DISENTANGLING SEDIMENT SIGNALS ON ALLUVIAL FANS IN SYNTHETIC APERTURE RADAR

Synthetic aperture radar (SAR) has the advantage of discerning fine-scale near-surface characteristics independent of atmospheric conditions. Thus, SAR images have been useful for studying not just the Earth, but Venus, Saturn’s moon Titan, and several other planetary bodies. Radar backscatter depends on wavelength-scale surface properties (mainly roughness and composition) and imaging geometry. For a dry alluvial deposit, surface roughness may be due to sediment properties of size, shape, sorting, and the organization of that sediment into channels, bars, and other mesoscale topography. Differentiating among these properties and other influences on radar backscatter is advanced by studies exploring a variety of field conditions.

We collect ground truth data at 30 study sites on alluvial fans around Death Valley, CA, to develop an empirical model of sediment-dependent SAR backscatter. With slight variations based on local conditions, we recorded triaxial size measurements and ordinal shape classes of 192 grains selected using a regularly spaced sampling grid over each 20 by 40 meter study area. Further surface roughness characterization was captured with structure-from-motion analysis of ground-based images centered around representative patches. For correlation with the field data, high resolution, tightly co-registered SAR images (~10 m/pixel) are being analyzed at multiple wavelengths: 5.5 cm (Sentinel-1 C-band) and 23.6 cm (PALSAR L-band). Mean intermediate-length axes measured in the field range from ~1 to ~23 cm (0.2 to 4 λ_C-band and 0.1 to 1 λ_L-band). Across these sizes, initial results show a ~12 times stronger backscatter dependence on wavelength-normalized grain size (D/λ) for L-band compared shorter C-band wavelengths, possibly due to the relevant scale and subsurface penetration.

Ongoing tests will assess how various sediment properties influence backscatter in alluvial environments compared to roughness-based approaches and scattering theory. These results ultimately have implications for interpreting SAR data where ground truth is not available, such as less accessible parts of Earth and other planetary bodies. For such distant locations, the characterization of unconsolidated sediment will support remote studies of sedimentological conditions and transport processes.