Previous projections show consistent increases in river flows of Asian Water Towers under future climate change. Here we find non-monotonic changes in river flows for seven major rivers originating from the Tibetan Plateau at the warming levels of 1.5 °C, 2.0 °C, and 3.0 °C based on an observation-constrained hydrological model. The annual mean streamflow for seven rivers at 1.5 °C warming level decreases by 0.1–3.2% relative to the present-day climate condition, and increases by 1.5–12% at 3.0 °C warming level. The shifting river flows for the Yellow, Yangtze, Brahmaputra, and Ganges are mostly influenced by projected increases in rainfall, but those for the Mekong, Salween, and Indus are dictated by the relative changes in rainfall, snowmelt and glacier melt. Reduced river flows in a moderately warmed climate threaten water security in riparian countries, while elevated flood risks are expected with further temperature increases over the Tibetan Plateau.

This study appraised and compared the performance of process-based hydrological SWAT (soil and water assessment tool) with a machine learning-based multi-layer perceptron (MLP) models for simulating streamflow in the Upper Indus Basin. The study period ranges from 1998 to 2013, where SWAT and MLP models were calibrated/trained and validated/tested for multiple sites during 1998–2005 and 2006–2013, respectively. The performance of both models was evaluated using nash–sutcliffe efficiency (NSE), coefficient of determination (R²), Percent BIAS (PBIAS), and mean absolute percentage error (MAPE). Results illustrated the relatively poor performance of the SWAT model as compared with the MLP model. NSE, PBIAS, R², and MAPE for SWAT (MLP) models during calibration ranged from the minimum of 0.81 (0.90), 3.49 (0.02), 0.80 (0.25) and 7.61 (0.01) to the maximum of 0.86 (0.99), 9.84 (0.12), 0.87 (0.99), and 15.71 (0.267), respectively. The poor performance of SWAT compared with MLP might be influenced by several factors, including the selection of sensitive parameters, selection of snow specific sensitive parameters that might not represent actual snow conditions, potential limitations of the SCS-CN method used to simulate streamflow, and lack of SWAT ability to capture the hydropeaking in Indus River sub-basins (at Shatial bridge and Bisham Qila). Based on the robust performance of the MLP model, the current study recommends to develop and assess machine learning models and merging the SWAT model with machine learning models.