EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Uncertainty‐Aware Machine Learning Models for Flash Flood Prediction

Jaqueline A. J. P. Soares, Allan K. S. Soares, Luiz F. Satolo, Graziela B. Scofield, Stephan Stephany, Leonardo B. L. Santos

Transactions in GIS · 2026-04 · 30 (2)

DOI: 10.1111/tgis.70254 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Transactions in GIS, online erschienen am 15.04.2026. Verfasst von Jaqueline A. J. P. Soares und 5 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 83 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Coordenação de Aperfeiçoamento de Pessoal de Nível Superior, Conselho Nacional de Desenvolvimento Científico e Tecnológico.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Flash floods have intensified in recent years, and machine learning (ML) models are increasingly used for real‐time prediction. However, most ML‐based flood‐forecasting studies remain largely deterministic and provide limited evidence on how input uncertainty propagates to forecasts, compromising reliability for operational early‐warning systems. This study addresses this gap through four main contributions: (i) a public 5‐year high‐resolution flood‐hazard dataset for benchmarking uncertainty‐aware ML flood forecasting, (ii) an uncertainty‐analysis approach based on synthetic noise injection to evaluate input uncertainty propagation, (iii) an uncertainty‐aware ensemble providing probabilistic predictions, and (iv) three new metrics to quantify robustness and uncertainty propagation. Five ML methods (LSTM, MLP, XGBoost, LightGBM, NuSVR) were evaluated for 12 lead times with and without synthetic noise. Multi‐criteria assessment revealed no single algorithm consistently outperformed others in predictive performance, computational cost, robustness, and uncertainty quantification. All data and code are publicly available to support reproducible research.

Förderung

  • Coordenação de Aperfeiçoamento de Pessoal de Nível Superior
  • Conselho Nacional de Desenvolvimento Científico e Tecnológico

Zitieren

APA

Soares, J. A. J. P., Soares, A. K. S., Satolo, L. F., Scofield, G. B., Stephany, S., & Santos, L. B. L. (2026). Uncertainty‐Aware Machine Learning Models for Flash Flood Prediction. Transactions in GIS, 30(2). https://doi.org/10.1111/tgis.70254

Vancouver

Soares JAJP, Soares AKS, Satolo LF, Scofield GB, Stephany S, Santos LBL. Uncertainty‐Aware Machine Learning Models for Flash Flood Prediction. Transactions in GIS. 2026;30(2). doi:10.1111/tgis.70254

BibTeX
@article{Soares2026,
  author = {Soares, Jaqueline A. J. P. and Soares, Allan K. S. and Satolo, Luiz F. and Scofield, Graziela B. and Stephany, Stephan and Santos, Leonardo B. L.},
  title = {Uncertainty‐Aware Machine Learning Models for Flash Flood Prediction},
  journal = {Transactions in GIS},
  year = {2026},
  volume = {30},
  number = {2},
  doi = {10.1111/tgis.70254}
}
Kostenlos starten →