«Medium-Range Weather Prediction Austin Woods Medium-Range Weather Prediction The European Approach The story of the European Centre for Medium-Range ...»
Weather forecasting is a valuable business. One Euro invested in meteorology is generally recognised as yielding ten to twenty Euro or more in terms of profits made, casualties avoided, harvests saved and so on. The more accurate the forecasts, the more they are worth. With a good fourday forecast, retailers can send their ice-cream orders to their sunniest outlets and adjust their window displays for rain or sun. Cinema audiences, medicine consumption, routes taken by ships and aircraft, and electricity consumption all vary according to the weather. Weather is the single most important factor in influencing price volatility, volume fluctuations and revenues in the energy industry. In winter, power companies can save perhaps €100,000 a day if they know in advance how high users will turn up their heating. In response to the deregulation of the power industry the weather derivatives market was developed. The companies involved needed a financial vehicle to help manage their exposures to weather risk. Re-Insurers and financial institutions soon entered the market, and the market expanded to include “end user industries” that are affected by the weather, such as beverage sales and agriculture. The more valuable the forecasts become, the more the commercial companies want to get into the business.
In the decades after the Second World War, funding from aviation supported in large part the development of many of the European National Meteorological Services (NMSs). For example, aviation required, and to an extent paid for, the expensive network of weather ships providing essential observational data over the Atlantic.
We need to look at the Centre’s involvement in commercial issues in the wider context. In Europe, commercial meteorological activity by non-governmental organisations, having been generally at a low level, started to pick up in the 1970s. There was a growing, and potentially profitable, demand for 221
222 Chapter 18applied weather services. As it happened, this coincided more or less with the years of the Centre’s establishment. Indeed, we have seen in Chapter 4 the study on the potential significant economic benefits of good-quality medium-range forecasts; this was one of the justifications for the establishment of the Centre.
The commercial meteorological sector in the USA was becoming increasingly active in the late 1960s. The private companies, both in Europe and the USA, were keenly interested in using ECMWF forecasts as soon as they became available. Some European NMSs also began commercial activities, so that governmental agencies found themselves competing with the private sector. The Centre therefore developed its data policy in the framework of some difficult discussions affecting the NMSs. The Centre and its Council were not in the forefront of these discussions, but were concerned by them.
The Centre watched with interest the meetings in WMO on the subject.
The commercial interests of Europe’s state-owned NMSs were, and continue to be, widely different. Some had a duty to increase revenue from selling their own and ECMWF forecasts, in part as a response to reduced government support. Others had no such duties. Their abilities to exploit commercial opportunities varied.
Some complained that the commercial companies had unfair advantages.
The US National Weather Service (NWS), according to the rules that govern it, is not allowed to sell information. In fact, under pressure from commercial agencies, the NWS had been obliged to stop producing “commercial value added services” for delivery to those interested in purchasing specialised forecasts. This had to be the role solely of the private sector in the USA. The NWS gave private companies the data it received from the WMO, and its own computer forecasts, without charge. NWS products, including predictions for the European area, were even distributed free of charge by the Freie Universität, Berlin. European private meteorological companies used them.
They sold forecasts based on the data, without having to invest in costly satellites and other observing infrastructure, or supercomputers.
On the other hand, the private companies complained that the NMSs had the advantage of easy access to ECMWF forecasts. They asked that these forecasts be made available to them without charge, following the American example. Some Member States funded the Centre from NMS budgets, and their income came partly from revenue raised by their commercial activities.
If ECMWF products were to be made available to all without payment, why should they not simply denounce the Convention, ending their obligation to contribute to the Centre’s budget, knowing that they would continue to get the valuable forecasts anyway? Or, should not private companies contribute Commercial issues 223 directly to the Member State contribution to the ECMWF budget?
Furthermore, we have seen in Chapter 14 that the cost of collecting suitable observational data on which the forecasts are based greatly exceeds the cost of processing the data and making the forecasts, by a factor of perhaps 100 or even more. Commercial issues indeed introduced complications into an already complicated world!
The United States government view on commercial meteorological matters was, to say the least, not widely shared by the European NMSs. There was much more private meteorological activity in the USA than in Europe.
The American companies were becoming active in Europe, competing not only with the European private sector but also with the European NMSs.
Discussions became strained. Some in the USA, with the best will in the world, found it difficult to comprehend the European point of view; the position of the USA was similarly beyond the comprehension of some in Europe. The Congress of WMO, its supreme body, noted in 1991 that “commercial meteorological activities (have) the potential to undermine the free exchange of meteorological data and products between National Meteorological Services”. A frightening abyss was facing the international world of meteorology. The consequences would have been serious for the entire world, not only for ECMWF.
Eventually WMO in 1995 passed “Resolution 40: WMO policy and practice for the exchange of meteorological and related data and products including guidelines on relationships in commercial meteorological activities”. The Resolution was wide-ranging, taking into account not only commercial matters, but data and products to be provided freely and without restriction to research, education and other users. The interests of developing countries whose NMSs could be affected by commercial sector’s commercial use of the data originating in their territory were considered. So were relations between the NMSs and the commercial sector. Resolution 40 was not perfect, but it did provide a framework for commercial activities.
Earlier, in May 1990, the ECMWF Council had set up the “Advisory Committee to consider, and make recommendations regarding, the establishment of a Meteorological Licensing Agency”. The immediate trigger for this was a meeting of the Western European Directors in April 1990, at which there had been a long discussion on the implications of commercialisation of weather information and competition in Europe. The Meteorological Licensing Agency would as one of its first steps concern itself with the sale and control of ECMWF products.
However there was a somewhat longer history of Council wrestling with the interesting, difficult and at times very complicated issue of commercialisation
224 Chapter 18of ECMWF forecasts. The Centre itself is an international organisation established by a Convention that does not include selling its forecasts as an objective. It is an organisation “owned” by its members and its role is to enhance their powers, rather than appropriating them. The Centre provides its forecasts to the NMSs of its Member States, and it is for the NMSs to decide what to do with them, commercially and otherwise.
As early as 1980 the Council had adopted “Rules governing the distribution of results from the Centre’s work”. For commercial organisations in non-Member States: “No data will be provided on any terms”; for those in Member States: “the request is to be submitted to the National Meteorological Service”.
In 1983, a distinction was made between distribution by the Centre, and by the NMSs of the Member States. Council “adopted, on a provisional basis, the draft guidelines governing the dissemination of ECMWF operational products”. ECMWF should pass on any request from commercial organisations to the NMS of the Member State concerned. If it was not clear which was the appropriate one, the request should be passed to all the NMSs. In turn the NMSs “should not distribute ECMWF products to bodies in a nonMember State” — this was in fact a complete ban on such distribution.
However the legal and political situation was not at all clear. In the international arena, there can be something of an ambiguity concerning the sovereignty of a State, and its rights and duties as a member of an international organisation. There was lack of agreement among the Member States on how commercialisation could be approached. Every year from 1981 to 1985 the Council found itself discussing dissemination of ECMWF products. It was not even clear whether this was an issue for the Council. In 1985, the position of Sweden was that the NMSs could not be bound by guidelines adopted by the Council. Each Member State had the freedom to do as it wished. France agreed that the Convention did not bind the Member States with respect to their national sovereignty. The problem of selling or not selling ECMWF products was not a problem for the Centre and should not be discussed by Council.
Such fundamental lack of clarity in a potentially important and divisive area was highly unsatisfactory, especially taking into account the growing activities of the private weather companies in Europe. In addition, the European Commission was taking a growing interest in commercial activity in the Common Market. The Single European Act, re-launching the single market by reducing trade barriers, was signed in 1986.
On 23 June 1987, a press release from Accu-weather Inc of State College, Pennsylvania announced “a major agreement with a consortium of Commercial issues 225 European Governments giving Accu-weather exclusive rights to market the European Model in North America”. Nordmet, a consortium of the NMSs of Denmark, Finland and Sweden, had negotiated the contract with Accuweather. The contract, to extend over a 15-month period, committed Nordmet to providing Accu-weather with information based on ECMWF forecasts. Neither Richard Hallgren, Director of the US National Weather Service, nor Lennart Bengtsson, the Director of the Centre, nor the ECMWF Council, had been informed in advance of this contract.
On 11–12 June 1987, less than two weeks before the Accu-weather press release, the Council had had a lengthy discussion on distribution and charging policy. EUMETSAT, the European meteorological satellite organisation supported by many of the ECMWF Member States, was considering adopting a policy on this subject. ECMWF Director Bengtsson had noted that it was essential to have a consistent policy between the two organisations, and “If revenue is generated from outside the Member States from sale of the Centre’s products, this should be shared pro rata among the Member States”.
In the discussion, many of the complications arising from commercial issues in the framework of long-standing international meteorological co-operation were raised. The Council asked Dr Heinz Reiser, president of DWD, the NMS of Germany, to convene a sub-group of NMS Directors to draw up a Report on the matter, to be considered in the autumn.
The press release from Accu-weather came as a bombshell into these delicate discussions. Hallgren sent a copy to Bengtsson, who immediately queried the Nordmet Directors. On 29 June, the Directors of the NMSs of the Centre’s other Member States were informed of the contract, and of its duration and scope, in a telex signed by the Directors of the NMSs of Denmark, Finland and Sweden. At the same time, Hallgren was formally informed of the contract. On 3 July, Bengtsson noted in a letter to the NMS Directors “the damaging repercussions this may have for international co-operation in meteorology”. Dr (later Sir) John T. Houghton, Director-General of the Met Office, noted that “from my conversations with Mr Hallgren it is clear that he and his service in the USA are seriously embarrassed: ‘how can I continue to defend free dissemination of US satellite data?’ was his reaction”. The Nordmet weather services, on the other hand, restated their belief that this “was not a matter for the ECMWF Council”.
The President of Council was requested by many Member States to convene an extraordinary session of the Council. The session was held on 4 September 1987. The Chairman Prof S. Palmieri of Italy noted that “the links of mutual trust, loyalty and co-operation among meteorologists within Europe and outside were very firmly established, enabling projects of
226 Chapter 18great value to humanity, to be established. Increasing interest in commercialisation was not a negative sign”. The Council after a wide-ranging discussion, asked that Dr Reiser’s Working Group meet on the following day and prepare Guidelines governing the dissemination of ECMWF realtime (as opposed to archive) products, to replace the existing Guidelines.
The new Guidelines were adopted in November 1987, stating that “Member States should distribute... products... to all other bodies in non-Member States only with the approval of the Council”. The contract with Accuweather, which was not at all as wide-ranging as the press release had implied, ran its course but was not renewed.
The commercialisation of ECMWF products continued to be a matter for Council discussion in the following years. As we noted, some Member States had a real interest in generating revenue from sale of the forecasts of their own NMSs as well as those of the Centre; others especially in the 1980s had little or no interest, and to an extent were observers rather than active participants in the discussions.
The problem was a difficult one to pose properly. It had many facets, and all had to be taken somehow into account
• How would ECMWF forecast products be distributed:
– within the NMS’s own State.
– to another Member State.