Study of the spatial distribution of the tension water storage capacity and hydrological simulation effects in the upper reaches of the Yangtze River

Abstract

Tension water storage capacity is a pivotal variable in conceptual hydrological models, and its spatial distribution significantly influences runoff calculations. However, accurately characterizing its spatial variability remains a significant challenge. This study, based on the physical significance of tension water storage capacity, quantitatively calculates the tension water storage capacity for 95 subwatersheds in the upper reaches of the Yangtze River. Utilizing the grid-based tension water storage capacity data, three types of tension water storage capacity curves were constructed: the calibrated parabolic linetype (CPL), the fitted parabolic linetype (FPL), and the fitted straight linetype (FSL). Theoretical comparisons and rainfall-runoff simulation experiments were carried out in eight actual subwatersheds. The results indicate that, in theoretical runoff calculations, FSL demonstrates the smallest relative error during grid-based runoff computation, with a relative error of −25.40 % before the intersection point and 1.00 % after the intersection point. Furthermore, during rainfall-runoff simulation validation, the overall average Nash-Sutcliffe Efficiency (NSE) for the eight subwatersheds using FSL was 0.85, compared to 0.80 for both CPL and FPL. This represents an improvement of 0.05 for FSL over CPL and FPL. Additionally, the average Kling-Gupta Efficiency (KGE) value for FSL was 0.83, while CPL and FPL achieved 0.77 and 0.75, respectively, indicating that FSL outperformed CPL and FPL by 0.06 and 0.08, respectively. Therefore, based on both theoretical runoff calculations and rainfall-runoff simulations, the FSL more accurately describes the spatial distribution of tension water storage capacity within the watershed. This study provides a new alternative method for the spatial distribution of tension water storage capacity in hydrological modeling, improves the calculation method of streamflow production, and contributes to the development of hydrological modeling, which is useful for the efficiency of water resources utilization and flood risk reduction.