Interannual Changes of the Upper Indus Basin Hydrological Cycle

Abstract:

River runoff generated in the Upper Indus Basin (UIB) by precipitation and seasonal melting of snow and glaciers constitutes the primary source of fresh-water supply to sustain agricultural activities and hydro-power generation of Pakistan. Rising atmospheric temperatures are causing fast glacier retreat and decreasing snow coverage in the region. Available estimates suggest that this process will continue in the 21^st century leading to a progressive decline in river streamflow. Decreasing river discharge, together with rising demand for water from an increasing population, will expose the region to water stress and food shortages. Improving our understanding of the hydrological regimes controlling river runoff generation in the region is, therefore, critical to local authorities for water management and resource allocation. Here, we use outputs from the University of New Hampshire Water Balance Model (WBM) forced by precipitation and temperature estimates from a suite of 10 gridded datasets based on rain gauge, satellite, and reanalysis data to simulate the water cycle of the UIB. We validate the model’s outputs by using time series of time-variable gravity from the NASA/DLR GRACE (Gravity Recovery and Climate Experiment) mission and streamflow data from river gauges available within the region. We find that the ECMWF ERA5 reanalysis provides the best estimates of the climatic conditions. We use it to identify the physical processes driving the inter-annual variation in terrestrial water storage (TWS) and river runoff (R). Our analysis demonstrates that variations in TWS are significantly correlated with energy input during the summer months but not with changes in precipitation. River discharge has a complex behavior. In the north-west, with more extensive glacial cover, river runoff variability correlates with summer temperatures, but in the south-east, the variability is determined by precipitation during the winter and spring months. In all considered cases, we find an increasing correlation between annual flow and summer precipitation during the first two decades of the twenty-first century. This process is associated with a simultaneous decrease in the correlation between yearly discharge and winter precipitation and summer temperature.