THE MALAYSIA PM10 ANALYSIS USING EXTREME VALUE

Abstract

The study of air quality is closely associated to air pollution. Air pollution is of the main concerns of the authority in view of the fact that it can generate damaging effects to human health, crops and environment. This paper assesses the use of Extreme Value Distributions (EVD) of the two-parameter Gumbel, two and three-parameter Weibull, Generalized Extreme Value (GEV) and two and three-parameter Generalized Pareto Distribution (GPD) on the maximum concentration of daily PM10 data recorded in the year 2010 - 2012 in Pasir Gudang, Johor, Bukit Rambai, Melaka and Nilai, Negeri Sembilan. Parameters for all distributions were estimated using the method of Maximum Likelihood Estimator (MLE). The goodness-of-fit of the distribution was determined using six performance indicators namely; the accuracy measures which include Prediction Accuracy (PA), Coefficient of Determination (R2), Index of Agreement (IA) and error measures that consist of Root Mean Square Error (RMSE), Mean Absolute Error (MAE) and Normalized Absolute Error (NAE). The best distribution was selected based on the highest accuracy measures which are close to 1 and the smallest error measures. The result showed that the Generalized Extreme Value (GEV) distribution was the best fit for daily maximum concentration for PM10 for all monitoring stations. The GEV gave the smallest errors (NAE, RMSE and MAE) and the highest accuracy measures (PA, R2 and IA) when compared to other distributions. The method gave the accuracy of more than 98% in PA, IA and R2 for all stations. The analysis demonstrated that the estimated numbers of days in which the concentration of PM10 exceeded the Malaysian Ambient Air Quality Guidelines (MAAQG) of 150 µg/m3 were between ½ and 2 days.

Keywords

• Extreme Value Theory
• PM10
• air pollution
• prediction