Characterization of anthropogenic methane plumes with the Hyperspectral Thermal Emission Spectrometer (HyTES): a retrieval method and error analysis
- 1Joint Institute for Regional Earth System Science & Engineering, University of California, Los Angeles, USA
- 2Jet Propulsion Laboratory, California Institute of Technology, Pasadena, California, USA
- 3Department of Applied Mathematics, University of Washington, Seattle, Washington, USA
Abstract. We introduce a retrieval algorithm to estimate lower tropospheric methane (CH4) concentrations from the surface to 1 km with uncertainty estimates using Hyperspectral Thermal Emission Spectrometer (HyTES) airborne radiance measurements. After resampling, retrievals have a spatial resolution of 6 × 6 m2. The total error from a single retrieval is approximately 20 %, with the uncertainties determined primarily by noise and spectral interferences from air temperature, surface emissivity, and atmospheric water vapor. We demonstrate retrievals for a HyTES flight line over storage tanks near Kern River Oil Field (KROF), Kern County, California, and find an extended plume structure in the set of observations with elevated methane concentrations (3.0 ± 0.6 to 6.0 ± 1.2 ppm), well above mean concentrations (1.8 ± 0.4 ppm) observed for this scene. With typically a 20 % estimated uncertainty, plume enhancements with more than 1 ppm are distinguishable from the background values with its uncertainty. HyTES retrievals are consistent with simultaneous airborne and ground-based in situ CH4 mole fraction measurements within the reported accuracy of approximately 0.2 ppm (or ∼ 8 %), due to retrieval interferences related to air temperature, emissivity, and H2O.