Extreme Rainfall Events Classification Using Machine Learning for Kikuletwa River Floods - Article summary

Aug 3, 2026

Machine Learning methods are leveraged to detect and classify extreme rainfall events that are difficult to model in earth systems, supporting rainfall-induced flood prediction. The study targets a sub-catchment within the Pangani River Basin in Northern Tanzania, using labeled data from five weather stations. Five algorithms are trained on historical records from 1979 to 2014 and evaluated with precision and recall, with Random Forest and XGBoost delivering the strongest overall results. Because of class imbalance, a generic Multi-layer Perceptron identifies heavy rainfall events most effectively.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages13

Reading Time33 min

AccessFree to download

Introduction

Motivation and flood impacts

Methodology

Data and study area

Machine learning algorithms

Results

Precision and recall evaluation

Discussion

Effect of class imbalance and heavy-rain detection

The Nelson Mandela AFrican Institution of Science and Technology NM-AIST Repository [https://dspace.mm-aist.ac.tz](https://dspace.mm-aist.ac.tz) Computational and Communication Science Engineering Research Articles [CoCSE] 2023-02-20 Extreme Rainfall Events Classification Using Machine Learning for Kikuletwa River Floods Mdegela, Lawrence Preprints [https://doi.org/10.20944/preprints202301.0558.v2](https://doi.org/10.20944/preprints202301.0558.v2) Provided with love from The Nelson Mandela African Institution of Science and Technology Disclaimer/Publisher’s Note: The statements, opinions, and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s) . MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions, or products referred to in the content. Preprints ([www.preprints.org) | NOT PEER-REVIEWED | Posted: 20 February 2023 ](www.preprints.org) | NOT PEER-REVIEWED | Posted: 20 February 2023 doi:10.20944/preprints202301.0558.v2)[ doi:10.20944/preprints202301.0558.v2](www.preprints.org) | NOT PEER-REVIEWED | Posted: 20 February 2023 doi:10.20944/preprints202301.0558.v2) Article Extreme Rainfall Events Classification Using Machine Learning for Kikuletwa River Floods Lawrence Mdegela 1,4 *, Esteban Municio 2, Yorick De Bock 1, Edith Talina Luhanga 3, Judith Leo 4 and Erik Mannens 1 1 University of Antwerp-imec; {lawrencenehemiah.mdegela,Yorick.DeBock, [erik.mannens}@uantwerpen.be](erik.mannens}@uantwerpen.be)[ ](erik.mannens}@uantwerpen.be)2 i2CAT Foundation; [esteban.municio@i2cat.net](esteban.municio@i2cat.net) 3 Carnegie Mellon University Africa; [eluhanga@andrew.cmu.edu](eluhanga@andrew.cmu.edu) 4 The Nelson Mandela African Institution of Science and Technology; [judith.leo@nm-aist.ac.tz](judith.leo@nm-aist.ac.tz) * Correspondence: [lawrencenehemiah.mdegela@uantwerpen.be](lawrencenehemiah.mdegela@uantwerpen.be); Tel.:+32494594736 Abstract: Advancements in Machine Learning techniques, availability of more data-sets, and increased computing power have enabled a significant growth in a number research areas. Predicting, detecting and classifying complex events in earth systems which by nature are difficult to model is one of such areas. In this work, we investigate the application of different machine learning techniques for detecting and classifying extreme rainfall events in a sub-catchment within Pangani River Basin, found in Northern Tanzania. Identification and classification of extreme rainfall event is a preliminary crucial task towards success in predicting rainfall-induced river floods. To identify a rain condition in the selected sub-catchment, we use data from five weather stations which have been labeled for the whole sub-catchment. In order to assess which Machine Learning technique suits better for rainfall classification, we apply five different algorithms in a historical dataset for the period of 1979 to 2014 . We evaluate the performance of the models in terms of precision and recall, reporting Random Forest and XGBoost as the ones with best overall performance. However, since the class distribution is imbalanced, the generic Multi-layer Perceptron performs best when identifying the heavy rainfall events, which are eventually the main cause of rainfall-induced river floods in the Pangani River Basin. Keywords: Heavy rainfall; River floods; Machine learning ; 1. Introduction Rainfall-induced river floods are among Earth’s most common and most catastrophic natural hazards [1] . Worldwide, flash floods account for more than 5000 deaths annually with a mortality rate more than 4 times greater than other types of flooding [2], and subsequently, their social, economic, and environmental impacts are significant. According to the Tanzania Meteorological Agency, in the last decade, the northern part of the country has experie

FAQs

What problem does the study address?

The study focuses on detecting and classifying extreme rainfall events to support prediction of rainfall-induced river floods in a sub-catchment of the Pangani River Basin.

Which data and time period are used for training the models?

The work uses data from five weather stations labeled for the sub-catchment, and models are trained on historical records covering 1979 to 2014.

Which machine learning techniques perform best overall?

Random Forest and XGBoost provide the best overall performance based on precision and recall metrics.

Aug 3, 2026

Upload by Lucas Martin