Evaluating Changes in SWE from 2020-21 NASA UAVSAR with Ground-Penetrating Radar at Cameron Pass, Colorado

InSAR-based methods are a promising approach for measuring snow water equivalent (SWE) at high spatial resolution (<100 m) from space, particularly at L-band frequencies (1-2 GHz) where backscatter is less sensitive to snow grain size. The NASA SnowEx 2020 and 2021 Time-Series Campaigns emphasized L-band SAR image acquisitions from the airborne UAVSAR instrument to create weekly to bi-weekly InSAR time-series for the evaluation of this method's potential for maintaining coherence and mapping changes in SWE over a range of recurrence intervals. We analyzed datasets collected in a high elevation (3100 m) continental seasonal snowpack near Cameron Pass, Colorado, which was one of 13 acquisition sites in 2020 and one of seven in 2021. For each flight date (±1 day), we surveyed changes in snow conditions (i.e., depth, density, and SWE) using L-band ground-penetrating radar (GPR) surveys, manually probed snow depths, and snow pit measurements. When both atmospheric delay and local incidence angle are accounted for, InSAR phase changes (Δϕ) in a snowpack are primarily driven by the addition or removal of SWE (ΔSWE) or liquid water content between acquisitions. For the survey dates evaluated here, conditions were dry at the field site. We binned GPR measurements of two-way travel time and probed snow depths to the UAVSAR grid and converted them to SWE using a bulk snow density measured in the snow pit. We surveyed a variety of land covers, including open meadows, sub-alpine spruce-fir forest, and burned areas from the 2020 Cameron Peak wildfire, covering an average of 500 UAVSAR pixels (4.7 m x 6.2 m) with coherence >0.6 per survey date. GPR SWE-retrievals and SWE from binned snow depths increased from 353 to 491 mm between February and March 2020 and from 140 to 371 mm between January and March 2021. Unwrapped Δϕ was converted to ΔSWE using the Guneriussen et al. (2001) equation and snow pit-measured densities. We explore relations between coherence and land cover and evaluate UAVSAR ΔSWE-retrievals with GPR ΔSWE-retrievals. Results are presented in context of the upcoming L-band SAR satellite, NISAR, a joint NASA-ISRO mission.