January-February 2000, Volume 10, Issue 1, Pages 4 - 14

Article

An air pollution model for use in epidemiological studies: evaluation with measured levels of nitrogen dioxide and benzene OLE RAASCHOU-NIELSEN1, OLE HERTEL2, ELISABETTA VIGNATI2, RUWIM BERKOWICZ2, STEEN S. JENSEN2, VISTI B. LARSEN1, CHRISTIAN LOHSE3 & JØRGEN H. OLSEN1 1Danish Cancer Society, Institute of Cancer Epidemiology, Strandboulevarden 49, DK-2100 Copenhagen Ø, Denmark     2National Environmental Research Institute, Frederiksborgvej 399, DK-4000 Roskilde, Denmark     3University of Odense, Department of Chemistry, Campusvej 55, DK-5230 Odense M, Denmark     Correspondence to: OLE RAASCHOU-NIELSEN, Danish Cancer Society, Institute of Cancer Epidemiology, Strandboulevarden 49, DK-2100 Copenhagen Ø, Denmark. Tel.: +45-35257617. Fax: +45-35257734. E-mail: ole@cancer.dk

Keywords

air pollution;   benzene;   epidemiology;   modelling;   nitrogen dioxide;   traffic

Abstract

The aim of the study was to evaluate the predictions derived from the Danish Operational Street Pollution Model (OSPM) when the input data are obtained by simple methods that could be used in large-scale epidemiological studies. The model calculations were thus compared with passive sampler measurements of nitrogen dioxide and benzene at 103 street locations in Copenhagen, Denmark, and at 101 locations in rural areas. Data on traffic and street configuration were collected by means of a simple registration scheme in which forms were filled out by local municipal authorities. Meteorological data were derived from routine measurements at Copenhagen airport, and data on background air pollution were based on a simple empirical model. Differences in air pollution levels between rural areas and Copenhagen and differences in nitrogen dioxide concentrations at various locations in Copenhagen were well reproduced by the OSPM. The correlation coefficients (r) between the measured and the predicted half-year average concentrations of nitrogen dioxide in Copenhagen were between 0.75 and 0.80 for various degrees of precision of the input data for the model. The results indicate that the OSPM used with the presented methods for generation of input data might be useful in assessing long-term exposure to air pollutants in epidemiological studies. Journal of Exposure Analysis and Environmental Epidemiology (2000) 10, 4-14.

Received 22 September 1998; Accepted 12 May 1999

^© Macmillan Publishers Ltd 2000