WMO Lead Centre for Wave Forecast Verification established at ECMWF

Thomas Haiden
Zied Ben Bouallègue
Richard Mládek
Jean-Raymond Bidlot

 

In 2016, the World Meteorological Organization (WMO) Commission for Basic Systems recommended that ECMWF become the Lead Centre for Wave Forecast Verification (LC‑WFV). With more than 20 years’ experience in wave forecast verification and wave model intercomparison (see ECMWF Newsletter No. 150, winter 2016/17), ECMWF was ideally placed to formally take on this role. Three years later, the LC‑WFV has reached a stage where most centres contributing to the original intercomparison are providing data to the new system, and where verification results are published regularly on the LC-WFV web page at https://confluence.ecmwf.int/display/WLW. The role of Lead Centre enables ECMWF to immediately identify weaknesses in its wave forecasts compared to others, which helps to inform further improvements to the wave model. Model intercomparison is based on the exchange of forecast fields rather than scores, making it more sustainable in the longer term and providing the necessary flexibility for introducing new scores and observation datasets in the future.

Acronym

Centre

Country

BoM

Bureau of Meteorology

Australia

DMI

Danmarks Meteorologiske Institut

Denmark

DWD

Deutscher Wetterdienst

Germany

ECCC

Environment and Climate Change Canada

Canada

ECMWF

European Centre for Medium-Range Weather Forecasts

Europe

FNMOC *

Fleet Numerical Meteorology and Oceanography Center

USA

JMA

Japan Meteorological Agency

Japan

KMA

Korea Meteorological Administration

Korea

LOPS *

Laboratoire D’Océanographie Physique et Spatiale

France

METEOAM

Servizio Meteorologico dell’Aeronautica Militare

Italy

METNO *

Norwegian Meteorological Institute

Norway

METFR

Météo-France

France

NCEP

National Centers for Environmental Prediction

USA

NIWA

National Institute of Water & Atmospheric Research

New Zealand

NZMS *

New Zealand MetService

New Zealand

PRTOS

Puertos del Estado

Spain

SHNSM

Department of Meteorology of the Naval Hydrographic Service

Argentina

UKMO

UK Met Office

UK

Contributing Centres. As of October 2019, 14 wave forecast centres regularly provide model fields to the Lead Centre. Data from another four centres, marked by an asterisk, are in preparation.

Collecting and archiving forecast data

ECMWF gathers and archives a set of selected model fields relevant to wave forecasting activities under an agreed format. The data are produced by operational global or regional forecasting systems. The following parameters are exchanged:

Atmospheric forcing

  • 10 m wind speed U and V components

Wave fields

  • Significant wave height
  • Peak period
  • Mean wave period based on the second moment of the frequency spectrum
  • Mean wave direction

The fields are provided on a regular latitude-longitude grid at the resolution that best matches the native resolution of each contributing model. The data are encoded in GRIB format using WMO-compliant templates. During the development phase, ECMWF assisted contributing centres in the conversion to GRIB. Participants are committed to guaranteeing the steady and reliable provision of their data and are asked to communicate any relevant changes in their systems as close to real time as possible.

Observations

Forecasts of the above parameters are evaluated against quality-controlled in-situ observations from about 400 buoys and platforms available to ECMWF. Most of these are located in the coastal areas of North America, Europe, Brazil, Japan, Korea, India and Australia. Others are part of different tropical buoy networks (TAO, TRITON, PIRATA, RAMA) or of the OceanSITES network. It is anticipated that more in‑situ observations will become available over time. They will be added following careful selection and quality control. Participants are strongly encouraged to promote the exchange of in‑situ wind and wave observations. Other verifying data available to ECMWF, such as altimeter wave heights, will also be considered in the future.

Scores

Scores computed on a regular basis include mean error (ME), root mean square error (RMSE), error standard deviation (SDEV), scatter index (normalised standard deviation of error), symmetric slope (variance ratio), and quantile-quantile (Q-Q) plots. These metrics have proven useful in previous ECMWF-led wave model intercomparison activities. Aggregated scores are generated for a wide range of domains, from global down to regional ones (e.g. the Mediterranean, or the North Sea). All results are shown on the Lead Centre’s web page, together with maps of data availability.

Contributing centres

As of October 2019, 14 wave forecast centres regularly provide model fields to the Lead Centre (see table). Four centres (marked by *) are in the process of setting up the data provision. Depending on the centre, the data are sent up to four times per day (from 00/06/12/18 UTC model runs). The lead time range varies between +72 and +288 hours with time steps from one‑ to six‑hourly. More details about each contribution can be found on the LC-WFV web page.

Locations of selected buoys providing wave height data for verification. The map shows the distribution of locations for which measurements of significant wave height were available during summer 2019 (June–July–August).

Technical aspects

The agreed data format is WMO-compliant GRIB edition 2. The ECMWF data acquisition system (ECPDS) is used for data exchange. The data processing before the actual verification consists of acquisition, validation (encoding and basic data quality checks), and archiving in MARS. It follows well-established workflows as used by other projects, such as TIGGE for medium-range weather forecasts or S2S for sub-seasonal to seasonal forecasts. This makes it possible to run the processing smoothly and in a flexible way using parallel processing and back archiving. Nevertheless, about 10 to 15 data issues need to be resolved on average every month to keep the archive in order and eliminate any data gaps. The most frequent issue is the transfer of incomplete or corrupted data, usually as a result of problems or technical changes in the processing chain on the provider side. In order for the Lead Centre to be able to sustain its activities in the longer term, it will be important to reduce the number of such incidents.

Outlook

Once the initial setup of the Lead Centre has been completed, ECMWF will explore possible extensions of the verification, for example the use of observations from drifting buoys, which has become possible now that fields rather than scores are exchanged. ECMWF would also like to take this opportunity to thank the contributing wave forecast centres for their efforts in implementing the data exchange and for their continued support for the Lead Centre’s activities.