Spatio-temporal Distribution Of Snow Cover And Snowmelt Runoff Modeling In Kaidu River Basin

Water resource is the key factor that restricts economic and social development and ecological security in arid region of Northwest China,the water resources in arid area mainly come from snow melt water and mountain precipitation.As climate warms,not only the types and quantity of supply but also the annual distribution of runoff are all impacted,besides,warming has also increased uncertainty over water resources and complicated hydrological processes,and increased the frequency of hydrological extremes.Therefore,Studies on the impact of climate on snow cover and runoff as well as snowmelt runoff modeling in arid region of Northwest China have a significant effect on the water resources management,ecological protection and economic and social development.Based on site data and multi-source Remote Sensing data,this paper analyzes the variation trends of annual snow cover and streamflow in Kaidu River,and discuss the relationship between climate change,snow cover and streamflow,and the snowmelt runoff model was established to simulate and forecast the snowmelt runoff under the climate change scenario.The main conclusions in this paper are as follows:(1)the total snow cover fraction showed a slightly increasing trend from 2000 to 2016(-0.017%/a),the annual maximum snow cover increased but the annual minimum snow cover showed a decreasing trend.Seasonally,the snow cover decreased in spring and summer,increased in autumn and winter.Snow cover days in most areas shower a decreasing or unchanging trend.In the past 45 years(1972-2016),the onset of snowmelt period has a shift of 10.35 days earlier while ending date has mover 7.56 days later.The correlation and sensitivity between cover and meteorological variables suggests that the snow cover was mostly influenced by temperature in spring and was mostly influenced by precipitation in winter.(2)Both the annual streamflow and peak streamflow exhibited an increasing trend from 2000 to 2016in a speed of 2.26×10~8m~3 per decade,the streamflow is sensitive to the variations of temperature and precipitation in summer,and to the variations of snow cover in spring.Increasing of temperature and precipitation was the primary factor leading the increasing of streamflow.(3)Constructed SRM model in the upstream of Kaidu River Basin,the results show that the overall accuracy needs to be improved,but the simulation effect of streamflow peak and daily change tendency is good,the Coefficient R~2 of determination is 0.66,and the volume difference D_V is 19.8%.There is slightly decreasing trend of snowmelt runoff simulated by using two RCP scenarios,and runoff will decrease significantly in June and July.