ANNEXES

ANNEX I - EXECUTIVE SUMMARY OF THE GTOS PLANNING GROUP REPORT

(The text below is from the June 1996 report of the ad hoc Scientific and Technical Planning Group for a Global Terrestrial Observing System. The Chair of the GTOS Planning Group was Dr David Norse of University College, London. The report was submitted to the GTOS Co-sponsors in December 1995. Bibliographic details are: GTOS Planning Group (1996). Global Terrestrial Observing System - GTOS. Turning a sound concept into a practical reality. pp xii + 114; 8 Annexes. London: UNEP, FAO, UNESCO, WMO, ICSU).

The objectives of the Planning Group's proposals are to:

• Make the case for the establishment of a Global Terrestrial Observing System: Why do we need a GTOS?

• Recommend its primary function: What will GTOS do?

• Present options for its operation: How will it be organized?

• Outline the main costs and benefits: How much cost, how much gain?

• Suggest what needs to be done: Where do we go from here?

WHY DO WE NEED A GTOS?

Terrestrial ecosystems are the foundation of social and economic well-being because they are the main source of food and other basic needs. They also play a vital role in the regulation of atmospheric, biogeochemical and hydrological processes. Yet we do not know how, where and over what time frame humankind is endangering terrestrial and freshwater ecosystems including coastal zones. We do not even fully understand the role of these ecosystems in global processes. In particular, we cannot answer five inter-related questions central to sustainability development:

• Food and renewable resources: can the land feed another 5-6 billion people?

• Fresh water: where, when and by how much will demand exceed supply so as to cause supra-national problems?

• Toxins: do they or will they cause major trans-boundary threats to human and environmental health and the capacity of ecosystems to detoxify them, and if so where and when?

• Biological diversity: where and what type of biological resources are threatened with loss, and where will these losses irreversibly damage ecosystem function or socio-economic progress?

• Terrestrial ecosystems: where, when and how much will they change in response to global atmospheric, climate and land-use changes, and how will this impair their capacity to support life?

Such questions have led nations to sign the climate change, biodiversity and desertification conventions, to adopt Agenda 21 and other actions relating to deforestation and environmental protection in general. Many of these conventions and actions require better terrestrial observation data, but the international community has yet to establish the means of obtaining them.

We lack spatially and temporally comprehensive data on the physical environment, on terrestrial ecosystem processes, and on the socio-economic driving forces that are changing them. There is no global mechanism for the collection of compatible data, and so there is a critical gap in the current observing systems for climate (GCOS) and the oceans (GOOS). Consequently, we cannot determine whether major policy changes are needed now and at high economic and social costs, or if both the impacts and the correction processes are longer term.

Without GTOS, we will continue to see major investment in non-compatible systems, a consequent data or data integration gap, and a lack of support for international conventions. With GTOS we will be better placed to identify pressure points and determine how best to use Earth's resources to achieve sustainable development.

WHAT WILL GTOS DO?

Given the above questions, the central mission of GTOS should be to provide the data needed to detect, quantify, locate and give early warning of changes (especially reductions) in the national or global capacity of terrestrial ecosystems to support sustainable development and improvements in human welfare. It should also help advance our understanding of such changes.

These objectives should be accomplished through an integrated and equitable partnership of data providers and users that meets both the short-term needs of national governments and the longer-term needs of the global change research community, GTOS's focus should be on five key development issues of global concern:

• Land use change, land degradation and the sustainability of managed ecosystems

• Water resources management

• Pollution and toxicity

• Loss of biodiversity

• Climate change.

GTOS should be directed at specific needs and at overcoming the deficiencies in the existing global and regional observational systems. It should not be directed at data collection for its own sake, and research should not be a major function, though it should identify research needed to improve observational and information systems. But GTOS should support research programmes and collaborate with IGBP, DIVERSITAS and others in the assembly of appropriate data sets.

To make a unique contribution to our ability to manage the planet wisely, GTOS must:

• Be global in scope, meaning both that its coverage is comprehensive (but regionally balanced and resolved) and that it should address phenomena that are global in their nature or impact.

• Provide continuity of information collection over the long-term - periods from years to decades which are consistent with the rate at which global processes occur in order to detect trends sensitively and in a timely fashion.

• Be an integrated system in which the separate pieces of information add to each other's value. For example, GTOS data must not only detect and describe changes, but allow them to be understood and predicted.

HOW WILL IT BE ORGANIZED

GTOS should operate on the basis of a partnership of partnerships formed largely from existing sites and networks (plus others like WHYCOS which are in the process of development), and on present and planned remote sensing systems. Implementation should be essentially bottom-up, with GTOS providing the framework within which the output from the spaced-based Earth Observing Systems and the existing databases such as GEMS/Water and the Global Runoff Data Centre (GRDC) can be integrated with in situ observations. Actions should be both direct and catalytic. The core of the proposed system is a hierarchical sampling strategy, with four tiers of decreasing complexity and frequency of in situ observations and a fifth tier to provide global coverage largely through satellite remote sensing. At one extreme, detailed data is collected almost continuously at a few large sites and, at the other, a large number of small systematically located sites are sampled at intervals of five years.

The establishment of such a hierarchy and the data management and exchange system to support it should be undertaken by a Central Coordinating Unit (CCU) with some international secretariat functions. The CCU should be linked to regional and national bodies of variable form and structure, since they should evolve in response to user and provider initiatives rather than being part of a pre-set structure. The CCU should have two guidance mechanisms - a Steering Committee for strategic considerations and a Technical Advisory Group - plus some 40-50 corresponding members who would contribute to and/or comment in writing on proposals for the implementation of GTOS. In addition, ad-hoc or permanent supporting bodies should be established to guide the development of operational plans for particular functional or thematic components of the programme. As far as possible, these bodies should be joint activities with GCOS and GOOS.

GUIDING PRINCIPLES

Guiding principles for GTOS's data management system should be common or compatible with those for GCOS and GOOS. The data management system should be constructed, as far as possible, using off-the-shelf application tools and existing or planned communication systems. It should be sufficiently flexible to incorporate or link to data sets originating outside GTOS, since GTOS may not hold much of the data but provide an access mechanism for dispersed data sets. Data links between the centre, the regions and other data centres would be primarily electronic with magnetic and optical disks as a parallel alternate system, especially during the transitional period.

HOW MUCH COST, HOW MUCH GAIN?

A substantial proportion of the required infrastructure is already in place and funded, so the incremental costs of the proposals will be low compared for a totally new system. The operating costs of existing terrestrial observing systems are over US$300 million. GTOS, on the other hand, will initially cost less than one million dollars per year possibly rising to about US$3.4 million after five years.

The costs of launching GTOS should be phased over five or ten years, and built up in a modular fashion. This would permit multiple financial mechanisms to operate, with individual donors supporting those modules that are consistent with their issues or regional priorities. In the early stages, average annual operating costs could be some US$700-850k for the CCU, guidance bodies, and initial actions to link existing monitoring sites and networks. During the first five years the operating costs of strengthening and extending these activities, and some capital costs would rise to about US$5 million. In the medium-term total costs would rise to around US$12 million if the resources can be found to extend regional activities, improve existing sites and to fill gaps in the spatial coverage of the sites.

The proposed activities would improve the returns from major investments in independent in situ observation systems by providing complimentary regional or global data, and in earth observation satellites and remote sensing devices by providing comprehensive ground truthing. The drawing together of existing but disparate databases, sites and networks into a common framework with the standardization or harmonization of measurements and terminology would increase substantially the usage and value of such data and information. The GTOS activities would support global change research programmes by contributing to the refinement, calibration and validation of the GCM, ecosystem and carbon cycle models.

The provision of globally comprehensive and timely data on anthropogenic impacts on terrestrial ecosystems will help UN agencies - and the secretariats of the Climate, Biodiversity, Desertification, Ozone and other conventions and treaties - to fulfil their mandates. It will also help them and multilateral donors advise their member governments on priorities for sustainable development.

National benefits include support to planning, natural resource management and environmental agencies, opportunities for staff training, promotion of contacts and interactions between scientists of participating nations, and greater access to new technology for environmental assessment. Strengthening national terrestrial ecosystem monitoring should contribute to socio-economic development by helping to identify opportunities for - and undesirable consequences of - developments project at all scales. GTOS could help countries add the global dimension to national environmental strategy formulation, obtain data for national global research programmes, develop better policy planning tools and meet reporting obligations under the post Rio conventions.

WHERE DO WE GO FROM HERE?

The Planning Group recommends the progressive implementation of GTOS, starting in 1996 with a five-year programme. Priority in 1996 should be given to:

• Finding an institutional home for GTOS

• Establishing the institutional framework and some of the operating mechanisms.

• Completing a dialogue with potential partners already involved in global data management and harmonization or operating global, regional, national or sectoral systems and networks or sites.

• Setting up a pilot framework drawing together a sub-set of existing observing systems.

• Launching data management and harmonization procedures.

• Developing the detailed operational plans for later stages of implementation.

• Obtaining the supplementary funding for future development of GTOS from multi- and bi-lateral donors and other institutions.

Over the following four years, a start should be made on upgrading existing sites and filling the most urgent of the gaps in the geographic, biome or crop coverage of natural and managed terrestrial ecosystems, and promoting regional bodies.

ANNEX II - SPONSORS GROUP FOR THE GLOBAL OBSERVING SYSTEMS

1. There are three Global Observing Systems: the Global Climate Observing System (GCOS); the Global Ocean Observing System (GOOS) and the Global Terrestrial Observing System (GTOS). Together the three are sponsored by six international organizations: the United Nations Environment Programme (UNEP); the Food and Agriculture Organization of the United Nations (FAO); the United Nations Educational, Scientific and Cultural Organization (UNESCO) and its Intergovernmental Oceanographic Commission (IOC); the World Meteorological Organization (WMO); and the International Council of Scientific Unions (ICSU). UNEP and ICSU sponsor all three Global Observing Systems. Each of the Global Observing Systems has found an institutional home in a different host agency (GCOS in WMO; GOOS in IOC; GTOS in FAO).

2. The three observing systems were founded independently and at different times, and so are at different stages of development with GCOS being the most advanced and GTOS being the least developed. Since they share many approaches, interfaces and common problems, the Co-sponsors felt that development of the Global Observing Systems should be guided by a common strategic framework and close working relationship. Consequently, the sponsoring organizations formed the Sponsors Group in late 1996. UNEP, working through its Earthwatch programme, assumed responsibility for providing secretariat support for the Sponsors Group. The first meeting of the Sponsors Group was held in January 1997. The group meets annually but additional meetings are held by mutual agreement should they be needed.

3. The Sponsors Group provides an important opportunity for the Global Observing Systems to be considered by their sponsoring organizations in a concerted fashion. The Group agreed that the secretariats and sponsors should share information with all of the members of the Sponsors Group regardless of whether they were formally a sponsor of the system in question. It has discussed at length many of the problems faced by the Global Observing Systems such as national and international visibility, financial support, and participation. Unfortunately, progress in these areas has so far been limited. Thus the sponsoring organizations have not yet been able to take proper advantage of the forum provided by the meetings of the Sponsors Group to further the three Global Observing Systems and bring them closer together in form and function. They have, however, identified a GOS Panel on Socio-economic Benefits as a priority need for GOS since a clear definition of the societal benefits from the GOSs will help to build and maintain support for all three GOSs. They have also examined the roles of the governing bodies of each sponsor since the support of governments is essential for maintaining the observing systems in the work programmes and budgets of the sponsoring organizations.

4. The Committee on Earth Observation Satellites (CEOS) and the International Group of Funding Agencies for global change research (IGFA) have called for an Integrated Global Observing Strategy (IGOS) as a joint product of all the agencies involved in the collection and analysis of space-based and in situ ground data. This need has been recognized by the Sponsors Group with the result that UNEP has prepared a draft IGOS strategy as a proposed umbrella for the three observing systems and any other international observing activities. CEOS and IGFA are also preparing a draft IGOS strategy. These will be considered at future meetings of the Sponsors Group.

ANNEX III - DATA SCALES

Spatial and temporal scales

A major problem for all terrestrial observing system is the spatial and temporal scales at which data are gathered and the relevance of data obtained at one scale to other data obtained at different scales. Spatial and temporal scaling issues are of the utmost importance to GTOS and the utilization of its data and products.

Spatial scales

Land observations are commonly made at a wide range of spatial scales from broad satellite coverage to site specific measurements on the ground. Thus spatial data may represent areas ranging from one metre or less to hundreds of kilometres. Terrain is inherently heterogeneous but the degree of apparent heterogeneity is a function of the spatial scale at which data gathering observations are made. The finer the spatial scale used, the greater the heterogeneity that can be recognized. Some global change questions can be dealt with adequately by using fairly coarse spatial scales, but others, such as those involving wetlands, or burning, require much finer resolutions to resolve adequately. The in situ site measurements necessary to understand and model processes are made at finer resolutions still. No single scale will be adequate for all GTOS activities and it is likely that studies will require extrapolation across several spatial scales. Integrated models that can be used at all scales are unlikely to be developed so that, for example, in global analyses it is more probable that layered models will be used to provide inputs into other models in a hierarchical approach.

Temporal scales

Temporal scale difficulties are even more difficult to resolve since it is usually not easy to extend information on biological functions such as photosynthetic rate, and uptake of water by plants, from daily to longer time scales in terms of years or decades. To do this satisfactorily requires large scale information across gradients on a wide range of phenomena (eg vegetation physiognomy, species composition, and ecosystem changes).

Upscaling

Site specific ground data, if sufficiently frequent spatially, can be clumped to develop useful but coarser resolution wider area data planes, and these in turn can be further clumped to form even more broad scale but still useful data planes of greater areal coverage. The reverse, however, is not true and it is not possible to derive meaningful site specific fine resolution data from broad scale data such as those normally available from resource assessment satellites. Thus it is possible to go from fine scale to broad scale in a meaningful way but not usually from broad scale to fine scale. In practical terms this means that useful fine scale data specific to small ground sites cannot normally be derived from broad scale satellite coverage unless the constituent pixels of that coverage are themselves very fine. However, the relationship between information content and spatial resolution of satellite data is usually non-linear and there are also resolution windows that are associated with any given terrain type. Also terrain heterogeneity is normally seasonally dependent which affects the scale at which it is useful to record data. Consequently, many workers have found that a resolution of 500m is a useful working compromise as a lower limit for deriving reliable site data from satellite coverage. GTOS will have to review upscaling in relation to its programmes and outputs.

Extrapolation of site specific data

Spatial scale changes are of importance to national planners and resource managers because they need to know the relevance of site or area specific data to wider geographical areas beyond the borders of the observation sites, and whether data obtained in one area can be used to develop or manage another of similar environment and ecology. This becomes even more important when site specific data originate beyond national borders and are supplied through an international system such as GTOS. This is an important aspect of the national level GTOS programme because many potential users of the system, such as national governments and convention secretariats, are concerned with the development and management of specific sites and localities which are themselves at the moment outside GTOS system coverage and are likely to remain outside in the future.

ANNEX IV - ACRONYMS