WHAT IS A CLEARING HOUSE?

In its original meaning (19th Century), a Clearing House was an establishment where financial institutions adjusted claims for cheques and bills and settled mutual accounts with each other. The term has extended in meaning to refer to mechanisms of sharing information (not only financial) through an independent intermediary for mutual benefit of the concerned parties. Thus a clearing house to "facilitate technical and scientific cooperation" (in biodiversity) is a facility for the deposit and open exchange of information for mutual benefit on the state of biological resources, and related data collection, research, and technology. The key characteristics are :

• independence (of the clearing house), that is the operators of the clearing house should have no vested interest in

• controlling or subverting the flow of information; and

• mutual benefit, which usually means, in practical terms, that no detailed accounting is made of transactions; equality of contribution and benefit is assumed to occur over a long period.

The modern clearing house is usually aided by computer networking and information management technology, to allow for the exchanges to take place rapidly and efficiently. One implication is that there is usually a store of information which has been contributed to the clearing house. With network technology this store may be conceptual, that is, the information continues to physically reside with the contributing agencies or subsidiary clearing houses, who continue to exercise expert custodianship.

The purpose of a biodiversity clearing house is to alleviate the challenges identified in Section 1.2 in an efficient manner – for instance, providers of information or technology should only need to provide it once, and not have to respond to repetitious requests, and information seekers should have a single window source. In this way the goal of "facilitating technical and scientific cooperation" would be achieved.

HOW DOES A CLEARING HOUSE MECHANISM WORK?

A background paper to the "Informal Consultation on how to establish a Clearing House Mechanism" hosted by the interim secretariat to the CBD, Geneva, August 26-27 1994, identified four models for a clearing house mechanism:

• information retrieval and referral services;

• demand based broker services where, upon request, an intermediary assists service users with particular needs by providing advice, finding technical or financial partners or both;

• technology development services which distribute technology to partners for local adaptation and feed back; and

• Policy development services which provide assistance upon request.

This background paper concentrates on the first of these - the information service clearing house, although in practice any clearing house usually operates to some extent in more than one of these modalities. It should be noted however that there are more proactive elements to brokerage, technology development and policy development services which greatly add to the complexity of operation and cost of infrastructure, than the more passive information retrieval and referral style of clearing house.

A very important aspect of any form of clearing house is the need for meta-information (or meta-data), that is information which describes information (eg; its nature, quality, source, formats, and accessibility). An information service which mainly holds meta-data which refers the query to the source of the information is often called a referral system. Bibliographic and abstracting services are of this nature, and INFOTERRA is a well-known example of a referral system in the environmental field.

An information retrieval service allows the user to obtain the desired data directly without (apparent) referral to another service or agency. In practice, many services may operate in both modes - for instance providing direct retrieval of smaller datasets or documents, while referring the user to the source agency for large databases or non-digital records, such as maps and books.

The key to a successful information retrieval and referral service is good meta-data, organised in a systematic and controlled manner. The meta-data must in turn be supported by controlled vocabulary, thesauri, data dictionaries and the like, which aid in

meeting the user query specification by providing information on coded fields, keywords, synonyms (multilingual equivalents, if possible), broader and narrower related terms etc.

Thus, the basic components of an information service clearing house are:

- meta-database
- thesaurus of terms
- data dictionary
- information registry and quality control system
- referral system
- information retrieval and management system
- user query interface
- electronic network connections.

Tools to update and maintain all of the above also need to be in place.

SCOPE AND NATURE OF BIODIVERSITY INFORMATION

The range of potential information types varies widely, with biodiversity factors placed in the context of a range of other types of information necessary for the management and understanding of biodiversity. The following paragraphs, adapted from the United Nations Environment Programme (UNEP) Guidelines for Country Studies (1993), identify some of the types of information which might be important, particularly for the initial assessment and strategy development.

Biological
This is the primary focus of biodiversity conservation - the core data which includes the requirements for species, ecosystems and genetic resources, covering issues ranging from status and distribution of resources to functional relationships and the development of tools to support the science.

Physical
Information on physical factors such as climate, topography and hydrology allows biological data to be placed within a physical context, and also allows for the development of predictive models (as the distribution of many species and vegetation types can be predicted by a combination of physical characteristics). Physical factors can also have a significant effect on potential use of resources, and on management options.

Socio-economic
The use and abuse of biological resources is essentially a function of socio-economic factors. Important data might range from monitoring of forestry or fisheries practice, to the impact of farming methods, or the distribution of population centres and transport routes. Perhaps as significant is accessibility to resources, and the use that local peoples make of the available natural resources, which often forms an essential, but perhaps invisible part of the local economy.

Costs and benefits
In order for management of biodiversity to be efficient it is necessary to know the true value of biodiversity, and the costs and benefits of management options. This needs to cover questions such as the costs of managing protected area systems, the level of income derived from visitation, and the value of indirect benefits derived from protection of the watershed. Some of these values require methods of assessing the true benefits of biological diversity which are only just being developed, and wider dissemination of information on the methodologies will be required.

Threats
Identifying and monitoring both potential and actual threats to biological diversity is likely to be an important component of any data collection and management programme oriented towards improved management of biological resources. However, information collection programmes may need to go beyond the obvious physical causes and their effects to also identify and record the underlying causes, including the responses to human activities (which links threats to socio-economic factors).

Management
Conservation of biological diversity is about effective and sustainable resource management. To assess that management, information will not only be required on the biodiversity, its status and its distribution, but also on current and past management activities, especially on the use of biodiversity resources. For example, information is likely to be required on a range of characteristics of protectedareas and on effective management methods and technologies in a range of habitats, both protected and unprotected.

Sources and contacts
Information is also required on sources of information models, standards, and technology, and on appropriate agencies and experts who can be contacted. This may include bibliographic information on who has published what, where, basic information on names and addresses of appropriately qualified experts, sources of information on reliable and appropriate models, and metadatabases.

Interrelationships
The above paragraphs, although brief, begin to indicate the extent of the interrelationships between the information that might be required in order to study and manage biodiversity more effectively. It is essential that these interrelationships are kept clearly in mind when planning both information collection and management strategies, and information on methods of assessing and forecasting effects of interrelationships must form an important component of any information sharing mechanism.

All this points to a vast scope for the information to be collected and exchanged, and clearly some selection must be made. The following is a suggested list of the most relevant categories:

• information which provides a practical baseline for monitoring effectiveness of action

• key information identified by biodiversity managers as being important for decision making

• socio-economic value of the local and national biodiversity, and of protected areas

• functions and benefits of biodiversity, particularly service functions of ecosystems and protected areas

• policy and conservation programmes, and the legislative framework and other institution- related matters

• technology useful in monitoring, assessing and improving the sustainable utilisation of biological resources

• genetic resources, including medicinal plants, landraces and wild ancestors of domestic breeds and cultivars

• species that could serve as indicators of ecosystem health and flagship species habitats, the conservation of which protect the diversity of other species and habitats

• alien or exotic species, the spread of which could threaten indigenous biological diversity

• threats to biodiversity and biodiversity known to be threatened

• species and habitat distribution and its change with time

FORMS OF BIODIVERSITY INFORMATION

The information required occurs in a range of forms and media, including:

Numeric
Numeric data is derived directly from many types of survey ranging from counts of species in particular locations, to measurements of rainfall, tree growth or the length of a bird's primary feathers (which might be used in identification and taxonomic work). Numeric data can also be generated automatically from recording machines, or derived from analysis of remote sensed data.

Because of its nature, numeric data lends itself to computer aided manipulation and analysis, and the derivation of further datasets based on such analyses. For example, the absolute altitudinal range of a protected area can be derived from subtracting the lower altitude from the upper. It is also extensively used in modelling. For example, information on the temperature, rainfall and altitude of a particular site (all numeric data) can be used to predict the Holdridge life zone within which it lies.

It is possible to structure numeric data quite strictly, for instance into database tables, and to exercise quality checking on data entry against norms, and various types of statistical analysis may be applied.

Categoric
A very common form of biological information is classified or coded non-numeric data, and might include indications of factors such as soil type, land cover, forest type, species, protected area designation, and so on. The data is structured, usually through a thesaurus or data dictionary, and can be checked for allowed values, and the like. Statistical analysis is not often appropriate, but spatial distribution is often important.

Text
Text is an extremely common form of biodiversity information, including descriptions of protected areas, descriptions of species, descriptions of threats, ecosystem status reports, "State of the Environment" reports, legislation, regulation, strategies and plans.

By its nature, it is much less structured, is often subjective, and difficult to appropriately search and retrieve unless careful controls are placed on its compilation. However, with recent developments in text manipulation using computers, such as the use of hypertext links, the value of text description has increased considerably.

Text, when attached to a database of numeric and categoric information, can provide valuable extra description on the quality and sources of the data. For example, it can be used to check more thoroughly the identity of a species or expand upon an ecozone classification once a database has been used to identify it.

Spatial data
Biodiversity information is intrinsically spatial in nature, including the locations of sample points, mappable phenomena such as climate, topography, habitat, species range, point located and distributed threats, vegetation and land use. This information is often contained on paper maps, in digital remotely sensed imagery or in computer-based geographic information systems.

Other Media
Biodiversity information also includes non-digital images (photographs, drawings) of landscapes, specimens, and technology (instruments, methodological flow diagrams and the like). Future consideration should also be given to moving images (such as video tapes recording wildlife behaviour) and sound recordings.

SUMMARY

The information that a clearing house must deal with has the characteristics of being:

• multi-sectoral - including many sectors of environmental and natural resource data, as well as socio-economic and policy information;

• spatial - often over very large areas of the globe;

• diversified - containing statistical data, quantitative scientific measurement, numerical and categoric classified data, and descriptive/narrative information; and

• very voluminous - millions of documents, terrabytes of data.

The technical challenges to an effective clearing house mechanism include:

• effective management of very large spatial databases;

• abstraction and summarisation in a manner which facilitates decision making;

• harmonisation of information from differing measurement regimes;

• integration of data from different environmental and socio-economic sectors;

• maintaining quality control and metadata on the quality and origins of the information; and

• integration of textual information with conventional databases.