Idowu et al. (2025) Open problems in uncertainty quantification for flood modelling: A systematic review

Identification

• Journal: Environmental Modelling & Software
• Year: 2025
• Date: 2025-11-27
• Authors: Jamiu Adekunle Idowu, Ayman Alfahid
• DOI: 10.1016/j.envsoft.2025.106799

Research Groups

• Department of Computer Science, University College London, United Kingdom
• Sahel AI, Sahel Group Inc., United States
• Department of Information Systems, College of Computer and Information Sciences, Majmaah University, Saudi Arabia

Short Summary

This systematic review identifies eight critical open problems in uncertainty quantification for flood modelling, highlighting a system-level mismatch between flood complexity and current fragmented modelling practices, and advocating for seamless, end-to-end probabilistic pipelines.

Objective

• To systematically review and identify open problems in uncertainty quantification for flood modelling.

Study Configuration

• Spatial Scale: Covers various spatial scales relevant to flood modelling, from local to regional and global, as identified in the reviewed literature.
• Temporal Scale: Encompasses diverse temporal scales, including short-term predictions, long-term prediction errors, and future projections up to 2050, as discussed in the reviewed literature.

Methodology and Data

• Models used: Systematic review of literature discussing various modelling approaches for uncertainty quantification, including multi-model and ensemble strategies (e.g., SMA, WMA, Bayesian frameworks), probabilistic inference, Monte Carlo sampling, surrogate/emulator modelling, data assimilation, and machine-learning methods.
• Data sources: Systematic review of literature discussing various data sources for flood modelling, including input data (e.g., precipitation, discharge, Digital Elevation Models), measurement/estimation processes, and environmental/climatic drivers.

Main Results

• Identified eight open problems in uncertainty quantification for flood modelling:
  • Long-term prediction errors.
  • Poor calibration of predictive intervals.
  • Incomplete representation of uncertainties.
  • Inadequate handling of spatial and temporal variability.
  • Non-linearity.
  • Data scarcity and integration issues.
  • High computational costs.
  • Failure to capture uncertainties in extreme events.
• Concluded that these challenges reflect a system-level mismatch between the dynamic complexity of floods and the fragmented nature of current modelling practice.
• Proposed a shift from siloed, modular workflows to seamless, end-to-end probabilistic pipelines for real progress in flood risk science.

Contributions

• Provides a comprehensive systematic review identifying and synthesizing eight critical open problems in uncertainty quantification for flood modelling.
• Highlights the systemic nature of challenges in flood risk science, moving beyond isolated technical issues.
• Proposes a strategic shift towards integrated, end-to-end probabilistic modelling pipelines to address pervasive uncertainties.

Funding

• Not specified in the provided text.

Citation

@article{Idowu2025Open,
  author = {Idowu, Jamiu Adekunle and Alfahid, Ayman},
  title = {Open problems in uncertainty quantification for flood modelling: A systematic review},
  journal = {Environmental Modelling & Software},
  year = {2025},
  doi = {10.1016/j.envsoft.2025.106799},
  url = {https://doi.org/10.1016/j.envsoft.2025.106799}
}