The ESA initiative EO4FLOOD develops a new generation of flood forecasting systems that synergistically integrates cutting-edge Earth observation data, advanced hydrological models, and artificial intelligence (AI) tools.

In today's era of rapid technological advancement, satellite observations offer unprecedented levels of accuracy and coverage, providing invaluable insights into environmental conditions. To fully leverage the potential of these satellite data, they still need to be fully integrated into operational flood forecasting systems. EO4FLOOD evaluates and analyses the most advanced satellite data products available today, identifies key areas for improvement and develops cutting edge approaches and techniques to improve their usability in early warning systems. By taking advantage of the strengths of satellite and in-situ data, the project demonstrates the maturity of satellite data products for the implementation of a robust and reliable forecasting framework capable of providing timely and accurate predictions of flood events.

The goals of EO4FLOOD are structured around three main pillars:

1. development of an advanced open Earth observation dataset (EO4FLOOD Dataset),
2. utilization of the EO4FLOOD dataset in flood forecasting models,
3. demonstration of the EO data and models’ utility for science and society.

DGFI-TUM contributes to the development and dissemination of the EO4FLOOD Dataset by providing water level time series for virtual river stations based on satellite altimetry. A main focus is on the development and assessment of near real-time (NRT) water levels with a latency of less than one day. Furthermore, we are responsible for quality control and contribute to a discharge algorithm comparison project.

More information about the project and the collaboration partners can be found on ESA's project website.