Novel Measurements of Tropospheric Ethane from the Cross-track Infrared Sounder

Atmospheric ethane is the most abundant non-methane hydrocarbon in the troposphere. It is emitted predominantly from the oil and gas sector and is an important tracer for partitioning anthropogenic methane emissions. Space-based measurements of tropospheric ethane would offer valuable new information for understanding its global sources and for constraining global methane budgets. Prior work in our group used a machine learning approach (Retrieval of Organics from CrIS Radiances; ROCR) to develop a high-sensitivity, low-noise, and high space-time resolution satellite-based measurement of atmospheric isoprene in the thermal infrared. In this work, we extend the ROCR framework to provide a satellite-based measurement of ethane with sensitivity to the lower troposphere. First, the CrIS detection of ethane is demonstrated via spectral analysis of both wildfire and fossil fuel plumes. We then apply the ROCR neural network retrieval to quantify ethane column abundances based on the corresponding CrIS-observed hyperspectral range index (HRI). The HRI methodology increases near-surface sensitivity while reducing impacts from interferants, a necessity given the relatively weak absorption of atmospheric ethane. To evaluate our retrieval, we compare our column retrievals to column FTIR observations of ethane from the NDACC network. Our retrieval is highly correlated with these measurements but underestimates column ethane by a factor of two. We use our retrieval to look at the strength and variability of multiple ethane sources around the globe, with special focus on the Permian Basin, which stands out as a unique hotspot for global ethane emissions.