Drought Analysis And Prediction Of Dongsheng Area In Ordos Under The Background Of Climate Change

Drought refers to the water shortage phenomenon formed by the imbalance of water balance caused by the lack of precipitation in the longer period.The occurrence of drought disaster gradually tends to normal,the occurrence frequency is higher,the intensity is larger,the scope is wider,its destructive,the abnormality is particularly prominent.Therefore,the prediction of drought risk in the context of climate change is one of the hot issues that many researchers pay close attention to at present,according to statistics,70%of natural disasters are derived from meteorological disasters,and because drought causes about 50%of meteorological disasters,the damage caused is also the most serious,so meteorological drought dominates in the event of drought.The meteorological drought affected the further deterioration of other droughts.By collecting meteorological data from 1988-2018 in Ordos Dongsheng District,Inner Mongolia,this paper analyzes the correlation of meteorological factors and the main factors affecting meteorological drought,finds out the applicable drought Index analysis Dongsheng District The situation and law of drought in recent 31 years,and establishes the prediction model of Dongsheng District drought.The drought forecast for the next 5 years in 2019-2023 is projected through modelling validation.The main conclusions of this paper are as follows:(1)According to the correlation of meteorological factors and principal component analysis,relative humidity,precipitation,temperature and evaporation are the main components affecting drought and the correlation is high in 8 meteorological factors.The temperature is positively correlated with the geothermal temperature,and the average annual change of sunshine hours is large,negatively correlated with precipitation,positively correlated with evaporation,and the precipitation is negatively correlated with evaporation,but the average annual change of precipitation is small.(2)According to 4 drought indexes:relative wetting index(MI),precipitation distance level percentage(PA),index of aridity(K),standardized precipitation evapotranspiration index(SPEI)analysis of drought in Dongsheng District scale,drought index reaction drought has a high consistency.The above drought indicators show that there have been about 10 droughts in the past 25 years,mainly in the light Drought,in 2000 and 2005 drought degree is more serious,in 2012 drought degree is the lowest.(3)When the drought analysis of Dongsheng District in the last 25-year scale was carried out by SPEI Index,the frequency of drought was as follows:summer>autumn>spring>winter,the intensity of drought was in turn:spring>autumn>summer>winter.The frequency of drought was analyzed by PA value:autumn>summer>spring>winter,the intensity of the drought was in turn:autumn>spring>summer>winter.The frequency and intensity of droughts in winter are slightly lower than in other seasons,and droughts occur more frequently in summer,but are less intense and prone to drought in spring and autumn.(4)Using multiple linear regression and BP neural network to establish the prediction model,it is found that the accuracy of both is high,and the forecast of drought in the next 5 years is predicted by 2019 to 2023.The results show that the prediction results of the two are roughly the same,there are drought phenomena in the 2020 and drought may occur in 2022,the degree of drought is light drought,based on the above seasonal analysis,it is expected that the drought phenomenon in spring and autumn is more obvious.