Rainfall and river flow ensemble verification

Providing a framework and evidence base for verifying and optimising current and future ensemble flood forecasting systems.

Documents

The outputs of this project are available on the Natural Environment Research Council (NERC) website.

Rainfall and river flow ensemble verification: prototype framework and demonstration: Phase 1 - report

Rainfall and river flow ensemble verification: Phase 2

Project summary

This project has developed a new ensemble verification framework to help interpret ensemble sets of forecasts of rainfall and river flow. Tools within the framework can be used to inform real-time decision-making during flood incidents.

Despite increasing skill in forecasting weather and flooding, there remains inherent uncertainties. Ensemble forecasts can help with understanding this uncertainty. Instead of a single forecast of the weather and river response, a large set of forecasts (“ensemble”) is produced, providing a range of possible outcomes

To assess past performance, verification of ensemble forecasts is required. This is useful in helping forecasters interpret forecast information during incidents as well as guiding improvements to future ensemble forecasting systems.

This project follows initial work carried out under the “Rainfall and River Flow Ensemble Verification” project (Phase 1) between September 2016 and July 2017 which provided the foundation for this Phase 2 project.

Approach

The project carried out a detailed review of different approaches that provide ensemble verification. Case studies were combined with scientific analysis to prototype, demonstrate, assess and refine a set of useable verification tools. This has resulted in a new ensemble verification framework, while a joint coding framework provides the basis for an operational system.

The project also co-developed and prototyped an ensemble forecast visualisation and verification system to allow forecasters to visualise verification information.

Results

The project’s main finding is that joint precipitation and river flow ensemble verification is possible and useful.

The project has:

  • succeeded in bringing together the meteorological and hydrological verification approaches to define, test and demonstrate a joint ensemble verification framework for ensemble precipitation and river flow forecasts

  • demonstrated how verification information can be used to enhance the use of ensemble forecasts for day-to-day flood risk decision-making

Recommendations

This project recommends implementing an interactive ensemble forecast visualisation and verification system to provide more informed decision-making at times of potential flood risk. The ensemble verification framework, joint coding framework and ensemble forecast visualisation and verification system may be used as a starting point for further development and incorporation into operational systems.

Project information

Project executive: Keith Fenwick, Flood Forecasting Centre Project manager: Dave Cox, Flood Forecasting Centre Research contractors: UK Centre for Ecology & Hydrology, Met Office

This was a collaborative project funded by the Environment Agency, the Flood Forecasting Centre (FFC) and the Scottish Environment Protection Agency (SEPA). The work was carried out by the Met Office verification team and the UK Centre for Ecology and Hydrology, with support from the FFC, the Environment Agency, SEPA, Natural Resources Wales and Deltares. Phase 1 was funded by the FFC and SEPA.

A Welsh translation of the summary is also available.

Published 1 March 2021
Last updated 25 November 2021 + show all updates
  1. Project summary published with links to the main reports on Natural Environment Research Council (NERC) website. The details section explaining objectives has been removed as it has been superseded by the summary. The project is labelled as complete. The project title has been changed to match the title used by the flood forecasting centre. previous title was: Getting the best out of Grid to Grid (G2G) river flow ensembles for flood forecasting

  2. First published.