Evaluation of the Best-Fit Probability of Distribution and Return Periods of River Discharge Peaks. Case Study: Awetu River, Jimma, Ethiopia

Abstract

The identification of the best distribution function is essential to estimate a river peak discharge or magnitude of river floods for management of watershed and ecosystems. However, inadequate estimation of the river peak discharge and floods magnitude may decrease the efficiency of water-resources management, resulting in soil erosion, landslides, environmental damage and ecosystem degradation. To overcome this problem in hydrology, different methods have been employed, applying a probability distribution.In this study to determine the suitable probability of distribution for estimating the annual discharge series with different return periods, the annual mean and peak discharges of the Awetu River (Jimma, Ethiopia) over a 24 years’ time period have been collected and used. After the homogeneity and consistency test, data were analyzed to predict extreme values and were applied in seven different probability distribution functions by using L-moment and easy fit methods. Then, three goodness of fit tests, Anderson-Darling (AD), Kolmogorov-Smirnov (KS), and Chi-Squared (x2) tests, were used to select the best probability distribution function for the study area. The obtained results indicate that, Log-normal distribution function is the best-fit distribution to estimate the peak discharge recurrence of the Awetu River. The 5-year, 10-year, 25-year, 50-year, 100-year and 1000-year return periods of discharge were calculated for this river. The results of this study are useful for the development of more accurate models of flooding inundation and hazard analysis. AVALIAÇÃO DA MELHOR PROBABILIDADE DE AJUSTE DE DISTRIBUIÇÃO E PERÍODOS DE RETORNO DOS PICOS DE DESCARGA FLUVIAL. ESTUDO DE CASO: AWETU RIVER, JIMMA, ETIÓPIAResumoAvaliação da melhor função de probabilidade de distribuição e de períodos de retorno de picos de descarga de rio. Estudo de caso: Rio Awetu, Jimma, Etiópia. A identificação da melhor função de distribuição é essencial para estimar um pico de descarga de rios ou a magnitude das inundações de bacias hidrográficas e ecossistemas, tendo em vista a gestão dos sistemas hídricos e dos ecossistemas. Entretanto, uma estimativa inadequada da magnitude do pico de vazão e inundações do rio pode diminuir a eficiência do gerenciamento dos recursos hídricos, resultando em erosão do solo, deslizamentos de terra, danos ambientais e degradação do ecossistema. Para superar esse problema na hidrologia, diferentes métodos foram empregados, aplicando funções de probabilidade de distribuição e retorno.Neste estudo, para determinar a probabilidade adequada de distribuição e para estimar séries de descarga anuais com diferentes períodos de retorno, foram usados dados de médias anuais de picos de descarga do Rio Awetu (Jimma, Etiópia) durante um período de 24 anos. Após o teste de homogeneidade e consistência, os dados foram analisados para prever valores extremos e foram aplicados a sete funções diferentes de probabilidade de distribuição, usando o momento L e métodos de ajuste fácil. Em seguida foram utilizados, três testes de qualidade de ajuste, Anderson-Darling (AD), Kolmogorov-Smirnov (KS), and Chi-Squared (x2), para selecionar a melhor função de probabilidade de distribuição para a área de estudo. Os resultados obtidos indicam que, a função de distribuição log-normal é a que mais se adequa para estimar a recorrência de picos de descarga do Rio Awetu. Os períodos de retorno de descarga de 5 anos, 10 anos, 25 anos, 50 anos, 100 anos e 1000 anos foram calculados para este rio. Os resultados deste estudo são úteis para o desenvolvimento de modelos mais precisos de inundação e análise de risco.Palavras-chave: Descarga de Rio. Qualidade de ajuste. Log Pearson Tipo III. Distribuição de probabilidade.

An Investigation on the Effect of Outliers for Flood Frequency Analysis: The Case of the Eastern Mediterranean Basin, Turkey

“…Considering the Chi-squared test, the most appropriate probability distribution with the data available for the 1712 SGS was acquired by Normal distribution, similar to many studies in the literature [48]. However, for the results of some other studies, the lowest performance was achieved with the Normal distribution [49,50]. After adjusting the outliers, Pearson Type III showed compatible distribution with similar studies [18].…”

Evaluation of generalized extreme value and Gumbel distributions for estimating maximum daily rainfall

“…Maximum precipitation series can be represented by more than one probability distribution, and it is important to select the best distribution to be used (Zhang et al, 2002;Mandal and Choudhury, 2015;Vivekanandan, 2015a;Feyissa and Tukura, 2019). We can select the best model based on tests comparing observed frequencies with theoretical frequencies.…”

“…However, we observe that most studies show only the best fit obtained, while in many cases, two or more models fit properly, with very small differences, and often less laborious and sometimes equally efficient model options are not shown. Feyissa and Tukura (2019) highlighted that the proper evaluation of flood frequency distribution is one of the main problems faced by hydrologists. This issue is very important as different distributions can produce significantly different estimates for the same return period (Coulson, 1991).…”

Surface runoff modeling in Ketar watershed, Ethiopia

“…The depth of this lake has decreased due to the sediment transported to the lake through surface runoff. For proper management of soil erosion, landslides, and environmental degradation, estimation of flood and peak discharge are the essential (Feyissa and Tukura 2019). Therefore, it is required to control surface runoff that results in erosion and deposing.…”