Improving runoff modelling through strengthened snowmelt and glacier module enhances runoff attribution in a large watershed in Central Asia

Understanding runoff changes at the catchment scale is important for water resources management. Particularly, accurate attribution of runoff changes poses a significant challenge for water resources management in cold and arid regions due to complex cryospheric processes. This study improved the physically based distributed Geomorphology-Based Hydrological Model (GBHM) through modified snow module and supplement of glacier melt module. The improved model was validated using streamflow observations and applied to simulate natural runoff in Central Asia’s Balkhash Lake Basin (1955–2020). By comparing simulated natural runoff with observed records, we quantified climate change and human activities impacts across three periods: Human activities dominated runoff reductions during 1970–1986 (80.11%) and 1987–2002 (67.55%), climate change and human activities showed comparable influences during the period of 2003–2020 (43.22% vs 56.78%). Scenario simulations demonstrated limited hydrological effects from land use changes, with no consistent directional trend. The improved GBHM provides a practical tool for cold region hydrology studies, offering critical insights for sustainable water management in transboundary basins over arid and cold region.