NASA/SERVIR Flash-Flood Forecasting Work Presented at AMS 2024

Jan 31, 2024. NASA/SERVIR project work on flash-flood forecasting in West Africa was presented at the 104th Annual Meeting of the American Meteorological Society (AMS 2024), held in Baltimore, MD, during January 28–February 1, 2024.

The presentation, titled “Advancing Flash Flood Forecasting Capabilities in West Africa with Machine Learning and Satellite Observations,” focused on the development of a flash-flood forecasting system for West Africa. The work is motivated by the region’s vulnerability to high-impact rainfall events and the difficulty of monitoring flash floods in areas where in situ rainfall observations are sparse and ground-based weather radar networks are limited.

As part of the NASA SERVIR program, the presented system combines the EF5 distributed hydrologic modeling framework, near-real-time satellite precipitation estimates from NASA IMERG, and machine-learning-based precipitation nowcasting. Together, these components are intended to support improved monitoring and forecasting of short-duration, high-intensity rainfall events that can lead to flash flooding.

The presentation also discussed key operational challenges, including satellite precipitation uncertainty, data-stream latency, and prediction lead time. In this context, the project’s machine-learning nowcasting efforts aim to improve short-term precipitation information by integrating satellite observations, including infrared imagery, to reduce the latency of observed precipitation fields and support downstream hydrologic forecasting.

This work is part of the broader NASA/SERVIR collaboration on strengthening flash-flood monitoring, forecasting, and decision-support capabilities in data-scarce and highly vulnerable regions of West Africa.

Congratulations to Dr. Nikolopoulos, our NASA/SERVIR Principal Investigator and the rest of the project team on presenting this work at AMS 2024!