Soil Moisture Evolution in Hyper-arid Regions: A Comparison of InSAR, SAR, Microwave, Optical, and Data Assimilation Results in the Arabian Desert

Abstract:

Estimation of the spatial and temporal changes in soil moisture after rain events in hyper-arid regions, such as the Arabian Desert, can improve our understanding of the climate and geomorphology of these regions. However, monitoring soil moisture in hyper-arid regions is difficult because they are often remote and sparsely populated. Here, we assess constraints on soil moisture from InSAR (Interferometric Synthetic Aperture Radar) coherence by analyzing data spanning two major cyclones that impacted the southern Arabian Peninsula in 2018. We compare InSAR-derived soil moisture metrics to remote sensing constraints derived from passive microwave, active microwave, and optical data, as well as soil moisture constraints based on meteorological models, and data assimilation approaches. We compare a total of 12 different soil moisture datasets, with a particular focus on our comparison of the InSAR-derived quantity to 3 of these 12 datasets: SMAP (Soil Moisture Active Passive), ASCAT (Advanced SCATterometer), and GLDAS (Global Land Data Assimilation System).

InSAR-derived measurements of soil moisture are higher spatial resolution (~30 m) compared to passive microwave observations, real aperture radar (RAR) active microwave observations, and data assimilation/meteorological model approaches (~10-36 km). InSAR coherence records a days-to-weeks long decay in soil moisture after a rain event, which was not evident in some of the other datasets, but does appear in the GLDAS model. This is one illustration of the difference in sensitivity to changes in soil moisture of InSAR data compared to other datasets examined here. The applicability of this technique is currently limited to hyper-arid regions where there is limited vegetation and where the time interval between individual events is long relative to the time required for the soil to dry out. Future work exploring the impact of soil moisture on SAR data at different polarizations and in vegetated regions could allow for a broader use of this analysis that helps to better constrain the impact of soil moisture on studies of ground deformation.