Application of extreme value theory for predicting violations of air quality standards for an urban road intersection

Extreme value theory (EVT) has been used for making predictions of the expe cted number; of violations of the National Ambient Air Quality Standards (N AAQS) as prescribed by the Central Pollution Control Board (CPCB), India, f or hourly- and eight-hourly average carbon monoxide(CO) concentration for a n Air Quality Control Region (AQCR) comprising of an urban road intersectio n. Almost 100% contribution to the CO pollution in the region is due to veh icular traffic. The parameters of the extreme value distribution (EVD) have been estimated by four methods viz, method of moments, maximum likelihood, least square fit and Gumbel's fitting method. Least square fit and Gumbel' s method gave the best fit and predictions. A comparison of the predicted v iolations of NAAQS and the exceedence of the maximum pollution concentratio n with that of the observed data suggests that the theory performs satisfac torily for the extreme air pollution events. (C) 1999 Published by Elsevier Science Ltd. All rights reserved.