ECMWF supports the South‐East European Multi-Hazard Early Warning Advisory System

Fredrik Wetterhall, Umberto Modigliani (both ECMWF), Milan Dacić, Sari Lappi (both WMO)


In January 2020, ECMWF became involved in the second phase of the project to build a ‘South‐East European Multi‐Hazard Early Warning Advisory System’ (SEE‐MHEWS‐A), which aims to strengthen the existing early warning capacity in the region. The project was initiated in 2014 by the World Meteorological Organization (WMO) and the US Agency for International Development (USAID). It is currently funded by the World Bank, through the Global Facility for Disaster Reduction and Recovery, and by the European Union. For 18 months, the project will test a prototype of a flood early warning system using local information and multiple models to better characterise the flood risk in selected catchments.

SEE-MHEWS-A setup. The diagram shows the suite of coupled meteorological, hydrological and marine/oceanographic prediction models in SEE-MHEWS-A as well as their links with a centralised observational database using SAPP technology and with a common information and communication platform.

ECMWF’s role in the project

ECMWF’s role in the project will be to support the implementation of the system, the tuning of the models, and the dissemination of the results through a web interface. The three areas in which ECMWF will be involved are:

  • supporting the installation and customisation of a pilot version of the Centralised Observational Database (CODB) based on ECMWF’s data acquisition and pre‐processing system (SAPP) and ECMWF’s Meteorological Archival and Retrieval System (MARS) to support the envisaged suite of coupled meteorological, hydrological and marine/ oceanographic models
  • supporting the implementation in an operational environment; the calibration and verification of the hydro‐meteorological modelling chain, building on tools that ECMWF is already operating for the Copernicus Emergency Management Service (CEMS); and further post‐processing of hydrological outputs
  • ​supporting the setting up of limited‐area model runs in the ECMWF environment and thus supporting the demonstration of cascading forecasts with multiple numerical weather prediction models at global, regional and local scales.

Countries covered by SEE-MHEWS-A. Albania, Bosnia and Herzegovina, Bulgaria, Croatia, Cyprus, Greece, Hungary, Israel, Jordan, Lebanon, Montenegro, Republic of Moldova, Romania, Serbia, Slovenia, North Macedonia, Turkey, Ukraine.

The work will be carried out by expert teams contracted by the World Bank and the WMO for the project. The role of ECMWF is to support these experts in their work in relation to ECMWF’s high‐performance computing facility and its IT/software frameworks (SAPP, ecFlow, and similar packages).

Technical setup

The project needs temporal and spatial observations of high resolution across the region. These data already exist but are not exchanged as this goes beyond the minimum required by WMO data exchange policies. Observations will be converted to BUFR using SAPP. These data will then be used in the data assimilation for the initial conditions of the meteorological and hydrological chain, and for tuning and calibration of the hydrological model.

The implementation of state‐of‐the‐art transboundary hydrological forecasting systems in an operational setup requires complex expertise, software and tools. This can be beyond the reach of national hydrological services. Through CEMS, ECMWF has gained solid expertise in setting up, running and verifying coupled hydrometeorological forecasting systems. SEE‐MHEWS‐A will provide an opportunity for ECMWF to transfer this expertise to Member and Co‐operating States in the region by supporting the implementation of such a complex state‐of‐the‐art system.

Further information

More details about the SEE‐MHEWS‐A project are available on the WMO project web page: