The European and other Weather Forecast Centres

What are their roles?

I get asked, from time to time about the differences between the various forecast centres that issue weather charts. Some sailors prefer to use ECMWF charts, others like the UK output, others that from Offenbach etc. The following may help to clear the air - or make it more foggy!

ECMWF

National Met Services

Met Office Output

Comparison with other Centres

How Good is the UK Met Office?

See also Calculating the weather and How weather forecasting has developed

ECMWF

The ECMWF is a research organisation supported by 18 European States - Belgium, Denmark, Germany, Spain, France, Greece, Ireland, Italy, Luxembourg, the Netherlands, Norway, Austria, Portugal, Switzerland, Finland, Sweden, Turkey, United Kingdom. There are co-operation agreements with Croatia, Czech Republic, Estonia, Iceland, Hungary, Romania, Slovenia and Serbia and Montenegro,, as well as the World Meteorological Organisation (WMO),, the European Organisation for the Exploitation of Meteorological Satellites (EUMETSAT),, the African Centre of Meteorological Applications for Development ( ACMAD), the, Joint Research Centre (JRC), Preparatory Commission for the Comprehensive Nuclear Test-Ban Treaty Organisation (CTBTO), Executive Body of the Convention on Long-Range Transboundary Air Pollution (CLRTAP and the European Space Agency (ESA).
. Initially it was charged with the

development of numerical methods for medium-range (up to 14 days ahead) weather forecasting;

preparation, on a regular basis, of medium-range weather forecasts for distribution to the meteorological services of the Member States;

scientific and technical research directed to the improvement of these forecasts;
collection and storage of appropriate meteorological data.

To this end ECMWF uses, and continues to develop, a state of the art numerical weather prediction model that goes way beyond the 6 or 7 day period for normal operational forecasting. They also experiment with seasonal and annual forecasting but not climate studies. Only the output to 7 days ahead is broadcast to the general public. This is, partly at least, because the output beyond that period is still regarded as experimental.
ECMWF runs its computer model from midnight and midday. A particularly useful feature of their presentation is that it is possible to select a day, say 3 days ahead and then compare forecasts verifying at that time from previous compute runs. This gives an idea of the reliability of the forecast.

National Meteorological Services

Several European National Weather Services, the UK, France, Germany and Italy run their own numerical weather prediction models. So does a group of countries known as the HIRLAM group (Scandinavian and Low countries, Iceland, Ireland, and Spain). These centres are more concerned with the next 24 - 48 hours ahead. They are more operational in their approach than ECMWF. . Their models are likely to be optimised for their prime areas of interest.
The operational weather services eg the UK Met Office , Offenbach etc amend
their forecast output subjectively and put fronts onto their forecast charts. In other words, they do allow experienced forecasters to modify the computer output. This is done partly in the interests of making the charts look realistic and partly to take account of any later information not used by the computer. The forecasters can also look at models from the other centres and may decide that these are giving better guidance than the national model.

Output from the UK Met Office

The UK uses a data cut off of about 3 hours ie they start their forecast computer runs at about H + 3 hours but run them to 5 or 6 days ahead twice a day ie from 00 and 12 hours UTC. They rune their models at 06 and 18 hours UTC but only out to 48 hours, All the data received before 00 (or 12 UTC) are used to derive a very good first guess of the actual situation, in effect a very short period forecast. They then combine all the available new data for 00 or 12 UTC with the first guess.
In addition, the UK uses and even earlier cut off time for a UK area very high
resolution model that rune out to 36 hours or so only. (Known as the Meso-scale model)
At about 1215 and 0015 the Met Office issues to the Internet forecast charts (based on the previous 0000 or 1200 UTC data) to 6 days ahead (144 hours from initial data time). These charts are a little difficult to use at first because they are output straight from the computer and contain, in effect, three charts on each page. However, the surface isobars can be seen fairly easily.

Comparisons between different centres

ECMWF output is very clear; it is high quality presentationally and scientifically. . Rather more "fussy" charts from the Met Office can be accessed from the USAF in Germany. These are, essentially, computer output with fronts added but no other subjective amendment. Charts that are easier to use are issued on the MetFax service and on the MetWEB site. Similar charts are available also from the German Met service (DWD). Comprehensive lists of weather sites can be found on various peoples web pages. A set of sites of particular interest to sailors can be found at Essential Weather Sites for sailors which contains some notes on the various forms of output. These should be of some help in deciding which to use and what are the differences between the different UK Met Office charts.
The national weather services and the HIRLAM outputs are likely to be as good with some benefits. UK Met Office charts via the USAF base and other sites are also very clear, they show fronts and isobars, are updated twice a day and extend to 5 days ahead.

How good is the UK Met Office?

The Met Office forecasts can be particularly useful during the hurricane season. The US Severe Storm Centre at Miami agrees that the UK Met Office model handles hurricanes better than any other forecasts system. The world's airlines use UK Met Office and Washington as the two world centres for aviation (and many airlines prefer the UK - even for the southern hemisphere). In effect, the UK Met Office is regarded as the best or equal best world-wide.
That is not to imply that they are perfect. Their predictions are based upon the available data and using the best science available. But, in the final analysis, the
atmosphere is still more complex than the most complex numerical prediction models. Errors can and will occur. Never just take one forecast as gospel. Always keep updating with the latest information - your own eyes, your own barometer, the latest forecast from the BBC, the Coastguard, the Internet etc etc. Just as you continually check your navigation and pilotage, do the same with the weather. You will never guarantee not to be caught out but you will minimise the risks. On another page, I have place a very brief review of numerical weather prediction and how forecasting has evolved over the ages.

Return to Top

Return to Home Page

Amended November 2005

ECMWF
The ECMWF is a research organisation supported by 18 European States - Belgium, Denmark, Germany, Spain, France, Greece, Ireland, Italy, Luxembourg, the Netherlands, Norway, Austria, Portugal, Switzerland, Finland, Sweden, Turkey, United Kingdom. There are co-operation agreements with Croatia, Czech Republic, Estonia, Iceland, Hungary, Romania, Slovenia and Serbia and Montenegro,, as well as the World Meteorological Organisation (WMO),, the European Organisation for the Exploitation of Meteorological Satellites (EUMETSAT),, the African Centre of Meteorological Applications for Development ( ACMAD), the, Joint Research Centre (JRC), Preparatory Commission for the Comprehensive Nuclear Test-Ban Treaty Organisation (CTBTO), Executive Body of the Convention on Long-Range Transboundary Air Pollution (CLRTAP and the European Space Agency (ESA). . Initially it was charged with the development of numerical methods for medium-range (up to 14 days ahead) weather forecasting;	preparation, on a regular basis, of medium-range weather forecasts for distribution to the meteorological services of the Member States; scientific and technical research directed to the improvement of these forecasts; collection and storage of appropriate meteorological data. To this end ECMWF uses, and continues to develop, a state of the art numerical weather prediction model that goes way beyond the 6 or 7 day period for normal operational forecasting. They also experiment with seasonal and annual forecasting but not climate studies. Only the output to 7 days ahead is broadcast to the general public. This is, partly at least, because the output beyond that period is still regarded as experimental. ECMWF runs its computer model from midnight and midday. A particularly useful feature of their presentation is that it is possible to select a day, say 3 days ahead and then compare forecasts verifying at that time from previous compute runs. This gives an idea of the reliability of the forecast.
National Meteorological Services
Several European National Weather Services, the UK, France, Germany and Italy run their own numerical weather prediction models. So does a group of countries known as the HIRLAM group (Scandinavian and Low countries, Iceland, Ireland, and Spain). These centres are more concerned with the next 24 - 48 hours ahead. They are more operational in their approach than ECMWF. . Their models are likely to be optimised for their prime areas of interest. The operational weather services eg the UK Met Office , Offenbach etc amend	their forecast output subjectively and put fronts onto their forecast charts. In other words, they do allow experienced forecasters to modify the computer output. This is done partly in the interests of making the charts look realistic and partly to take account of any later information not used by the computer. The forecasters can also look at models from the other centres and may decide that these are giving better guidance than the national model.
Output from the UK Met Office
The UK uses a data cut off of about 3 hours ie they start their forecast computer runs at about H + 3 hours but run them to 5 or 6 days ahead twice a day ie from 00 and 12 hours UTC. They rune their models at 06 and 18 hours UTC but only out to 48 hours, All the data received before 00 (or 12 UTC) are used to derive a very good first guess of the actual situation, in effect a very short period forecast. They then combine all the available new data for 00 or 12 UTC with the first guess. In addition, the UK uses and even earlier cut off time for a UK area very high	resolution model that rune out to 36 hours or so only. (Known as the Meso-scale model) At about 1215 and 0015 the Met Office issues to the Internet forecast charts (based on the previous 0000 or 1200 UTC data) to 6 days ahead (144 hours from initial data time). These charts are a little difficult to use at first because they are output straight from the computer and contain, in effect, three charts on each page. However, the surface isobars can be seen fairly easily.
Comparisons between different centres
ECMWF output is very clear; it is high quality presentationally and scientifically. . Rather more "fussy" charts from the Met Office can be accessed from the USAF in Germany. These are, essentially, computer output with fronts added but no other subjective amendment. Charts that are easier to use are issued on the MetFax service and on the MetWEB site. Similar charts are available also from the German Met service (DWD). Comprehensive lists of weather sites can be found on various peoples web pages. A set of sites of particular interest to sailors can be found at Essential Weather Sites for sailors which	contains some notes on the various forms of output. These should be of some help in deciding which to use and what are the differences between the different UK Met Office charts. The national weather services and the HIRLAM outputs are likely to be as good with some benefits. UK Met Office charts via the USAF base and other sites are also very clear, they show fronts and isobars, are updated twice a day and extend to 5 days ahead.
How good is the UK Met Office?
The Met Office forecasts can be particularly useful during the hurricane season. The US Severe Storm Centre at Miami agrees that the UK Met Office model handles hurricanes better than any other forecasts system. The world's airlines use UK Met Office and Washington as the two world centres for aviation (and many airlines prefer the UK - even for the southern hemisphere). In effect, the UK Met Office is regarded as the best or equal best world-wide. That is not to imply that they are perfect. Their predictions are based upon the available data and using the best science available. But, in the final analysis, the	atmosphere is still more complex than the most complex numerical prediction models. Errors can and will occur. Never just take one forecast as gospel. Always keep updating with the latest information - your own eyes, your own barometer, the latest forecast from the BBC, the Coastguard, the Internet etc etc. Just as you continually check your navigation and pilotage, do the same with the weather. You will never guarantee not to be caught out but you will minimise the risks. On another page, I have place a very brief review of numerical weather prediction and how forecasting has evolved over the ages.