SPATIO-TEMPORAL TRENDS OF AIR POLLUTION DUE TO WILDFIRES IN CALIFORNIA: INFERRED FROM MODIS MAIAC AND SENTINEL-5P

California has a long history of large-scale severe fire events, and smoke plumes emitted from these fires have resulted in heavy regional air pollution. With the advancement in Earth observation satellites and cloud computing capabilities in remote sensing, long-term analysis of air pollution has become possible using satellite images. This study utilizes the Google Earth Engine (GEE) platform to navigate its geospatial datasets of MODIS MCD19A2 and Sentinel-5P products to investigate the impact of (2010-2020) wildfires on air quality in California. MODIS MCD19A2 (Version 6, level 2) uses an advanced MAIAC algorithm to produce 1-km resolution images, which retrieves Aerosol Optical Depth (AOD) at 470 and 550 nm wavelengths. In this study, the MODIS 1-km MYD14A1 (V6) dataset is used to identify the historical fire locations, while the ground-based sun photometers AERONET AOD measurements are used to validate the MODIS AOD measurements. The mean absolute error, the relative mean bias, and the root mean square error were extracted to check for uncertainty in the analysis, meanwhile, the correlations between MODIS and AERONET measurements were investigated through linear regression. The average monthly MODIS MAIAC AOD at 470 and 550 nm tended to overestimate AOD by 19 and 4%, respectively, compared to the AERONET AOD values. The correlation coefficient and the adjusted R-squared value varied from 0.78 to 0.80 and from 0.60 to 0.65, respectively, for AOD measurements at 550 and 470 nm. Better correlations were observed during the fire season (May-October), with correlation coefficients above 0.8 and adjusted R-squared values above 0.65 at both wavelengths. Sentinel-5P data showed that the 2020 wildfires significantly raised the NO₂ concentration in its surrounding areas. These observations of spatio-temporal trends of air pollution due to wildfires in California can help in making informed decisions and can improve prevention and mitigation programs.