Analysis Of Water Resources Evolution And Vulnerability Assessment Under Climate Change In The Chushandian Reservoir Basin

Runoff is an important basis of the terrestrial hydrological cycle,with global warming,natural runoff processes and water resources characteristics have changed significantly,leading to increased uncertainty in studying water resources.In this context,it is significant to investigate the response of runoff to climate change for the scientific management of water resources.In this study,the impact of climate change on runoff processes was explored based on the SWAT model in the Chushandian Basin,the uncertainties arising from multiple factors in hydrological simulation predictions were identified and quantified,and the water resources vulnerability under historical and future scenarios was assessed based on hydro-climatic conditions,the economic and social system,the major conclusions were obtained as follows:(1)The sensitive parameters of the hydrological model were screened based on sequential uncertainty stochastic fitting,and a scheme to determine the parameter’s ranges and values were proposed,the final model performance was stable and efficient.The results show that the sensitivity of the parameters identified using the three objective functions of NSE,R~2 and PBIAS is different,the PBIAS has the highest uncertainty in evaluating the performance level,the NSE has the best comprehensive ability,and the sensitive parameters identified using PBIAS are:CN2,CANMX,SOL＿BD,GW＿DELAY,GWQMN,GW＿REVAP,ESCO,RCHRG＿DP and SOL＿AWC;the SWAT performance and precipitation distribution are related,the more abundant the precipitation,the better the simulation.(2)The hydrological distribution characteristics in the study area were evaluated based on the SWAT model,and the climate change effects on flow production were discussed qualitatively and quantitatively.The results show that the runoff intra-annual distribution has obvious seasonality and heterogeneity,in which total runoff is maximum in summer and minimum in winter,reaching a maximum in July;there is no obvious trend in precipitation and runoff during 1961-2014,while an increasing trend in temperature;precipitation is the main influencing factor for runoff changes,with a 2.0%precipitation decrease and 0.28°C temperature increase,runoff decreases by 4.5%,where precipitation produced more than 90%contribution.(3)The Haegeraves and Penman-Monteith evaporation algorithms,Variable Storage and Muskingum river routing were compared in their hydrological applicability in the study area,the multifactor Analysis of Variance was used to quantify the module algorithm,model parameters,GCMs and SSPs on runoff prediction uncertainty contributions.The results indicate that runoff will increase in June and August during 2021-2060 and all months during 2061-2100;the annual mean streamflow(Q_m)and annual extreme streamflow(Q_p)increase by 1.8%and 2.6%during 2021-2060 and by 14.7%and 18.6%during 2061-2100;except for the SSP1-2.6 scenario,Q_m and Q_p under SSP2-4.5,SSP3-7.0,and SSP5-8.5 scenarios will increase with high probability during 2061-2100,among which the probability of Q_p increase is as high as 97%;the evapotranspiration algorithm is the main uncertainty source in Q_m simulation,contributing more than 50%,and GCM and SSP are the main uncertainty sources in Q_p simulation,contributing about 70%.The uncertainty of the parameters is small and negligible.(4)Based on three indicators of climate exposure,sensitivity and adaptability,the climate risk to the water resources,the vulnerability and resilience facing the risk in the study area were analyzed.The vulnerability of water resources under historical and future scenarios was assessed;to cope with the adverse impacts or potential impacts of climate change and to promote green and sustainable development of water resources,macro and micro adaptive water resources management measures were proposed.