The Himalayas are critical for supplying water for ∼2 billion people who live downstream, and available water is highly sensitive to climate change. The role of the groundwater system in sustaining the northern Himalayan rivers remains unknown, and this compromises Asia's future water sustainability. Here, we quantify the spatiotemporal contribution of groundwater to river flows in the Yarlung Zangbo Basin (upper reaches of Brahmaputra). Our results show that the groundwater recharge represents ∼23% of mean annual precipitation, translating into ∼30 km³/yr of baseflow, which contributes ∼55% of the total river discharge in the upstream reaches to ∼27% in the downstream reaches. The percentage of groundwater contribution is inversely related to topographic steepness and total precipitation, with the steepest topography and highest precipitation in the eastern Himalayas. This study fills a knowledge gap on groundwater in the Himalayas and is a foundation for projecting water changes under climatic warming.

Key Points: Shallow groundwater (<10 m deep) of alluvial plain responds to land surface regimes and dominates seasonal water budget variations Groundwater system in mountainous upstream impacts the evapotranspiration and surface‐groundwater interactions in alluvial plain downstream Groundwater sustains ecosystem by contributing to evapotranspiration of vegetation in the growing season and buffering lateral flow

Mountainous areas are referred to as “water towers” since they are the source of water for many low‐lying communities. The hydrologic budgets of these areas, which are particularly susceptible to climate change, are typically poorly constrained. To address this, we analyzed the partitioning between baseflow and mountain block recharge (MBR) using a regional groundwater model of the northern Qinghai‐Tibet Plateau run with multiple scenarios. We found that ~19% of precipitation is recharged, approximately 35% of which becomes MBR, while 65% discharges as baseflow. This partitioning is relatively independent of the recharge rate but is sensitive to exponential depth decrease of hydraulic conductivity (K). The MBR is more sensitive to this exponential decrease in K than baseflow. The proportion of MBR varies from twice to half of baseflow as the decay exponent increases by more than fivefold. Thus, the depth dependence of K is critical for quantifying hydrologic partitioning in these sensitive areas.

A modified Water-Table Fluctuation (WTF) method is developed to quantitatively characterize the regional groundwater discharge patterns in stressed aquifers caused by intensive agricultural pumping. Two new parameters are defined to express the secondary information in the observed data. One is infiltration efficiency and the other is discharge modulus (recurring head loss due to aquifer discharge). An optimization procedure is involved to estimate these parameters, based on continuous groundwater head measurements and precipitation records. Using the defined parameters and precipitation time series, water level changes are calculated for individual wells with fidelity. The estimated parameters are then used to further address the characterization of infiltration and to better quantify the discharge at the regional scale. The advantage of this method is that it considers recharge and discharge simultaneously, whereas the general WTF methods mostly focus on recharge. In the case study, the infiltration efficiency reveals that the infiltration is regionally controlled by the intrinsic characteristics of the aquifer, and locally distorted by engineered hydraulic structures that alter surface water-groundwater interactions. The seasonality of groundwater discharge is characterized by the monthly discharge modulus. These results from individual wells are clustered into groups that are consistent with the local land use pattern and cropping structures.