Comparison of regression models with land-use and emissions data to predict the spatial distribution of traffic-related air pollution in Rome

Abstract

Spatial modeling of traffic-related air pollution typically involves either regression modeling of land-use and traffic data or dispersion modeling of emissions data, but little is known to what extent land-use regression models might be improved by incorporating emissions data. The aim of this study was to develop a land-use regression model to predict nitrogen dioxide (NO 2 ) concentrations and compare its performance with a model including emissions data. The association between each land-use variable and NO 2 concentrations at 68 locations in Rome in 1995 and 1996 was assessed by univariate linear regression and a multiple linear regression model that was constructed based on the importance of each variable. Traffic emissions (particulate matter, carbon monoxide, nitrogen oxides, and benzene) were estimated for 164 areas of the city based on vehicle type, traffic counts and driving patterns. Mean NO 2 concentration across the 68 sites was 46.8 mg/m 3 (SD 9.8 mg/m ). The most important predicting variables were the circular traffic zones (main ring road, green strip, inner ring road, traffic-limited zone), distance from busy streets, size of the census block, the inverse population density, and altitude. A multiple regression model including these variables resulted in an R 2 of 0.686. The best-fitting model adding an emission term of benzene resulted in an R 2 of 0.690, but was not significantly different from the model without emissions (P ¼ 0.147). In conclusion, these results suggest that a land-use regression model explains the traffic-related air pollution levels with reasonable accuracy and that emissions data do not significantly improve the model.

A land use regression model for estimating the NO2 concentration in shanghai, China

“…Limited by data accessibility of traffic intensity, traffic counts and traffic emissions, which was also a problem in many other studies with the LUR model, we relied on road lengths and distance to the nearest road in this study. Previous studies have successfully applied the use of these two predictor variables, and some of them have reported that the performance of LUR models only using the lengths of roads were comparable with those using the traffic intensity or the traffic emission data for explaining the small-scale variability of pollutants concentrations (Henderson et al, 2007;Rosenlund et al, 2008). Population density is another predictor variable used frequently in LUR models for predicting NO 2 concentration.…”

Traffic-related air pollution in relation to respiratory symptoms, allergic sensitisation and lung function in schoolchildren

“…The NO 2 concentration in the census block of residence (a small area of about 500 inhabitants per census block) was estimated for each child by a land use regression model described in detail elsewhere 24. This model was built by assessing the association between variables describing land use and traffic information and NO 2 concentrations measured at 68 sites in Rome.…”

Air pollution forecasting in Ankara, Turkey using air pollution index and its relation to assimilative capacity of the atmosphere

“…Degradation of urban air quality attracted attention because of its adverse effects on human health (Makri and Stilianakis 2008;O'Connor et al 2008;Bernstein et al 2008;Kampa and Castanas 2008;Brunekreef 2007), its impact on material (Lorenzo et al 2007;Vallet et al 2006) and on visibility (Rokjin et al 2006;Young et al 2006;Hand et al 2002). Pollutant emissions, particularly from combustion for space heating and transportation (Braniš et al 2009;Tsikardani et al 2006;Bellander et al 2001;Rosenlund et al 2008;Costabile and Allegrini 2008;Yuval et al 2008;Fenger 1999), local meteorology (Fisher et al 2001(Fisher et al , 2006Aldrin and Haff 2005), and topography (Sládek et al 2007;Kadja et al 1998), are the main contributors to urban air pollution episodes.…”