Analysis On Characteristics Of Drought And Flood Alternating Events In Sihu Lake Basin Based On Palmer Drought Model

Due to the changes of global climate change and underlying conditions and the impact of violent human activities,the situation of drought and flood in China has been widely distributed and frequent.It is a serious threat to water security and food security in China.In this paper,the daily meteorological data from 1956-2013 and the soil moisture data from 1960-1962 and 1981-2002(total 25 years)of Jingzhou station were selected to study the occurrence regularity and the statistical characteristics of the drought and flood alternating events in Sihu Lake Basin.Based on the Palmer drought model,which is used to identify the events,three characteristic variables of the drought and flooding events are separated from the daily Palmer index by using the run theory.The Archimedean Copula is used to construct the two-variable the triple-variable joint probability distribution,and calculate the conditional probability and the reproduction period,in order to deal with regional drought and flood disaster prevention and control work to provide a basis.This paper mainly draws the following conclusions:(1)The Palmer drought model which modified in this paper is more accurate than the original Palmer drought index.The PMDI indicator has a clear physical meaning,and it not only considers a variety of meteorological factors,but also takes the soil moisture,the water of the early period into account,with comparable in time and space.(2)Drought and flood caused by varying degrees,light drought to light water frequency occurred the highest frequency,accounting for half of all drought and waterlogging events,followed by drought and waterlogging and light drought to waterlogging,respectively,18.00%and 12.99%The frequency of drought and heavy waterlogging was the lowest,only 0.59%.The frequency of light and waterlogging was the highest,accounting for 48.34%of all the flood and drought,followed by flooding and light flooding,accounting for 18.58%And 12.43%,the frequency of heavy drought and heavy water occurred the lowest,only 0.68%.(3)The drought-flooding event is a random hydrological event,and its characteristic variables have a dependency relationship.The two-variable joint distribution constructed by GH Copula function and the three-variable joint distribution fit are satisfactory.Based on the GH Copula function,the two?dimensional conditional probability and the multivariate combination reproduction period of three drought-flooding characteristic variables are obtained,which can provide the basis for the regional water conservancy project planning and risk assessment.