Artigo Folsomia Toxicidade

7/23/2019 Artigo Folsomia Toxicidade

1/11

Please cite this article in press as: Winkler, D., Collembolan response to red mud pollution in Western Hungary. Appl. Soil Ecol. (2013),http://dx.doi.org/10.1016/j.apsoil.2013.07.006

ARTICLE IN PRESSGModel

APSOIL-1892; No.of Pages11

Applied Soil Ecology xxx (2013) xxxxxx

Contents lists available at ScienceDirect

Applied Soil Ecology

journal homepage: www.elsevier .com/ locate /apsoi l

Collembolan response to red mud pollution in Western Hungary

Daniel Winkler

Institute of Wildlife Management and Vertebrate Biology, University of West Hungary, Bajcsy-Zs. str. 4., 9400 Sopron, Hungary

a r t i c l e i n f o

Article history:

Received 7 February 2013Received in revised form 24 June 2013Accepted 8 July 2013

Keywords:

Collembola communitiesRed sludgeSoil pollutionSodium alkalinityHeavy metalsBioavailability

a b s t r a c t

Effects ofred mud pollution on the community structure ofCollembola were studied in soils from opengrassland and forest habitats following the red mud disaster in Western Hungary. Nearby unpollutedcontrol plots ofeach habitat types were selected for comparative purposes. Analyses revealed that soil

became strongly alkaline and, even nine months after the disaster, pH exceeded a value of 9.0 in thepolluted forests. Water soluble Na content found to be 50160 times greater in the polluted area, andtotal content ofmetals (e.g. Fe, Al, Mn, Zn, As, Cr, Cu, Ni, Pb, Zn) also increased considerably. Nevertheless,owing to the high alkalinity and red muds adsorption capacity, bioavailable forms ofheavy metals werelower in comparison to the acid control soils. Collembola species richness was about the same in thepolluted and control forests (31 and 32, respectively), but lower in the polluted meadows compared tothe control plots (21 and 27, respectively). Total community abundance changed differently in the openhabitat and in the forest. Its value dropped by 45% in the polluted meadows, while almost tripled inthe polluted forests. Changes in the abundance ofindividual species involved both decrease/eliminationof sensitive species (e.g. Isotomiella minor, Sminthurinus aureus) and displacement of species tolerantto pollution (e.g. Micranurida pygmaea) into higher abundance classes. Certain species (e.g. Folsomiamanolachei, Sphaeridiapumilis), following the pollution, showed a reverse pattern ofabundance in thetwo habitat types; increasing in the forest while decreasing in the meadow. This study has suggestedthat soil alkalinity and salt (Na) toxicity were presumably the two most important factors determiningthe structure of Collembola communities in the area affected by red mud pollution. Despite the hightoxicity risk associated with this accident, no adverse effect has been observed in Collembola abundance.

Nevertheless, as a consequence ofsoil re-acidification, re-mobilisation offixed metals may occur in thelong term, constituting to a potential risk to soil Collembola.

2013 Elsevier B.V. All rights reserved.

1. Introduction

Legacy of the 20th century has left us with two significant eco-logical questions for the new century:

How can we validate an ecologically conscientious view in eco-nomical decision making?

How can we put a stop to the dramatic decline of biodiversity?

The strong interconnection between these two questions hasenforced the need for rapid action especially in light of detrimentalindustrial disasters. Unfortunately, large scale industrial disasterscontinue to occur even in the 21st century. The most well knowninternational incident of the new era is the Japanese Fukushimadisaster; however, there have also been local incidents, where theaffects were localised, but their ecological impact could not beneglected. One such localised disaster occurred in 2010 in Hungary

Tel.: +36 70 3170807; fax: +36 99 518350.E-mail address: [email protected]

which became known as the red mud incident in internationalnews. The nature of this disaster spillage of red mud from a reser-voir has directed focus onto soil conservation and protection ofsoil biodiversity. One major reason for the increased interest in soilbiodiversitywasrelatedto thefactthatfromveryearlyon itbecameapparent that the red mud contaminated soil can only be utilisedand inhabited after complete soil removal. This point of view con-sidered thebenefits toHomosapiens alone.Butwhathappenstothesoil biodiversity in those areas where the contaminatedsoil cannotbe removedfor example: in forested areas? The answerwill prob-ably be found in soil fauna recolonisation which will most likelyoriginate from surrounding habitats where contamination did notreach and from so called survival spots. However, how soon thiscan happen and how will all this take place nobody can predict foras long as this sadly uniquein situ experimentation does notstop.

On October 2010, in Western Hungary, one of the dykes of ared mud reservoir of an aluminium processing factory breachedand more than 1.3 millionm3 of toxic sludge broke free, flood-ing the surrounding area (Szpvlgyi, 2011). This event destroyednearly1000 hectares of land, including nearbyrivers, where practi-cally all aquatic life was destroyed (Mayes et al., 2011); forests and

0929-1393/$ see front matter 2013 Elsevier B.V. All rights reserved.

http://dx.doi.org/10.1016/j.apsoil.2013.07.006
http://localhost/var/www/apps/conversion/tmp/scratch_5/dx.doi.org/10.1016/j.apsoil.2013.07.006http://localhost/var/www/apps/conversion/tmp/scratch_5/dx.doi.org/10.1016/j.apsoil.2013.07.006http://localhost/var/www/apps/conversion/tmp/scratch_5/dx.doi.org/10.1016/j.apsoil.2013.07.006http://www.sciencedirect.com/science/journal/09291393http://www.elsevier.com/locate/apsoilmailto:[email protected]://localhost/var/www/apps/conversion/tmp/scratch_5/dx.doi.org/10.1016/j.apsoil.2013.07.006http://localhost/var/www/apps/conversion/tmp/scratch_5/dx.doi.org/10.1016/j.apsoil.2013.07.006mailto:[email protected]://www.elsevier.com/locate/apsoilhttp://www.sciencedirect.com/science/journal/09291393http://localhost/var/www/apps/conversion/tmp/scratch_5/dx.doi.org/10.1016/j.apsoil.2013.07.006http://localhost/var/www/apps/conversion/tmp/scratch_5/dx.doi.org/10.1016/j.apsoil.2013.07.006


2/11




2 D. Winkler / Applied Soil Ecologyxxx (2013) xxxxxx

agriculturalfields, where thesoil surface wascovered with an aver-age 510cm thick red mud layer after the flood(Anton et al., 2012).This red mud disaster turned out to be one of the most severeecological disasters ever experienced in the country (Burai et al.,2011), leaving behind long-term environmental impacts which areimpossible to assess at the present time.

The red mud is a fine fraction by-product of the so-called Bayeralumina process which uses sodium hydroxide for recovering sub-stantially pure alumina from bauxite (Liu et al., 2007; Power et al.,2011). Red mud is therefore strongly alkaline with a pH of 912.5(Milacic et al., 2012). Its main components typically include resid-ual minerals and oxides, such as hematite, goethite, boehmite,quartz, sodium aluminosilicates, titaniumdioxide, calcium carbon-ate/aluminate and magnesium oxide (Mayes et al., 2011). Furthercomponents in lower concentrations also include heavy metalssuch as copper, zinc, chromium, cadmium, arsenic, mercury, lead,nickel and vanadium; and a few rare-earth metal elements (Cablik,2007).

To assess the environmental impact of the red mud disaster,some investigations have already been carried out. Gelencsr et al.(2011) focused on the potential health effects of red mud dust,while other studies evaluated the impact of red sludge pollutionon water quality and aquatic life (Harka, 2011; Mayes et al., 2011;Klebercz et al., 2012). Effects of red mud on plant growth andplantcompositionhavebeenextensivelystudied(Friesletal.,2003;Koulikourdiset al., 2005; Ruyters et al., 2011a); however, in regardsof soil fauna, no investigations have been carried out.

It is without doubt that red mud had considerable effect on soilproperties and soil biodiversity (Ruyters et al., 2011b; Anton et al.,2012). The high sodium content and the extremely fine grain sizeoftheredmudcan deteriorate soil structure (Ruyters et al., 2011a).Further risks associated with red mudthat affect soil properties are

related to alkalinity, heavy metal content and radioactive materialcontamination (Klauber et al., 2011; Anton et al., 2012). High pHitself can be a source of direct toxicity to soil organisms (Mertens,1975; Hutson, 1978), as well as, heavy metal pollution also tendsto affect, often adversely, soil animal communities (Hodson, 2013).Effects of individual pollutants on soil Collembola have been widelystudied (e.g. Hgvar and Abrahamsen, 1990; Hopkin, 1994; Gilletand Ponge, 2003; Lock et al., 2003; Smit and van Gestel, 1996);however, studies examining the toxic affect of complex mixturesof pollutants, such as sludges, on these organisms are less widelydocumented (Cole et al., 2001; Domene et al., 2008, 2010; Natal-da-luz et al., 2009).

The objectives of the present research were to (i) determinewhether there are signs of revitalisation and recolonisation byCollembola of the selected red mud polluted area not affectedby remediation; (ii) evaluate the changes in Collembola commu-nity species composition and abundance in polluted compared tounpolluted plots selected as control from nearbylocations; and(iii)collect new information on pH tolerance ofCollembola species infield.Thestudywasconducted9monthsaftertheredmuddisaster.

2. Materials and methods

2.1. Study area

The study was conducted in the area of the Torna stream val-ley, which was the main area affected by the toxic red mud flood.The site is located near the village of Tskevr (Fig. 1); it is charac-terised by a variety of semi-natural and agricultural habitats suchas forests, grasslands and arable lands. Sampling was carried out inboth forest and open meadow habitats. Of each habitat type, threepolluted andthreeunpolluted plots were selected forsamplingand

Fig. 1. Study area, location of sampling plots.
http://localhost/var/www/apps/conversion/tmp/scratch_5/dx.doi.org/10.1016/j.apsoil.2013.07.006http://localhost/var/www/apps/conversion/tmp/scratch_5/dx.doi.org/10.1016/j.apsoil.2013.07.006


3/11




D. Winkler / Applied Soil Ecologyxxx (2013) xxxxxx 3

comparativeanalyses.Meanplotsizewas3.38ha(SE 0.22, n =15).The polluted plots were situated in the formerly flooded area, nextto the Torna stream, while nearby unpolluted control plots werelocated at slightly higher elevations which were unaffected by theflood. Plant species composition and vegetation structure charac-teristic were taken into account in selecting control sampling plots,so as to make the most accurate comparisons possible. The sam-pled forests were characterised bya tree layer composed ofQuercusrobur, Fraxinus excelsiorand scattered trees ofCarpinus betulus andUlmus laevis. The sampled grasslands were tall herb communitieswith dominance ofArrhenatherum elatius, Deschampsia caespitosaand Dactylis glomerata.

2.2. Soil chemistry

SimultaneouslywiththeCollembola sampling,inAugustof2011,five soil core samples were collected (after litter removal) at eachof the polluted and unpolluted plots in order to determine the soilproperties and the degree of heavy metal pollution.

Soil pH was determined potentiometrically in H2Oan d1MKClusing a soil-to-solution ratio of 1:5. Percentage quantity of thesoil organic matter (Som) content was evaluated using the potas-sium dichromate oxidation method. Soil hygroscopicity (hy) wasdetermined by using Kurons method (Stefanovits, 1992). Watersoluble content of Na was measured following shaking method.Arseniccontentwasquantifiedintheaquaregiasolution.Totalcon-tent of heavy metals was measured with flame atomic absorptionspectrometer (AAS) in hydrochloric solution resulted by diges-tion of soil samples in HClO4HNO3 mixture. Bioavailable formsof selected heavy metals were determined using extraction with0.1M Na2EDTA.

2.3. Sampling and extraction of Collembola

Fromeachoftheselectedplots,fivesoilcoresof100cm 3 volume(3.6cm in diameter and10.0cm in depth)were sampled randomly;at least 10m from the edges to ensure that edge effects do not bias

theresults.Springtails were extractedfrom thetotalof 60 soil sam-pleswithin14daysusingamodifiedTullgrenapparatus.Specimenswere identified at species level following principally taxonomi-cal keys by Deharveng (1982), Fjellberg (1980, 1998), Babenkoet al. (1994), Zimdars and Dunger (1994), Weiner (1996),Jordanaet al. (1997), Pomorski (1998), Bretfeld (1999), Potapov (2001) andThibaudetal.(2004). Taxonomicclassificationisprimarilybasedonthe most recent classification byJanssens and Christiansen (2011).

2.4. Data analyses

In the analysis of the collembolan communitystructure, speciesabundance distributions were compared with three correspond-ing theoretical distributions: geometric (Motomura, 1932), broken

stick (MacArthur, 1957) and logarithmic series (Fisher et al., 1943).For comparison of observed rank abundance curves and expectedtheoretical curves Chi-square tests were performed.

In addition to the observed species richness nonparametricrichness estimators (abundance-based estimators Chao1 and ACEand incidence-based estimators Chao2 and ICE) were evaluatedusing v2.1 of the Species Richness Estimators Eco-Tool (Russell,2011), which implements the techniques described inColwell andCoddington (1994) as well as thoseofColwell et al. (2004). Single-tons and doubletons (number of species represented by one or twoindividuals) were also verified.

Three measures of species diversity were calculated for quan-titative evaluation of collembolan species diversity in the pollutedand control areas: the ShannonWeavers index (Shannon and

Weaver, 1949), the Simpson index (Simpson, 1949) and Pielous

(1966) evenness index. The Shannon indices were comparedaccording to the modified t-test proposed by Hutcheson (1970).Rnyi diversity profiles (Tthmrsz, 1997) were used for partialranking of the Collembola communities based on diversity. A com-munity of higher diversity has a profile consistently above theprofile of a less diverse community. In case the diversity profilescross each other, the communities are not comparable, and thusthe diversity comparison carried out by using t-test gets overruled.

Dominance was measured using the community dominanceindex (McNaughton, 1968) calculated as the percentage of abun-dance that is contributed by the two most abundant specieswithin the community. To evaluate the similarities of species com-position, Jaccards similarity coefficient (Jaccard, 1901) and theMorositaHorn index were used (Magurran, 1988).

Effect of selected environmental variables on collembolancommunity composition was evaluated by canonical correspon-dence analysis (CCA) performed with the software Past ver. 2.17b(Hammeretal.,2001). Inthe final analysis only those variableswereincluded (Som, hy, pH, Na content), which equalled to the mea-sured Collembola sample size (n = 15 per habitat). Species found inless than 5 samples and those represented by less than 10 individ-uals were excluded from theanalysis (33from total 59 species) dueto uncertain relationship with environmental factors. UnrestrictedMonte Carlo permutation tests, with 1000 randomisations, werecarried out for the total inertia of the CCA, thus, providing an over-all test on the effect of the constraining parameters used in theanalysis (ter Braak, 1986).

3. Results

3.1. Chemical analysis of soil

The control forest and meadow in this study presented strongacid to moderately acid alluvial soils, with average pH (in H2O)values of 5.2 and 5.8, respectively (Table 1). Soil became alkalineafter the red sludge flooding and, even nine months after the disas-

ter, pH reached an average of 9.20 (very strongly alkaline) in thered mud polluted forest and 8.20 (strongly alkaline) in the pollutedopen grassland.Hygroscopic moisture content (hy), which also rep-resents soil texture, washigherin the polluted habitats. Na contentfound to be almost 50 greater in the polluted forest compared tothe control, and this ratio was even higher (160) in relation tothe meadows. Total content of Fe and Al, which represent majorred mud components, was remarkably higher in the polluted soilsamples. As, Cr and Ni concentrations were found to be consider-ably high in the polluted sites, however; these did not exceed thethreshold concentrationsdefined by Hungarian legislation for toxictrace metals in soil (MSZ-21470-50, 1998). Concentrations of otherheavymetalssuchasZn,PbandCd,althoughincreasedaftertheredmud pollution, did not show extreme levels. Bioavailable contentof heavy metals as extracted by EDTA showed an opposite pattern:Fe, Mn was orders of magnitude lower in the polluted soils wherealso Cu and Zn were found in lower concentrations in comparisonto the control samples.

3.2. Collembola species richness, diversity and abundance

A total of 4339 specimens representing 15 families, 35 gen-era and 59 species of Collembola were collected and identified(Table 2). A particular species of the genus Lepidocyrtus, namelyLepidocyrtus tomosvaryi, was also found and proved to be new toscience (Winkler and Traser, 2012). Structural characteristics ofthe communities are presented in Table 3. The highest observedspecies richness was identified from the control forest (32) fol-

lowedbytheredmud pollutedforest (31), while theopen grassland


4/11





Table 1

Soil parameters (meanSE)in the polluted and control habitats.

Forest Meadow

Polluted Control Polluted Control

pH H2O* 9.2 0.09 5.2 0.15 8.8 0.12 5.8 0.12pH KCl* 8.1 0.11 4.6 0.12 7.6 0.15 5.1 0.10Som (%)* 12.3 0.25 11.1 0.21 8.7 0.17 8.8 0.14hy* 2.8 0.16 2.2 0.08 1.8 0.16 1.51 0.09Na (mg/kg)a,* 883 34.9 18.5 0.71 543 14.9 3.4 0.30As (mg/kg)b 9.8 0.39 4.1 0.08 8.7 0.27 4.4 0.15Total Ca (mg/kg) 854 59 1764 87 1214 41 1371 52Total Al (mg/kg) 17,658 833 6894 353 19,473 1067 7470 334Total Fe (mg/kg) 67,356 2599 11,432 766 64,023 3257 10,514 492Total K (mg/kg) 1173 65 994 58 1372 116 1018 78Total Mn (mg/kg) 1264 171 328 80 1422 106 247 38Total Cd (mg/kg) 0.57 0.09 0.32 0.04 0.49 0.08 0.24 0.04Total Cr (mg/kg) 57.8 7.94 14.1 1.75 54.4 7.00 15.2 1.44Total Cu (mg/kg) 14.3 1.48 6.9 1.30 12.3 0.88 7.8 0.87Total Ni (mg/kg) 38.6 2.04 11.6 1.79 37.4 1.83 11.2 1.13Total Pb (mg/kg) 41.3 5.98 17.2 1.45 44.2 3.98 12.9 1.12Total Zn (mg/kg) 84.3 8.0 44.3 6.7 79.3 10.5 41.4 4.4EDTA Fe (mg/kg) 60 10.3 587 130 40 4.2 237 78.9EDTA Mn (mg/kg) 8.4 1.9 694 137 43.6 13.0 241 67.8EDTA Cu (mg/kg) 2.37 0.39 3.54 0.79 1.85 0.41 2.83 0.35EDTA Zn (mg/kg) 1.60 0.49 9.07 1.39 1.98 0.48 5.46 1.19

Som: Soil organic matter; hy: hygroscopicity; n =3 (consisting of five pooled samples) with theexception of *n =15.

a Water soluble.b Aqua regia solution.

habitats (control and polluted) had lower number of species; 28and 21, respectively. Similarly, estimated species richness was thehighest in the control forest spanning from 39 (incidence-basedICE) to 41 (abundance-based Chao1) compared to the 32 actu-ally collected. Chao1 estimator also gave the highest estimate ofspecies richness in the polluted meadow, while maximum speciesrichness was predicted by ICE in the polluted forest and controlmeadow. The Shannon and Simpson diversity indices and speciesevenness indicatedthe highest collembolan diversityin the controlmeadow. According to the t-test proposed by Hutcheson (1970),Shannon diversity was significantly lower in both polluted habitat

types (p < 0.001), which is reflected also in Rnyis diversityprofiles(Fig. 2).

Collembola abundance was the highest in the polluted for-est, exceeding 90,000 ind./m2, while abundance was significantly

Fig.2. ComparisonofCollemboladiversitiesinredmudpollutedandcontrolhabitats

using Rnyis diversity profiles.

lower (t-test, P


5/11





Table 2

Species ofCollembola and their abundance (number of individuals/m2) inthered mud polluted andunaffected control habitats.Data aremeans followed by standard errors.Significant differences between thepaired polluted and control habitats (P< 0.05) areindicated in italicised letters.

Forest Meadow

Polluted (n = 15) Control (n = 15) Polluted (n = 15) Control (n =15)

Neanuridae

1. Frisea truncata Cassagnau, 1958 6745 3333 0 02. Deutonura conjuncta (Stach, 1926) 6745 3333 0 100723. Neanura muscorum (Templeton, 1835) 3333 13377 0 04. Micranurida pygmaea Brner, 1901 933425 3333 3333 05. Pseudachorutes parvulus Brner, 1901 3333 10053 0 06. Pseudachorutes pratensis Rusek, 1973 0 0 3333 3333

Brachystomellidae

7. Brachystomella parvula (Schffer, 1896) 1167585 0 1467559 3333Hypogastruridae

8. Ceratophysella luteospina (Stach, 1920) 233108 0 0 09. Ceratophysella succinea (Gisin, 1949) 3333 0 0 010. Hypogastrura vernalis (Carl, 1901) 133133 0 1667822 1533127611. Xenylla corticalis Brner, 1901 3333 0 0 0

Onichiuridae

12. Protaphorura armata (Tullberg, 1869) 0 3333 367260 103326913. Protaphorura campata (Gisin, 1952) 0 267267 0 014. Protaphorura gisini (Haybach, 1960) 0 0 1567371 900214

Tullbergidae

15. Mesaphorura critica Ellis, 1976 0 0 13377 30011816. Mesaphorura hylophila Rusek, 1982 0 3333 0 0

17. Mesaphorura krausbaueri Brner, 1901 0 0 0 1007218. Mesaphorura macrochaeta Rusek, 1976 0 467204 13377 163357219. Metaphorura riozoi Castano-Meneses,

Palacios Vargas and Traser, 20000 0 0 6745

20. Stenaphorurella lubbocki (Bagnall, 1935) 0 6767 0 0Tomoceridae

21. Pogonognathellus flavescens (Tullberg,1871)

0 0 3333 0

Isotomidae

22. Cryptopygus bipunctatus (Axelson, 1903) 267267 41001268 733381 2233115423. Folsomia candida Willem, 1902 52333263 6745 333324. Folsomia manolachei Baggnall, 1939 28,5336256 74331405 267 128 100022925. Folsomia quadriocuculata (Tullberg, 1871) 30,2336761 10,167 2331 233145 103330326. Isotoma caerulea (Bourlet, 1839) 0 0 0 26714527. Isotoma viridisBourlet 1839 0 0 0 333328. Isotomiella minor(Schffer, 1896) 0 600208 0 029. Isotomodes productus (Axelson, 1906) 0 0 0 6767

30. Isotomodes sexsetosus da Gama, 1963 0 0 0 333331. Parisotoma notabilis (Schffer, 1896) 76671867 500176 367192 500183Entomobryidae

32. Entomobrya dorsalis Uzel, 1891 0 3333 0 033. Entomobrya handschini Stach, 1922 0 0 567381 26712834. Entomobrya muscorum (Nicolet, 1842) 10053 3333 0 035. Lepidocyrtus cyaneus Tullberg, 1871 1667490 400298 0 43318236. Lepidocyrtus cf. arrabonicus Traser, 2000 0 0 0 676737. Lepidocyrtus tomosvaryi Winkler and

Traser, 2012167105 1267 338 0 3333

38. Lepidocyrtus paradoxus Uzel, 1890 6745 0 0 1005339. Pseudosinella alba (Packard, 1873) 10072 0 0 040. Pseudosinella cf. bohemica Rusek, 1979 0 833270 0 041. Seira sp. Juv 0 3333 0 042. Heteromurus major(Moniez, 1889) 200118 0 0 043. Heteromurus nitidus (Templeton, 1835) 1033226 26783 0 333344. Orchesella albofasciata Stach, 1960 0 0 233200 045. Orchesella cincta (Linnaeus, 1758) 10072 0 0 046. Orchesella flavescens (Bourlet, 1839) 867424 233128 3333 047. Orchesella multifasciata (Stscherbakow,

1898)1800951 667446 0 6767

48. Orchesella spectabilis Tullberg, 1871 15331498 300266 0 0Paronellidae

49. Cyphoderus sp.juv 0 0 3333 0Neelidae

50. Megalothorax minimus Willem, 1900 6745 0 0 0Sminthurididae

51. Sphaeridia pumilis (Krausbauer, 1898) 68002005 2033517 267168 23331394Katiannidae

52. Sminthurinus aureus(Lubbock, 1862) 0 267145 0 43321753. Sminthurinus elegans (Fitch, 1863) 467251 16780 0 233128

Arrhopalitidae

54. Arrhopalites caecus Tullberg, 1871 333116 0 0 0Sminthuridae


6/11





Table 2 (Continued )

Forest Meadow

Polluted (n = 15) Control (n = 15) Polluted (n = 15) Control (n =15)

55. Allacma fusca (Linnaeus, 1758) 0 167135 0 056. Spatulosminthurus flaviceps (Tullberg,

1871)0 500276 0 0

57. Lipothrix italica (Cassagnau, 1968) 6745 10072 0 0Bourletiellidae

58. Fasciosminthurus strigatus (Stach, 1922) 0 0 6745 059. Fasciosminthurus virgulatus (Skorikow,

1899)0 0 10053 0

Total 90,03316,467 31,367 5062 83671284 14,9003240

Table 3

Structural indices ofCollembola communities in thestudied polluted and control habitats.

Forest Meadow

Polluted Control Polluted Control

S 31 32 21 28Singletons/Doubletons 4/5 8/2 6/1 6/4ACE 34 40 27 34ICE 35 39 26 36Chao1 32 41 29 31Chao2 33 40 25 35Number of presumed species 14 79 48 38H 1.899 2.186 2.385 2.659D 0.769 0.813 0.874 0.908

J 0.553 0.631 0.783 0.798CDI 65.27 56.11 38.64 30.65

S: total number of species; ACE, ICE, Chao1 and Chao2: nonparametric richness estimators; H: ShannonWeavers diversity index; D: Simpsons diversity index;J: Pielousevenness index; CDI: community dominance index (%).

The CCA analysis reflects a more detailed response of collem-bolan communities to environmental changes (Fig. 5). Theeigenvalue of axis 1, as well as the explained variance of bothaxes were relatively high (axis 1: eigenvalue: 0.445, explained

variance: 63.84%; axis 2: eigenvalue: 0.193; explained variance:27.69%). Monte Carlo permutation testing was used to test thenull hypothesis of no association between species and soil envi-ronmental factors, and rejected this hypothesis for both axis 1

Fig. 3. Changes in theabundance of Collembolan species after thered mud pollution.


7/11





Table 4

Pairwise comparison of thepolluted and control habitats using Jaccards similarity coefficient (normal letters) and theMorisitaHorn index (italicised letters).

Forest control Forest polluted Meadow control Meadow polluted

Forest control 0.92 0.51 0.20Forest polluted 0.47 0.38 0.16Meadow control 0.30 0.31 0.57Meadow polluted 0.23 0.24 0.36

Fig. 4. Species rank abundance of the red mud polluted and unaffected controlhabitats.

and axis 2 (P< 0.01 and P< 0.05, respectively). The first axis of this

data set represents mainly soil organic-matter content (Som) andhigroscopicity (hy), whereas the second axis represents mainly thechemical factors (soil pH, water soluble Na). Axis 1 clearly sepa-rated the forest and meadow habitats while axis 2 separated thepolluted and control plots.

Species typical for open grassland are those far from the originof the positive side of axis 1 (B. parvula, H. vernalis, Protaphoruraarmata, Protaphorura gisini, Mesaphorura critica and Entomobryahandschini), while species found exclusively or predominantly inforests (e.g. P. cf. bohemica, Orchesella spectabilis, Arrhopalites cae-cus) were projected onthe negative sideof the sameaxis. As shownby their position in Fig. 5, the most sensitive species to red mudpollution and by this to remarkable pH change appeared to be I.minor, P. cf. bohemica, L. tomosvaryi, S. aureus and Spatulosminthurus

flaviceps, whereas other species like B. parvula, M. pygmaea, F. can-dida showed tolerance to the increased pH.

4. Discussion

4.1. Soil properties

The red mud disaster caused considerable changes in soil phys-ical and chemical properties. As the increased soil hygroscopicityindicates in the polluted area, red mud modified the particle sizestructure, mostlyby raising clay fraction. This has also been provedby laboratory experiments, modelling the effects of red mud flood(Anton et al., 2012). Because of the high content of sodium hydrox-ide in the sludge, the soil became strongly alkaline and total

concentration of several trace metals increased considerably. The

potential toxicity of heavy metals in soils depends on not onlytheirconcentration but principallyon theirbioavailability, stronglyinfluenced by various soil properties including pH, organic matterand clay contents (Crommentuijn et al., 1997; Frische et al., 2002;Van Gestel and Koolhaas, 2004; Van Gestel, 2008). Usually, withincreased clay content, bioavailability and toxicity of metals arereduced (Van Gestel, 2008); while the opposite is true for salin-ity(Owojori, 2009). Although previous studies pointed out that forcertain soil organisms acute toxicity in soils polluted with sludgeswas primarily related to water-soluble elements (Domene et al.,2008, 2010); total concentrations of toxic heavy metals should notbe neglectedeither aspollutant uptakeby specific groupsofCollem-bola can also be more directly related to the solid phase of soil

(Vijver et al., 2001).In acidic soil environments, the availability and mobility of

metal ions are higher due to the chemical form in which these ionsare present in soil solutions (Reddy et al., 1995; Takc et al., 2009;Violante et al., 2010). From the obtained data, it is well reflected inthehigherEDTAFe,Mn andZncontentoftheacidcontrolsoilscom-pared to levels of the same elements in the red mud polluted soils.Ithastobenoted,thatredmud itself is a highlycomplex substance.Despite its heavy metal content and thus its potential adverseeffects, red mud has shown some encouraging results as heavymetal sorbent, decreasing the bioavailability of elements such asAs, Cd, Cu, Ni, Pb, Zn in polluted soils (Lombi et al., 2002; Bertocchiet al., 2006; Garau et al., 2007, 2011). Owing to its hyper-alkalinity,metals can be precipitated as hydroxides; and its high Al and Fe

content, in the form of oxides and hydroxides, can provide surfacesforthe adsorption of potentially toxic elements (Anton et al., 2012).The results clearly revealed that the pH increase caused by the redsludge flood was responsible for increasing of the residual frac-tion of heavy metals. Focusing on the pore-water hypothesis (VanGestel, 1997), at the time of the sampling, soil alkalinity and salt(Na) toxicity were presumably the most important factors deter-miningthestructureofCollembola communitiesintheareaaffectedby red mud pollution.

Soil pollution can affect Collembola directly and indirectly. Incase of the red mud disaster, the flood of the highly alkaline slurrycan be considered as primary direct toxic effect. The high sodiumcontent can induce osmotic stress in haemolymph (Witteveenet al., 1987) and can negatively affect collembolan reproduction

(Hutson, 1978; Domene et al., 2010). Further direct toxic effect islinked with uptake of polluted food, as observed in the digestivetract of several, mainly epigeic and hemiedaphic species collectedfrom the polluted area which clearly showed the presence of redsludgehemorganic humus mixture.

Indirect effects of soil pollutions are mainly associated withchanges in food resources (Rusek and Marshall, 2000). Numbersand diversity of microbial communities can decrease as a responseto pollution, which tends to precede changes in the collembolancommunity composition by affecting groups that primarily feed onsoil fungi and bacteria (Cole et al., 2001). Although soil microbialparameters were not investigated in this study, red mud does notnecessarily decrease bacterial populations; on the contrary,Garauet al. (2007) found significant increase in their number in red mudamended soil experiments.


8/11





Fig. 5. Ordination biplot of canonical correspondence analysis (CCA) with mean abundances ofCollembola species and selected soil variables (Som: soil organic mattercontent; hy: hygroscopicity; pH of H2O; Na content) in the polluted and control area (FRM: red mud polluted forest; FC: control forest; MRM: redmud polluted meadow; MC:control meadow). Collembola species name abbreviations consistof theinitial letter of thegenus name followed by thefirst three letters of species.

4.2. Collembola diversity and abundance

Soil pollution by a wide range of contaminants can affectCollembola species richness, diversity and total abundance, andcan determine the presence or absence and dominance of certainspecies (e.g. Filser et al., 1995; Kuznetsova and Potapov, 1997; Coleet al., 2001; Fountain and Hopkin, 2004a; Santamara et al., 2012).Among the most important factors influencing community struc-tures, changes in food resources and in competition patterns meritspecial mention (Hgvar, 1990; Rusek and Marshall, 2000; Gilletand Ponge, 2003), along with the direct toxicity effects alreadydiscussed.

As expected, the two surveyed habitat types (forest and grass-

land) have shown distinct differences in collembolan communitycharacteristics, and community responses to red mudpollution asa result also showed differences. The contrast between Collem-bolan communities in forest and open habitats is well documented(e.g. Pozo et al., 1986; Ponge, 1993; Setala et al., 1995; Pongeet al., 2003, 2008; Winkler and Tth, 2012); however, the unfor-tunate red mud accident has raised the important question as tohow do these different communities change in a forest-grasslandhabitat mosaic in response to the same pollution source. Higherspecies richness (both observed and estimates) was associatedwith the riverine forests versus meadows but after the pollu-tion accident it decreased in both habitats, especially in themeadows.

The Shannon and Simpson indices showed lower diversityin the forests, although the intersections observed between the

community diversity profiles (Fig. 2) of the sampled forests andmeadows indicated that open and forest areas contain very dif-ferent collembolan communities that are not comparable solelybased on diversity (Tthmrsz, 1997). Decrease in diversity wasfound in the polluted areas, both in forests and in meadows. Nev-ertheless, the diversity profiles of communities from the pairedcontrol-polluted habitats,especially concerningtheforest sites, runclose to each other, suggesting no fundamental change in diver-sity. Nonetheless, other studies have revealed that there is notnecessarily an evident relationship between Collembolan diver-sity and soil pollution and thus species diversity indices should beused with caution when evaluating toxicity (Van Straalen, 1997;Cortet et al., 1999; Fountain and Hopkin, 2004a,b; Santorufo et al.,

2012).Total community abundance changed differently in the openhabitat and in the forest. Its value dropped by 45% in the pollutedmeadows (8000ind./m2), while almost tripled in the pollutedforests reaching a value higher than 90,000ind./m2. The collecteddataoncommunity-levelabundanceinthepollutedforestsroughlycorrespond to those obtained by Gillet and Ponge (2003) frompoplar forest sites heavily polluted by metals. The low speciesrichness and abundance in the polluted meadow sites might beexplained, among other factors, by the different, less complexhabitat structure, and in consequence, by the lower patchiness insoil, which is essential for the survival of species in polluted sites(Bengtsson, 1997; Filser et al., 2000; Niklasson et al., 2000)

By analysing the species abundance distribution, marked dif-ference was observed only between the forest and meadow


9/11
http://localhost/var/www/apps/conversion/tmp/scratch_5/dx.doi.org/10.1016/j.apsoil.2013.07.006


10/11
http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0245http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0245http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0245http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0245http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0245http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0245http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0245http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0245http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0245http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0245http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0245http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0245http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0245http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0245http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0245http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0245http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0245http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0245http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0245http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0240http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0235http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0235http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0235http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0235http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0235http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0235http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0235http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0235http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0235http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0235http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0235http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0235http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0235http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0235http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0235http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0235http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0235http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0235http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0235http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0230http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0230http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0230http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0230http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0230http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0230http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0230http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0230http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0230http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0230http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0230http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0230http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0230http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0230http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0230http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0230http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0230http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0225http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0225http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0225http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0225http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0225http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0225http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0225http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0225http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0225http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0225http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0225http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0225http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0225http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0225http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0225http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0225http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0225http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0225http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0225http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0220http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0220http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0220http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0220http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0220http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0220http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0220http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0220http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0220http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0220http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0220http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0220http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0220http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0220http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0220http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0220http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0215http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0215http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0215http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0215http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0215http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0215http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0215http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0215http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0215http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0215http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0215http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0215http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0215http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0215http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0215http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0215http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0215http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0215http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0215http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0215http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0210http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0210http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0210http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0210http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0210http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0210http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0210http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0210http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0210http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0210http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0210http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0210http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0210http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0210http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0210http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0210http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0210http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0210http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0205http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0205http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0205http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0205http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0205http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0205http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0205http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0205http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0205http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0205http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0205http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0205http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0205http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0205http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0205http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0205http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0205http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0200http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0200http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0200http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0200http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0200http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0200http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0200http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0200http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0200http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0200http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0200http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0200http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0200http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0200http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0200http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0200http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0200http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0200http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0200http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0195http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0195http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0195http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0195http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0195http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0195http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0195http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0195http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0195http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0195http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0195http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0195http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0195http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0195http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0195http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0195http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0190http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0190http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0190http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0190http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0190http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0190http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0190http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0190http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0190http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0190http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0190http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0190http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0190http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0190http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0190http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0190http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0190http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0190http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0190http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0190http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0190http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0185http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0185http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0185http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0185http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0185http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0185http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0185http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0185http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0185http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0185http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0185http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0185http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0185http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0185http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0185http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0185http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0185http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0185http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0185http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0185http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0185http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0185http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0180http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0180http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0180http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0180http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0180http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0180http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0180http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0180http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0180http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0180http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0180http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0180http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0180http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0180http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0180http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0180http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0180http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0180http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0180http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0180http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0180http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0180http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0180http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0180http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0180http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0175http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0175http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0175http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0175http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0175http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0175http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0175http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0175http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0175http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0175http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0175http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0175http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0175http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0175http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0175http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0175http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0175http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0175http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0175http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0175http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0175http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0175http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0175http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0175http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0175http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0175http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0175http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0175http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0165http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0165http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0165http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0165http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0165http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0165http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0165http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0165http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0165http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0165http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0165http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0165http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0165http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0165http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0165http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0165http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0165http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0165http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0165http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0165http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0165http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0160http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0160http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0160http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0160http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0160http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0160http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0160http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0160http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0160http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0160http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0160http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0160http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0160http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0160http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0160http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0160http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0160http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0160http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0160http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0160http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0160http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0160http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0160http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0160http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0155http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0155http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0155http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0155http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0155http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0155http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0155http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0155http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0155http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0155http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0155http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0155http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0155http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0155http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0155http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0155http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0155http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0155http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0155http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0155http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0155http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0150http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0150http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0150http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0150http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0150http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0150http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0150http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0150http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0150http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0150http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0150http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0150http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0150http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0150http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0150http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0150http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0150http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0140http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0140http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0140http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0140http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0140http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0140http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0140http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0140http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0140http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0140http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0140http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0140http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0140http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0140http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0140http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0140http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0140http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0140http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0140http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0140http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0140http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0140http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0140http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0140http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0140http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0140http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0140http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0135http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0135http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0135http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0135http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0135http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0135http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0135http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0135http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0135http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0135http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0135http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0135http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0135http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0135http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0135http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0130http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0130http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0130http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0130http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0130http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0130http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0130http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0130http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0130http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0130http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0130http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0130http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0130http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0130http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0130http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0125http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0125http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0125http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0125http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0125http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0125http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0125http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0125http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0125http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0125http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0125http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0125http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0125http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0125http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0125http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0125http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0125http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0125http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0125http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0125http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0125http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0120http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0120http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0120http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0120http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0120http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0120http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0120http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0120http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0120http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0120http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0120http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0115http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0115http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0115http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0115http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0115http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0115http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0115http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0115http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0115http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0115http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0115http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0115http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0115http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0115http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0115http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0115http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0115http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0115http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0115http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0115http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0115http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0115http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0110http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0110http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0110http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0110http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0110http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0110http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0110http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0110http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0110http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0110http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0110http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0110http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0110http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0110http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0110http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0110http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0110http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0110http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0110http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0110http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0110http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0110http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0110http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0110http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0105http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0105http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0105http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0105http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0105http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0105http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0105http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0105http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0105http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0105http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0105http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0105http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0105http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0105http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0105http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0105http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0105http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0105http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0100http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0100http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0100http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0100http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0100http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0100http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0100http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0100http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0100http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0100http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0100http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0100http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0100http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0100http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0100http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0100http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0100http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0100http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0100http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0100http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0100http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0100http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0100http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0095http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0090http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0090http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0090http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0090http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0090http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0090http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0090http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0090http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0090http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0090http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0090http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0090http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0090http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0090http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0090http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0090http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0090http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0090http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0090http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0090http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0090http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0090http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0090http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0090http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0090http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0090http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0090http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0090http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0085http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0085http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0085http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0085http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0085http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0085http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0085http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0085http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0085http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0085http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0085http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0085http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0085http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0085http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0085http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0085http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0085http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0080http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0080http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0080http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0080http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0080http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0080http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0080http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0080http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0080http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0080http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0080http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0080http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0080http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0075http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0075http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0075http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0075http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0075http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0075http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0075http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0075http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0075http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0075http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0075http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0075http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0075http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0075http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0075http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0075http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0070http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0070http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0070http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0070http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0070http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0070http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0070http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0070http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0070http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0070http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0070http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0070http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0070http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0070http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0070http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0070http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0070http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0070http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0070http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0070http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0070http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0070http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0070http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0070http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0065http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0065http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0065http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0065http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0065http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0065http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0065http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0065http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0065http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0065http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0065http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0065http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0065http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0065http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0065http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0065http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0065http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0065http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0065http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0065http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0060http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0060http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0060http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0060http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0060http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0060http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0060http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0060http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0060http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0060http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0060http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0060http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0060http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0060http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0060http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0055http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0055http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0055http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0055http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0055http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0055http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0055http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0055http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0055http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0055http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0055http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0055http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0055http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0055http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0055http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0055http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0055http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0050http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0050http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0050http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0050http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0050http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0050http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0050http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0050http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0050http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0050http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0050http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0050http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0050http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0050http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0050http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0050http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0050http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0050http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0050http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0050http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0050http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0050http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0045http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0045http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0045http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0045http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0045http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0045http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0045http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0045http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0045http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0045http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0045http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0045http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0045http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0045http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0045http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0045http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0045http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0035http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0035http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0035http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0035http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0035http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0035http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0035http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0030http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0030http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0030http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0030http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0030http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0030http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0030http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0030http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0030http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0030http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0030http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0030http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0030http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0030http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0030http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0030http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0030http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0030http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0030http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0030http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0030http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0030http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0030http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0030http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0030http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0025http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0020http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0020http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0020http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0020http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0020http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0020http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0020http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0020http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0020http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0020http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0020http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0020http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0020http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0020http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0020http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0015http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0015http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0015http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0015http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0015http://refhub.elsevier.com/S0929-1393(13)00193-5/sbref0015http://refhub.elsevier.com/S0929-1393(13)00193-5/sbr

Documents

Artigo Folsomia Toxicidade