by Pirkko Saarikivi, Daniel Söderman and Harry Newman
Weather Service Finland Ltd.

As the new century begins, meteorology in Europe is at a crossroads. Over the past twenty years, many National Meteorological Services (NMS’s) have begun to commercialize access to the observational data they monitor and collect. Though seemingly satisfying short-term budgetary necessities, the restrictions on the free flow of information that result from these policies and practices are having a serious effect on weather research and forecasting across the Continent – and, unless they are remedied soon, will compromise European meteorology for years to come.

The last two decades have seen the rise of private sector weather services throughout Western Europe. Guided by an entrepreneurial spirit and unencumbered by the bureaucratic inertia all too common in governmental agencies, these companies have been able to utilize advances in computing technology, numerical modelling, and the fundamental understanding of weather patterns to create unique and valuable products for a wide range of applications.

Though still only a fraction of the market, the 25 – 30 private weather companies are fast becoming crucibles for the most exciting developments in meteorology today. As a whole, the private sector is demonstrating that it is able to respond quicker and more creatively to new technological challenges such as the emergence of Internet-linked mobile WAP-phones, and to provide more imaginative and accurate value-added products than the existing national agencies. These trends have not gone unnoticed by the governmental weather services.

It is clear that a new and more competitive era is dawning in meteorology in Europe. However, in order to function at their highest, private weather companies – as well as other researchers and forecasters – need access to the raw data that only the National Meteorological Services can provide. Instead of reducing the obstacles to obtaining this information and thus collaborating in the creation of the new structures that are arising, the NMS’s have resorted to the time-honored tradition of resisting change by erecting barriers and attempting to centralize control. Or as the Americans would put it, they are “circling the wagons.”

This behavior on the part of the governmental weather services, which should be the standard-bearers of weather science in Europe, is disappointing. Without free and open collaboration between the providers of essential data and the leading research and forecasting talents (in and out of the private sector), Europe runs the risk of falling behind in the basic research and applications of meteorology. Quite simply, we will lose the ability to compete in the field. Such a development would have serious repercussions for meteorology not just in Europe, but worldwide.

One of the guiding principles of the international meteorological community from its origins has been the open and unrestricted exchange of observational data produced and collected around the world. This practice is a natural consequence of the fact that weather is global and all nations need data from many countries to meet their individual research and forecasting needs. To limit the exchange of data in any one region could reduce the value of the analysis and research coming out of that region and end up affecting many other regions as well.

Furthermore, in an era where global and regional atmospheric models have become the norm for operational forecasting, the need for comprehensive worldwide weather information has become essential. An increasing number of countries nowadays require real-time access to observational data on a global basis. like the weather systems it studies, meteorology is now an irretrievably interdependent and global phenomenon.

Unfortunately, old habits die hard and the recognition of new realities takes time. In Western Europe, the National Meteorological Services have been the central bodies for the acquisition and exchange of national and international meteorological information for decades. In this privileged position, they have been able to determine the direction and form of research, the quality and distribution of data, and significantly influence governmental spending and decision-making. Towards these ends, they have also managed to take control of two intergovernmental agencies established by European countries, the European Centre for Medium-Range Weather Forecasts (ECMWF) and the European Meteorological Satellite Agency (EUMETSAT) and use them to consolidate their position and authority.

In addition, at the end of 1995, in conjunction with the directorate-General for Competition of the European Commission (DG-IV), the National Meteorological Services joined together to form a European Economic Interest Grouping called ECOMET. ECOMET is a quasi-commercial enterprise whose objectives (as stated in its charter) are:

1. to maintain and increase the free exchange of meteorological information;
2. to improve the availability of all data and products from the NMS’s for commercial use on the European Market;
3. to create the conditions for fair competition under European rules; and
4. to have commercial users of meteorological services contributing to the financing of the infrastructure of the National Meteorological Services.

On the surface, it would seem that these goals would go a long way towards making data more readily available throughout Europe. In practice, however, ECOMET’s impact has been the reverse. Among weather scientists, ECOMET is primarily known for the price list it controls of meteorological products based on data provided by the NMS’s. Rather than serving as a means for encouraging and supporting a healthy environment for research, ECOMET functions primarily as a price-fixing cartel for the national weather bureaucracies.

Prices are kept artificially high and have been known to change without much notice. For instance, in autumn 1999, the cost to users of essential radar data was increased by almost one order of magnitude. This kind of pricing wreaks havoc on research budgets and serves only to stifle exploration and ideas. As a result of these and other policies, use of basic observational data and model products outside the NMS’s themselves remains low. In fact, companies and university researchers are increasingly turning directly to the United States (via the Internet) for the information they need.

The formation of ECOMET is only the latest step in a process that started in the 1980’s when the weather services (and all national agencies in Europe) were required by their governments to justify the money spent on them and to become more responsible for raising revenue. Under this previously unheard of financial pressure, the national meteorological services turned to the only commodity they could put their hands on: the data they were collecting. These came to be viewed as much for the income they could provide as for their scientific value.

For the members of ECOMET, on the other hand, use of observational data and the model products derived from them remains more or less free of charge. In other words, information gathered and analyzed at taxpayers’ expense is being utilized purely for the profit of the national agencies. Private sector weather services, which also contribute as taxpayers to their respective governments, are denied the same level of access to this material. Since each NMS develops and sells its own value-added products from the raw data it obtains supposedly for the public good, this is an inherent conflict of interest, if not an unfair trade practice.

The above practices and policies combined with the inefficiency of governmental organizations in general create other problems. Contrary to public mandate, most of the potentially available weather data never reaches third parties under any circumstances. There are several reasons for this: sufficient information about available data is not made known to potential users; only a fraction of the total information is ever made available to non-NMS users; distribution channels are inefficient and inadequate.

When faced with the same pressures to streamline practices and develop new sources of revenues, the national meteorological services in the United States took a different path. As summarized in an excellent paper by Weiss and Backlund (1996), “United States policy is based on the premise that government information is a valuable national resource and that the economic benefits to society are maximized when government information is available in a timely and equitable manner to all. Policies such as charging no more than the cost of dissemination and prohibitions against restrictions on the reuse or re-dissemination of government information are aimed at achieving this goal.”

As an example of the above policy, the National Oceanographic and Atmospheric Agency (NOAA) operates U.S. Government servers that maintain meteorological and hydrological data. In addition, the servers access a selection of current official weather observations, forecasts, and warnings from U.S. Government sources for use by the national and international community. There is no regulation or restriction on use of any kind.

In the United States, the private sector research and business communities fully support the government’s policies on freedom of information and have actively sought to have them implemented on an international basis. In regards to meteorology, the Coalition for the Open Exchange of Global Data, representing the private weather industry in the U.S., recently authored a position paper in favor of an open data exchange policy in Europe, which was approved by more than 25 companies. The key arguments they present in opposition to the data restrictions imposed by ECOMET are:

1. That worldwide science, education, research, and industry would suffer.
2. Data restrictions would create immediate conflict between nations.
3. Data restrictions would undermine U.S. policy and laws.
4. Data restrictions harm healthy competition.
5. Restrictions on data have time sensitive repercussions.
6. Entrepreneurs would ultimately defeat restrictions on data and information.

These differing positions between Western Europe and the United States were discussed at the 11^th World Meteorological Organization (WMO) Congress in 1991 and a working group was set up to study the question in depth. The work of this group led to the adoption at the 12^th World Congress in 1995 of WMO Resolution 40, which defined a policy and procedures for the international exchange of data. Of particular relevance were the establishment of two categories of data: essential data, on which no exchange restrictions could be placed, and “additional” data, on which conditions for use could be defined by the producer. Resolution 40 also stipulates that research and teaching activities have access free of charge to WMO data and products.

Though the resolution in general favors the free and open exchange of meteorological information, one major restriction was introduced. Conditions were set for the re-exportation for commercial purposes of data classified as “additional.” For instance, if a private company wishes to commercialize in Europe information of European origin classified as “additional” and obtained free of charge from the United States, it must ask for the authorization from the agency which originally produced the information.

All member states are required by the WMO to make a declaration as to what is “essential” and what is “additional” data. As soon as Resolution 40 was adopted, Western European countries began to designate, seemingly arbitrarily in some cases, which information was “additional” and hence subject to conditions on use and reuse. Some went so far as to remove data from WMO supervision entirely. In the case of the U.S., all data have been declared essential and hence subject to unrestricted access internationally.

The difference between the information policy in the United States and its European counterparts could not be larger. In Europe, all essential data are strictly controlled by the national meteorological services and normally only made available selectively and at significant cost to other parties. In the U.S., all basic observational data as well as model output are in the public domain, and directly accessible via operational servers and satellite data feeds.

In an effort to bring the European data system more into line with the American approach to information policy, the leading private sector weather services in the fall of 1999 banded together to form the Association of Environmental Data Users in Europe. In collaboration with the older Association of Independent Weather Services, the group has begun to campaign for a more sensible and open policy on the exchange of essential data in Europe.

From a legal perspective, such efforts would appear to be unnecessary. The first directive regarding freedom of access to information on the environment in Europe was adopted in 1990 by the European Community (directive 90/313/EEC) for implementation by the end of 1992. The key articles of this document state:

Furthermore the European Commission published in 1998 a Green Paper on Public Sector Information in the Information Society, which in general supports the views and concerns of the private sector, as highlighted in this paper.

Unfortunately, political considerations and the influence of governmental bureaucracies on policy and practice make the situation much more complex. On 21 October 1999, the joint sale of meteorological products by ECOMET was authorized by the European Commission, stating that it “has satisfied itself that there is still adequate competition between the ECOMET members, particularly on prices, since their commercial divisions compete with each other, and that the grouping operates in such a way as to permit fair competition with independent service providers.”

The main factors upon which the Commission purportedly based its decision were:

It should be kept in mind that the preparatory work leading to this unfortunate decision started in 1995 and thus was performed mainly under the previous Commission. However, it is very difficult from a private sector perspective to understand how an entity that to all intents and purposes acts as a monopoly in its control and distribution of data - data that is essential to all parts of society - can formally be declared by the EC as acceptable within a European free market environment.

In addition, the specifics of the EC decision are biased towards the state-owned sector to an almost absurd degree. For instance, the infrastructure that revenue from data sales will pay for has in most cases already been funded in full through the state budget. Furthermore no reasonable person would believe that there is no cross-subsidization if the public and commercial activities of a governmental agency are located on the same premises.

Despite this blatantly protectionist action, the emergence of a fully-formed private sector meteorological industry in Europe is inevitable. Recent strengthening of Competition Laws and their application both by the EC and on the national level clearly require removal of all restrictions on access to essential observational and model data imposed by the national meteorological services. A clean separation between the commercial and non-commercial wings of the NMS’s is also essential and is already happening in some countries.

If these steps are not taken, the NMS’s will find themselves subject to sustained (and probably successful) legal appeals within their respective countries and, if necessary, in the EC itself. The national meteorological services are in a highly indefensible position, having misrepresented to the Commission for more than five years the actions of ECOMET (particularly in regards to the staggering price increases) and using their control of observational and model data to maintain an unfair trade advantage over their competitors, in this case private sector weather services in Europe and the United States.

In consequence, the only feasible solution for Western Europe in the longer term can be summarized as follows:

1. all NMS’s within European Commission countries must strictly separate their public and commercial activities if they want to continue in a fair and competitive environment as regulated by national and EC laws (so far such a change has been completed in The Netherlands);
2. all basic observational and model data should be made available on the same conditions and at the same prices, ideally at nominal or no cost, to all players in the market place (including any commercial wings of the NMS’s);
3. an efficient mechanism for real-time access to both metadata and essential data must be established in all countries; and
4. the basic observational and model data referred to above should be given the status of ‘essential facilities’ and the agencies responsible for their production, maintenance and distribution guaranteed sufficient resources for the foreseeable future to do their job properly.

If the above cannot be achieved, ECOMET practices will continue to significantly slow down research and development activities in Europe in what is one of the most economically, socially, and scientifically important fields. By contrast, the open policies in the United States – European meteorology’s chief international rival – will continue to stimulate rapid developments. This will further widen the gap between the United States and Europe in the field of environmental research and applications, and also have a detrimental impact on the competitive abilities of industries dependent on environmental data and products.

In summary, data access problems in the U.S. have been solved neatly, as all data held by governmental agencies belong to the people by law. Therefore, basic meteorological information is openly and freely made available to anyone. Value-added services are provided by the private sector, without competition from the government.

In Europe, the state-owned weather services view the basic information as a set of products to be exploited and sold. This approach is questionable for many reasons. First, the data has already been funded by taxpayers and is being monopolised for further profit by the state. Second, because of their monopoly status, state weather services may apply unreasonable, if not discriminatory, prices and conditions on the information they sell. Third, and most troubling, state-owned services also produce the derived information mentioned above. In other words, they are using at no cost to themselves the basic data they are selling to their competitors in the private sector.

Comparing the volume of the commercial weather business in Europe, the United States, and Japan clearly shows the negative influence of restrictive data policies towards the weather industry.  In the U.S. and Japan, where the use of public information has been open and effective, the annual sales of weather services total $430 and $300 million, respectively, whereas the European sales are lagging well behind at $170 million. Adjusting for the gross national products of these market areas, the level of the European business should in fact be three times higher or more.

The private sector weather industry in Europe is as yet relatively poorly developed and struggles for survival, due mainly to the monopolistic practices of governmental weather services, which set high data prices and create other difficulties to getting the basic information necessary for commercial activity. This situation has prevailed for decades. Besides limiting industrial development, it also means that the citizens of Europe have not been getting all the benefits they are entitled to through innovative and modern weather services. 

However, competition in the European weather industry cannot be turned back. Indeed, applying to Europe the U.S. policy of free access to basic data seems the only way to substantially expand the European weather market. Further liberalization of weather data and services along these lines would increase overall revenue and strengthen the infrastructure for both state-owned and private weather services. It is only a matter of time before such changes take place, but like the weather itself, the speed of change is difficult to predict.

References

# # #