Improvement of the SWAT Model for Snowmelt Runoff Simulation in Seasonal Snowmelt Area Using Remote Sensing Data

The SWAT model has been widely used to simulate snowmelt runoff in cold regions thanks its ability of representing the effects and permafrost on generation confluence. However, a core method model, temperature index method, assumes both dates for maximum minimum factors threshold, which leads inaccuracies simulating seasonal regions. In this paper, we present development application an improved (SWAT+) daily area Northeast China. improvements include introduction total radiation modification factor variation formula, changing threshold according snow depth derived from passive microwave remote sensing data area. Further, SWAT+ is applied study climate change impact future (2025–2054) under scenarios including SSP2.6, SSP4.5, SSP8.5. Much simulation obtained as result, supported by several metrics, such MAE, RE, RMSE, R2, NSE calibration validation. Compared with baseline period (1980–2019), March–April ensemble average shown decrease SSP8.5 scenario during 2025–2054. This provides valuable insight into efficient utilization spring water resources areas.