Quantifying the Impact of Environmental Factors on the Methane Point-Source Emission Algorithm

Methane (CH4) emissions in coal-energy-rich regions are characterized by hidden emission point sources and highly variable emission rates. While the Matched Filter (MF) method for detecting the CH4 point source using hyperspectral satellite sensors has been validated for high-emission concentrations, the accurate inversion of low-concentration emissions in complex environments remains challenging. In this study, an ‘end-to-end’ experiment—from emission simulations to satellite spectra and inversion results—has been designed to quantify the impact of internal payload parameters and environmental parameters for CH4 emission inversions, and perform real-scenario calculations. The study reveals several key findings: (1) Under ideal conditions, 15% of satellite spectral noise contributes to a 13% bias in CH4 detection inversion, and a spectral resolution of 10–14 nm allows the detection of CH4 emissions with concentrations as low as 350 ppb, above the background level of 1900 ppb. (2) For near-surface aerosols at 2100 nm, an aerosol optical depth (AOD) of 0.1 leads to a low bias of −51.6% with water-soluble aerosols and a strong bias of −69.2% with black carbon aerosols, while dust aerosols induce a medium bias of up to −60.7%. (3) The height of the aerosol layer affects the accuracy of methane inversion, which is up to 7.3% higher under aerosol conditions at 3 km than under aerosol conditions near the ground. (4) When the CH4 emission source and its diffuse plume are located above a high-reflectance (bright) surface, while the background CH4 concentration is associated with a low-reflectance (dark) surface, the significant reflectance contrast between the two surfaces leads to a rapid degradation in inversion accuracy. This contrast makes it impossible to effectively extract CH4 signals when the reflectance difference reaches 0.2. (5) Under harsh conditions, where multiple parameters are present (AOD = 0.2, albedo = 0.2, aerosol layer height (ALH) = 2), the MF method is still able to detect CH4 emissions, but with a significant error of 74.65%. (6) External environmental variables, particularly atmospheric pressure and water vapor content, significantly influence the inversion accuracy of methane (CH4) concentrations. Variations in atmospheric pressure induce deviations in the CH4 concentration distribution, resulting in an average inversion error of −12.06%. Similarly, elevated water vapor levels can lead to a maximum error of −16.2%. These findings highlight the substantial challenges in accurately detecting low-concentration CH4 emissions. The results offer critical insights for refining CH4 detection algorithms and enhancing the precision of satellite-based inversions for low-concentration CH4 point-source emissions.