Accurate runoff estimation in ungauged basins is critical for water resource management but often relies on static parameters like the runoff threshold (P0), derived from the Soil Conservation Service Curve Number method, which fail to capture spatiotemporal soil moisture variability. This study proposes an automated methodology utilising Google Earth Engine to dynamically adjust P0 by integrating daily soil moisture data from SMAP L4, land cover from MODIS, and precipitation from GSMaP. Unlike traditional approaches that use antecedent precipitation as a proxy, this method classifies moisture conditions using historical percentiles to update the threshold daily. The methodology was validated in two sub-basins within the Guadiana River basin (Spain). The results highlight a stark contrast between methods: while static regulatory values remained invariant (36 and 48 mm), the proposed dynamic model revealed significant fluctuations, with P0 values ranging from over 50 mm in dry periods down to less than 14 mm during saturation. Conversely, the proposed dynamic method effectively captures real-time soil saturation, exhibiting adaptability with reductions in P0 of up to 72% immediately following rainfall events. This satellite-based approach provides a scalable, physically consistent alternative for assessing runoff potential in data-scarce regions, significantly enhancing the reliability of hydrological modelling compared to conventional regulatory standards.
Pachón-Acuña et al. (2026) studied this question.