Your Cart

Your cart is empty

Home  >  Volume 45

41. COMPARATIVE EVALUATION OF FLOOD ANALYSIS MODELS OF RAINFALL IN WARRI METROPOLIS by Audu H.A.P. and Oviri E. D.Volume 45 (March, 2018), pp 311 – 318
Sale price: $5.00
Quantity:

Abstract

This study carried out comparative evaluation of flood analysis models for the rainfall of Warri in Delta State, Nigeria. The study became very necessary due to the perennial flood menace experience in Warri Metropolis, especially during the wet season. It is aimed at determining the best flood analysis model that would aid in taking optimum decision in the selection of the best flood analysis model for the rainfall modelling of the study Area. The daily rainfall data of thirty one (31) years (1983-2013) of the study area were obtained from Nigerian Meteorological Agency (NIMET) Office in Warri. The extreme values of the yearly rainfall data were extracted from the rainfall data and used for the analysis. In order to determine the best probability fitting for the extreme values, the three most commonly used probability frequency distributions models, which were evaluated and compared in this study, included Logistics, Gumbel and Lognormal distributions models. The fittings were done using CumFreq software. The return periods and their frequency of occurrence were computed from the software. The frequency of occurrence of the rainfall in the study area, which was within the magnitude of 868-1023mm was 98.18%, 98.08% and 96.30% for Logistics probability distribution model, Gumbel probability distribution model and lognormal probability distribution model respectively. This implies that the highest expected rainfall for the study area is 1023mm and that the frequency of occurrence of rainfall of this magnitude may hardly occur. The return period for the rainfall magnitude was 51 years, 46 years and 27 years for the Logistics, Gumbel and Lognormal distributions respectively. The average absolute difference between the calculated and observed return period for Logistics, Gumbel and Lognormal distributions was 0.97, 0.95 and 0.63 respectively. The results have shown that the Logistics probability distribution model gave the best fitting for the rainfall pattern of Warri.

Keywords: Flood Disaster, Comparative Evaluation, Probability Distribution Models and Cumulative Frequency 

Close
Loading...