Analysis Of The Evolution Characteristics Of Hydrological Elements And Attribution Of Runoff Changes In Yulin City Under Changing Environment

Under the combined effects of climate change and human activities,the water resources situation in Yulin City is becoming increasingly severe,and the shortage of water has seriously limited the local economic and social development.Precipitation and runoff are two important parts of the regional water cycle,and a proper grasp of their evolutionary characteristics and an indepth investigation of the influence of each factor on runoff changes can provide a valuable reference for the scientific management and rational allocation of water resources in Yulin.Based on the above,the research is carried out on the topic of "Analysis of the evolution characteristics of hydrological elements and attribution of runoff changes in Yulin City under changing environment",the main results of the study are as follows:(1)The performance of the National Tibetan Plateau Data Center(TPDC)precipitation dataset in Yulin City was evaluated using the correlation coefficient(CC),root mean square error(RMSE),mean absolute error(MAE),NashSutcliffe efficiency coefficient(NSE),and Kling-Gupta efficiency coefficient(KGE)based on the measured precipitation at the stations,then the particle swarm optimization algorithm(PSO),genetic algorithm(GA),sparrow search algorithm(SSA)and ant colony algorithm(ACO)were used to optimize the BP neural network to correct the TPDC precipitation data,respectively.The results showed that the precipitation accuracy of TPDC data in winter was relatively high;the correction effect of the SSA＿BP neural network was better than the other three models.Taking Bobai station as an example,the model improved the CC,NSE,and KGE values of the original TPDC precipitation by 0.24,0.41 and0.25,respectively,and reduced the RMSE and MAE values by 32.8 and 30.6,respectively.(2)The annual and seasonal precipitation and runoff evolution characteristics of Yulin city were studied by using trend synthesis analysis,mutation synthesis analysis,Morlet wavelet analysis,and R/S analysis,etc.Taking the annual precipitation of Yulin city and the annual runoff of Bobai station as examples,the results showed that: the historical annual precipitation showed a non-significant increasing trend,the mutation year was 1965,the first and second main periods were 37 a and 23 a respectively,and the future annual precipitation will maintain an increasing trend;the historical annual runoff showed a non-significant decreasing trend,the mutation year was 1966,the first and second main periods were 22 a and 9a respectively,and the future annual runoff will maintain a decreasing trend.(3)The intra-annual distribution characteristics of precipitation and runoff in Yulin City were studied by using uneven coefficient,full adjustment coefficient,surplus expectation coefficient,etc.The results showed that the precipitation and runoff in Yulin were mostly concentrated in May-August,accounting for about 60% of the whole year;and over 40% in summer,which is significantly higher than the other three seasons;with the passing of time,the intra-annual distribution of precipitation gradually tends to be uniform,but the tendency is not significant,the intra-annual distribution of runoff tends to be significantly uniform,the maximum monthly precipitation and the maximum monthly runoff depth occurred at a non-significantly later time.(4)Qualitative analysis of climate factors which causing runoff changes was conducted by using partial correlation analysis,multiple correlation analysis,and path analysis,etc.The contribution of climate change and human activities on the runoff changes was quantified by using the slope change ratio of cumulative quantity(SCRCQ)and double mass curve(DMC).The results showed that: the overall combined effect of climate factors on runoff changes was found to be facilitative,and there was a significant positive correlation between precipitation and runoff,and the influence of precipitation on runoff was always higher than that of potential evapotranspiration;taking annual runoff as an example,the contribution of climate change calculated by the two methods of SCRCQ and DMC were 49.5% and 36.1%,respectively,and the contribution of human activities were 50.5% and 63.9%,respectively.