Validation Of Satellite-Derived Aerosol Optical Depth Using Long-Term Sunphotometer Data In Malaysia

Abstract

Aerosol Optical Depth (AOD) is a dimensionless measure of the extinction of solar radiation by aerosols in the vertical column of the atmosphere and it is widely used as an indicator of air quality. It can be measured from space using satellite sensors such as MODIS, which observes the total atmospheric column from the top of the atmosphere to the surface, or from the ground using AERONET sunphotometers, which measure the attenuation of direct solar radiation from the ground to the atmospheric boundary layer. In Malaysia, satellite-derived AOD retrieval is often challenging due to frequent cloud cover, high humidity, surface heterogeneity, and spectral mismatch between MODIS wavelengths (470 nm and 550 nm) and AERONET wavelengths (440 nm and 500 nm). These differences in measurement time, spatial scale, and wavelength often cause MODIS to overestimate AOD, especially during high aerosol loading. AOD is closely linked to fine particulate matter (PM2.5) making it important for assessing air quality impacts on health and the environment. Globally, accurate AOD information supports climate and haze studies, while in Malaysia, it is essential for monitoring seasonal haze and urban pollution. This study applied spatial, temporal, and spectral correction techniques to minimise discrepancies between MODIS MAIAC AOD and AERONET Level 2.0 data from USM Penang and Kuching, Malaysia improving the reliability of satellite AOD for air quality monitoring in tropical conditions. Long-term MODIS MAIAC data (MCD19A2.061) from 2012 to 2024 for Penang and from 2012 to 2020 for Kuching were used, along with AERONET data from USM and Kuching stations. Spatial correction was applied using regression on a 3×3 MODIS pixel grid. Spectral correction used interpolation to estimate AOD at 470 nm and 550 nm. Temporal correction was performed by filtering AERONET data between 1000h until 1100h to match the MODIS Terra overpass. The corrected MODIS AOD values showed improved agreement with AERONET. In Penang, the 550 nm result showed R = 0.72 and RMSE = 0.08, while in Kuching, R = 0.97 and RMSE = 0.15. Seasonal and frequency analysis revealed that MODIS tends to overestimate AOD, especially during haze events. This study demonstrates that calibrated satellite AOD data using long-term observations can support more reliable air quality monitoring in urban Malaysian settings.