SEASONAL VARIATION OF PARTICLE NUMBER COUNT (PNC) AT THE COASTAL ENVIRONMENT IN PENINSULAR MALAYSIA USING BOOSTED REGRESSION TREES (BRTS)
DOI:
https://doi.org/10.46754/umtjur.v2i3.165Keywords:
Particle number count, concentration, seasonal analysis, boosted regression treeAbstract
The concentrations of airborne particulate matter (PM) is often measured as a mass concentration. However, the other way to express particulate matter is by using the Particle Number Count ([PNC]) concentrations. This study aims to analyse the seasonal variation of airborne particulate matter in terms of [PNC] by using R packages and the Boosted Regression Trees (BRTs) technique. The study was conducted at IOES, Universiti of Malaya in Bachok, Kelantan. The monitoring was important to understand the variability of seasonal effects due to different seasons. In this work, only the datasets for three seasons (Inter Monsoon, North East Monsoon and South-West Monsoon) were analysed involving 25,958 data. The air quality monitoring equipment involved was the particle counter Environment Dust Monitor GRIMM Model 180 and a weather station for recording the meteorological parameters. The data analysis was completed by using R software and its package for evaluating seasonal variability and providing the statistical analysis. The relationship between variables was studied by using the Boosted Regression Tree (BRT) technique. The interaction between independent variables towards the [PNC] in different seasons was discussed. The best setting result of BRT model evaluation R² is 0.22 (North-East Monsoon), 0.87 (Intern monsoon 1), and 0.59 for South West Monsoon which indicated that the model developed is acceptable except for NEM and intern monsoon seasons. Temperature (57 %) and wind direction (67%) were found to be the highest factor influenced by the formation of [PNC] concentrations in this area. Finally, good results indicated that BRT technique is an acceptable way to analysed air pollution data.
References
AQEG 2005. (2005). Particulate Matter in the United Kingdom. Department for Environment, Food and Rural Affairs, UK.
Carslaw, D. C., & K. Ropkins, (2012). Openair – an R package for air quality data analysis. Environmental Modelling & Software.
Carslaw, D.C. & Taylor, P.J. (2009). Analysis of air pollution data at mixed source location using boosted regression trees. Atmospheric Environment, 43, 7053-7063.
Chen R. J., Chen B. H., & Kan H. D. (2010). A health-based economic assessment of particulate air pollution in 113 Chinese cities. China Environmental Science, 30(3), 410–5.
Friedman, J. H. (2001). Greedy Function Approximation: A Gradient Boosting Machine. The Annals of Statistics, 29, 1189- 1232.
Friedman, J. H. (2002). Stochastic gradient boosting. Computational Statistics & Data Analysis, 38, 367-378.
Harisson, R. M. (2000). Studies of the source apportionment of airborne particulate matter in the United Kingdom. Journal of Aerosol Science, 31, 106-107.
Linda Smith. (2009). Ambient Air Quality Standards (AAQS) for Particulate Matter. California Environmental Protection Agency. Air Resources Board.
Noor Zaitun Yahaya, Siew Moi Phang, Azizan Abu Samah, Intan Nabila Azman & Zul Fadhli Ibrahim. (2018). Analysis of Fine and Course Particle Number Count Concentrations Using Boosted Regression Tree Technique in Coastal Environment. Journal of Environment Asia, 11(3), 221- 234.
Yahaya, N., & Zulfadhli, I. (2019). The used of the Boosted Regression Tree Optimization Technique to analyse an Air Pollution data. Journal of Recent Technology and Engineering (IJRTE), 8(4), 1565-1575.
Pope III, C. A., Ezzati, M., & Dockery, D. W. (2009). Fine-particulate air pollution and life expectancy in the United States. New England Journal of Medicine, 360, 376–86.
R Development Group. (2008). R: A Language and environment for Statistical Computing. In: Computing. R. F. F. S. (ed.). Vienna, Austria.
Ridgeway, G. (2010). GBM: Generalized Boosted Regression Models. . R Package Version 1.6-3.1.
Yahaya, N. Z., Tate, J. E., & Tight, M. R. (2011a). Studying Particle Number Concentrations (PNC) in an Urban Street Canyon: Using Boosted Regression Trees BRT. Proc. The International Conference on Humanities, Social Sciences and Science Technology 2011, Manchester University UK.
Yahaya, N. Z., Tate, J. E., & Tight, M. R. (2011b). Analyzing Roadside Particle Number Concentrations using Boosted Regression Trees (BRT). Proc. Presented The European Aerosol Conference 2011, Manchester University.
Yahaya, N. Z. (2013). Spatial and Temporal variations of ultrafine particles in urban Environment. Phd Thesis Institute for Transport Studies, University of Leeds, United Kingdom.
