17. Poluarea Aerului Ozonul Indicatori_1999

8/3/2019 17. Poluarea Aerului Ozonul Indicatori_1999

1/58

Topic report No 3/1999

Air pollut ion by ozone in Europein 1997 and summer 1998

Prepared by:Frank de Leeuw, Rob Sluyter and Tim de Paus

Project manager:Gabriel Kielland

European Environment Agency


2/58

Cover design: Rolf Kuch ling, EEA

Legal notice

The contents of this report do not necessarily reflect the official opinion of the European

Communities or other European Communities institutions. Neither the European Environment

Agency nor an y person or com pan y acting on th e beh alf of the Agency is responsible for th e use that

may be made of the information contained in this report.

A great deal of additional information on the Euro pean Union is available on th e Intern et.

It can be accessed throu gh the Europa server (h ttp:/ / europa.eu.int)

EEA, Copenhagen, 1999

Reprodu ction is authorised pro vided th e source is acknowledged

Printed in Italy

Printed on recycled and chlorin e-free bleached pap er

ISBN 92-9167-123-1

European Environment Agency

Kon gens Nytorv 6

DK-1050 Copenhagen K

Denmark

Tel: +45 33 36 71 00

Fax: +45 33 36 71 99

E-mail: ee [email protected]


3/58

3

Table of contents

Content s............................................................................................. 3

Preface ............................................................................................. 4

Summary............................................................................................. 4

Part I Exceedance of ec ozone threshold values in Europe in 1997. 7

1. Introduction ................................................................................ 8

2. Data report ing .......................................................................... 112.1. Introduction................................................................................................11

2.2. Data handling................................................................................. 12

3. Survey of reported data for 1997............................................. 15

4. Discussion ................................................................................. 224.1. Geographical coverage of monitoring stations.............................. 224.2. Annual statistics, 1997 ................................................................... 264.3. Exceedances of thresholds in 1997................................................294.3.1.Exceedances of the threshold value for protection of human health ........294.3.2.Exceedances of the threshold values for information and warning

of the population........................................................................................344.3.3.Exceedances of the daily threshold value for protection of vegetation ....374.3.4.Exceedance of the hourly threshold value for protection of vegetation....40

4.4. Preliminary review of exceedances of the revised guidelines ofWHO .............................................................................................. 42

4.5. Data reported for 1994-1997......................................................... 445. Conclusions and recommendations........................................... 47

6. References................................................................................ 49

Annex 1: Observed exceedances and annual stat ist ics ..................... 51

Annex 2: Phenomenology of ozone concentrat ions .......................... 56

Part II Information document concerning air pol lution by ozone ... 59

1. Int roduction .............................................................................. 60

2. Availabilit y of data.................................................................... 62

3. Summary of data reported for summer 1998............................ 653.1. Geographical distribution .............................................................. 693.2. Comparison with earlier years........................................................ 73

4. Main ozone episodes................................................................ 75

5. Conclusions............................................................................... 79

6. References................................................................................ 80


4/58

4

Preface

This report is based upon data submitted to the European Commission un der

the Directive 92/ 72/ EEC on air pollution by ozone. The Commission hasrequested th e European Environmen t Agency (EEA) to assist in this annu alrep orting by EU Member States since 1995. The volun tary submission to th eEEA of information on ozone levels by other European coun tries made itpossible to presen t an ozone assessmen t on a wider European scale. The datacollection in Phare coun tries was greatly facilitated by the Ph are Topic Link onAir Quality working as part of the European Topic Centre on Air Quality. Twoseparate reports, one covering the year 1997, the o ther covering summer 1998,were originally produced by the European Topic Centr e on Air Quality andpresented to the Member States in their meeting of November 1998. Theseparate reports are available in electronic form on the EEA hom epage on the

Internet (http:/ / eea.eu.int).

In th is EEA Topic Report, the annual 1997 and summer 1998 reports arereprodu ced in the ir original form as presented to the Commission. Thedifferences in available data, the set of threshold values, the nu mber ofstations, the location of stations and the status of the information ( based onvalidated 1997 data and non-validated 1998 data) justifies such an approach .

A major asset of this report is the timeliness of its de livery. The assessment ofozone episodes in 1998 was based up on data measured on ly two month searlier, while the n ormal production time from field measuremen ts to

validated assessmen t reports is rarely less than eighteen mon ths. This timelyrep orting has only been possible with the suppor t of the individual contactpoints within each cou ntry and the efficient comm un ication established.

The harm ful effects of tropospher ic ozone on human health an d well-being aswell as damage to ecosystems is now being recognised as a major concernthroughout the European Union . The European Community has taken stepsto address the p roblem through Directive 92/ 72/ EEC on ambient ozone ,Directive 96/ 62/ EEC on ambient air quality assessmen t and management (th eFramework Directive) , and developmen t of an ozone daughter Directive, aswell as the decision to develop a Community strategy for the redu ction of

ozone pollution. The measures necessary to abate pollution remain however arespon sibility of each Member State an d require political decisions with costimplications and consequences for the developm ent of activities in the society.In th is political process objective and re liable information on the exten t andseverity of the problem is essential.

It is the inten tion of EEA to con tinue the yearly reporting and assessmen t ofground level ozone in Europe in close co-operation with the EuropeanCommission, Member States and other countries.

Gordon McInn es

Programme manager


5/58

5

Summary

This report summarises the an nu al information on ozone mon itoring

stations and exceedances of ozone th reshold values dur ing 1997 and gives afirst evaluation of the observed exceedan ces of the th resholds duringSummer 1998 (April-August) . According to the Cou ncil Directive92/ 72/ EEC on air pollution by ozone, EU Member States have to provideinformation on ozone levels (statistical parameters, number and durationof exceedances of specified th reshold values) on an an nual basis before 1July of the n ext year. Additionally, exceedances of the threshold values forpopu lation information an d warn ing, as set in the Directive, must bereported to the Commission within one m onth after occurrence.

The analysis for the year 1997 presented in th is report is based oninformation made available before 12 August 1998. By then , informationfor the calendar year 1997 was received from all Member States and, on avolun tary basis, from 5 o ther European coun tries (Switzerland, CzechRepublic, Latvia, Norway and Poland) . Information on the situation inSlovakia was received one month later; unfortun ately too late to includefully in th is report. All information h as been submitted in computerreadable form.

For the summer 1998-period, all 15 EU Member States providedinformation on the observed exceedan ces in time (th e dead line for

transmitting data was set at 20 September 1998), or indicated that n oexceedances were observed. It is greatly appreciated by the Commissionthat Member States were able to transmit August exceedan ce data beforethe formal deadline as set in th e Directive. Some coun tries submitted fileswhich were not formatted according to the prescribed Commissionrequirements. These files by exception were converted at the EuropeanTopic Centre on Air Quality (ETC-AQ) for furth er p rocessing.

From an evaluation of the exceedan ces and an nual statistics, the followingconclusions are drawn:

In 1997 the th reshold value set for the protection of human h ealth (110g/ m 3 for 8-hourly average concentrations) was exceeded substantiallyand in all reporting coun tries: the total number of exceedancesreported for stations within in th e EU approached 20,000; on theaverage, th is threshold was exceeded p er station on more th an 23 daysper year.

In 1997 the threshold value of daily average con centrations set for theprotection of vegetation (65 g/ m 3) was exceeded substantially (by upto a factor 3) , widely ( in all reporting coun tries) and frequ ently: at 52

stations located in 10 different coun tries exceedances have beenreported during more than 200 days.


6/58

6

The threshold value of hourly average concentrations set for theprotection of vegetation

(200 g/ m3) was exceeded largely and widely (reported by 10 EUMember States and in three ou t of the six other Europ ean countr ies)

on a limited number of days ( in total 909 exceedance days werecounted at EU monitoring stations).

In 1997 the threshold value for providing information to the popu lation(180 g/ m 3 for hourly values) was exceeded in 15 countries of which 12EU Member States during a limited number of days: in EU MemberStates a total of 2336 exceedance days was counted. During Summer1998 this threshold was exceeded in all Member States with th eexception of Ireland, Finland , Denmark and Sweden . The number ofdays on which at least on e exceed ance was observed ran ged from 7 inBelgium to 74 in Spain. 61 % of all stations reported on e or more

exceedance. On average 3.9 exceedan ces occurr ed th is year at stationswhich recorded at least on e exceedance; the average exceedanceduration was 2.7 hours. The average maximum hourly concentrationduring an exceedan ce of the threshold this year was 201 g/ m

3.

Exceedance of the threshold value for warning of the population wasreported in 1997 from one station ( on 18 Jun e 1997, 13.00 an hourly

ozone concen tration of 383 g/ m 3 was measured at Lykovrissi, Athens,Greece). The threshold was exceeded on three days during summer1998 in the Athens conurbation (2 and 3 July 1998 at 2 stations, 29 July

1998 at 1 station) and at 1 station on 7 August 1998 in France.

Ozone monitoring data for the year 1997 were r eceived from 984stations within th e EU and from 100 stations in o ther countries; du ringSummer 1998 a comparable number of stations was operational.

Spatial coverage and documentation on mon itoring data quality needimprovemen t. Depen ding on the local situation , the ozone mon itoringstations are characterised as rural, urban, street or other (e.g.industrial) . The p resent subset of rural stations is no t representative for

the land area of the EU: the subset is estimated to cover about 40-50%of the area. The geograph ical coverage of the rural stations is ratheradequate in Nor th West Europe but in other regions gaps are noted.

A limited presentation of the percentile values observed in the period1989-1997 is given for four Member States for which this informationwas available. Based on the reported data n o firm conclusionconcern ing a tren d in percentile values can be given.

DISCLAIMER

The information presented in part II,Information document concerning air pollution by

ozone, is partly based on non -validated mon itoring data and hen ce shou ld beregarded as preliminary


7/58


8/58

8

1. Introduction

Ozone, O3, is a stron g ph otochemical oxidant which causes serious health

problems and d amage to materials and ecosystems. Human exposure toelevated levels of ozone concentr ations can give r ise to inflammatoryresponses and decreases in lung function. Symptoms observed are cough,chest pain, d ifficulty in breath ing, headache and eye irritation. Bothlaboratory and epidemiological data indicate large variations betweenindividuals in r esponse to ep isodic ozone exposure, the effects seem to bemore pron oun ced in children than in adults. Studies indicate that

exposure to ozone concentrations in the ran ge 160-360 g/ m 3 for a p eriodof 1-8 hou rs - concentr ations often observed in ambient air over Europe -red uces various pulmonary functions.

Ozone exposure of ecosystems and agricultural crops results in visible foliarinjury and in red uctions in crop yield and seed prod uction. For vegetationa long-term, growing season averaged exposure rath er th an a ep isodicexposure is of concern. Adverse effects on vegetation can be noted atrelatively low ozone levels. Within the fram ework of th e UN-ECEConven tion on Lon g-Range Transboundary Air Pollution the critical level

1

for ozone is expressed as the accumulated ozone exposure above a

threshold of 40 ppb (corr espon ding with 80 g/ m 3) . Guideline values ofthis accumu lated ozone exposure of 3000 ppb.h an d 10,000 ppb.h aregiven for crop s and forest, respectively. The World Health Organization in

Europe ( WHO ) came forward with similar guidelines (WHO, 1996b).

It is known that ozone affects materials such as natural and syntheticrubbers, coatings and textiles. However, th ere are today serious gaps inknowledge on the mechan isms of damage, the attribution of ozone todamage in comp arison to other factors and th e economic evaluation ofsuch damage. As far as un derstood there is no no-effect level of ozone formaterial corrosion ; it is assumed that dose-respon se relations for materialsare linear or n early linear und er ambient conditions. Recently, synergisticeffects of ozone in combination with th e acidifying components SO

2and

NO 2 have been reported to lead to increased corrosion on buildingmaterials like steel, zinc, copper, aluminium and bronze.

In view of the harmful effects of photoch emical pollution in th e lowerlevels of the atm osphere, the Council adopted in 1992 the Directive92/ 72/ EEC on air pollution by ozone. The Directive came into force inMarch 1994. It established procedures for h armon ised mon itoring ofozone con centrations, for exchan ge of information, for comm un icationwith and alerting of the population regarding ozone and to optimise theaction needed to reduce ozone formation.

1

Critical levels are defined as concentrations of pollutants in t he atmosphere above which direct adverse

effects on receptors, such as plants, ecosystems or materials may occur according to present knowledge

(Bull, 1991).


9/58

9

Article 6 of the Directive specifies how the information on mon itoringresults must be p rovided by the Member States to th e Comm ission.Regarding the time frame, two main types of reporting can bedistinguished. Information on exceedances of the so-called informationthreshold (article 6 sub 2) an d warning th reshold (article 6 sub 3) for th eozone concen tration is to be provided within one mon th after occurrence.Information on exceedances of all threshold values given in Article 6 mu stbe provided within six months following the annual reference period(article 6 sub 1). Article 7 of the Directive stipulates that the Commissionshall at least on ce a year evaluate th e data collected un der th e Directive.The present rep ort gives an overview of ozone mon itoring results of 1997.Similar overviews of the 1994, 1995 and 1996 annual data have beenprep ared by the European Topic Centre on Air Quality (d e Leeuw et al.,1995; 1996; 1997). Prior to th e curren t report an overview on ozonethreshold exceedances during Summer 1997 (April-August) has been

presented to the Commission (de Leeuw et al., 1997).

Harmful ozone con centrations are observed over th e whole of Europe.Formation of ozone takes place at various space an d time scales: the h ighemission density of reactive precursors in urban ised areas might lead tohigh ozone levels within the city or at shor t d istances downwind. But ozoneprecursors may also be transported over distances of hundreds tothousands kilometres, resulting in ozone formation far from the sources.For improving the insight in curr ent ambient ozone levels over Europe,countries outside the European Un ion have been asked by the EuropeanEnvironment Agency (EEA) to provide information on ozone exceedan cesin line with th e O zone Directive.

The d ata reported here do n ot cover all ozone mon itoring stations in theEuropean Union. For inclusion in th is report, the data must satisfy certaincriteria stipulated in the Directive, concerning inter alia measuringmeth ods, sampling meth ods, station siting, quality assurance anddocumentation. Formats on the transfer of data have been defined by theExper t Group on Photochemical Pollution. Th is group , established by theCommission following Article 7 of the Directive, had several meetings to co-ordinate the work within the Member States and the Commission in the

framework of the Directive.

Backgroun d information on th e curren t experience and knowledgeconcern ing ph otochemical air pollution, dealing in particular with theph enomen ology of ozone, the scientific understand ing as based onexperimen ts and theory, and the insights from modelling studies on th erelation between ozone levels and precursor emissions, may be foun d inBorrell and Van den Hou t (1995), Derwent and Van den Hou t (1995),Barrett and Berge (1996) and in Tropospheric ozone in the European Union.The consolidated report(Becket al., 1998) which is curren tly in preparation

following Article 8 of the Directive. In the process of prepar ing the newOzone Daughter Directive, a comprehensive review of almost all aspectsconcern ing ozone pollution is curren tly prep ared by an expert groupestablished by the European Commission. This so-called Position Paper on


10/58

10

Ozone will be published in 1999. Some aspects of the ozoneph enomenology are briefly discussed in Annex II.

In th e very near futu re a n ew ozone daughter directive will be pu t forwardby the Commission. In this new proposal long-term objectives and targetvalues for the protection of human health and vegetation will be definedbased on the revised gu idelines of WHO (WHO , 1996). The target valuesshould be attained, as far as possible, in 2010. In addition, the draft ozonedirective proposes alert thresholds for informing sensitive and gen eralinformation . Whereas for humans an exposure period 8 hours isconsidered, for the protection of vegetation the relevant p eriod extendsover th e whole growing season (th ree m onths). As a ru le of thu mb onemight say that the targets for protection of vegetation will be more difficultto meet than the h uman health related targets.

While the revised WHO guidelines are also relevant in the con text ofpreparing the new daugh ter directive for ozone an attemp t has been madeto evaluate the ozone data submitted und er the framework of the curren tdirective against these guideline levels, see section 4.5.


11/58

11

2. Data report ing

2.1. Introduction

According to th e Ozone Directive, the EU Member States have to providethe following information for th e annu al reference period:

maximum, median and 98 percen tile value of 1h- and 8h- averageozone concen trations;

the n umber, date and duration of periods during which thresholdvalues as presented in Table 1 are exceeded and the maximumconcentrations recorded during each occurren ce.

In addition to th is ann ual repor t, Member States must inform the

Commission on a mon thly basis in case exceedances of the informationand warn ing threshold values are observed. In this repor t only datareported on an ann ual basis will be considered.

Table 1: Threshold values for ozone concentrations (in g/m3)

Threshold for: Concentrat ion Averagingperiod

health protectionvegetation protection

population information

population warning

11020065180

360

8 h1 h24 h1 h

1 h

A group of experts from th e Members States have followed the practicalimplementation of the Directive. Among other items th is group d iscussedprocedures for data reporting. The formats for information and dataexchanges have been defined in th e documen t Coun cil Directive92/ 72/ EEC on air p ollution by ozone. Information and dataexchange/ formats, Doc.Rev. 11/ 243/ 95. In general terms, the requestedinformation consists of two parts:

1. information on stations and measuremen ts techn iques (O zoneDirective, article 4.2, inden ts 1 and 2);2. information on ozone concentration: annual statistics and threshold

exceedances (Ozone Directive, article 6.1).

Based on the experiences in processing the data for the 1994 annualreport, the European Topic Centre on Air Quality (ETC-AQ) providedremarks concerning data transmission and suggestions for improvementwhich were discussed in th e Expert Group on Photochemical Pollution.Considering the increasing amount of data requ iring processing, as well asthe improvemen t of the transfer of data relating to th e implementation of

the Directive, the Commission h as prepared an update ( April 1996) of thedata exchange format. The major changes concern the transfer ofadditional information:


12/58

12

type of station: definition of the location of stations as recommen ded inthe d ecision on Exchanges of Information (97/ 101/ EC);

altitude of stations as recommended by the Exper t Group ; NO

xand VOC data, according to Annex 2.3 of the ozone directive;

file names: it is recommend ed to define un ique n ames for all files inorder to improve the man agement an d tran sfer of the d ata files.

For submission of the 1997 data no furth er modifications in d atarequ irements and data exchange formats have been made. Non-EUcoun tries have been requested to submit their data in agreement with thesedata exchan ge formats.

2.2. Data handling

According to the Ozone Directive the requested information over 1997

should h ave been forwarded to th e Commission before 1 July 1998. All dataforwarded by the Commission an d received at the European Topic Centreon Air Qu ality (ETC-AQ) before 12 August 1998 has been processed is thisreport. In th is report the definition of the data formats as described in th edocument Doc.Rev. 11/ 243/ 95 and its April 1996 amend men t is used as areference. If necessary, files were converted at the ETC-AQ for furtherprocessing.

Non-EU Member States were contacted by the European EnvironmentAgency, through th e Phare Topic Link on Air Quality or, in case ofSwitzerland, directly and asked to voluntarily submit data following thestandard formats on data exchange. Information was received fromSwitzerland, the Czech Republic, Latvia, Norway and Poland

2.

The summary of received information presented in Table 2 shows that theinformation for the EU is nearly complete. However, analysing the datarevealed a n umber of errors, ambiguities or missing data. It was attemptedto con tact nation al data suppliers to clarify these points but generally thiswas not successful. Lack of e-mail addresses or fax-numbers hampered arapid d ialogue.

Spain has not submitted statistical information and on ly submitted anincomplete set of exceedances.

The data received from Italy was of a low quality. Typical errors frequentlyfound in the Italian information are:

incomplete set of station information (co-ordinates, un ique coding,station type, etc. is missing);

incomplete date of exceedan ce (month but n ot day or h our is given);

2

At a very late moment (18 September 1998) data from 8 stations in Slovakia was received. Although the data

was delivered according to t he agreed formats, the limited time and resources did not allow to fully include the

data in t his report. Data from Slovakia are included only in the overview tables 2-5 presented in Chapter 2 and

3.


13/58

13

inconsistent information on exceedances of the 180 g/ m 3 and 200g/ m

3threshold: for a number of stations, exceedances of the 200 g/ m 3

threshold are not qu oted as exceedance of the 180 g/ m 3 threshold; there are clear indications that for a number of stations, in case of

exceedance of the daily threshold value of 65 g/ m3 the quotedmaximum con centration r efers to an hourly averaged concentration instead o f the daily mean .

The data from France contain inconsistent information on exceedances of

the 180 g/ m 3 and 200g/ m 3 threshold: for a number of stations,exceedances of the 200 g/ m 3 threshold are not quoted as exceedance ofthe 180 g/ m 3 threshold.

The statistical information of the Netherlands was in an incorrect format;

information on data coverage was not given. This precluded the further useof the Dutch statistical information in the an alysis.

A measurement techniques file was not received from a number of coun tries.Voluntary data on precursor concentrations was received for nine MemberStates and two additional countries.

For 1997 information on ozone concen trations (an nu al statistics and / orexceedance information) was received for 1076mon itoring stations ofwhich 984 were located in EU Member States. Figure 1 indicates that th enumber of reporting stations is still growing. The large increase compared

to 1996 is mainly due to the fact that over 1996 only 13 Member Statessubmitted information .

Figure 1: Number of stations report ing ozone data wit hin the framework ofthe ozone directive.

0

200

400

600

800

1000

1200

94 95 96 97

numberofstations

non-EU other/unspecstreet urbanrural

Note that 1996 refers to the actual reporting Member States. For stations in non-EU countries a station

classification is known but not resolved in this figure to maintain consistency with earlier years for EU Member

States.


14/58

14

Table 2: Summary of information received for 1997

count ry stat ioninfo

meas.tech

stat ist ics excee-dance

NO2

NOx

VOC

AT 9 9 9 9 8 8 8

BE 9 8 9 9 9 8 9DE 9 8 9 9 8 8 8

DK 9 9 9 9 9 9 8

ES 9(4)

9(4)

8 9(3)

8 8 8

FI 9 9 9 9 9 9 8

FR 9 9 9 9 8 8 8

GB 9 9 9 9 9 9 8.

GR 9 9 9 9 9 9(6)

8

IE 9 9 9 9 8 8 8

IT 9 9 9 9 9 9 9

LU 9 9 9 9 9 9 9

NL 9 8 9(1) 9(2) 9(1) 9(1) 8PT 9 9 9 9 9 9 8

SE 9 9 9 9 8 8 8

CH 9 9 9 9 9 9 8

CZ 9 9 9 9 9 9 9

LV 8 8 9(5)

9(5)

8 8 8

NO 9 8 9 9 8 8 8

PO 9 8 9 9 8 8 8

SK 9 9 9 9 8 8 89: delivered;

8: not delivered;

(1) wrong format;(2) originally data in wrong format; updated file received 18 August 1998;

(3) only period March-October; only exceedances of 180 g/ m3 (1h-averaged);(4) no 1997-update received; information submitted for 1996 has been used;

(5) hourly data submitted for period April-December 1997; exceedances and statistics calculated by ETC/AQ;

(6) Information on NO2and NO (not NOx) has been received.


15/58

15

3. Survey of report ed data for 1997

The location of monitoring stations in EU Member States which are used

for th e implemen tation of the Ozone Directive and which are reportingover 1997, is presented in Map 1. Stations located in oth er Europeancoun tries which have reported ozone data on a volun tary basis are shown aswell. In total information for 984 stations in 15 Member States has beenreceived; for 100 stations in the six reporting non-EU coun triesinformation has been received. All reporting coun tries use the referencemeth od ( UV absorption ) as prescribed in Annex V of the Ozone Directive.However, at a limited number o f stations chemiluminescence is used.

In 1997 only one exceedance of the threshold value of 360 g/ m 3 forhourly values has been reported

3: on 18 Jun e, 13.00 a hourly ozone

concentration of 383 g/ m 3 was measured at Lykovrissi (Athens, Greece) .This day was the start of an ozone episode in Athens. The station Lykovrissi

reported exceedan ces of the hu man health related th reshold value of 110 g/ m

3for 8-hourly averaged during th e period 18-24 Jun e. During the whole

mon th of Jun e, concen trations were high in Athens: on 17 days at least one

(out of seven) stations reported an hourly value exceeding 180 g/ m 3. On22 Jun e all seven stations reported an exceedance with an maximum of 229

g/ m 3. For a more extensive description o f this episode see de Leeuw et al.(1997).

A summ ary of the m aximum con centration measured at any of thereporting stations where exceedan ce

4of a th reshold value is observed is

presented in Table 3. When no exceedances of a threshold have beenreported by a coun try this is indicated with a dash (-) .

3

On 20 June 1997 the Italian station Via Venezia in Caltanissetta (Sicilia) reported an exceedance of the

threshold value of 200 g/ m3 during two hours; as maximum concentration during this exceedance a value of

379 g/ m3

was reported. This value has not been reported as an exceedance of the 360 g/ m3

level. Accordingto information from the Italian data supplier, this value is in error and should be disregarded; another

exceedance, observed at nighttime in the winter at the same station, should be disregarded as well (Mr. P.

Lagrotta, personal communication).4

In this report exceedances are counted on a daily basis, that is, a day on which at least one 1h- or 8h-

concentration exceeds the threshold value, is marked as an exceedance.


16/58

16

Map 1: Location of ozone monitoring stations as reported by MemberStates in the framework of the Ozone Direct ive for the referenceperiod 1997. Stations for which other European countriessubmitted information are shown as well.


17/58

17

Table 3: Maximum ozone concentrations (in g/ m3) measured during aperiod of exceedance of t hreshold values (reference period 1January - 31 December 1997). A dash () indicates that noexceedances have been observed; na indicates that no data areavailable.

180(1h)

200(1h)

360(1h)

110(8h-

(a))

110(8h-

(b))

65(24h)

AT 228 228 166 171 153

BE 277 277 205 219 171

DE 253 253 221 215 214

DK 181 141 155 133

ES 279 279(1)

na na na

FI 147 136 119

FR 199(2)

270 184 210 170

GB 314 314 188 195 140

GR 383 383 383 221 246 138IE 151 131 133

IT 353 353 207 230 353(3)

LU 203 203 166 174 148

NL 266 266 183 199 121

PT 271 271 204 230 174

SE 141 144 128

EU-15 383 383 383 221 246 353(3)

CH 274 274 182 208 166

CZ 200 200 177 182 148

LV 115 127 102

NO 162 162 162

PL 222 222 163 168 135

SK 135 149 122

all 383 383 383 221 246 353(3)

(a) based on three non-overlapping eight hourly values between 0.00-8.00; 8.00-16.00; 16.00-24.00;

(b) based on the eight hourly value between 12.00 - 20.00;

(1) exceedances of the threshold of 200 g/ m3 have not been report ed; a maximum concentration of279 g/ m3 is deduced from the information on exceedances of the threshold of 180 g/ m3;

(2) for a number of stations exceedances of the 200 g/ m3 threshold have not been reported asexceedances of the 180 g/ m3 threshold;

(3) there are clear indications that the maximum concentration refers to an hourly, not daily averaged value.

As the n umber of mon itoring stations differs widely from coun try tocoun try, the absolute n umber of exceedan ces is less suitable forcomparison. Therefore, the concept of occurren ce of exceedances isintroduced here. Occurren ce of exceedan ces is defined as the averagenu mber of observed exceedances per country, that is, the total number ofexceedan ces summed over all the stations of a coun try divided by the totalnumber of reporting stations. A summary of occurrence of exceedances ispresented in Table 4. Still, Table 4 has to be interpreted carefully as thereare add itional reasons which h amper a comp arison between the countries.Firstly, the local environmen t ( in particular NO

xsources) influen ce the

ozone levels; the d ifferences between coun tries partly result from th edifferences in the ratios of street, urban an d rural stations. Secondly, about20% of the reporting stations show a data coverage of less than 75% (ordata coverage is not known); this might result in an u nderestimation of the


18/58

18

number of exceedances. Also the differen ces in definition of anexceedance ( some coun tries coun t an exceedan ce when the concentrationis GREATER THAN the threshold value, other coun tries count it when th econcentrations is EQUAL TO O RGREATER THAN the threshold) mayintroduce differen ce in occurren ce of exceedances.

Adverse effects of ozone on human health an d vegetation will not onlydepend on the frequency by which a threshold is violated but also on theseverity of the exceedance. The severity of an exceedan ce can be expressedby theAccumulated exposure Over a Threshold (AOT). AOT-values arecalculated on basis on hourly concentration according to:

( ) AOT C T ii

N

i=

where i = 0 for con centration Ci

less than threshold Tan d i = 1 when Ci

exceeds the th reshold T; Nis the nu mber of valid measurements. The AOTis expressed in concentration x time un its, for example inppm.h or mg.h.m

-3.

From the received information it is no t possible to calculate an AOT-levelas during a period of exceedance on ly the m aximum concentration isgiven . As an app roximation of the exposure we can calculate an exceedancerate which is defined as:

( ) ER C T d

AVi

i

NET

= +

,max 12

where th e summation is now over th e total number of exceedan ceNET anddis the duration of the exceedan ce expressed in nu mber of averagingperiods,AV. Similar to AOT, the ER is expressed in concentration x timeun its, for example inppm.h or mg.h.m

-3. In Table 4 next to the occurrence

of exceedan ces, the exceedan ce rate averaged over th e number ofexceedances, is given. Examination of this table shows some - but certainlyno perfect - correlation between the n umber of exceedances and theseverity of exceedance. Th e data suggests that in southern coun tries(Greece, Italy, Portugal) th e con centrations are gen erally much higherduring an exceedance of the information threshold than in the n ortherncoun tries. The steep fall-off in the n umber of exceedan ces of levels of 180,

200 and 240 g/ m 3 in n orthern countries compared to the more gradualdecrease in southern coun tries as shown in Figure 2, leads to a similarconclusion.

For each of the coun tries the lowest and highest 50-, 98- and 99.9-percen tilevalues observed at ind ividual stations are presented in Table 5. In this tablealso information on the maximum values is included. Note that th emaximum 8-hourly concentration , as reported in Table 5 is based on amoving average an d may therefore d iffer from th e values in Table 3 whichare based on fixed 8-hourly periods. The Czech Republic has not submitted

percen tile values for m oving eight-hourly concentr ations; Spain hassubmitted no statistical information at all. Statistics for Latvia and theNetherlands are not included in the table. In case of Latvia, the datacoverage for the submitting station is below the requ ired 75%; this low data


19/58

19

coverage is largely caused by missing data for the per iod 1 Janu ary - midApril. The information on m aximum concen trations and on exceedanceswill therefore be r eliable. In the case of the Netherlands no information ondata coverage has been submitted and therefore the validity of thepercentile values can n ot be judged. Full details on percen tile values andthe nu mber of exceedan ces at the individual stations is presented in TablesI.3 of Annex I

5.

Figure 2: Relative change in exceedances for the threshold values of 180,

200 and 240 g/ m3. The number of exceedances of 180 g/ m3 isset to unity; reference period 1 January - 31 December 1997. For

Spain no informat ion on exceedances of 200 g/ m3 has beensubmit ted. For all countries, the number of exceedances of 240 g/m

3has been extracted f rom t he informat ion on exceedances of

the threshold of 180 g/m3.

Relative change in exceedances for dif ferent

averaged over all

0.0

0.2

0.4

0.6

0.8

1.0

AT BE CH CZ DE DK ES FI FR GB GR IE IT LU LV NL NO PL PT SE

180 ug/m3

200 ug/m3

240 ug/m3

As ozone con centrations show a high au to-correlation, the corre lationbetween 1-h and 8-h percen tile values is no surpr ise. Median values for

hourly and moving 8-hourly concentrations are very similar. Th e 98-percentile for 8-h values is generally abou t 10% lower than thecorresponding 1-h value although this ratio is slightly different for th evarious types of stations and h as also a meteorological depend ence.Detailed information on the percentile values and maximumconcentrations observed at individual monitoring stations is given in Ann exI, Table I.3.

For a large number of countries additional statistical information on NOx,

NO2

and VOC concentrations has been received. This information ispr imarily used here for th e classification of stations. Being precursors ofozone, information on ambient levels of NO

xand VOC is essen tial to

5

Annex I is only available in computer readable form from the ETC-AQ web-site (URL:

http ://www.etcaq.rivm.nl). In this report a description of the tables in Annex I is provided.


20/58

20

evaluate th e effectiveness of ozone abatement strategy. The informationvolun tary submitted in th e framework of the O zone Directive is, however,not sufficien t for such an evaluation. Th e limited time series (four years orless) and generally lack of information on the local environment of thestations, hampers an an alysis of a possible trend in precursor NO

xand VOC

emissions. As the reported NO2

and O3

concentrations are not measuredsimultaneously, it is not possible to improve the insight in spatial variabilityof ozone concentration based on mapp ing of oxidan t (sum of NO

2an d

ozone) concentrations. Oxidant concentrations are representative for alarger area as oxidant is less depen den t on local condition andmeteorological cond itions than either ozone or NO

2(see also Annex II).

An overview of the reported NOx, NO

2and VOC con centrations is given in

Table I.4 and I.5 of Annex I.

Table 4: Occurrence of exceedances (OoE; in days) and average

exceedance rate (ER; in (g.h)/m3 for the threshold value of 180 g/ m

3and in (mg.h)/m

3for the threshold values of 110 and 65

g/ m3); reference period 1 January - 31 December 1997. na = no

data available.

Threshold value (in g/m3)number of 180-1h 110-8h(a) 65-24hstations (b) OoE ER OoE ER OoE ER

AT 111 0.1 2.7 30.5 2.9 119 55.4BE 25 3.2 151.7 23.1 4.7 50.4 19.4DE 373 1.4 52.2 25.5 3.4 64.0 24.2

DK 5 0.2 0.3 8.4 0.9 78.4 18.3ES 36 4.2 141.9 na na na naFI 11 0.0 0.0 6.9 0.5 128 42.7FR 151 1.2 13.1 21.9 2.8 64.6 24.4GB 71 0.7 27.5 6.5 1.0 33.0 9.1GR 13 10.1 710.1 32.0 5.2 64.6 22.7IE 6 0 0.0 2.7 0.3 73.3 20.7IT 121 12.3 863.9 33.8 5.4 68.0 51.3LU 5 1.0 17.2 26.2 4.7 70.4 34.2NL 39 1.2 44.1 14.5 2.4 31.5 10.2PT 11 3.1 160.9 22.8 4.1 52.8 23.1SE 6 0 0 9.5 0.9 166 56.0

EU-15 984 2.6 136 23.2 3.1 60.3 24.8

CH 13 3.7 130 58.5 9.3 135 67.2CZ 48 0.2 3.4 28.2 2.9 112 46.9LV 1 0 0 6.0 0.5 55.0 12.5NO 14 0 0 6.7 1.1 147 97.3PL 16 0.5 11.7 15.9 1.6 79.1 29.0SK 8 0 0 4.8 0.3 30.3 7.6(a) based on the eight hourly value between 12.00-20.00.

(b) note that d ifferences in the number of stations reporting for each of t he threshold levels may occur.


21/58

21

Table 5: Range in reported 50-, 98- and 99.9-percent ile values andmaximum observed values (based on hourly and moving eight -hourly average concentrations, period : 1997) observed at

individual monit oring stat ions in repor t ing countries (in g/ m3); na= range in percentile values not available;? = range in percent ile values can not be calculated as datacoverage is below 75%, or information on data coverage islacking; * = addit ional information submit ted on a voluntary basis.

1997 1h-P50 1h-P98 1h-P99.9* 1h-MAXrange min max min max min max min maxAT 15 97 90 143 117 176 124 228BE 18 56 96 154 150 203 166 277DE 8 102 63 193 99 225 109 253DK 44 62 90 117 116 153 124 181ES na na na na na na na naFI 35 74 86 112 106 140 118 179FR 11 97 82 164 111 214 26 270GB 10 64 68 124 102 194 74 314GR 18 56 82 167 117 266 137 383IE 40 66 82 106 na na 114 173IT 11 86 51 186 98 237 121 271LU 20 62 79 146 115 182 131 203NL ? ? ? ? ? ? 67 266PT 10 65 44 157 63 177 77 271SE 59 69 96 116 114 147 118 169

CH 14 87 116 163 142 205 150 274CZ 21 80 82 140 106 171 127 200

LV ? ? ? ? ? ? 141NO 52 70 88 111 102 146 108 162PL 22 82 90 127 127 185 136 222SK 50 113 130 122 161

1997 8h-P50 8h-P98 8h-P99.9* 8h-MAXrange min max min max min max min maxAT 15 97 83 141 108 167 112 174BE 20 56 88 175 134 202 143 226DE 14 102 55 180 82 219 93 225DK 44 62 86 114 109 147 112 158ES na na na na na na na naFI 35 75 81 110 92 133 102 147FR 14 97 75 159 101 194 15 211GB 12 66 62 118 92 178 60 200GR 20 57 72 147 97 203 109 252IE 39 66 79 104 na na 110 162IT 48 48 129 129 162 162 94 230LU 22 62 72 140 103 166 110 175NL ? ? ? ? ? ? 57 199PT 11 65 38 154 53 165 57 233SE 59 69 96 113 112 142 114 147

CH na na na na na na 135 209CZ 24 80 74 134 97 160 107 187

LV ? ? ? ? ? ? 129NO 51 70 88 108 101 142 102 147PL 23 81 81 124 114 157 125 175SK 50 105 120 108 149


22/58

22

4. Discussion

4.1. Geographical coverage of monitoring stations

For 1997 information of 984 stations in EU Member States and of 92stations in five

6oth er Europ ean coun tries was received; for n early all of

them geographical co-ordinates are available. For the interpretation ofozone data it is essential to have an indication of the imm ediatesurroun dings of the station as the ozone concentration m ay be stronglyinfluen ced by local conditions. For example, the ozone con centrations maybe scavenged by locally emitted nitrogen oxides or by enhanced drydeposition as might be the case un der a forest canopy; see Annex II for abrief discussion on ozone phenomenology. For n early all stations

information on immediate surroundings was made available.

As was observed in previous reports, not only the relative fractions of street,urban and rural stations but also th e station den sity varies stron gly betweenthe reporting countries (De Leeuw et al., 1997; Becket al., 1998), as Figure3 shows. Here the station density as function of various indicators is given.The nu mber of stations per capita or per km

2land area varies strongly. The

station density relative to the Gross Domestic Product seems to be slightlymore constant. When comparing th e number of stations with ozoneprecursor emissions (h ere the sum of NOx and twice the VOC emissions is

selected as indicator for ozone formation , see Sluyter, 1995), again a widespread is observed. More detailed indicators (n umber of urban stations(excluding street stations) per urban dweller, number of rural stations perkm

2show even larger spread . It has to be stressed that a certain variation in

densities is expected over Europ e as national conditions with respect tometeorology, topography/ orography, economic structure , ph ysicalplanning etc. are different over Europe but the large variations observedhere suggest that basic pr inciples in defining a mon itoring strategy differsstron gly between the countries. In all cases the d ensities are normalised toa un it density averaged over all 20 reporting coun tries.

Map 2 shows the location of backgroun d monitoring stations. Tentatively,the area for which th e measurem ents at these stations are assumed to berep resentative is indicated with a circle using a radius of 100 km. Thisradius of representativeness might be different for th e various regions inEurope - it might even depend on the wind d irection - and should be basedon more detailed analysis of the ozone phenomen ology at the stations.Although th e iden tification of the background stations is by far notcomplete and the radius of representativeness may differ from the assumed100 km, Map 2 suggests that th e p resent set of stations covers 40 to 50% ofthe area of the reporting coun tries. As is shown in Map 2 the geograp hical

coverage of backgroun d stations is rather adequate in north -west andCentral Europe but in southern regions gaps are noted . In Scandinavia the

6

In this chapter data from Slovakia has not been included.


23/58

23

coverage by background stations seems to be inadequate but the smallgradients in ozone concentrations over this region suggest a larger radiusof representativiness than th e assumed 100 km.

To comp are th e monitoring r esults in the various coun tries it is importantthat in th e national or regional networks similar criteria are used to classifythe stations. Further work on harmonisation of station classificationschem es is un dertaken by ETC/ AQ in co-operation with DGXI (Larssen etal., 1998). A second requ irement for obtaining Europ e wide comparableresults is the h armonisation of air quality measurements. This work isconducted on behalf of the Commission by the European ReferenceLaboratory of Air Pollution (ERLAP) at the Joint Research Centr e in Ispra.Recently the results of a compar ison of calibration p roced ures used bynational laboratories in EU and EFTA coun tries were p ublished ( De Saegeret al., 1997). 21 national laboratories made use of the facilities of ERLAP to

check and compare their ozone calibration m ethods with the ERLAPreference methods. During the inter-laboratory tests the most commonlyused technique for automatic measurements was based on UV absorptionph otometry. With the exception of two analysers, the inter-comparisonsshowed that measurements of various ozone concentrations were in goodagreement (within 3%) with the standard value. The good agreementbetween the calibration techn iques and results obtained by theparticipating laborator ies shows the h igh level to which techniques for themeasuremen t of ambient ozone concen trations are harm onisedthroughout Europe. However, periodical checks and mainten ance of theanalysers is essential to avoid inter ference with o ther water vapour andother trace gases.


24/58

24

Figure 3: Station density (the average density for all countries isnormalised to unity) relative to population, land area, GrossDomestic Product and precursor emissions. The second and thirdgraph on the bot tom line show the density of urban stat ionrelat ive to the urban population and the number of rural stationsper km2, respectively. Average densities for the reportingcountries are respectively: 2.5 station per million inhabitants; 0.27station per 1000 km

2; 0.14 station per 10

9US $ (1990 prices);

0.026 stat ion per emitt ed kton; 1.26 urban station per millionurban inhabitants; 0.08 rural station per 1000 km

2.

population

0.1 1 10

AT

BE

DE

DK

ES

FI

FR

GB

GR

IE

IT

LU

NL

PT

SE

EU15

CH

CZ

LV

NO

PO

relative density

area

0.1 1 10

AT

BE

DE

DK

ES

FI

FR

GB

GR

IE

IT

LU

NL

PT

SE

EU15

CH

CZ

LV

NO

PO

relative density

GDP

0.1 1 10 100

AT

BE

DE

DK

ES

FI

FR

GB

GR

IE

IT

LU

NL

PT

SE

EU15

CH

CZ

LV

NO

PO

relative density

NOx + 2* VOC

0.1 1 10

AT

BE

DE

DK

ES

FI

FR

GB

GR

IE

IT

LU

NL

PT

SE

EU15

CH

CZ

LV

NO

PO

relative density

urb pop

0.1 1 10

AT

BE

DE

DK

ES

FI

FR

GB

GR

IE

IT

LU

NL

PT

SE

EU15

CH

CZ

LV

NO

PO

relative density, urban stations

area

0.01 0.1 1 10

AT

BE

DE

DK

ES

FI

FR

GB

GR

IE

IT

LU

NL

PT

SE

EU15

CH

CZ

LV

NO

PO

relative density, rural stations


25/58

25

Map 2: Locat ion of background ozone monitoring stations. The area forwhich the ozone measurements might be representative istentatively indicated wit h a circle with a radius of 100 km.


26/58

26

4.2. Annual statist ics, 1997

The geograp hical distribution of 98-percentile values calculated on thebasis of hourly concentrations is presented in Map 3 for backgroun d

stations and in Map 4 for urban, street and other stations.

Similar to the observations in p revious years (de Leeuw et al., 1997) the 98-percen tiles at backgrou nd stations show in gen eral low values in the Nordiccoun tries, and an increase from Northwest to Cen tral Europe. In particularfor the stations in Austria, the elevated location of the monitoring stationsmay play a role. Similar pattern s have been estimated from measurementsmade within the framework of EMEP (H jellbrekke, 1996). The relativelylow values observed in the EMEP data over th e Iber ian pen insula can notbe con firmed (or rejected) with the p resent observations. Information for

Spain is not available; in Portugal there is on ly one rural station availablefor comparison . Here a 98-percen tile of 157 g/ m 3 is observed. Th is isslightly higher than the observed ran ge of 116-149 g/ m 3 at EMEP station sin Spain and Portugal in 1995; the limited comparison and the differen cein measuring period precludes any conclusion.

For urban and other stations (Map 4) no large scale concentrationgradient in 98-percen tiles is recogn ised; high values are observed all overEurope. The local conditions seems to be more impor tant than Europ ean-wide smog episodes (at stations downwind of the urban area relatively highozone values might be observed whereas at stations with NO

xsources, such

as traffic, in their immediate surrounding relatively low ozone levels will bemeasured, see Ann ex II).

The 98-percen tile values based on moving eight-hourly averageconcentr ations show a stron g correlation with th e hourly 98-percen tile

values: on the average, 8-h percen tiles are about 7 g/ m 3lower than th ecorresponding 1h value. The relation between the two statistics depen ds onthe type of the stations and also h as a meteorological depend ence. Thegeographical distribution of the 8-h percen tile values is very similar to thedistribution of the 1-h percen tile values.


27/58

27

Map 3: 98 percenti les (based on hourly concentrations; g/ m3) measuredat background stat ions, period 1 January - 31 December 1997


28/58

28

Map 4: 98 percentiles (based on hourly concentrat ions; g/m3) measuredat urban, street and other stations, period 1 January - 31December 1997


29/58

29

4.3. Exceedances of thresholds in 1997

4.3.1. Exceedances of the threshold value for protection of human health

The th reshold value for protection of hu man h ealth ( 110 g/ m 3) is basedon eight-hourly values. According to th e Ozone Directive, four eight-hourlyperiod s have to be considered: 0.00-8.00; 8.00-16.00, 16.00-24.00 and 12.00-20.00. Based on th e average d iurn al profile of ozon e (see Ann ex II) thehighest eight-hourly values are generally expected for the 12.00-20.00period; only exceedances of the threshold values for this period have beenconsidered h ere.

No clear definition of the time reference (local time or universal time,summer- or winter time) is given. This might lead to inconsistencies

between the Member States. The differen ces between the maximumconcentrations quoted for the (8h ) con centration between 12.00 and 20.00(Table 3) and for the moving (8h) average concentrations (Table 5)indicates that small errors might be introduced by the ambiguity on timereference.

In 1997 exceedan ces of this threshold value have been observed in allreporting coun tries. A maximum concentration exceeding twice th ethreshold value have been observed in three Member States; on a moving

8-h average base, six countries report con centrations above 220 g/ m 3. The

data submitted by Spain d id not include information on exceedances ofthis threshold. The frequency by which the threshold level of 180 g/ m 3 hasbeen exceeded, suggest that, on th e averaged , the level of 110 g/ m 3 (8h)has been exceeded 20-30 times at a Spanish station .

Figure 4 shows the frequency distribution of eight-hourly ozoneconcentrations in excess of the threshold value u sing so-called Box-Jenkinsplots. For each country the Box-Jenkins plot indicates the minimum (here

the minimum is of course 110 g/ m 3), the maximum, the 25 percen tileand the 75 percentile value of the concentrations during exceedance.

Although peaks of more than 200

g/ m

3are observed in 7 out of 19

reporting Member States, Figure 4 shows that, in m ore than 75% of all

exceedance situations, the concentrations are below 165 g/ m 3 ( that is,150% of the threshold value).


30/58

30

Figure 4: Frequency distribution of ozone concentrations (eight-hourlyvalues; period 12.00-20.00; 1 January - 31 December 1997) in

excess of the 110 g/ m3 threshold for hour ly values. For eachcountry the total number of observed exceedances is given in row#Ex, the number of stations is given in row #St. Frequencydistr ibut ions are presented as Box-Jenkins plots indicating theminimum, the 25-Percentile, the 75-percentile and the maximumvalue.

8h values (12.00 - 20.00 h) > 110 g/m3

110

130

150

170

190

210

230

250

AT BE DE DK ES FI FR GB GR IE IT LU NL PT SE CH CZ LV NO PL

na


#ex 3360 577 9530 42 na 69 3218 446 384 16 1353 131 567 228 57 761 1299 6 94 223

#st 110 25 373 5 na 10 147 69 12 6 40 5 39 10 6 13 46 1 14 14

n.a = number of exceedances not available

The geograph ical distribution of the n umber of days the th reshold value

was exceeded is shown in Map 5 for backgroun d stations and in Map 6 forurban, street and other stations. A comparison of Map 5 and Map 6 showsthat exceedan ces are gen erally more frequen tly observed at rural stations.Th is is also demonstrated more clearly in Figure 5 where for each stationtype (rural, urban and street stations) the average exceedance rate isplotted. The average exceedance rate generally decreases in the orderrural - urban - stree t. Exceptions are foun d in Greece an d Italy. It might bethat this is caused by a bias in the network con figuration s: in both coun triesthere are much more urban th an ru ral stations, moreover, in Italy the ruralstations are all located in the most north ern part on the country. Note thatother stations and stations for which the type has not been specified are

not presented in Figure 5.

Exceedances most frequen tly occur in th e summer months (May-August) .In the winter mon ths ( January-March and September-December)exceedances have been observed very occasionally.


31/58

31

Figure 5: The exceedance rate of the threshold value for prot ect ion of

human health (110 g/ m3 for eight-hourly values (in mg.h/m3),period 1 January - 31 December 1997) averaged over all report ingrural, urban, and street stations. No information is available for

Spain.

Exceedance rates (mg.h.m-3) averaged per subset of station

threshold 110 g/m3 (8h, 12:00-20:00)

0

2

4

6

8

10

12

14


Rural

Urban

Street

na


32/58

32

Map 5: Number of exceedances of the threshold value for protect ion of

human health (110 g/ m3 for eight hourly values) observed atbackground stations; 1 January - 31 December 1997; eight -hourlyaverage values for the period 12.00-20.00.


33/58

33

Map 6: Number of exceedances of the threshold value for protection of

human health (110 g/ m3 for eight-hourly values) observed aturban, st reet and other stations; 1 January - 31 December 1997;eight-hourly average values for the period 12.00-20.00.


34/58

34

4.3.2. Exceedances of the threshold values for information and warning of the

population

The threshold value for warning of the popu lation ( 360 g/ m 3, hour ly

value) was exceeded in 1997 on 18 June, 13.00 at one station (Lykovrissi) inAthens, Greece. Th is day was the start of an ozone episode in Athens. Thestation Lykovrissi reported exceedances of the human health related

threshold value of 110 g/ m 3 for 8-hourly averaged during the period 18-24 Jun e. During the whole month of Jun e, concentrations were high inAthens: on 17 days at least one (out of seven) stations reported an h ourly

value exceeding 180 g/ m 3. On 22 Jun e all seven station reported anexceedance with an maximum of 229 g/ m 3. For a more extensivedescription of this episode see d e Leeuw et al. (1997).

The geograph ical distribution of the n umber of exceedances of the

threshold value for information of the public (180 g/ m 3, hourly value) ispresented in Map 7 for background stations and in Map 8 for u rban, streetand other stations. Exceedances are repor ted by 15 coun tries (12 Member

States) . In Ireland and in the most nor thern coun tries the 180 g/ m 3 levelwas not reached.

Figure 6 shows the frequ ency distribution o f concentrations in excess of thethreshold value. Althou gh incidentally the threshold value may beexceeded by more th an a factor of two, in almost all of the cases theexceedances are less extreme: Figure 6 shows that in nearly all Member

States on 75% of the days on which the th reshold value was exceeded, thelevel of 240 g/ m 3 ( that is 33% above the threshold value) was no t reached.

Figure 6: Frequency dist ribut ion of ozone concentrat ions (hourly values; 1January - 31 December 1997) in excess of the 180 g/ m3 thresholdfor hourly values. For each country the total number of observedexceedances is given in row #Ex, t he number of stations is givenin row #St . Frequency distribut ions are presented as Box-Jenkinsplots indicating the minimum, the 25-Percentile, the 75-percentileand the maximum value.

1h values > 180 g/m3

180

230

280

330

380

430


AT BE DE DK ES FI FR GB GR IE IT LU NL PT SE CH CZ LV NO PL#ex 15 81 534 1 151 0 181 48 111 0 11285 47 34 0 48 8 0 0 7

#st 111 25 373 5 36 11 148 71 11 6 92 5 38 11 6 13 48 1 14 14


35/58

35

Map 7: Number of exceedances of the threshold value for information ofthe population (180 g/ m3 for hourly values) observed atbackground stations; 1 January - 31 December 1997


36/58

36

Map 8: Number of exceedances of the threshold value for informat ion of

the population (180 g/ m3 for hourly values) observed at urban,street and other stations; 1 January - 31 December 1997


37/58

37

4.3.3. Exceedances of the daily threshold value for protection of vegetation

As Table 4 shows, exceedances of the daily threshold of 65 g/ m 3 werefrequently observed in all reporting coun tries. All coun tries have stationswhere th e threshold was exceeded on more th an 100 days in 1997 with the

exception of Latvia and the Nether land s. The geograph ical distribution ofthe nu mber of exceedan ces of the daily threshold is presented in Map 9 forthe backgroun d stations. In Map 9 an attempt h as been made to quan tifythe area where exceedances are observed. For th e backgroun d stations arepresentative radius of 100 km is assumed, see also Map 2. When therep resentative areas of two or more stations overlap, the number ofexceedances in this location is estimated by a distance-weighted

interpolation . At all rural stations exceedan ces of the 65 g/ m 3 level werereported; the lowest number ( 5 days of exceedance) was reported for theGerman station Ffm. Greisheim located near Frankfurt am Main.Exceedances on more th an 150 days are reported both by northern andsouth ern Member States.

For exceedances of a daily average concentration the differences betweenrural and urban stations are more pron oun ced than is the case for hour lyconcentrations. In u rban areas the low concentrations during the n ight(caused by interaction with NO

xemissions) reduce the daily average

concentrations; in rural areas the d ecrease in ozone con centrations duringthe n ight is generally less stron g. In NW Europe, however, th e h igh NO

x

emission d ensity might cause also some qu enching of ozone at ru ral siteswhich explains the relative low number of exceedan ces in th is region

compared both to North and South Europe.

Figure 7 shows the frequen cy distribution of daily values in excess of 65 g/ m

3. Although high values of more than 200 g/ m

3have been observed,

in n early all coun tries for 75% of the exceedan ces the daily average

concentration falls between 65 and 98 g/ m 3 ( that is 150% of the thresholdvalue) . The extremely high values reported for Italy are most likely inerror; there are clear indications that at several stations the maximumobserved 1h concentration was rep orted as daily averaged . Theexceedances are most frequently observed d ur ing the growing season ( May-

July): more than 45% of the exceedances were r epor ted in this three-mon th period. About 18% of the exceedances occurred during the wintermon ths ( January-March and October-December ).

In Figure 8 th e exceedance rates, averaged per subset of stations types, aregiven for each country. This figure clearly shows that especially at ruralstations this threshold value is frequen tly and largely exceeded. H ighexceedance rates are observed both in nor thern and south ern countr ies.


38/58

38

Figure 7: Frequency dist ribut ion of ozone concentrations (24h values; 1

January - 31 December 1996) in excess of the 65 g/m3 thresholdfor daily values. For each country the total number of observedexceedances is given in row #Ex, t he number of stations is givenin row #St . Frequency distr ibut ions are presented as Box-Jenkins

plots indicating the minimum, the 25-Percentile, the 75-percentileand the maximum value.


65

115

165

215

265

315

365


na


#ex 1322

4

1261 2373

7

392 na 1407 9305 2209 840 440 3466 352 1198 475 997 1759 5356 55 2057 1265

#st 111 25 371 5 na 11 144 67 13 6 51 5 38 9 6 13 48 1 14 16

Figure 8: The exceedance rate of the threshold value for protection of

vegetation of 65

g/m

3for daily values (in mg.h/m

3), period 1

January - 31 December 1997) averaged over all report ing rural,urban, and street stat ions. No information is available for Spain.

exceedance rates (mg.h.m-3) averaged per subset of station types

threshold 65 g/m3 (24h)

0

20

40

60

80

100

120

140

160

180

200


Rural

Urban

Street

na


39/58

39

Map 9: Number of exceedances of the threshold value for vegetation (65g/ m3 for daily values) observed at background stations; 1 January- 31 December 1997. Data is interpolated using inverse distanceweight ing and a cut-off distance of 100 km


40/58

40

4.3.4. Exceedance of the hourly threshold value for protection of vegetation

Exceedances of the h ourly threshold values for p rotection of vegetation

(200 g/ m 3) were r eported by 10 Member States; no exceedances were

observed in Denmark, Finland, Ireland and Sweden, see Table 4. Thegeographical distribution of the n umber of days on which th is hourlythreshold value h as been exceeded is presented in Map 10 for backgroundstations. The map shows that roughly above 55

0N this threshold value has

not been exceeded. The h ighest number of exceedances at a rural station(13 times) is observed at the German station Schwarzwald-Sd, see Map 10.

The frequency distributions of exceedan ces of the hourly threshold arepresented in Figure 9. For abou t 75% of the exceedances the ozone levels

fall between 200 and 275 g/ m 3.

Figure 9: Frequency dist ribut ion of ozone concentrat ions (hourly values; 1

January - 31 December 1997) in excess of the 200 g/m3 thresholdfor hourly values. For each country the total number of observedexceedances is given in row #Ex, t he number of stations is givenin row #St . Frequency distr ibut ions are presented as Box-Jenkinsplots indicating the minimum, the 25-Percentile, the 75-percentileand the maximum value.


200

220

240

260

280

300

320

340

360

380

400


na


#ex 3 25 143 0 na 0 110 12 61 0 531 1 10 13 0 11 1 0 0 2

#st 111 25 373 5 na 11 151 71 10 6 82 5 38 11 6 13 48 1 14 14


41/58

41

Map 10: Number of exceedances of t he threshold value for vegetation (200

g/ m3 for hourly values) observed at background stat ions; 1January - 31 December 1997. Data is interpo lated using inversedistance weighting and a cut-off distance of 100 km


42/58

42

4.4. Preliminary review of exceedances of the revised guidelines ofWHO

As men tioned in Chapter 1 the World Health Organization in Europe

produced a revision of its guidelines for ozone (WHO , 1996a,b), see Table6. According to the Fifth Environment Action Programme these guidelinesshall serve as a basis for setting air quality standards in the Community.While these values are also relevant in the context of preparing the newdaugh ter directive for ozone an attempt h as been made to evaluate theozone data submitted un der the framework of the curren t directive againstthese guideline levels. This can only be a very preliminary evaluation as theinformation available un der the directives deviates from th e guidelinerequ irements on averaging time an d p eriod given in Table 6.

Table 6: An overview of ozone guideline levels of WHO.

Descript ion level Time period Const raints

human health 120 g/m Floating 8h mean

for vegetation:

Crops and semi-naturalvegetation

AOT40*= 3 ppmh 3 months

(May-July)

Forests AOT40 = 10 ppmh 6 months(April-September)

Crops (visible injury) AOT40= 0.2 ppmhAOT40= 0.5 ppmh

5 days5 days

Humid air condit ionsDry

air conditions

* calculated from 1h mean values during daylight hours; 1ppmh = 1000g/ m3.h

Referring to the guideline for human h ealth, the exceedances reported forthe 8h p eriod between 12.00 and 20.00 has been used to evaluate

exceedances of the th reshold value of 120 g/ m 3 for moving 8-h values. In1997, at one out of every five stations the 8h -averaged concentration

between 12.00 and 20.00 is in excess of a level of 120 g/ m 3 on more than20 days. As ozone con centrations generally reach their maximum in thelate aftern oon , the mean over th e fixed period 12.00-20.00 is expected tobe a good approximation of the highest daily 8-h values calculated from amoving averaged . Figure 10 shows a Europe wide exceedan ce of thisthreshold value. However, evaluation of hour ly data has shown that whencomparing the nu mber of exceedan ce days calculated from moving 8-hourly mean s with such based on the fixed 12-20 h period , the latter isabou t 15% lower, in urban areas even 25% lower ( Martin Lutz, privatecommunication). Figure 10 might therefore give an un derestimation of theexceedances of the WHO guideline value.


43/58

43

Figure 10: The occurrence of exceedances of the WHO guideline for

prot ect ion of human health (120 g/m3 for moving 8-hourlyvalues; calculations are, however, based on ozone levelsreport ed for the fixed time period 12.00-20.00h, see text),averaged over all stations; per iod 1 January - 31 December1997.

Occurrence of exceedances

120 g/m3 (8h, 12:00-20:00)

0

5

10

15

20

25

30

35

40

AT BE CH CZ DE DK ES FI FR GB GR IE IT LU LV NL NO PL PT SE

na

Evaluation on the AOT40 values for vegetation is not p ossible using theavailable information . A crude estimate might be based on the information

on exceedance of a daily concentration of 65 g/ m 3. The exceedance rate(see Eq 2) of a thr eshold of 80 g/ m 3 (ER40) an d the AOT40 values differ,however, on the time window for integration: the AOT40 is based onhourly values during the daylight-period (8.00-20.00 CET) whereas theER40 is based on 24h per day. A longer integrat ion time might lead to

higher values. On the other h and, days with a daily average below 80 g/ m 3

do not contribute to the ER40 but may contribute to the AOT40 when

hourly averages peak above 80 g/ m 3. Due to these differences theexceedan ce rate may either over- or underestimates the AOT40. Apreliminary analysis using hourly data for a limited n umber of stations inLatvia, the Netherlands, and th e United Kingdom showed th at the ER40 is

lower th an the corr esponding AOT40; in some cases AOT40 is even afactor 3-8 higher th an th e ER40.

An exceedan ce rate of 6000 g/ m 3.h was exceeded in 1997 on 26% of allreporting stations; limiting the an alysis on ly to th e rural stations, thispercentage increases to 41%. In view of the d iscussion above th is will give asevere un derestimation of the stations where the WHO critical level for

protection of crops and semi-natural vegetation (AOT40 = 6000 g/ m 3.h) isexceeded. Both observations as well as model calculations made withinEMEP (EMEP MSC-W, 1998) show that south of 55-57

oN an AOT40 of 6000

g/ m3

.h is exceeded in more than 90% of the area.


44/58

44

It must be concluded th at the information currently collected und er theOzone Directive is not adequate to assess exceedan ces of the revised WHO -guidelines for ozone.

4.5. Data reported for 1994-1997

Tren ds in ozone concen trations are expected to result from trends inprecursor emissions in Europe and from th e increasing trend inhemisph eric backgroun d concen trations (Borrell and van den H out,1995). Th e magnitude an d even the sign of a possible tren d m ight differfrom location to location. In a study of trends in concentrations of ozoneand related species in the Netherland s and nearby countries Roemer(1996) concluded that the ground level oxidant (sum of ozone and NO

2)

concentr ations have decreased significantly in the Netherlands from 1981to 1994 with an average d ecrease of about 1% per year. For German y a

slightly downward (in the northern part) or slightly upward (in thesouth ern part) trend was no ted but probably none of these trends issignificant at the 95% confidence interval (Roemer, 1996) .

From an analysis of ozone concentr ations in un polluted air massesarriving over th e Nor th Atlantic region at th e Irish station Mace Head. itappeared that over th e 1990-1996 period there is an upward trend of 0.092ppb per year (or 0.2% per year) in background ozone concen trations(PORG, 1997) . Model calculations on a global scale support the findings ofa ongoing change in chemical composition of tropospher ic backgroundair. In the late 1990s daily mean concentrations of ozone are three to four

times higher than in the pre-industrial era as a result of the tremend ousgrowth in NO

xemissions from industry and the transport sector.

Th e year to year variation in 50-percen tile values are relatively small (2 -7 g/ m

3on a P50-value of about 30 to 60 g/ m 3) when compared to the

variation s in the 98-percen tile values (a variation of 11-18 g/ m 3 for P98-values ranging from 76-145 g/ m 3) . Peak values of ozone are stron glycorrelated with temperature mainly because th e conditions leading to hightemperatures (e.g. strong solar radiation, low wind speeds, continentalflows) also trigger photoch emical formation. Meteorological fluctuations

may cause variations in peak ozone levels that are much larger than thevariations due to chan ges in precursor emissions. A yearly fluctuation ofabou t 15% in 98-percen tile value is not excep tional.

The time series reported in th e framework of the O zone Directive, nowcovering a period of four years (1994-1997) , are still too short or toovariable in terms of station configuration to give conclusive answers on anozone tren d in the whole of the EU. Figure 11 shows the nu mber ofoccurren ce of exceedance (th at is, the n umber of exceedances averaged

per station) for th e 110 g/ m 3 and the 65 g/ m 3 level over the four-year

per iod. Judging th e exceedances of the 110 g/ m3

threshold, 1997 is a yearwith generally a low number o f exceedan ces and a decreasing tren d isobserved in some coun tries (e.g. AT, DE) but n ot in all (see, for exampleIT) . A different p icture is found for exceedances of the daily threshold of


45/58

45

65 g/ m 3. Here, 1997 is not a year with an exception ally low number ofexceedan ces; for a number of coun tries (AT, BE, FR, IT, PT, SE) 1997 iseven the year with th e h ighest nu mbers. Data for DK, GB, and GR suggeston th e other h and, a decreasing trend.

For various Member States data reports are available for years previous to1994. Longest time series are available for Belgium, Luxembourg, theNetherlands, and United Kingdom. At the level of the individual stations inthese countries, a possible trend in ozone levels has been examined usingthe non-parametr ic Mann -Kendall test (Gilbert, 1987). On ly those stationsfor which at least data for 5 years are available have been included in thetests. The results for trends in 50- and 98-percen tile values are summar isedin Table 6. For most of the stations no significant trend in 50-percentilevalues is observed. For the stations with a significant trend, the results are

mixed. An upward tren d of 1-3 g/ m 3 per year is seen at a few stations inBelgium and Luxembourg. Trends in th e United Kingdom are bothpositive or negative. The same is true for the Netherlands, although herethe nu mber of stations with a downward trend outweighs the nu mber ofstations with a positive trend. In the analysis of the 98-percentiles also noclear picture for a coun try or region emerged. On ly for the Netherlandsthe majority of stations show a downward tren d but h ere artefacts caused bythe introduction o f a new ozone mon itor during the winter of 1990/ 1991might not have been excluded (RIVM, personal commun ication).

Figure 11: Occurrence of exceedances of t hreshold values of 110 (8h)

and 65 (24h) g/ m3

in EU Member States during the period1994-1997.

OoE of 110 ug/m3 (8h)

0 10 20 30 40 50 60 70

AT

BE

DE

DKES

FI

FR

GB

GR

IE

IT

LU

NL

PT

SE

97

96

95

94

OoE of 65ug/m3 (24h)

0 50 100 150 200

AT

BE

DE

DKES

FI

FR

GB

GR

IE

IT

LU

NL

PT

SE

97

96

95

94


46/58

46

Table 6: Trend est imates for 50- and 98-percentile values for individualstations in Belgium, United Kingdom, Luxembourg and theNetherlands; period 1989-1997.

member state number of stations(1)

slope range

(g/m3

per year)(2)

50-percentile

BE 7(3,0) 1.6 3.3 (0.0 3.3)GB 24 (2,2) -1.0 1.4 (-1.0 1.4)LU 5 (2,0) 0.9 2.6(-1.8 2.6)NL 32(3,10) -1.6 0.9 (-1.6 1.0)98 percentile

BE 7 (1,0) 1.5 (-2.5 8.3)GB 24 (0,2) -6.7 -6.0 (-6.7 6.0)LU 5 (1,0) 1.8 (-0.3 3.0)NL 32 (0,20) -6.3 -2.6 (-6.3 3.0)

(1) in parentheses the number of stations with a significant upward, respectively downward trend (at an

= 0.10 significant level) is given;(2) range in estimated annual trends for stations with significant t rend; results for all stations including

those where no significant trend has been observed are given in parentheses.


47/58

47

5. Conclusions and recommendat ions

1. Data for 1997 was received by the European Commission from all

Member States. In addition, information for 6 o ther Europeancountries was received.

Th e15 EU Member States provided information on ozoneconcen trations for a total of 984 monitor ing stations. Switzerland,the Czech Republic, Latvia, Norway, Poland and Slovakia providedinformation on ozone con centrations for 100 stations.

2 The threshold value set for the protection of human health wasexceeded substantially in all reporting countries.

The th reshold value of 110 g/ m3

(8h-average) was exceededsubstantially ( in abou t25% of the repor ted exceedances the 8h-

average con centrations exceeded 165 g/ m 3) .

3. The threshold values set for the protection of vegetation wereexceeded substantially and in almost all EU Member States.

The th reshold value of 65 g/ m3

(24h average) is reported to beexceeded substantially (by up to a factor 3) , widely ( in all reportingcountries) and frequently (n early all countries report exceedancesdur ing more th an 100 days at one or more of their stations). Thethreshold value of 200 g/ m

3(hourly average) is exceeded largely

and widely ( in total 13 countries, 10 EU Member States) on a limited

number of days.

4. The threshold value for information of the population wasexceeded in almost all EU Member States during a limited number

of days.

Exceedance of the information threshold value of 180 g/ m3

(1 haverage) was rep orted for stations in 15 countries (12 EU MemberStates). For on e station (Lykovrissi, Athens) an exceedance of thewarn ing level of 360 g/ m

3(1h average) was reported .

5. The spatial coverage and quality documentation of the data need

further improvement.Depend ing on the local situation, the ozone m onitoring stations arecharacterised as rural, urban , street or other ( e.g. industrial) .Further work towards a harmonised classification system isrecommended.

The present subsets of rural stations and urban stations are n otrepresentative for the land area of the EU and the urbanpopu lation , respectively.

Spatial coverage and documentation of the monitoring data qualityneed improvemen t if the level of protection of human h ealth andecosystems in Europe to elevated ozone levels is to be fully assessed .Member States are encouraged to reconsider th eir ozone measuring


48/58

48

networks in the light of the spatial coverage of ozone monitor ingstations. It is recommend ed to improve the d ocumen tation on th erepresentativity and on the surrou ndings of the existing stations.

6. It is recommended to improve the reporting of ozone precursors

(NOx, NO2 and VOC).NO

xmeasuremen ts should be co-located with the ozone mon itoring

stations as NOx

can be u sed as an indicator of the stationrep resentativity. Moreover, pr ecursor con centrations will be neededfor testing of compliance with VOC an d NO

xemission reduction

programs.

Acknowledgement

The authors are indebted to Jaroslav Fiala and H ana Livorova at thePHARE Topic Link on Air Qu ality and Jana Ostatnicka at the CzechHydrometeorological Institute for th e cartograph ic contributions andfurther GIS suppor t .


49/58

49

6. References

Anderson H.R., Spix C., Medin a S., Schouten J.P., Castellsague J., Rossi G., Zmirou

D., Tou loum i G., Wojtyniak B., Ponka A., Bacharova L., Schwarts J. andKatsouyann i K. (1997) Air pollution and daily admissions for ch ron icobstructive pulmonary disease in 6 Europ en cities: results from the APHEAproject.Eur. Respir J. 10, 1064-1071.

Barrett K. and Berge E. (eds) (1996) Transboundary air pollution in Europe. Part1: Estimated dispersion of acidifying agen ts and of near surface ozone.EMEP/ MSC-W Report 1/ 96. The Norwegian Meteorological Institute, Oslo,Norway.

Beck J.P., Kryzanowski M. and Koffi B. (1998) Tropospheric ozone in theEuropean Union . The consolidated Report, European Topic Centre on AirQuality, Report presented to the Steering Group on Ambient Air Quality,

Brussels,

Borrell, P. and van d en Hout D. (1995) Troposph eric ozone - a review of curren tun der stand ing. Doc.Ref XI/ 363/ 95, Europ ean Comm ission, Brussels.

Borrell P., Builtjes P., Grenn felt P. and Hov ( eds) ( 1997) Photo -oxidants,acidification and tools. Policy application of EUROTRAC results. Springer-Verlag, Berlin.

Bull K.R. (1991) Th e critical load/ level approach to gaseous pollutant emissioncontrol.Environ. Pollut., 69, 105-123.

De Leeuw F., Sluyter R. Van Zantvoort E. and Larssen S. (1995) Exceedance ofOzone threshold values in th e Europ ean Commun ity in 1994. Summary basedon the information repor ted in th e framework of the Council Directive92/ 72/ EEC on air pollution by ozone. European Topic Centre on Air Quality.

De Leeuw F., Sluyter R. and Van Zantvoort E. (1996) Air Pollution by Ozone inthe European Union , Exceedance o f ozone threshold values in 1995 andsummer 1996. EEA Topic Report 29/ 1996. European Environmen t Agency,Copenhagen.

De Leeuw F., Sluyter R. and Van Zantvoort E. (1997) Air Pollution by Ozone inthe European Union , Exceedance o f ozone threshold values in 1996 andsummer 1997. EEA Topic Report 7/ 1997 European Environmen t Agency,Copenhagen.

De Saeger E., Noriega Guerra A., Gerboles M., Rau H ., Aman tini L. and PerezBallesta P. (1997) Harmonisation of directive 92/ 72/ EEC on air pollution byozone; inter-comparisons of calibration procedures for ozone measurements.Report EUR 17662, European Commission, Brussels.

Derwent, D. and van d en Hou t D. (1995) Computer modeling of ozone formationin Europ e. Doc.Ref XI/ 364/ 95, Europ ean Commission, Brussels.

EC (1994) . Information Document con cern ing Air Pollution by Ozone - Overviewof the situation in Europe dur ing Summ er 1994. European Commission,Brussels.

Gilbert R.O (1987) Statistical methods for environmental pollution monitoring.

Van Nostran d Rein hld, New York.

Hjellbrekke A.G.(1996) Ozone Measurements 1993-1994, EMEP/ CCC rep ort1/ 96, NILU, Kjeller, Norway.


50/58

50

Hov O. (Editor) (1997) Tropospheric Ozone Research, Springer Verlag, Berlin.

Larssen S., Sluyter R., and Helmis C. (1998) Cr

Documents

17. Poluarea Aerului Ozonul Indicatori_1999