2023BIO Web of ConferencesOpen access

A comparison of generalized extreme value, gumbel, and log-pearson distributions for the development of intensity duration frequency curves. A case study in Costa Rica

Maikel Méndez-Morales, Luis-Alexander Calvo-Valverde, Jorge-Andrés Hidalgo-Madriz, Andrés Araya-Obando

Open full text 5 citations

Abstract

Global warming has already affected frequency and intensity of extreme rainfall events. This makes the evaluation of current and alternative statistical distributions used in the formulation of Intensity Duration Frequency curves (IDF) curves highly relevant. This study aims to evaluate the suitability of applying the Generalized Extreme Value (GEV) and the Log-Pearson type 3 (LP3) probability distributions against the traditionally used Gumbel (EV1) distribution to derive IDF curves for a flood prone area located in northern Costa Rica. A ranking system based on a normalized total-score from five metrics was implemented to identify the best distribution. GEV proved to be the most suitable distribution for most storm-durations and was therefore selected for development of the IDF curves with return periods ranging from 2 to 100 years. As return periods get longer however, deviations between rainfall estimates obtained get more prominent. Hence, a meticulous analysis of adjustment to select the most adequate probability distribution to estimate extreme events with return periods of 50 years or more should be undertaken, regardless of GEV or any other distribution. Results also reinforce the need to identify the distribution that best fits observed data for a particular weather station, especially when time-series are asymmetric.

Open-access reader

About this research paper

What this paper is about

Global warming has already affected frequency and intensity of extreme rainfall events. This makes the evaluation of current and alternative statistical distributions used in the formulation of Intensity Duration Frequency curves (IDF) curves highly relevant. This study aims to evaluate the suitability of applying the Generalized Extreme Value (GEV) and the Log-Pearson type 3 (LP3) probability distributions against the traditionally used Gumbel (EV1) distribution to derive IDF curves for a flood prone area located in northern Costa Rica. A ranking system based on a normalized total-score from five metrics was implemented to identify the best distribution. GEV proved to be the most suitable distribution for most storm-durations and was therefore selected for development of the IDF curves with return periods ranging from 2 to 100 years. As return periods get longer however, deviations between rainfall estimates obtained get more prominent. Hence, a meticulous analysis of adjustment to select the most adequate probability distribution to estimate extreme events with return periods of 50 years or more should be undertaken, regardless of GEV or any other distribution. Results also reinforce the need to identify the distribution that best fits observed data for a particular weather station, especially when time-series are asymmetric.

Why it matters

OpenAlex reports 5 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Global warming has already affected frequency and intensity of extreme rainfall events. This makes the evaluation of current and alternative statistical distributions used in the formulation of Intensity Duration Frequency curves (IDF) curves highly relevant. This study aims to evaluate the suitability of applying the Generalized Extreme Value (GEV) and the Log-Pearson type 3 (LP3) probability distributions against the traditionally used Gumbel (EV1) distribution to derive IDF curves for a flood prone area located in northern Costa Rica. A ranking system based on a normalized total-score from five metrics was implemented to identify the best distribution. GEV proved to be the most suitable distribution for most storm-durations and was therefore selected for development of the IDF curves with return periods ranging from 2 to 100 years. As return periods get longer however, deviations between rainfall estimates obtained get more prominent. Hence, a meticulous analysis of adjustment to select the most adequate probability distribution to estimate extreme events with return periods of 50 years or more should be undertaken, regardless of GEV or any other distribution. Results also reinforce the need to identify the distribution that best fits observed data for a particular weather station, especially when time-series are asymmetric.

Key concepts: Gumbel distribution, Generalized extreme value distribution, Return period, Extreme value theory, Statistics, Flood myth, Intensity (physics), Mathematics

Related papers

Back to paper searchBrowse research topicsOriginal source
A comparison of generalized extreme value, gumbel, and log-pearson distributions for the development of intensity duration frequency curves. A case study in Costa Rica — Research Paper | ScholarLens