Evolving hydrological flood forecasting systems with globally-configured on demand high-resolution services

Abstract

High-intensity, short-duration extreme precipitation events are causing severe natural disasters worldwide, resulting in major infrastructure and property damages (mill. €/flood event) and loss of human life. The DE_330_MF project takes advantage of recent advances in short-term forecasting of these extremes, driven by expanding monitoring capabilities and enhanced high-resolution numerical weather prediction models. Breakthroughs, relative to the capabilities of established services, are under development with a focus on predictions at sub-km scales that should improve the description of precipitation extremes down to local scales.Here we present our vision for how operational hydrological forecasters across Europe may interact with the on-demand weather-induced extremes digital twin (DT). We rely on the hydrology use case within DE_330_MF, and focus on the actions needed to ensure an efficient integration of hydrological systems with the DT. To ensure relevance for flood risk management, robustness and applicability across Europe, workflows are co-designed for both the DT software service infrastructure and the real-time flood warning decision-making for nine selected hydrological use cases representing different European countries/national warning services. These events were chosen to cover a broad range of flood types, geographic locations, spatial impact scales and resulting damages/fatalities. They are key in the co-design process, and to ensure sufficient on-demand capabilities and configurability options are demonstrated for flood risks across Europe.Different levels of interaction with the DT are required to enable the ingestion of the globally-configured DT data into the diversity of existing nationally-driven hydrological modelling prediction systems. The added value of the DT data can be demonstrated by comparing hydrological simulation results using the DT forecasts with current resolution forecasts, and analysing how model output accuracy improves. In parallel, the pan-European Hydrological Predictions for the Environment (HYPE) model will be incorporated directly within the DT technical service platform, thereby demonstrating how hydrological models can be embedded in the DT structure. This will allow the DT to provide complementary hydrological information to all national warning services across Europe, which can be further evaluated by comparing with the nine national models implemented in the first step.Action plans will also be co-created with societal users to ensure advancements produced by the Extremes DT can be implemented via the proposed actionable response scenarios. Use-case relevance and progress over existing capabilities will be co-evaluated with local partners/responsible agencies, giving options for enhancing the way the DT interacts with users and their ability to request on-demand information on extreme flood events, demonstration and user requests of the tailored workflows.This work is funded by the EU under agreement DE_330_MF between ECMWF and Météo-France. The on-demand capability proposed by the Météo-France-led international partnership is a key component of the weather-induced extremes DT, which ECMWF will deliver in the first phase of Destination Earth, launched by the EC.This work is funded by the service contract 2022/DE_330_MF, an international partnership led by Météo-France under the digital twins ECMWF will deliver in the first phase of Destination Earth, launched by the EC.