Concentrations and Trends of Nitrogen Oxides, Ozone and Volatile Organic Compounds in the Helsinki Metropolitan Area - NOVOC

M2T0244

Concentrations and Trends of Nitrogen

Oxides, Ozone and Volatile Organic

Compounds in the Helsinki

Metropolitan Area - NOVOC

Tuomas Laurila, Jaakko Kukkonen, Harri Pietarila,

Hannele Hakola, Heidi Hellén, Virpi Tarvainen and

Mari Kauhaniemi

Raportointiaika Tammikuu 2003 Raportointikausi 1.1.2002 – 31.12.2002

Projektin koodi

M2T0244

Projektin nimi

Typenoksidien, otsonin ja haihtuvien hiilivetyjen pitoisuudet ja niiden ajallinen kehitys pääkaupunkiseudulla

Vastuuorganisaatio

Ilmatieteen laitos

Projektin vastuuhenkilö

Jaakko Kukkonen

Projektin yhteyshenkilö Osoite

Tuomas Laurila Sahaajankatu 20E, 00880, Helsnki

Puhelinnumero Telefax Sähköpostiosoite

(09) 19295510 (09) 19295403 Tuomas.Laurila@fmi.fi Muut tahot:

Organisaatio Yhteyshenkilö Puhelinnumero Sähköpostiosoite

YTV Tarja Koskentalo 09 1561358 Tarja.Koskentalo@ytv.f

i

VTT Energia Juhani Laurikko 09 4565463 Juhani.Laurikko@vtt.fi

IL Harri Pietarila 09 19295432 Harri.Pietarila@fmi.fi

Hankkeen alkamisaika Hankkeen suunniteltu kesto Hankkeen suunniteltu päättymisaika

1.5.2002 2002-2003 31.12.2003

Projektin rahoitus (k€)

Organisaatio 1999 2000 2001* 2002* 2003* Yhteensä

MOBILE2 -rahoitus 15 000 15 000 30 000

Muu rahoitus (YTV) 20 000 30 000 50 000

IL (oma rahoitus) 15 300 10 000 25 300

Yhteensä 60 300 45 000 105 300

*suunniteltu

Hankkeen tavoite

Viimeisen vuosikymmenen aikana typenoksidien päästöt sekä niiden ja otsonin pitoisuudet ilmassa ovat selkeästi muuttuneet. Typpidioksidi ja otsoni ovat keskeisiä ilmansaasteita, joilla on sekä terveydellisiä että luontovaikutuksia, ja näiden pitoisuuksien vähentämiseen pyritään erilaisin säädöksin. Merkittävistä ilmansuojelullisista toimenpiteistä huolimatta pitoisuudet eivät ole laskeneet toivotulla tavalla; itse asiassa otsonin pitoisuudet ovat olleet nousussa. Tässä tutkimuksessa selvitetään mainittujen pitoisuuksien muuttumisen määrää ja syitä, sekä havaintojen että mallitustarkastelujen avulla.

Haihtuvat hiilivedyt ovat merkittävässä osassa valokemiallisen otsonin muodostuksen kannalta. Näistä yhdisteistä erityisesti bentseeni on myös sinällään haitallinen. Tutkimuksessa keskitytään tämän vuoksi myös haihtuvien hiilivetyjen mittauksiin ja mallikehitykseen bentseenin osalta.

Projektin julkaisuluettelo

(MOBILE2 –julkaisut ja muut julkaisut projektiin liittyen)

Hellén, H., Hakola, H., Laurila, T., Hiltunen, V. and Koskentalo, T., 2002. Aromatic hydrocarbon and methyl tert-butyl ether measurements in ambient air of Helsinki (Finland) using diffusive sampling. Sci-ence of Total Environ., 298, 55-64.

Hellén, H., Laurila, T. and Hakola, H., 2002. Source apportionment study of NMHCs in Helsinki using chemical mass balance receptor model. In: Atmospheric Chemistry within the Earth System: 7th Scien-tific Conference of the International Global Atmospheric Chemistry Project, 18–25 Sept. 2002, Herak-lion, Greece. abstract.

Hellén, H., Hakola, H., Laurila, T., 2003 Determination of source contributions of NMHCs in Helsinki (60ºN, 25ºE) using chemical mass balance and UNMIX multivariate receptor models. Atmospheric Envi-ronment, accepted.

Pietarila, H., Alaviippola, B., Hellén, H., Salmi, T., Laurila, T. and Hakola, H., 2002. The preliminary assessment under the EC air quality directives in Finland: Carbon monoxide and benzene. Report, Fin-nish Meteorological Institute. 60 p.

Laurila, T., Hakola, H., Lindfors, V. and Tuovinen, J.-P., 2002. Increasing ozone trends in Finland in the 1990s. In: Midgley, P. M. and Reuther, M. (Eds.). Transport and Chemical Transformation in the Troposphere: Proceedings of the EUROTRAC-2 Symposium 2002, Margraf Verlag, Weikersheim. 4p.

Seminaarit

(Seminaarit ja konferenssit joissa projektia on esitelty, ml. MOBILE2-seminaarit)

Laurila, T., 2002. Trends of ozone, nitrate and VOCs in Finland during the 1990’s and ozone scenarios for 2050 in Europe. EUROTRAC-2 TOR-2 (Tropospheric Ozone Research) Workshop, Moscow, 8-12 September, 2002. Abstracts, MAKS Press, Moscow.

Hellén, H., Laurila, T. and Hakola, H., 2002. Source apportionment study of NMHCs in Helsinki using chemical mass balance receptor model. In: Atmospheric Chemistry within the Earth System: 7th Scientific Conference of the International Global Atmospheric Chemistry Project, 18–25 Sept. 2002, Heraklion, Greece

Opinnäytteet hankkeeseen liittyen

FK Heidi Hellén valmistelee väitöskirjaa hankkeeseen liittyen haihtuvien hiilivetyjen pitoisuuksista ja päästöistä Helsingin seudulla.

Mari Kauhaniemi tekee pääosin CAROS-hankkeeseen liittyen, mutta osittain myös tästä aiheesta DI – työtä, joka valmistuu kevään 2003 kuluessa.

1.

INTRODUCTION AND OBJECTIVES

A major fraction of nitrogen oxides released from traffic is commonly nitrogen monoxide (NO); however, the NO2/NOx fraction in exhaust emissions can range from less than one per

cent to more than half of the total emissions. The released NO is chemically transformed into the more harmful nitrogen dioxide (NO2); the reaction timescale is typically a few minutes in

urban daylight conditions (Härkönen et al., 1998). The transformation rate of NO into NO2

depends on emission characteristics, meteorological conditions, solar radiation, local environment and background concentrations of various pollutants (NO, NO2, O3 and

hydrocarbons).

In most major European cities, the flow of vehicular traffic has continuously increased during the 1990’s. Despite the increasing traffic flows, the corresponding emissions of nitrogen oxides (NOx, defined as the sum of NO and NO2) have gradually decreased in most European

countries. In Finland, the emissions of nitrogen oxides originated from traffic and energy production have substantially decreased since the late 1980’s. For instance, the emissions of nitrogen oxides originating from vehicular traffic in Finland have decreased from 140 kt to 123 kt during the period 1993 to 1997 (Mäkelä, 1999). This improvement has been achieved mainly by the increased use of catalytic converters and improvements in engine technology (e.g., Kousa et al., 2001).

Despite of the significant emission reductions, only slight decrease of NO2 concentrations has

been observed in the Helsinki Metropolitan area during the past ten years. Continuous measurements over an urban measurement network in the Helsinki Metropolitan Area during the 1990’s have shown that the atmospheric concentrations of NO have decreased substantially in this area; however, the concentrations of NO2 have shown a substantially

smaller decrease. For instance, at the monitoring sites of Töölö and Vallila, in central Helsinki, during the period from 1993 to 1997, the annual average NO concentrations have decreased by approximately 40 and 33 %, respectively, while the NO2 concentrations have

correspondingly decreased by only 14 and 27 % (Kousa et al., 2001).

One reason for the slower decrease in the NO2 concentrations is that the formation of NO2

from NO is characteristically limited by the availability of O3, and the urban concentrations of

O3 actually slightly increased in the course of the above-mentioned time period. For instance,

there has been no decreasing trend in the ozone concentrations in Nordic countries during the episodes of high concentrations in summer (Sohlberg et al., 2002). Another reason is the increased use of oxidation catalysts in diesel vehicles, as these cause a higher ratio of NO2 to

NOx in emissions (Kousa et al., 2001).

An objective of this project is therefore to investigate in more detail the factors that influence the trends of the atmospheric concentrations of ozone and nitrogen oxides.

The information has been very scarce on the emission sources and concentrations of benzene (Bz) and other volatile organic compounds (VOC) in the Helsinki Metropolitan Area. The other main objective of this project is therefore to conduct the first basic survey of the Bz and VOC concentrations in the area, both by using measurements and by model computations. We

will also apply source apportionment methods (receptor analysis) in order estimate the contributions from various source categories.

2.

TRENDS OF THE CONCENTRATIONS OF OZONE AND

NITROGEN OXIDES

Ozone concentrations close to the earth surface are influenced by concentrations in the troposphere, photochemical formation, the titration of ozone due to NO emissions and other oxidative processes, and deposition to the ground. Fresh NO emissions are oxidised by ozone; these deplete ozone concentrations as follows:

NO +O3 -> NO2

Ozone concentrations in street canyons and urban centres can be substantially lower than in surrounding areas due to ozone depletion (e.g., Kukkonen et al., 2001 and 2003; Kousa et al., 2001).

For a larger spatial scale, the same titration effect is observed in winter in Northern Europe, when there is no photochemical ozone formation. Ozone concentrations are therefore low in the air masses that have been long-range transported from the more densely populated parts of Europe (e.g., Laurila, 1999). On the contrary, in spring and summer high ozone concentrations are observed in such air masses.

In Western Europe, NOx emissions have decreased since the 1980’s as a result of air pollution

abatement measures. This decline of NOx emissions is partly responsible for the increased

ozone concentrations during winter in monitoring stations in Southern Finland. The cumulative ozone concentration distributions during the winter half-year (defined as the period from October to March) have been presented in Figure 1 for years from 1989 to 2001. The curves were drawn based on hourly concentrations that were measured at a remote island site of Utö in the Finnish Gulf. The values clearly show the gradual increase of ozone concentrations.

The cumulative ozone concentration distributions for each year at an urban site of Töölö have been presented in Figure 2. The gradually increasing trend of the ozone concentrations is clear.

The NOx emissions originated from traffic have been decreasing steadily in Helsinki during

the 1990’s. Emissions from energy production in power plants have been decreasing even more steeply, especially from 1991 to 1993 (Figure 3). The rates of decrease of annual average NOx concentrations at these measurement sites in central Helsinki are approximately the same

as that of the emissions. As NOx concentrations decrease, there is more ozone available for

oxidising NO (Figure 4). The NO2 concentrations therefore decrease more slowly, compared

Figure 1. Cumulative ozone concentration distributions during the winter half-year (from October to March) at the station of Utö. The values were computed on a yearly basis, from 1989 to 2001.

Figure 2. Annual cumulative ozone concentration distributions at an urban site of Töölö in central Helsinki. The values were computed on a yearly basis, from 1990 to 2001.

Töölö 0 20 40 60 80 100 120 140 0.1 1 10 30 50 70 90 99 99.9 O3 (µgm m-3) Percentile 1990 2001

Annual concentration frequency distributions

Utö Ozone October-March

O₃ (µgm-3) 0 20 40 60 80 100 120 140

Percentile

Figure 3. Annual NOx emissions from traffic (squares) and power plants (blue triangles) in

Helsinki and average NOx concentrations at two urban monitoring sites in Helsinki.

Figure 4. Annual NO2 concentrations at urban sites Töölö and Vallila in Helsinki and

suburban sites Tikkurila (Vantaa) and Leppävaara (Espoo) and at the rural (regional background) site of Luukki in Espoo.

Annual averages 1988 1990 1992 1994 1996 1998 2000 2002 NOx (ppb) Töölö 0 50 100 150 200

NOx (ppb) Vallila Helsinki traffic emissions 10

8 g/year 0 20 40 60 80 100 120 140

Töölö NOx (ppb), left axis Vallila NOx (ppb), right axis NOx emissions Helsinki, traffic NOx emissions Helen

Annual NO₂ averages 1988 1990 1992 1994 1996 1998 2000 2002 NO 2 (ppb) 0 5 10 15 20 25 30 Töölö Vallila Leppävaara Tikkurila Luukki

3.

MEASUREMENTS OF THE CONCENTRATIONS

OFBENZENE AND VOLATILE ORGANIC COMPOUNDS

Concentrations of benzene were measured at four sites in Helsinki Metropolitan Area using passive samplers (Figure 5). These measurements show that concentrations are well below the limit values given in the CO and benzene directive (Hellén et al., 2002). Concentrations at the Luukki site, which is a regional background station, are relatively high compared with the urban sites.

Active adsorbent tube and canister samples were collected and analysed at the urban moni-toring sites for source apportionment analysis (Hellén et al., 2003), which showed that one third of the mass originates from outside the Helsinki Metropolitan area. Additional meas-urements to support source contributions analysis showed that the contribution of small-scale wood combustion emissions is substantial. However, their influence is more important in the suburban areas, where the small-scale wood combustion mainly takes place.

Figure 5. Bi-weekly averages of benzene concentrations at four sites in the Helsinki Metro-politan area. The limit value and the upper and lower assessment thresholds are also shown.

4. MODELLING THE EMISSIONS AND ATMOSPHERIC

CONCENTRATIONS OF BENZENE

The results presented here have also been part of the Preliminary Assessment in Finland for carbon monoxide and benzene under the EC Air Quality Directives; for a more detailed discussion, the reader is referred to Pietarila et al. (2002). The assessment is based on a combination of air quality measurements, emission inventories and dispersion modelling. The assessment is based on five years of air quality measurement data covering the years 1997– 2001. 0 1 2 3 4 5 6

tammi helmi maalis huhti touko kesä heinä elo syys loka

(µ

g/m

3 )

Töölö Leppävaara

Luukki Ruskeasanta

Raja-arvo Ylempi arviointikynnys

The emission data from the Finnish national emission database of the Finnish Environmental Institute (SYKE) and the traffic emission calculation system of the Technical Research Centre of Finland (VTT) were used in the assessment. The benzene emissions in Finland were estimated based on the SYKE´s NMVOC inventory and the source profiles were estimated at the Finnish Meteorological Institute (FMI). Also more detailed data from individual emission inventories prepared by the FMI and the Helsinki Metropolitan Area Council (YTV) were used in the assessment.

FMI have carried out dispersion modelling studies for benzene in the Helsinki Metropolitan area, using the CAR-FMI model. The emissions from traffic as well as the background concentrations were taken into account in the calculations.

Evaluation of the total Bz emissions in Finland in 2000 has been presented in Figure 6. The small-scale wood combustion contributes approximately 60 % of the benzene emissions. About 20 % of the emissions are caused by road traffic and about 10 % by other mobile sources and machinery.

Figure 6. The emissions of benzene in Finland in 2000 (Pietarila et al., 2002). Kotitalouksien, …” denotes the small scale wood combustion, “Tieliikenne” is road traffic, and “Muu liikenne” is other mobile sources and machinery. The total amount of the emissions was 137 t/a.

The results of emission and dispersion model computations have been presented in Figure 7. The results include the contribution from vehicular traffic, and the regional background concentrations; the latter were evaluated based on the Bz measurements at the station of Luukki.

Figure 7. The spatial distribution of the annual average Bz concentrations (µg/m³) in the Helsinki Metropolitan Area in 2000 (Pietarila et al., 2002). Raja-arvo = EU limit value, YAK = upper assessment threshold, AAK = lower assessment threshold. The location of the measurement stations has also been shown, for stations that are directly influenced by the traffic, and for urban and regional background stations.

According to the modelled data, the benzene concentrations are well below the limit value and the upper assessment threshold set by EU. The lower assessment threshold may be exceeded in the vicinity of major roads or streets, and their junctions.

5. SOME CONCLUSIONS AND AIMS FOR THE YEAR 2003

The emission survey of benzene showed that the contribution of benzene emissions from small-scale wood combustion is unexpectedly high. According to both the measurements and model computations for benzene, the concentrations are well below the limit value and the upper assessment threshold set by EU. According to the model computations, the lower as-sessment threshold may be exceeded in the vicinity of major roads or streets, and their junc-tions.

The measurements of benzene using passive sampling will be continued in 2003, especially in order to obtain data that is well suitable for the evaluation of the model computations. This

concerns particularly to emissions from small-scale wood combustion. A preliminary com-parison of the predicted results and the measurements has been presented by Pietarila et al. (2002). We have planned to conduct a more thorough model evaluation against measured data in 2003.

For the source apportionment analysis of VOC’s, new families of species will be included. Aldehydes and ketones will be measured using the new LC/MS analysis system at the Finnish Meteorological Institute. We will also start measurements of halogenated species. These measurements can be used to study, which ozone precursors are the most important ones in the Helsinki urban plume.

In Western Europe, the NOx emissions have decreased since the 1980’s as a result of air

pollution abatement measures. Analysis of the measurement data shows that the ozone concentrations have gradually increased since the 1990’s, both at the regional background stations and at urban stations in central Helsinki. The measured data also shows that the NO2

concentrations have decreased substantially more slowly, compared with those of NOx. We

have presented in this report a qualitative analysis of the reasons for these trends.

We will analyze further the trends of NO, NO2 and ozone in urban air and at the FMI rural

monitoring sites. We have planned to study in more detail the influence of the temporal trend of the ratio of NO to NO2 in vehicular emissions. A simple numerical analysis of the influence

of the tailpipe NO2/NOx fraction and meteorological conditions on the ambient air NO2

concentrations was presented previously by Härkönen et al. (1998).

ACKNOWLEDGEMENTS

In addition to the MOBILE research programme, we are grateful for financial support from Helsinki Metropolitan Area Council, Maj and Tor Nessling Foundation and the Academy of Finland. We would also like thank Lic. Tech. Tarja Koskentalo for a good co-operation.

REFERENCES

Härkönen, J., Kukkonen, J., Valkonen, E. and Karppinen, A., 1998. The influence of vehicle emission characteristics and meteorological conditions on urban NO2 concentrations. International Journal of Vehicle Design, Vol. 20, Nos. 1-4, p. 125-130.

Hellén, H., Hakola, H., Laurila, T., Hiltunen, V. and Koskentalo, T., 2002. Aromatic hydrocarbon and methyl tert-butyl ether measurements in ambient air of Helsinki (Finland) using diffusive sampling. Science of Total Environ., 298, 55-64.

Hellén, H., Laurila, T. and Hakola, H., 2002. Source apportionment study of NMHCs in Helsinki using chemical mass balance receptor model. In: Atmospheric Chemistry within the Earth System: 7th Scientific Conference of the International Global Atmospheric Chemistry

Hellén, H., Hakola, H., Laurila, T., 2003. Source contributions of NMHCs in Helsinki using chemical mass balance (CMB) and multivariate (Unmix) receptor models. Atmospheric Environment, accepted.

Kousa, A., J. Kukkonen, A. Karppinen, P. Aarnio, T. Koskentalo, 2001. Statistical and diagnostic evaluation of a new-generation urban dispersion modelling system against an extensive dataset in the Helsinki Area. Atmos. Environ., Vol 35/27, pp 4617-4628.

Kukkonen, Jaakko, Esko Valkonen, Jari Walden, Tarja Koskentalo, Päivi Aarnio, Ari Karppinen, Ruwim Berkowicz and Raimo Kartastenpää, 2001. A measurement campaign in a street canyon in Helsinki and comparison of results with predictions of the OSPM model. Atmos. Environ. 35-2, pp 231-243.

Kukkonen, Jaakko, Leena Partanen, Ari Karppinen, Jari Walden, Raimo Kartastenpää, Päivi Aarnio, Tarja Koskentalo and Ruwim Berkowicz, 2003. Evaluation of the OSPM model com-bined with an urban background model against the data measured in 1997 in Runeberg Street, Helsinki. Atmospheric Environment (in print).

Laurila, T., 1999. Observational study of transport and photochemical formation of ozone over northern Europe. J. Geophys. Res. 104(D21): 26 235-26 243.

Mäkelä, K., 1999. LIPASTO, Calculation system for traffic emissions and energy consumption. http://www.vtt.fi/rte/projects/lipastoe/index.htm. VTT Communities and Infrastructure, Espoo, Finland.

Pietarila, H., Alaviippola, B., Hellén, H., Salmi, T., Laurila, T. and Hakola, H., 2002. The preliminary assessment under the EC air quality directives in Finland Carbon monoxide and Benzene. Report, Finnish Meteorological Institute.

Solberg S., Bergström R., Langner J., Laurila T., Sjöberg K. & Lindskog A., 2002. Changes in ozone episodes due to emission reductions. A Nordic study, EMEP/CCC-Report 10/2002