In this study, a backpropagation artificial neural network snow simulation model (BPANNSIM) is built using data collected from the National Climate Reference Station to obtain simulation data of China's future daily snow depth in terms of representative concentration pathways (RCP4.5 and RCP8.5). The input layer of the BPANNSIM comprises the current day's maximum temperature, minimum temperature, snow depth, and precipitation data, and the target layer comprises snow depth data of the following day. The model is trained and validated based on data from the National Climate Reference Station over a baseline period of 1986-2005. Validation results show that the temporal correlations of the observed and the model iterative simulated values are 0.94 for monthly cumulative snow cover duration and 0.88 for monthly cumulative snow depth. Subsequently, future daily snow depth data (2016-2065) are retrieved from the NEX-GDPP dataset (Washington, DC/USA: the National Aeronautics and Space Administration(NASA)Earth Exchange/Global Daily Downscaled Projections data), revealing that the simulation data error is highly correlated with that of the input data; thus, a validation method for gridded meteorological data is proposed to verify the accuracy of gridded meteorological data within snowfall periods and the reasonability of hydrothermal coupling for gridded meteorological data.

Human activities have substantially altered present-day flow regimes. The Headwater Area of the Yellow River (HAYR, above Huanghe'yan Hydrological Station, with a catchment area of 21,000 km(2) and an areal extent of alpine permafrost at similar to 86%) on the northeastern Qinghai-Tibet Plateau, Southwest China has been undergoing extensive changes in streamflow regimes and groundwater dynamics, permafrost degradation, and ecological deterioration under a warming climate. In general, hydrological gauges provide reliable flow records over many decades and these data are extremely valuable for assessment of changing rates and trends of streamflow. In 1998-2003, the damming of the Yellow River by the First Hydropower Station of the HAYR complicated the examination of the relations between hydroclimatic variables and streamflow dynamics. In this study, the monthly streamflow rate of the Yellow River at Huanghe'yan is reconstructed for the period of 1955-2019 using the double mass curve method, and then the streamflow at Huagnhe'yan is forecasted for the next 20 years (2020-2040) using the Elman neural network time-series method. The dam construction (1998-2000) has caused a reduction of annual streamflow by 53.5-68.4%, and a more substantial reduction of 71.8-94.4% in the drier years (2003-2005), in the HAYR. The recent removal of the First Hydropower Station of the HAYR dam (September 2018) has boosted annual streamflow by 123-210% (2018-2019). Post-correction trends of annual maximum (Q(Max)) and minimum (Q(Min)) streamflow rates and the ratio of the Q(Max)/Q(Min) of the Yellow River in the HAYR (0.18 and 0.03 m(3).(-)s(-1).yr(-1) and -0.04 yr(-1), respectively), in comparison with those of precorrection values (-0.11 and -0.004 m(3).s(-1).yr(-1) and 0.001 yr(-1), respectively), have more truthfully revealed a relatively large hydrological impact of degrading permafrost. Based on the Elman neural network model predictions, over the next 20 years, the increasing trend of flow in the HAYR would generally accelerate at a rate of 0.42 m(3).s(-1).yr(-1). Rising rates of spring (0.57 m(3).s(-1).yr(-1)) and autumn (0.18 m(3).s(-1).yr(-1)) discharge would see the benefits from an earlier snow-melt season and delayed arrival of winter conditions. This suggests a longer growing season, which indicates ameliorating phonology, soil nutrient availability, and hydrothermal environments for vegetation in the HAYR. These trends for hydrological and ecological changes in the HAYR may potentially improve ecological safety and water supplies security in the HAYR and downstream Yellow River basins.