Assessment Of Meteorological Drought Risk In Xinjiang And Its Impact On Agriculture And Vegetation Based On Copula Method

Global climate change is one of the greatest threats to the earth in the 21st century.Due to climate change,the frequency and severity of extreme events such as droughts and high temperatures are expected to increase further in the future.Drought is a major natural disaster event that can have a significant impact on the ecological environment,water resources,agriculture and other related fields in any region of the world,and has a profound impact in an increasingly globalized world.Xinjiang is one of the main agricultural production bases in China,and it is also a drought-prone area.At the same time,due to the impact of extreme events such as drought and high temperature,the vegetation in Xinjiang has undergone significant changes.Due to serious land desertification,severe evaporation of soil water,coupled with the impact of extreme climate,the overall vegetation ecological environment in Xinjiang is extremely fragile.Therefore,understanding the occurrence and variation of drought events in historical and projected future climates,and assessing the impact of drought on agriculture and vegetation loss are critical for the effective management of natural resources and the development of related policies.Therefore,this study developed a set of risk assessment and quantification method systems based on probability and statistical analysis to reveal the historical drought evolution trends and predict the future drought changes in Xinjiang.Through constructing joint probability distribution functions between the drought index and the agricultural crop yield and vegetation characteristic index,the impacts of drought on crop yield and the vulnerability of vegetation ecosystems are analyzed,which provided important decision support for drought prevention and disaster mitigation in Xinjiang.In details:(1)The characteristics of drought duration,severity and mean time interval are identified based on SPEI(Standardized precipitation evapotranspiration index)using run theory.The joint distribution of drought duration and severity extracted from 66 stations in Xinjiang is constructed based on bivariate copula statistical probability model.Then the drought characteristic risk is evaluated from multiple variables perspectives respectively.The results show that Xinjiang experienced severe drought events in the early 1980s and early 21st century.In terms of the spatial distribution of drought events,the Tarim Basin region in southern Xinjiang,which mainly includes the junction of Hotan Prefecture,Aksu Prefecture and Bayingolin Mongol Autonomous Prefecture,is the area with high incidence of severe drought events.(2)The temporal and spatial patterns of drought characteristic risks under different shared socioeconomic paths(SSPs:SSP126,SSP245,SSP370 and SSP585)in the future are analyzed and identified at grid scale based on the copula function,coupling with CMIP6(Coupled Model Intercomparison Project Phase 6)multi-model ensembles,as well as statistical downscaling and bias correction methods.Results show that the drought risk of high severity and long duration increased gradually in Xinjiang.(3)A copula-based bivariate probabilistic framework model is developed to quantify the impacts of drought events on crop yields,taking into account three crops(wheat,maize,and cotton)and SPEI at multi-scales(SPEI-1,SPEI-3,SPEI-6 and SPEI-12).The model has advantages in quantifying the complex nonlinear relationship between crop yield anomalies and drought,identifying the response sensitivities of different crops to drought time-scales,and quantifying the probabilities of crop yield loss under different wet and dry scenarios.Results show that wheat and maize yields are susceptible to long-term drought time-scales,while cotton is sensitive to short-term drought time-scale.Results deepen our understanding of drought impacts on crops at different time-scales during the growing seasons,thereby providing insights for rational crop irrigation management to reduce drought risk providing general guidance and ultimately promote sustainable agricultural development.(4)The spatial correlation between NDVI(Normalized Difference Vegetation Index)and SPEI at the grid scale in Xinjiang is identified based on the copula conditional probability model framework,and the risk of vegetation vulnerability under different drought scenarios is quantified.The model can effectively assess the relationship between vegetation response to drought.Results show that the response sensitivities of vegetation to drought during different growing seasons changes as summer>growing season>autumn>spring,that is,the risk of vegetation vulnerability to drought is the greatest in summer.(5)Considering the hot extreme scenarios,the risk of vegetation loss under the combined dry-hot extreme event is analyzed.A conditional probability model based on the vine copula is developed to quantify the effects of the dry-hot compound event,and the response of the summer vegetation in Xinjiang under the influence of the combined dry-hot event is identified.The model can flexibly construct the correlation dependencies between vegetation status,drought and hot events,and quantify the risk of dry-hot extreme compound events,providing researchers with certain insights and ideas in the field of high-dimensional risk analysis..In summary,this thesis has explored the evolution trends of historical and future drought occurrence and change in Xinjiang,assessed the risk of drought occurrence under climate change scenarios,and revealed the spatio-temporal pattern of drought in the future.Meantime,a conditional probabilistic statistical model framework is established based on copula method to identify the impact of different drought time scales on crop yield and quantify vegetation vulnerability risk under different drought scenarios.A three-dimensional copula model is constructed to identify and quantify the impact of dry-hot compound events on vegetation,providing important decision support for decision-makers to understand the response mechanism of vegetation to drought and to formulate related risk management and early warning schemes.