Advancing the prediction of extreme hydrological events
About the project
Floods and droughts can have serious effects on communities, especially in regions where early warning systems are less reliable, such as parts of the Global South. SEED-FD addresses this challenge by using satellite data and local micro-sensors to improve the accuracy of flood and drought forecasts.
Building on the Copernicus Emergency Management Service (CEMS), SEED-FD develops enhanced prediction systems that can be used worldwide. With more accurate forecasts, we save lives, protect essential resources, and help communities become more resilient in a rapidly changing climate.
Project aims
Enhance the CEMS hydrological model
for better representing the range of hydroclimatic processes worldwide.
To improve flood and drought predictions, SEED-FD refines the CEMS hydrological model by incorporating complex processes like groundwater flow and wetland impacts. This will make forecasts more accurate, even in data-scarce areas.
Enhancing CEMS hydrological simulations and forecasts globally
SEED-FD demonstrates the added value of satellite data and innovative local micro-sensors to improve the accuracy of global flood and drought predictions in the CEMS by:
- Using Earth Observation data (collected by satellites) for more realistic initial conditions, instead of estimations based on modelled reanalysis meteorological data
- Continuously adjusting forecasts through real-time correction, data assimilation (integrating real-time measurements into the model), and post-processing, making predictions more reliable
- Developing Artificial Intelligence (AI) algorithms to combine forecasts and observations, reducing errors and strengthening early warnings globally
Expanding the CEMS early warning system forecast product portfolio for floods and droughts
SEED-FD develops tools to predict a broader range of extreme flood and drought events, covering areas previously lacking specific warnings:
- Flash Drought forecast product
- Seasonal Drought forecast product
- Drought tracking product
- Flash Flood forecast product
These tools enable earlier detection, supporting communities in vulnerable regions worldwide.
Enhance the CEMS hydrological model
for better representing the range of hydroclimatic processes worldwide.
To improve flood and drought predictions, SEED-FD refines the CEMS hydrological model by incorporating complex processes like groundwater flow and wetland impacts. This will make forecasts more accurate, even in data-scarce areas.
Enhancing CEMS hydrological simulations and forecasts globally
SEED-FD demonstrates the added value of satellite data and innovative local micro-sensors to improve the accuracy of global flood and drought predictions in the CEMS by:
- Using Earth Observation data (collected by satellites) for more realistic initial conditions, instead of estimations based on modelled reanalysis meteorological data
- Continuously adjusting forecasts through real-time correction, data assimilation (integrating real-time measurements into the model), and post-processing, making predictions more reliable
- Developing Artificial Intelligence (AI) algorithms to combine forecasts and observations, reducing errors and strengthening early warnings globally
Expanding the CEMS early warning system forecast product portfolio for floods and droughts
SEED-FD develops tools to predict a broader range of extreme flood and drought events, covering areas previously lacking specific warnings:
- Flash Drought forecast product
- Seasonal Drought forecast product
- Drought tracking product
- Flash Flood forecast product
These tools enable earlier detection, supporting communities in vulnerable regions worldwide.
Use cases
The project is composed of two distinct phases – a scientific development phase and a scale-up validation phase – linked together by a prototyping stage. Use cases are essential in SEED-FD, providing real-world scenarios to test and refine our hydrological models and forecasting tools. By focusing on specific regions, the project ensures its solutions are effective, adaptable, and reliable, ultimately enhancing disaster preparedness and resilience across diverse environments.
Development Phase: Danube and Bhima
In the development phase, SEED-FD focuses on the Danube and Bhima River Basins. These data-rich regions serve as testbeds to implement and evaluate evolutions before they reach maturity, benchmarking against the existing CEMS EWS capability for forecasting floods and droughts. The Danube flows through diverse climates in Central and Eastern Europe, while the Bhima in India experiences monsoon-driven weather, enhancing model accuracy and adaptability.
Validation Phase: Juba-Shebelle, Niger and Paraná
During the validation phase, SEED-FD expands efforts to the Juba-Shebelle, Niger and Paraná River Basins. The Juba-Shebelle Basin in the Horn of Africa faces semi-arid conditions and frequent droughts. The Niger River in West Africa experiences seasonal floods, while the Paraná River in South America has significant hydrological variability. These areas represent diverse hydrological challenges, allowing to fine-tune the forecasting models and enhance prediction accuracy.
Validation Phase: World
In the final step of the validation phase, SEED-FD demonstrates the global applicability of new hydrometeorological extreme event detection and prediction products. By applying advanced algorithms to existing CEMS datasets, SEED-FD compares results with documented real-life events. This aims to enhance global disaster preparedness, providing precise flood and drought predictions around the world.
Partners
The SEED-FD project is a collaborative initiative carried out by a diverse and competent consortium of organisations, each contributing their unique expertise:
Magellium (France, prime), ECMWF (European Centre for Medium-Range Weather Forecasts, science leader), CNR-IRPI (Italy, Research Institute for Geo-Hydrological Protection of the Italian National Research Council), ICPAC (Kenya, Intergovernmental Authority on Development (IGAD) Climate Prediction and Application Center), IIASA (Austria, International Institute for Applied Systems Analysis), vorteX-io (France), POLIMI (Italy, Politecnico di Milano), Design & Data (Germany), JRC (European Union, Joint Research Centre)
To read more about each consortium member and their specific roles in the project, please click here.
ContaCT
The SEED-FD project consortium is coordinated by:
Magellium SAS
1 Rue Ariane
31520 Ramonville-Saint-Agne
France
If you have any questions, you can contact the SEED-FD team via email.
