Compound Hydrologic Risk Assessment Over The Loess Plateau Under Climate Change

Under the background of global climate change,the basic laws of regional hydrological cycle have also changed.The water resources have been redistributed in different time and space scales,and the hydrological conditions in some areas have changed significantly.The characteristics of extreme weather or climate events such as extreme precipitation and drought are also in the process of dynamic change.Both extreme precipitation and drought events can pose a serious threat to agriculture,socioeconomics,ecosystems and life safety,and are the focus and research objects of agricultural scientists,hydrologists,ecologists and meteorologists.The Loess Plateau is an important agricultural production area in China,and the precipitation in this area is unevenly distributed in time and space.Climate change would affect the available water resources and the degree of soil erosion in the Loess Plateau.Therefore,the research on the occurrence and development of extreme precipitation and drought in the context of climate change in this region can provide scientific basis for regional hydrological risk management and sustainable water resources management.This paper aims to gain in-depth knowledge of the responses of extreme events such as extreme precipitation and drought under climate change over the Loess Plateau.The research focused on the following aspects:the multi-dimensional risk change characteristics of extreme precipitation events on the Loess Plateau;the development of the downscaling method for climate model outputs;the impact of future climate change on the coupled events of drought and high temperature on the Loess Plateau;the changing characteristics of future hydrological drought at different time scales in the Jinghe River Basin of the Loess Plateau.The specific contents of this paper are as follows:(1)The joint risk and its spatial variation characteristics of multi-dimensional extreme precipitation index combinations on the Loess Plateau were quantitatively evaluated,and the multi-dimensional joint risks(including the three-dimensional index combination)were evaluated.The second return periods based on the Monte Carlo random sampling algorithm for two and three-dimensional indicator combinations were innovatively applied to extreme precipitation risk assessment to meet the needs of different risk assessment in hydrological design.By using the forward and inverse calculations,the multi-dimensional risks of extreme precipitation indicators in the Loess Plateau were studied comprehensively.The calculation results can not only reflect the severity of the event,but also get reliable risk assessment results within a reasonable range,avoiding excessively large or too small joint weights.It was found that the imbalance of the spatial and temporal distribution of precipitation in the Loess Plateau has intensified.The northwest region was in a severe drought state for most of the year,and the annual precipitation might be highly concentrated in several strong precipitation events.However,the risk of flood disasters which caused by heavy rain in the southeast region has increased.(2)An assemble downscaling method has been developed,which can synchronously modify the bias of the output results of multi climate models.The information sharing between the output of different climate models can weaken the impact of the climate model with poor simulations on the results.The combined simulation results can reflect the uncertainty in climate simulations.The developed method was used to correct the deviation of precipitation outputs from seven regional climate models.Compared with other traditional methods,the method developed in this study has lower requirements on the quality of the climate model outputs and shows more stable model performance.It can provide method support for solving the problem that the simulation results of climate models in areas with complex terrain and climate types which are generally poor.The model was further applied to the aggregate forecast of future precipitation,and the results showed that the predicted precipitation would be significantly reduced under the medium and high emission scenarios(RCP4.5 and RCP8.5)in the central and southern Loess Plateau.The possibility of future drought would increase,and the eastern region would be more prone to flood disaster in the future.(3)Considering that when drought and extreme high temperature events occur at the same time,the synergistic effect of the two events can aggravate the severity of each other.This paper has established a hot drought identification system,which can quantitatively assess the combined risk of drought and high temperature.Two regional climate models was used to study the impact of future climate change on the coupled risk of drought and heat wave in the Loess Plateau.Two levels of drought and high temperature indexes were set to construct four combined research schemes.The results showed that the extreme high temperature duration of the Loess Plateau would significantly increase in the RCP4.5 and RCP8.5 scenarios,while the serious drought events were mainly distributed in the southwest and south central of the Loess Plateau.In the southwest and central-south regions of the Loess Plateau,long-term high temperature accompanied by moderate drought events would occur frequently at some stations in the future.Under the combined effect of drought and heat waves,it may have a serious impact on local agricultural production and even human health.(4)A coupling system of regional climate model and hydrological model was constructed.The system can identify the future water resources of the river basin under climate change and the changes of hydrological drought risks at different time scales,and provide information support for the sustainable management of regional water resources and the formulation of long-term disaster prevention and mitigation measures.The possible impacts of future climate change on runoff and hydrological drought characteristics of Jinghe River Basin in the Loess Plateau were studied.The results showed that the average monthly runoff under RCP8.5 scenario would be higher than that under historical period;the annual precipitation under future RCP4.5 scenario would also be higher than that under historical scenario,but the average annual runoff was smaller,which might be affected by the increase of evaporation due to the increase of temperature.In the future,the imbalance of runoff distribution would be intensified,and the average runoff in rainy season of RCP4.5 scenario would increase significantly compared with the historical stage,while it would decrease significantly in other months.Compared with the historical and RCP8.5 scenarios,the annual drought is more serious under the RCP4.5 scenario,but its severity gradually decreases with time.In the future,the monthly drought situation would be more severe,and the duration and severity of drought events would increase significantly.