Study On The Propagation Characteristics And Laws Of Drought In Meteorology-Agriculture-Disaster Loss

In the background of global climate warming,the water cycle is intensified,and drought events are more and more frequent,which has a great impact on people's life.At present,agricultural drought,as a major factor directly threatening food security,has attracted much attention.However,the evolution process and propagation law of drought have not been fully revealed,which makes drought resistance work lack of theoretical support.Therefore,it is of great significance to explore the propagation law of meteorological drought to agricultural drought in the changing environment.The main stream of Wei River and its two tributaries(Jing River and Beiluo River)are taken as the research object in this paper.The Standardized Precipitation Index(SPI)and Standardization Soil Moisture Index(SSMI)are used to characterize meteorological drought and agricultural drought respectively,and the characteristics of spatiotemporal evolution of drought are analyzed.Based on the Bayesian theory framework,the Copula function is used to quantify the seasonal propagation time and trigger threshold of meteorological drought to agricultural drought,and the trigger thresholds of different levels of drought losses under agricultural drought stress are calculated.In addition,the dynamic evolution law and influencing factors of propagation time are explored.The main research results of this paper are as follows:(1)It is revealed that the temporal and spatial variation characteristics of meteorological drought and agricultural drought by using Mann-Kendall non-parametric rank correlation test and Morlet wavelet analysis,heuristic segmentation algorithm and run theory.In terms of time,it is shown that the SPI and SSMI sequences of the three sub-regions have significant downward trends on the whole.The SPI sequences of the three regions have the main period of 18,10 and 17 years in the 51-year evolution process respectively,and SSMI series have 18,6 and 18 years.The SPI sequence of Beiluo River has a variation point in 1965,and the SSMI sequences of three subregions have variation point in 1990,1969 and 1985,respectively.In space,the average duration and severity of agricultural drought are significantly greater than that of meteorological drought,and the agricultural drought in the main stream is the most serious.(2)Based on the framework of Bayesian theory,it is calculated that the propagation time from meteorological to agricultural drought.The results show that the static propagation time of meteorological drought to agricultural drought in spring is the longest,which is 7.3,8.0 and 7.3 months,respectively.The propagation time in autumn of Jing River is the shortest,which is 2.7 months.In addition,the spring and autumn propagation time of meteorological drought to agricultural drought in the main stream is longer than that in summer and winter,and the spring and winter propagation times of drought in Jing River and Beiluo River are longer than that in summer and autumn.Based on the moving window of 25 years,it is found that the propagation times from meteorological drought to different grades agricultural drought of the four seasons of main stream,the summer and winter of Jing River and the spring of the Beiluo River are significantly shortened,and the autumn propagation time of the Beiluo River is significantly longer.(3)It is quantified that the seasonal triggering thresholds of meteorological drought to different grades of agricultural drought by using conditional probability.The results show that the trigger thresholds of drought increase with the increase of agricultural drought level.According to the absolute value of SPI,it is found that the seasonal triggering threshold of meteorological drought to agricultural drought in the main stream is the smallest on the whole,and that in Beiluo River is the largest.From the perspective of accumulated precipitation loss,the trigger thresholds of the spring and summer in the main stream and the autumn and winter in Jing River are the smallest,while the spring in Jing River and summer,autumn and winter in Beiluo River are the largest.(4)It is explored that the influencing factors of the propagation time from meteorological drought to agricultural drought of different grades by Partial correlation analysis,and it is analyzed the role of potential evapotranspiration(PET)in drought propagation based on the standardized precipitation evapotranspiration index SPEI.The results indicate that the main influencing factors are human water use,ENSO,precipitation and PDO in the main stream,PDO,sunspot and AO in Jing River,and ENSO,PET,sunspot and precipitation in Beiluo River in the propagation process of meteorological drought to agricultural drought at the same level.At different levels,the PDO and temperature are the main influencing factors in the main stream,and PDO,soil moisture and temperature in Jing River,while PDO,PET and sunspot in the Beiluo River.Among them,PET can accelerate the propagation of meteorological to agricultural drought.(5)Based on the framework of Bayesian theory,it is constructed that the probabilistic evaluation model of loss-causing threshold of agricultural drought,and it is proposed that the trigger thresholds of agricultural drought to different levels of drought losses.The results show that the drought resistance in Shaanxi is the strongest,followed by Gansu and Ningxia.Among them,it is found that the absolute value of trigger threshold in autumn of Shaanxi is the largest among the three provinces,and Ningxia spring is the smallest.In addition,the loss-causing thresholds of agricultural drought in corresponding seasons in Shaanxi-Gansu-Ningxia ranged from small to large are spring of Ningxia,spring of Gansu,summer of Gansu,spring of Shaanxi and autumn of Shaanxi.