UCLA

Slow-moving landslides are destabilized by precipitation. The continuous soil moisture monitoring aid the understanding of landslide processes. Here we study time-series soil moisture using NASA's Uninhabited Aerial Vehicle Synthetic Aperture Radar (UAVSAR) at 6-m resolution for a seasonally active grassland landslide in the northern California Coast Ranges, USA. A physically based radar scattering model is used to retrieve the near-surface (5-cm depth) soil moisture. The forward model is developed for polarization HH and VV. The soil moisture retrieval using HH&VV shows unbiased RMSE (ubRMSE) of 0.058 m3/m3. From the Freeman-Durden decomposition for UAVSAR's time-series data, the surface scattering and double bounce dominate the landslide area which suggests the strong correlation with soil moisture for the data. The physical-model based algorithm can be applied to other grassland covered landslides in California to retrieve soil moisture.