Detection of elevated CO₂ and CH₄ emissions from power plant and landfill sites using airborne and spaceborne imaging spectroscopy

Accurate quantification of atmospheric carbon dioxide (CO₂) and methane (CH₄) emissions from point sources is crucial in advancing our understanding of atmospheric composition and mitigating climate change. In this study, we have established an advanced remote sensing framework using imaging spectrometer data to detect and estimate anomalous concentrations of CO₂ and CH₄ in the atmosphere from targeted point sources, employing the spectral normalized matched filter (SNMF) algorithm. This algorithm utilizes the distinct absorption features of CO₂ and CH₄ in the shortwave infrared (SWIR) bands, enabling precise discrimination of emission sources from the atmospheric background. The practical benefits of SNMF are robust detection, background suppression, and better adaptiveness. The SNMF algorithm was rigorously tested at two key sites in tropical atmospheric conditions: Kota Super Thermal Power Station (KSTPS) in Rajasthan, India, for CO₂ emission detection and Pirana landfill site in Ahmedabad, India, for CH₄ emission detection. Multiple cloud-free satellite observations from PRISMA and EMIT have been used here, and airborne AVIRIS-NG data were also included in the study. The analysis revealed the robustness of the SNMF algorithm for CO₂ and CH₄ detection in different seasons at the same site, and a comparison was also made between PRISMA and EMIT retrievals. We estimated annual CO₂ emissions from KSTPS of 6.35 ± 0.65 megatons and CH₄ emissions from the Pirana landfill of 0.80 ± 0.14 megatons. Wind data have been incorporated for uncertainty analysis to enhance the robustness of the emission estimation, and external emission data has been analyzed for validation.