Determining Flood Return Periods in River Swat Through Statistical Approaches

Flood frequency analysis is essential for understanding extreme hydrological events and supporting effective hydraulic design and flood risk management. This study employs diverse statistical distributions to estimate flood return periods at Chakdara and Khwaza Khela gauge stations along River Swat, Pakistan. Eight probability distributions were applied to historical flood data, including Log-Normal, Cauchy, Log-Normal 3P, Log-Pearson Type III, LogLogistic, Log-Logistic 3P, Generalized Extreme Value, and Gumbel. The performance of these distributions was evaluated using Goodness of Fit (GOF) tests, namely the Kolmogorov-Smirnov (K-S), Anderson-Darling (A-D), and Chi-Square tests. The ranking of models was based on their overall GOF performance. Results indicate that the Cauchy distribution provided the best return period estimation at Chakdara, predicting a 111-year return period for the extreme 2010 flood. In contrast, the Log-Logistic 3P distribution was identified as the best fit for Khwaza Khela, forecasting an 89-year return period for the 2022 flood. These findings offer valuable insights into flood risk assessment in River Swat, aiding policymakers, hydrologists, and disaster management authorities in devising effective flood mitigation and preparedness strategies.