Aerosol-Calibrated Matched Filter Method for Retrievals of Methane Point Source Emissions Over the Los Angeles Basin

Abstract

Methane, with a global warming potential roughly 86 times greater than carbon dioxide over a 20-year timeframe, plays a crucial role in global warming. Remote sensing retrieval is a pivotal methodology for identifying methane emission sources, with accuracy influenced largely by surface and atmospheric properties, including aerosols. In this study, we propose an Aerosol-Calibrated Matched Filter (ACMF) algorithm to improve the traditional Matched Filter (MF) method. Our new approach incorporates an aerosol scattering correction factor to reduce the aerosol-induced bias on methane retrievals. Validating our algorithm through simulated spectra, we demonstrate that considering the aerosol scattering effect significantly reduces retrieval errors compared to MF methods by an average of approximately 90%. We apply our newly developed algorithm to hyperspectral data obtained from the Airborne Visible/Infrared Imaging Spectrometer—Next Generation in the Los Angeles Basin and focus on 11 plumes identified through case studies. Our results reveal that ACMF estimates of emission rates and inversion uncertainties exhibit an average reduction of approximately 4% compared to corresponding MF results, with deviation increasing with aerosol optical depth (AOD).