Improving Aerosol Optical Depth (AOD) data via Regression Methods derived from Mean Combustion Efficiency (MCE) and Aerosol Index (UVAI) Data Products

Abstract

Aerosol optical depth (AOD) is a valuable parameter that reflects the density and characteristics of atmospheric aerosols, providing critical insight for studies into processes like climate change or ecological situation assessment. However, AOD data limitations arise from dense cloud cover, rendering a lack of variables over certain coverage areas. To combat this, a foray into developing a regression model that interpolates AOD data based on Mean Combustion Efficiency trends and UV Aerosol index (UVAI) correlation has been proposed. This study focuses on the California August Complex fire from August 20 to August 31, 2020, comparing UVAI, AOD, and MCE data to discern optimal regression models. Models are evaluated with correlation coefficient (CC), mean bias (MB) and normalized root mean square error (NRMSE).