Development of capabilities for an urban flood forecasting service
Building an operational urban and flash flood early warning service, designed around the people who run it and the communities who rely on it.
Project background
Theme: Early Warning Systems
Status: Ongoing
Partner countries: Antigua and Barbuda, Barbados, Comoros, Guatemala, Haiti
Region: Caribbean, Central America, and the South-West Indian Ocean
Timeline: March 2026 – March 2028
Project team
PI: Humberto Vergara, University of Iowa
Co-I: Yagmur Derin, University of Iowa
Co-I: Adam Pollak, University of Iowa
Co-I: Efthymios Nikolopoulos, Rudgers University
Co-I: Andrew Kruczkiewicz, Columbia University
Co-I: Michael Souffront, Tethys Geoscience Foundation
Scope
The project helps national services forecast urban and flash floods and turn those forecasts into impact-based warnings. The approach is user-centered: the system is built with each country's forecasters and disaster managers, shaped by their data, their daily operations, and the decisions they need to make. The goal is a service that local teams can run, maintain, and trust, so that warnings reach communities in time to act.
Outcomes
A National-scale automated urban and flash flood forecasting system.
Implement state-of-the-art AI models for flood inundation mapping.
Impact-based forecasting procedure and standard operating procedures.
Training materials and local capacity to operate and maintain UFFIS.
Developed System
UFFIS, the Urban Flood Forecasting Information System.
Training and resources
Recorded sessions and webinars held during the project were designed for expert and non-expert end-users: [Link to YouTube channel]
The following HydroLean module were developed as part of the capacity building activities:
Course | Description |
This module gives early-career hydrologists a compact, hands-on introduction to how forecasters actually judge whether a streamflow model is doing its job. It moves you from continuous metrics (RMSE, NSE, KGE, PBIAS) and Moriasi performance ratings, through categorical event verification (POD, FAR, POFD, CSI, ROC), to probabilistic ensemble verification (Brier score, BSS, reliability diagrams, rank histograms, CRPS). At every step you read a real hydrograph, score it, and decide whether the result is fit for an operational decision. The hands-on activities use the HyMOD conceptual rainfall-runoff model on the Leaf River catchment through a sequence of four Python notebooks, runnable directly in Google Colab. | |
Learn the three major satellite precipitation families (GPM IMERG, H SAF, and the NOAA Enterprise Rain Rate / SCaMPR), download each for your own region with Python notebooks, and compare them on a common grid with difference maps, scatter plots, and hyetographs. You finish by writing a one-page satellite rainfall briefing for your area of responsibility. | |
Learn to represent rainfall uncertainty as a mean-preserving multiplicative ensemble, propagate it through the HyMOD rainfall-runoff model, read the resulting streamflow ensemble as spaghetti plots, fan charts, and exceedance probabilities, and attribute the spread to its sources. The hands-on activities use two Python notebooks on the Leaf River catchment (Mississippi, USA), with every quoted number reproducible from fixed random seeds. | |
Learn what HEC-RAS 2D does and follow the workflow from digital elevation model to flood map: terrain, mesh, boundary conditions, gridded rainfall forcing, simulation, and results in RAS Mapper. Then learn to acquire model-ready inputs automatically from open data (USGS 3DEP elevation, NOAA MRMS radar rainfall, GOES satellite rainfall) in a Colab notebook, and see how the RAS Commander Python package automates model runs and GIS exports. You finish by drafting a defensible one-page model setup plan for a flood hazard mapping job. | |
Learn how impact-based forecasting turns a rainfall forecast into an actionable warning. You will work with the UNDRR and WMO risk vocabulary (hazard, exposure, vulnerability, impact, risk), read and apply an impact matrix, map the actor landscape around a warning, combine hazard, exposure, and vulnerability layers into Yellow, Amber, and Red warning zones in a notebook, and finish as the duty forecaster assigning warning levels in an escalating scenario. | |
https://edx.hydrolearn.org/courses/course-v1:UniversityofIowa+WMO_06+2026/about Link says page not found |
Acknowledgements
This work is supported by:
Early Warning System for Floods, United States Department of State (EWS-F US DoS).
Supporting Regional Cooperation to Strengthen Operational Multi-Hazard Forecasting and Early Warning Systems at National Level in the South-West Indian Ocean (CREWS South-West Indian Ocean).