Ensemble weather-runoff forecasting models for reliable flood early warning systems

Flood early warning systems often rely on a single hydro-meteorological forecast, which can limit reliability. Recent advances in deep learning (DL) offer promising improvements due to their low computational cost, enabling the generation of ensemble forecasts. This study investigates how to process...

Full description

Saved in:
Bibliographic Details
Main Authors: Alberto de la Fuente, Carolina Meruane, Viviana Meruane
Format: Article
Language:English
Published: Elsevier 2025-04-01
Series:Progress in Disaster Science
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2590061725000171
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849330178405171200
author Alberto de la Fuente
Carolina Meruane
Viviana Meruane
author_facet Alberto de la Fuente
Carolina Meruane
Viviana Meruane
author_sort Alberto de la Fuente
collection DOAJ
description Flood early warning systems often rely on a single hydro-meteorological forecast, which can limit reliability. Recent advances in deep learning (DL) offer promising improvements due to their low computational cost, enabling the generation of ensemble forecasts. This study investigates how to process multiple weather-runoff forecasts to improve model performance in predicting extreme events. We applied DL-based weather-runoff forecasting in river stations located at the foot of the Andes Mountains in Chile. The models couple a near-future global weather forecast with short-range runoff forecasting systems based on Long Short-Term Memory (LSTM) cells. Meteorological and geomorphological input variables commonly used in hydrological models were selected. Training and validation used ERA5 data, while NCEP-GFS data were used for testing and real-time operation. Model performance was evaluated using the Kling-Gupta efficiency (0.6–0.8) and Nash-Sutcliffe efficiency (greater than 0.9). The threat score index, which assesses the model's ability to predict threat peak flow exceedance, ranged between 0.6 and 0.8. The best-performing models were analyzed probabilistically to quantify uncertainty. Finally, we introduced the concept of conditional probability to estimate the likelihood of exceeding a threat peak flow, providing a basis for raising alerts and improving decision-making under uncertain conditions.
format Article
id doaj-art-5783724e119f49088a778bdaedbf6bba
institution Kabale University
issn 2590-0617
language English
publishDate 2025-04-01
publisher Elsevier
record_format Article
series Progress in Disaster Science
spelling doaj-art-5783724e119f49088a778bdaedbf6bba2025-08-20T03:47:02ZengElsevierProgress in Disaster Science2590-06172025-04-012610042010.1016/j.pdisas.2025.100420Ensemble weather-runoff forecasting models for reliable flood early warning systemsAlberto de la Fuente0Carolina Meruane1Viviana Meruane2Departamento de Ingeniería Civil, Universidad de Chile, Santiago, Chile; Corresponding author at: Departamento de Ingeniería Civil, Universidad de Chile. Avda. Blanco Encalada 2002. Santiago, Chile.Departamento de Ingeniería Civil, Universidad de Chile, Santiago, Chile; Modelación Ambiental SpA, ChileDepartamento de Ingeniería Mecánica, Universidad de Chile, Santiago, ChileFlood early warning systems often rely on a single hydro-meteorological forecast, which can limit reliability. Recent advances in deep learning (DL) offer promising improvements due to their low computational cost, enabling the generation of ensemble forecasts. This study investigates how to process multiple weather-runoff forecasts to improve model performance in predicting extreme events. We applied DL-based weather-runoff forecasting in river stations located at the foot of the Andes Mountains in Chile. The models couple a near-future global weather forecast with short-range runoff forecasting systems based on Long Short-Term Memory (LSTM) cells. Meteorological and geomorphological input variables commonly used in hydrological models were selected. Training and validation used ERA5 data, while NCEP-GFS data were used for testing and real-time operation. Model performance was evaluated using the Kling-Gupta efficiency (0.6–0.8) and Nash-Sutcliffe efficiency (greater than 0.9). The threat score index, which assesses the model's ability to predict threat peak flow exceedance, ranged between 0.6 and 0.8. The best-performing models were analyzed probabilistically to quantify uncertainty. Finally, we introduced the concept of conditional probability to estimate the likelihood of exceeding a threat peak flow, providing a basis for raising alerts and improving decision-making under uncertain conditions.http://www.sciencedirect.com/science/article/pii/S2590061725000171Hydrological deep learning modelEnsemble weather-runoff forecastingForecasts uncertainty quantification
spellingShingle Alberto de la Fuente
Carolina Meruane
Viviana Meruane
Ensemble weather-runoff forecasting models for reliable flood early warning systems
Progress in Disaster Science
Hydrological deep learning model
Ensemble weather-runoff forecasting
Forecasts uncertainty quantification
title Ensemble weather-runoff forecasting models for reliable flood early warning systems
title_full Ensemble weather-runoff forecasting models for reliable flood early warning systems
title_fullStr Ensemble weather-runoff forecasting models for reliable flood early warning systems
title_full_unstemmed Ensemble weather-runoff forecasting models for reliable flood early warning systems
title_short Ensemble weather-runoff forecasting models for reliable flood early warning systems
title_sort ensemble weather runoff forecasting models for reliable flood early warning systems
topic Hydrological deep learning model
Ensemble weather-runoff forecasting
Forecasts uncertainty quantification
url http://www.sciencedirect.com/science/article/pii/S2590061725000171
work_keys_str_mv AT albertodelafuente ensembleweatherrunoffforecastingmodelsforreliablefloodearlywarningsystems
AT carolinameruane ensembleweatherrunoffforecastingmodelsforreliablefloodearlywarningsystems
AT vivianameruane ensembleweatherrunoffforecastingmodelsforreliablefloodearlywarningsystems