Streamflow from forested mountain watersheds is critical to aquatic ecosystems and social development in watersheds. However, understanding the intra-annual variability of streamflow is limited by the lack of observation of terrestrial water storage (TWS) in large-scale watersheds. This study developed a monthly Budyko framework incorporating TWS from the Gravity Recovery and Climate Experiment (GRACE). The extended Budyko framework was applied using four classic Budyko equations in the Qinba Mountains. The results showed that the extended Budyko framework could competently represent the relationship between monthly water supply and demand, with better performance than the original Budyko framework. Based on the extended Budyko framework, this study further quantified the contributors of streamflow variability using the variance decomposition method. The dominant contributor to intra-annual streamflow variability was precipitation (50%), followed by TWS (11%) and their covariance (-21%) in this region. Specifically, precipitation played a dominant role on streamflow variability in summer and autumn, while evapotranspiration and TWS significantly impacted streamflow in spring and winter, respectively. Furthermore, the hydrologic effects of rainfall intensity and vegetation were investigated to explain streamflow variability. As the rainfall intensity decreases, more precipitation is partitioned into evapotranspiration and TWS, while the increase of rainfall intensity leads to the partitioning of precipitation into streamflow. Similarly, monthly vegetation promotes the partitioning of precipitation into TWS, while inhibiting the partitioning of precipitation into streamflow. The opposite effect of vegetation on streamflow and TWS is weakened due to the neglect of TWS at an annual timescale, which may lead to an overestimation of the effect of annual vegetation on streamflow. The results have implications for improving the performance of the Budyko framework to reveal the relationship between monthly water supply and demand and understanding streamflow variability at an intra-annual timescale.

来自森林覆盖的山区流域的溪流对流域的水生生态系统和社会发展至关重要。然而，由于缺乏对大规模流域陆地蓄水量 (TWS) 的观察，了解流量的年内变化受到限制。本研究开发了一个月度 Budyko 框架，其中包含来自重力恢复和气候实验 (GRACE) 的 TWS。扩展的 Budyko 框架在秦巴山脉使用四个经典的 Budyko 方程。结果表明，扩展后的 Budyko 框架能够很好地表示月供水和需求之间的关系，性能优于原始 Budyko 框架。基于扩展的 Budyko 框架，本研究使用方差分解方法进一步量化了流量变化的贡献者。该地区年内流量变化的主要贡献者是降水 (50%)，其次是 TWS (11%) 及其协方差 (-21%)。具体而言，降水对夏季和秋季的流量变化起主导作用，而蒸散和 TWS 分别对春季和冬季的流量产生显着影响。此外，还研究了降雨强度和植被的水文效应，以解释流量变化。随着降雨强度的降低，更多的降水被分配到蒸散和 TWS 中，而降雨强度的增加导致降水被分配到水流中。相似地，每月植被促进降水分配到 TWS 中，同时抑制降水分配到溪流中。由于在年时间尺度上忽略了 TWS，植被对径流和 TWS 的相反作用减弱，这可能导致高估年植被对径流的影响。结果对改进 Budyko 框架的性能具有影响，以揭示月供水和需求之间的关系，并了解年内时间尺度的流量变化。这可能会导致高估一年生植被对河流流量的影响。结果对改进 Budyko 框架的性能具有影响，以揭示月供水和需求之间的关系，并了解年内时间尺度的流量变化。这可能会导致高估一年生植被对河流流量的影响。结果对改进 Budyko 框架的性能具有影响，以揭示月供水和需求之间的关系，并了解年内时间尺度的流量变化。