Page 67 - Special Topic Session (STS) - Volume 2
P. 67

STS459 Norshahida S. et al.

                           Probability distribution model for predicting
                            ozone (O3) exceedances at two air quality
                          monitoring sites in Malaysia during dry season
                                                                    1
                               Norshahida Shaadan  1,2,3 ; Nabihah Jasri
              1  Centre for Statistical and Decision Science Studies, Faculty of Computer & Mathematical
                                            Sciences, UiTM
             2  Advanced Analytic Engineering Center, Faculty of Computer & Mathematical Sciences, UiTM
                                 3  Business Datalytic Research Group, UiTM

            Abstract
            Tropospheric Ozone (O3) is one of the strongest atmospheric oxidants and has
            become  an  important  criteria  pollutant  in  the  Malaysia  environment  other
            than PM10. Many studies worldwide have proven that, high concentration of
            O3 contributes to a certain environmental problem including health problems,
            vegetation and materials, as well as climate changing. Thus due to the facts, it
            is necessary to gain a good understanding of the characteristics of O3 pollution
            so that the information can be the input for managing the problem. In this
            study, several probability distributions including Gamma, Lognormal, Normal,
            and Weibull were compared with the aim to find the best distribution that can
            fit the O3 at two selected air quality monitoring stations in Malaysia located in
            Shah Alam and Putrajaya. Based on the two years (2013 and 2014) period of
            hourly recorded data, the model parameters were estimated using the method
            of maximum likelihood estimator (MLE). The best distribution was determined
            using the plot of cumulative distribution function (CDF) and the goodness of
            fit  statistic  including  the  Kolmogorov-Smirnov,  Cramer-von  Mises  and
            Anderson-Darling. The study results have shown that Weibull is found to be
            the  best  model  for  Shah  Alam  while  Gamma  model  is  for  Putrajaya  with
            expected exceedance return period of seven days.

            Keywords
            Ozone; Prediction Model; Probability Distribution; Pollutant Exceedances; Air
            Quality

            1.  Introduction
                Nowadays, air pollution has become a global problem. Over exposed to
            polluted air that contained mixed hazardous gases, dust or hazes has been
            proven  to  be  negatively  affect  human’s  health,  animals,  plants  and  the
            surroundings  (Verma  et  al.,  2015;  Felzer  et  al.,  2007).  The  substances  that
            caused air pollution are called pollutants. In the Malaysia environment, Ozone
            (O3) has been identified as the second most dominant pollutant other than
            particulate matter (PM10). Ozone (O3) is defined as a secondary type of gas
            pollutant  and  also  known  as  photochemical  oxidant  that  is  formed  via

                                                                56 | I S I   W S C   2 0 1 9
   62   63   64   65   66   67   68   69   70   71   72