Identification of Sedimentary Processes on Alluvial Fans Using Aerial Thermal Images and Ground Truth

Introduction: We have previously demonstrated the viability of using thermal infrared images for mapping surface grain size distributions on terrestrial allu-vial fans [1, 2, 3]. For a well characterized alluvial fan in Owens Valley, CA we have used a quantitative analysis between aerial thermal images and ground-based grain size counts as well as a qualitative analysis between our thermal images and a published surface feature map [1]. Here, we combine aerial thermal images with field-based ground truth observations to study sedimentary processes on a wide assortment of fans in both Owens Valley and Death Valley, CA. High-resolution thermal infrared mapping, sensitive to the grain size of the top ~10 cm, in conjunction with visible images, provides insight into the processes that have acted on the surfaces of alluvial fans that is not available using visible images alone. Several sedimen-tary processes both build and rework the surfaces of alluvial fans, acting to sort grains by their size. Although the particle size distribution at the surface of most alluvial fans is the result of secondary weathering processes, remnant fluvial or debris-flow deposits still present at the surface can be revealed by thermal imaging. By understanding the thermal signatures of sedi-mentary deposits on terrestrial alluvial fans, we can utilize similar thermal infrared data sets to study proposed Martian alluvial fans [4, 5, 6]. To our knowledge , this study is the first thermal infrared survey of terrestrial alluvial fans. Methods: Individual daytime and pre-dawn thermal images of the target alluvial fans were taken from a light aircraft to give an overhead perspective similar to that seen from orbit, but at higher spatial resolution. We selected times of day during which surfaces of different thermal inertias had the greatest separation in temperature-3:30 to 5:00 AM local time for pre-dawn images (also a time when topographic effects are muted [7]) and 11:00 to 12:30 AM local time for daytime images (images have high signal to noise). Visible images were acquired during the same 24-hour period from the same airborne vantage points as the thermal images by using an aviation GPS navigation unit. Sophisticated multivariable analytical models for the diurnal thermal behavior of terrestrial surfaces have been developed (e.g., Watson, 2000), but here we make the simplifying assumption that thermal inertia is inversely proportional to the change in surface temperature between the pre-dawn and daytime im