta nella pianta e la concentrazione

dell’elemento Cisuolo nel suolo (K= Cipianta / Cisuolo).

Il fattore di assimilazione è influenzato dal potenziale ionico dei diversi elementi

considerati, dal pH, dalla temperatura e dalle caratteristiche della pianta.

Infine sono state eseguite analisi di bucce e vinaccioli che hanno mostrato un generale

arricchimento in metalli, ed in particolare in metalli alcalino-terrosi, per cui la permanenza dei

residui a contatto con i succhi può influire sulla concentrazione in metalli.

CONCLUSIONI

Nell’area veneta indagata, i vigneti a DOC e DOCG devono l’elevata qualità e le

caratteristiche organolettiche alla natura vulcanica dei suoli. Grazie all’analisi del rapporto fra

concentrazione dei macro e micronutrienti nei suoli e nei vini sono stati definiti i coefficienti

di assimilazione per le due varietà Verdicchio e Refosco p.r. e l’impronta geochimica del

suolo. La metodologia proposta non solo consente di verificare il rispetto dei contenuti in

metalli e tutelare la salute dei consumatori, ma anche di evitare la commercializzazione di

prodotti qualitativamente non idonei.

La qualità della produzione vitivinicola in suoli vulcanici ad elevate concentrazioni di

metalli di transizione richiede il controllo del rispetto dei limiti normativi e del corretto valore

nutrizionale in termini di macro e micronutrienti, caratteristiche riscontrate per Refosco p.r. e

Verdicchio dei Lessini che hanno basse ma corrette concentrazioni di Cromo, Nichel, Cobalto

e Vanadio. Va segnalato che il Refosco p.r. è risultato più sensibile rispetto al Verdicchio

all’assimilazione di Cr, Ni e V per cui, in casi di forti anomalie del contenuto di questi

metalli, il Verdicchio fornisce maggiori garanzie del rispetto dei parametri nutrizionali.

Le misure di radioattività naturale e antropica hanno messo in evidenza l’assenza di rischio.

Eventuali futuri monitoraggi potranno aiutare a definire la stabilità composizionale e quindi

il mantenimento delle caratteristiche geochimiche riscontrate per uva e vino.

BIBLIOGRAFIA

Caputo R. & Bosellini A. 1994- La flessura pedemontana del Veneto centrale: anticlinale di

rampa a sviluppo bloccato da condotti vulcanici. The pedealpine flexure Zone of central

Venetian Alps:a ramp anticline halted by volcanic conduits. Atti Tic. Sc. Terra 1994 (Serie

speciale), 1: 255-268.

Ferioli D.G. 2010 Tesi di Dottorato in Scienze della Terra - Tracciabilità delle provenienze e

valorizzazione dei prodotti alimentari attraverso nuovi sistemi di caratterizzazione

geochimica.

D.M. 29 dicembre 1986 - Caratteristiche e limiti di alcune sostanze contenute nei vini.

D.M. 25 ottobre 1999, n.471

Regolamento recante criteri, procedure e modalita' per la messa in sicurezza, la bonifica e il

ripristino ambientale dei siti inquinati, ai sensi dell'articolo 17 del decreto legislativo 5

febbraio 1997, n. 22, e successive modificazioni e integrazioni.

Decreto Legislativo 3 aprile 2006, n. 152 Norme in materia ambientale.

Regolamento (CEE) N. 1881/2006 della Commissione del 19 dicembre 2006 che definisce i

tenori massimi di alcuni contaminanti nei prodotti alimentari.

4 - 112



Fig. 1 Concentrazioni in ppm dei metalli nei suoli normalizzati ai valori limiti D.Lgs.

152/2006 per uso verde pubblico, privato e residenziale.

Questi superamenti sono da imputare a cause naturali essendo le elevate concentrazioni in

elementi di transizione caratteristiche dei suoli di origine vulcanica basica. Tale anomalia non

viene trasferita alle uve e ai vini dei Lessini che hanno infatti concentrazioni

significativamente inferiori ai limiti previsti dal Decreto Ministeriale 29 dicembre 1986

“Caratteristiche e limiti di alcune sostanze contenute nei vini”, dal regolamento europeo

2006/1881 CE, che definisce i tenori massimi di alcuni contaminanti nei prodotti alimentari e

dalle linee guida dettate dal O.I.V. (Office International de la Vigne et du Vin). A parità di

concentrazione nei suoli veneti inoltre il Refosco assimila i metalli con un ordine di grandezza

superiore (scala logaritmica) rispetto al Verdicchio.

Le concentrazioni degli elementi di interesse ambientale nei suoli dei vigneti a Verdicchio e

a Refosco dei Lessini sono state inoltre confrontate con quelli di Loreto (AN - Marche)

coltivati anch’essi a Verdicchio e Refosco p. r. Nelle Marche, uve e vini di Refosco e

Verdicchio mostrano basse concentrazioni di questi elementi essenziali “Fig. 2”.

(a)

(b)

Fig.2. Confronto dei trends di concentrazione dei metalli (a) e di quelli di interesse

ambientale (b) nel Refosco dei vigneti del Veneto e Marche.

Il rapporto fra i vari metalli nei suoli e nelle cultivars consente di discriminare i contributi

dovuti al suolo e i rapporti legati alle capacità di assimilazione delle piante. Ad esempio le

concentrazioni di Magnesio e Stronzio nelle uve e nei vini per la stessa cultivar ottenute nel

Veneto e nelle Marche sono confrontabili, nonostante le forti differenze riscontrate fra i suoli

vulcanici dei Lessini e sedimentari carbonatici delle Marche per cui essi sono caratteristici e

non sono significativi ai fini della tracciabilità geografica “Figg. 2, 3”.

(a)

(b)

Fig.3. Confronto dei trends di concentrazione dei metalli(a) e di quelli di interesse

ambientale (b) nel Verdicchio dei vigneti del Veneto e Marche.

Le due varietà sono state caratterizzate mediante i coefficienti di assimilazione K espressi

come rapporto fra concentrazione dell’elemento Cipianta nella pianta e la concentrazione

dell’elemento Cisuolo nel suolo (K= Cipianta / Cisuolo).

Il fattore di assimilazione è influenzato dal potenziale ionico dei diversi elementi

considerati, dal pH, dalla temperatura e dalle caratteristiche della pianta.

Infine sono state eseguite analisi di bucce e vinaccioli che hanno mostrato un generale

arricchimento in metalli, ed in particolare in metalli alcalino-terrosi, per cui la permanenza dei

residui a contatto con i succhi può influire sulla concentrazione in metalli.

CONCLUSIONI

Nell’area veneta indagata, i vigneti a DOC e DOCG devono l’elevata qualità e le

caratteristiche organolettiche alla natura vulcanica dei suoli. Grazie all’analisi del rapporto fra

concentrazione dei macro e micronutrienti nei suoli e nei vini sono stati definiti i coefficienti

di assimilazione per le due varietà Verdicchio e Refosco p.r. e l’impronta geochimica del

suolo. La metodologia proposta non solo consente di verificare il rispetto dei contenuti in

metalli e tutelare la salute dei consumatori, ma anche di evitare la commercializzazione di

prodotti qualitativamente non idonei.

La qualità della produzione vitivinicola in suoli vulcanici ad elevate concentrazioni di

metalli di transizione richiede il controllo del rispetto dei limiti normativi e del corretto valore

nutrizionale in termini di macro e micronutrienti, caratteristiche riscontrate per Refosco p.r. e

Verdicchio dei Lessini che hanno basse ma corrette concentrazioni di Cromo, Nichel, Cobalto

e Vanadio. Va segnalato che il Refosco p.r. è risultato più sensibile rispetto al Verdicchio

all’assimilazione di Cr, Ni e V per cui, in casi di forti anomalie del contenuto di questi

metalli, il Verdicchio fornisce maggiori garanzie del rispetto dei parametri nutrizionali.

Le misure di radioattività naturale e antropica hanno messo in evidenza l’assenza di rischio.

Eventuali futuri monitoraggi potranno aiutare a definire la stabilità composizionale e quindi

il mantenimento delle caratteristiche geochimiche riscontrate per uva e vino.

BIBLIOGRAFIA

Caputo R. & Bosellini A. 1994- La flessura pedemontana del Veneto centrale: anticlinale di

rampa a sviluppo bloccato da condotti vulcanici. The pedealpine flexure Zone of central

Venetian Alps:a ramp anticline halted by volcanic conduits. Atti Tic. Sc. Terra 1994 (Serie

speciale), 1: 255-268.

Ferioli D.G. 2010 Tesi di Dottorato in Scienze della Terra - Tracciabilità delle provenienze e

valorizzazione dei prodotti alimentari attraverso nuovi sistemi di caratterizzazione

geochimica.

D.M. 29 dicembre 1986 - Caratteristiche e limiti di alcune sostanze contenute nei vini.

D.M. 25 ottobre 1999, n.471

Regolamento recante criteri, procedure e modalita' per la messa in sicurezza, la bonifica e il

ripristino ambientale dei siti inquinati, ai sensi dell'articolo 17 del decreto legislativo 5

febbraio 1997, n. 22, e successive modificazioni e integrazioni.

Decreto Legislativo 3 aprile 2006, n. 152 Norme in materia ambientale.

Regolamento (CEE) N. 1881/2006 della Commissione del 19 dicembre 2006 che definisce i

tenori massimi di alcuni contaminanti nei prodotti alimentari.

4 - 113



APPLICATION OF ORGANIC CARBON STATUS INDICATORS ON VINEYARD SOILS: THE CASE STUDY OF DOC PIAVE

(VENETO REGION, ITALY)

G. Manni(1), G. Concheri(1), A. Garlato(2), I. Vinci(2), P. Marcuzzo(3)(1)Università degli Studi di Padova - Dipartimento di Biotecnologie Agrarie

Viale dell’Università 16, 35020 Legnaro (PD), Italia [email protected]

(2)ARPAV - Agenzia Regionale per la Prevenzione e Protezione Ambientale del Veneto - Servizio Suoli Via Santa Barbara 5/a, 31100 Treviso, Italia

[email protected](3)Centro di Ricerca per l’Agricoltura-Viticoltura Via XXVIII Aprile 26, Conegliano (TV), Italia

[email protected]

ABSTRACT According to the Kyoto Protocol objectives, it’s necessary to identify alternative carbon

dioxide sinks, and vineyard soils could be a significant opportunity. A set of soil organic carbon status indicators, proposed by JRC (Stolbovoy, 2006), was

tested on vineyard soils of DOC Piave area (Veneto region) to validate it. Information available in the regional soil database for the study area (Soil Maps of Treviso

and Venice provinces at 1:50,000 scale with 614 soil profiles on about 150,000 ha, 5% of which with vineyards) was analysed to point out significant relationships between soil organic carbon content, soil type and land uses. An approach for functional soil groups was adopted: the soil typological units were grouped on the basis of texture, coarse fragments, drainage and physiography (Manni, 2007). The highest value, which differs statistically from the others, was observed in fine texture and poorly drained soils. Furthermore, vineyard soils showed higher content than crop soils, especially on the first 30 cm. But no significant differences were observed. Then, for each functional group and separately for vineyard and crop topsoil and subsoil, a set of soil organic carbon status indicators were defined. The results showed higher capacity to sequestrate carbon on vineyard topsoil.

The present study allows an overview of the DOC Piave area carbon pool and highlights priorities areas where policy interventions should be concentrated.

KEYWORDSoil organic carbon – sequestration – vineyard – indicator – functional group.

INTRODUCTION According to the Kyoto Protocol objectives to reduce greenhouse gases emissions, in the

last years the joined countries are been elaborating footprint wine carbon calculators. In 2007, the Australian law imposed to the main wineries and winemakers to count emissions; they are asked to communicate greenhouse gases emissions, productions and energy uses. Also Italy is working on the first wine carbon calculator called Ita.Ca® and, in the next program versions, vineyard soils’s role of CO2 sequestering into organic matter will be introduced (Battaglene et al., 2010). So viticulture represents a carbon sink.

In a previous study, conducted on the alluvial Veneto plain of Brenta and Piave rivers, soil organic carbon (SOC) of different land uses was considered (Manni, 2007). The OC trend of

the topsoil (0-30 cm) was compared between orchard, meadow, vineyard, vegetable and corn. Vineyard soils showed a significantly higher OC content (49 t/ha).

To study vineyards’ capacity to sequestrate OC, the DOC Piave area has been investigated (Fig. 1). The study area is located in Treviso and Venice provinces and is near 150,000 ha extended. The main land uses are crop and, for about 5% of the area, vineyard.

In the present study a set of Soil Organic Carbon Status Indicators (SOCSI), developed by the JRC-Ispra to support the EU policies related to SOC, are been considered to investigate OC trends in different soil types and land uses. The SOCSI are knowledge-based and can be derived from available soil data at regional scale. SOC content results from combination of Soil Typological Unit (STU) and land use/management. Each combination has specific SOC margins

so SOC content change is limited. Moreover, potential for the change depends on the actual OC content.

The present study aims to test the SOCSI in the DOC Piave area and validate them against empirical observations. There is an urgency to make the SOCSI instrumental for supporting authorities to setup policy decisions regarding carbon management (Stolbovoy, 2008).

MATERIALS AND METHODS To describe the different soil types of DOC Piave area, soil maps of Treviso and Venice

provinces at 1:50,000 scale were considered (Soil Service-ARPAV Treviso, 2008); also, to distinguish between the different land uses, the covered soil map of Veneto Region at

1:10,000 scale (Regione Veneto, 2009) was used. The study area includes 614 soil profiles, 153 of

which on vineyard (Fig. 2); each of them is geographically referred with a GIS system (ArcView 3.3®) and is been reconnected to the STUs defined for the soil maps. Soil data are been collected in a regional database managed with Access® 2003. It includes soil profiles and STUs description, soil chemical and physical analysis for each horizon. In particular, the OC value considered is been analyzed according to 14,235 ISO method. Bulk density, necessary to convert OC % into t/ha, is been measured by a soil sample ring of known dimension (100 cc) (ISO 11,272-core method) or is been estimated by pedotransfer specific for plain soils.

Through a set of query, topsoil (0-30 cm) and subsoil (0-100 cm) OC values, are been calculated by weighting each horizon’s OC content. The OC trend, between different soil types and land uses, has been represented by bar charts, then significant differences are been investigated by STATISTICA 8® program (LSD test, p=0.05).

Since data for each of 70 STUs were sometimes too little to work out statistical analysis, the functional soil groups defined for the alluvial plain of Brenta and Piave rivers (Manni, 2007) were adopted. STUs were grouped into functional groups (Tab.1) on the basis of texture, coarse fragments content, drainage and physiography. But not all the groups differed significantly.

Fig. 1. DOC Piave area.

Fig. 2. Soil profiles on DOC Piave area.

4 - 114



APPLICATION OF ORGANIC CARBON STATUS INDICATORS ON VINEYARD SOILS: THE CASE STUDY OF DOC PIAVE

(VENETO REGION, ITALY)

G. Manni(1), G. Concheri(1), A. Garlato(2), I. Vinci(2), P. Marcuzzo(3)(1)Università degli Studi di Padova - Dipartimento di Biotecnologie Agrarie

Viale dell’Università 16, 35020 Legnaro (PD), Italia [email protected]

(2)ARPAV - Agenzia Regionale per la Prevenzione e Protezione Ambientale del Veneto - Servizio Suoli Via Santa Barbara 5/a, 31100 Treviso, Italia

[email protected](3)Centro di Ricerca per l’Agricoltura-Viticoltura Via XXVIII Aprile 26, Conegliano (TV), Italia

[email protected]

ABSTRACT According to the Kyoto Protocol objectives, it’s necessary to identify alternative carbon

dioxide sinks, and vineyard soils could be a significant opportunity. A set of soil organic carbon status indicators, proposed by JRC (Stolbovoy, 2006), was

tested on vineyard soils of DOC Piave area (Veneto region) to validate it. Information available in the regional soil database for the study area (Soil Maps of Treviso

and Venice provinces at 1:50,000 scale with 614 soil profiles on about 150,000 ha, 5% of which with vineyards) was analysed to point out significant relationships between soil organic carbon content, soil type and land uses. An approach for functional soil groups was adopted: the soil typological units were grouped on the basis of texture, coarse fragments, drainage and physiography (Manni, 2007). The highest value, which differs statistically from the others, was observed in fine texture and poorly drained soils. Furthermore, vineyard soils showed higher content than crop soils, especially on the first 30 cm. But no significant differences were observed. Then, for each functional group and separately for vineyard and crop topsoil and subsoil, a set of soil organic carbon status indicators were defined. The results showed higher capacity to sequestrate carbon on vineyard topsoil.

The present study allows an overview of the DOC Piave area carbon pool and highlights priorities areas where policy interventions should be concentrated.

KEYWORDSoil organic carbon – sequestration – vineyard – indicator – functional group.

INTRODUCTION According to the Kyoto Protocol objectives to reduce greenhouse gases emissions, in the

last years the joined countries are been elaborating footprint wine carbon calculators. In 2007, the Australian law imposed to the main wineries and winemakers to count emissions; they are asked to communicate greenhouse gases emissions, productions and energy uses. Also Italy is working on the first wine carbon calculator called Ita.Ca® and, in the next program versions, vineyard soils’s role of CO2 sequestering into organic matter will be introduced (Battaglene et al., 2010). So viticulture represents a carbon sink.

In a previous study, conducted on the alluvial Veneto plain of Brenta and Piave rivers, soil organic carbon (SOC) of different land uses was considered (Manni, 2007). The OC trend of

the topsoil (0-30 cm) was compared between orchard, meadow, vineyard, vegetable and corn. Vineyard soils showed a significantly higher OC content (49 t/ha).

To study vineyards’ capacity to sequestrate OC, the DOC Piave area has been investigated (Fig. 1). The study area is located in Treviso and Venice provinces and is near 150,000 ha extended. The main land uses are crop and, for about 5% of the area, vineyard.

In the present study a set of Soil Organic Carbon Status Indicators (SOCSI), developed by the JRC-Ispra to support the EU policies related to SOC, are been considered to investigate OC trends in different soil types and land uses. The SOCSI are knowledge-based and can be derived from available soil data at regional scale. SOC content results from combination of Soil Typological Unit (STU) and land use/management. Each combination has specific SOC margins

so SOC content change is limited. Moreover, potential for the change depends on the actual OC content.

The present study aims to test the SOCSI in the DOC Piave area and validate them against empirical observations. There is an urgency to make the SOCSI instrumental for supporting authorities to setup policy decisions regarding carbon management (Stolbovoy, 2008).

MATERIALS AND METHODS To describe the different soil types of DOC Piave area, soil maps of Treviso and Venice

provinces at 1:50,000 scale were considered (Soil Service-ARPAV Treviso, 2008); also, to distinguish between the different land uses, the covered soil map of Veneto Region at

1:10,000 scale (Regione Veneto, 2009) was used. The study area includes 614 soil profiles, 153 of

which on vineyard (Fig. 2); each of them is geographically referred with a GIS system (ArcView 3.3®) and is been reconnected to the STUs defined for the soil maps. Soil data are been collected in a regional database managed with Access® 2003. It includes soil profiles and STUs description, soil chemical and physical analysis for each horizon. In particular, the OC value considered is been analyzed according to 14,235 ISO method. Bulk density, necessary to convert OC % into t/ha, is been measured by a soil sample ring of known dimension (100 cc) (ISO 11,272-core method) or is been estimated by pedotransfer specific for plain soils.

Through a set of query, topsoil (0-30 cm) and subsoil (0-100 cm) OC values, are been calculated by weighting each horizon’s OC content. The OC trend, between different soil types and land uses, has been represented by bar charts, then significant differences are been investigated by STATISTICA 8® program (LSD test, p=0.05).

Since data for each of 70 STUs were sometimes too little to work out statistical analysis, the functional soil groups defined for the alluvial plain of Brenta and Piave rivers (Manni, 2007) were adopted. STUs were grouped into functional groups (Tab.1) on the basis of texture, coarse fragments content, drainage and physiography. But not all the groups differed significantly.

Fig. 1. DOC Piave area.

Fig. 2. Soil profiles on DOC Piave area.

4 - 115



Tab. 1. Functional groups description. FUNCTIONAL

GROUP LANDSCAPE COARSE

FRAGMENTS DRAINAGE TEXTURE

A1 1-15% A2 High plain >15% B2 <1% excessively drained, somewhat

excessively drained, well drained Corse loamy, coarse silty

B3 <1% excessively drained, somewhat excessively drained, well drained

Fine loamy, fine silty, clayey

B4 <1% moderately well drained Sandy, coarse loamy, coarse silty

B5 <1% moderately well drained Fine loamy, fine silty, clayey

B6

Low plain

<1% somewhat poorly drained, poorly drained, very poorly drained


O2 Soils with mollic horizon


All textures

A set of SOCSI, proposed by JRC (Stolbovoy, 2006), has been applied for each functional group. The set includes: • data-derived parameters (mean, minimum and maximum values); • knowledge-derived parameters (CSP-Carbon Sequestration Potential, PCL-Potential

Carbon Loss, CSR-Carbon Sequestration Rate, CLR-Carbon Loss Rate and capability classes for OC change).

The minimum and maximum values represent the margins of the OC range of change. Potential for the change depends on the actual OC content (Fig. 3).

Fig. 3. OC range of change (Stolbovoy, 2006).

Following the JRC procedure, the low/medium/high capability classes of CSP and PCL were defined for each functional group:

• Low (L): < [Min + (Max – Min)/3]• Medium (M): between [Min + (Max – Min)/3] and [Min + 2(Max – Min)/3]• High (H): > [Min + 2(Max – Min)/3

These SOCSI could be drawn on maps (one for CSP and one for PCL) to show areas in low, medium or high PCL/CSP classes.

RESULTS AND DISCUSSION Bar charts of functional groups’ topsoil (0-30 cm) and subsoil (0-100 cm) OC content (t/ha

and %) on vineyard and crop were drawn. The most significant results are on vineyard’s OC (t/ha) graphics. Both the topsoil and the subsoil bar charts (Fig. 4 and 5) show a close relationship between soil properties and SOC content:

• in the high plain, soils with lower coarse fragments content (A1 than A2) have higher fine earth volume and so higher OC content;

• in the low plain (from B2 to B6), OC trend increases according to clay content and in opposition to the soil drainage (except for B2 group on topsoil): sandy soils result in more ready oxidation of organic matter compared with heavier soils because have lower moisture content and are more aerated;

• the O2 group has obviously the highest OC content, in fact soils are characterized by mollic horizon and are very poorly drained.

Fig. 4. Vineyard’s topsoil OC content (t/ha) of the functional groups.

OC100_tha-Functional Groups

0

50

100

150

200

250

A1 A2 B2 B4 B5 B6 O2

Functional Groups

OC

100_

tha

Fig. 5. Vineyard’s subsoil OC content (t/ha) of the functional groups.

The only significant differences are between B4-B6 and between O2-all the other groups; if subsoil OC is considered, also A2-B5 and A2-B6 differ significantly.

Then, vineyard and crop land uses were compared. Vineyard shows higher values than crop on topsoil for all the functional groups (there are no observations for B3 group on vineyard), except for O2 group (Fig. 6). But no significant differences are observed.


0

20

40

60

80

100

120

140

A1 A2 B2 B4 B5 B6 O2

Functional Groups

OC3

0_th

a

4 - 116



Tab. 1. Functional groups description. FUNCTIONAL

GROUP LANDSCAPE COARSE

FRAGMENTS DRAINAGE TEXTURE

A1 1-15% A2 High plain >15% B2 <1% excessively drained, somewhat

excessively drained, well drained Corse loamy, coarse silty

B3 <1% excessively drained, somewhat excessively drained, well drained


B4 <1% moderately well drained Sandy, coarse loamy, coarse silty

B5 <1% moderately well drained Fine loamy, fine silty, clayey

B6

Low plain



O2 Soils with mollic horizon


All textures

A set of SOCSI, proposed by JRC (Stolbovoy, 2006), has been applied for each functional group. The set includes: • data-derived parameters (mean, minimum and maximum values); • knowledge-derived parameters (CSP-Carbon Sequestration Potential, PCL-Potential

Carbon Loss, CSR-Carbon Sequestration Rate, CLR-Carbon Loss Rate and capability classes for OC change).

The minimum and maximum values represent the margins of the OC range of change. Potential for the change depends on the actual OC content (Fig. 3).

Fig. 3. OC range of change (Stolbovoy, 2006).

Following the JRC procedure, the low/medium/high capability classes of CSP and PCL were defined for each functional group:

• Low (L): < [Min + (Max – Min)/3]• Medium (M): between [Min + (Max – Min)/3] and [Min + 2(Max – Min)/3]• High (H): > [Min + 2(Max – Min)/3

These SOCSI could be drawn on maps (one for CSP and one for PCL) to show areas in low, medium or high PCL/CSP classes.

RESULTS AND DISCUSSION Bar charts of functional groups’ topsoil (0-30 cm) and subsoil (0-100 cm) OC content (t/ha

and %) on vineyard and crop were drawn. The most significant results are on vineyard’s OC (t/ha) graphics. Both the topsoil and the subsoil bar charts (Fig. 4 and 5) show a close relationship between soil properties and SOC content:

• in the high plain, soils with lower coarse fragments content (A1 than A2) have higher fine earth volume and so higher OC content;

• in the low plain (from B2 to B6), OC trend increases according to clay content and in opposition to the soil drainage (except for B2 group on topsoil): sandy soils result in more ready oxidation of organic matter compared with heavier soils because have lower moisture content and are more aerated;

• the O2 group has obviously the highest OC content, in fact soils are characterized by mollic horizon and are very poorly drained.

Fig. 4. Vineyard’s topsoil OC content (t/ha) of the functional groups.


0

50

100

150

200

250

A1 A2 B2 B4 B5 B6 O2

Functional Groups

OC

100_

tha

Fig. 5. Vineyard’s subsoil OC content (t/ha) of the functional groups.

The only significant differences are between B4-B6 and between O2-all the other groups; if subsoil OC is considered, also A2-B5 and A2-B6 differ significantly.

Then, vineyard and crop land uses were compared. Vineyard shows higher values than crop on topsoil for all the functional groups (there are no observations for B3 group on vineyard), except for O2 group (Fig. 6). But no significant differences are observed.


0

20

40

60

80

100

120

140

A1 A2 B2 B4 B5 B6 O2

Functional Groups

OC3

0_th

a

4 - 117



OC30_tha-LAND USE

0

20

40

60

80

100

120

140

160

180

200

220

A1 A2 B2 B3 B4 B5 B6 O2

Functional Groups

OC

30_t

ha VineyardCrop

Fig. 6. Topsoil OC content (t/ha) of vineyard and crop land uses.

Topsoil OC content on vineyard is higher because of the conservative tillage adopted by the wineries included in the DOC Piave zoning. Information about vineyard’s management are been collected by interviews to the about 90 wineries involved in the present study. In particular, they were asked to give information about soil manure (chemical or organic), irrigation, soil tillage, presence of grass covered soil. Almost all the wineries apply also organic manure on soil, have a reduced tillage and, at least, an inter-row grass covered soil. These agricultural practices maintain or increase soil organic matter content, especially on the first 30 cm.

The typical topsoil OC trend is, instead, lost in the subsoil OC content: only A1, A2 and B5 groups have higher content on vineyard.

Following a procedure proposed by the JRC (Stolbovoy, 2006), the minimum, mean and maximum OC functional groups values were found (for topsoil and subsoil, in t/ha and %), separately for vineyard and crop land uses. Then, the capability classes of PCL and CSP were calculated. The most significant results are shown by OC (%) on vineyard topsoil, where some functional groups have low potential to loss and high potential to gain carbon (L/H); subsoil, instead, has medium potentials (M/M) for all the groups. No H/L classes were found.

The groups showing major potentials of changing (with L/H classes) are A1, B2, B5 and B6 if OC(%) is considered. So, applying an appropriate vineyard’s management, it’s possible to increase topsoil organic matter for these functional groups.

The crop capability classes show, in opposition to vineyard classes, more possibilities to change in the subsoil than in the topsoil, so interventions to increase SOC on crops is more complicated. A conversion from crop to vineyard, instead, could allow to increase SOC pool acting on topsoil.

The capability classes founded could be drawn on maps to better highlight the priority areas where interventions on soil protection should be concentrated.

CONCLUSIONS Since topsoil OC (t/ha) was significantly higher on vineyard than on the other main land

uses of the Veneto plain (Manni, 2007), the DOC Piave area is been investigated. Information available in the regional soil database was analysed to point out significant

relationships between SOC content on different soil types and land uses (vineyard-crop).

Functional soil groups were created and bar charts of topsoil (0-30 cm) and subsoil (0-100 cm) OC content (% and t/ha) of the different functional groups were obtained, separately for vineyard and crop land uses. In the high plain, vineyard’s soils with lower coarse fragments content (A1 than A2) have higher OC (t/ha) values; in the low plain (from B2 to B6), OC trend increases according to clay content and in opposition to the soil drainage (except for B2 group on topsoil); the mollic and very poorly drained soils (O2) have obviously the highest OC content. Furthermore, vineyard shows higher values than crop on topsoil for all the functional groups, except for O2 group. Topsoil OC content on vineyard is higher because of the conservative tillage adopted by the wineries involved in the present study. However, the functional groups will be created again on the basis of other factors to better highlight differences between different soil types on different land uses.

Following a procedure proposed by the JRC (Stolbovoy, 2006), a set of SOCSI were calculated. Some functional groups, on topsoil vineyard, showed low potential to loss and high potential to gain OC. So, applying an appropriate vineyard’s management, it’s possible to increase topsoil organic matter for these soil groups. The capability classes founded could be drawn on maps to highlight the regional areas where policy measures and interventions should be concentrated to guarantee soil protection.

ACKNOWLEDGMENTS The authors thank the Council for Agriculture Research and Experimentation-Research

Centre for Viticulture of Conegliano and the wineries involved in the DOC Piave zoning for the furnished information about vineyard’s management.

BIBLIOGRAPHYARPAV, 2005. Carta dei suoli della Regione Veneto alla scala 1:250.000. Treviso.ARPAV, 2008. Carta dei suoli della provincia di Treviso. Treviso. ARPAV, 2004. Carta dei suoli del bacino scolante in laguna di Venezia. Treviso. ARPAV, 2008. I suoli della provincia di Venezia. Padova. Battaglene T., Savage C., 2010. Importante un metodo comune per valutare l’impronta

carbonica. Vite&Vino (supplemento a L’Informatore Agrario n.13), n:11-13. Jones R.J.A., Hiederer R., Rusco E., Loveland P.J., Montanarella L., 2004. Function of

OC/OM in soils. In: The map of Topsoil Organic Carbon in Europe: Version 1.2-september 2003. Ispra. European Communities. n. 6.

Manni G., 2007. Andamento del carbonio organico nei suoli di pianura del Veneto in funzione del tipo di suoli e del loro uso. Università degli Studi di Padova.

Manni G., 2008. Towards soil organic carbon status indicators in the Veneto Region. In: EUROSOIL 2008 Book of Abstracts. Vienna. n. 83.

Manni G., Piccolo S., Concheri G., Vinci I., 2009. I suoli vitati e il Protocollo di Kyoto: il caso della DOC Piave (Regione Veneto, Italia). In: 32nd World Congress of vine and wine-Book of Abstracts. Zagabria. n. 249.

Regione del Veneto, 2009. Carta della Copertura del Suolo del Veneto. Edizione 2009. Stolbovoy V., 2008. Application of soil organic carbon for policy-decision making in the

EU. In. EUROSOIL 2008 Book of Abstracts. Vienna. n. 83. Zdruli P., Jones R.J.A., Montanarella L., 2004. Effect of soil properties. In: Organic matter in

the soils of southern Europe. Ispra. European Communities. n. 8. http://eusoils.jrc.ec.europa.eu/esbn/Plenary_esbn_2007/ESBN_2007/ESBN_Stolbovoy_Ca

rbon.pdf

4 - 118



OC30_tha-LAND USE

0

20

40

60

80

100

120

140

160

180

200

220

A1 A2 B2 B3 B4 B5 B6 O2

Functional Groups

OC

30_t

ha VineyardCrop

Fig. 6. Topsoil OC content (t/ha) of vineyard and crop land uses.

Topsoil OC content on vineyard is higher because of the conservative tillage adopted by the wineries included in the DOC Piave zoning. Information about vineyard’s management are been collected by interviews to the about 90 wineries involved in the present study. In particular, they were asked to give information about soil manure (chemical or organic), irrigation, soil tillage, presence of grass covered soil. Almost all the wineries apply also organic manure on soil, have a reduced tillage and, at least, an inter-row grass covered soil. These agricultural practices maintain or increase soil organic matter content, especially on the first 30 cm.

The typical topsoil OC trend is, instead, lost in the subsoil OC content: only A1, A2 and B5 groups have higher content on vineyard.

Following a procedure proposed by the JRC (Stolbovoy, 2006), the minimum, mean and maximum OC functional groups values were found (for topsoil and subsoil, in t/ha and %), separately for vineyard and crop land uses. Then, the capability classes of PCL and CSP were calculated. The most significant results are shown by OC (%) on vineyard topsoil, where some functional groups have low potential to loss and high potential to gain carbon (L/H); subsoil, instead, has medium potentials (M/M) for all the groups. No H/L classes were found.

The groups showing major potentials of changing (with L/H classes) are A1, B2, B5 and B6 if OC(%) is considered. So, applying an appropriate vineyard’s management, it’s possible to increase topsoil organic matter for these functional groups.

The crop capability classes show, in opposition to vineyard classes, more possibilities to change in the subsoil than in the topsoil, so interventions to increase SOC on crops is more complicated. A conversion from crop to vineyard, instead, could allow to increase SOC pool acting on topsoil.

The capability classes founded could be drawn on maps to better highlight the priority areas where interventions on soil protection should be concentrated.

CONCLUSIONS Since topsoil OC (t/ha) was significantly higher on vineyard than on the other main land

uses of the Veneto plain (Manni, 2007), the DOC Piave area is been investigated. Information available in the regional soil database was analysed to point out significant

relationships between SOC content on different soil types and land uses (vineyard-crop).

Functional soil groups were created and bar charts of topsoil (0-30 cm) and subsoil (0-100 cm) OC content (% and t/ha) of the different functional groups were obtained, separately for vineyard and crop land uses. In the high plain, vineyard’s soils with lower coarse fragments content (A1 than A2) have higher OC (t/ha) values; in the low plain (from B2 to B6), OC trend increases according to clay content and in opposition to the soil drainage (except for B2 group on topsoil); the mollic and very poorly drained soils (O2) have obviously the highest OC content. Furthermore, vineyard shows higher values than crop on topsoil for all the functional groups, except for O2 group. Topsoil OC content on vineyard is higher because of the conservative tillage adopted by the wineries involved in the present study. However, the functional groups will be created again on the basis of other factors to better highlight differences between different soil types on different land uses.

Following a procedure proposed by the JRC (Stolbovoy, 2006), a set of SOCSI were calculated. Some functional groups, on topsoil vineyard, showed low potential to loss and high potential to gain OC. So, applying an appropriate vineyard’s management, it’s possible to increase topsoil organic matter for these soil groups. The capability classes founded could be drawn on maps to highlight the regional areas where policy measures and interventions should be concentrated to guarantee soil protection.

ACKNOWLEDGMENTS The authors thank the Council for Agriculture Research and Experimentation-Research

Centre for Viticulture of Conegliano and the wineries involved in the DOC Piave zoning for the furnished information about vineyard’s management.

BIBLIOGRAPHYARPAV, 2005. Carta dei suoli della Regione Veneto alla scala 1:250.000. Treviso.ARPAV, 2008. Carta dei suoli della provincia di Treviso. Treviso. ARPAV, 2004. Carta dei suoli del bacino scolante in laguna di Venezia. Treviso. ARPAV, 2008. I suoli della provincia di Venezia. Padova. Battaglene T., Savage C., 2010. Importante un metodo comune per valutare l’impronta

carbonica. Vite&Vino (supplemento a L’Informatore Agrario n.13), n:11-13. Jones R.J.A., Hiederer R., Rusco E., Loveland P.J., Montanarella L., 2004. Function of

OC/OM in soils. In: The map of Topsoil Organic Carbon in Europe: Version 1.2-september 2003. Ispra. European Communities. n. 6.

Manni G., 2007. Andamento del carbonio organico nei suoli di pianura del Veneto in funzione del tipo di suoli e del loro uso. Università degli Studi di Padova.

Manni G., 2008. Towards soil organic carbon status indicators in the Veneto Region. In: EUROSOIL 2008 Book of Abstracts. Vienna. n. 83.

Manni G., Piccolo S., Concheri G., Vinci I., 2009. I suoli vitati e il Protocollo di Kyoto: il caso della DOC Piave (Regione Veneto, Italia). In: 32nd World Congress of vine and wine-Book of Abstracts. Zagabria. n. 249.

Regione del Veneto, 2009. Carta della Copertura del Suolo del Veneto. Edizione 2009. Stolbovoy V., 2008. Application of soil organic carbon for policy-decision making in the

EU. In. EUROSOIL 2008 Book of Abstracts. Vienna. n. 83. Zdruli P., Jones R.J.A., Montanarella L., 2004. Effect of soil properties. In: Organic matter in

the soils of southern Europe. Ispra. European Communities. n. 8. http://eusoils.jrc.ec.europa.eu/esbn/Plenary_esbn_2007/ESBN_2007/ESBN_Stolbovoy_Ca

rbon.pdf

4 - 119



ON-THE-GO RESISTIVITY SENSORS EMPLOYMENT TO SUPPORT SOIL SURVEY FOR PRECISION VITICULTURE

M.C. Andrenelli, E.A.C. Costantini, S. Pellegrini, R. Perria, and N. Vignozzi CRA-ABP- Centro per l'Agrobiologia e la Pedologia,

Piazza M. D’Azeglio, 30 50121, Firenze, Italy [email protected]

ABSTRACT There is an increasing need in agriculture to adopt site-specific management (precision

farming) because of economic and environmental pressures. Geophysical on-the-go sensors, such as the ARP (Automatic Resistivity Profiling) system, can effectively support soil survey by optimizing sampling density according to the spatial variability of apparent electrical resistivity (ER).

The aim of this work was to test the sensitivity of the ARP methodology in supporting soil survey for precision viticulture. In particular, an optimization procedure for coupled geoelectrical and soil surveys is illustrated.

The research was carried out in a vineyard located in Tuscany (central Italy) affected by low yield due to soil salinity; the investigation was simultaneously conducted by soil survey and resistivity measurements. The ARP method consists in the electric current injection into the ground and in the continuous measure of the resulting potential, simultaneously providing three georeferenced values of ER related to 50, 100 and 170 cm depths for each point.

Forty-nine soil samples were taken at 10-30 cm depth and analyzed for moisture, particle size distribution and electrical conductivity. The best correlation (R2 = 0.609; P <0.01) was obtained between clay content and ER referred to the 0-50 cm depth (ER50).

The evaluation of the density reduction effect for both ARP and soil survey was expressed in terms of ER50 and clay predictability. Doubling the ARP swaths width (12 m) the ER50 accuracy was substantially in agreement with that obtained for the highest ARP survey density (22 swaths 6 m spaced); the further width doubling (24 m) provided a moderate accuracy. With regard to clay content prediction k accuracy values ranged between 0.87 and 0.49 for the 22 swaths/25 soil samples and 10 swaths/12 soil samples combination, respectively.

KEYWORDARP – ER – accuracy – precision viticulture – GIS – clay INTRODUCTIONViticultural precision farming needs detailed soil information, which can be obtained by means

of remote as well as proximal sensors, besides traditional invasive soil survey. The understanding of the nature, extent and causes of vineyard variability may help grape-growers and winemakers to use precision viticulture tools to better target their management (irrigation, rate of fertilizers, pruning and harvesting). Nevertheless, the use of the new technologies is still in its infancy, because of their costs and the lack of knowledge about the detail actually needed for the viticultural husbandry.

Several authors (Bramley and Proffitt, 1999) demonstrated that traditional soil surveys can not succeed in exhaustively explaining the reasons of variability in vineyard performance. The authors find more efficient the evaluation of soil properties by sampling at points selected

according to, for instance, electromagnetic measurements (EM38). Actually, soil electrical properties can be considered as an alternative but also complex source of information for assessing the spatial and temporal variability of many soil physical and chemical properties (i.e. structure, texture, water content and salinity). EMI represents the most widespread geophysical technique employed in agriculture, anyway, it is noteworthy that electrical surveys performed by means of this instrumentation require a calibration every time it is used (Taylor, 2004). With this regard, Dabas et al. (2001) prefer electrical current (i.e., device that injects electrical current into the soil) as the calibration is more constant and less sensitive to error from soil heterogeneity, though the limitation regarding the main drawback to the DC sensors (direct current) are problems when the soil exhibits a high contact resistance, that is when it is either very dry or frozen (Dabas and Tabbagh, 2003; Luck and Eisenreich, 2001).

With the aim to reduce the mobile sensor surveys costs Farahan and Flynn (2007) studied the different quality of maps provided by widening the swath width for the Veris 3100 sensor (Veris Technologies, Salina, KS). These authors assess the density effect on the possibility of providing acceptable prediction of the conductivity (ECa) map compared to the densest survey.

Since the density effect of the geoelectrical survey on its reliability to support traditional soil survey in vineyard was not fully investigated, the aim of the work was to statistically test the possibility of combining an optimized strategy for both geoelectrical and soil sampling, able to provide significant information accuracy of the soil spatial variability.

MATERIALS AND METHODS The study vineyard, sized 3.5 ha, is located in central coast of Tuscany (Central Italy),

cultivated with Cabernet Sauvignon and Cabernet Franc in the past. Actually, soil salinity problems strongly reduced the wine production and induced the wine growers to remove the vineyard.

The survey identified three main soil typologies, according to the WRB classification system (FAO, IUSS, ISRIC, 2006): Endostagnic Cambisols (Calcaric, Sodic) on marine clays; Haplic Cambisols (Eutric) and (Calcaric) on conglomerates.

Soil sampling at 10-30 cm was carried out on a regular grid sampling scheme (35-40 m per 20 m), simultaneously to the measurement of soil resistivity executed by the ARP equipment. (a direct current sensor). Laboratory analyses for moisture determination was carried out with the gravimetric method while the texture analysis was performed with hydrometer, identifying five fractions (coarse and fine sand, coarse and fine silt, clay percentages); a 1:5 soil water suspension was then employed for the electrical conductivity determination, expressed as mScm-1. The ARP survey was carried on 22 passages, 6 m spaced from each other. In each sampling point, the device simultaneously provides 3 georeferenced values of electrical resistivity values (Ohm.m) related to different soil depths of investigation (0-50; 0-100 and 0-170 cm). Actually, only the surface ER data (ER50) have been considered compared to the deeper ones provided by the ARP machinery because such value was expected to be more linked to soil properties of 10-30 cm depth.

In order to find a relationship between ARP information and soil properties, resistivity data were spatialized over the whole study area in ARC/VIEW GIS environment (ESRI ArcView 3.2(R)) by means of inverse distance weighted interpolation (IDW2-3) algorithm. Such an algorithm employees 2 neighbours and a 3 power function to interpolate the data. Successively, a buffer of 3 m radius around each soil sample was created and by means of the ArcView tool

4 - 120



ON-THE-GO RESISTIVITY SENSORS EMPLOYMENT TO SUPPORT SOIL SURVEY FOR PRECISION VITICULTURE

M.C. Andrenelli, E.A.C. Costantini, S. Pellegrini, R. Perria, and N. Vignozzi CRA-ABP- Centro per l'Agrobiologia e la Pedologia,

Piazza M. D’Azeglio, 30 50121, Firenze, Italy [email protected]

ABSTRACT There is an increasing need in agriculture to adopt site-specific management (precision

farming) because of economic and environmental pressures. Geophysical on-the-go sensors, such as the ARP (Automatic Resistivity Profiling) system, can effectively support soil survey by optimizing sampling density according to the spatial variability of apparent electrical resistivity (ER).

The aim of this work was to test the sensitivity of the ARP methodology in supporting soil survey for precision viticulture. In particular, an optimization procedure for coupled geoelectrical and soil surveys is illustrated.

The research was carried out in a vineyard located in Tuscany (central Italy) affected by low yield due to soil salinity; the investigation was simultaneously conducted by soil survey and resistivity measurements. The ARP method consists in the electric current injection into the ground and in the continuous measure of the resulting potential, simultaneously providing three georeferenced values of ER related to 50, 100 and 170 cm depths for each point.

Forty-nine soil samples were taken at 10-30 cm depth and analyzed for moisture, particle size distribution and electrical conductivity. The best correlation (R2 = 0.609; P <0.01) was obtained between clay content and ER referred to the 0-50 cm depth (ER50).

The evaluation of the density reduction effect for both ARP and soil survey was expressed in terms of ER50 and clay predictability. Doubling the ARP swaths width (12 m) the ER50 accuracy was substantially in agreement with that obtained for the highest ARP survey density (22 swaths 6 m spaced); the further width doubling (24 m) provided a moderate accuracy. With regard to clay content prediction k accuracy values ranged between 0.87 and 0.49 for the 22 swaths/25 soil samples and 10 swaths/12 soil samples combination, respectively.

KEYWORDARP – ER – accuracy – precision viticulture – GIS – clay INTRODUCTIONViticultural precision farming needs detailed soil information, which can be obtained by means

of remote as well as proximal sensors, besides traditional invasive soil survey. The understanding of the nature, extent and causes of vineyard variability may help grape-growers and winemakers to use precision viticulture tools to better target their management (irrigation, rate of fertilizers, pruning and harvesting). Nevertheless, the use of the new technologies is still in its infancy, because of their costs and the lack of knowledge about the detail actually needed for the viticultural husbandry.

Several authors (Bramley and Proffitt, 1999) demonstrated that traditional soil surveys can not succeed in exhaustively explaining the reasons of variability in vineyard performance. The authors find more efficient the evaluation of soil properties by sampling at points selected

according to, for instance, electromagnetic measurements (EM38). Actually, soil electrical properties can be considered as an alternative but also complex source of information for assessing the spatial and temporal variability of many soil physical and chemical properties (i.e. structure, texture, water content and salinity). EMI represents the most widespread geophysical technique employed in agriculture, anyway, it is noteworthy that electrical surveys performed by means of this instrumentation require a calibration every time it is used (Taylor, 2004). With this regard, Dabas et al. (2001) prefer electrical current (i.e., device that injects electrical current into the soil) as the calibration is more constant and less sensitive to error from soil heterogeneity, though the limitation regarding the main drawback to the DC sensors (direct current) are problems when the soil exhibits a high contact resistance, that is when it is either very dry or frozen (Dabas and Tabbagh, 2003; Luck and Eisenreich, 2001).

With the aim to reduce the mobile sensor surveys costs Farahan and Flynn (2007) studied the different quality of maps provided by widening the swath width for the Veris 3100 sensor (Veris Technologies, Salina, KS). These authors assess the density effect on the possibility of providing acceptable prediction of the conductivity (ECa) map compared to the densest survey.

Since the density effect of the geoelectrical survey on its reliability to support traditional soil survey in vineyard was not fully investigated, the aim of the work was to statistically test the possibility of combining an optimized strategy for both geoelectrical and soil sampling, able to provide significant information accuracy of the soil spatial variability.

MATERIALS AND METHODS The study vineyard, sized 3.5 ha, is located in central coast of Tuscany (Central Italy),

cultivated with Cabernet Sauvignon and Cabernet Franc in the past. Actually, soil salinity problems strongly reduced the wine production and induced the wine growers to remove the vineyard.

The survey identified three main soil typologies, according to the WRB classification system (FAO, IUSS, ISRIC, 2006): Endostagnic Cambisols (Calcaric, Sodic) on marine clays; Haplic Cambisols (Eutric) and (Calcaric) on conglomerates.

Soil sampling at 10-30 cm was carried out on a regular grid sampling scheme (35-40 m per 20 m), simultaneously to the measurement of soil resistivity executed by the ARP equipment. (a direct current sensor). Laboratory analyses for moisture determination was carried out with the gravimetric method while the texture analysis was performed with hydrometer, identifying five fractions (coarse and fine sand, coarse and fine silt, clay percentages); a 1:5 soil water suspension was then employed for the electrical conductivity determination, expressed as mScm-1. The ARP survey was carried on 22 passages, 6 m spaced from each other. In each sampling point, the device simultaneously provides 3 georeferenced values of electrical resistivity values (Ohm.m) related to different soil depths of investigation (0-50; 0-100 and 0-170 cm). Actually, only the surface ER data (ER50) have been considered compared to the deeper ones provided by the ARP machinery because such value was expected to be more linked to soil properties of 10-30 cm depth.

In order to find a relationship between ARP information and soil properties, resistivity data were spatialized over the whole study area in ARC/VIEW GIS environment (ESRI ArcView 3.2(R)) by means of inverse distance weighted interpolation (IDW2-3) algorithm. Such an algorithm employees 2 neighbours and a 3 power function to interpolate the data. Successively, a buffer of 3 m radius around each soil sample was created and by means of the ArcView tool

4 - 121



Zonal Statistics, the mean value of ER50 grids within each buffer was associated to the relative soil sample information. The resolution of the raster layer was 3 meter. Regression analysis was employed to correlate ER50 information to soil physical and chemical data.

In addition, to evaluate the possibility of reducing the ARP survey cost, the accuracy of predicting ER50 values for the swath widths of 6, 12 and 24 m was assessed. Similar approach was employed to evaluate the opportunity of reducing soil samplings by assessing soil properties predictability for decreasing sampling closeness. Those localizations were in turn selected according to the observed ER50 variability for the diverse densities of the ARP survey.

Such accuracy analysis was carried out in ARC/VIEW GIS environment by means of the ArcView tool Kappa analysis which elaborates a confusion matrix containing categorical similarities between the observed values and the predicted ones All the previous statistical elaborations were then implemented in a excel spreadsheet to elaborate graphs and tables.

RESULTS AND DISCUSSIONBefore investigating the probable relation between resistivity signal and soil properties it was

evaluated the possibility of reducing the costs of the ARP survey. With that aim the spatialization of ER50 values for different densities of the geoelectrical survey was compared. In particular, three different swaths width were investigated: actual 6 m, 12 m and 24 m, relative to 22, 10 and 5 passages, respectively (Fig. 1).

22 swaths 10 swaths 5 swaths

ER50 (Ohm.m)

0 - 13.56 13.56 - 16.22 16.22 - 19.62 19.62 - 60 No Data

Figure 1. IDW 2-3 interpolation of ER50 values for 22, 10 and 5 swaths. (Scale 1:5,000).

The truthfulness of the ER50 values calculated for different ER survey densities was evaluated comparing the predicted values with those interpolated starting from the more dense survey (22 rows). In particular, for the K analysis a pixel by pixel comparison was applied; in such a way the evaluation was extended over the whole area starting from the ER50 values transformed into four equal dimensional classes. Tab. 1 illustrates for each density survey, expressed in terms of both number of measurements per ha and of swaths width, the statistics of ER50 values calculated over the whole area, the accuracy parameters (Landis and Koch ,1977) for ER50 prediction.

Table 1. Summary of ER50 statistics for different resistivity survey densities. Swaths number

22 10 5 Width swath (m) 6 12 24 ER sample points per ha 667 417 276

ER50 sampling points statistics: Mean (Ohm.m) 17.34 17.5 17.78 Standard deviation (Ohm.m) 6.74 7.01 7.72 Overall accuracy of ER50 prediction over the whole area 74% 63% Theta value 0.25 0.25 K value 0.65 0.51 Agreement classification Substantial Moderate

It is noteworthy, that only 10 rows,

corresponding to a reduction of almost 40% of the sampling points (from 667 to 417 per ha), may provide a reliable accuracy in ER50 prediction equivalent to a substantial agreement, compared to 5 rows. Actually, the further swath width enlarging to 24 m reduces significantly the ER50 predictability becoming characterized by a moderate agreement.

As rule, the resistivity maps are employed as surrogate information of soil variability to selecting the soil sampling localization. With the aim to obtain a unique ER50 map representing the resistivity variability of the study area, the mean value of ER50 grids among 22 and 10 swaths ER50(22-10 swaths) was calculated to identify three different densities of soil survey (25, 12 and 6 points), which in turn had to be compared with the denser scheme (49 samples) (Fig. 2).

Actually, ER50 grids provided by only 5 swaths were excluded from the successively elaboration because of its moderate agreement respect to 22 swaths results. For all the soil survey intensities, the procedure of sample localization/identification consisted in

selecting points uniformly distributed over the area and able to explain the whole ER variability.

Summary statistics of ER50(22-10swaths) values for different soil survey densities.

Sampling points (n) (Ohm.m) 49 25 12 X 6 Minimum 8.82 8.82 8.82 8.82 Maximum 31.59 31.59 31.59 26.93 Mean 17.17 17.99 16.95 16.1Standard deviation 4.97 6.28 5.60 6.24

3

Figure 2. ER50(22-10swaths) grid values, localization of the sampling points (Scale 1:3,500) and ER50(22-

10swaths) statistics for different soil survey density.

Endostagnic Cambisols (Calcaric, Sodic) Haplic Cambisols

(Calcaric)

Haplic Cambisols (Eutric)

4 - 122



Zonal Statistics, the mean value of ER50 grids within each buffer was associated to the relative soil sample information. The resolution of the raster layer was 3 meter. Regression analysis was employed to correlate ER50 information to soil physical and chemical data.

In addition, to evaluate the possibility of reducing the ARP survey cost, the accuracy of predicting ER50 values for the swath widths of 6, 12 and 24 m was assessed. Similar approach was employed to evaluate the opportunity of reducing soil samplings by assessing soil properties predictability for decreasing sampling closeness. Those localizations were in turn selected according to the observed ER50 variability for the diverse densities of the ARP survey.

Such accuracy analysis was carried out in ARC/VIEW GIS environment by means of the ArcView tool Kappa analysis which elaborates a confusion matrix containing categorical similarities between the observed values and the predicted ones All the previous statistical elaborations were then implemented in a excel spreadsheet to elaborate graphs and tables.

RESULTS AND DISCUSSIONBefore investigating the probable relation between resistivity signal and soil properties it was

evaluated the possibility of reducing the costs of the ARP survey. With that aim the spatialization of ER50 values for different densities of the geoelectrical survey was compared. In particular, three different swaths width were investigated: actual 6 m, 12 m and 24 m, relative to 22, 10 and 5 passages, respectively (Fig. 1).

22 swaths 10 swaths 5 swaths

ER50 (Ohm.m)

0 - 13.56 13.56 - 16.22 16.22 - 19.62 19.62 - 60 No Data

Figure 1. IDW 2-3 interpolation of ER50 values for 22, 10 and 5 swaths. (Scale 1:5,000).

The truthfulness of the ER50 values calculated for different ER survey densities was evaluated comparing the predicted values with those interpolated starting from the more dense survey (22 rows). In particular, for the K analysis a pixel by pixel comparison was applied; in such a way the evaluation was extended over the whole area starting from the ER50 values transformed into four equal dimensional classes. Tab. 1 illustrates for each density survey, expressed in terms of both number of measurements per ha and of swaths width, the statistics of ER50 values calculated over the whole area, the accuracy parameters (Landis and Koch ,1977) for ER50 prediction.

Table 1. Summary of ER50 statistics for different resistivity survey densities. Swaths number

22 10 5 Width swath (m) 6 12 24 ER sample points per ha 667 417 276

ER50 sampling points statistics: Mean (Ohm.m) 17.34 17.5 17.78 Standard deviation (Ohm.m) 6.74 7.01 7.72 Overall accuracy of ER50 prediction over the whole area 74% 63% Theta value 0.25 0.25 K value 0.65 0.51 Agreement classification Substantial Moderate

It is noteworthy, that only 10 rows,

corresponding to a reduction of almost 40% of the sampling points (from 667 to 417 per ha), may provide a reliable accuracy in ER50 prediction equivalent to a substantial agreement, compared to 5 rows. Actually, the further swath width enlarging to 24 m reduces significantly the ER50 predictability becoming characterized by a moderate agreement.

As rule, the resistivity maps are employed as surrogate information of soil variability to selecting the soil sampling localization. With the aim to obtain a unique ER50 map representing the resistivity variability of the study area, the mean value of ER50 grids among 22 and 10 swaths ER50(22-10 swaths) was calculated to identify three different densities of soil survey (25, 12 and 6 points), which in turn had to be compared with the denser scheme (49 samples) (Fig. 2).

Actually, ER50 grids provided by only 5 swaths were excluded from the successively elaboration because of its moderate agreement respect to 22 swaths results. For all the soil survey intensities, the procedure of sample localization/identification consisted in

selecting points uniformly distributed over the area and able to explain the whole ER variability.

Summary statistics of ER50(22-10swaths) values for different soil survey densities.

Sampling points (n) (Ohm.m) 49 25 12 X 6 Minimum 8.82 8.82 8.82 8.82 Maximum 31.59 31.59 31.59 26.93 Mean 17.17 17.99 16.95 16.1Standard deviation 4.97 6.28 5.60 6.24

3

Figure 2. ER50(22-10swaths) grid values, localization of the sampling points (Scale 1:3,500) and ER50(22-

10swaths) statistics for different soil survey density.

Endostagnic Cambisols (Calcaric, Sodic) Haplic Cambisols

(Calcaric)

Haplic Cambisols (Eutric)

4 - 123



For that purpose, a buffer of 3 m radius was created around all the soil sampling points and the mean value of ER50(22-10 swaths) grids was calculated within each buffer. In such a way the interpolation effect was averaged and the resistivity attribution to each sampling point became more reliable respect to soil properties distribution.

The results illustrated in Fig. 2 outline that each selection guarantees the whole ER variability along with the uniformly spatial distribution of the diverse soil selections.

In order to assess the reliability provided by the different soil survey densities in terms of characterization of soil properties variability over the study area, the possibility of discovering a relation between ER50 values and some of soil properties was investigated. For that purpose, once again, the mean values of ER50 grids provided by the different ARP survey densities was averaged within each soil sampling buffer and related to the soil properties.

Among all the analyzed soil parameters only the clay content was always linked to the ER50 values (i.e., separately provided by 22 and 10 swaths) with an high level of significance(p<0.001) (Tab. 2); therefore the clay content was employed to test/compare the performances provided by different soil survey densities.

All the relations between ER50 and clay for the diverse densities of soil and ARP surveys assumed the exponential form. Here after, Tab. 3 illustrates the parameters of the regressions employed to assess the clay content starting from the resistivity signal, for different soil and ARP survey densities.

Table 2. Correlation coefficient among soil parameters versus the mean values of ER50 for two ARP survey densities and for all the soil samples

(49). ER50 (22ARPswaths) ER50 (10ARPswaths) W -0.270 N.S. -0.205 N.S. E.C.(1:5) -0.408 ** -0.498 *** Clay -0.750 *** -0.818 *** Total Sand

The cells depicted in grey colour represent the comparison term respect to all the other combinations between soil survey points and ARP swaths. Despite the high value of the determination coefficient, all the regressions involving solely 6 samples are less significant because of the few degrees of freedom (df).

0.446 ** 0.565 *** Fine Sand 0.370 ** 0.487 *** Coarse Sand 0.154 N.S. 0.124 N.S. Total Silt 0.059 N.S. -0.038 N.S. Fine Silt -0.04 N.S. -0.149 N.S. Coarse Silt 0.244 N.S. 0.283 * *** Significant at 0.001 probability level; ** Significant at 0.01;* Significant at 0.05 probability level; N.S. Non significant.

Table 3. Parameters of the regressions.

22 ARP swaths 10 ARP swaths Soil sample number R2 df Significance level R2 df Significance level

49 0.610 47 *** 25 0.670 23 *** 0.755 23 *** 12 0.799 10 *** 0.902 10 *** 6 0.828 4 * 0.906 4 **

In order to evaluate the consistency of the clay content assessment over the whole study area

only the more significant regressions (***) were implemented in ARC View GIS environment, starting from ER50 values for different ARP survey densities. In such a way it was possible to

compare the results provided by 49 samples-22 ARP swaths on the one hand, with all the other combinations of soil samples number and ARP swaths and therefore evaluate the corresponding clay predictability. Once again clay values were transformed into categorical classes being employed into the confusion matrix for accuracy analysis (Tab. 4).

Table 4. Results of the confusion matrix for the clay accuracy determination.

22 ARP swaths 10 ARP swaths soil

samples (n) Overall

accuracy theta value K value Agreement

class overall


class 25 0.92 0.37 0.87 Almost perfect 0.76 0.37 0.62 Substantial 12 0.80 0.33 0.70 Substantial 0.65 0.32 0.49 moderate

The predictability of clay content ranged between 0.87 and 0.49, 22 ARP swaths provided

always excellent accuracy for both the analyzed soil sample sizes. Conversely, the more spaced ARP survey guaranteed a substantial accuracy only with 25 soil samples.

CONCLUSIONS For optimizing the use of ARP technology to support soil survey for precision viticulture two

possible strategies were indicated. With the highest geoelectrical survey density the soil samples number may be reduced to twelve, at the most, for assuring at least a substantial accuracy in clay prediction. Conversely a combined reduction of both costs (ARP and soil survey), able to assure the same clay accuracy, may be provided by 10 ARP swaths with 25 soil samples for 3.5 ha, equivalent to less than 3 swaths and 7.5 samples by ha, respectively.

BIBLIOGRAPHY

Bramley R.G.V. and Proffitt A.P., 1999. Managing variability in viticultural production. Grapegrower and Winemaker. July 1999, 427:11-16.

Dabas M., Tabbagh J. and Boisgontier D., 2001. Multi-depth continuous electrical profiling (MuCep) for characterization of in-field variability. In: G. Grenier and S. Blackmore (eds.). Proc. Third European Conference on Precision Agriculture. Montpellier, France, 361-366.

Dabas M. and Tabbagh J., 2003. Comparison of EMI and DC methods for soil mapping in Precision Agriculture. In: J.V. Stafford and A. Werner (eds.). Proc. Fourth European Conference on Precision Agriculture. Berlin, Germany.

FAO, IUSS, ISRIC, 2006. World Reference Base for soil resource. World Soil Resource Report n.103, FAO, Rome, Italy.

Farahani H.J. and Flynn R.L., 2007. Map Quality and Zone Delineation as affected by Width of Parallel Swaths of Mobile Agricultural Sensors. Biosystems Engineering, 96 (2):151-159.

Landis J.R. and Koch G.G., 1977. The measurement of observer agreement for categorical data. Biometrics, 33: 159-174.

Luck E. and Eisenreich M., 2001. Electrical Conductivity Mapping For Precision Agriculture. In: G. Grenier and S. Blackmore (eds.). Proc. Third European Conference on Precision Agriculture. Montpellier, France, 425-429.

Taylor J.A., 2004. Precision Viticulture and Digital Terroir. PhD Thesis. The University of Sydney.

4 - 124



For that purpose, a buffer of 3 m radius was created around all the soil sampling points and the mean value of ER50(22-10 swaths) grids was calculated within each buffer. In such a way the interpolation effect was averaged and the resistivity attribution to each sampling point became more reliable respect to soil properties distribution.

The results illustrated in Fig. 2 outline that each selection guarantees the whole ER variability along with the uniformly spatial distribution of the diverse soil selections.

In order to assess the reliability provided by the different soil survey densities in terms of characterization of soil properties variability over the study area, the possibility of discovering a relation between ER50 values and some of soil properties was investigated. For that purpose, once again, the mean values of ER50 grids provided by the different ARP survey densities was averaged within each soil sampling buffer and related to the soil properties.

Among all the analyzed soil parameters only the clay content was always linked to the ER50 values (i.e., separately provided by 22 and 10 swaths) with an high level of significance(p<0.001) (Tab. 2); therefore the clay content was employed to test/compare the performances provided by different soil survey densities.

All the relations between ER50 and clay for the diverse densities of soil and ARP surveys assumed the exponential form. Here after, Tab. 3 illustrates the parameters of the regressions employed to assess the clay content starting from the resistivity signal, for different soil and ARP survey densities.

Table 2. Correlation coefficient among soil parameters versus the mean values of ER50 for two ARP survey densities and for all the soil samples

(49). ER50 (22ARPswaths) ER50 (10ARPswaths) W -0.270 N.S. -0.205 N.S. E.C.(1:5) -0.408 ** -0.498 *** Clay -0.750 *** -0.818 *** Total Sand

The cells depicted in grey colour represent the comparison term respect to all the other combinations between soil survey points and ARP swaths. Despite the high value of the determination coefficient, all the regressions involving solely 6 samples are less significant because of the few degrees of freedom (df).

0.446 ** 0.565 *** Fine Sand 0.370 ** 0.487 *** Coarse Sand 0.154 N.S. 0.124 N.S. Total Silt 0.059 N.S. -0.038 N.S. Fine Silt -0.04 N.S. -0.149 N.S. Coarse Silt 0.244 N.S. 0.283 * *** Significant at 0.001 probability level; ** Significant at 0.01;* Significant at 0.05 probability level; N.S. Non significant.

Table 3. Parameters of the regressions.

22 ARP swaths 10 ARP swaths Soil sample number R2 df Significance level R2 df Significance level

49 0.610 47 *** 25 0.670 23 *** 0.755 23 *** 12 0.799 10 *** 0.902 10 *** 6 0.828 4 * 0.906 4 **

In order to evaluate the consistency of the clay content assessment over the whole study area

only the more significant regressions (***) were implemented in ARC View GIS environment, starting from ER50 values for different ARP survey densities. In such a way it was possible to

compare the results provided by 49 samples-22 ARP swaths on the one hand, with all the other combinations of soil samples number and ARP swaths and therefore evaluate the corresponding clay predictability. Once again clay values were transformed into categorical classes being employed into the confusion matrix for accuracy analysis (Tab. 4).

Table 4. Results of the confusion matrix for the clay accuracy determination.

22 ARP swaths 10 ARP swaths soil

samples (n) Overall


class overall


class 25 0.92 0.37 0.87 Almost perfect 0.76 0.37 0.62 Substantial 12 0.80 0.33 0.70 Substantial 0.65 0.32 0.49 moderate

The predictability of clay content ranged between 0.87 and 0.49, 22 ARP swaths provided

always excellent accuracy for both the analyzed soil sample sizes. Conversely, the more spaced ARP survey guaranteed a substantial accuracy only with 25 soil samples.

CONCLUSIONS For optimizing the use of ARP technology to support soil survey for precision viticulture two

possible strategies were indicated. With the highest geoelectrical survey density the soil samples number may be reduced to twelve, at the most, for assuring at least a substantial accuracy in clay prediction. Conversely a combined reduction of both costs (ARP and soil survey), able to assure the same clay accuracy, may be provided by 10 ARP swaths with 25 soil samples for 3.5 ha, equivalent to less than 3 swaths and 7.5 samples by ha, respectively.

BIBLIOGRAPHYBramley R.G.V. and Proffitt A.P., 1999. Managing variability in viticultural production.

Grapegrower and Winemaker. July 1999, 427:11-16. Dabas M., Tabbagh J. and Boisgontier D., 2001. Multi-depth continuous electrical profiling

(MuCep) for characterization of in-field variability. In: G. Grenier and S. Blackmore (eds.). Proc. Third European Conference on Precision Agriculture. Montpellier, France, 361-366.

Dabas M. and Tabbagh J., 2003. Comparison of EMI and DC methods for soil mapping in Precision Agriculture. In: J.V. Stafford and A. Werner (eds.). Proc. Fourth European Conference on Precision Agriculture. Berlin, Germany.

FAO, IUSS, ISRIC, 2006. World Reference Base for soil resource. World Soil Resource Report n.103, FAO, Rome, Italy.

Farahani H.J. and Flynn R.L., 2007. Map Quality and Zone Delineation as affected by Width of Parallel Swaths of Mobile Agricultural Sensors. Biosystems Engineering, 96 (2):151-159.

Landis J.R. and Koch G.G., 1977. The measurement of observer agreement for categorical data. Biometrics, 33: 159-174.

Luck E. and Eisenreich M., 2001. Electrical Conductivity Mapping For Precision Agriculture. In: G. Grenier and S. Blackmore (eds.). Proc. Third European Conference on Precision Agriculture. Montpellier, France, 425-429.

Taylor J.A., 2004. Precision Viticulture and Digital Terroir. PhD Thesis. The University of Sydney.

4 - 125



INFLUENCE OF SOIL TYPE ON JUICE QUALITY IN A VINEYARD FROM DO CA RIOJA

Unamunzaga, O.1; Castellón, A.1; G. Besga1, Gallejones, P.2; , Usón, A.3 and Aizpurua, A.1

1 Neiker-Tecnalia. Basque Institute for Agrarian Research and Development; 48.160 Derio, Spain 2 BC3 Basque Research Centre for the Climate Change. C/ GranVía, Bilbao, Spain

3Agricultural and Chemical Engineering School; University of Zaragoza Huesca, Spain

[email protected]

ABSTRACT

Soil plays an important role in wine quality, especially its water holding capacity because it affects the balance between vigour and grape yield. The aim of this work was to study the influence of different soil types on the must quality in a vineyard at DO Ca Rioja. The study was carried out during 2006 and 2007 in a vineyard of eight hectares, located in Oyón in Northern Spain. Four soil types were established according to topography and parent material: deposition (deeper than 110 cm and irregular distribution of organic matter in depth), calcareous red argillite (depth of 85-100 cm, with a heavy clay layer with reddish colour at 85-100 cm), calcareous lutite (depth of 50-100 cm) and finally sandstone (depth of 25-80 cm, and high sand content in depth). Grape samples were collected at 190 grapevines distributed through the whole vineyard for analysing , potential alcohol, total tartaric acid, pH, and K, and anthocyanins concentrations and polyphenols and colour indexes. The influence of soil type on juice quality varied according to the year. In 2006, in the soils with the lower water content (Sandstones) the potential alcohol was the highest (12.92 º), while in 2007, the Red argillite soil (greater water availability) got the greatest potential alcohol (13.72 º). The highest acidity was obtained in Depression soil (5.51 g L-1) and was higher in 2007 (5.48 g L-

1) than in 2006 (5.07 g L-1). Potassium juice concentration (3068 mg L-1) was higher in the Red argillite soil type due to its higher soil K content, and this caused also the higher pH (3.48) shown in this soil. The anthocyanins content, and polyphenols and colour indexes reached higher values in the Sandstone soil (803 mg L-1, 64 and 24 respectively).

KEYWORDSTerroir, Potential alcohol, poliphenols, colour index, anthocyanins, acidity

INTRODUCTION The aim of modern winemaking is to produce wines of high quality and tipicity that can compete in an increasingly broad and competitive market (Ubalde et al., 2007). This quality is closely related to the specific soil, climate, agronomical practices and training system conditions, which in turn are related to the cultivar. The concept of "terroir" is often used to describe this relationship (Deloire et al., 2004) and is usually defined as the ecosystem interaction taking place in a given area, including climate, soil, variety and vineyard management (Seguin, 1988). Thus, soil plays an important role in wine quality, especially its physical properties such as drainage, depth and texture which derive in a good soil water holding capacity. Soil water content is a main factor for the development of vineyards and wine composition. Increased water availability can increase yield (Williams and Matthews, 1990), while it is considered that a moderate stress as a result of reduced water availability

improves quality. The reduced vegetation caused by the hydric stress results in a better bunch exposure to light and also in a smaller fruit size which end in an improvement of the grape quality (McCarthy et al., 2000). The objective of this work was to study the influence of the soil type on the juice quality in a vineyard from DO Ca Rioja.

MATERIAL AND METHODS The study was conducted during 2006 and 2007 on an eight hectare vineyard called "Costanillas", owned by Zuazo Gastón Winery. The vineyard was located in Oyón (Northern Spain) in the Denominación de Origen Calificada (DO Ca) Rioja, and the vines were ” Tempranillo” (Vitis Vinifera L.) trained in a double cordon system. The soil surface is periodically cultivated to limit weed growth. The soil of the vineyard is calcareous (average carbonate content of 155 g kg-1 from 0 to 30 cm), with a high pH (8.6). Four soil types were identified thanks to a soil survey based on the description and analysis of 12 soil pits and 27 soil observations made by a hand auger. The soils were classified according to their depth, texture, organic matter vertical distribution and parent material: 1 - Depression: Soil depth greater than 110 cm, with a clay content of 250-310 g kg-1 at 70-100 cm. The main feature of this soil is the irregular distribution of organic matter in depth, the content drops at 62-100 cm, and increases again from this depth. This change in the downward trend is due to the erosion processes occurring in the vineyard, since this land is located in a low area where two slopes converge and the eroded soil from both of them accumulates there. 2 – Calcareous red argillite: Soil with a depth of 85-100 cm and a layer of calcareous argillites at 60-74 cm depth. This layer is characterized by a high clay content (450-500 g kg-1) and soil K (72-76 mg kg-1 ) and Mg (2,7-4,1 mg kg-1) contents higher than in the other soil types at the same depth. 3 – Calcareous lutite. Soil depth of 50-100 cm, and clay content in depth of 270-380 g kg-1. 4 - Sandstone. Soil depth is between 25 and 80 cm. It is characterized by a high sand content in depth (about 300 g kg-1) and the lower clay content of the vineyard (230 g kg-1). Climatology. The average annual rainfall was 399 mm and average temperature 13.5 º C according to the meteorological station Agoncillo, close to the plot. The climatic conditions of the years 2006 and 2007 are shown in Table 1. The plot has a drip irrigation system with emitters with a flow of 2.5 L h-1 and a distance between emitters of 0.7 m. In the year 2006 the vineyard was watered and in 2007, irrigation was applied at 29 and 30 July with a total dose of 57 mm. A sampling mesh of 24x14.4 m was designed, marking 190 vines. Grape samples were taken from these vines to measure the following quality parameters: - Quality parameters related to the pulp. The yield of each vine was squeezed and afterwards analyzed. Potential alcohol (PA) was measured with a refractometer, total tartaric acidity and pH by automatic potentiometry. - Juice properties related to the skin. The yield of each vine was weighted and small fragments of every harvested bunch were taken. These fragments were cut at the top, middle and bottom of the bunch. Then all the berries were separated and 100 of them were weighted, and another 200 were separated and blended in a Mixer for two minutes obtaining a slurry. Potassium and anthocyanin concentration, and polyphenols and colour indexes were measured in this slurry. Potassium was measured by flame atomic absorption, the colour index by spectrophotometry (420 +520 +620 nm), the total polyphenol index by spectrophotometry at 280 nm and anthocyanins using the method of bleaching with sodium bisulfite (quote).

4 - 126



INFLUENCE OF SOIL TYPE ON JUICE QUALITY IN A VINEYARD FROM DO CA RIOJA

Unamunzaga, O.1; Castellón, A.1; G. Besga1, Gallejones, P.2; , Usón, A.3 and Aizpurua, A.1

1 Neiker-Tecnalia. Basque Institute for Agrarian Research and Development; 48.160 Derio, Spain 2 BC3 Basque Research Centre for the Climate Change. C/ GranVía, Bilbao, Spain

3Agricultural and Chemical Engineering School; University of Zaragoza Huesca, Spain

[email protected]

ABSTRACT

Soil plays an important role in wine quality, especially its water holding capacity because it affects the balance between vigour and grape yield. The aim of this work was to study the influence of different soil types on the must quality in a vineyard at DO Ca Rioja. The study was carried out during 2006 and 2007 in a vineyard of eight hectares, located in Oyón in Northern Spain. Four soil types were established according to topography and parent material: deposition (deeper than 110 cm and irregular distribution of organic matter in depth), calcareous red argillite (depth of 85-100 cm, with a heavy clay layer with reddish colour at 85-100 cm), calcareous lutite (depth of 50-100 cm) and finally sandstone (depth of 25-80 cm, and high sand content in depth). Grape samples were collected at 190 grapevines distributed through the whole vineyard for analysing , potential alcohol, total tartaric acid, pH, and K, and anthocyanins concentrations and polyphenols and colour indexes. The influence of soil type on juice quality varied according to the year. In 2006, in the soils with the lower water content (Sandstones) the potential alcohol was the highest (12.92 º), while in 2007, the Red argillite soil (greater water availability) got the greatest potential alcohol (13.72 º). The highest acidity was obtained in Depression soil (5.51 g L-1) and was higher in 2007 (5.48 g L-

1) than in 2006 (5.07 g L-1). Potassium juice concentration (3068 mg L-1) was higher in the Red argillite soil type due to its higher soil K content, and this caused also the higher pH (3.48) shown in this soil. The anthocyanins content, and polyphenols and colour indexes reached higher values in the Sandstone soil (803 mg L-1, 64 and 24 respectively).

KEYWORDSTerroir, Potential alcohol, poliphenols, colour index, anthocyanins, acidity

INTRODUCTION The aim of modern winemaking is to produce wines of high quality and tipicity that can compete in an increasingly broad and competitive market (Ubalde et al., 2007). This quality is closely related to the specific soil, climate, agronomical practices and training system conditions, which in turn are related to the cultivar. The concept of "terroir" is often used to describe this relationship (Deloire et al., 2004) and is usually defined as the ecosystem interaction taking place in a given area, including climate, soil, variety and vineyard management (Seguin, 1988). Thus, soil plays an important role in wine quality, especially its physical properties such as drainage, depth and texture which derive in a good soil water holding capacity. Soil water content is a main factor for the development of vineyards and wine composition. Increased water availability can increase yield (Williams and Matthews, 1990), while it is considered that a moderate stress as a result of reduced water availability

improves quality. The reduced vegetation caused by the hydric stress results in a better bunch exposure to light and also in a smaller fruit size which end in an improvement of the grape quality (McCarthy et al., 2000). The objective of this work was to study the influence of the soil type on the juice quality in a vineyard from DO Ca Rioja.

MATERIAL AND METHODS The study was conducted during 2006 and 2007 on an eight hectare vineyard called "Costanillas", owned by Zuazo Gastón Winery. The vineyard was located in Oyón (Northern Spain) in the Denominación de Origen Calificada (DO Ca) Rioja, and the vines were ” Tempranillo” (Vitis Vinifera L.) trained in a double cordon system. The soil surface is periodically cultivated to limit weed growth. The soil of the vineyard is calcareous (average carbonate content of 155 g kg-1 from 0 to 30 cm), with a high pH (8.6). Four soil types were identified thanks to a soil survey based on the description and analysis of 12 soil pits and 27 soil observations made by a hand auger. The soils were classified according to their depth, texture, organic matter vertical distribution and parent material: 1 - Depression: Soil depth greater than 110 cm, with a clay content of 250-310 g kg-1 at 70-100 cm. The main feature of this soil is the irregular distribution of organic matter in depth, the content drops at 62-100 cm, and increases again from this depth. This change in the downward trend is due to the erosion processes occurring in the vineyard, since this land is located in a low area where two slopes converge and the eroded soil from both of them accumulates there. 2 – Calcareous red argillite: Soil with a depth of 85-100 cm and a layer of calcareous argillites at 60-74 cm depth. This layer is characterized by a high clay content (450-500 g kg-1) and soil K (72-76 mg kg-1 ) and Mg (2,7-4,1 mg kg-1) contents higher than in the other soil types at the same depth. 3 – Calcareous lutite. Soil depth of 50-100 cm, and clay content in depth of 270-380 g kg-1. 4 - Sandstone. Soil depth is between 25 and 80 cm. It is characterized by a high sand content in depth (about 300 g kg-1) and the lower clay content of the vineyard (230 g kg-1). Climatology. The average annual rainfall was 399 mm and average temperature 13.5 º C according to the meteorological station Agoncillo, close to the plot. The climatic conditions of the years 2006 and 2007 are shown in Table 1. The plot has a drip irrigation system with emitters with a flow of 2.5 L h-1 and a distance between emitters of 0.7 m. In the year 2006 the vineyard was watered and in 2007, irrigation was applied at 29 and 30 July with a total dose of 57 mm. A sampling mesh of 24x14.4 m was designed, marking 190 vines. Grape samples were taken from these vines to measure the following quality parameters: - Quality parameters related to the pulp. The yield of each vine was squeezed and afterwards analyzed. Potential alcohol (PA) was measured with a refractometer, total tartaric acidity and pH by automatic potentiometry. - Juice properties related to the skin. The yield of each vine was weighted and small fragments of every harvested bunch were taken. These fragments were cut at the top, middle and bottom of the bunch. Then all the berries were separated and 100 of them were weighted, and another 200 were separated and blended in a Mixer for two minutes obtaining a slurry. Potassium and anthocyanin concentration, and polyphenols and colour indexes were measured in this slurry. Potassium was measured by flame atomic absorption, the colour index by spectrophotometry (420 +520 +620 nm), the total polyphenol index by spectrophotometry at 280 nm and anthocyanins using the method of bleaching with sodium bisulfite (quote).

4 - 127



Statistical analysis. Difference in quality properties caused by soil types and years were studied using the analysis of variance, and mean separation was made by Duncan's test. SAS Statistical program was used (SAS, 1998).

RESULTS AND DISCUSSION A significant interaction was observed between "year" and "soil type" and therefore the behaviour of the potential alcohol was studied within each soil type and for each of the years (Tab. 2). Based on the behaviour of the potential alcohol in the different soil types in the two years two groups were made, which mainly differ in soil water holding capacity. On the one hand the Depression and Red argillite soils, with a high water holding capacity and on the other side Sandstone and Calcareous lutite soils with lower water retention capacity. In 2006, the Sandstone and Calcareous lutite soils presented the highest potential alcohol (12.92 ° and 12.88 °, respectively) (Tab. 2), while in 2007 the opposite occurred, i.e. Red argillite and Depression soils showed the highest PA. These results are related to the rainfall distribution in the two study years. In 2006, heavy rainfall occurred in the early stages of development of the berries, and exceeded in 54 mm the historical average, while the period from veraison to maturity was characterized by being drier (41 mm less rainfall). The synthesis and accumulation of sugars are higher when water availability is low from bloom to veraison (Rhul and Alleweldt, 1985; Rhul, 1988) and there is high water availability during maturation (Smart et al., 1974, Hardie and Considine 1976; Rühl and Allewedt, 1985, García-Escudero 1991, 1994 and Gutierrez-Granda Sipiora 1998, Esteban et al., 1999; Intrigiololo and Castel 2010). In this way we can explain the different behaviour of the potential alcohol Sn the different soil types each year. In 2006, excessive water availability from bloom to veraison prevailed over the relatively dry maturation period. So the soils with lower water retention capacity (Sandstone and Calcareous lutite) showed the higher PA In 2007 the rainfall was high during maturation, therefore the soils with the greater soil water holding capacity (Red argillite and Depression) obtained the highest PA in their musts. Studying the reasons for the favourable or unfavourable effect of soil water at different times of the growing cycle, Van Zyl (1985) and Rhul (1988) suggested that high water availability in the period between bloom and veraison causes an excessive vegetation, which decreased the light received by the bunches. This induces a delay in the maturation so the desired sugar content is not reached at the time of harvest. Regarding the effect of water availability in maturation García-Escudero (1991), Esteban et al., (1999) and Deloire et al., (2004) argue that the plant activity is adversely affected by water deficit in maturation, which results in a decrease in the potential alcohol, because the plant is not able to meet the needs of sugar accumulation by the reduction of photosynthetic activity. Furthermore, Smart et al., (1974), Hardie and Considine (1976), Neja et al. (1977) and Bravdo (1984) note that a severe water deficit in the maturation period is related to an inadequate maturation. Coipel et al. (2006) conducted a study in 15 vineyards located at west of the river Rhone (France) and within “Appelation controllee” Côtes du Rhône (France). Five soil types were established primarily by soil depth, identifying deep soils (deeper than one meter), and shallow soils with a depth of 60 to 80 cm. They found that soil water availability was inversely related to the sugar content of the grapes, with shallower soils showing higher sugar contents. In our case, however, the behaviour of the sugar content in the different soil types depends on the climatic conditions of the year, more precisely the distribution of rainfall plus irrigation. Generally the PA was higher in 2007 except in Sandstone soils, probably because in this year the water availability was higher in the period of maturation, promoting the synthesis and accumulation of sugars.

Significant differences were observed for juice total acidity between different soil types and different years of study (Tab. 3), but no interaction was detected between the two factors. Depression soils showed about 0.5 g L-1 higher total tartaric acidity, compared to the other soils. This difference is mainly due to vines vigour in this soil that induces on one hand a certain delay in the maturation process, and on the other hand, an increase in the shading of the bunches. This is the result of enlarged vegetative mass, which has a positive influence on the total acidity (García-Escudero, 1991). Kliever and Schultz (1964) observed that the greater shading of bunches in vineyards with irrigation, as a result of intense vegetative growth, gives juices with higher acidity, since acid degradation by combustion is lower. Similarly, in 2007 the acidity was higher respect to 2006 and that is probably due to a difference in the vine vigour as in the case of the differences between soil types. Regarding the pH, the only difference was observed in the Red argillite soil in the year 2007, which had a pH significantly higher than the other soils (Tab. 2). The juice K concentration in this type of soil, as shown below, is higher due to higher soil K content. There is a clear relationship between K and pH of the juice, since high levels of K cause the precipitation of tartaric acid to potassium bitartrate (Mpelasoka et al., 2003). Consequently, there is a decrease of tartaric acid levels and in turn an increase in pH (Boulton 1980, Gawel et al., 2000), because this acid is among the organic acids, which exerts more force on the pH. Significant differences were found between the years of study and the soil types regarding to juice K, anthocyanin concentrations and polyphenol and colour indexes, but the interaction was not significant. Usually the same trend was observed for all these properties related to the berry skin, being the highest values at the Sandstone soil(Tab. 3). The only exception was the juice K concentration, which was higher on the Red argillite soil. The higher soil K content in this soil at 60-75 cm had more influence on juice K concentration than other soil properties. Sandstone soils had the lowest water holding capacity due to the shallower depth and the coarser texture. So we could say that low water availability has led to an improvement of the berry skin related properties. These results are consistent with several research studies (Van Leeuwen et al., 2004; Coipel et al., 2006), whose authors argue that the soils with lower soil water holding capacity are the ones which produces better colour properties in the must. The influence of water on these parameters , is related to the influence on the vine vigour, linked with more vegetative growth and therefore a lower lightening of the bunch, which disfavours the synthesis of polyphenols and anthocyanins. Besides, soil water content in the first stages of growth (until approximately mid-July) has a strong influence on berry size (Hardie and Considine, 1976; Matthews and Anderson, 1988;, Jackson and Lombard, 1993; Esteban et al., 1999; Roby et al., 2004; Ollat et al., 2002), with higher berry weight when water availability is greater. The bigger the berry size, the lower the ratio skin:pulp and therefore the concentration of the parameters commented. However, in our case there was not a clear influence of soil type on berry weight (Tab. 2). This is probably due to the difficulty of sampling correctly, and we hypothesize that perhaps the way of taking the sample was not sufficiently representative to reflect differences between different soil types. Subsamples of each vine should have been taken to assess the accuracy of the measurement. In 2007 must K, anthocyanins, and polyphenols concentrations and colour index were higher than in 2006 (Tab. 3). These differences can be related to the difference in berry weight (Tab. 2) found. Berry weight was generally higher in 2006 than in 2007 due to the rainfall from bloom to veraison in 2006, which, clearly has a positive influence on the berry weight. These results agree with Garcia-Escudero (1991), Esteban et al., (1999) Intrigliolo et al., (2010), who found a negative relationship between berry size and must colour, related to the irrigation

4 - 128



Statistical analysis. Difference in quality properties caused by soil types and years were studied using the analysis of variance, and mean separation was made by Duncan's test. SAS Statistical program was used (SAS, 1998).

RESULTS AND DISCUSSION A significant interaction was observed between "year" and "soil type" and therefore the behaviour of the potential alcohol was studied within each soil type and for each of the years (Tab. 2). Based on the behaviour of the potential alcohol in the different soil types in the two years two groups were made, which mainly differ in soil water holding capacity. On the one hand the Depression and Red argillite soils, with a high water holding capacity and on the other side Sandstone and Calcareous lutite soils with lower water retention capacity. In 2006, the Sandstone and Calcareous lutite soils presented the highest potential alcohol (12.92 ° and 12.88 °, respectively) (Tab. 2), while in 2007 the opposite occurred, i.e. Red argillite and Depression soils showed the highest PA. These results are related to the rainfall distribution in the two study years. In 2006, heavy rainfall occurred in the early stages of development of the berries, and exceeded in 54 mm the historical average, while the period from veraison to maturity was characterized by being drier (41 mm less rainfall). The synthesis and accumulation of sugars are higher when water availability is low from bloom to veraison (Rhul and Alleweldt, 1985; Rhul, 1988) and there is high water availability during maturation (Smart et al., 1974, Hardie and Considine 1976; Rühl and Allewedt, 1985, García-Escudero 1991, 1994 and Gutierrez-Granda Sipiora 1998, Esteban et al., 1999; Intrigiololo and Castel 2010). In this way we can explain the different behaviour of the potential alcohol Sn the different soil types each year. In 2006, excessive water availability from bloom to veraison prevailed over the relatively dry maturation period. So the soils with lower water retention capacity (Sandstone and Calcareous lutite) showed the higher PA In 2007 the rainfall was high during maturation, therefore the soils with the greater soil water holding capacity (Red argillite and Depression) obtained the highest PA in their musts. Studying the reasons for the favourable or unfavourable effect of soil water at different times of the growing cycle, Van Zyl (1985) and Rhul (1988) suggested that high water availability in the period between bloom and veraison causes an excessive vegetation, which decreased the light received by the bunches. This induces a delay in the maturation so the desired sugar content is not reached at the time of harvest. Regarding the effect of water availability in maturation García-Escudero (1991), Esteban et al., (1999) and Deloire et al., (2004) argue that the plant activity is adversely affected by water deficit in maturation, which results in a decrease in the potential alcohol, because the plant is not able to meet the needs of sugar accumulation by the reduction of photosynthetic activity. Furthermore, Smart et al., (1974), Hardie and Considine (1976), Neja et al. (1977) and Bravdo (1984) note that a severe water deficit in the maturation period is related to an inadequate maturation. Coipel et al. (2006) conducted a study in 15 vineyards located at west of the river Rhone (France) and within “Appelation controllee” Côtes du Rhône (France). Five soil types were established primarily by soil depth, identifying deep soils (deeper than one meter), and shallow soils with a depth of 60 to 80 cm. They found that soil water availability was inversely related to the sugar content of the grapes, with shallower soils showing higher sugar contents. In our case, however, the behaviour of the sugar content in the different soil types depends on the climatic conditions of the year, more precisely the distribution of rainfall plus irrigation. Generally the PA was higher in 2007 except in Sandstone soils, probably because in this year the water availability was higher in the period of maturation, promoting the synthesis and accumulation of sugars.

Significant differences were observed for juice total acidity between different soil types and different years of study (Tab. 3), but no interaction was detected between the two factors. Depression soils showed about 0.5 g L-1 higher total tartaric acidity, compared to the other soils. This difference is mainly due to vines vigour in this soil that induces on one hand a certain delay in the maturation process, and on the other hand, an increase in the shading of the bunches. This is the result of enlarged vegetative mass, which has a positive influence on the total acidity (García-Escudero, 1991). Kliever and Schultz (1964) observed that the greater shading of bunches in vineyards with irrigation, as a result of intense vegetative growth, gives juices with higher acidity, since acid degradation by combustion is lower. Similarly, in 2007 the acidity was higher respect to 2006 and that is probably due to a difference in the vine vigour as in the case of the differences between soil types. Regarding the pH, the only difference was observed in the Red argillite soil in the year 2007, which had a pH significantly higher than the other soils (Tab. 2). The juice K concentration in this type of soil, as shown below, is higher due to higher soil K content. There is a clear relationship between K and pH of the juice, since high levels of K cause the precipitation of tartaric acid to potassium bitartrate (Mpelasoka et al., 2003). Consequently, there is a decrease of tartaric acid levels and in turn an increase in pH (Boulton 1980, Gawel et al., 2000), because this acid is among the organic acids, which exerts more force on the pH. Significant differences were found between the years of study and the soil types regarding to juice K, anthocyanin concentrations and polyphenol and colour indexes, but the interaction was not significant. Usually the same trend was observed for all these properties related to the berry skin, being the highest values at the Sandstone soil(Tab. 3). The only exception was the juice K concentration, which was higher on the Red argillite soil. The higher soil K content in this soil at 60-75 cm had more influence on juice K concentration than other soil properties. Sandstone soils had the lowest water holding capacity due to the shallower depth and the coarser texture. So we could say that low water availability has led to an improvement of the berry skin related properties. These results are consistent with several research studies (Van Leeuwen et al., 2004; Coipel et al., 2006), whose authors argue that the soils with lower soil water holding capacity are the ones which produces better colour properties in the must. The influence of water on these parameters , is related to the influence on the vine vigour, linked with more vegetative growth and therefore a lower lightening of the bunch, which disfavours the synthesis of polyphenols and anthocyanins. Besides, soil water content in the first stages of growth (until approximately mid-July) has a strong influence on berry size (Hardie and Considine, 1976; Matthews and Anderson, 1988;, Jackson and Lombard, 1993; Esteban et al., 1999; Roby et al., 2004; Ollat et al., 2002), with higher berry weight when water availability is greater. The bigger the berry size, the lower the ratio skin:pulp and therefore the concentration of the parameters commented. However, in our case there was not a clear influence of soil type on berry weight (Tab. 2). This is probably due to the difficulty of sampling correctly, and we hypothesize that perhaps the way of taking the sample was not sufficiently representative to reflect differences between different soil types. Subsamples of each vine should have been taken to assess the accuracy of the measurement. In 2007 must K, anthocyanins, and polyphenols concentrations and colour index were higher than in 2006 (Tab. 3). These differences can be related to the difference in berry weight (Tab. 2) found. Berry weight was generally higher in 2006 than in 2007 due to the rainfall from bloom to veraison in 2006, which, clearly has a positive influence on the berry weight. These results agree with Garcia-Escudero (1991), Esteban et al., (1999) Intrigliolo et al., (2010), who found a negative relationship between berry size and must colour, related to the irrigation

4 - 129



provided in the first growth phase of the berry called herbaceous growth. The water availability at this period increases the berry size, decreasing the ratio skin:pulp.

CONCLUSIONS Soil variability caused by different parent materials and erosive processes in a vineyard influences the qualitative properties of the must. The effects of climate and soil type on the must quality are largely explained by the relationship they have to soil water content. Interaction between soil type and year is observed sometimes, which means that properties such as the potential alcohol behave differently in different soil types under different climatic conditions. The qualitative properties related to the skin such as must anthocyanins, polyphenol and colour indexes, showed the highest values in the soils with the lower water retention capacity (Sandstone soils). Soils with higher soil K content (Calcareous red argillite soils) showed higher juice K concentration.

ACKNOWLEDGMENTS The authors gratefully acknowledge “Bodegas y viñedos Zuazo Gastón” winery where the field work was carried out. This work was financially supported by European Economic Community through Interreg IIIb program and the Department of Agriculture, Fisheries and Food of the Basque Government. We also want to thank the oenological laboratory Casa del Vino from Laguardia for making the analysis.

BIBLIOGRAPHY Boulton, R. 1980. The relationhips between total acidity, titratable acidity and pH in wine. Am J Enol. Vitic. 31:1. 76-80. Bravdo, V., Hepner, Y., Loing, S., Cohen, S., Tabacmen, H. 1984. Effect de l'irrigation de l'alimentation minérale the sur la qualité du mout et des vins provenant des vignobles of Cabernet Sauvignon Carignane et aux rendements soar in Israel. Bull. O.I.V., 643-644, 729-740. Coipel, J., Rodriguez-Lovelle, B., Sipp, C., and C. Van Leeuwen. 2006. Terroir effect, as a result of Environmental stress, depends more on soil depth than on soil type (Vitis vinifera L. cv. Grenache Noir, Côtes Du Rhône, France, 2000). J. Int Sci Vigne Vin. 40 (4):177-185. Deloire, A., Carbonneau, A., Wang, Z.P. and Ojeda, H. 2004. Vine and water: a short review. J. Int Sci Vigne. Vin. 38:1-13. Esteban, M.A., Villanueva, M.J. and Lisarrague, J.R. 1999. Effect of irrigation on changes in berry composition of Tempranillo during Maturation. Sugars, organic acids and mineral elements. Am J Enol. Vitic. 50 (4):418-434. EVE. 1991. Geological Map of the Basque Country. García-Escudero, E. 1991. Influencia de la dosis y del momento de aplicación del riego sobre la producción, desarrollo vegetativo, calidad del mosto y nutrición mineral de la vid (Vitis vinifera L.). Tesis Doctoral. Universidad Politécnica de Madrid. Thesis. Gawel, R., Ewart, A., and Cirami, R. 2000. Effect of rootstock on must and wine composition and sensory properties of the Cabernet Sauvignon grown at Langhorme Creek, South Australia. Australian and New Zealand Wine Industry Journal 15:67-73. Hardie, W.J., and Considine, J.A. 1976. Response of grapes to water deficit stress in particular stages of development. Am J Enol. Vitic., 27: 55-61. Intrigliolo, D.S. and Castel, J.R. Response of grapevine cv. 'Tempranillo' to timing and amount of irrigation: water relations, vine growth, yield and berry and wine composition. Irrig. Scie. 28:113-125.

Jackson, DI., and Lombard, PB. 1993. Environmental and management practices affecting grape composition and wine quality-A review. Am J Enol. Vitic. 44:409-430. Kliewer, W.M. 1964. Influence of environment on metabolism of organic acids and carbohydrates in Vitis vinifera I. Temperature. Plant. Physiol., 39: 869-871. Lambert JJ, Mcelrone A., Battany M., Dahlgren R., and Wolpert, AJ (2008). Influence of soil type and soil solution chemistry Changes in Growth parameters on vine and grape and wine quality in a vineyard central California coast. VIIth International Terroir Congress. ACW, Agroscope Changins-Wädenswill. 1, 38-44. Matthews, M.A., and Anderson, M. 1989. Reproductive development in grape (Vitis vinifera L.) responses to seasonal water deficit. Am J Enol. Vitic. 40: 52-60. Mpelasoka, B., Shachtman, D., Treeby, M.T. and Thomas, M. 2003. A review of potassium nutrition in grapevine with special emphasis on berry accumulation. Aust. J. of Grape and Wine Research 9:157-168. Neja, RA, Wildman, WE, Ayers, RS, Kasimatis, AN 1977. Grapevine response to irrigation and trellis Treatments in the Salinas Valley. Am J Enol. Vitic., 28: 16-26. Ollat, N., Diakou-Verdin, P., Carde, JP, Barrieu, F., Gaudillère, JP, and Moing, A. Grape berry development: A review. J. Int Sci Vigne. Vin. 36 (3) :109-131. Rhul, E.H., Alleweldt, G. 1985. Investigations Into the Influence of time of irrigation on yield and quality of Grapevine. Acta Horticulturae, 171: 457-462. Rhul, E.H. 1988. Expérience avec l 'irrigation Cuvée des raisins en Allemagne Occidentale. The Australian Grapegrower and Winemaker, Australia, April: 99-102. SAS Institute. 1998. SAS / STAT User's Guide, version 8. SAS Institute, Cary, NC. Sipiora, M.J. Granda and Gutierrez M.J. 1998. Effects of Pre-veraison irrigation cut-off and skin contact time on the composition, color, and phenolic content of young Cabernet Sauvignon wines in Spain. Am J Enol. Vitic. 49:2 152-162. Smart, R.E., 1974. Aspects of water relations of the grapevine (Vitis vinifera). Am J Enol., 25: 84-91. Ubalde J.M., X. Sort, Poch R.M. and Porta M. (2007). Influence of climatic factors on grape quality in Conca de Barbera Vineyards (Catalonia, Spain). International Journal of Vine Wine Sciences. 41, No. 1, 33-41. Van Leeuwen, C., P. Friant, Chon, X., Tregoat, O., Koundouros, S. and Dubourdieu, D. 2004. Influence of climate, soil and cultivar on terroir. Am J Enol. Vitic. 55:3, 207-217. Van Leeuwen, C., and Seguin, G. 2006. The concept of terroir in viticulture. J. Wine Research. 17 (1): 1-10. Van Zyl, J.L. 1985. Influence de l'irrigation sur la croissance et la qualité des vignes et raisins of Colomabar. Bull. O.I.V., 648-649, 173-188. Table 1. Precipitation collected at the meteorological station of Agoncillo at different times of the cycle, for the years 2006 and 2007 and the historical mean (years from 1973 to 2007), together with the irrigation applied. Rainfall (mm) 2006 2007 Historical mean Budbreak-Bloom 90 125 88 Bloom-Veraison 131 85 77 Veraison-Harvest 50 43+57 91 Growing cycle 270 254 221 Budbreak-Bloom = 1st April-31st May; Bloom-Veraison: 1st June – 31st July; Veraison-Harvest- 1st August- 12th October; Growing cycle; 1st April- 12th October. Figures in italics are the irrigation dose for each period.

4 - 130



provided in the first growth phase of the berry called herbaceous growth. The water availability at this period increases the berry size, decreasing the ratio skin:pulp.

CONCLUSIONS Soil variability caused by different parent materials and erosive processes in a vineyard influences the qualitative properties of the must. The effects of climate and soil type on the must quality are largely explained by the relationship they have to soil water content. Interaction between soil type and year is observed sometimes, which means that properties such as the potential alcohol behave differently in different soil types under different climatic conditions. The qualitative properties related to the skin such as must anthocyanins, polyphenol and colour indexes, showed the highest values in the soils with the lower water retention capacity (Sandstone soils). Soils with higher soil K content (Calcareous red argillite soils) showed higher juice K concentration.

ACKNOWLEDGMENTS The authors gratefully acknowledge “Bodegas y viñedos Zuazo Gastón” winery where the field work was carried out. This work was financially supported by European Economic Community through Interreg IIIb program and the Department of Agriculture, Fisheries and Food of the Basque Government. We also want to thank the oenological laboratory Casa del Vino from Laguardia for making the analysis.

BIBLIOGRAPHY Boulton, R. 1980. The relationhips between total acidity, titratable acidity and pH in wine. Am J Enol. Vitic. 31:1. 76-80. Bravdo, V., Hepner, Y., Loing, S., Cohen, S., Tabacmen, H. 1984. Effect de l'irrigation de l'alimentation minérale the sur la qualité du mout et des vins provenant des vignobles of Cabernet Sauvignon Carignane et aux rendements soar in Israel. Bull. O.I.V., 643-644, 729-740. Coipel, J., Rodriguez-Lovelle, B., Sipp, C., and C. Van Leeuwen. 2006. Terroir effect, as a result of Environmental stress, depends more on soil depth than on soil type (Vitis vinifera L. cv. Grenache Noir, Côtes Du Rhône, France, 2000). J. Int Sci Vigne Vin. 40 (4):177-185. Deloire, A., Carbonneau, A., Wang, Z.P. and Ojeda, H. 2004. Vine and water: a short review. J. Int Sci Vigne. Vin. 38:1-13. Esteban, M.A., Villanueva, M.J. and Lisarrague, J.R. 1999. Effect of irrigation on changes in berry composition of Tempranillo during Maturation. Sugars, organic acids and mineral elements. Am J Enol. Vitic. 50 (4):418-434. EVE. 1991. Geological Map of the Basque Country. García-Escudero, E. 1991. Influencia de la dosis y del momento de aplicación del riego sobre la producción, desarrollo vegetativo, calidad del mosto y nutrición mineral de la vid (Vitis vinifera L.). Tesis Doctoral. Universidad Politécnica de Madrid. Thesis. Gawel, R., Ewart, A., and Cirami, R. 2000. Effect of rootstock on must and wine composition and sensory properties of the Cabernet Sauvignon grown at Langhorme Creek, South Australia. Australian and New Zealand Wine Industry Journal 15:67-73. Hardie, W.J., and Considine, J.A. 1976. Response of grapes to water deficit stress in particular stages of development. Am J Enol. Vitic., 27: 55-61. Intrigliolo, D.S. and Castel, J.R. Response of grapevine cv. 'Tempranillo' to timing and amount of irrigation: water relations, vine growth, yield and berry and wine composition. Irrig. Scie. 28:113-125.

Jackson, DI., and Lombard, PB. 1993. Environmental and management practices affecting grape composition and wine quality-A review. Am J Enol. Vitic. 44:409-430. Kliewer, W.M. 1964. Influence of environment on metabolism of organic acids and carbohydrates in Vitis vinifera I. Temperature. Plant. Physiol., 39: 869-871. Lambert JJ, Mcelrone A., Battany M., Dahlgren R., and Wolpert, AJ (2008). Influence of soil type and soil solution chemistry Changes in Growth parameters on vine and grape and wine quality in a vineyard central California coast. VIIth International Terroir Congress. ACW, Agroscope Changins-Wädenswill. 1, 38-44. Matthews, M.A., and Anderson, M. 1989. Reproductive development in grape (Vitis vinifera L.) responses to seasonal water deficit. Am J Enol. Vitic. 40: 52-60. Mpelasoka, B., Shachtman, D., Treeby, M.T. and Thomas, M. 2003. A review of potassium nutrition in grapevine with special emphasis on berry accumulation. Aust. J. of Grape and Wine Research 9:157-168. Neja, RA, Wildman, WE, Ayers, RS, Kasimatis, AN 1977. Grapevine response to irrigation and trellis Treatments in the Salinas Valley. Am J Enol. Vitic., 28: 16-26. Ollat, N., Diakou-Verdin, P., Carde, JP, Barrieu, F., Gaudillère, JP, and Moing, A. Grape berry development: A review. J. Int Sci Vigne. Vin. 36 (3) :109-131. Rhul, E.H., Alleweldt, G. 1985. Investigations Into the Influence of time of irrigation on yield and quality of Grapevine. Acta Horticulturae, 171: 457-462. Rhul, E.H. 1988. Expérience avec l 'irrigation Cuvée des raisins en Allemagne Occidentale. The Australian Grapegrower and Winemaker, Australia, April: 99-102. SAS Institute. 1998. SAS / STAT User's Guide, version 8. SAS Institute, Cary, NC. Sipiora, M.J. Granda and Gutierrez M.J. 1998. Effects of Pre-veraison irrigation cut-off and skin contact time on the composition, color, and phenolic content of young Cabernet Sauvignon wines in Spain. Am J Enol. Vitic. 49:2 152-162. Smart, R.E., 1974. Aspects of water relations of the grapevine (Vitis vinifera). Am J Enol., 25: 84-91. Ubalde J.M., X. Sort, Poch R.M. and Porta M. (2007). Influence of climatic factors on grape quality in Conca de Barbera Vineyards (Catalonia, Spain). International Journal of Vine Wine Sciences. 41, No. 1, 33-41. Van Leeuwen, C., P. Friant, Chon, X., Tregoat, O., Koundouros, S. and Dubourdieu, D. 2004. Influence of climate, soil and cultivar on terroir. Am J Enol. Vitic. 55:3, 207-217. Van Leeuwen, C., and Seguin, G. 2006. The concept of terroir in viticulture. J. Wine Research. 17 (1): 1-10. Van Zyl, J.L. 1985. Influence de l'irrigation sur la croissance et la qualité des vignes et raisins of Colomabar. Bull. O.I.V., 648-649, 173-188. Table 1. Precipitation collected at the meteorological station of Agoncillo at different times of the cycle, for the years 2006 and 2007 and the historical mean (years from 1973 to 2007), together with the irrigation applied. Rainfall (mm) 2006 2007 Historical mean Budbreak-Bloom 90 125 88 Bloom-Veraison 131 85 77 Veraison-Harvest 50 43+57 91 Growing cycle 270 254 221 Budbreak-Bloom = 1st April-31st May; Bloom-Veraison: 1st June – 31st July; Veraison-Harvest- 1st August- 12th October; Growing cycle; 1st April- 12th October. Figures in italics are the irrigation dose for each period.

4 - 131



Table 2 Year and soil type based means for potential alcohol, pH and berry weight. Potencial alcohol (%) pH Berry weight (g) 2006 2007 Prob. 2006 2007 Prob. 2006 2007 Prob. Sandston 12,92 a A 12,81 a B *** 3,45 a A 3,48 a B * 2,42 a A 1,91 b A *** Red argilite 12,01 a B 13,72 a A *** 3,50 a A 3,58 a A * 2,02 a B 2,10 a A n.s. Depression 12,45 b AB 13,16 a

AB *** 3,45 a A 3,46 a B * 2,39 a A 2,06 b A ***

Calcareous lutite

12,88 b A 13,45 a AB

*** 3,49 a A 3,49 a B * 2,47 a A 2,05 b A ***

Prob. * * *** *** * n.se.

*** Prob.<0.001; ** Prob.<0.01, * Prob.<0.10. Means with the same capital letter refers to differences inside the same column and means with the same small letters to differences inside a row.

Table 3. Year based and soil type based means for total acidity, K, anthocyanins, poliphenols index, colour index and pruning weight. Año Suelo 2006 2007 Prob Sandstone Red

Argillite Depression Calcareous

lutite Prob

Total acidity acidity (g L-1)

5,07b 5,48a *** 5,09b 5,07b 5,51a 5,14a ***

K (mg l-1), 2495b 2765a *** 2762b 3068a 2484c 2717bc *** Anthocyanins (mg l-

1), 643b 720a * 803a 729ab 734ab 734ab ***

Poliphenols index 50b 56a *** 64a 55b 49b 49b *** Colour index 18b 21a *** 24a 20b 18b 18b *** Pruning weight (kg vine-1).

0,56c 0,76b 0,88a 0,68bc ***

*** Prob.<0.001; ** Prob.<0.01, * Prob.<0.10.

VULNERABILITY OF VINEYARD SOILS TO COMPACTION: THE CASE STUDY OF DOC PIAVE (VENETO REGION, ITALY)

S. Piccolo(1), M. Bertaggia(1), G. Concheri(1), I. Vinci(2) 1 Padua University - Department of Agricultural Biotechnology

Viale dell’Università 16 – 35020 Legnaro (PD) – Italy [email protected] [email protected] [email protected]

2 ARPAV – Regional Agency for Environmental Prevention and Protection – Regional Soil Observatory Via S. Barbara 5/A – 31100 Treviso – Italy

[email protected]

ABSTRACT The objective of this work is to study the vulnerability of vineyard soil to compaction. The process of soil compaction represents one of the eight threats to soil identified by European

Commission. It is important to know which soil is susceptible to compaction in order to be able to apply

proper soil use and cultivation and to prevent real compaction. From this point of view, the evaluation of soil susceptibility to compaction on European level was done.

The DOC Piave area has been chosen for this study because it is one the most important of the north Italy and involves a great variety of soils.

The model used considers as significant factors drainage, surface organic carbon content and texture. It results that soils with low organic carbon content, medium fine or fine and moderately well drained to very poorly drained have high vulnerability to compaction.

A large part of the vineyard soil of the DOC Piave area has at least moderate vulnerability to compaction.

KEYWORD vulnerability – compaction – vineyard – organic carbon – texture – drainage

INTRODUCTION Compaction can be defined as compression from an applied force that rearranges and destroys

aggregates, increasing bulk density and reducing porosity. It produces meaningful changes in structural properties, in soil behaviour, in the hydraulic and thermal conductivity.

It causes a greater mechanical resistance to radical growth and deepening, a reduction of porosity, with consequent conditions of asphyxia. This can slow down the development of the plants, with negative effects on the productivity of the agricultural cultivations and it can reduce water infiltration in the ground.

Soil compaction results from the combination of natural forces and man induced forces. Compaction depends on the used farm machineries and on the soil’s water content at the

moment of passage of the machine. This phenomenon was studied in the DOC Piave area, in the Veneto region (Fig. 1), one of the

most extensive DOC areas of the north Italy (ESAV, 1996) and one of the most national

4 - 132



Table 2 Year and soil type based means for potential alcohol, pH and berry weight. Potencial alcohol (%) pH Berry weight (g) 2006 2007 Prob. 2006 2007 Prob. 2006 2007 Prob. Sandston 12,92 a A 12,81 a B *** 3,45 a A 3,48 a B * 2,42 a A 1,91 b A *** Red argilite 12,01 a B 13,72 a A *** 3,50 a A 3,58 a A * 2,02 a B 2,10 a A n.s. Depression 12,45 b AB 13,16 a

AB *** 3,45 a A 3,46 a B * 2,39 a A 2,06 b A ***

Calcareous lutite

12,88 b A 13,45 a AB

*** 3,49 a A 3,49 a B * 2,47 a A 2,05 b A ***

Prob. * * *** *** * n.se.

*** Prob.<0.001; ** Prob.<0.01, * Prob.<0.10. Means with the same capital letter refers to differences inside the same column and means with the same small letters to differences inside a row.

Table 3. Year based and soil type based means for total acidity, K, anthocyanins, poliphenols index, colour index and pruning weight. Año Suelo 2006 2007 Prob Sandstone Red

Argillite Depression Calcareous

lutite Prob

Total acidity acidity (g L-1)

5,07b 5,48a *** 5,09b 5,07b 5,51a 5,14a ***

K (mg l-1), 2495b 2765a *** 2762b 3068a 2484c 2717bc *** Anthocyanins (mg l-

1), 643b 720a * 803a 729ab 734ab 734ab ***

Poliphenols index 50b 56a *** 64a 55b 49b 49b *** Colour index 18b 21a *** 24a 20b 18b 18b *** Pruning weight (kg vine-1).

0,56c 0,76b 0,88a 0,68bc ***

*** Prob.<0.001; ** Prob.<0.01, * Prob.<0.10.

VULNERABILITY OF VINEYARD SOILS TO COMPACTION: THE CASE STUDY OF DOC PIAVE (VENETO REGION, ITALY)

S. Piccolo(1), M. Bertaggia(1), G. Concheri(1), I. Vinci(2) 1 Padua University - Department of Agricultural Biotechnology

Viale dell’Università 16 – 35020 Legnaro (PD) – Italy [email protected] [email protected] [email protected]

2 ARPAV – Regional Agency for Environmental Prevention and Protection – Regional Soil Observatory Via S. Barbara 5/A – 31100 Treviso – Italy

[email protected]

ABSTRACT The objective of this work is to study the vulnerability of vineyard soil to compaction. The process of soil compaction represents one of the eight threats to soil identified by European

Commission. It is important to know which soil is susceptible to compaction in order to be able to apply

proper soil use and cultivation and to prevent real compaction. From this point of view, the evaluation of soil susceptibility to compaction on European level was done.

The DOC Piave area has been chosen for this study because it is one the most important of the north Italy and involves a great variety of soils.

The model used considers as significant factors drainage, surface organic carbon content and texture. It results that soils with low organic carbon content, medium fine or fine and moderately well drained to very poorly drained have high vulnerability to compaction.

A large part of the vineyard soil of the DOC Piave area has at least moderate vulnerability to compaction.

KEYWORD vulnerability – compaction – vineyard – organic carbon – texture – drainage

INTRODUCTION Compaction can be defined as compression from an applied force that rearranges and destroys

aggregates, increasing bulk density and reducing porosity. It produces meaningful changes in structural properties, in soil behaviour, in the hydraulic and thermal conductivity.

It causes a greater mechanical resistance to radical growth and deepening, a reduction of porosity, with consequent conditions of asphyxia. This can slow down the development of the plants, with negative effects on the productivity of the agricultural cultivations and it can reduce water infiltration in the ground.

Soil compaction results from the combination of natural forces and man induced forces. Compaction depends on the used farm machineries and on the soil’s water content at the

moment of passage of the machine. This phenomenon was studied in the DOC Piave area, in the Veneto region (Fig. 1), one of the

most extensive DOC areas of the north Italy (ESAV, 1996) and one of the most national

4 - 133



productive DOC areas in terms of hectolitres producted (134,228 hl in 2007) (Consortium for Piave Wines Protection). The DOC Piave lies in the provinces of Treviso and Venice but Treviso is predominant with 12,700 ha, of which 4,327 ha of vineyard (website Veneto Region).

Figure 1. The DOC Piave area in the Veneto region.

MATERIALS AND METHODS The basis for this work has been the Soil Map of the Province of Treviso (ARPAV, 2008). This soil map is structured in four hierarchical levels: district, landscape over-unit (soil system),

landscape unit and cartographic unit. Districts are distinguished on the basis of large geographical areas and afferent river basins. Soil systems are identified according to the genetic processes that have carried to the formation of the different surfaces and to the age at which these processes have finished. Soil systems for the plain are differentiated on the basis of the morphology and the texture of the parental material (sand, silt and clay) while for the mountain the dominant factor is the morphology. The soil systems of the DOC Piave area are represented in figure 2.

Landscape units identify different shapes on the land (levee, depression and modal plain). In every landscape unit there are cartographic units, homogeneous areas characterized by the

same soil set. Every cartographic unit has one or two predominant soil type, soil typologic unit (UTS), a soil group with similar features and organization in horizons.

In this study the predominant soil typologic unit in the cartographic unit has been chosen. The UTS, in which land use includes vineyard soil, have been selected. So only the cartographic units with vineyard soils are represented in this work and, as consequence, a wide part of the DOC Piave area isn’t embodied, especially that in province of Venice.

Figure 2. Soil systems in the DOC Piave area.

The model used in this study (Agriculture and Agri-Food Canada, adapted from Martin and Nolin, 1991) considers three soil characteristics: organic carbon (C.O.) content of A horizon, surface layer texture and drainage class (Tab. 1).

The A horizon is the top layer of a soil profile, also called topsoil; it may be darker in colour than deeper layers because of the greater organic material content or lighter because contains less clay or sesquioxides.

Soil vulnerability to compaction is the probability that soil becomes compacted when exposed to compaction risk. It can be nil to low, moderate or high (Tab. 2).

Table 1. Organic carbon content, texture (Agriculture Canada, 1976) and drainage classes C.O. content of A horizon DRAINAGE TEXTURE class of A horizon Low < 1.7% Very rapidly drained 1 Coarse Sand, loamy sand Moderate 1.7 – 4% Well-drained 2 Medium coarse Sandy loam High 4 – 9% Moderately well-drained 3 Medium Loam, silt loam, silt Very high 9 – 17% Imperfectly drained 4 Medium fine Sandy clay loam,

clay loam, silty clay loam Poorly drained 5 Fine Sandy clay, clay,

silty clay Very poorly drained P Peaty ≥ 17% C.O.

4 - 134



productive DOC areas in terms of hectolitres producted (134,228 hl in 2007) (Consortium for Piave Wines Protection). The DOC Piave lies in the provinces of Treviso and Venice but Treviso is predominant with 12,700 ha, of which 4,327 ha of vineyard (website Veneto Region).

Figure 1. The DOC Piave area in the Veneto region.

MATERIALS AND METHODS The basis for this work has been the Soil Map of the Province of Treviso (ARPAV, 2008). This soil map is structured in four hierarchical levels: district, landscape over-unit (soil system),

landscape unit and cartographic unit. Districts are distinguished on the basis of large geographical areas and afferent river basins. Soil systems are identified according to the genetic processes that have carried to the formation of the different surfaces and to the age at which these processes have finished. Soil systems for the plain are differentiated on the basis of the morphology and the texture of the parental material (sand, silt and clay) while for the mountain the dominant factor is the morphology. The soil systems of the DOC Piave area are represented in figure 2.

Landscape units identify different shapes on the land (levee, depression and modal plain). In every landscape unit there are cartographic units, homogeneous areas characterized by the

same soil set. Every cartographic unit has one or two predominant soil type, soil typologic unit (UTS), a soil group with similar features and organization in horizons.

In this study the predominant soil typologic unit in the cartographic unit has been chosen. The UTS, in which land use includes vineyard soil, have been selected. So only the cartographic units with vineyard soils are represented in this work and, as consequence, a wide part of the DOC Piave area isn’t embodied, especially that in province of Venice.

Figure 2. Soil systems in the DOC Piave area.

The model used in this study (Agriculture and Agri-Food Canada, adapted from Martin and Nolin, 1991) considers three soil characteristics: organic carbon (C.O.) content of A horizon, surface layer texture and drainage class (Tab. 1).

The A horizon is the top layer of a soil profile, also called topsoil; it may be darker in colour than deeper layers because of the greater organic material content or lighter because contains less clay or sesquioxides.

Soil vulnerability to compaction is the probability that soil becomes compacted when exposed to compaction risk. It can be nil to low, moderate or high (Tab. 2).

Table 1. Organic carbon content, texture (Agriculture Canada, 1976) and drainage classes C.O. content of A horizon DRAINAGE TEXTURE class of A horizon Low < 1.7% Very rapidly drained 1 Coarse Sand, loamy sand Moderate 1.7 – 4% Well-drained 2 Medium coarse Sandy loam High 4 – 9% Moderately well-drained 3 Medium Loam, silt loam, silt Very high 9 – 17% Imperfectly drained 4 Medium fine Sandy clay loam,

clay loam, silty clay loam Poorly drained 5 Fine Sandy clay, clay,

silty clay Very poorly drained P Peaty ≥ 17% C.O.

4 - 135



Table 2. Soil compaction vulnerability assessment model Drenaggio

Very rapidly drained to well drained

Nil to low

Organic soils

Moderately well drained to imperfectly drained

Peaty Nil to low

Poorly to very poorly drained

Peaty (fibric and mesic) Nil to low

Very poorly drained Peaty (humic) Moderate Mineral soils C.O. content of A horizon

Texture class of A horizon

Low Moderate High Very high

Moderately well to 1 and 2 Nil to low

imperfectly drained 3 Moderate Nil to low Nil to low Nil to low 4 and 5 High Moderate Nil to low Nil to low

Poorly to 1 and 2 Moderate Nil to low Nil to low Nil to lowvery poorly drained 3 High Moderate Nil to low Nil to low 4 and 5 High High Moderate Nil to low

Data concerning soil have been taken from the soil map of the province of Treviso, carried out by Regional Soil Observatory of Treviso. It carefully describes cartographic units with its prevalent UTS. In the explanation of the soil typologic unit, land use, drainage and the horizons with all the characteristic, texture and C.O. content are reported.

RESULTS AND DISCUSSION Applying the table 2, it results that soils with low C.O. content (< 1.7%), fine or medium fine

(clay soils) in the surface horizon, moderately well-drained or imperfectly drained and loam soils with low organic carbon content poorly drained have a high vulnerability to compaction. Whereas soils with moderate held in C.O. (1.7 – 4%), fine, moderately well-drained and loam poorly drained have a moderate vulnerability. Generally grounds with high organic carbon content (> 4%) have vulnerability nil to low.

In the DOC Piave area, the most vulnerable zones are localized in three zone: one south east of Treviso, from Silea to Roncade, one in the centre, in the east of Oderzo, between Ponte di Piave and Salgareda and another one from Gaiarine to Motta di Livenza (Fig. 3).

Highly vulnerable soils are these: soils of the ancient low plain of the Piave river: soils Marteggia and San Fior, silty clay

loam, with mediocre drainage, soils Lutrano and Borin, silty clay and soils Olmi, silty clay loam, imperfectly to poorly drained; soils of the ancient low plain of the Tagliamento river: soils Cinto Caomaggiore, silty

clay, poorly drained; soils of the recent plain of the Monticano and the Meschio rivers: soils Termen, silty

clay, poorly drained; hydromorphic soils of spring lowlands: soils Meolo and Biancade, of the reclaimed

wetlands, silty clay loam, poorly drained. All these soils have a moderated C.O. content in the surface layer.

Figure 3. Vulnerability of vineyard soils to compaction in the DOC Piave area. h i g hm o d e r a ten i l

t o

lo w

4 - 136



Table 2. Soil compaction vulnerability assessment model Drenaggio

Very rapidly drained to well drained

Nil to low

Organic soils

Moderately well drained to imperfectly drained

Peaty Nil to low

Poorly to very poorly drained

Peaty (fibric and mesic) Nil to low

Very poorly drained Peaty (humic) Moderate Mineral soils C.O. content of A horizon

Texture class of A horizon

Low Moderate High Very high

Moderately well to 1 and 2 Nil to low

imperfectly drained 3 Moderate Nil to low Nil to low Nil to low 4 and 5 High Moderate Nil to low Nil to low

Poorly to 1 and 2 Moderate Nil to low Nil to low Nil to lowvery poorly drained 3 High Moderate Nil to low Nil to low 4 and 5 High High Moderate Nil to low

Data concerning soil have been taken from the soil map of the province of Treviso, carried out by Regional Soil Observatory of Treviso. It carefully describes cartographic units with its prevalent UTS. In the explanation of the soil typologic unit, land use, drainage and the horizons with all the characteristic, texture and C.O. content are reported.

RESULTS AND DISCUSSION Applying the table 2, it results that soils with low C.O. content (< 1.7%), fine or medium fine

(clay soils) in the surface horizon, moderately well-drained or imperfectly drained and loam soils with low organic carbon content poorly drained have a high vulnerability to compaction. Whereas soils with moderate held in C.O. (1.7 – 4%), fine, moderately well-drained and loam poorly drained have a moderate vulnerability. Generally grounds with high organic carbon content (> 4%) have vulnerability nil to low.

In the DOC Piave area, the most vulnerable zones are localized in three zone: one south east of Treviso, from Silea to Roncade, one in the centre, in the east of Oderzo, between Ponte di Piave and Salgareda and another one from Gaiarine to Motta di Livenza (Fig. 3).

Highly vulnerable soils are these: soils of the ancient low plain of the Piave river: soils Marteggia and San Fior, silty clay

loam, with mediocre drainage, soils Lutrano and Borin, silty clay and soils Olmi, silty clay loam, imperfectly to poorly drained; soils of the ancient low plain of the Tagliamento river: soils Cinto Caomaggiore, silty

clay, poorly drained; soils of the recent plain of the Monticano and the Meschio rivers: soils Termen, silty

clay, poorly drained; hydromorphic soils of spring lowlands: soils Meolo and Biancade, of the reclaimed

wetlands, silty clay loam, poorly drained. All these soils have a moderated C.O. content in the surface layer.

Figure 3. Vulnerability of vineyard soils to compaction in the DOC Piave area. h i g hm o d e r a ten i l

t o

lo w

4 - 137



CONCLUSIONS There are a lot of vineyard soils that have low C.O. content, fine or medium fine texture in the

surface, moderately well-drained or loam soils with low organic carbon content, poorly drained and so they are very vulnerable to compaction.

In fact, the major part of the vineyard soil of the DOC Piave area (about 44% of the cartographic units) has moderate vulnerability, about 33% high and the rest (23%) nil to low.

This means that nearly the 80% of vineyard soils has at least moderate vulnerability to compaction, so in these soils we have to adopt appropriate tillage techniques and adequate soil management to preserve the land and to maintain the crop yield.

To reduce compaction levels, the best method is increasing organic matter levels: this maximizes the aggregation of soil particles and consequently increases soil stability. In addition, rotation crops provide a variety of root types and patterns in the soil that break up compacted layers.

ACKNOWLEDGMENTS Thanks to CRA-VIT (Research Centre for Agriculture-Viticulture) of Conegliano (TV) for

some data, in particular the PhD Diego Tomasi. The present study is a part of the research activity of PhD in Viticulture, Oenology and

Marketing of Wine Companies (Padua University) with the supervision of the prof. Giuseppe Concheri.

BIBLIOGRAPHYAgriculture Canada, 1976. Glossary of terms in soil science. Canada Department of Agriculture,

Publication 1459. Ottawa: Information Canada. ARPAV, 2008. Carta dei suoli della provincia di Treviso. Castelfranco Veneto (TV). ARPAV, 2005. Carta dei suoli del Veneto. Castelfranco Veneto (TV). Regione del Veneto, ESAV, 1996. I suoli dell’area a DOC del Piave - Provincia di Treviso. Serie

pedologica 2.

Websites Agriculture and Agri-Food Canada: http://www.agr.gc.caVeneto Region: http://www.regione.veneto.it/MondoAgricolo/NewsView.aspx?idNews=1185Veneto Agricoltura: http://www.venetoagricoltura.it/basic.php?ID=1955

RAPPORTI TRA DIVERSE TIPOLOGIE DI TERRENO E RISPOSTEPRODUTTIVE E QUALITATIVE DELLE UVE MERLOT E

CARMENÈRE NELL’AREA DOC PIAVE.Soil as it effects qualitative and quantitative performance of Merlot and

Carmenere grapes in the DOC Piave Area

D. Tomasi 1, P. Marcuzzo 1, A. Garlato 2, F. Gaiotti 1, L. Lovat 11 CRA – VIT : Centro di Ricerca per la Viticoltura, Viale XXVII Aprile 26 31015 Conegliano (TV) – Italia.2 ARPAV – Agenzia Regionale per la Prevenzione e Protezione Ambientale del Veneto– Servizio OsservatorioSuolo, Via Baciocchi 9 - 31033 Castelfranco Veneto (TV) – Italy.

ABSTRACTGiving the important effects of the environmental factors on the vine productivity and grape

quality, a branch of viticulture research has been focusing on the relation between vines andtheir ecosystems for years.

The DOC Piave area, located in the eastern part of the Veneto region, was the object of aspecific zoning study from 2007 to 2009.

The study compared two different types of soils, one located in the Southern part of theDOC Area has clay-loam texture, the other located further Nord has a gravelly texture. Forboth varieties the trellising system was Sylvoz and the vines were grafted on Kober 5bb.Sugar accumulation, pigments amount, total acidity and pH were determined along withvegetative and productive parameters.

The results confirmed that there exist a close relationship between soil and grape quality,but each variety responds in a different way: Merlot had the most interesting quality whengrown clay-loam soils, while a different behaviour was found in Carmenere. The winesensory score confirmed the grape analysis for Merlot, but only partially for Carmenere.

RIASSUNTODa anni la ricerca viticola sta orientando le sue attenzioni verso lo studio della vocazionalità

degli ecosistemi viticoli, perché fulcro della produttività della vite e qualità dei suoi frutti. Dal2007 anche l’area a DOC del Piave, situata nella parte orientale della regione Veneto, èoggetto di uno specifico studio.

Il lavoro ha messo a confronto due diverse tipologie di suolo, uno a tessitura fine (limoso –argilloso) più a sud dell’area DOC Piave e l’altro a tessitura più grossolana (ghiaioso –ciottoloso) nella zona più a nord. Entrambe le varietà coltivate erano allevate a Sylvoz,innestate su Kober 5BB. Lo studio ha verificato nella bacca il contenuto di sostanze coloranti,il contenuto in solidi solubili, dell’acidità totale, del pH oltre ai parametri produttivi evegetativi quali: n° grappoli/vite, produzione uva/vite, peso medio del grappolo e il legno dipotatura.

I risultati ottenuti nel triennio, hanno permesso di evidenziare come le caratteristiche delterreno abbiano influenzato nettamente sia le rese produttive sia la qualità delle uve. Qualitàche per la varietà Merlot è stata superiore nei suoli limoso – argillosi, al contrario ilCarmenère ha trovato il miglior adattamento nei suoli ghiaioso – ciottolosi. L’analisisensoriale ha confermato i dati analitici del Merlot ma non pienamente quelli del Carmenère.

4 - 138



CONCLUSIONS There are a lot of vineyard soils that have low C.O. content, fine or medium fine texture in the

surface, moderately well-drained or loam soils with low organic carbon content, poorly drained and so they are very vulnerable to compaction.

In fact, the major part of the vineyard soil of the DOC Piave area (about 44% of the cartographic units) has moderate vulnerability, about 33% high and the rest (23%) nil to low.

This means that nearly the 80% of vineyard soils has at least moderate vulnerability to compaction, so in these soils we have to adopt appropriate tillage techniques and adequate soil management to preserve the land and to maintain the crop yield.

To reduce compaction levels, the best method is increasing organic matter levels: this maximizes the aggregation of soil particles and consequently increases soil stability. In addition, rotation crops provide a variety of root types and patterns in the soil that break up compacted layers.

ACKNOWLEDGMENTS Thanks to CRA-VIT (Research Centre for Agriculture-Viticulture) of Conegliano (TV) for

some data, in particular the PhD Diego Tomasi. The present study is a part of the research activity of PhD in Viticulture, Oenology and

Marketing of Wine Companies (Padua University) with the supervision of the prof. Giuseppe Concheri.

BIBLIOGRAPHYAgriculture Canada, 1976. Glossary of terms in soil science. Canada Department of Agriculture,

Publication 1459. Ottawa: Information Canada. ARPAV, 2008. Carta dei suoli della provincia di Treviso. Castelfranco Veneto (TV). ARPAV, 2005. Carta dei suoli del Veneto. Castelfranco Veneto (TV). Regione del Veneto, ESAV, 1996. I suoli dell’area a DOC del Piave - Provincia di Treviso. Serie

pedologica 2.

Websites Agriculture and Agri-Food Canada: http://www.agr.gc.caVeneto Region: http://www.regione.veneto.it/MondoAgricolo/NewsView.aspx?idNews=1185Veneto Agricoltura: http://www.venetoagricoltura.it/basic.php?ID=1955

RAPPORTI TRA DIVERSE TIPOLOGIE DI TERRENO E RISPOSTEPRODUTTIVE E QUALITATIVE DELLE UVE MERLOT E

CARMENÈRE NELL’AREA DOC PIAVE.Soil as it effects qualitative and quantitative performance of Merlot and

Carmenere grapes in the DOC Piave Area

D. Tomasi 1, P. Marcuzzo 1, A. Garlato 2, F. Gaiotti 1, L. Lovat 11 CRA – VIT : Centro di Ricerca per la Viticoltura, Viale XXVII Aprile 26 31015 Conegliano (TV) – Italia.2 ARPAV – Agenzia Regionale per la Prevenzione e Protezione Ambientale del Veneto– Servizio OsservatorioSuolo, Via Baciocchi 9 - 31033 Castelfranco Veneto (TV) – Italy.

ABSTRACTGiving the important effects of the environmental factors on the vine productivity and grape

quality, a branch of viticulture research has been focusing on the relation between vines andtheir ecosystems for years.

The DOC Piave area, located in the eastern part of the Veneto region, was the object of aspecific zoning study from 2007 to 2009.

The study compared two different types of soils, one located in the Southern part of theDOC Area has clay-loam texture, the other located further Nord has a gravelly texture. Forboth varieties the trellising system was Sylvoz and the vines were grafted on Kober 5bb.Sugar accumulation, pigments amount, total acidity and pH were determined along withvegetative and productive parameters.

The results confirmed that there exist a close relationship between soil and grape quality,but each variety responds in a different way: Merlot had the most interesting quality whengrown clay-loam soils, while a different behaviour was found in Carmenere. The winesensory score confirmed the grape analysis for Merlot, but only partially for Carmenere.

RIASSUNTODa anni la ricerca viticola sta orientando le sue attenzioni verso lo studio della vocazionalità

degli ecosistemi viticoli, perché fulcro della produttività della vite e qualità dei suoi frutti. Dal2007 anche l’area a DOC del Piave, situata nella parte orientale della regione Veneto, èoggetto di uno specifico studio.

Il lavoro ha messo a confronto due diverse tipologie di suolo, uno a tessitura fine (limoso –argilloso) più a sud dell’area DOC Piave e l’altro a tessitura più grossolana (ghiaioso –ciottoloso) nella zona più a nord. Entrambe le varietà coltivate erano allevate a Sylvoz,innestate su Kober 5BB. Lo studio ha verificato nella bacca il contenuto di sostanze coloranti,il contenuto in solidi solubili, dell’acidità totale, del pH oltre ai parametri produttivi evegetativi quali: n° grappoli/vite, produzione uva/vite, peso medio del grappolo e il legno dipotatura.

I risultati ottenuti nel triennio, hanno permesso di evidenziare come le caratteristiche delterreno abbiano influenzato nettamente sia le rese produttive sia la qualità delle uve. Qualitàche per la varietà Merlot è stata superiore nei suoli limoso – argillosi, al contrario ilCarmenère ha trovato il miglior adattamento nei suoli ghiaioso – ciottolosi. L’analisisensoriale ha confermato i dati analitici del Merlot ma non pienamente quelli del Carmenère.

4 - 139



INTRODUZIONELa valutazione delle potenzialità produttive e qualitative di un ecosistema viticolo passa

necessariamente per lo studio accurato dei rapporti che si instaurano tra genotipo e ambiente.Per tale ragione si tende sempre più a parlare di qualità innata e qualità acquisita.Quest’ultima dipende dalle tecniche colturali di conduzione del vigneto e dalle praticheenologiche applicate in cantina, la qualità innata nasce invece dall’interazione tra i due fattorinaturali dell’agrosistema vitivinicolo: il fattore biologico (vitigno e portainnesto) e il fattoreambientale (suolo, clima e paesaggio) (Toninato et al., 2005).

Per quanto sopra esposto, ed essendo ormai acquisito che i risultati produttivi e qualitativi diuna zona sono il frutto dell’interazione tra genotipo e ambiente, appare essenziale capire comela componente eco-pedologica di un terroir eserciti la sua influenza nel caratterizzarel’espressione di un vitigno (Falcetti et al., 1993; Morlat et al., 1996; Iacono, Scienza, 1999,Tomasi et al., 2006;). Per tale ragione uno studio accurato del terroir deve essere la base perdefinire la vocazionalità di una zona viticola permettendo di identificare le aree che meglioesprimono le potenzialità di un determinato genotipo (Reynolds et al., 1996; Konduras et al.,2006).

In Veneto, regione del Nord – Est d’Italia, è stato condotto un importante lavoro dicaratterizzazione che ha coinvolto quasi tutte le aree viticole a DOC (Tomasi et al 2004,Tomasi et al, 2008; Scienza, 2008). Dal 2007 anche l’area DOC del Piave, localizzata nellaparte orientale della regione, è oggetto di uno specifico studio di zonazione. Scopo delpresente lavoro è quello di indagare e verificare come le caratteristiche fisico – chimiche didue diversi suoli (uno argilloso – limoso e l’altro ghiaioso – ciottoloso) influenzino la qualitàdi due cultivar a bacca rossa, il Merlot e il Carmenere.

MATERIALI E METODIL’indagine è stata condotta nel triennio 2007 – 2009 nella pianura trevigiana in vigneti

situati in due ambienti con la stessa natura geologica ma caratterizzati da una diversapedologia attuale, entrambi sono appartenenti allo stesso bacino mesoclimatico.

Secondo la classificazione di Koppen, la zona ha clima temperato umido con estati calde. Letemperature medie annue sono pari a 12,6 °C, che si alzano a 18,1 °C nella stagionevegetativa (aprile/ottobre), a cui corrisponde un Indice di Huglin di 2605 unità e un valore diWinkler di 1835. Le piogge annuali sono pari a 1090 mm di cui il 70 % è disponibile nelcorso del periodo vegetativo (i dati climatici riportati sono la media di 19 annate dal 1991 al2009).

Il suolo G – C (ghiaioso – ciottoloso) è caratteristico dell’alta pianura trevigiana, costituitoda conoidi ghiaiosi di origine fluvio–glaciale. Sono suoli moderatamente profondi a tessituramedia, con abbondante scheletro nel substrato, calcarei con drenaggio buono, permeabilitàmoderatamente alta, con falda assente e bassa capacità di riserva idrica.

Il suolo L – A (limoso – argilloso) è tipico della bassa pianura trevigiana, costituito sempreda depositi di origine fluvio-glaciale ma di granulometria più fine. Sono suoli moderatamenteprofondi, con scheletro assente, calcarei con drenaggio mediocre, permeabilitàmoderatamente alta, con accumulo di carbonati in profondità e buona capacità di riservaidrica.

Anche se l’origine dei suoli è la stessa, diverso è lo stato attuale, la granulometria delmateriale di deposizione è completamente differente (vedi tab1) e questo ha chiaramente unriflesso determinante sull’attività vegetativa e produttiva della pianta.

Per la varietà Carmenère sono stati individuati 7 vigneti guida (4 nel suolo G – C e 3 nelsuolo L - A), mentre per la varietà Merlot sono stati identificati 6 vigneti campione (3 per

ogni tipologia di suolo) un totale quindi di 13 vigneti, sui quali sono stati condotti i rilieviqualitativi e produttivi per tre annate. Tutti i vigneti guida presentavano le stesse tipologie diimpianto: allevamento a Sylvoz, sesto di impianto “medio” di 1,4 X 3,1; portainnesto kober5BB, con terreno inerbito sugli interfilari e diserbato sulla fila.

I rilievi effettuati comparativamente nei due siti, hanno riguardato la qualità dell’uva (ilcontenuto zuccherino, l’acidità titolabile, il contenuto di antociani e il pH), la produzione almomento della vendemmia (la produzione per ceppo, il peso del grappolo e il numero digrappoli per pianta,) e il peso del legno di potatura. Inoltre nelle ultime due annate si è volutoverificare anche la componente estraibile della frazione colorante. Infine, nel triennio, da ognisito e per tutte e quattro le tesi a confronto è stata microvinificata una quantità d’uva pari a150 Kg e degustata da un panel esperto di 8 degustatori.

Per definire il rapporto tra la diversa granulometria e l’acqua accumulata, in due dei tredicivigneti (uno per tipologia di suolo) sono stati posizionati 8 pozzetti di misura dell’umidità (4per ogni tipologia di suolo). Mediante l’utilizzo dello strumento TDR è stata monitorata,durante la stagione vegetativa e per due annate, l’umidità dei suoli (v/v) a 60 cm diprofondità.

Tab. 1 Analisi fisico-chimica del terreno effettuate nei primi 30 centimetri. In alto immagine delsuolo limoso argilloso, in basso foto del suolo sabbioso ciottoloso.

Suolo G – C L - AScheletro abbondante assente

Sabbia (%) 56,0 34,7Argilla (%) 8,7 25,6Limo (%) 35,3 39,7pH in acqua 8,05 7,73Calcare totale (%) 29,7 22,3Calcare attivo (%) 2,1 2,4Capacità di scambio cationico(meq/ Kg) 17,3 22,9Sostanza organica (%) 3,17 2,19Azoto totale (%) 0,19 0,15Potassio scambiabile (mg/ kg) 124 211Fosforo assimilabile (mg/ kg) 21 19Magnesio scambiabile (mg/kg) 380 586Calcio scambiabile (mg/ kg) 2751 3560Ferro assimilabile (mg/ kg) 25 36Boro solubile (mg/ kg) 0,63 0,65Rapporto Mg/K 10,2 12,4Rapporto C/N 9,6 8,8

RISULTATI E DISCUSSIONIStato idrico dei suoli. Nel grafico 1 sono riportati i dati relativi allo stato idrico dei suoli

rilevati nelle annate 2008 e 2009. In entrambe le annate l’acqua è riuscita a discriminare ilcomportamento dei due suoli, ma un discorso più approfondito merita il 2009. Infatti nelcorso del periodo primaverile di quest’annata la disponibilità di acqua non è stata limitantegrazie alle sufficienti precipitazioni ma è a partire dalla fioritura che l’umidità del terreno hainiziato a differenziare i suoli sulla base della loro diversa capacità di ritenzione idrica. Infattida giugno in poi nei suoli G – C la percentuale di acqua si è sempre mantenuta su valoricompresi tra il 10% e il 15%. Diversamente nei suoli L – A la disponibilità idrica è intorno al

4 - 140



INTRODUZIONELa valutazione delle potenzialità produttive e qualitative di un ecosistema viticolo passa

necessariamente per lo studio accurato dei rapporti che si instaurano tra genotipo e ambiente.Per tale ragione si tende sempre più a parlare di qualità innata e qualità acquisita.Quest’ultima dipende dalle tecniche colturali di conduzione del vigneto e dalle praticheenologiche applicate in cantina, la qualità innata nasce invece dall’interazione tra i due fattorinaturali dell’agrosistema vitivinicolo: il fattore biologico (vitigno e portainnesto) e il fattoreambientale (suolo, clima e paesaggio) (Toninato et al., 2005).

Per quanto sopra esposto, ed essendo ormai acquisito che i risultati produttivi e qualitativi diuna zona sono il frutto dell’interazione tra genotipo e ambiente, appare essenziale capire comela componente eco-pedologica di un terroir eserciti la sua influenza nel caratterizzarel’espressione di un vitigno (Falcetti et al., 1993; Morlat et al., 1996; Iacono, Scienza, 1999,Tomasi et al., 2006;). Per tale ragione uno studio accurato del terroir deve essere la base perdefinire la vocazionalità di una zona viticola permettendo di identificare le aree che meglioesprimono le potenzialità di un determinato genotipo (Reynolds et al., 1996; Konduras et al.,2006).

In Veneto, regione del Nord – Est d’Italia, è stato condotto un importante lavoro dicaratterizzazione che ha coinvolto quasi tutte le aree viticole a DOC (Tomasi et al 2004,Tomasi et al, 2008; Scienza, 2008). Dal 2007 anche l’area DOC del Piave, localizzata nellaparte orientale della regione, è oggetto di uno specifico studio di zonazione. Scopo delpresente lavoro è quello di indagare e verificare come le caratteristiche fisico – chimiche didue diversi suoli (uno argilloso – limoso e l’altro ghiaioso – ciottoloso) influenzino la qualitàdi due cultivar a bacca rossa, il Merlot e il Carmenere.

MATERIALI E METODIL’indagine è stata condotta nel triennio 2007 – 2009 nella pianura trevigiana in vigneti

situati in due ambienti con la stessa natura geologica ma caratterizzati da una diversapedologia attuale, entrambi sono appartenenti allo stesso bacino mesoclimatico.

Secondo la classificazione di Koppen, la zona ha clima temperato umido con estati calde. Letemperature medie annue sono pari a 12,6 °C, che si alzano a 18,1 °C nella stagionevegetativa (aprile/ottobre), a cui corrisponde un Indice di Huglin di 2605 unità e un valore diWinkler di 1835. Le piogge annuali sono pari a 1090 mm di cui il 70 % è disponibile nelcorso del periodo vegetativo (i dati climatici riportati sono la media di 19 annate dal 1991 al2009).

Il suolo G – C (ghiaioso – ciottoloso) è caratteristico dell’alta pianura trevigiana, costituitoda conoidi ghiaiosi di origine fluvio–glaciale. Sono suoli moderatamente profondi a tessituramedia, con abbondante scheletro nel substrato, calcarei con drenaggio buono, permeabilitàmoderatamente alta, con falda assente e bassa capacità di riserva idrica.

Il suolo L – A (limoso – argilloso) è tipico della bassa pianura trevigiana, costituito sempreda depositi di origine fluvio-glaciale ma di granulometria più fine. Sono suoli moderatamenteprofondi, con scheletro assente, calcarei con drenaggio mediocre, permeabilitàmoderatamente alta, con accumulo di carbonati in profondità e buona capacità di riservaidrica.

Anche se l’origine dei suoli è la stessa, diverso è lo stato attuale, la granulometria delmateriale di deposizione è completamente differente (vedi tab1) e questo ha chiaramente unriflesso determinante sull’attività vegetativa e produttiva della pianta.

Per la varietà Carmenère sono stati individuati 7 vigneti guida (4 nel suolo G – C e 3 nelsuolo L - A), mentre per la varietà Merlot sono stati identificati 6 vigneti campione (3 per

ogni tipologia di suolo) un totale quindi di 13 vigneti, sui quali sono stati condotti i rilieviqualitativi e produttivi per tre annate. Tutti i vigneti guida presentavano le stesse tipologie diimpianto: allevamento a Sylvoz, sesto di impianto “medio” di 1,4 X 3,1; portainnesto kober5BB, con terreno inerbito sugli interfilari e diserbato sulla fila.

I rilievi effettuati comparativamente nei due siti, hanno riguardato la qualità dell’uva (ilcontenuto zuccherino, l’acidità titolabile, il contenuto di antociani e il pH), la produzione almomento della vendemmia (la produzione per ceppo, il peso del grappolo e il numero digrappoli per pianta,) e il peso del legno di potatura. Inoltre nelle ultime due annate si è volutoverificare anche la componente estraibile della frazione colorante. Infine, nel triennio, da ognisito e per tutte e quattro le tesi a confronto è stata microvinificata una quantità d’uva pari a150 Kg e degustata da un panel esperto di 8 degustatori.

Per definire il rapporto tra la diversa granulometria e l’acqua accumulata, in due dei tredicivigneti (uno per tipologia di suolo) sono stati posizionati 8 pozzetti di misura dell’umidità (4per ogni tipologia di suolo). Mediante l’utilizzo dello strumento TDR è stata monitorata,durante la stagione vegetativa e per due annate, l’umidità dei suoli (v/v) a 60 cm diprofondità.

Tab. 1 Analisi fisico-chimica del terreno effettuate nei primi 30 centimetri. In alto immagine delsuolo limoso argilloso, in basso foto del suolo sabbioso ciottoloso.

Suolo G – C L - AScheletro abbondante assente

Sabbia (%) 56,0 34,7Argilla (%) 8,7 25,6Limo (%) 35,3 39,7pH in acqua 8,05 7,73Calcare totale (%) 29,7 22,3Calcare attivo (%) 2,1 2,4Capacità di scambio cationico(meq/ Kg) 17,3 22,9Sostanza organica (%) 3,17 2,19Azoto totale (%) 0,19 0,15Potassio scambiabile (mg/ kg) 124 211Fosforo assimilabile (mg/ kg) 21 19Magnesio scambiabile (mg/kg) 380 586Calcio scambiabile (mg/ kg) 2751 3560Ferro assimilabile (mg/ kg) 25 36Boro solubile (mg/ kg) 0,63 0,65Rapporto Mg/K 10,2 12,4Rapporto C/N 9,6 8,8

RISULTATI E DISCUSSIONIStato idrico dei suoli. Nel grafico 1 sono riportati i dati relativi allo stato idrico dei suoli

rilevati nelle annate 2008 e 2009. In entrambe le annate l’acqua è riuscita a discriminare ilcomportamento dei due suoli, ma un discorso più approfondito merita il 2009. Infatti nelcorso del periodo primaverile di quest’annata la disponibilità di acqua non è stata limitantegrazie alle sufficienti precipitazioni ma è a partire dalla fioritura che l’umidità del terreno hainiziato a differenziare i suoli sulla base della loro diversa capacità di ritenzione idrica. Infattida giugno in poi nei suoli G – C la percentuale di acqua si è sempre mantenuta su valoricompresi tra il 10% e il 15%. Diversamente nei suoli L – A la disponibilità idrica è intorno al

4 - 141



25%, con punte del 35% nel mese di settembre in concomitanza di fenomeni piovosi (mm dipioggia di settembre: 63).

Questo diverso comportamento ha chiaramente un riflesso importante sulle dinamiche di

maturazione delle uve e sarà importante per interpretare le differenze riscontrate.Aspetti produttivi. La produzione per ceppo è una variabile strettamente dipendente dal

genotipo e dall’ambiente di coltivazione, con ripercussioni sulla fertilità delle gemme e sulpeso del grappolo. Dai dati riportati in tabella 2 si nota che l’annata 2008 è risultata piùproduttiva per entrambe le varietà, conseguenza del maggior numero di grappoli per ceppo.L’annata 2007, caratterizzata da un’estate siccitosa, ha differenziato i due suoli in entrambele cultivar: i suoli G – C hanno risentito in maniera particolare delle scarse disponibilitàidriche con pesi del grappolo notevolmente inferiori rispetto ai suoli L – A. Nel complesso,vedi tab.3, il suolo non ha influito sulla risposta produttiva delle due varietà, anche se i vigneticoltivati sui suoli G – C hanno mostrato una minore adattabilità alle avverse condizioniclimatiche (annata 2007). Più marcate sono invece le differenze tra le due cultivar (tab. 3). Idati sul peso dell’acino e sul legno di potatura caratterizzano infatti in maniera univoca le duevarietà. In conclusione, le differenze più importanti sembrano dovute più al genotipo che non

0

5

10

15

20

25

30gi

u-08

lug-

08

ago-

08

set-0

8

%um

idità

dels

uolo

(v/v

)

20

40

60

80

100

120

140

piov

osità

(mm

)

G - C L - A piovosità

05

101520253035404550

mar

-09

apr-0

9

mag

-09

giu-

09

lug-

09

ago-

09

set-0

9

ott-0

9

%um

idità

dels

uolo

(v/v

)

20

30

40

50

60

70

80

90

100

piov

osità

(mm

)G - C L - A piovosità

Graf.1 Piovosità mensile e contenuto di umidità nei due suoli.In alto i dati relativi all’annata 2008, in basso i dati relativi all’annata 2009

all’ambiente di coltivazione che non ha influito in nessuna delle due varietà sulle reseproduttive.

Aspetti qualitativi. Di maggiore interesse sono i dati relativi all’aspetto qualitativo. Infattiper entrambe le varietà i suoli a tessitura L – A si sono contraddistinti per una maggioregradazione (tabella 2 e 3) e solo nel 2007, specificatamente per il Carmenère, c’è stato unmaggiore accumulo nei suoli G – C, comportamento influenzato però dal maggiore caricoproduttivo dei suoli L - A (12,0 Kg vs 7,5 Kg). Come per le prestazioni produttive c’è statauna diversa risposta della vite nelle diverse annate ed è stato il 2009 l’annata con i maggioriaccumuli zuccherini. L’acidità titolabile e il pH non sembrano invece essere influenzate ne dalsito di coltivazione ne dalla varietà, anche se l’acidità titolabile parrebbe leggermentesuperiore nel Merlot rispetto al Carmenère. La sostanza colorante ha marcato la diversità tra ledue varietà ed è il Carmenère a raggiungere i maggiori contenuti. Interessante notare lasignificatività dell’interazione tra suolo e cultivar relativamente al contenuto di antociani (tab.3) e mentre per il Merlot i suoli a tessitura più fine permettono una superiore intensità

Tab. 2 Prestazioni produttive e qualitative nel triennio 2007 – 2009 nei due siti per la varietà Carmenère eMerlot.

(dati raccolti al momento della vendemmia)Carmenère Merlot

2007 2008 2009 2007 2008 2009G - C L - A G - C L - A G - C L - A G - C L - A G - C L - A G - C L - A

grappoli / vite 52 62 63 68 45 44 55 47 71 58 59 53produzione uva / vite (kg) 7,5 12,0 11,5 9,4 8,2 4,5 8,7 9,3 11,1 10,1 9,5 8,1peso grappolo (g) 144 196 172 133 183 112 157 200 157 181 163 159peso acino (g) 2,18 2,14 1,88 1,96 2,35 2,15 1,88 1,57 1,56 1,62 1,62 1,47zuccheri (°Brix) 19,2 18,2 18,8 20,0 19,0 21,6 19,1 20,6 18,4 20,6 20,0 22,2Acidità titolabile (g/L) 4,4 5,5 5,5 5,2 4,8 6,1 4,9 4,5 6,3 5,3 6,1 6,1pH 3,39 3,34 3,33 3,34 3,74 3,54 3,41 3,38 3,14 3,18 3,67 3,43legno di potatura / vite (Kg) 1,97 1,40 1,88 1,46 1,53 1,97 0,69 0,61 0,66 0,61 0,57 0,79antociani estrabili ( mg/Kg uva) - - 516 421 792 303 - - 298 310 302 703antociani totali (mg/Kg uva) 1259 938 885 749 1115 959 463 668 595 705 791 1029% estraibilità - - 58 60 71 32 - - 50 44 38 68Uva / legno 4,3 8,5 6,6 6,5 5,9 2,3 13,2 11,5 22,7 16,8 17,4 9,3

Tab. 3 Influenza del sito e della varietà nelle prestazioni produttive e qualitative alla vendemmia; mediadel triennio 2007 -2009. (dati raccolti al momento della vendemmia)

zonagrappoli/ vite

produzioneuva / vite(kg)

pesograppolo(g)

pesoacino(g)

zuccheri (°Brix)

Aciditàtitolabile(g/L) pH

legno dipotaturavite (Kg)

antocianiestrabili(mg/Kguva)

antocianitotali(mg/Kguva)

G – C 59 9,7 163 1,97 19,1 5,4 3,45 1,20 477 899L – A 56 9,0 165 1,77 20,6 5,4 3,37 1,12 400 814significatività ns ns ns ns ** ns ns ns ns ns

Carmenère 56 9,1 158 2,13 19,4 5,2 3,44 1,72 508 1040Merlot 59 9,6 168 1,63 20,1 5,6 3,39 0,65 378 672significatività ns ns ns *** * ns ns *** ns ***interazione ns ns ns ns ns ns ns ns ns *

*, **, ***, ns: Significatività rispettivamente a p< di 0,05; 0,01; 0,001 o non significativo

4 - 142



25%, con punte del 35% nel mese di settembre in concomitanza di fenomeni piovosi (mm dipioggia di settembre: 63).

Questo diverso comportamento ha chiaramente un riflesso importante sulle dinamiche di

maturazione delle uve e sarà importante per interpretare le differenze riscontrate.Aspetti produttivi. La produzione per ceppo è una variabile strettamente dipendente dal

genotipo e dall’ambiente di coltivazione, con ripercussioni sulla fertilità delle gemme e sulpeso del grappolo. Dai dati riportati in tabella 2 si nota che l’annata 2008 è risultata piùproduttiva per entrambe le varietà, conseguenza del maggior numero di grappoli per ceppo.L’annata 2007, caratterizzata da un’estate siccitosa, ha differenziato i due suoli in entrambele cultivar: i suoli G – C hanno risentito in maniera particolare delle scarse disponibilitàidriche con pesi del grappolo notevolmente inferiori rispetto ai suoli L – A. Nel complesso,vedi tab.3, il suolo non ha influito sulla risposta produttiva delle due varietà, anche se i vigneticoltivati sui suoli G – C hanno mostrato una minore adattabilità alle avverse condizioniclimatiche (annata 2007). Più marcate sono invece le differenze tra le due cultivar (tab. 3). Idati sul peso dell’acino e sul legno di potatura caratterizzano infatti in maniera univoca le duevarietà. In conclusione, le differenze più importanti sembrano dovute più al genotipo che non

0

5

10

15

20

25

30

giu-

08

lug-

08

ago-

08

set-0

8

%um

idità

dels

uolo

(v/v

)

20

40

60

80

100

120

140

piov

osità

(mm

)


05

101520253035404550

mar

-09

apr-0

9

mag

-09

giu-

09

lug-

09

ago-

09

set-0

9

ott-0

9

%um

idità

dels

uolo

(v/v

)

20

30

40

50

60

70

80

90

100

piov

osità

(mm

)


Graf.1 Piovosità mensile e contenuto di umidità nei due suoli.In alto i dati relativi all’annata 2008, in basso i dati relativi all’annata 2009

all’ambiente di coltivazione che non ha influito in nessuna delle due varietà sulle reseproduttive.

Aspetti qualitativi. Di maggiore interesse sono i dati relativi all’aspetto qualitativo. Infattiper entrambe le varietà i suoli a tessitura L – A si sono contraddistinti per una maggioregradazione (tabella 2 e 3) e solo nel 2007, specificatamente per il Carmenère, c’è stato unmaggiore accumulo nei suoli G – C, comportamento influenzato però dal maggiore caricoproduttivo dei suoli L - A (12,0 Kg vs 7,5 Kg). Come per le prestazioni produttive c’è statauna diversa risposta della vite nelle diverse annate ed è stato il 2009 l’annata con i maggioriaccumuli zuccherini. L’acidità titolabile e il pH non sembrano invece essere influenzate ne dalsito di coltivazione ne dalla varietà, anche se l’acidità titolabile parrebbe leggermentesuperiore nel Merlot rispetto al Carmenère. La sostanza colorante ha marcato la diversità tra ledue varietà ed è il Carmenère a raggiungere i maggiori contenuti. Interessante notare lasignificatività dell’interazione tra suolo e cultivar relativamente al contenuto di antociani (tab.3) e mentre per il Merlot i suoli a tessitura più fine permettono una superiore intensità

Tab. 2 Prestazioni produttive e qualitative nel triennio 2007 – 2009 nei due siti per la varietà Carmenère eMerlot.

(dati raccolti al momento della vendemmia)Carmenère Merlot

2007 2008 2009 2007 2008 2009G - C L - A G - C L - A G - C L - A G - C L - A G - C L - A G - C L - A

grappoli / vite 52 62 63 68 45 44 55 47 71 58 59 53produzione uva / vite (kg) 7,5 12,0 11,5 9,4 8,2 4,5 8,7 9,3 11,1 10,1 9,5 8,1peso grappolo (g) 144 196 172 133 183 112 157 200 157 181 163 159peso acino (g) 2,18 2,14 1,88 1,96 2,35 2,15 1,88 1,57 1,56 1,62 1,62 1,47zuccheri (°Brix) 19,2 18,2 18,8 20,0 19,0 21,6 19,1 20,6 18,4 20,6 20,0 22,2Acidità titolabile (g/L) 4,4 5,5 5,5 5,2 4,8 6,1 4,9 4,5 6,3 5,3 6,1 6,1pH 3,39 3,34 3,33 3,34 3,74 3,54 3,41 3,38 3,14 3,18 3,67 3,43legno di potatura / vite (Kg) 1,97 1,40 1,88 1,46 1,53 1,97 0,69 0,61 0,66 0,61 0,57 0,79antociani estrabili ( mg/Kg uva) - - 516 421 792 303 - - 298 310 302 703antociani totali (mg/Kg uva) 1259 938 885 749 1115 959 463 668 595 705 791 1029% estraibilità - - 58 60 71 32 - - 50 44 38 68Uva / legno 4,3 8,5 6,6 6,5 5,9 2,3 13,2 11,5 22,7 16,8 17,4 9,3

Tab. 3 Influenza del sito e della varietà nelle prestazioni produttive e qualitative alla vendemmia; mediadel triennio 2007 -2009. (dati raccolti al momento della vendemmia)

zonagrappoli/ vite

produzioneuva / vite(kg)

pesograppolo(g)

pesoacino(g)

zuccheri (°Brix)

Aciditàtitolabile(g/L) pH

legno dipotaturavite (Kg)

antocianiestrabili(mg/Kguva)

antocianitotali(mg/Kguva)

G – C 59 9,7 163 1,97 19,1 5,4 3,45 1,20 477 899L – A 56 9,0 165 1,77 20,6 5,4 3,37 1,12 400 814significatività ns ns ns ns ** ns ns ns ns ns

Carmenère 56 9,1 158 2,13 19,4 5,2 3,44 1,72 508 1040Merlot 59 9,6 168 1,63 20,1 5,6 3,39 0,65 378 672significatività ns ns ns *** * ns ns *** ns ***interazione ns ns ns ns ns ns ns ns ns *

*, **, ***, ns: Significatività rispettivamente a p< di 0,05; 0,01; 0,001 o non significativo

4 - 143



colorante, il Carmenère trova il migliore habitat nei suoli grossolani in situazioni dove l’acquasi trova spesso in situazioni di leggera carenza (graf. 2), pur raggiungendo nei suoli L – Agradazioni superiori.

Qualità del vino. Il giudizio sensoriale relativo alle micro vinificazioni dei vini è riportato nelgraf 3 e soprattutto per il Merlot è risultato determinante l’effetto del suolo. Per questo vitignoi suoli L – A hanno conferito maggiore piacevolezza, intensità e persistenza olfattiva. Sono

prevalse, in misura minore, anche note di fruttato, floreale ed eleganza. Non ci sono invecedifferenze riguardo ai descrittori come il corpo e l’astringenza. Nei vini ottenuti dal vitignoCarmenère, diversamente dal Merlot, dominano note speziate e vegetali ma le differenze tra idue vini non sono però così nette, anche se rimane una leggera preferenza dei vini ottenuti neisuoli L – A dove si è riscontrato un maggior corpo e una superiore intensità olfattiva,risultando comunque meno astringenti.Anche l’analisi sensoriale conferma quindi i risultati ottenuti dalle analisi di laboratorioidentificando nei suoli a tessitura limoso – argillosa i più indicati per la coltivazione delMerlot; entrambi gli ambienti hanno invece una buona interazione con il vitigno Carmenère.

CONCLUSIONILa diversa composizione granulometrica tra i due ambienti oggetto della prova ha influito

sulle proprietà idrologiche dei siti in osservazione. Il sito G – C ha riportato una permeabilitàdecisamente superiore rispetto all’altro suolo (si vedano i dati di umidità del suolo) e questopone la vite in situazioni di maggiore stress idrico, che viene aggravata in annate

18,5

19

19,5

20

20,5

21

21,5

G - C L - A

Zucc

heri

(°B

rix)

Carmenere Merlot

500

600

700

800

900

1000

1100

1200

G - C L - A

Ant

ocia

ni(m

g/K

gdi

uva)

Carmenere Merlot

Graf. 2. Accumulo zuccherino a sinistra e contenuto in sostanza colorante a destra delle uve coltivatenei due ambienti (media delle annate 2007 – 2009).

2,0

3,0

4,0

5,0

6,0

7,0Astringente

Corpo

Eleganza

Equilibrio

FlorealeFruttato

Intensita olfattiva

Persistenza

Piacevolezza

G - C L - A

2,0

3,0

4,0

5,0

6,0

7,0Astringente

Corpo

Eleganza

Equilibrio

FruttatoIntensità olfattive

Persistenza

Speziato

Vegetale

G - C L - A

Graf. 3. Risultati dell’analisi sensoriale dei vini (media delle annate 2007 – 2008).A sinistra il vino Merlot e a destra il vino Carmenère.

particolarmente difficili (vedi ad es. 2007). Proprio l’analisi degli andamenti stagionaliconferma l’importanza delle caratteristiche idrogeologiche e la reattività del terreno, infatti leprestazioni produttive per entrambe le varietà nell’annata 2007, particolarmente siccitosa,hanno visto un calo produttivo nei suoli G – C che si sono dimostrati meno “elastici”. E’ statodimostrato che il calo è dovuto principalmente al minor peso del grappolo risultato quasi del30% inferiore. Nelle annate 2008 e 2009 invece, i suoli G – C sono stati i più produttivi.Possiamo affermare che la costanza produttiva viene maggiormente garantita dai suolilimoso-argillosi perché in grado di fronteggiare carenze idriche permettendo alla pianta disopperire ad eventi anomali come quelli dell’annata 2007.Emerge chiaro anche l’effetto dell’interazione vitigno – terreno; ogni vitigno ha infatti unapropria espressione quali – quantitativa che induce a valutare caso per caso le diverse scelte: isuoli L – A si sono dimostrati favorevoli alla coltivazione del Merlot sia in termini diaccumuli zuccherini che di antociani; a supporto dei rilievi analitici anche le degustazioniconfermano la preferenze dei vini Merlot prodotti in questi suoli. La varietà Carmenère hadato invece i risultati migliori nei suoli G – C producendo uve con maggiori quantità insostanze coloranti, in tutte e tre le annate, pur con produzioni sensibilmente superiori;superiorità nei parametri analitici che però non è del tutto emersa con le degustazioni dove ilpanel non ha espresso una netta preferenza per uno dei due vini degustati.Questo lavoro ha voluto sottolineare ancora una volta l’importanza del substrato pedologiconell’indirizzare la qualità futura del vino, trasmettendo la consapevolezza che solo accurati epluriennali studi di zonazione possono decretare l’adattabilità di un vitigno e la suainterazione con determinati tipi di suolo.

BIBLIOGRAFIAFalcetti M., Bertamini M., Porro D., 1993. Determinazione dell’effetto del suolo e della produzione sulle

caratteristiche organolettiche dei vini. Vignevini, 20 (7 - 8): 78 – 82.Iacono F., Scienza A., 1999. Il rapporto vite territorio alle soglie del Duemila. Vignevini, 26 (9): 25 – 33.Konduras S., Marinos V., Gkoulioti A., Kotseridis Y., Van Leeuwen C., 2006. Influence of Vineyard Location

and Vine Water Status on Fruit Maturation of nonirrigated Cv. Agiorgitiko (Vitis vinifera L.). Effects onWine Phenolic and Aroma Components. J. Agric. Food Chem., 54: 5077 – 5086.

Lorenzoni A., Tomasi D, 2008. Il soave oltre la zonazione. Dalla ricerca ai Cru. Padova: Veneto AgricolturaMorlat R. 1996. Eléments importants d’une méthodologie de caractérisation des facteurs naturels du terroir, en

relation avec la réponse de la vigne à travers le vin. Les terroirs viticoles : concept, produit, valorisation In:Actes di 1er colloque international. Angers, France : 17 – 31.

Reynolds A. G., Wardle D. A., Dever M., 1996. Vine Performance, Fruit Composition, and Wine SensoryAttributes of Gewurztraminer in Response to Vineyard Location and Canopy Manipuulation. Am. J. Enol.Vitic., 47 (1): 77 – 91.

Scienza A., Toninato L., Corrazzina E., Mariani L., Minelli R., Marangon A., Tosi E., Pastore R. , 2008. Lazonazione del Bardolino – Manuale d’uso del territorio. Padova: Veneto Agricoltura.

Tomasi D., Belvini P., Pascarella G., Sivilotti P., Giulivo C., 2006. L’effetto del suolo e sulla qualità dei vitigniCabernet Sauvignon, Cabernet franc e Merlot. Vignevini, 3: 59 – 65.

Tomasi D., Cettolin C., Calò A., Bini C., 2004. I suolied i climi della fascia collinare de lcomune di Coneglianoe loro attitudine alla coltivazione del vitigno Prosecco (Vitis sp). Comune di Conegliano(TV).

Tomasi D., Gaiotti F., 2008. Gambellara terre e colli da vino. Vicenza: Camera di Commercio IndustriaArtigianato Agricoltura Vicenza.

Toninato L., Bernava M., Cricco J, Brancadoro L. , 2005. Caratterizzazione dei terroir di Vinci e Cerreto Guidimediante le risposte del Sangiovese. Informatore Agrario, 2: 63 – 66.

4 - 144



colorante, il Carmenère trova il migliore habitat nei suoli grossolani in situazioni dove l’acquasi trova spesso in situazioni di leggera carenza (graf. 2), pur raggiungendo nei suoli L – Agradazioni superiori.

Qualità del vino. Il giudizio sensoriale relativo alle micro vinificazioni dei vini è riportato nelgraf 3 e soprattutto per il Merlot è risultato determinante l’effetto del suolo. Per questo vitignoi suoli L – A hanno conferito maggiore piacevolezza, intensità e persistenza olfattiva. Sono

prevalse, in misura minore, anche note di fruttato, floreale ed eleganza. Non ci sono invecedifferenze riguardo ai descrittori come il corpo e l’astringenza. Nei vini ottenuti dal vitignoCarmenère, diversamente dal Merlot, dominano note speziate e vegetali ma le differenze tra idue vini non sono però così nette, anche se rimane una leggera preferenza dei vini ottenuti neisuoli L – A dove si è riscontrato un maggior corpo e una superiore intensità olfattiva,risultando comunque meno astringenti.Anche l’analisi sensoriale conferma quindi i risultati ottenuti dalle analisi di laboratorioidentificando nei suoli a tessitura limoso – argillosa i più indicati per la coltivazione delMerlot; entrambi gli ambienti hanno invece una buona interazione con il vitigno Carmenère.

CONCLUSIONILa diversa composizione granulometrica tra i due ambienti oggetto della prova ha influito

sulle proprietà idrologiche dei siti in osservazione. Il sito G – C ha riportato una permeabilitàdecisamente superiore rispetto all’altro suolo (si vedano i dati di umidità del suolo) e questopone la vite in situazioni di maggiore stress idrico, che viene aggravata in annate

18,5

19

19,5

20

20,5

21

21,5

G - C L - A

Zucc

heri

(°B

rix)

Carmenere Merlot

500

600

700

800

900

1000

1100

1200

G - C L - A

Ant

ocia

ni(m

g/K

gdi

uva)

Carmenere Merlot

Graf. 2. Accumulo zuccherino a sinistra e contenuto in sostanza colorante a destra delle uve coltivatenei due ambienti (media delle annate 2007 – 2009).

2,0

3,0

4,0

5,0

6,0

7,0Astringente

Corpo

Eleganza

Equilibrio

FlorealeFruttato

Intensita olfattiva

Persistenza

Piacevolezza

G - C L - A

2,0

3,0

4,0

5,0

6,0

7,0Astringente

Corpo

Eleganza

Equilibrio

FruttatoIntensità olfattive

Persistenza

Speziato

Vegetale

G - C L - A

Graf. 3. Risultati dell’analisi sensoriale dei vini (media delle annate 2007 – 2008).A sinistra il vino Merlot e a destra il vino Carmenère.

particolarmente difficili (vedi ad es. 2007). Proprio l’analisi degli andamenti stagionaliconferma l’importanza delle caratteristiche idrogeologiche e la reattività del terreno, infatti leprestazioni produttive per entrambe le varietà nell’annata 2007, particolarmente siccitosa,hanno visto un calo produttivo nei suoli G – C che si sono dimostrati meno “elastici”. E’ statodimostrato che il calo è dovuto principalmente al minor peso del grappolo risultato quasi del30% inferiore. Nelle annate 2008 e 2009 invece, i suoli G – C sono stati i più produttivi.Possiamo affermare che la costanza produttiva viene maggiormente garantita dai suolilimoso-argillosi perché in grado di fronteggiare carenze idriche permettendo alla pianta disopperire ad eventi anomali come quelli dell’annata 2007.Emerge chiaro anche l’effetto dell’interazione vitigno – terreno; ogni vitigno ha infatti unapropria espressione quali – quantitativa che induce a valutare caso per caso le diverse scelte: isuoli L – A si sono dimostrati favorevoli alla coltivazione del Merlot sia in termini diaccumuli zuccherini che di antociani; a supporto dei rilievi analitici anche le degustazioniconfermano la preferenze dei vini Merlot prodotti in questi suoli. La varietà Carmenère hadato invece i risultati migliori nei suoli G – C producendo uve con maggiori quantità insostanze coloranti, in tutte e tre le annate, pur con produzioni sensibilmente superiori;superiorità nei parametri analitici che però non è del tutto emersa con le degustazioni dove ilpanel non ha espresso una netta preferenza per uno dei due vini degustati.Questo lavoro ha voluto sottolineare ancora una volta l’importanza del substrato pedologiconell’indirizzare la qualità futura del vino, trasmettendo la consapevolezza che solo accurati epluriennali studi di zonazione possono decretare l’adattabilità di un vitigno e la suainterazione con determinati tipi di suolo.

BIBLIOGRAFIAFalcetti M., Bertamini M., Porro D., 1993. Determinazione dell’effetto del suolo e della produzione sulle

caratteristiche organolettiche dei vini. Vignevini, 20 (7 - 8): 78 – 82.Iacono F., Scienza A., 1999. Il rapporto vite territorio alle soglie del Duemila. Vignevini, 26 (9): 25 – 33.Konduras S., Marinos V., Gkoulioti A., Kotseridis Y., Van Leeuwen C., 2006. Influence of Vineyard Location

and Vine Water Status on Fruit Maturation of nonirrigated Cv. Agiorgitiko (Vitis vinifera L.). Effects onWine Phenolic and Aroma Components. J. Agric. Food Chem., 54: 5077 – 5086.

Lorenzoni A., Tomasi D, 2008. Il soave oltre la zonazione. Dalla ricerca ai Cru. Padova: Veneto AgricolturaMorlat R. 1996. Eléments importants d’une méthodologie de caractérisation des facteurs naturels du terroir, en

relation avec la réponse de la vigne à travers le vin. Les terroirs viticoles : concept, produit, valorisation In:Actes di 1er colloque international. Angers, France : 17 – 31.

Reynolds A. G., Wardle D. A., Dever M., 1996. Vine Performance, Fruit Composition, and Wine SensoryAttributes of Gewurztraminer in Response to Vineyard Location and Canopy Manipuulation. Am. J. Enol.Vitic., 47 (1): 77 – 91.

Scienza A., Toninato L., Corrazzina E., Mariani L., Minelli R., Marangon A., Tosi E., Pastore R. , 2008. Lazonazione del Bardolino – Manuale d’uso del territorio. Padova: Veneto Agricoltura.

Tomasi D., Belvini P., Pascarella G., Sivilotti P., Giulivo C., 2006. L’effetto del suolo e sulla qualità dei vitigniCabernet Sauvignon, Cabernet franc e Merlot. Vignevini, 3: 59 – 65.

Tomasi D., Cettolin C., Calò A., Bini C., 2004. I suolied i climi della fascia collinare de lcomune di Coneglianoe loro attitudine alla coltivazione del vitigno Prosecco (Vitis sp). Comune di Conegliano(TV).

Tomasi D., Gaiotti F., 2008. Gambellara terre e colli da vino. Vicenza: Camera di Commercio IndustriaArtigianato Agricoltura Vicenza.

Toninato L., Bernava M., Cricco J, Brancadoro L. , 2005. Caratterizzazione dei terroir di Vinci e Cerreto Guidimediante le risposte del Sangiovese. Informatore Agrario, 2: 63 – 66.

4 - 145

