Forecasting of particulate matter (PM10) concentration based on gaseous pollutants and meteorological factors for different monsoons of urban coastal area in Terengganu

Samsuri Abdullah, . and Si Yuen Fong, . and Marzuki Ismail, . Forecasting of particulate matter (PM10) concentration based on gaseous pollutants and meteorological factors for different monsoons of urban coastal area in Terengganu. pp. 3-17. ISSN 1823-8556

Particulate Matter (PM10) is the most conspicuous pollutant in Peninsular Malaysia having the highest air pollutant index (API) value contrasted with the other criteria contaminations. Long-term experience of PM10 may lead to a marked reduction in life expectancy due to increase in cardiopulmonary and lung disease mortality. Compelling forecasting models at local level to predict PM10 status is essential as the information generated permits the authority and local community to take prudent steps and lessen the effect of particulate contamination. The study aim to develop Multiple Linear Regression (MLR) and Principal Component Regression (PCR) models for PM10 concentration prediction in the East Coast of Peninsular Malaysia for the two main monsoon seasons. Daily observations of PM10 concentrations meteorological variables and gaseous pollutants in Kuala Terengganu Malaysia from January 2000 to December 2014 (14 years) were selected for predicting PM10 concentration level. Model comparison statistics through performance indicators using Coefficient of Determination (R2) Index of Agreement (IA) Normalized Absolute Error (NAE) Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) shows that MLR is a better model in forecasting next day PM10 concentration compared to PCR.

Full text available from: Official URL: http://jssm.umt.edu.my/wp-content/uploads/sites/51...

Abstract

Particulate Matter (PM10) is the most conspicuous pollutant in Peninsular Malaysia having the highest air pollutant index (API) value contrasted with the other criteria contaminations. Long-term experience of PM10 may lead to a marked reduction in life expectancy due to increase in cardiopulmonary and lung disease mortality. Compelling forecasting models at local level to predict PM10 status is essential as the information generated permits the authority and local community to take prudent steps and lessen the effect of particulate contamination. The study aim to develop Multiple Linear Regression (MLR) and Principal Component Regression (PCR) models for PM10 concentration prediction in the East Coast of Peninsular Malaysia for the two main monsoon seasons. Daily observations of PM10 concentrations meteorological variables and gaseous pollutants in Kuala Terengganu Malaysia from January 2000 to December 2014 (14 years) were selected for predicting PM10 concentration level. Model comparison statistics through performance indicators using Coefficient of Determination (R2) Index of Agreement (IA) Normalized Absolute Error (NAE) Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) shows that MLR is a better model in forecasting next day PM10 concentration compared to PCR.

[error in script]

Item Type:	Article
AGROVOC Term:	Coastal area
AGROVOC Term:	Forecasting
AGROVOC Term:	Pollutants
AGROVOC Term:	Monsoon climate
AGROVOC Term:	Contamination
AGROVOC Term:	Mortality
AGROVOC Term:	Sustainability
AGROVOC Term:	Environmental pollution
AGROVOC Term:	Air pollution
AGROVOC Term:	Wind speed
Depositing User:	Ms. Suzila Mohamad Kasim
Last Modified:	24 Apr 2025 06:29
URI:	http://webagris.upm.edu.my/id/eprint/24993

Actions (login required)

View Item