Effects of land surface temperature on PM10 local variations using [Landsat 8-TIRS] and [GWR approach] / Norziana Ab.llah

Air pollution is one the environmental issues that brings global warming. The main factors that lead to the deterioration of air quality are the development and industrial growth and increasing energy consumption. The conventional regression approach such as Ordinary Lease Square (OLS) is not pos...

Full description

Bibliographic Details
Main Author: Ab.llah, Norziana
Format: Thesis
Language:English
Published: 2019
Subjects:
Online Access:http://ir.uitm.edu.my/id/eprint/22760/
http://ir.uitm.edu.my/id/eprint/22760/1/TD_NORZIANA%20AB.LLAH%20AP%20R%2019.5.PDF
Description
Summary:Air pollution is one the environmental issues that brings global warming. The main factors that lead to the deterioration of air quality are the development and industrial growth and increasing energy consumption. The conventional regression approach such as Ordinary Lease Square (OLS) is not possible to represent the effects of temperature to pollutant parameter at local level. The aim of this study is to determine the effects of land surface temperature (LST) on PMIO local variations using Landsat 8- Thermal Infrared Sensor (TIRS) and Geographically Weighted Regression (GWR) approach in 2015 at Penang. The objectives of this study are to generate land surface temperature of study area, to extract value of LST and PMIO on virtual stations, and to determine the relationship between LST and PMIO. The concentrations of PMIO were taken based on CAQM stations at Universiti Sains Malaysia (USM), and two stations at Prai while land surface temperature was derived from a thermal band of Landsat 8-TIRS. By using kiiging interpolation on virtual stations, there were 565 locations utilized to extract the values of LST and PMIO. The strong relationship between LST and PMIO is found based on the local regression analysis of the GWR, (r^=0.606). The result also improved the global regression analysis (r^= 0.026) in the relationship between both variables. In the local variation of the relationship between LST and PMIO, there were 70 locations foxind statistically significant. In conclusion, the regression between satellite remote sensing and local regression approach such as Landsat 8-TIRS and GWR is possible to be used to determine the effect of land surface temperature towards air pollution.