Det Norske Meteorologiske Institutt
Niels Henrik Abels Vei 40
N 0313 – Oslo (Norway)
Dear Prof. Eliassen,
ECMWF archive data - access and pricing
In addition to the access and charging problems associated with operational real-time data discussed in the previous letter, there is another major problem area which has recently emerged in earnest: ECMWF archive data, with particular reference to historical sets of analyses and forecasts for extended periods.
The ECMWF archives are potentially of great value for European meteorology, in particular as they go back as far as 1979, and even to 1957 if re-analysis data are included. In addition to pure scientific applications, these data are increasingly being used for the development of methods and applications that can be utilised commercially both by governmental and private meteorological services.
However, the current cost level applied by ECMWF is once again excessive, thereby limiting the use of these valuable data for the benefit of the ECMWF Member States. According to a recent quote by ECMWF, a re-analysis set of 8 parameters at 1.125º resolution every six hours for a 45 year period would cost as much as £750,000, including a handling charge of some £75,000 (note: this handling charge is particularly difficult to justify in view of the fact that the costs of the Centre's telecommunications facilities including the links to the Member States are fully covered by the ECMWF budget).
In any case the above numbers are astounding and would, unless lowered substantially, put a stop to using ECMWF data for the planned purpose and similar ventures. Ironically a corresponding re-analysis data set can be acquired from the United States at a cost of some $1500.
Any commercial division of a national meteorological service should because of the level playing field requirement specified by the European Commission in its Comfort Letter on ECOMET dated 21 October 1999 pay compensation for any historical data potentially available for commercial applications. In view of the enormous amount of data and associated costs this is in many cases unrealistic.
Therefore technical arrangements should be introduced whereby both governmental and private sector actors are given selective access to the total set of historical data on the same technical and financial conditions. However, such arrangements would take many years to fully implement.
In view of the above the best solution at least in the relatively short term would seem to be making ECMWF archive data available at no license cost to users in the Member States, in line with the ECMWF Convention. This would also automatically and immediately create the level playing field required by the European Commission, whilst minimising the administrative effort for all concerned parties.
This matter and others related to it are likely to be addressed more fully in a further communication.
Yours sincerely,
Daniel Soderman
PRIMET Secretary General
Tel.: +358-9-8193 4488
| < Prev | Next > |
|---|