1. The UN General Assembly, in its resolution 44/224 of 1989, was convinced that early warning of emerging environmental threats and degradation would help Governments to take preventive action, and emphasized the importance of broader participation in Earthwatch in order to strengthen its capacity to make authoritative assessments, anticipate environmental degradation and issue early warnings to the international community.

2. While the concept of early warning is an attractive one, it has proven difficult to implement in practice. It is not easy to determine when (a) a problem is sufficiently important, (b) the evidence in support of it is adequately substantiated and convincing, and (c) the risks it represents are large enough, to warrant an early warning, or even to decide what form such a warning should take.

3. Yet the UN system must find a way to respond to increasing demands for effective ways to make operational its early warning responsibility in the environmental field as mandated for Earthwatch. Any operational mechanism to achieve sustainability requires this environmental input. Governments are no longer satisfied with reports only on the past and present state of the environment. Given the inertia in many global environmental processes, and the long lead time required to implement an effective response in society, decision-makers want future projections of those problems that are coming. The UNEP Governing Council, for instance, has specifically asked for 20 year projections to 2015. It is therefore worth considering what processes are involved in environmental early warning, what gaps exist, and how these can be filled, taking advantage as far as possible of both new information technologies and existing structures and mechanisms.

4. Put simply, an issue must be identified as a problem, the evidence for it must be collected and the scientific mechanisms involved must be determined, the risks and implications or impacts must be assessed, and if these are serious enough, an appropriate form of early warning should be communicated to those who need to respond to it. This will then set in motion other steps such as determining policy implications, planning and implementing responses, and organizing monitoring to follow the evolution of the problem. This note focusses only on the early warning dimension of this larger process.

Identification of a problem

5. In the environmental field, most problems are initially identified by the scientific community. Any early warning mechanism must therefore follow debate in the scientific community and spot emerging concerns or newly discovered problems. The interface withscientists and their procedures for peer review and consensus is critical. The difficulty with the scientific process, in this perspective, is that it is inherently diffuse and unfocussed. Discoveries and observations are reported at meetings and published, usually after some delay, in a wide variety of peer-reviewed journals, often accessible only to a small circle of specialists. A controversial observation will require confirmation by other workers, and even where a majority opinion emerges, alternative or even opposite interpretations can persist almost indefinitely. Such differences can be used by opponents of action to block any solution to the problem. Significant advances in understanding are generally pieced together from many bits and pieces of research in a whole series of technical publications. It is hard to distil out of this process a specific conclusion of policy relevance that political leaders can rely on to take hard decisions.

6. In addition, global environmental problems may arise from interactions among many human driving forces and environmental processes. These may be large scale problems that build slowly, but with increasing inertia, until some crisis threshold is reached, by which time serious damage may be inevitable. It is not certain that the discipline-based scientific community will be sufficiently rapid in anticipating or identifying these large-scale problems in time for an effective societal response. This is where modelling, scenario building, and projections can provide additional tools for early warning, at least by suggesting where observations need to be focussed to confirm or disprove predicted trends.

7. A third potential input should be through the observations of and concerns expressed by the people directly involved (workers, residents, indigenous inhabitants, etc.) who may discern significant changes in their own local environment before they attract the attention of the scientific community. Some of these warning signals may be picked up first by NGOs, so channels of communication that can reach down to the grass roots and listen to their concerns will be important. This could be linked to participatory observation programmes.

Determination of the mechanisms and processes involved

8. Once a problem is identified or at least formulated as a question, then a whole range of scientific processes need to be set in motion, including data collection, basic research, targeted studies, experimentation, modelling, etc. to document the problem and determine the processes or mechanisms involved. This could involve both the scientific research community and governmental observation programmes.

9. For an early warning mechanism, the important point is to address the right questions to the scientific community in order to stimulate their response. It is then necessary to collect the results of the scientific work so stimulated and to evaluate them.

Assessment of risks and implications

10. The most difficult element in the process is to assess the probability and magnitude of the problem and the resulting risks to society and the environment, often in the face of considerable uncertainty and scientific debate. The customary approach has been to rely on expert judgements by consultants, panels or task forces, ranging from the single author of a report to elaborate mechanisms like the Intergovernmental Panel on Climate Change. Tools to make the process more systematic, such as risk assessments and model projections, are beginning to be incorporated into these processes. The best assessments undergo some typeof peer review or other quality control. For politically sensitive issues, governments often insist on their involvement, such as through government-nominated expert groups.

11. Assessment mechanisms have generally been ad hoc, with a new process created for each assessment or state of the environment report. However, there is now a tendency for some assessment mechanisms to be perpetuated under the major conventions (ozone, climate change, etc.).

12. These specialized assessment procedures are not sufficient to cover the continuing Earthwatch mandate to watch out for and to give early warning of emerging environmental problems that require a response at the international level. Present mechanisms work slowly, may be inefficient or intermittent, and leave major gaps in coverage.

Formulation and delivery of an early warning

13. A principal weakness in the UN system identified in the in-depth study of Earthwatch is in the delivery of the results of the assessment process. The traditional approach has been to prepare and document the findings in a comprehensive report of impressive size, and then to pay little attention to its distribution. Even if such reports reached senior civil servants, advisors and political leaders, they would never have the time to read them. Reports to intergovernmental bodies are usually read by someone in the major capitals to prepare a briefing for their delegation, but the results do not go beyond this narrow circle and they are seldom read again.

14. For early warnings to be effective, there must be new, short and more focused delivery of the principal conclusions to decision makers, governments, intergovernmental bodies, the media, NGOs and the general public. This cannot be an afterthought, but must be given sustained attention by the responsible organizations, assisted by communications specialists.

SOME OPTIONS FOR OPERATIONAL IMPLEMENTATION

15. What has always been lacking in the UN system, despite its general mandate for environment assessment, has been a continuing mechanism to define the right questions, assess the available information, and bring the results to the attention of the international community. Even within UNEP, this has been an ad hoc process over the years involving senior advisers, consultants, and expert meetings such as those at Villach on global warming. Thus, while many supporting elements for assessment are in place within the UN system, the intellectual leadership function on environmental issues which it should be providing to the international community is inadequate to the complex and increasing challenge from global change.

16. What is needed is an early warning mechanism that has the respect and support of both the scientific community and governments, with the highest level of intellectual rigor and objectivity, able to review the full range of environmental problems, to spot those of particular concern, and to draw the conclusions necessary for early warning from the often uncertain and sometimes contradictory base of scientific information.

17. The following are some possible options to respond to this need. They are not necessarily mutually exclusive and could be combined or developed sequentially. The Earthwatch Working Party may wish to consider these and any others that may be submitted for discussion, and to propose some possible steps forward to fill this gap.

Expert panels

18. Since no one person can master all the fields of knowledge that need to be included in environmental early warning, some type of consultative expert process will be required. Some national academies of science or scientific societies have taken on this role at the national level, providing reports that distil the best scientific opinion on critical issues. Usually a separate panel is convened for each issue or report. The report may be requested by a government body, or may be produced at the initiative of the scientific institutions. Usually these reports reflect a national perspective, and they are thus not necessarily appropriate by themselves as guides to international action.

19. The only equivalent structure internationally is the International Council of Scientific Unions and its Scientific Committee On Problems of the Environment (SCOPE). SCOPE has used expert panels to review and prepare significant statements on international environmental problems and activities. It is well placed to propose scientific agendas and to assemble a balanced international scientific perspective on environmental issues, but its resources are very limited. It could be mandated to undertake an early warning function for Earthwatch if the necessary support could be found. However, its views would only reflect those of the scientific community, and it would be necessary to build government confidence in the results of this process.

20. Expert panels have also been used frequently by international organizations to prepare reports in their sectoral or geographic areas, usually on an ad hoc basis. Since such reports are expensive, they are not prepared very often in any one field, and each panel usually has to start over again from scratch. There is little opportunity to learn from past experience, making the process relatively inefficient.

21. There is an additional disadvantage to the use of ad hoc expert panels where future perspectives or projections are concerned. Since a panel is disbanded when its report is completed, it is no longer around to observe the accuracy of its vision and to make adjustments where required in the light of new evidence. If the process of early warning is to guide policy-making effectively and earn the confidence of decision-makers, it should involve some continuing accountability and be able to up-date its forecasts whenever required.

22. Furthermore, there is government disillusion with standing high level advisory committees, perhaps because they tend to produce weighty reports of limited impact.

Regional approaches

23. Many environmental problems may have characteristics specific to a particular region, or become evident first in a regional situation. Regional organizations or the regional offices of international organizations may be well placed to implement an environmental early warning mechanism for the region, either independently, or as an intermediate stage in aglobal early warning process. In most instances this would be done through some kind of regional expert panels, with the same advantages and limitations as those at the national or global level.

24. Unfortunately, the scientific community is less well organized at the regional level. Most scientific bodies are either national or international. Special scientific structures might have to be created to implement such regional early warning mechanisms.

A scientific "tribunal" or hearing process

25. One possibility would be to draw on the model of a tribunal or high court. A law court is a legal mechanism to determine the truth in an area of uncertainty and contradictory views by weighing the evidence according to clear standards and precedents. Transposed to a scientific context, this could become a concrete and specialized form of peer review intended to produce policy relevant outcomes.

26. A scientific tribunal or panel of assessors for Earthwatch would consist of highly-respected scientific experts charged with preparing short policy-oriented opinions, assessments and warnings on international environmental problems. They would define the critical questions and identify issues that they should take under advisement, either on their own initiative or based on submissions from governments, international organizations or scientific bodies. They would review the scientific literature on the issue, and organize public hearings where the scientists most involved could present their evidence and conclusions. Such hearings would foster transparency in the peer review process necessary to build political confidence in the outcome. They could also become media events which would help to educate the public to the issues involved and to the state of scientific thinking. The panel would then weigh the evidence available and issue an opinion or assessment designed to provide the basis for policy formulation and management action. These opinions would crystallize the best scientific thinking of the moment as an input to the political process. As with any science, these opinions would be subject to revision by similar open processes in the light of new evidence.

27. It would be important for the panel members or assessors not to have any personal involvement in the outcomes of issues they take under deliberation, a problem with many present forms of scientific peer review. Initially, panel members could be selected on an ad hoc basis, perhaps through a body such as the Scientific Committee on Problems of the Environment (SCOPE) of ICSU, with expertise appropriate to the issue under review. For the results to be accepted in political circles, some attention to geographic balance and other forms of balance would be necessary.

An Earthwatch assessment panel

28. If the initial trials of this approach prove effective, consideration could be given to establishing a permanent high level Earthwatch assessment panel or tribunal. As with a high court, the appointment of permanent members of distinguished attainments and proven qualities of scientific judgement and the ability to relate science to the policy process would free the panel members from career pressures that might otherwise bias the assessments. Such a standing panel could include 5 to 9 experts representative of the major environmentaldisciplines (with regard to due geographic balance), named perhaps by the Executive Director of UNEP after wide consultation with ICSU and the interagency partners in Earthwatch, and confirmed by the UNEP Governing Council, to ensure their acceptability to both the scientific and political communities. Members of the panel could be named to serve until retirement, comparable to high court justices, or for set non-renewable terms of at least 5 years.

29. The panel would be expected to operate with full transparency. In addition to organizing open hearings on the scientific evidence as described above, it could hold debates in international scientific congresses and request reviews by existing expert networks and bodies including those of ICSU, SCOPE, IGBP, IUCN, etc. It could also use the new electronic media to collect scientific opinion and to involve a wide range of experts in its debates. It would thus organize an open process to review the state of scientific knowledge on any issue under consideration, after which it would deliberate and issue an authoritative opinion, citing the key lines of scientific evidence and their weighting of them, with the possibility of minority dissenting views where appropriate. It could also review and modify earlier opinions in the light of new evidence.

30. If it was determined that issues required more specialist scientific expertise than could be available on a small panel, then the legal model could be imitated further by assembling, for critical issues, a jury of scientific peers to hear the evidence and to provide their combined judgement to the panel for incorporation in the final opinion.

31. Such an approach would build on the principles of open debate and peer review that are at the heart of science, but that are presently embodied in the slow incremental process of scientific publication, in a way that would allow conclusions to be drawn suitable for policy formulation and management action. If the panel were a standing body, it could respond rapidly to emerging issues, and could provide the kind of frequent assessment statements desired by various agency heads. It would not need to duplicate existing assessment processes where they exist, but would build on and complement them, looking for instance at interactions that might be missed in more sectoral or single issue assessments. It could also be incorporated into other assessment processes, and be used to review the results and prepare conclusions where this is now done on an ad hoc basis. It could, for instance, provide an open mechanism to evaluate, draw conclusions from and build consensus for the Global Environmental Outlook modelling/scenario efforts. It could also give a high public profile to the UN system-wide Earthwatch and the whole environment assessment process, and its hearings and opinions would contribute to public education on the application of science to critical issues.

32. While the creation of such an Earthwatch assessment panel would require a significant commitment of resources, it would not be just another high-level advisory committee. It would organize and implement a practical mechanism for environment assessment that would not just collect information but deliver policy-oriented results. It could thus fill a critical gap between science and action at the international level.