Abstract

The vast equatorial sand seas of Saturn's moon Titan are of particular interest for understanding surface processes and the budget of organic materials on Titan's surface. Previous studies have described sand dune morphology, transport, collection, and distribution patterns utilizing Cassini Imaging Science Subsystem (ISS), Visual and Infrared Mapping Spectrometer (VIMS), and Synthetic Aperture Radar (SAR) imagery. Studies have also shown evidence both for and against a global interconnected system of sand transport between Titan's major sand seas. This study worked both to create a higher resolution dataset of sand distribution using trimodal thresholding techniques applied to SAR images, as well as provide more evidence toward determining whether Titan's sand seas are a global interconnected system or separate, individually confined sand seas with limited interaction. Trimodal image thresholding was utilized on the basis of the tendency for SAR imagery to show three main groupings of grayscale pixel brightness within sand sea areas: radar-dark linear dunes, radar-bright non-dune terrain, and a middle value that visually correlates with radar-dark interdunes and other smooth terrain. Image histograms resemble normal distributions, and three-peak Gaussian curve fitting was applied to SAR swaths that had sufficient dune material and good resolution. Boundary values were determined, and images were then thresholded into three groups: Dark pixels, Middle pixels, and Bright pixels. Within sand sea areas, these populations are interpreted as Dark = sand dunes, Middle = thinning of dune forms downslope and some sandy interdunes, and Bright = non-sandy interdunes and non-dune terrain. This interpretation is strengthened by comparison with thresholded radar imagery from Earth showing strong correlation of the Dark and Middle populations with sandy material. Comparison of areal percentages between four major sand seas (Fensal/Aztlan, Senkyo, Belet, and Shangri La) showed between 45-54% coverage for the Dark population, 30-32% coverage for the Middle population, and 14-22% for the Bright population. Sand seas showed similarity to one another in these relative proportions. The Dark and Middle populations tend to increase toward the centers and south-central portions of sand seas and in topographic basins where previous studies also observed thickening of organic sand deposits, suggesting collection of sand and possible decrease in sand transport rates in these regions. Sand also accumulates upwind of topographic obstacles and within corridors between large topographic obstacles. These corridors, such as in Fensal/Aztlan or Senkyo, can be found connecting larger areas of sand to each other within and between sand seas. The Dark and Middle populations tend to decrease toward the edges of sand seas - especially toward the northern edges - and mirror an increase in the Bright populations toward these edges. This suggests either (or both) an increase in sand transport rates near the edges of sand seas, or a decrease in sand material availability. Finally, a decrease in the variability of the three populations (in particular the Middle population) and in the visual periodicity of percentage results from the western edge of Fensal/Aztlan eastward to the eastern edge of Shangri La agrees well with results from previous studies showing a decrease in dune form statistical variability eastward. This evidence, along with the presence of corridors of thick sand dunes and material between sand seas, lends weight to the argument for a global interconnected system of sand transport, rather than individually confined sand seas.

Degree

College and Department

Computational, Mathematical, and Physical Sciences; Geological Sciences

Rights

https://lib.byu.edu/about/copyright/