Remote Sensing of Soil Water Retention Signatures Using Sentinel-2 Time-Series and Exponential Decay Fitting Model

Soil water retention capacity (SWRC) is vital for agriculture and watersheds, but traditional measurements are hindered by destructive sampling and spatial discontinuity. This study selected Youyi and Heshan farm in Heilongjiang Province as the study area, using the time-series Normalized Difference Water Index (NDWI) from Sentinel-2 during the snowmelt-to-bare-soil window as a soil water retention signature (SWRS) for monitoring SWRC. The exponential decay fitting model (EDFM) was used to construct a Soil Moisture Decay Index (SMDI) to analyze the spatial patterns of the SWRC. Results showed that: (1) time-series NDWI exhibited distinct exponential decay signatures varying with soil textures and degradation gradients; (2) the EDFM effectively fitted the time-series NDWI (R² = 0.84-0.99), extracting decay rate and stable level to quantify SWRC; (3) SMDI showed high consistency with in situ soil moisture (R = 0.82-0.88) and measured field capacity (Youyi Farm: R² = 0.56; Heshan Farm: R² = 0.59), and correlated significantly with soil organic matter (R² = 0.61-0.71) and texture (R² =0.50-0.64), confirming the physical controls on water retention; and (4) SMDI spatial distribution revealed distinct degradation patterns across varying topographic and soil conditions. This study innovatively transformed point-scale static SWRC measurements into spatially continuous monitoring, offering new tools for precision water management and degraded-soil restoration, with strong theoretical and practical value.