Study On Hydrologic Drought Characteristics Of Main Rivers In Weihe River Basin

The charateristics of runoff series of main rivers in Weihe River basin were studied in this paper, which involved runoff traits, high and low streamflow and encounter between high and low streamflow in different situations. Combination with the results of probability distribution of hydrologic drought variables which have been known and using run theory, extreme theory and stochastic theory, the probability distribution and charateristic value of critical hydrologic drought factors were analyzed, which the underlying hydrologic series were followed by different distributions. Meanwhile, Comparison between simulation and analytical methods was completed. Bivariate probability model of hydrologic drought was established by Copula function, which was tested through monthly streamflow series of Jinghe river and Beiluohe river. The main results are as follows:(1) Runoff charateristics of main rivers in Weihe River basin were analyzed. Runoff of main rivers in Weihe River basin had a descend tendency. The annual runoff existed strong auto-correlation and persistence. Change between high and low flow exhibited good synchronization. The distribution of annual runoff in month was unequal, which concentrated in July and August.(2) Transition between high and low flow of mai rivers in Weihe River basin was studied. The tendency of transition to low existed, the transitional probability of low flow was bigger than high flow, which was harm to water utilization. The probability analysis of several low or high flow showed that one-year high low occurred with the mean of 5.82 times, which was larger than one-year low flow with 5.27 times. More than 2-year low flow took place with the mean of 5.81 times which was higher than 4.91 times of more than 2-year high flow. The run probability analysis indicated that the probability of 2-year high flow was higher slightly than 2-year low flow, which were 24% and 22%,respectively. However, the probability of 3-year high flow with 13% was lower than 14% of 3-year low flow, which more than 3-year situation had the same results. The consecutive years of high flow group were between 1.57 and 2.60 years, the mean was 2.16 years. The consecutive years of low flow group were between 0.96 and 3.86 years, the mean was 2.83 years. The intensity of high flow group were from 0.26 to 6.62, the mean was 2.40. The intensity of low flow group were from 0.15 to 5.42, the mean was 1.88. In Weihe River basin, the probability of one-year high or low flow was higher than consecutive year high or low flow, the probability of one-year high flow was bigger than one-year low flow. Otherwise, the probability of consecutive year high flow was lower than consecutive year low flow, which showed that Weihe River basin was easy to threated by drought and there was a drought occurred in less than 2 years.(3) Encounter of high flow and low flow of main rivers in Weihe River basin was considered. The results indicated that low flow state that three encounter stations which were Linjiacun-Qianyang, Heiyukou-Yingge and Zhuangtou-Zhangjiashan,were in the same year, was easy to happen, and low-high combination frequency was small. The analysis of conditional probability indicated that the conditional probability of low flow was bigger than high flow when the runoff state of foremore station was known.(4) The probability distribution of hydrologic drought variables and its affected factors were analyzed. The results indicated that the formulations of negative acrossing and probability distribution of drought duration were accuracy except some errors in few stations due to the length of runoff series, the results from simulation and analytical method were coherent, the expected values of critical drought duration and severity of underlying hydrologic series with log-norm distribution were bigger than gamma distribution and norm distribution, which can be used to estimate the scope of critical drought variables in a certain period when the distribution of runoff series was unknown, the expected values of critical drought variable existed positive correlation with first-order auto-coefficient and skewed coefficient of runoff series. With application of above results in Weihe River basin, the results showed the expected values of critical drought duration in 50 years were from 5 years to 9 years, which were between 6 years and 12 years in 100 years.(5) The probabilty distribution of bivariable hydrologic drought was developed by Copula function, and the drought risk was measured through tail-correlation of Copula function. There was a high correlation between drought duration and drought severity, which marginal distribution is difficult to reflect real drought characteristic. The bivariable hydrologic drought model by Copula function avioded the deficiency mentioned above, which included different combination of drought severity and drought duration. The supervisor can predict the probability of drought severity with a certain level when drought duration with the same level happened through tail correlation analysis, which was very important in risk management. For instance, it was said that the probability of drought severity more than 0.95 percentail was 34.48% when drought duration more than 0.95 percentail according to tail correlation analysis.