期刊名称：Analele Universităţii din Craiova. Seria Geografie
印刷版ISSN：1224-4112
电子版ISSN：2069-6191
出版年度：2017
卷号：2017
语种：English
出版社：Universitatea din Craiova
摘要：The main objective of the study is to use the statistical models to analyse the probability of flood and the joint of temporal variability of the flood peaks and their seasonality. The case study refers to the 2013 spring floods on the Jiu River, Filiași-Craiova sector. Thus, in order to obtain several features of the maximum discharge, the maximum values, the increasing and decreasing time also the frequency of occurrence, one of largest floods is analysed. The data used were recorded at 3 hydrometrical stations on the Jiu River for a common period of 10 years.The hydrological analysis is based on two methods of the maximum flow values: Mann-Kendal test and Gumbel distribution. In the first part, we have identified if it’s a trend or not in the data series of the annual maximum discharge using Mann-Kendall trend test and the results were obvious, both statistically and graphically. The annual trend at two hydrometrical stations is of decrease (Test Z values: -1.58, Filiași hydrometrical station and -0.81, Răcari hydrometrical station), while at the hydrometrical station that closes the analysed sector is an increasing trend (Test Z value: 0.62, Podari hydrometrical station).The Gumbel distribution is used to process the data from a long observation period of the maximum values. Applying this method to the flood data registered at the three hydrometric stations along the Jiu river are corresponding certain return periods (years) as follows: 1,140 cm/s registered at Filiași hydrometrical station has a 15 years return period; 1,447 cm/s reached at Răcari hydrometrical station has a 24 years return period; 1,309 cm/s recorded at Podari hydrometrical station has a 27 years return period. The above-mentioned values confirm the magnitude of the maximum flow at Răcari and Podari hydrometrical stations and anticipate the potential damage.The flood frequency analysis using statistical models is important and also required in Romania given that the statistics of extreme events plays an important role in the engineering practice of water resources management.
