2017-issue-4-item03

FEWS news: issue 4, 2017

 

 

 

Global flood forecast viewer to increase flood preparedness worldwide

Floods are the most frequent natural disasters that affect millions of people and properties worldwide. To manage flood risks, preparedness by the means of flood forecasting and early warning services is vital. Many areas that are at risk of flooding still lack adequate forecasting services, though. In these areas, global flood forecasts can contribute to the solution. To this end, Deltares shares results from its global storm surge and flood forecasting models through GlobalFloodForecast.com.

Forecasts for the upcoming 7-10 days
GlobalFloodForecast.com presents the results of the global flood and storm surge models of Deltares. These models provide the forecasts for the upcoming 7-10 days on soil moisture, discharge, water level and surge. The viewer was launched by Deltares at Global Flood Partnership 2017 conference in Tuscaloosa, USA at end of July.

Technical background
The information shown on GlobalFloodForecast.com is generated by two operational Delft-FEWS systems, running global forecasting models. These systems are the global storm surge system (GLOSSIS) and the global flood forecasting system (GLOFFIS).

For the global storm surge information system the Global Tide and Surge Model (GTSM) is used, which is a DFlowFM model using an unstructured grid to have higher resolution results near the coast. The resolution varies from 50 km offshore and 5 km grid cells near coastal areas. It currently produces forecasts on water levels, currents and storm surge and work is ongoing to also include wave forecasts. Furthermore data is collected on over 600 locations around the entire world to validate the model and work towards data assimilation to further improve the model results.

The underlying hydrological models of GLOFFIS are the PCRaster Global Water Balance model (PCRGLOB-WB), World-Wide Water Resources Assessment model (W3RA) and WFLOW models. The W3RA and WFLOW models presently apply a resolution of ~50 km, and efforts are underway to increase the resolution to 5 km. Both the W3RA and PCRGLOB-WB models are run in deterministic and/or ensemble mode using meteorological forecasting from GFS, GEFS and ECMWF-EPS (latency 2 days). Besides running these forecasts operationally, GLOFFIS will be used for experiments into predictability of floods (and droughts) and their dependency on initial state estimation (e.g. through data assimilation), meteorological forcing and the hydrologic models used.

Who can use it?
The viewer offers information for the following stakeholders in particular:

  • Authorities in areas lacking forecasting services
    Global flood forecasts can be crucial in areas that do not have an operational forecasting system in place. They can provide boundary conditions for more refined local models so that the authorities can take early actions such as issuing early warning.
  • Humanitarian aids and donor agencies
    Humanitarian aids and donor agencies can use this tool for strategic planning and mobilisation of human resources, supplies and investments to reduce the impacts of potential disaster.
  • Researchers
    The viewer is a product of our interest and research on global flood forecasting. We encourage other researchers to use the data and work together towards better and more meaningful flood forecasting on a global scale. 

In addition, any interested parties can use the viewer for getting up-to-date information on the model results including water levels and discharges.

Next steps
The system is under constant development to increase the quality and resolution of the models and the results. Our vision is to work towards impact based forecasting, where we can make a connection between the storm surge and the hydrological model. Based on these results and detailed terrain maps we want to create inundation maps, risk maps and eventually work towards the impact based forecasting on a global scale. 
Whereas results on globalfloodforecast.com are provided free of charge, the underlying data will be distributed as a data service, to be launched in the fall of 2017. For more information, contact helpdeskdata@deltares.nl.