Shomaru SUDO A Study on the Estimation of Low-Flow Recession Characteristics and Soil Evapotranspiration Minjiao LU Securing water resources is essential for daily life. In Japan, drought damage is frequently reported. Low-flow recession characteristics are commonly expressed by recession equations, including the exponential recession equation for groundwater discharge from confined aquifers and the fractional recession equation for discharge from unconfined aquifers. Some studies have suggested that the fractional form is more suitable for the Japanese archipelago. However, conventional recession equations do not represent rapid decreases in river discharge or streamflow cessation, and they do not account for evapotranspiration. Takagi analyzed recession processes using a recession equation that incorporates evapotranspiration. In this formulation, the seasonally varying recession constant is separated into a basin-specific recession constant determined by basin characteristics (e.g., geology) and seasonally varying evapotranspiration. By incorporating evapotranspiration, the time when river discharge becomes zero can be estimated. For 55 forested basins across Japan, basin-specific recession constants were analyzed, and an exponential approximation using provisional soil evapotranspiration enabled estimation of these constants for each basin. In this study, the estimated basin-specific recession constants were applied to the evapotranspiration-inclusive recession equation, and soil evapotranspiration for each recession segment was estimated from observed data. The results showed that the estimates were strongly influenced by the basin-specific recession constants. Therefore, the estimation method was revised so that the recession constant and soil evapotranspiration could be estimated simultaneously. When analyzed by single geological classes, the revised recession constants showed smaller standard deviations than conventional values. Comparisons with estimates based on geological composition ratios were distributed more closely along the 1:1 line than before. For mixed-geology basins, some estimates deviated from the 1:1 line, but detailed examination revealed problems such as extracted data not representing true recession segments; excluding these improved agreement. Estimated annual soil evapotranspiration was about 17.1% of annual precipitation, and combined with canopy evaporation (about 20%), total annual evapotranspiration was about 40% of precipitation, which is reasonable. These results demonstrate the potential to estimate both recession constants and soil evapotranspiration using a recession equation that accounts for soil evapotranspiration.