Volume 14, No. 4, 2020
Received: 2020/09/29, Accepted:
Authors:
Ihsan F. Hasan;
Abstract:
In this study, thirteen different probability distributions were used to analyze flood frequencies based on the annual maximum daily flood discharge data for AMF from 10 stream gauging stations in the Tigris River basin in the northeast of Iraq. Methods of moments, maximum likelihood and L-moments were used to estimate the parameters of these distributions. L-moment ratio diagram and goodness-of-fit tests were used in order to determine the best fitting probability distribution for AMF. The results showed that the Log-Pearson type-3 with method of moments, Log-Logistic type-3 with the maximum likelihood method and Generalized Extreme value with L-moments method were identified as the best-fit distributions for the selected gauging stations rather than the other distributions. A comparison was performed between the above three best-fitted distributions by using Root Mean Square Error. It has been observed that Log-Pearson type-3 with method of moments for parameter estimation was the most suitable distribution that fitted the annual maximum daily flood data for the study area.
Keywords:
Flood frequency analysis, Log-Pearson 3, Probability distributions, Tigris river