Statistical modeling of annual maximum precipitation in Oued El Gourzi Watershed, Algeria

Abstract

This study aims to model annual maximum precipitation based on extreme value theory for the Oued El Gourzi Watershed, Algeria. A generalized extreme value (GEV) distribution was used to determine the probability distribution of extreme values and their dependency on time for the five stations distributed across the watershed. The non-stationary models are used to represent the GEV parameters assumed an invariant shape parameter and linear functions as location and scale parameters. The best model was selected using Akaike’s information criterion and Bayesian information criterion. Stationary and non-stationary return levels for different return periods have been proposed for the study area.