Temporal And Spatial Evolution Characteristics And Trend Of Runoff In The Source Area Of The Yellow River

The source area of the Yellow River(SAYR)only accounts for about 1/6 of the area of the Yellow River Basin,but provides about 1/3 of the water volume of the Yellow River Basin,which affects the water safety and social and economic development of people who live in the basin.In recent years,the water resources in SAYR have undergone major changes because of global warming and human activities.Therefore,it is important to study the evolution and development trend of runoff in SAYR.This study conducted research on the characteristics of runoff evolution,influencing factors,hydrological response,and development trends in SAYR.The results are as follows:(1)The runoff in SAYR has shown a downward trend in the past few decades.The MK trend test shows that the change trend of the hydrological stations in the main stream is not obvious;the catastrophe analysis shows that the abrupt years of the runoff at each station in SAYR are mainly happened in the early 1990 s and mid-2000 s.During this period,the trend of runoff changes was "up-down-up";cycle analysis shows that there are small-scale cycles of5-10 a,mesoscale cycles of 11-16 a,and large-scale cycles of 23-27 a at each station in SAYR.And the degree of mesoscale periodic oscillation is the strongest.The specific changes to reduce runoff during the year of SAYR mainly to reduce the runoff of flood season,and also in the decrease in the peak value in September after the 1990 s.Spatial analysis shows that the main runoff generating interval in SAYR is in the Mentang-Maqu interval,and the spatial cause of runoff reduction is mainly the decrease in runoff in the runoff generating interval.(2)Trend analysis shows that precipitation and temperature in SAYR showed a significant upward trend from 1960 to 2018,passing the 95% and 99% significance tests respectively.Regional trend analysis shows that the upward trend of precipitation mainly for large area west and north of scarce rainfall areas,and abundant precipitation in the southeastern region of the upward trend is not obvious;the air temperature was rising in most parts of the basin and its trend is significant.In the past few decades,the total area of various land use types in SAYR has not changed much,on the contrary,spatial exchanges have been frequent.(3)Apply the SWAT model to SAYR for runoff simulation,and add topography and snowmelt factors,using the 1970 s and 1980 s as the regular and verification periods,respectively.The Nash-Suttcliffe efficiency coefficients Ens for the regular and verification periods are respectively 0.86 and 0.87,meet the requirements of the model.(4)The analysis of the importance of random forest features shows that climate change impact on runoff of SAYR is greater than human activities.The SWAT model contribution rate analysis also shows that the runoff change process in SAYR in the past few decades is mainly controlled by climate,and underlying surface changes' impact on runoff changes is much smaller than climate change.The analysis of abnormal precipitation-runoff relationship shows that the reason for the lower runoff in the 2000 s than in the 1980 s was likely to be the decrease in snowmelt and the increase in actual evaporation.(5)Land use under extreme runoff simulation scenarios shows that the outflow of the basin under scenario A(decreasing vegetation coverage)increases significantly,while the outflow of basin under scenario B(increasing vegetation coverage)is smaller than scenario A.And in the annual distribution of internal flow,the runoff of scenario a is larger than scenario B in snowmelt and flood season,while in the rest period,it is less than scenario B.It shows that the high vegetation coverage can effectively prevent the loss of water and adjust the distribution of runoff during the year in SAYR.Runoff simulations under different climate scenarios show that precipitation,evaporation,and will increase runoff under all climate scenarios in 2020-2099 in SAYR.The change rate of hydrological elements under the high-concentration emission scenario is relatively higher than that under the low-concentration emission scenario.Except for the RCP2.6 scenario,as the emission concentration increases,the changes in hydrological elements will become more drastic.