TROPICAL FORESTRY SERVICES Ltd.
John Palmer and Mary Marshall
3 Beechcroft Road, Summertown, Oxford OX2 7AY, U.K.
telephone: [UK +44] Oxford (0 1865) 554 004; fax: (0 1865) 311 505
E-mail tropical.forestry@rmplc.co.uk
ÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄ
FINAL DRAFT
Invited paper for Topic 1 - The basis for certification
Part A - Ecological
for
Economic, Social and Political Issues
in Certification of Forest Management
UBC-UPM Conference on Certification
12-16 May 1996, Kuala Lumpur (Putrajaya, Kajang), Malaysia
J. R. Palmer, D. Curtin and C. Graham
final draft, updated 28 July 1996
Sponsored by the Faculty of Forestry at the University of British Columbia
and the Universiti Pertanian Malaysia, Serdang
ÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄ
Registered in England no.2888992 VAT registration no.630 6574 46
Registered office : St.John's House, 5 South Parade, Summertown, Oxford OX2 7JL, U.K.
Table of Contents
Abstract
Introduction - the case of BSE
Monitoring to satisfy whose interests ?
Audiences for forest monitoring
Commercial processors and retailers of forest products
Consumer and environmentalist organizations
The general public
Indigenous and traditional peoples
Forest operators
Other stakeholders
Spatial and temporal scales of stakeholder interests
Frequency of monitoring
Anticipated detail in monitoring
Assurance of credibility
Communicating the results of monitoring
What to monitor for the different stakeholders ?
Monitoring the forest for national/international statistical reporting
Monitoring for the forest owner and manager
Components of the forest management system
Norms for indicators
Characteristics and examples of indicators
Monitoring with remote sensors
Snapshots versus trends
Costs of monitoring
Table of Contents (continued)
Priority setting or weighting among criteria and indicators
Problem of non-equivalence
Hazard analysis by critical control points
Analysis of criteria and/or indicators in relation to the certification decision
Research needs for improved monitoring
References
Annex 1 - Some characteristics and selection criteria for indicators
Annex 2 - Tracing of supplies from the global market by Timbmet Ltd., U.K.
UBC-UPM Conference on Certification
12-16 May 1996, Kuala Lumpur (Putrajaya, Kajang), Malaysia
Sponsored by the Faculty of Forestry at the University of British Columbia
and the Universiti Pertanian Malaysia, Serdang
Invited paper for Topic 1 - The basis for certification
Part A - Ecological
for
Economic, Social and Political Issues
in Certification of Forest Management
Monitoring forest practices
---------------------------
John R Palmer
Tropical Forest Services Ltd., Oxford OX2 7AY, U.K.
Daniel Curtin
Independent consultant, Oxford, U.K.
Catherine Graham
Group Environmental Coordinator, Timbmet Ltd., Oxford OX2 9PP, U.K.
Abstract
Consumer-responsive or treaty/convention-obligated certification requires mutual reciprocity of recognition, which in turn requires equivalence in the indicators whose values contribute to the certification decision. Reciprocity requires also equivalence in the audit procedures for verification of the operation of forest management systems; well established principles of accreditation facilitate equivalence in audits. The forest stewardship standards which are currently in operation or under development have excessive numbers of criteria and indicators of the quality of forest management. Evaluation indicators should be separated from forest-specific checks on the implementation of management prescriptions. Indicators should be as globally relevant and as comparable as possible, while being meaningful and relevant at the level of the local forest management unit. Procedures for assimilating indicator values to derive a Yes/No forest certification decision need improvement to reduce the influence of subjective judgements and to increase transparency of the decision-making process. An incremental approach to globally relevant indicators should first aim to satisfy consumer expectations.
Note
In this paper, Standards (with an initial capital S) and FSS (forest stewardship standards) mean a designed set of principles, criteria, indicators and norms for assessment of the quality of forest management, or at least some combination of this hierarchical set. 'Good forest stewardship' is preferred in this paper to 'sustainable forest management', because the latter can only be assessed retrospectively
This paper was prepared during an unprecedented agricultural crisis in Britain. A fatal neurological disorder, bovine spongiform encephalopathy (BSE), has been caused by feeding cattle with rations compounded from a huge variety of organic substances, including cardboard and the carcases of sheep which died of the scrapie brain disease, as well as carcases of cattle which died with rather similar symptoms. There are fears that a small proportion of human consumers of beef and beef products from animals affected by BSE may contract Creutzfeldt-Jakob Disease (CJD), a somewhat similar and fatal affliction of the human brain. The sale and use of cattle feed containing ruminant protein was banned after July 1988. However, the consumption by cattle of old stocks of that compound feed has continued for years afterwards, because the risks involved were not sufficiently appreciated.
The nature of BSE is complex and difficult to research. To have applied the "precautionary principle" when BSE was first associated with compounded rations would have caused an outcry by farmers because of the increased cost of feeding cattle. The chain of cause and effect was too weakly understood to make a convincing scientific case for a ban on such rations in the face of financial arguments for no ban.
While research was being conducted on BSE, links between veterinary and human health monitoring were too weak to allow rapid detection of possible transmission to humans. When that possibility was admitted, the consequences for public health caused diverse reactions. On the one hand, governments have initiated a variety of controls, and the European Commission imposed a world-wide ban on the sale of British beef. The cost of loss of business, including international trade, runs into astronomical figures.
The reactions of consumers in Europe have been interestingly different, depending on distinct views of risk. The announcement of a possible link between BSE and CJD in Britain was followed by a very sharp fall in sales of beef and readily identifiable beef products. Supermarkets which halved the retail price of beef reported a rapid recovery in sales. Supermarkets which have short supply chains and tight control over the production methods of farmers have been able to retain a high proportion of their normal sales. Shops which can advertise beef as coming from healthy herds reared on a diet of natural forage and cereal have also lost only a small proportion of their normal sales. There does not seem to have been any switch in consumer purchasing behaviour, from cheap beef derived from animals reared on compound feeds to more expensive beef reared on natural grassland and forage. On the continent of Europe, sales of beef have fallen by about 50 per cent in Germany and 40 per cent in Italy, even though the reported incidence of BSE has been extremely small (a total of 129 animals in Switzerland, 67 in France, and lesser numbers in other countries; this is very much a "British" disease). Smaller declines in dairy products and beef sales have been noted elsewhere in Europe.
One might expect that organic cattle farmers would stand to gain a large market share from the conventional producers of dairy and beef products. However, terms of trade are still significantly stacked against the organic farmers. In Britain, conventional producers using compound feeds for their animals receive substantial government grants, much larger than for those operating to organic standards. Organic meat producers may obtain a "green premium" of perhaps 10 per cent on the farm gate price but the major difference is in the retail price. Retailers explain the large difference in premium between farmgate price and retail price by the small volumes of organic farm produce and consequent lack of economies of scale, extra transport costs, and higher wastage due to shorter shelf life and a higher proportion of visually unacceptable produce. The 10 percent "green premium" is a poor stimulant to most farmers, compared with the relatively generous subsidies available to conventional farmers through the operation of the European Common Agricultural Policy, There is no evidence that British supermarkets, which in effect determine the retail prices for food, will improve the premium which they offer to organic producers.
The case of BSE and organic farming is cited here because it emphasises the paradoxes of evaluations of, and responses to, risk at the family and government levels. There are weaker but probably instructive parallels in sustainable forest management. Because so little is still known about the long-term effects of forest management regimes on ecosystem functions, ideally the more intensive regimes should be more intensively monitored. But monitoring is expensive, the relative value of forest products is low compared with cattle, and risks to ecosystems are naturally perceived as less critical than risks to human health. Choice of what to monitor in the forest is thus difficult. The precautionary principle implies that we should do more monitoring when uncertainty is high, while forest operators argue that conservative and monitoring should not have negative impacts on their costs; exactly as the British cattle farmers argued.
The BSE case introduces this background paper because it illustrates the different responses of veterinary and medical scientists, farmers, food retailers, politicians and the general public to a superficially simple epidemic in cattle with a single identified cause and a superficially simple solution. As knowledge increases about the subtleties of the disease and its actual and potential ramifications, so the implicit needs for monitoring, and analysis of returns from monitoring, increase enormously. As human health may be at risk, there is not currently much scrutiny of the costs of the control measures and long-term monitoring which will be required. It would be unreasonable to expect the same tolerant attitude to rocketing costs when forest practices are being monitored.
The BSE case is also relevant to this UBC-UPM conference because it shows the hazards of delaying action. There is considerable evidence that controls on cattle feeding practices were postponed because scientists could not prove a connection between neurological disorders in cattle and in humans. Trying to "get all of the science right" before making policy decisions was, in this case, a fatal approach. Moreover, in relation to forestry, it is an unhistorical approach. If Dietrich Brandis had not taken an incremental approach in India in the 1850s, making conservative extrapolations from forest management practices in Europe, who knows how long formal management of tropical forests would have been delayed.
We should be aiming to transform forest practices from CATNAP (currently available technology narrowly avoiding prosecution) to BATNEEC (best available technology not entailing excessive costs); we are not inventing these terms, they are well known in quality assurance systems.
Certification is a market process, not an academic exercise. This background paper is concerned principally with independent, third-party audit and voluntary certification of the quality of forest management at the level of the local forest management unit (LFMU). The paper does not deal with inter-governmental reporting schemes such as the Helsinki and Montreal Processes or the FAO Global Forest Assessment, or with whole-country certification. This last idea is unacceptable to consumer and environmentalist groups, precisely because of notorious failures of national forest services in the past and the public perception that they cannot be impartial, disinterested evaluators of what they should be regulating anyway. Moreover, as quality of forest management is not (yet) linked directly to public health & safety, there is no obvious reason why government should intervene in market-mediated decisions (contrast GATT Uruguay Round's Agreement on Sanitary and Phytosanitary Measures - see paper by Markku Simula in this volume). The background paper is also not concerned with high-specification FSS for the "boutique" market.
From the 1980s, at least, surveys of public opinion have indicated an expressed willingness to pay substantially above the ruling price for products which come from forests under environmentally-sensitive management (MORI 1990, Gerstman & Meyers 1991, Botting 1994). If the whole of that extra value were to be returned intact to the forest producers, the amount would be ample to pay for a vast improvement in the quality of forest management (OFI 1991). Unfortunately, the morally sensitive response to pollsters is not confirmed by surveys of actual purchasing behaviour. Buyers for home improvement/DIY (Do-It-Yourself) sheds would pay a small premium if they could be sure of the forest management quality and the security of the chain of custody. In contrast, the policy of builders' merchants was to source from the cheapest supplier for a given quality of timber (Botting 1994).
We concentrate on the commercial users of forest products (such as construction companies) and on retailers, because information about the views of primary processors is sparse. The sawmillers who responded to the unpublished Soil Association/Trees for People study on chains of custody in the Black Forest of Germany in 1995, commissioned by the European Commission, tended to adopt a wait-and-see attitude. The enterprises which dealt direct with consumers were more positive about certification but lacked knowledge of what they might monitor in the supply forests.
The usual explanation for the more positive attitude is that the home improvement/DIY sheds are closer, and must be more responsive, to the demands of the final consumers, while the builders' merchants are often small and traditional companies, and separated by one or more links in the marketing chain from the final consumer. The home improvement/DIY businesses tend to be relatively new companies, large, image conscious, and increasingly having public policies of commitment to ethical business. These policies or ethical charters often contain commitments against purchases from suppliers which are ecologically and socially irresponsible. Negative factors mentioned in the ethical charters of major U.K. companies include: use of child labour or forced prison labour; discrimination in recruitment of workers against women and local people; lack of respect for the claims and rights of indigenous and traditional peoples; production processes which are environmentally insensitive, polluting, or wasteful of raw materials or energy. Some companies also subscribe to the principles of fair trade: promotion of sustainable development by helping supplier enterprises in developing countries to be more successful; educational campaigns to explain to consumers how they can make a difference by selective purchases; and promotion of more equitable terms of trade between suppliers and buyers.
Influence is brought to bear on the commercial companies by publicity campaigns mounted by consumer and environmentalist groups, including contact with and through members of the Boards of Directors and shareholders. Positive signals can be sent by selective purchase of the products of ethical companies and purchase of their stocks and shares. Negative signals can be sent by unfavourable publicity, boycotts and sale of shares. Businesses subscribing to ethical charters are particularly influenced by the opinions of consumer and environmental journalists working for the major broadsheet newspapers.
Companies which take an ethical stance involve themselves necessarily in substantial extra commitments to monitor both their internal procedures and processes, as well as those of their suppliers. Most companies promoting an ethical charter state that it is simply a formalization of their obligations under the implicit Social Contract between the private sector and the general public. They acknowledge that they hope it will be good for business, mainly in terms of market share, much less in terms of "green premium". The ever-increasing volume of health and safety regulation in the European Union, and the monitoring obligations which that legislation implies, lessens the additional voluntary burden implicit in an ethical business stance.
Nevertheless, private sector companies such as those in the WWF-UK's 1995 Plus Group and Timbmet have found it no easy task to apply ethical principles to their practices and suppliers of forest products. As a first step, the companies have tried to identify the exact production locations of the forest products which they use or sell. Annex 2 describes the timber tracing process developed by Timbmet, which is Europe's largest importer of hardwood. Timbmet draws from a continuously fluctuating pool of about 200 direct suppliers, which fan out to a potentially enormous number of primary sources.
Tracing the origin of the forest product is clearly an important stage in being able to respond to the concerns of consumer and environmentalist individuals and organizations. That increasing numbers of major commercial enterprises do respond to public concerns is shown by the coincidence between the demands of consumer and environmentalist pressure groups and the companies' ethical charters. Tracing is at least theoretically possible for any purchaser, through the chains of purchase orders and invoices, and is within the administrative expertise of most companies. Responding to other concerns involves local and specialist knowledge which is normally not available within the purchasing enterprise.
In order to verify the technical assurances provided by suppliers, purchasers traditionally use the services of third-party certification bodies. "Third-party" means that the certification body is not associated directly with either the supplier ("first-party") or the purchaser ("second-party"), and does not have a stake, or stand to benefit financially, from either a positive or a negative result of the audit. The particular reason why purchasers do not turn to government agencies for verification of the suppliers's assurances about forest products is because most governments have a stake in the outcome. Nearly all governments have some kind of legislation governing forest harvesting. Most governments have legislation covering many other aspects of forest land use and production systems. The legislation is often acknowledged to be inadequate and to be only partly implemented, but citation of specific examples is often embarrassing to the responsible government agencies. Consumer and environmentalist organizations thus do not trust the impartiality of government audits, and hence the interest in third-party services.
Paradoxically, the need for public confidence in third-party certification bodies brings in governments at a different level. Accreditation of certification bodies may be conducted by an authority established by mutual consent between certification bodies and other stakeholders, or by governments themselves. More usually, governments set the frameworks within which accreditation authorities or agencies must operate.
The similarity of the demands or expectations of consumer and environmentalist organizations with the policy statements in the ethical charters of commercial companies is an indication of the influence which the pressure groups can exert. One can also interpret the similarity as the commercial companies allowing the pressure groups to do the companies' thinking for them, especially in areas where the commercial companies naturally lack expertise. Whether the thinking is adopted or not is largely determined by the cost of implementation. Demands which fit easily with better business practices are obviously more likely to be adopted than those which require much extra effort and cost. The WWF-UK 1995 Plus Group exemplifies the success of a gradualist approach to the promotion of certification, fostered by an environmentalist group. This approach merged with changes to business practice in response to awareness of the commercial benefits of positively addressing consumer interests. The 62 member companies of this group, representing some 20 per cent (US$ 3 billion annually) of the U.K. trade in forest products, have agreed to purchase wood and wood products after December 1999 only from forests which are certified to be managed in accordance with the principles and criteria of the Forest Stewardship Council.
The non-government (NGO) consumer and environmentalist organizations range from the most extreme preservationist to the liberal near-centrist. Action-minded organizations such as Greenpeace capture media attention but are often too extreme to attract a supportive base in either commerce or science. The broader base and greater knowledge of more mainstream groups such as the Audubon Society, Friends of the Earth, Sierra Club and World Wide Fund for Nature (World Wildlife Fund in the USA) allow easier interaction with government and commercial circles, because they are perceived to adopt a less threatening approach.
Those involved directly in forest management may object to the simplification of particular concerns to a set of simple slogans or sound bites. However, experience in many fields shows that such simplification is necessary to overcome natural inertia to change, especially when real needs are comprised of factors which are not easily grasped by non-specialists. The commitments in ethical charters are single-phrase or single-sentence distillates of often complex issues, condensed to essentials which can be grasped by shareholders, company directors and politicians. One strength of successful consumer and environmentalist organizations is their ability to blend professional understanding of subject matter with an appreciation of what and how much the non-specialist can appreciate and support.
The importance of consumer and environmental journalists in influencing the behaviour of commercial enterprises has been noted above. The primary contacts for the journalists are usually the consumer and environmental NGOs, not those engaged in forestry research or forest operations. Public understanding of forest monitoring procedures, and the results of monitoring, thus tends to be mediated through two layers of communication. Ensuring that the NGOs understand the choice of monitoring procedures, and the limitations on interpretation of the results of monitoring, is a major task for forest operators.
NGOs have to be careful to maintain the interest of their subscribers by constant repetition of threats which (only) they, the NGOs, can counteract. However, the threat must not appear so remote or generalized that the supporter feels emotionally detached. Although a subscribing supporter of a NGO is probably better informed than the general public, public understanding of forest problems appears to be quite limited. Casual interviews by U.K. radio and television reporters with the "person-in-the-street" and with politicians about forest problems suggest that the following social and environmental aspects are important:
social -
- the claims and rights of indigenous groups should be respected, and their traditional ways of life should be sustainable;
- employees in forest operations should enjoy the same legal rights and protection for employment, health and safety as in any urban industry;
environmental -
- there should be no, or minimal, use of persistent pesticides and other artificial organic and inorganic chemicals, such as fertilizers;
- water flows and water quality should be "tap water clear";
- no visible soil erosion or sedimentation;
- no burning, or only traditional burning in small patches;
- forest landscapes should appear as continuous cover or parkland from a distance, but not so thick that the casual walker might lose sight of his/her car; freedom to roam, and an abundance of gently graded, well marked and signposted paths;
- stands should be heavily thinned or naturally open, and regeneration should be by single-tree or patch clearings, not by clear felling;
- there should be an abundance of "charismatic megafauna", brightly coloured and easy to see (but not dangerous) wildlife; biodiversity should be conserved or enhanced;
- in peri-urban forests and tourist areas, information centres with simple literature and signboards should be complemented by knowledgeable ranger staff to interpret the forest.
Very obviously, this is an urban-centric, non-industrial view of the forest, derived largely from the television and cinema screen and to be expected in an increasingly urbanised world. Some of the concerns and preferences are contradictory. Most of the preferences are shared by professional foresters, although they would be expressed in different terms. Some of the preferences, such as blanket opposition to clear felling and burning, are based on lack of knowledge about the variety of regeneration processes in the natural state.
It will be evident from the list above that interest is mostly in the state of the forest, as reflecting the outcomes of forest practices, rather than in the forest practices themselves. This is generally true in all monitoring systems and for all classes of stakeholder. It is, however, only partly true if the operation of an environmental management system (EMS) is being audited, when the process may be as important as the outcome. The emphasis in EMS on policies and procedures permits some judgement about the commitment of a forest operator to quality of forest management in the future. This aspect of good forest stewardship cannot be checked if the monitoring system is concerned only with the current outcomes of forest management practices in the past.
The third numerically-major group of stakeholders (i.e., those with a legitimate interest or stake in the outcome of forest management activities) are the indigenous and traditional peoples. These peoples have more or less strong spiritual attachments to the forests in their home ranges, as well as more or less strong dependence on forest goods and services for subsistence and income. Much attention has been devoted by anthropologists and social scientists to identifying and categorising the value systems, needs and expectation of the often marginalized indigenous and traditional peoples. By far the prominent issue in common between groups is the security of inter-generational access rights to forest resources. Stronger dependence on the forest is associated with greater need for legal protection against the dilution of resource rights and against the transfer of these rights to other communities. In general, measures which favour the conservation of deep-forest species of plants and animals favour also the indigenous and traditional peoples.
Although numerically strong in global total, the indigenous and traditional peoples are often politically weak and socially marginalized. Pierce Colfer (1995) and Pierce Colfer et al. (1995) developed a six-dimensional framework within which forest dependency and influence over the forest could be assessed. However, the major influence exerted by these stakeholders is the de facto power to clear the forest for farming, either temporarily for rotational agriculture or permanently for settled cropping.
Owners and managers of forest enterprises, and their contractors, do not comprise a homogeneous group. Some larger operators have extensive series of long-term sample plots to monitor tree growth and forest yield, and may also monitor water quality and population dynamics of wildlife. Such operators may provide encouragement and logistic support to observational and experimental research on the effects of different forest management regimes.
At the other end of the scale, there are more or less legally approved operators whose interests are simply in mining of the forest resource at least possible cost. They will often have good reasons for opposing any form of monitoring.
Other forest stakeholders are, numerically, much less important than the five groups described above (commercial entities at the consumer end of the processing and marketing chain, consumer and environmental groups, the general public, indigenous and traditional peoples, and forest operators or producers). The other stakeholders include policy makers at national and international levels; legally potent government services; hydroelectric power managers; and environmental tourists. The stakeholder groups are by no means discrete. A person may be a government forester during the working week, an environmental activist at the weekend and an eco-tourist on holiday.
The spatial and temporal dimensions of interests in the forest obviously vary greatly between the stakeholder groups. Indigenous and traditional groups are concerned about the inter-generational durability of resource rights over forests in their home ranges. Loggers and other harvesters think about the concessions which they rent from the legal forest owners, often for short periods and with scant security of tenure. Commercial processors of forest products may take the whole world as their supply base, seeking the cheapest source for a given technical specification; regular and long-term bulk supplies of reliable quality may be more important than short supply chains. Government forest services have to take a long-term and national view in applying an increasing volume of complex legislation, some of which flows from international agreements such as the Framework Convention on Biological Diversity. Other government departments may take a short-term view of forests as land banks to relieve the pressure of excess human populations, as dumps for toxic wastes, and as potential reservoirs for hydropower schemes. Consumer and environmentalist organizations will tend to take long-term and geographically broad approaches to forest management but will often focus on specific short-term issues and particular localities for campaigning reasons.
The brief review of forest stakeholders and their different interests indicates the great variety in the amount of and kind of monitoring of forest change which would be required to respond adequately to each category. During the colonial period, British foresters and their national colleagues spent much effort in devising a set of standard reporting forms (BEFC 1948). These forms were used in annual state and national reports and in the approximately quinquennial Commonwealth Forestry Conferences. The forms summarised changes in areas by forest types and administrative divisions, concession areas granted and operated, progress on management planning, annual outturn/exports/imports by main types of products, information on forest revenue and expenditure. Other forms gave details of quantities of and taxes levied on marketed non-wood forest products, and of reported offences against forest regulations, forest fire occurrences, etc. The national returns were compiled through a simple hierarchical administrative system devised for the Indian Forest Service in the 1800s but having its origins in Roman times.
Area information was derived from surveys on foot until aerial photographs were introduced in the 1920s. Monitoring of forest quality was not attempted because of the difficulty of devising standard scales and the paucity of baseline information except in experimental areas. In spite of great improvements over the last several decades in the techniques for monitoring forests, publication of regular annual reports by national forest services has become a rarity in many countries.
The need to reintroduce elementary national reporting, to demonstrate progress towards the Year 2000 Objective of the International Tropical Timber Organization (ITTO), has led to the development of the more complex and detailed forest resource accounting - FRA (IIED and WCMC 1994). FRA includes concepts of quality and condition assessment. IIED and WCMC point out that FRA reporting could be developed from the level of the individual forest upwards, or from whole-country overviews downwards. Frequency of monitoring would depend on need, but it was anticipated that there would be annual national-level returns to the International Tropical Timber Council on progress towards the Year 2000 Objective. Uptake of FRA seems to be slow, partly because it is not obviously associated with incentives. The IIED/WCMC reports on FRA do not prescribe any methodology or minimum standards; FRA is simply a reporting system.
Annual and quinquennial reports providing snapshots or summarising trends are obviously useful for government planning and inter-governmental reporting (for example, as part of the post-UNCED process), but cannot easily reflect the diversity between forests or pick out trends which are important in any one forest. Many of the suggestions for items to be included in FRA have re-appeared in the criteria and indicators of forest management for the Helsinki and Montreal inter-governmental reporting systems. The FRA suggestions parallel also the concerns of the general public which were listed above.
Forest owners and managers will require, for their own business accounting, more detailed monitoring of the implementation of the forest management system. Those who receive grant subsidies or tax incentives will normally be required to submit annual returns to the grant-administering agency. This agency is often the national forest service, and these returns aid the compilations necessary for statistical reporting to the political administration and in respect of inter-governmental obligations. More detailed reporting may be required at the end of one grant cycle (3-5 years) or before the start of another cycle.
Purchasers of forest products, from processors to consumers, and non-consumptive users such as tourists, book readers and cinema-goers, are usually seeking frequent but generalised reassurance about the quality of forest management. Indigenous and traditional peoples also seek frequent reassurance, but with respect to particular products and services and particular locations. Local NGOs, and national/international NGOs running local campaigns, are more likely to want the same spatial and temporal degree of monitoring as an active forest manager.
The monitoring detail expected by stakeholders is generally proportional to the spatial and temporal scales. That is, for national and inter-governmental reporting, little detail is usually expected, because of the sheer volume of data and the difficulty in comprehending and acting on a mass of detail. The nature of reports on progress towards ITTO's Year 2000 Objective is left to member countries. In contrast, the UNCSD Secretariat closely prescribes the format and level of detail required for post-UNCED reporting.
Stakeholders remote from the forest do not usually have the technical knowledge to make use of detailed monitoring and therefore look to indices compounded from indicators, or to simple Yes/No certification decisions.
Stakeholders directly concerned with forest management, such as national forest services, private forest owners, forest managers, contractors and concessionaires will need monitoring as detailed as is required by the nature and scale of the forest management system which they are supervising or operating. Some of the debate in Europe is now centred on what is a reasonable level of monitoring detail which a private owner or manager can be expected to provide, relative to the monitoring undertaken by the state or national forest service, and who should pay for the monitoring.
At the national and inter-governmental levels, credibility of monitoring is attained mainly by inter-country comparisons. Countries which deviate markedly from the median values for a group of countries with roughly the same characteristics would be expected to explain the differences. The same kind of test should be applied to forest managers at the local forest management unit (LFMU) level, and this is indeed the procedure used by the main certification bodies now operating in forestry.
As noted in the previous section on detail in monitoring, stakeholders remote from the forest management process need a summary decision about the quality of management in a source forest. This implies trust in the certification process, with regard to the nature of the forest stewardship standards, the reliability of the certification bodies, and the audit process. Commercial enterprises dealing in products from certified forests have to acquire some knowledge of what the certificate means, if only in relation to the nature of the claims that can be made in public advertising. Consumer trust is not usually gained by direct evaluation of certification schemes but by the endorsement of those schemes by consumer and environmentalist organizations, relayed and publicised by broadsheet journalists. The debate on the Internet over the last several months about the quality of forest management in teak plantations in Costa Rica shows how important is professional competence on the part of the managers and certification bodies, trust between all concerned parties, and care in the public claims which are made in relation to the certificate of management quality.
A number of national forest services believe that their monitoring of forest practices is adequate and credible. They undertake monitoring sufficient in kind and intensity to satisfy legislation and the political administration. They argue that this should be sufficient to satisfy all other stakeholders. Currently, the Forestry Commission of Great Britain (FCGB) does not allow third-party, independent certification bodies to have access to information which would enable them to audit forests owned and managed by the Forest Enterprise wing of the FCGB. This stance overlooks the democratic right of consumer choice but, as there is no Freedom of Information Act in the U.K., the FCGB stance cannot be challenged legally.
The consumer and environmentalist objection to "Nanny-knows-best" is that "Nanny's" commissioned independent study (Lorrain-Smith and Walker 1993) showed that actual management in privately-owned, grant-aided woodlands in Great Britain often did not match the standards prescribed by the FCGB. A declining budget will make it more difficult in future for the FCGB to insist that its monitoring is effective and sufficient, even to its own standards.
More generally, the mandate of national forest services and their supporting legislation does not cover all the areas of concern which are encompassed within the current concept of good forest stewardship. Cultural and social issues are almost invariably outside the mandate of national forest services so, just on technical grounds, a government certification scheme would require unusual coordination between government agencies.
Disputes in Canada between the Provincial Forest Services, the commercial forest products companies and consumer/environmentalist organizations, about the adequacy of forest monitoring, have led to the involvement of the Canadian Standards Agency (CSA). Based on work in the private sector, the CSA has developed an environmental management systems approach. This approach has fed into both the Montreal Process for inter-governmental standards of forest management and an informal international study group to examine the possibility of a sector-specific standard under the auspices of the International Standards Organisation (ISO). Problems with the CSA approach are discussed in the section below on what to monitor.
In summary, the credibility of government monitoring of the quality of forest management is low because government is not viewed as being independent and impartial, and/or because its managements standards are not sufficiently enforced. Also, governments tend to take a narrower view of the concept of forest management than do consumer and environmentalist organizations. It is generally agreed that government monitoring is necessary but is not currently sufficient.
North American efforts at wildlife and wildland conservation, particularly of species close to extinction, has shown how important is frequent communication with stakeholders about positive progress as well as setbacks. NGOs, especially, need to maintain high levels of interest and consequent financial support. The NGO experience shows that a relentlessly optimistic view is not obligatory for maintaining public credibility and offers too many pitfalls to entrap the organization when setbacks are encountered. The multi-facetted nature of modern concepts of forest management means that it is only too easy to detect faults in some aspect of management in any forest.
The poor reputation of a number of forest products companies in private ownership stems partly from their failures to communicate with stakeholders. Having no shareholders does not mean having no stakeholders.
Government forest services tend to communicate the results of monitoring to stakeholders via the written word. In contrast, public limited companies in the private sector, and consumer/environmentalist organizations, take a much more multi-media approach and recognize the power of the visual image. Easily-obtained films and photographs of falling trees, burning forests and muddy, rutted extraction tracks convey strongly negative impressions which are difficult to counter with images of well-managed forests. The better-funded conservation organizations and state/national forest services show that well-chosen photographs can be effective transmitters of favourable impressions. However, the value of professionally-shot video and film is still underrated by smaller forest services and the private sector. In contrast, some environmentally-sensitive retailers use videos to train their staff to answer customers' question about sources of forest products and quality of forest management.
The analysis above showed the great variety of stakeholders who are associated with the current broad concepts of forest management, as distinct from a narrow focus on timber production. The increasing appreciation of the variety of stakeholders is a re-discovery of what was well understood and well articulated in the forest policy of India in the 1890s (IFS 1894). With such a broad constituency, disagreements about priorities and directions for forest management are inevitable. The crippling effects of frequent legal challenges to forest managers have been one of the factors propelling the US Forest Service to adopt an ecosystem management approach. A major feature of this approach is the development of a scenario of what a particular forest should produce, in goods and services, in the long term. Agreement on the scenario to be worked towards requires a long process of conflict resolution, and an iterative process during which current understanding of the production capacity of the forest is matched to the various expectation and needs of stakeholders. The complexity of the social questions to be addressed is indicated in Emery and Paananen (1995), and a wide-ranging questionnaire has been issued by the U.S. National Research Council to capture ideas which could be used to build scenarios for non-federal forests (Kirk Baer 1996).
The US ecosystem management approach will not eliminate disagreements about priorities but should provide a more workable basis for daily forest management, by fostering a more or less democratic process to agree on what is wanted. The CSA process likewise calls for early definition of values and goals, which are equivalent to scenarios in US ecosystem management. In principle, there should be better reconciliation of local, regional and national needs for goods and services and better mechanisms to adjust to changes in those needs as they change over time - without such frequent recourse to litigation.
Greater response to local needs and to the diversity of each forest may imply the monitoring of a greater diversity in indicators of good forest stewardship/sustainable forest management). There must be a balance between measuring everything that could be measured, and the effect of such measurement on the costs of the goods and services from that forest when the cost of monitoring is included. Ideally, the desirability of monitoring will be self-evident and necessary to the forest manager, and technically-feasible and economic procedures will be incorporated conventionally into the local forest management system. The additional cost of the monitoring would be negligible, but the cost of certification to demonstrate good forest stewardship or sustainable forest management is an unavoidable extra. Experience to date of third-party certification suggests that the external audit itself will add only about 1-2 per cent to retail prices.
Clearly, if the forest is being moved from an unmanaged state to a managed state, there will be significant management costs. Upgrading the quality of existing management will be much less shocking. Contrast the surprise in the Brazilian Amazon about the cost of introduction of even simple management (Verissimo et al. 1992) with the biophysical and financial benefits of reduced-impact logging in Sabah (Pinard et al. 1995, Pinard and Putz in press).
Inter-governmental reporting, for example, for the FAO Global Forest Assessment and to UNCSD and ITTO, mainly involves area statistics and estimates of percentage change from baseline dates. As noted earlier in this paper, changes in forest quality have not been included in the past. The methods for monitoring and the levels of precision are rarely prescribed; in contrast with the detailed prescriptions for monitoring of CFC gas emissions under the Montreal Protocol. Counts and areas are included in the FRA proposals by IIED and WCMC (1994), as well as in the Helsinki and Montreal Process reporting protocols. It is left to national agencies to interpret "significant deviation from the historic range of variation" and similar wording which occurs frequently in the Montreal Process (CFS 1995).
If methods of assessment, definitions of classifications, and levels of accuracy and precision are not prescribed, great problems of comparability and interpretation are inevitable, as FAO has noted at each decennial global forest assessment. The declassification of military technology, following the end of the Cold War, offers the near-future prospect of sub-metre resolution from instruments mounted on earth-orbiting satellites. Greater uniformity in global assessments, through more centralized and automated classifications, will reduce the historical problems caused by the idiosyncrasies of methods used within individual countries. Moreover, although differences between countries in assessment systems are regrettable, no direct economic penalty is incurred as a result.
If criteria and indicators (C/Is) of good forest stewardship are required only for monitoring of activities and outcomes at the level of the LFMU, such as an internal audit of a forest management system, then it would be quite appropriate to select indicators and associated norms (CSA "objectives') which are highly situation-specific and which match the policies and objectives specified in the local forest management system and management plan; such norms or target levels of indicators.
However, if the monitoring is to be used to demonstrate conformity with externally-set standards, then those standards are usually accompanied by instructions about methods, accuracy, precision, and the spatial and temporal aspects of sampling. The indicators to be assessed and the methods to be used must be standardized, if demonstration of conformity is a requisite for certification of the quality of forest management. This standardization is also necessary if the certification is to be mutually recognized by certification bodies which are accredited to certify in relation to a specified set of external standards. This equivalence is necessary for natural justice and consumer/environmentalist confidence. Otherwise, a certificate issued in one place could imply quite different management standards from those in another. The tight equivalence of C/Is should be associated with tight regulation of accreditation of certification bodies.
It is, obviously, a major challenge to devise an assessment system for quality of forest management which can be applied impartially, objectively and equitably to all kinds of forests globally. Indicators to be monitored should be meaningful and sensitive to change at the local level, and have the same applicability and meaning in any other forest. As a first approximation to this ideal, selected indicators should be those which operate at the highest level of generalization or abstraction while still being directly relevant to the local forest management unit. The administrative desirability of equivalence in C/Is is just the opposite of the need for adaptation of forest management systems (FMS) to site-specific circumstances. The compromise is selection from a menu of indicators for any one criterion, the indicators being as global as possible while still being relevant and meaningful at the LFMU level.
Proponents of environmental management systems (EMS) point to the huge number of indicators which would have to be monitored if the needs and expectations of all classes of stakeholders were to be taken into account. Over 700 criteria and indicators are listed in the full set of forest stewardship standards of the Initiative Tropenwald (Hahn-Schilling 1994). These are useful as a checklist when preparing the management plan for a LFMU but would be extremely expensive to apply thoroughly in monitoring of any one forest. The disadvantage of attempting to be comprehensive is that the diversity of forests and their human situations will always defeat the attempt; no single list can encompass all the situations, and even if it could, problems of comparability and weighting for the certification decision become overwhelming. Moreover, very complex systems are difficult for most stakeholders to comprehend. A common reaction of consultees in the FSC-U.K. process to the first public draft of standards compatible with the global principles and criteria of the Forest Stewardship Council (FSC ) has been "keep it simple and cheap, so that our stakeholders are not discouraged". Consultees have been more concerned about the potential amount of extra documentation, and the associated cost, than about possible changes in the forest management systems themselves and the associated monitoring.
Under the EMS approach, much attention is given to rational setting of policies and objectives, together with documented support for those policies and objectives and the assignment of resources for their attainment. The main objection from the consumer/environmentalist point of view to a pure EMS is that the standards to be followed can be set internally, without comparability to those in use in adjacent or other forests and without reference to the needs and expectations of the range of stakeholders. A common expression of the objection to a pure EMS is "an excellent EMS would not of itself prevent ecologically inappropriate harvesting by clear felling". There is no theoretical impediment to an EMS being mated with externally-set forest stewardship standards, a point reiterated frequently by the Executive Director of the FSC.
Emphasis on documentation of policies, objectives, procedures and monitoring within the ISO formulation of an EMS makes the development and implementation of an EMS a major positive factor when monitoring forest practices at the LFMU level. Given that current concepts of good forest stewardship/sustainable forest management involve a much more holistic view of the forest than simply timber production, and given the natural desire for additional costs of management and certification to be kept to the irreducible minimum, the choice of globally meaningful indicators to be monitored becomes critically important. This choice is discussed below, with components of the forest management system.
There appear to be no major issues in monitoring at the national/international levels which do not arise also in monitoring at the level of the local forest management unit. The current state of the reporting requirements under the Helsinki and Montreal protocols represent a progressive compromise between the wide ranging implications of the Global Forest Principles and Agenda'21, and the kinds of national statistics which national forest services can assemble without major extra expense. At this stage in both protocols, the individual countries are left to devise their own methods and standards for compiling the statistics.
Like proponents of certification at the LFMU level, participants at the inter-governmental level recognize the need for progressive and incremental improvements. Reports of concern about incompatibility between the two levels seem to reflect the worry of national forest services that independent certification of the quality of forest management might erode their mandates and, by extension, their budgets.
It is a regrettable fact that many countries have had excellent forest working plans which have been allowed to become out-of-date and irrelevant. A working plan by itself is not evidence of good forest stewardship, but most sets of FSS require plans to be prepared and implemented. In principle, it would be possible to assess the quality of forest management over a LFMU which was controlled only by an unwritten plan, whose details might be held in the memory of village elders or landowners. However, the cost of verifying the details and checking for contradictory interpretations would make it doubtful if such an operation could ever be certified at bearable financial cost.
Comparison between forest enterprises which currently practise good forest stewardship with those which are generally accepted to be deficient suggests that the presence or absence of a plan is not the major indicator to be assessed. Rather, the assessor should look for evidence of high level commitments within the enterprise to policies and objectives which foster good forest stewardship, and the dedication of resources to implement those policies. In other words, an EMS.
The CIFOR test team at Forstamt Bovenden in November 1994 proposed a framework with eight major components, which correspond to the main elements of operational forest management systems and concepts of good forest stewardship. The eight components listed below are slightly modified from the team's proposal, following discussions in the U.K. during December 1994 and January 1995. The correspondence is shown between these components and the Montreal Process criteria and indicators for use at the national/inter-governmental level (CFS 1995).
Institutional issues and decision-making processes
a. policy framework
b. regulatory framework (laws, rules, guidelines)
- these two components (a and b) are summarised in the Montreal Process criterion 7 "existence of a legal, institutional and economic framework for forest conservation and sustainable management"
c. decision-making processes and associated activities, including:
setting of priorities and targets; mechanisms for implementation; control of implementation; audit/monitoring of effects; detection and correction of mistakes; feedback into development (update/upgrade) of improved management and communication systems; documentation and record keeping.
d. training and research.
Factual basis for evaluation of good forest stewardship
e. productive resource base
- Montreal Process criterion 2 "maintenance of productive capacity of forest ecosystems" and criterion 4 "conservation and maintenance of soil and water resources"
f. ecosystem processes (cycles and interactions)
- Montreal Process criterion 3 "maintenance of forest ecosystem health and vitality" and criterion 5 "maintenance of forest contribution to global carbon cycles"
g. biological diversity
- Montreal Process criterion 1 "conservation of biological diversity"
h. well being of the forest-dependent people (including social and economic factors)
- Montreal Process criterion 6 "maintenance and enhancement of long-term multiple socio-economic benefits to meet the needs of societies".
The successive drafts of Helsinki and Montreal Process criteria and indicators for sustainable forest management have progressively focused on the indicators to be assessed and have avoided prescription about how the implied objectives should be attained. This contrasts with most of the FSS developed by NGOs, which tend to mix prescription and evaluation.
The great difference between the inter-governmental reporting systems of the Helsinki and Montreal Process types, and the FSS developed for use at the LFMU level, is that for the latter there are explicit or implied norms for the indicators to be assessed in the monitoring process.
A useful distinction is between threshold levels of indicators and target levels (using terminology now current in discussion of sustainable development indicators). A threshold level is the value or level of the indicator, beyond which the system may change significantly and damagingly to a state which is effectively irreversible in a human lifetime. Depending on the nature of the indicator, the threshold value may represent a cusp or inflexion on a graph. In contrast, a target value is the level of the indicator which the implementation of the forest management system should strive to attain or surpass during the operational period of the FMS. Some environmental NGOs would like to see the development of target values for indicators which would be globally meaningful and relevant at the LFMU level. It is not clear how feasible is this aim. The development of target values for the emission of CFCs under the Montreal Protocol/Vienna Convention suggests that some indicators of this nature may be found, for physical and biochemical factors related to soil and water qualities.
Policy, planning, and social indicators seem, at present, unlikely to be susceptible to evaluation against quantitative norms. However, within a given FSS, these indicators should be assessed with uniform scoring systems, or at least with scoring systems which are mutually compatible. The two-fold test should be - do the systems produce repeatable scores, and are the results comprehensible to stakeholders.
A preliminary step is to devise a minimum set of indicators for monitoring which conform to the following framework:
a. having the attributes of good criteria/indicators proposed by the Free University of Amsterdam and the World Bank (see annex 1), and by CSA (document Z808, draft of January 1996, pages 17-18);
b. having definable attributes such as "orientation" (relating to an input, a process or activity, or an outcome of forest management), geographic level (national, regional, local), and being objectively assessable or quantifiably measurable;
c. fitting within the 8 components of a forest management system proposed above, or within the 4 groups of forest functions (regulatory, carrier, productive and informative);
d. operating at the highest level of generalization/abstraction while still being directly relevant to the local forest management unit;
e. relating to an explicitly declared concept of good forest stewardship, such as the FSC Principles and Criteria (FSC 1994) and/or ITTO Guidelines (ITTO 1990). This point is important to avoid the criticism directed at purely ISO type schemes, in which the operator is permitted to devise internal Standards without any absolute frame of reference. Internally-generated Standards would not meet the expectations of conservation NGOs or consumer groups.
The CIFOR team at the Forstamt Bovenden test provided the following preliminary examples of indicators which might be monitored:
(criterion) National land use policy is implemented at the level of the local forest management unit.
(indicator) the forest management plan interprets and specifically refers to the national land use policy in the spatial assignment of forestry activities.
(criterion) Forest management respects all national and local laws and administrative regulations, including the national interpretation of international obligations (such as the CITES, ILO, ITTA and the Conventions on Biological Diversity and Climate Change).
(indicator 1) the operations manual (or equivalent document) requires the staff of the forest management unit to comply with all legal requirements.
(indicator 2) the documentation of the forest management unit confirms compliance with all legal requirements.
(criterion) Legal security of resource rights (usufruct, tenure) is guaranteed for the forest owner and operator of the management unit.
(indicator 1) document of title and/or usufruct licence has been granted by the appropriate authority and is legally registered.
(indicator 2) conflicts over resource rights are resolved by mutual agreement between the parties, or by arbitration or legal process, and the decisions are respected and upheld, as registered in records of the parties in conflict, arbitrators and/or court records.
(criterion) All legally prescribed charges are paid promptly and fully.
(indicator) management records confirm that dated payments have been made and acknowledged.
(criterion) Sufficient quantitative data are available for the whole area of the local management unit to allow rational planning and operation of forestry activities.
(indicator 1) statistically valid volume (or weight) and growth & yield data are current for the various kinds of forest products generated in the productive area of the unit, classified by operational sub-unit (such as the compartment or stand) and by the age- and/or size-class distribution. These data can be displayed in both tabular and map form.
(indicator 2) non-traditional forest products and services (such as perennial flows of clean water) are similarly quantified as regards their standing stock and production possibilities (flow rates).
(indicator 3) areas within the management unit which are not in the production working circle (or equivalent classification) are designated and displayed on all relevant maps. If they are subject to protection measures, critical points on the boundaries of these areas should also be marked with warning signs on the ground.
(criterion) Offtake rates of goods and services do not exceed the sustainable rates of production.
(indicator) offtake rates are continuously monitored and correctly recorded, and are compared with statistically valid estimates of production possibilities.
(criterion) Hydrological regime is not adversely affected by forest management operations.
(indicator 1) (riverine strips are large enough so that) no surface runoff from roads and tracks passes directly into natural watercourses.
(indicator 2) (riverine strips are large enough so that) no road construction material passes directly into natural watercourses.
(indicator 3) no artificial ponding is caused by construction of roads or tracks or by inadequate culverts or bridges.
(indicator 4) viable breeding populations of aquatic organisms are maintained.
(indicator 5) perennial supplies of clean water are maintained.
(indicator 6) quality of outflow water is not less than the quality of inflow, unless there are endogenous changes caused by natural features of the geology or soils within the LFMU, such as peatswamps
(criterion) The IUCN Red List category status of forest-dependent species is not adversely affected by forest management operations.
(indicator 1) viable healthy breeding populations of "keystone" species are maintained.
(indicator 2) there is no deliberate reduction in genetic diversity.
(indicator 3) there is special provision for rare species.
These examples include indicators of process (in relation to the regulatory framework criteria) and of outcome (in relation to the forestry practices in the field). As reiterated above, they are intentionally generalized, partly to allow mutual recognition of certificates of quality of forest management. In addition to the generalized indicators, it is likely that forest owners/managers will develop indicators to demonstrate compliance with the enterprise- and forest-specific prescriptions documented in their forest management systems. The distinction between the globally-generalizable and the forest-specific indicators might help to keep small the number of global indicators and so reduce costs for small enterprises with simple management systems.
Because of the difficulty of devising indicators which are truly global in applicability and simultaneously locally relevant, forest managers might be offered the option of proposing alternative indicators for any one criterion, which would be in keeping with the particular circumstances of their own forests and compatible with the generalized indicators. Some consultees in the FSC-U.K. process for developing national Standards felt that this option was too liable to abuse in some countries and would give rise to endless arguments about the comparability of certificates issued under any one set of FSS. These objectors included Greenpeace.
The CSA proposal (document Z808, draft of January 1996) opts for a well structured and rigorous EMS combined with indicators which must be selected locally through public consultation. CSA acknowledged that "In very small areas, selecting indicators becomes [impracticable]". CSA was also frank that public consultation could involve a long and/or complex reiterative process of indicator selection and discard.
It will be evident from a consideration of the components of a forest management system and the indicators exemplified above that imagery from remote sensing instruments on earth-orbiting satellites can provide only a fraction of the desired information for certification. AVHRR imagery provides 1 km resolution, older Landsat 80 metres, while the newer Landsat and SPOT both provide 30 metres. Remote sensing and GPS literature is looking forward to the declassification of military technology as part of the "Peace Dividend" at the ending of the Cold War, or the development of civilian technology, to allow resolution to improve to 1 metre or less. Leaving aside the stupendous problems of storing such data streams, and of selectively retrieving the desired information over periods as long as tree rotation lengths, remote sensing can provide useful information about spatial changes in atmospheric temperature profiles and the reflectance characteristics of ground cover.
A geometrically rectified print of a single Landsat scene costs between US$ 1100 and 1600 from the Brazilian national space agency INPE. Although some NGOs in developing countries are equipped to use computer-compatible tapes of remotely sensed imagery, most national and state forest services, forest enterprises and environmentalist organizations in developing countries lack the training to make full use of such imagery. The cost of hardware and materials is rarely an obstacle for an organization which can make use of the information. Moreover, through the Internet, an increasing volume of imagery is becoming available for little more than the cost of telecom time to download the data or images. Experience in Latin America suggests that the main barrier to efficient use of remotely sensed imagery is inadequate training and consequent lack of confidence.
The much publicised attractions of remotely sensed imagery, and its relatively low cost, have tended to diminish the use of aerial photographs at scales of 1:2000 to 1:60,000. There has been surprisingly little use for monitoring in the tropics of video imagery from cameras mounted in booms attached to light aircraft or, on a more restricted scale, of videos from ultralight aircraft. For special situations which need inspection of areas which are difficult or dangerous to reach on the ground, a video camera mounted in a miniature helicopter costs upwards of US$ 3,000 per day. The Hover-Cam is restricted to a range of about 400 metres in line of sight from the pilot controlling the robot helicopter from the ground. This would be too expensive to use in audits of most forest enterprises, but is an indication that absence of technology is not a major problem in assessment of those factors which can be evaluated through remotely sensed imagery.
Snapshot information (indicator data assessed at a particular date or time) is of limited use by itself. Sustainable forest management can be evaluated only in retrospect. One can say that a particular forest has been managed so that the productive capacity has been maintained, enhanced or degraded in relation to a previously assessed baseline, or in relation to a trend determined through repeated assessments of the same indicators using the same methodology. Snapshot information related to the same indicators from similar forests might suggest the relative quality of forest management in a particular forest, but would be much less useful than comparison with trend data from the same forest.
Forest stewardship is a more workable and equitable concept than sustainable forest management. Stewardship is assessed by reference to the commitment to sustainable management (as expressed in an enterprise's policy statements), to the specific objectives and mechanisms (as expressed in the documented forest management system), and to the matching of indicator values to target values registered in the FMS documents.
It should not be possible to certify the sustainability of forest management until an enterprise has established a track record and has accumulated, or inherited, baseline and/or trend data for indicators. However, this is patently unfair if an enterprise's tenure over a forest is too short for it to obtain such information. Given that management tenure, especially through the concession system, is often limited to years rather than decades, certification of the quality of forest stewardship is more equitable. The same indicators or kinds of indicators should be assessed for both concepts.
With very few exceptions, long-term trend data in the tropics is confined to tree growth and survival in yield plots and silvicultural experiments. Although some long-term plots date to the 1930s, the principal series were started in the late 1950s. Some of these plots have been lost because of inadequate initial mapping (Sarawak), government policy decisions in favour of a change in land use (Peninsular Malaysia), damage during re-entry before the end of the regeneration cycle (many countries), civil war and other disorders (Philippines and Uganda). Data from some plot series have been difficult to interpret, because of official or unofficial changes in assessment practices, sometimes documented (Ghana and Uganda). Registers of long-term plot series have been compiled by the ASEAN Institute of Forest Management, by FORTECH at the Australian National University, and by the S4.02 subject group of IUFRO.
Most other long-term records in the tropics are restricted in duration to the working lifetime of the enthusiast who started the series. This applies especially to ornithological censuses, which may be the best available long-term sets for assessment of changes in biodiversity. BirdLife International is compiling a register of such sets. Detailed bird records may also be good indicators of changes in quality of forest habitat, and similar claims are made for dung beetles (Nigel Stork at the Cooperative Research Centre - Tropical Rainforest Ecology and Management, James Cook University, Australia), moths (Jeremy Holloway at the Natural History Museum, U.K.) and nematodes (Nigel Price at the CAB International Institute of Parasitology, U.K.).
It is not clear that simultaneous modelling of the behaviour of multiple aspects of forest management will be helpful, in relation to a certification decision, until there is greater clarity about the indicators which satisfy a consensus of stakeholders. Model results, projected from past trends, may indicate that the current management regime is sustainable. Moir and Mowrer (1995) stress that models cannot (yet) measure or incorporate the probabilities of increasing stochasticity, critical threshold or extreme-value events, or chaotic system behaviour. They advocate a conservative approach to management, maintaining/enhancing biodiversity, and avoiding operations which might create overcompensating feedbacks.
It seems to be difficult to disaggregate the overall costs of monitoring which national forest services undertake. In the consultations for the development of FSC-compatible Standards for the U.K., the FCGB has expressed concern about the additional costs of improved management (including self-monitoring) and of external audit. At present, the probable costs of external audit by third-party certification bodies are commercially confidential. Various statements by one FSC-accredited certification body suggest that audit costs might increase retail prices by about 1-2 per cent, including audit of the chain of custody from forest gate to retailer. The FCGB is apparently not able to separate from its accounts the costs of the monitoring of grant-aided private sector forestry by its woodland officers. NGOs such as the Royal Society for the Protection of Birds have expressed concern that the FCGB's guidelines for and monitoring of both state and private forestry may be inadequate in relation to the U.K.'s international obligations (Cosgrove and Turner 1995).
If monitoring is undertaken to improve the quality of forest management, as a tool within an enterprise or a national forest service, it would be rational to vary the importance of different indicators in accordance with the circumstances and objectives of different forests. When monitoring is used as a tool to aid inter-governmental reporting or mutually recognized certification, then standardization becomes essential. When certification is the issue, and the decision is a simple Yes/No - the forest is under good stewardship, conducive towards sustainable forest management, or it is not - then the ranking of criteria and indicators and/or the weighting of the indicator values becomes extremely important.
The difficulty is obvious. If the monitoring system is sensitive to the circumstances of a particular forest, the ranking and weighting may be quite inappropriate for any other forest. If tenure over forest resources is a major issue in one forest but is not critical in an adjacent forest, how should the respective rankings be managed so as to arrive at a certification decision which is comparable with and comprehensible to stakeholders everywhere else ? Are social factors more important than pollution records ? Does maintenance of bio-geo-physico-chemical water quality and perennial flows of clean water rank higher than enhanced biodiversity ?
Certification bodies currently accredited by FSC have taken different approaches to this difficult issue. The Smart Wood programme of the Rainforest Alliance (RALL 1993) emphasises consensual agreement between stakeholders and the audit team at the start of the inspection. The criteria are refined and weighted to account for a variety of factors such as scale and intensity of the forest management and the scatter of the forests within the applicant enterprise. Indicators are scored on a subjectively-applied scale from 0 = not applicable, through 1 = strongly unfavourable, to 5 = strongly favourable. Smart Wood has acknowledged that the system is unsatisfactory because of its non-equivalence between forests or enterprises and will refine its procedures as part of the conditionality attached to the first year of accreditation by the Forest Stewardship Council.
The Forest Conservation Program of Scientific Certification Services (SCS 1994) collects indicator data and interviews stakeholders as input to the judgement of the relative importance of each criterion with three major elements (timber resource sustainability, forest ecosystem maintenance, and financial and socio-economic considerations). The relative importance of each criterion is determined through a formal, numerical, multiple attribute ranking algorithm. The certification decision is presented as percentage scores in each of the three major elements. There is obviously still a major role for subjective judgement and for possible bias in the sampling of the stakeholder opinions.
It may help to reduce the areas for disagreement about the equivalence of certificates if the number of indicators could be reduced and ranking or weighting eliminated. Developers of FSS have taken a variety of approaches to listing of criteria and indicators. RALL Smart Wood have only a short list, generalized in a manner which should facilitate compatibility with an EMS. ITW chose the other extreme, with a highly detailed checklist. All the CIFOR test teams have noted that the FSS being compared are not strictly comparable, because they have been developed from different philosophical standpoints and constituencies. Not surprisingly, however, there are many elements in common to all the FSS, including those used in inter-governmental reporting formats.
One way to reduce the number of indicators to be monitored could be a HACCP - Hazard Analysis by Critical Control Points (Maurice 1994, Gibson 1995). This system for focusing on critical aspects of a system was devised initially to ensure the maximum safety of food for US astronauts - gastric disorder in a cramped spacecraft could be most unpleasant. The HACCP was a review of the entire system, from growing the food through harvesting and transport, processing, packaging, and storage. The review identified those points in the system at which pollution was most likely to occur, and then concentrated control measures at those points. NASA felt that it was impossible to attempt to monitor all aspects of the food chain, even when financial resources were almost unlimited.
The parallel with the forest management system is that both that and the astronauts' food chain have many interacting elements from a huge and varying list of inputs and processes (both physical activities and managerial decisions). The difference is that NASA had a quite restricted range of outcomes (different types of food), while the range of stakeholders means that the forest owner/ manager has a much larger variety of outcomes.
There may be useful parallels also for foresters in the selection and condensation of sustainable development indicators into politically usable indices (Hammond et al. 1995). However, at the level of the LFMU, the forest owner/manager may have problems in interpretation of the causes of changes in highly aggregated indicators, unless those changes are quantifiably related to specific causes.
After reducing to a minimum the subjectivity of indicator assessments and the number of indicators, there remains the problem of assimilating personnel recruitment scores, fish population numbers and extraction track areas (or other indicator values) to a single certification decision. The SCS procedure has been mentioned above. The "analytic hierarchy process" described by Kangas (1992) uses a pair-wise analysis similar to that of SCS but also relies on the consistency and narrow span of stakeholder opinions. The multicriteria analyses approach favoured by some specialists in sustainability indicators (for example, Munasinghe and McNeely 1995) follows a comparable philosophy which seems to produce consensual answers when the main factors can be reduced to a common currency which is, usually, money. It seems likely that more consensus is required on methods for holistic valuation of forests before a multicriteria analysis in support of certification decisions would be generally acceptable.
Less than a decade has passed since certification of the quality of forest management emerged as an important issue for consumer and environmentalist organizations. The UNCED process and the publication of the report "Our Common Future" by the World Commission on Environment and Development (the Brundtland Commission - WCED 1987) coincided with widespread dissatisfaction with government and industry monitoring of the state of forests. IIED's report to ITTO in 1988, on the tiny proportion of forest in ITTO producer countries under a recognizable form of sustainable management (Poore et al. 1989) stimulated demand for market incentives for better management.
There are currently over 25 sets of FSS under development, mostly for application within a particular country but some devised with the intention of generic applicability. Government agencies and NGOs developing FSS have made most progress when taking an incremental approach, involving broad consultations with a wide range of stakeholders. It is a reasonable concern of national forest services, forest managers and the forest products industries that certification should not add significantly to the price of forest products, either through improvements to field forest management or in the audit process.
It is mainly the responsibility of the developers of FSS to justify the criteria and indicators which they propose, and to show how these relate to critically important aspects of a holistic view of managed forests. Nearly all the current sets of FSS have indicators which are so numerous or so costly to measure that their wholesale incorporation into forest management systems could disadvantage those operators which are seeking to respond transparently to public concerns. Most of the current developers of FSS are committed publicly to progressive refinement of their Standards as knowledge and experience accumulate.
The two aspects of indicators which need most research are:
a. rationalized choice of a minimum set of indicators;
b. methods for assimilating the values of the chosen indicators into a Yes/No certification decision.
The minimum set should be large enough to cover all the major components of a forest management system, and should be selected to be as globally applicable as possible yet still meaningful and relevant at the LFMU level. The aim should be to replace the current reliance on best professional judgement with impartial, objective and transparent assessments. Because of the large numbers of forest stakeholders, and the breadth of their concerns, it is difficult to set priorities for composing and refining the set of indicators relevant to certification. Given that certification is a market-driven process, which relies on consumer demand, an early concentration on satisfying the concerns of the general public (as summarised earlier in this paper) could provide direction. Such an approach would also be conducive towards establishing public credibility for the certification process, a major theme of the organizers of this conference.
At this early stage in certification, it may be found that the criteria and indicators preferred by consumer and environmental groups are not cost-efficient or cost-effective, or are too difficult to evaluate with scientifically acceptable methods, or lack adequate technology. Such findings would be helpful in developing market understanding of the complexity of forest management and the extraordinary difficulty of making definitive judgements.
It should be self-evident that the results of monitoring ought to feed back into the process of dynamic forest management. However, as shown by the accumulations of unanalysed data from long-term sample plots in many countries, the feedback does not happen automatically. One of the strengths of an EMS is the requirement for demonstration of feedback loops in management. Many developing countries lack the skills for managing monitoring data, and require substantial training. However, there should also be improvement in the techniques for communicating and using the results of monitoring, including dissemination to interested but non-technical stakeholders.
BEFC (1948) 'Consolidated report on the 5th British Empire Forestry Conference'. London, U.K.; Forestry Commission of Great Britain.
BLINN, Charlie (1995) 'On-line bibliographic database of ecosystem management literature citations'. Notice of availability posted on Internet <forest@listserv.funet.fi> on 27 November.
BOTTING, Mike (1994) 'Green issues: a hard sell ?'. TTJ Supplement on Timber and the Environment 368 (6084) 29 January 1994, p4-5.
CFS (1995) 'Criteria and indicators for the conservation and sustainable management of temperate and boreal forests. The Montreal Process'. Hull, Canada; Canadian Forest Service. ISBN 0-662-23109-0. 27p.
COLFER, Carol J. Pierce (1995) 'Who counts most in sustainable forest management ?'. Working Paper number 7. Bogor, Indonesia; Center for International Forestry Research. 16p.
COLFER, Carol J. Pierce with PRABHU, Ravi; WOLLENBERG, Eva (1995) 'Principles, criteria and indicators: applying Ockham's razor to the people-forestry link'. Working Paper number 8. Bogor, Indonesia; Center for International Forestry Research. 16p.
COSGROVE, Peter; TURNER, Roger (1995) 'Implementation in the United Kingdom of the Helsinki guidelines for sustainable forest management'. Edinburgh, U.K.; Royal Society for the Protection of Birds. 63p.
CSA (1996) 'A sustainable forest management system'. Guidance document, draft of 19 January 1996, number Z808. Ottawa, Canada; Canadian Standards Association. 55p.
DYCK, W.J.; COLE, D.W.; COMERFORD, N.B. (editors) (1994) 'Impacts of forest harvesting on long-term site productivity'. London, U.K.; Chapman & Hall. ISBN 0-412-58390-9. 371p.
EMERY, Marla; PAANANEN, Donna M. (1995) 'Humans, forests, and global environmental change: planning a social science research agenda'. General Technical Report number NE-212. Radnor, Pennsylvania; USDA Forest Service, Northeastern Forest Experiment Station. 22p.
FSC (1994) 'Principles and criteria for natural forest management'. FSC Document number 1.2. Oaxaca, Mexico; Forest Stewardship Council. 8p.
GERSTMAN and MEYERS (1991) 'National survey of consumer opinion'. Quoted in SCS (1994) 'The forest conservation program. Program description and 1994 operations manual'. Oakland, California; Scientific Certification Systems. p2.
GIBSON, Donald (1995) 'Hazardous food'. Letter to the Editor. New Scientist 145 (1960), p47.
HAHN-SCHILLING, Bernd (1994) 'Criteria for an evaluation of sustainable management of tropical forests'. In: HEUVELDOP, Jochen (compiler) (1994) 'Assessment of sustainable forest management (a contribution to the development of concept and procedure)'. Mitteilungen der Bundesforschungsanstalt für Forst- und Holzwirtschaft, Hamburg number 178, p37-165.
HAMMOND, Allen; ADRIAANSE, Albert; RODENBURG, Eric; BRYANT, Dirk; WOODWARD, Richard (1995) 'Environmental indicators: a systematic approach to measuring and reporting on environmental policy performance in the context of sustainable development'. Washington, D.C.; World Resources Institute. ISBN 1-56973-026-1. 8+51p.
IFS (1894) 'Forest policy'. Circular number 22 F. New Delhi, India; Indian Forest Service. 8p.
IIED and WCMC (1994) 'Forest resource accounting: stock-taking for sustainable forest management'. Final report of an activity for the International Tropical Timber Organization "Forest resource accounting: monitoring forest condition and management". IIED Forestry and Land Use Series number 1. London, U.K. International Institute for Environment and Development, and Cambridge, U.K.; World Conservation Monitoring Centre. ISBN 0-905347-98-6. 51p + 30p annexes.
INNES, John L. (1993) 'Forest health: its assessment and status'. Wallingford, U.K.; CAB International. ISBN 0-85198-793-1. 656p.
ITTO (1990) 'ITTO guidelines for the sustainable management of natural tropical forests'. ITTO Technical Series number 5. Tokyo, Japan; International Tropical Timber Organization. 9+18p.
KANGAS, Jyrki (1992) 'Multiple-use planning of forest resources by using the Analytic Hierarchy Process'. Scandinavian Journal of Forest Research 7:259-268.
KIRK BAER, Charlotte (1996) 'NRC Committee on Prospects and Opportunities for Sustainable Management of America's Nonfederal Forests'. Questionnaire posted on Internet <forest@listserv.funet.fi> on 3 May.
LeVAN, Susan L. (1994) 'Life cycle analysis: assessing environmental impact'. Keynote address to Section 5, Environmental Aspects, of the International Research Group on Wood Preservation. 25th Annual Meeting, Bali, Indonesia, 29 May - 3 June 1994. Document number IRG/WP/94-50034. Stockholm, Sweden; IRG Secretariat. 12p.
LORRAIN-SMITH, Roy; WALKER, William (1993) 'An assessment of standards in grant-aided woodland planting. report of a survey carried out for the Forestry Commission'. Whitchurch, U.K.; CANOPY, the Environmental Consultancy Division of Scottish Woodlands Ltd. 2 volumes. 38p + appendices.
McKENZIE, Daniel H.; HYATT, D. Eric; McDONALD, V. Janet (1992) 'Ecological indicators'. London, U.K.; Chapman & Hall. ISBN 0-412-73970-4. 2 volumes. 1567p + appendices.
MAURICE, John (1994) 'The rise and rise of food poisoning'. New Scientist 144 (1956), p28-33.
MOIR, William H.; MOWRER, H. Todd (1995) ' Unsustainability'. Forest Ecology and Management 73:239-248.
MORI (1990) 'Tropical rainforests and the environment. A survey of public attitudes. Research study conducted for WWF, the World Wide Fund for Nature'. London, U.K.; Market & Opinion Research International. 24p + appendices.
MUNASINGHE, Mohan (1993) 'Environmental economics and sustainable development'. World Bank Environment Paper number 3. Washington, D.C.; World Bank. 112p.
MUNASINGHE, Mohan; McNEELY, Jeffrey (1995) 'Key concepts and terminology of sustainable development'. In: MUNASINGHE, Mohan; SHEARER, Walter (editors) (1995) 'Defining and measuring sustainability: the biophysical foundations'. Washington, D.C.; World Bank and United Nations University. ISBN 0-8213-3134-5. p19-56.
MUNDA, G. (1994) 'Fuzzy information in multicriteria environmental evaluation models'. Ispra, Italy; European Commission/Joint Research Centre, Institute for Systems Engineering and Informatics. Grant number EUR 15602 EN. 6+233p + bibliography.
OFI (1991) 'Incentives in producer and consumer countries to promote sustainable development of tropical forests'. Pre-project report to the International Tropical Timber Organization. Document number PCM, PCF, PCI (V) /1/Rev.3. Oxford, U.K.; Oxford Forestry Institute in association with the Timber Research and Development Association. 71p.
PASTOR, John; MLADENOFF, David J. (1993) 'Modeling the effects of timber management on population dynamics, diversity, and ecosystem processes'. In: Le MASTER, Dennis C.; SEDJO, Roger A. (editors) 'Modeling sustainable forest ecosystems: proceedings of a conference, Washington D.C., 18-20 November 1992'. American Forests 16-19.
PINARD, Michelle A.; PUTZ, Francis E.; TAY, John; SULLIVAN, Thomas E. (1995) 'Creating timber harvest guidelines for a reduced-impact logging project in Malaysia'. Journal of Forestry 93 (10), p41-45.
PINARD, Michelle A.; PUTZ, Francis E. (1996) 'Retaining forest biomass by reducing logging damage'. Biotropica (in press).
POORE, Duncan; BURGESS, Peter; PALMER, John; RIETBERGEN, Simon; SYNNOTT, Timothy (1989) 'No timber without trees: sustainability in the tropical forest. A study for ITTO'. London, U.K.; Earthscan Publications Ltd. ISBN 1-85383-050-X.
RALL (1993) 'Generic guidelines for assessing natural forest management'. Revised draft October 1993. New York, New York; Rainforest Alliance, Smart Wood Program. 8p.
SCS (1994) 'Program description and 1994 operations manual'. Oakland, California; Scientific Certification Systems, Forest Conservation Program. 33p + 8 appendices.
SCHUSTER, Ervin G.; LEEFERS, Larry A.; THOMPSON, Joyce E. (1993) 'A guide to computer-based analytical tools for implementing National Forest plans'. General Technical Report number INT-296. Ogden, Utah; USDA Forest Service, Intermountain Research Station. 269p.
VERISSIMO, A.; BARRETO, P.; MATTOS, M.; TARIFA, R.; UHL, C. (1992) 'Logging impacts and prospects for sustainable forest management in an old Amazonian frontier: the case of Paragominas'. Forest Ecology and Management 55, p169-199.
WCED (1987) 'Our common future'. Report of the World Commission on Environment and Development - the Brundtland Commission. Oxford, U.K.; Oxford University Press. ISBN 0-19-282080-X paperback. 15+400p.
WHITE, James D.; ESTERHOLDT, Karen (1995) 'D.E.M.O.: a research/NSF cooperative effort'. Forest Research West August 1995. Fort Collins, Colorado; USDA Forest Service, Rocky Mountains Forest Experiment Station. p4-7.
The CIFOR field team at the first comparative test of selected FSS at Forstamt Bovenden in 1994 reviewed papers by the World Bank, World Resources Institute and Free University of Amsterdam on indicators or environmental quality and sustainable development. The following table summarised the review:
Desirable characteristics of indicators
1. represent components or processes of real world systems / explicit or implicit models / represent reversible and manageable processes.
2. are empirical specification of concepts which cannot be fully operationalized on the basis of generally accepted rules / a compromise between scientific accuracy and the demand for concise information but still having a scientific basis. If the phenomenon is not easily expressed in scientific terms (as in some social phenomena), the indicators should at least be well defined and clearly expressed.
3. quantify information, more than words or pictures alone, through a fixed (prescribed, rather than unchanging) methodology of measurement / include reference or threshold values / numerical values have significance beyond the numerical value itself (surrogate for complex variables).
4. have predictive meaning, providing direct information about a possible or likely future of the system represented by the indicator.
5. simplify information about complex phenomena, so as to improve communication.
6. provide information without social bias (not subject to different interpretations according to social group).
Criteria for selection of indicators
a. sensitivity to changes in time.
b. sensitivity to changes across space.
c. sensitivity to change over social distribution.
d. sensitivity to reversibility of effects.
e. sensitivity to controllability of effects.
f. predictive ability.
g. integrative ability.
h. ease of data collection.
i. relative ease of application.
j. relevance to vulnerability of ecosystems and assessment of risk / detection of cusps in environmental response surfaces.
As Europe's largest hardwood importer, Timbmet (Oxford, U.K.) has had much call to consider its position with regard to environmental issues. Timbmet is increasingly being asked by customers and environmental groups for verification that its timber comes from well managed and 'sustainable' sources. As a result, the post of Environmental Coordinator was created at Timbmet with the brief of identifying all timber stocked to its source at forest level. Information will be collated from all of Timbmet's suppliers. They will then be allocated a rating, based on the details and documented evidence that they can provide in relation to their environmental standards. This information will be used by Timbmet's purchasing director to purchase preferentially from those suppliers which demonstrate the highest environmental credentials, with the aim of eliminating all suppliers which cannot demonstrate good forest management. At present Timbmet stocks about 60 species from hundreds of suppliers and forest worldwide. The task of verifying the sources all of these timbers and gaining meaningful information on all suppliers is long and complex.
A number of routes are being used to obtain information on the source of Timbmet's timber. By approaching various different sources, it is hoped that general information will be accumulated on the producing countries as well as specific details of Timbmet's direct suppliers.
Information is being requested from Timbmet's suppliers through the medium of questionnaires. These are being sent out in order to obtain information concerning the sources of Timbmet's timber, and also the route that the timber has taken prior to reaching the company's yard. By the time the timber reaches Timbmet's gate, it has already passed through a number of links in the wood chain. The questionnaires are designed so that each of the linking enterprises, through which the timber has passed, should answer a section of the questionnaire before passing the questionnaire onto that link's own suppliers. Questions addressed to Timbmet's agents, shippers and the primary mills request information on how they may be contacted. Other questions are designed to assess the supplier's commitment to environmental improvement and, in particular, the good management of the world's forests. Suppliers are asked, for example, whether they have an environmental policy, or a policy to purchase timber only from well managed forests.
In order to obtain information on the forest source of Timbmet's timber, the questionnaires are to be sent to the forest owners or concession holders. As well as questions concerning contacts and evidence of their commitment to environmental issues, the respondents are asked more detailed questions concerning the management of their forests. Timbmet is asking if a management plan is being implemented, what type of forest is being managed, if there is a reforestation programme, as well as questions designed to assess whether sustained yield forestry is being implemented. Timbmet asks questions designed to elicit whether the forest is being managed in such a way which would make it eligible for application for certification by a FSC-accredited certification body.
The questionnaires are tailored to each of the countries to which they are sent. For example, the forest managers will be asked if they are members of any trade association in their country. As Timbmet gains information on the trade association, this one simple question to the supplier will yield a large amount of information regarding the commitments and requirements. It is hoped that such an approach to the design of the questionnaires will help the applicant relate the questionnaire to their own situation.
Only a few responses have been received so far to the questionnaires. It obviously takes time for the questionnaires to filter back down the wood supply chain and for all of the parties concerned to complete their respective sections. Most of the information we have managed to collate so far has come from North American suppliers. The amount of detail they have been able to provide to Timbmet has varied. Some suppliers have been very encouraging, including details of their forest management. Others are however vague, with broad geographical regions being given for forest location, and only limited information on forest owners. Only time will tell if the gaps in the information can be filled in as Timbmet returns to ask for more of the information. A number of Timbmet's agents have expressed doubts about the likelihood of determining the exact forest from where all of Timbmet's timber originated. This has been the main reason for trying to assess all of the stages in the wood supply chain through which Timbmet's timber has passed. Even if Timbmet only manages to obtain information from the first stages back along the wood supply chain, the company will at least have some indication of the level of commitment exhibited as a starting point. Timbmet will know if timber from well managed forests is being sourced, and will be able to identify weak links in the chain.
Many of the gaps in the information may not be due to the impossibility of provision by the suppliers, but to the perceived impracticality of the process. Imagine the wood supply chain as a fan with Timbmet at the epicentre of the fan. Timbmet has over 200 direct suppliers in the form of agents or shippers. They in turn may deal up to one hundred suppliers with mills. Each of these mills may use timber from up to a hundred forests. In the worst case scenario, Timbmet is therefore asking for information on two million forest areas worldwide. Moreover, this figure is not static, because mills process timber from different forests at different times. At the mill level, timber from a large number of forests is combined and sawn to meet Timbmet's specifications. It is therefore a very complex process to find out from which forest the timber was derived unless, as is sometimes the case, the mills own their own forest, or only use timber from a limited number of forest owners. That said, money has changed hands and so it should be possible to trace back to the timbers source through the succession of related invoices. Clearly, it would be simpler to follow timber from the forest through the wood supply chain, rather than working back down the chain from Timbmet's mill. This would be the approach taken by FSC-accredited certification bodies, through their chain of custody verification. In that process, each of the processors and handlers of the timber must be certified as suppliers of certified timber. Obligations include physical separation of timber from certified sources, from non-certified timber.
A further complication in tracing timber sources is that, in areas which are cut over once and then are to be left to regenerate for a number of decades, the information Timbmet receives will only be valid for limited periods, becoming rapidly outdated. This is another reason for assessing the level of commitment throughout the chain. If Timbmet knows that a shipper, for example, is committed to sourcing timber from well managed forests, the company may rely on that shipper to act responsibly even if Timbmet does not have detailed information from each of those forests at all times.
As a customer of the suppliers, Timbmet can put pressure on them to meet high environmental standards by giving preference to those who can demonstrate commitment and provide evidence that they are promoting and practising good forest management. By giving preferential treatment to those who can demonstrate good forest management, Timbmet provides a market incentive for the suppliers to maintain or improve their environmental standards.
By putting this pressure on the suppliers however, there is the danger that they will provide just the information which they think that Timbmet wants to hear. Verifying that the received information is correct and complete is the next step in the collation of environmental information.
Cross referencing of information is one way in which the authenticity of the received information will be assessed. Documentation from national forestry services of producer countries provides information on their national forest policy. Similarly, information collated from trade associations details members' commitments and requirements. The combined information should indicate how the forests should be managed. Timbmet can use this combined information to assess compliance and non-compliance with regulations by comparison with the answers received from suppliers. Cross-referencing ('triangulation') can also be carried out by collecting data from various suppliers, who in turn source timber, and therefore data, from the same set of upstream suppliers. Some of the claims made by Timbmet's suppliers can be verified by directors of the company, and by Timbmet's agents, when they visit Timbmet's suppliers. Until there is a mechanism for independently tracing timber from the forest to the end user, Timbmet will have to rely on the information that Timbmet's various sources of information can provide, cross-referencing as much as possible in order to substantiate claims.
Through this process, Timbmet expects to obtain vast amounts of information and to be able to assess, to some extent, its reliability. Based on the amount of information deemed reliable, and the level of commitment demonstrated, Timbmet's suppliers will be graded. Purchasing will be directed towards those with the highest environmental standards. It is recognized that while studying written information received in Oxford, from forests all around the world, Timbmet staff are not qualified or able to determine if a forest is being sustainably managed. Timbmet aims to identify indicators of good forest management and to promote its application. Assessment of sustainable forest management needs to be carried out in the forests by people qualified to do so. As more timber becomes available from certified forests, Timbmet will preferentially purchase it, so promoting sustainable management of the world's forests. Until then, the next best option is to verify the sources of the timber, thus promoting trade in timber from well managed forests, reducing trade in timber from inappropriately managed sources, and putting into place the systems and framework for allowing the marketing of certified, sustainably produced timber.
Clearly there are pitfalls with the process of tracing timber. There are however very many positive points. Through this process, Timbmet is gaining large amounts of information from its suppliers about their awareness of environmental issues and commitment to improvement, as well as about the source and route that the timber has taken prior to reaching the company's yard. Checking on the sources of the timber is an interim measure, prior to greater availability of forest certification and chain of custody verification. Although it is not faultless, verification of sources is an essential process leading to the marketing of certified, sustainably produced timber and will lead to the increase in trade of timber from those suppliers who make noteworthy progress in good forest management.