Extreme value theory (EVT) has been used for making predictions ofthe expected number of violations of the National Ambient Air QualityStandards (NAAQS) as prescribed by the Central Pollution ControlBoard (CPCB), India, for hourly- and eight-hour1y average carbonmonoxide(CO) concentration for an Air Quality Control Region (AQCR)comprising of an urban road intersection. Almost l00/100 contributionto the CO pollution in the region is due to vehicular traffic. Theparameters of the extreme value distribution (EVD) have beenestimated by four methods viz. method of moments, maximum likelihood,least square fit and Gumbel's fitting method. Least square fit andGumbel's method gave the best fit and predictions. A comparison ofthe predicted vio1ations of NAAQS and the exceedence of the maximumpollution concentration with that of the observed data suggests thatthe theory performs satisfactorily for the extreme air pollutionevents.
展开▼