Urban Roadside Monitoring, Modeling and Mapping of Air Pollution

Abstract

Our study in its primal stage focused on modeling CO emissions emanating from Mathura road, with the use of CALINE4 model. For the execution of modeling process a set of emission factors employing simplified vehicle classification methodology were utilized. The predicted CO concentrations at the receptor location (722350mN, 3160100mE) were compared with actual monitored concentrations and a fair agreement between the two datasets was observed. 95% confidence interval was deduced for both monitored and predicted concentrations. The upper and lower confidence limits for predicted concentration were determined as 6351µg/m³ and 6243µg/m³ and for the monitored concentration, the values were obtained as 6680µg/m³and 6509µg/m³. Moreover, CALINE4 model’s performance was evaluated by determining root mean square error (RMSE) in µg/ m³ as 302 and coefficient of correlation (r) as 0.87. The study in its terminal stage involved an integration of predicted CO emissions with ArcGIS enabling GIS to post process the data in the form of digital maps. These maps highlighted the hot spots around our site which may enable planners and policy makers to formulate better pollution control strategies, lay down stringent air quality standards and address the major environmental impact assessment factors.