Simplified Fully Automatic Atmospheric Correction for THEMIS Infrared Data

The Thermal Emission Imaging System on Mars Odyssey spacecraft has acquired 10s of thousands of high spatial resolution (~100 m/pixel), multispectral (10 bands) thermal infrared images that have been used to investigate minerals across the Martian surface. In order to extract the spectral features of the surface mineralogy, the atmospheric component has to be well constrained and removed from the thermal radiance. However, the complexity of atmosphere-surface separation and the lack of automatic processing technique have limited the usage of THEMIS data in geological studies of Mars. In this work, we present a fully automated technique to simplify the traditional atmospheric correction method for THEMIS data using Davinci software package (developed and maintained by Arizona State University). The fully automated method uses individual Thermal Emission Spectrometer (TES) pixels to isolate the atmospheric component for the local area that is directly covered by the TES footprints within a THEMIS image. Each atmospheric component is then removed from the local THEMIS thermal radiance and the THEMIS image is modified to only areas where TES footprints meeting sufficient quality control constraints are available. The method has been applied to several locations (ranging from olivine-rich unit to quartz-bearing terrain) that have been well documented in previous studies, to assess its capability of determining surface spectral features. The fully automatic method demonstrates its simplicity, viability, and robustness by which more consistent results have been generated for the same area across multiple THEMIS images and for areas with large topographic gradients, possessing the potential to characterize Martian surface compositions more accurately. The tool developed in this work will broaden the utilization of THEMIS infrared spectral data to wider planetary science community.