High-resolution drought monitoring with Sentinel-1 and ASCAT: A case-study over Mozambique

This study investigates the potential of active microwave remote sensing to develop high-resolution drought indicators based on surface soil moisture (SSM) in Mozambique. A 500-meter resolution SSM product is derived from Sentinel-1 C-band radar backscatter using a change detection model (SSM_s1) and validated with state-of-the-art products including ASCAT, ERA5-Land, and SMAP. Results show that SSM_s1 provides consistent moisture information with greater spatial resolution compared to existing datasets. Two drought indicators are derived from SSM_s1: the soil water deficiency index (SWDI_s1) using SSM_s1 with auxiliary soil properties from Soilgrids, and the Z-score_s1, combining ASCAT long-term climatology with Sentinel-1 spatial resolution to develop an anomaly-based indicator. Both SWDI_s1 and Z-score_s1 are compared against precipitation and vegetation anomalies in six regions of Mozambique. Precipitation anomalies show low regional variability and often fail to capture drought dynamics during the dry season, yet correlate with Surface Soil Moisture (SSM) anomalies during the rainy season. The vegetation indicator detects drought with a temporal delay compared to both SWDI and Z-score_s1, suggesting that SSM provide information on earlier drought development, prior to observable vegetation anomalies. The results highlight the complementary strengths of these datasets and suggest their combined use in early warning systems. Combining information from precipitation, vegetation, and SSM_s1 enables complete monitoring of drought development. This research contributes to the development of strategies to monitor droughts and the improvement of early warning systems to improve the resilience of smallholder farmers as communities in Mozambique often rely on rain-fed agriculture.