Quantifying Atmospheric Methane Point Sources with High-Resolution Satellite Observations

Abstract

The GHGSat-D demonstration satellite instrument launched in 2016 and operated by GHGSat Inc. has unlocked the possibility of monitoring atmospheric methane emissions at the facility level from space. Previous satellite missions for methane focused on improving global- and regional-scale understanding of concentrations and emissions, which required instruments with global coverage and kilometer-scale pixels. Among these were the SCIAMACHY and GOSAT instruments, which dramatically improved methane source and sink accounting, but were by design unable to resolve individual point sources. The 2017 launch of TROPOMI aboard the Sentinel-5P satellite enabled global methane mapping at unprecedented 7-km resolution, but this is still too coarse to detect all but the most extreme methane point sources. In contrast, the GHGSat measurement concept trades spatial coverage for orders of magnitude finer imaging resolution. The demonstration instrument observes methane at <50-m resolution over targeted 12×12 km2 domains. These measurements provide new capabilities for national emission reporting and global methane budget analyses. Here we develop source rate retrieval algorithms for interpreting high-resolution satellite observations of methane point source plumes in terms of facility-level emissions, and evaluate the algorithms' performance on synthetic plume datasets generated by large eddy simulation (Chapter 1). We apply these algorithms to the GHGSat-D discovery of anomalously large methane emissions in an oil/gas field in Central Asia to estimate emissions from individual oil/gas production facilities (Chapter 2). We then develop wind-adjusted time-averaging techniques to improve the GHGSat detection threshold by combining plume observations from multiple satellite passes, and use these techniques to estimate time-averaged methane emissions from individual coal mine vents in the United States, Australia, and China (Chapter 3).