Copula-based Multivariate Probability Distribution Of Water Environmental Variables And Its Applications

Facing with the growing severe River Basin water pollution,the State Council of the People's Republic of China has released "the Action Plan for Prevention and Treatment of Water Pollution",which clearly put forward to strengthening the management of water quality standard,stricting and prevention of risk control of water environment.To closely meet with this demand,my thesis mainly focus on the Copula based joint distribution of water environment variables in the Huai River Basin(HRB),which contains the present situation of water in environment in HRB,a Copula-based multivariate probability distribution model of water environment variables considering the different surface streamflow scenarios,a multivariate L-comoments based regional water environment risk assessment model and regional seasonal drought/flood combined risk model,as well as the impact of a extreme flood events on the surface water quality.The main research contents and achievements are represented as follows:(1)Research of the spatio-temporal variablility of water environmental variables.We have investigated the spatio-temporal variability of water environmental variables of DO,CODMn,NH3-N and TP based on the seasonal Mann-Kendall test using the historical water quality data from 2003 to 2015.Results shows that,On the yearly time scale,the water quality DO,CODMn and NH3-N has been significantly improved,while TP concentration suffers a deteriorated trend,the percentage of water quality observation stations with an increasing trend of DO and TP are 50%and 60%,respectively,while 80%stations experienced a decreasing trend of both water quality CODMn and NH3-N.On the monthly time scale,the concentration of both CODMn and NH3-N are experiencing decreasing trend in each month of the entire year,with a more significantly trend in the non-flooding season,the concentration of TP is increase during August to November,and concentration of DO is decrease during June to August.Under the condition of multiple-year average concentration of the water quality variables,the DO concentration can be reach to surface water qulity standard class I or II in all the stations,while the percentage of water quality observation stations can only reach 55%of CODMn,NH3-N and TP considering the water quality class III,the water quality in tributaraies Shaying River and Guo River are relatively poor,of which the NH3-N is most polluted,and TP comes the second,while the CODMn is the lightesd polluted.General fluctuation and non-nomality pattern can be observed of all the four water quality indexes in all the selected 20 stations in both the mainstream and tributaries Shaying River and Guo River,of which,the concentration of TP varies most significantly,and the concentration of CODMn and NH3-N fluctuated more significantly in the mainstream compared to the two tributaries Shaying River and Guo River.At last,owing to the fluctuation of water quality varibles,a combined dynamic weight of the modified entropy weight coefficient and the pollution contribute rate coefficient base Euclidean distance model was employed to evaluate the comprehensive condition of water quality in HRB.(2)The construction of Copula-based multivariate probability distribution models of water environment variables with consideration of the different surface streamflow scenarios.The Copula based joint distribution models of bivariate water environmental indexes(CODMn and NH3-N)were established under each streamflow scenarios(S1,S2 and S3,each represented the low,medium and high streamflow scenario)in the three represented hydrological stations Jieshou,Lutaizi and Bengbu in the middle reach of HRB.Based on these models,a quantitative evaluation of the influence of river streamflow on the joint occurrence of multiple interdependent water quality variables was presented.Results shows that,in both the main stream of HRB(Lutaizi and Bengbu)and the tributary Shaying River(Jieshou),the low streamflow scenario had very high joint probabilities of exceeding the water quality standards.For the higher streamflow scenario,the corresponding probabilities decreased;What's more,high streamflow scenarios helped to improve the joint probability of water quality combinations within higher water quality grades(most often Grade ? and ?)but decreased that within lower grades(most often Grade ? or inferior ?).Among all the three selected the stations,Jieshou was consistently at the highest risk of poor water quality conditions under any combination of water quality index and streamflow scenario.Under the low streamflow scenario S1,the joint water quality exceeding probability was much higher at Bengbu compared with Lutaizi,indicating that Bengbu was exposed to a greater risk of water pollution during times of low flow.However,with the higher streamflow scenarios S2 and S3,the risk at Bengbu was lower than that at Lutaizi.(3)The establishment of multivariate regonal water quality combined risk model based on the the multivariate L-comoments and Copula functions.We evaluated the joint probabilities of water quality indexes(COD and NH3-N)by developing the bivariate Copula function and L-comoments based regional water environmental risk model in the mainstream and the tributaries Shaying River and Guo River on middle reach of HRB.The results shows that the tributary Shaying River and Guo River suffer a general higher risk than that in the mainstream of Huai River of each given water quality grade;What is more,of both of the above two water quality combined events,a joint probability of decreasing tendency can be observed along Shaying River from upstream to downstream,while trend in the mainstream of HRB is more complicated.(4)The construction of regional seasonal drought/flood combined risk model based on the multivariate L-comoments and Copula functions.We evaluated the joint probability characteristics of seasonal continuous droughts(DD),seasonal continuous floods(FF),as well as the seasonal drought/flood alteration(ADFE)events by developing L-comoments and bivariate copula function-based multivariate regional frequency model using the standard streamflow index(SSI)of spring,summer and autumn from 31 hydrological stations in the upper and middle reaches of HRB.The results shows that:in both spring-summer and summer-autumn,FF events shows the highest frequency,follows by the DD events.The average probability of FF events are 13.78%and 17.06%during the spring-summer and summer-autumn,while they are 11.27%and 13.79%for DD events,respectively.Additionally,the seasonal ADFEs are found to most likely to occur near the mainstream of HRB,and seasonally,the summer-autumn droughts and floods occur more frequently than that in spring-summer.At last,we have investigated the impact of a extreme flood(in July 2007)on the wate environment,the extreme flood in July 2007 can further improve the water environmtal condition of the tributaries Shaying River and Guo River,while significant deteriorated water environment in the mainstream.