1. UKM Learning and Research Repository
2. PHD Thesis / Tesis PHD
3. Faculty of Science and Technology / Fakulti Sains dan Teknologi

Title:	Variation of surface ozone concentration and its precursors around the Klang Valley, Malaysia
Authors:	Negar Banan (P62662)
Supervisor:	Mohd Talib Latif, Prof. Dr.
Keywords:	Surface O3 concentration Secondary pollutant Linear regression Universiti Kebangsaan Malaysia -- Dissertations Dissertations, Academic -- Malaysia
Issue Date:	18-Nov-2015
Description:	Surface O3 is one of the key secondary pollutant with adverse effects on both human beings and vegetation. This study specifically conducted to determine spatial-temporal pattern of surface O3 concentration, its precursors in the Klang Valley and to forecast the daily maximum surface O3 concentration using linear regression. The variations of surface O3 and its precursor concentrations in the Klang Valley were using a nine years period database, between 2003 and 2011 recorded by the Malaysian Department of the Environment (DOE). The multivariate techniques of hierarchical agglomerative cluster analysis (HACA) was used to explore the spatial patterns, and principal component analysis (PCA) to determine the possible sources of the air pollution. The artificial neural network (ANN) and the nonlinear autoregressive exogenous (NARX) as non-linear models, and multiple linear regression (MLR) as a linear model, were utilised to forecast the daily maximum surface O3 concentration in year 2011. The overall results indicate that the Cheras station showed a daily maximum average value of 196 ± 35 ppbv, while the Shah Alam showed the daily maximum average value of 172 ± 26 ppbv. Shah Alam was found to have the highest number of hours and days for O3 concentrations, while Cheras had the highest maximum O3 concentration in the urban area compared to other stations in year 2004. Among all available precursors, NO has the most influence on the maximum O3 concentration. Pollution patterns indicate that, except Cheras, there is a high surface O3 concentration between February and April, while a low surface O3 concentration is observed almost in January, July and December. It was noticed that wind direction during different monsoons influences the concentration of O3 around the Klang Valley. PCA result for cluster one with several stations indicates that the total variance of the first factor (F1) is 28.4% confirming strong positive factor loadings for NOx (0.834) and NO (0.796). The second factor, (F2) shows 43.948 of total variance with a strong positive loadings on PM10 (0.675) but strong negative loadings on AT (-0.717). The PCA result shows the main sources of air pollution as motor vehicle exhaust emissions and gases released from industrial activities in cluster 1. Observed O3 and forecasted O3 by NARX model confirmed that Cheras and Shah Alam stations experienced the highest number of unhealthy O3 days (100 ppbv >) in 2011 compare to other sampling stations. Tanjung Malim and Putrajaya stations approximately had close number of good and moderate O3 days. Nilai, Klang and Petaling Jaya stations experienced the good air quality since more than 200 good O3 days were observed in these stations in year 2011. The best correlations between observed and forecasted O3 for all sampling stations obtained by NARX model at hidden neurons with the value of 255. It concluded that the heavy traffic caused by many daily commutes within the stations, vast residential and commercial areas, large number of population and industrial activities are the reasons for high number of moderate and unhealthy O3 days.,Ph.D.
Pages:	269
Publisher:	UKM, Bangi
Appears in Collections:	Faculty of Science and Technology / Fakulti Sains dan Teknologi

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.