Comparing flood frequency analysis using annual peak rainfall data

Project Abstract

Flood frequency analysis is a critical component of understanding and managing flood risk in various regions. In this study, we compare flood frequency analyses based on annual peak rainfall data from multiple locations. The research aims to evaluate the effectiveness of different methods in estimating flood frequency and exploring the variability in results across different regions. The study utilizes historical annual peak rainfall data collected from several gauge stations located in diverse hydrological settings. Different flood frequency analysis techniques, including the log-Pearson Type III, Generalized Extreme Value, and regional frequency analysis methods, are applied to the data. The analysis involves fitting probability distributions to the annual peak rainfall data and deriving flood frequency curves for each location. Results show that the choice of flood frequency analysis method significantly impacts the estimated return periods and flood quantiles. The log-Pearson Type III method tends to provide lower return periods compared to the Generalized Extreme Value method, especially for high return period events. Regional frequency analysis demonstrates the importance of considering spatial variability in rainfall patterns when estimating flood frequency. Furthermore, the study examines the sensitivity of flood frequency analysis results to the length of the data series. By subsetting the data into different periods, the research assesses how changing climate patterns and data availability affect flood frequency estimates. The analysis reveals that longer data series generally lead to more stable and reliable flood frequency results, highlighting the importance of using extended data records where possible. The comparison of flood frequency analysis methods highlights the trade-offs between simplicity and accuracy in estimating flood risk. While simpler methods like the log-Pearson Type III are easier to apply, they may lead to underestimation of extreme flood events. On the other hand, more complex methods like the Generalized Extreme Value approach provide better representation of tail behavior but require more data and computational efforts. Overall, this research contributes to the understanding of flood frequency analysis using annual peak rainfall data and provides insights into the implications of method selection on flood risk assessment. The findings have implications for flood management practices, emphasizing the need for robust methodologies that can capture the variability in extreme rainfall events across different regions.

Project Overview

<p> </p><div><p><strong>INTRODUCTION &nbsp; &nbsp; </strong><br><strong>1.1 BACKGROUND TO THE STUDY</strong><br>&nbsp;<br>Flood is a basic phenomenon of nature, it occurs during or immediately after an intensive rainfall or large and sudden snow-melt due to increase in temperature (in the glacial/temperate countries). Also, flood is a high water stage in which overflows its natural or artificial banks onto normally dry land such as a river inundated its flood pain. When flood spreads to area of steep slope, it accelerate the runoff, the potential energy immediately takes on an additional and high kinetic energy due to increased velocity.<br>The effects of flood on human well being range from unqualified blessings to catastrophes. The regular seasonal spring flood of the Nile River prior to construction of the Aswan high dam, for example were depends upon to provide moisture for the fertile flood pains of its delta. The uncontrolled flood of the Yantze River and the Huang Ho, have however, repeatedly wrought disaster when these rivers habitually re-chart their courses. Uncontrollable floods likely to cause considerable damage commonly result from excessive rainfall over brief period of time, as, for example the Omiyale flood of the Ogunpa river in Ibadan, Nigeria (1958, 1973 and 1980), the flood of Paris, France (1658 and 1910), of wars haw, England (1861 and1964), potentially disastrous floods may, however, also result from ice jams during the spring rise, as with the Danube, Switzerland (1342, 1402, 1501 and 1830), from storm tides such as those of 1099 and 1953 that the coast of England, Belgium, and the Netherlands; and from tsunamis, the mountainous sea waves cause by earthquakes, as in Lisbon (1755) and Hawaii, U.S.A (Hilo, 1946).<br>Flood can be measured by height, peak discharge, area inundated, and volume of &nbsp;flow, these factors are important to judicious land use, construction of protective levees and storage reservoirs, and, indirectly, the implementation of programs of soil and forest conservation to retard and absorb runoff from storm and more recently monitoring of river flow by computers using specifically designed software. The discharge volume of an individual storm is often highly variable from month to month and year to year. A particularly striking example of this variability is the flash flood, sudden unexpected torrent of muddy and sporadic rainfall: it is uncommon, of relatively brief duration and generally the result of summer thunderstorms in mountains. A flash flood can take place in a singles tributary while the rest of the drainage basin remains dry. The suddenness of its occurrences causes a flash flood to be extremely dangerous.<br>&nbsp;<br><strong>1.2 &nbsp; &nbsp;STUDY AREA</strong></p><p>Ibadan is the capital city of Oyo State and the third largest metropolitan area in Nigeria, after Lagos and Kano, with a population of 1,338,659 according to the 2006 census. Ibadan is also the largest metropolitan geographical area. She is located in south-western Nigeria, 128 km inland northeast of Lagos and 530 km southwest of Abuja, the federal capital, and is a prominent transit point between the coastal region and the areas to the north.</p><p></p></div><h3></h3><br> <br><p></p>