Using Machine Learning and Satellite Data for Improving Flood Prediction in the Lake Tanganyika Basin: A Case Study of the Rusizi Sub-Basin

Abstract Floods are among the most destructive natural hazards, particularly in Africa, where climate change, rapid urbanization, and data scarcity amplify their impacts. The Rusizi sub-basin has experienced recurrent floods that have caused significant disruptions and socioeconomic effects from 2014 to 2024. This study investigates the potential of machine learning, specifically Long Short-Term Memory (LSTM) and Convolutional LSTM (ConvLSTM) models, combined with satellite and hydrological datasets, to enhance flood prediction in the Rusizi Basin. Using ERA5 reanalysis meteorological data, river discharge , and satellite derived DEMs, the ConvLSTM model achieved high predictive skill (R2 = 0.96, NSE = 0.96, RMSE ≈ 0.0023), outperforming standard LSTM (R2 = 0.96). Flood risk maps were also generated for water levels of 4 m, 10 m, and 12 m. Results demonstrate that integrating AIbased models with satellite data significantly improves flood forecasting, providing valuable decision support tools for early warning and disaster risk reduction. Future integration with real-time monitoring and socioeconomic vulnerability data could further strengthen climate resilience strategies.

Authors

Institutions

Publication Details

Journal
Research Square
Published
2026-11-17
DOI
https://doi.org/10.21203/rs.3.rs-11753750/v1
Primary Topic
Hydrological Forecasting Using AI
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Using Machine Learning and Satellite Data for Improving Flood Prediction in the Lake Tanganyika Basin: A Case Study of the Rusizi Sub-Basin

N'tcha M'po Yèkambèssoun, Jean Claude Hakizimana, Freddy Hounoundaho, Emmanuel Egnide Lawin
4 links
Research Square
Hydrological Forecasting Using AI
preprint

Using Machine Learning and Satellite Data for Improving Flood Prediction in the Lake Tanganyika Basin: A Case Study of the Rusizi Sub-Basin

N'tcha M'po Yèkambèssoun, Jean Claude Hakizimana, Freddy Hounoundaho, Emmanuel Egnide Lawin
preprint en

Abstract

Abstract Floods are among the most destructive natural hazards, particularly in Africa, where climate change, rapid urbanization, and data scarcity amplify their impacts. The Rusizi sub-basin has experienced recurrent floods that have caused significant disruptions and socioeconomic effects from 2014 to 2024. This study investigates the potential of machine learning, specifically Long Short-Term Memory (LSTM) and Convolutional LSTM (ConvLSTM) models, combined with satellite and hydrological datasets, to enhance flood prediction in the Rusizi Basin. Using ERA5 reanalysis meteorological data, river discharge , and satellite derived DEMs, the ConvLSTM model achieved high predictive skill (R2 = 0.96, NSE = 0.96, RMSE ≈ 0.0023), outperforming standard LSTM (R2 = 0.96). Flood risk maps were also generated for water levels of 4 m, 10 m, and 12 m. Results demonstrate that integrating AIbased models with satellite data significantly improves flood forecasting, providing valuable decision support tools for early warning and disaster risk reduction. Future integration with real-time monitoring and socioeconomic vulnerability data could further strengthen climate resilience strategies.

Research Square
Université d'Abomey-Calavi (BJ), National University of Sciences, Technologies, Engineering and Mathematics (BJ)
Clean water and sanitation, Climate action, Industry, innovation and infrastructure, Sustainable cities and communities
Hydrological Forecasting Using AI
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon