Advice to the Lands and Forest Service
on Timber Supply and Management

Alberta Forest Management Science Council
C.R.James (Chairman), V. Adamowicz, S.Hannon, W.Kessler,
P.Murphy, E.Prepas, J.Snyder, G.Weetman, M.A.Wilson

September 10, 1997

Table of Contents

SCIENTIFIC CHALLENGE 

COUNCIL RESPONSE

A. DEFINITION OF SUSTAINABLE FOREST MANAGEMENT 

B. ELEMENTS OF SUSTAINABLE FOREST MANAGEMENT

1. Ecological Integrity and Inherent Disturbance
2. Desired Future Forest 
3. Social and Economic Values and Public Involvement 
4. Scales-Spatial and Temporal 
5. Adaptive Management 

C. CONCLUSION 

D. REFERENCES 

SCIENTIFIC CHALLENGE

The Government of Alberta, as a signatory to the Canada Forest Accord in 1992, committed itself to work to implementing Sustainable Forest Management (SFM). It began the process by working through the Canadian Council of Forest Ministers to define SFM through Criteria and Indicators, published in 1995(CCFM 1995) and through the development of the Alberta Forest Conservation Strategy (AFCS), recently forwarded to the Minister of Environmental Protection (May 1997). In 1996 it created the Alberta Forest Management Science Council (AFMSC) through the Minister for Environmental Protection. The group has reviewed a number of Forest Management issues. In the fall of 1996 the Director of Forest Management requested:

"That the Alberta Forest Management Science Council provide advice to the Land and Forest Service on Timber Supply Protocols, particularly the science base required to change from Sustained Yield Management to Sustainable Forest Management. I request that the Council develop:

A clear, science-based definition of Sustainable Forest Management for application in Alberta that would meet our obligation to the Canada Forest Accord, incorporate the principles, goal and vision of the current Alberta Forest Conservation Strategy and that would be recognized internationally.

The elements of a Timber Supply Protocol that would meet the definition and be applicable to Alberta's forests.

An assessment of the science-base required to change from our current sustained yield forest management"

COUNCIL RESPONSE

The following response of the AFMSC was drafted in September 1997, following a year of review of documents, presentations and discussions. The response comprises a definition of SFM and description of five important elements.

The elements described provide government a science-base for a new timber supply protocol that conforms with the definition of Sustainable Forest Management.

The Council recognizes that uncertainty exists in terms of the science base of ecological integrity as well as social and economic values. Uncertainty also exists in their forecast. The elements of the protocol provide a framework under which uncertainty can be incorporated into forest management.

A. DEFINITION OF SUSTAINABLE FOREST MANAGEMENT

The Alberta Forest Management Science Council defines Sustainable Forest Management as:

" The maintenance of the ecological integrity of the forest ecosystem while providing for social and economic values such as ecosystem services, economic, social and cultural opportunities for the benefit of present and future generations"(1).

B. ELEMENTS OF SUSTAINABLE FOREST MANAGEMENT

The Council regards the maintenance of ecosystem integrity as an essential goal for the sound and sustainable management of Alberta's forests. Knowledge about ecosystem processes, including disturbance ecology and resilience to disturbance, provides the necessary context to identify a Desired Future Forest based on a range of possible scenarios. Social and economic values may be identified, quantified, and addressed through public involvement processes. Analysis and planning to reconcile social and economic values with ecological realities must occur at a variety of temporal and spatial scales. Our ability to achieve the desired future forest is constrained by inherent uncertainties within ecological and human systems as well as our limited understanding; therefore, an adaptive management approach is required for the sustainable future of Alberta's forests.

1. Ecological Integrity And Inherent Disturbance

THE CONSERVATION OF ECOLOGICAL INTEGRITY OF THE FOREST IS A NECESSARY CONDITION FOR THE SOUND AND SUSTAINABLE MANAGEMENT OF THE FOREST.

Ecological management of the forest develops and applies understanding of how forest ecosystems sustain themselves over long periods of time. It involves examination of growth, development, and the inherent disturbances that underlie the ecological integrity, dynamics, biological diversity and resilience of forest ecosystems. The knowledge enables managers to develop approaches that work with, rather than against, the processes that underlie forest ecosystem sustainability.

1.1 Rationale

1.1.1 Inherent Disturbance

Alberta's forests have developed since the retreat of the glaciers. The plants, animals and micro-organisms that comprise Alberta's forest ecosystems have adapted within environments characterized by recurrent and often severe disturbances. Inherent sources of disturbance include wildfire, wind, floods, climate change, mass wasting, insects, disease and indigenous human use. 

Moreover, the current level of human activity affects inherent disturbance processes, leading to conditions in forest structure, composition and landscape pattern that in some cases lie outside the typical range of variation of the system. For example, fire suppression during the last 40 years is believed to have changed the rate of fire disturbance of Alberta's forests (Murphy 1985). Understanding the relationships between disturbance types and managing to maintain ecological processes within the typical range of variation (Haila et al. 1994, Haila 1994), are essential to the long-term sustainability of Alberta's diverse forest ecosystems.

Changes to Alberta's forest landscape are not limited to those activities associated with inherent disturbance. Further changes are being wrought by agricultural clearing, urban expansion, livestock grazing, oil and gas development, mining, road construction, introduction of exotic species, climate change and airborne contaminants. The rate of change is fuelled by Alberta's human population growth - an increase of 50-fold in the last 100 years - and the accompanying escalating increases in consumption of resources.
 
1.1.2 Ecological Management 

An ecological approach aims to maintain forest structures, patterns, diversity and processes within the range of variation characteristic of the inherent disturbance regime of the forest. Science has key roles to develop understanding of the disturbance regime of Alberta's forest regions, and to identify the range of variation of key processes within which a variety of management objectives may be pursued.

Ecological management enlarges the focus of management from sustaining yield of specific tree fibre to the long-term maintenance of the system's ecological integrity and productivity. An ecological approach includes the following objectives:

• to maintain key ecosystem processes characteristic of the forest 
• to conserve native biodiversity characteristic of the forest 
• to manage human disturbance to maintain ecological patterns and processes within their typical range of variation.
• to manage fire disturbance within an acceptable level of risk to human life and property.

1.2 Implications

• Sustainable management includes a comprehensive approach to disturbance that strives to maintain the ecological integrity of the forest while effectively managing risk to human life, property and values, including timber supply.
• In ecological management, patterns and ranges of inherent disturbances, both terrestrial and aquatic, may provide a conceptual model for management strategies. For example, current practices such as two pass harvesting and reforestation can be modified to more closely approximate the change to forests resulting from fire and forest succession.
• Although fire is part of the forest, wildfires pose threats to human life and property, especially in the urban interface. Ecological management will require a broader approach to fire management including landscape solutions. For example, harvesting schedules need to be arranged to remove high risk stands first or spatially arrange harvests to reduce risk of loss.
• Management for ecological integrity requires an understanding of forest change with age and site characteristics, particular to each forest region. Management systems such--as tenure, planning and operations would be designed to maintain ecological integrity. e.g., the Boreal Mixedwood would be managed as a mixedwood forest, not separate deciduous and coniferous allotments.

1.3 Science Requirements

• Research is underway in Alberta in both aquatic and terrestrial systems that compares ecosystem structure and process resulting from harvesting activities to inherent disturbances and subsequent renewal. The research should be continued and relevant results implemented.
• Knowledge about the inherent disturbance regime, such as wildfire, should be augmented so that a better understanding of its importance in the maintenance of ecological integrity is achieved. Surrogate approaches to replicate desired disturbance effects are needed for application in areas where the inherent disturbance regime must be changed.
• Since most forest stands originated as a result of fire, post-fire conditions and long-term successional patterns require study to gain insights into design of forest harvesting and silvicultural treatments. These are required to approximate, where appropriate, the effects of fire that most contribute to forest renewal and maintenance of biological diversity.
• Effective ways of monitoring ecological integrity, including the conservation of biodiversity, should be developed at a provincial level and applied.
• The science base of whole landscape management that maintains ecological integrity while allowing management for various human values (including social, economic and ecological values) needs further development. 

2. Desired Future Forest 

DEFINING A VISION OF A DESIRED FUTURE FOREST IN ALBERTA IS A NECESSARY STEP IN IMPLEMENTING A MORE SUSTAINABLE FOREST MANAGEMENT PROGRAM. 

The characteristics of a desired future forest are determined through the identification of social, economic and ecological values to be sustained. The characteristics of a desired future forest are then forecast from the existing forest (including current commitments) and an understanding of the processes under which it has evolved. The forecast includes trends in human use and changes in other elements of disturbance such as wildfire and predicts outcomes in term of forest structure, composition, ecosystem flows and benefits. The forecast includes transition flows and states from the existing to the desired future forest.

The timber supply, other resource uses and values such as quality of land, air and water, are determined within the context of the desired future forest.

Forest plans need to include an implementation section that clearly outlines the activities to be undertaken, their cost and the committed funding sources.

2.1 Rationale

A desired future forest represents a dynamic forecast selected from a range of possible outcomes. This process differs from the current practice of static zoning and prescriptive rules.

Forecasts allow for the determination of flows of benefits such as ecosystem services, timber, access, habitat, biodiversity and other values. These flows can be assessed on their economic, ecological and social impacts. Forecasts can test the sensitivity of benefit flows to various assumptions and constraints and assess the cumulative impact of multiple resource use.

A desired future forest forecast provides mechanisms to integrate various intensities of human use (including landscapes protected from certain types of use, lands used extensively, lands used intensively and lands used for facilities) with overall ecological integrity.

A future forest goal allows updating of the forecast as better data become available and social objectives change. The specification of a desired future forest allows for a better assessment of success or failure.

2.2 Implications

• In the selection of the desired future forest scenario, uses of the landbase such as oil and gas, parks, agriculture, aboriginal use and recreation would be factors. Local community representatives and the general public would be involved in the determination of the selection. 
• The desired future forest defines timber supply and provides a basis for evaluation of the sustainability of forest values. 
• Fundamental issues in planning of a future forest are: i) The values sought in the forest must be explicitly described and measured. Caution should be taken not to-let this focus on measurement systematically exclude important values that are difficult to measure. ii) The values sought in the forest should be functionally related to forest characteristics. For example each value should be related to stand type, stage of stand development, and/or configuration within a landscape such as a watershed. 
• A desired future forest would be feasible to attain biologically, economically and socially starting from the present. Public presentations must clearly present various future forest states, the costs and benefits of these states and the paths taken to get there. 
• Provincial forecasting has not had explicit linkages to the local level. In the selection process, provincial forecasts need to be spatially explicit and, should allow for aggregation from local levels. 
• The achievement of a desired future forest requires the integration of management principles and sustainability ideas from many disciplines but, in particular, from ecology and economics. 
• Industry will be challenged to become more innovative in attaining forest objectives. As a result the provincial regulatory agency could then become less prescriptive.

2.3 Science Requirements

• Forecasts require specific quantitative assumptions about the development of individual forest inventory polygons (e.g. ecological processes, yield, risk of loss, stand structure, species composition etc...). Forecasting requires critical analysis and rigorous data collection.
• Timber management focuses on easily measured and well understood features of stand type, stage of stand development and geographic pattern. By extension, if timber management influences the availability of non-timber values such as recreation or carbon storage, those values are also related to type/ stage / pattern. As a result, an emphasis should be placed on the establishment of functional relationships between non-timber values and the pattern of types/stages in the forest.
• Emphasis needs to be placed on better forest characterization. This is the means by which the forest is described using attributes from which the state of forest values can be inferred and the effect of management actions can be expressed. To enable the manager to prescribe actions that result in a forest condition which provides the desired values of the owner, it is essential that the values be defined in terms of appropriate (and obtainable) forest characteristics and that effects of management actions be defined in terms of the same forest characteristics. 
• The scientific basis and protocol for areas of intensive forest management, such as increasing wood yields, are required.
• The development of forest level models that recognize both temporal and spatial change are essential. These models must be designed to be used to readily inform the public.
• Population demographics (urban, rural, aboriginal etc...) along with changing consumptive demands need to be developed and incorporated into the forecast.
• The development of new and/or the modification of current computer simulation models for the Alberta forests are needed.
• Balancing the functional relationship between human values, their expression in quality and intensity of use and the maintenance of ecological integrity is the essence of defining the desired future forest and the transition required from the existing forest state and benefit flows. As an example, any enhancement of timber supply through more intensive forest management (public or private) must be balanced between demand for timber and the maintenance of ecological integrity. Balancing models and mechanisms require further definition.

3. Social and Economic Values and Public Involvement

SOCIAL AND ECONOMIC VALUES ARE INTEGRAL TO THE SELECTION AND ATTAINMENT OF THE DESIRED FUTURE FOREST.

Timber and non-timber values, activity levels and existing commitments of the landbase should be quantified and integrated into the projection of the future forest condition and flows using scientific methods to determine market and non-market conditions, trends and the evolution of these trends.

3.1 Rationale

Timber values, and the impact of alternative forest management approaches on these values, can be quantified using economic assessment methods. These methods summarize a form of public interest in forest management; interest reflected by participation in the market for forest products.

Some non-timber values, including some values of ecosystem services, are quantifiable using established scientific techniques and data collection methods that rely on structured public input. Other non-timber values are not easily quantified and are not easily integrated into quantitative assessments of landscape conditions. Public involvement should be the key component in identifying, analysing and weighing these non-timber values in context with quantifiable timber and non-timber values.

Since forest resources in Alberta are predominantly owned by the Crown in the right of Alberta, involvement of the public is needed to evaluate the benefits arising from these resources and in the setting of goals and objectives for the management of multiple benefits. Public involvement is a recognized component of Sustainable Forest Management as described by the Canadian Council of Forest Ministers, the Alberta Forest Conservation Strategy, and the principles set out by most of the major Forest Certification schemes worldwide. However, it should be recognized that there are several levels of "public" interest in forest management, ranging from interest by consumers of forest products worldwide, to members of the general public within the province who may or may not be users of the forest landbase, to members of local communities who are commonly users of the forest resource base and are interested in the environmental and economic impacts of forestry on local social and ecological conditions. Information from the various "scales" of the public is needed for a more complete view of public demands on forest lands, however, defining and eliciting information From these diverse groups is scientifically challenging.

Non-timber value assessments can be used to establish goals in the future forest evaluation process. In order to be effective, however, these non-timber valuation efforts must be conducted in a trade-off or compensatory analysis framework in order to reveal the relative merits of one future forest scenario over another. Furthermore, these approaches should make explicit any existing responsibilities, rights and commitments and the degree of latitude in these initial conditions so that the process focuses on problems with realistic scope and does not generate unrealistic expectations. Finally, input from the various scales of publics must be integrated and balanced.

3.2 Implications to Timber Supply and Other Factors

• The explicit incorporation of social and economic values into the definition of the future forest may have little effect on the timber supply in the short term because of existing commitments; however, over the medium term (5-20 years) other values may take precedence over timber values or enhance the value of timber. 
• The integration of social and economic values, at various levels, will aid in the assessment of timber supply levels. Timber supply and associated economic returns are only some of the benefits generated by the forest.

3.3 Scientific Base Required

• At a global level, human values and choice can be viewed as the marketplace or economic consequence of alternative decisions by consumers and firms. Improved knowledge is needed of the market opportunities for sale of products, including trends in the demand, supply and product substitution, as well as the linkage between environmental performance and market access via certification. The market for products is seldom viewed as public involvement, yet it is one of the most clearly defined forms of public participation in forest resource use decisions. Furthermore, if certification schemes operate effectively, the market will provide individuals worldwide with the opportunity to express their preferences for environmental aspects of the products as well as the material aspects.
• At the federal, provincial and local level the legitimacy of the democratic process of governance must be recognized as an important element of the valuation and public involvement process. Public involvement is a critical element for the collection of information on non-timber values, particularly those related to activities in the forest and to ecosystem services. Additional science is required to further integrate non-timber activity levels with forest or ecosystem service flows and to improve the predictive nature of the models of these activities. A significant gap in this area of study is the inventory of activities and uses in a spatial and temporal context. Improvement of this inventory will require additional efforts in survey research, often in a variety of cultural contexts aimed at different components of the public, to capture the fundamental elements of non-timber related activities. This is critical in the case of northern residents of Alberta, e.g., Aboriginal peoples.
• The assessment of the preferences of the public for alternative future forest conditions and value flows, - scenario analysis- must be reached through public participation. For the broader public, assessment of preferences can be based on science techniques for preference elicitation (e.g. Mitchell and Carson, 1989; Keeney and Raiffa,1976) within a realistic framework of current resource use and ecological integrity. At the local level similar techniques can be employed. An alternative approach relies on the use of group discussion conflict resolution techniques (e.g. Von winterfeldt and Edwards, 1986; McDaniels, 1996). Public involvement should be designed to be inclusive, transparent in terms of its role in decision making, have the potential for revisiting the issue on a formalized time scale (e.g. every 5 years), be oriented towards achieving public objectives rather than satisfying consistent demands, be integrated with public education to increase understanding, awareness and trust levels between all parties involved, and be flexible within the cultural context (Murphy, 1996; Leiss and Chociolko 1994).
• The use of future forest scenarios forms a solid basis for the assessment of social and economic value through public involvement. This approach provides a scientific assessment of the implications of the alternative scenarios. Such an approach also serves as an education mechanism and provides public feedback on the scientific assessments. However, efforts should be made to separate values from technical aspects and recognize values when they are presented as part of the future forest scenario.
• Adaptive management is associated with ecological management (AFCS 1997). It should be employed in social systems as well. New public involvement approaches, or development of new social institutions for the allocation of resources, could be tested on a regional basis and can become an explicit part of resource management designed to investigate the efficacy of new mechanisms.

4. Scales-Spatial and Temporal

THE TEMPORAL AND SPATIAL SCALES USED TO MANAGE THE FOREST MUST BE CONSISTENT WITH THE SCALES OF DISTURBANCES AND PROCESSES INHERENT TO THE FOREST AND SOCIAL SCALES RELEVANT TO FOREST RESOURCE USE.

4.1 Rationale

Ecological processes occur on several spatial and temporal scales. In the boreal forest, processes may occur on spatial scales from 1 cm (e.g. photosynthesis within a leaf) to hundreds of km (e.g. large fires on landscapes). Temporal scales may vary from months (e.g. seasonal changes in weather) to thousands of years (e.g. geological processes) (Holling 1992). Elements of biodiversity (composition, structure and function) can also be organized in a hierarchy of scales: genetic, population/species, community/ecosystem, and regional/landscape (Noss 1990). Ecological organization at the highest levels has unique properties that cannot be explained by simply "scaling up" processes at lower scales (Alien and Starr 1982).

In planning the future forest, relevant spatial scales will be multiple and might include within-stand habitats (-1-100m2), stands or non-forested units (e.g. peatlands) (-10-100ha), landscape units (>10,000ha), and forest regions (e.g. boreal mixedwood forest). Landscape units follow ecological boundaries and contain major ecological Rows and elements of interactions and will likely be different for aquatic and terrestrial systems (e.g. watershed, landform, fire regime). Within-stand habitats are unique or localized habitats which support unique or rare species. Stand management may include retaining trees to provide structure and downed woody material. Landscape management would approximate inherent patterns of stand sizes, amount of edge, juxtaposition of different stand types, landscape connectedness and other landscape metrics. Landscape management would include the total forest landscape including protected areas.

Temporal scales used in planning the future forest landscape should mesh ecologically relevant scales (seasonal and disturbance intervals e.g. wildfire, floods) with those meaningful for human planning. They may span 1-5 yr for short-term planning up to 3-4 fire return intervals for longer term planning. Seasonal planning would include scheduling potentially disruptive activities in the forest to periods with the lowest impact. Longer term planning would include variation in rotation ages of stands to approximate inherent age class patterns and to account for different successional patterns with site type. Deterministic models must incorporate output that includes stochasticity of events (flood, fire) so that risk can be incorporated into planning.

Social processes such as public involvement also occur at several scales. At the global scale, markets for forest products influence the demand for forest products. Shifts in global markets will have significant impacts on resource requirements and the economic viability of operations in Alberta. These global level effects will in turn affect national, provincial and local levels economically. Conversely, changes made at the local level may, through international trade linkages, have global repercussions (Sedjo, 1995). A second social scale is the people of Alberta. Alberta's citizens, while not all users of forest lands directly, have preferences for broad strategic objectives of forest resource use such as more recreational use or more specific values such as biodiversity, wildlife conservation, economic diversity and other aspects of forest ecosystem services and products. Direct users of the forest, including local communities and Aboriginal groups, form another social scale where concerns may be for local interests, including economic development, as well as for the broader strategic objectives. Effective forest resource management will employ information from all of these scales as well as the linkages between these scales.

Related to the various social scales are the temporal dimensions associated with social values. Demographic trends, technological change and continuing human development will generate changing demands for forest ecosystem services and forest products. Evidence suggests that in the developed world, the value of non-timber goods and services will increase, partly due to the relative scarcity of these goods and partly due to changing preferences of individuals in society. Furthermore, time scales for industrial aspects of forest product production, associated with international capital markets and global competition, are substantially different than those for ecosystems.

4.2 Implications

• Through an understanding of the spatial and temporal scales of natural systems dynamics, conservation of biodiversity will be realized more readily.
• Through an understanding of spatial and temporal scales, the public will better understand the basis for a sustainable forest.
• At the provincial scale, management will cross tenure boundaries, and should involve coordination and cooperation of the government, citizens of Alberta-FMA and quota holders, woodlot owners and other users of the forest.
• The province should facilitate planning and management at the landscape unit and larger spatial scales, provide for integration at the provincial level.
• Models for forest values must be spatially and temporally explicit and multiscale.
• Impacts of resource users should be considered at relevant scales and be integrated with forest management.
• The cumulative effects of management decisions, land use and inherent disturbance over space and time need to be addressed.
• Trends and time scales, as well as continued scientific efforts to forecast and understand trends are required for accurate depiction of the future forest condition and the path towards this future forest.

4.3 Science Base Required

• Planning tools, such as spatially explicit resource supply and landscape models, must be developed for large-scale (landscape and above) management.
• The scientific basis for ecologically and socially relevant landscape units for planning (e.g. define fire patterns, watersheds, boundaries of ecological flows) and for tenure should be established.
• The impact on users (forestry, oil and gas, parks and others) of any change to management units should be assessed.
• Not all variables can be aggregated from the small to the large scale. Research and guidelines must be developed to determine which variables can be scaled up or down and what transformations are necessary.

5. Adaptive Management

ADAPTIVE MANAGEMENT MONITORS PROGRESS TOWARDS THE DESIRED FUTURE FOREST, CONTINUALLY IMPROVES THE KNOWLEDGE BASE AND ADJUSTS ACTIONS TO CORRECT FOR DEVIATIONS

Adaptive management is a process of hypothesis testing at the scale of whole systems. It continually evaluates and adjusts management relative to predicted responses, objectives and predetermined thresholds of acceptable change.

Adaptive management includes improvement of the data and analyses on which forest management predictions are based, and testing of the assumptions underlying the management practices carried out on forest lands.

5.1 Rationale

The process of implementing a land management plan is fraught with unknowns, stemming both from our imperfect knowledge and from the inherent unpredictability of nature. Although the best available information may be used to predict responses, many uncertainties remain. The unknowns will be greatest where management includes new or seldom-applied management options in response to a broader array of social and ecological objectives. Furthermore, treatments that were found, workable within the controlled environment of experimental plots may require testing in operational settings. Adaptive management requires that scientists and managers in collaboration, establish a framework for scientific testing of the concepts, methods, and assumptions applied to the land.

The adaptive management framework has two components. First, there is the need to design and apply specific prescriptions as experiments, including suitable controls and replications. The design of these experiments is adjusted on the basis of the ongoing analysis of results. These experiments may include extreme applications of a proposed approach - similar to the concept of "testing to failure" used by engineers. In other words, to learn rapidly we may need to try the extreme versions of management practices that we conjecture are "correct".

The second component of adaptive management addresses the cumulative effects of management in space and time. These are evaluations of whether the system overall is responding as predicted, and whether the management appears to be on a path leading to the desired outcomes specified in the plan. In other words, adaptive management includes ongoing evaluation of whether the various actions and practices that are applied to the land appear to be "adding up" to the desired future conditions and outcomes. Key elements in these evaluations include monitoring of selected indicators, identification of ranges of variation and/or thresholds for those indicators, and feedback mechanisms for adjusting the assumptions, models, or management practices that account for the deviations. For example, a key question in ecosystem-based management is whether habitat and other habitat elements will 'add up" to landscape conditions capable of maintaining biological diversity. Monitoring individual treatments will not answer this whole-system question. Instead, indicators will be required to evaluate the cumulative effects of these treatment strategies in space and time.

Adaptive management requires the establishment of reference areas to allow interpretation of research and monitoring results. The functions of these reference areas are twofold:

1. Reference areas are required as spatial controls for the experiments carried out in adaptive management. This requires reference areas that are representative of the experimental area in terms of stand age classes, species composition of plants and animals, and spatial layout and size of stands. 
2. Another function of reference areas is to measure background change that is occurring regionally or globally; for example, changes in air quality or plant and animal populations that may be associated with global atmospheric or temperature conditions. In adaptive management, the purpose of monitoring is to measure change and effects that result from the management under evaluation. It is essential to be able to distinguish the effects associated with management from background change that emanates from other sources.

5.2 Implications

• All projections of forest management outcomes, including timber supply, are tentative.
• Managers must acknowledge the limitations of our understanding of forest ecosystems, and strive to reduce uncertainty through monitoring and research.
• Adaptive management and large-scale research require close collaboration of scientists and managers.
• The cost of monitoring and research must be accepted as an integral cost of the business of forest management
• There must be feedback mechanisms that link the results of monitoring and research to the forest policy arena (Lee 1993).
• Adaptive management will require greater flexibility in our social systems.

5.3 Science Requirements

• Changing the focus from individual uses to a system perspective requires an interdisciplinary research environment that encourages interaction and cooperation of scientists from diverse disciplines and areas of specialization.
• Knowledge to support forest management must derive from two kinds of science: the science of parts, emerging from traditions of reductionism; and science of the integration of parts, which derives from a whole-system perspective (Waiters and Holling 1990).
• Alternative statistical approaches suited for large-scale experiments, both deliberate and unplanned (i.e. retrospective studies), need to be applied in adaptive forest management (see, for example, Likens 1985, Schindler 1987, and Carpenter et al. 1989).
• Evaluating how ecosystems and associated values and properties (such as biological diversity and resilience) may respond to the collective, cumulative, and long-term effects of management requires scientific examination at multiple (including large) scales, and from a whole-system perspective.
• The scientific basis and protocol for the establishment of reference areas to assess sustainable forest management is required.

C. CONCLUSION

Achievement of Sustainable Forest Management is a complex process, but may be approached more systematically by considering the five elements that have been described here: Ecological Integrity and Inherent Disturbance, Desired Future Forest, Social and Economic Values and Public Involvement, Scales - Spatial and Temporal, and Adaptive Management. While these are consistent with the definition of Sustainable Forest Management, they are qualitatively different from any other timber supply protocol in Canada.

The Council's definition of Sustainable Forest Management is consistent with but not the same as the definitions of the Canadian Council of Forest Ministers or the Alberta Forest Conservation Strategy. The Council has changed the definition so that a clear connection between ecological integrity and social and economic values can be made. The Council advises the Land and Forest Service to raise this difference to the Minister of Environmental Protection and the Minister of Natural Resources Canada to ensure that this protocol is consistent with Alberta's and Canada's approach to Sustainable Forest Management nationally and internationally.

It is not clear to the Council that the scope of what is proposed in this document is feasible to undertake. Particularly in the area of non-timber values and their linkage to forest characteristics at various scales, the costs and benefits need to be determined and applied through a pilot process. Another important feasibility issue is "testing the system to failure" as proposed in the Adaptive Management element. Also, the criteria for reference areas needs to be clarified.

Finally, a trade-off analysis framework (Section 3) as between timber and non-timber values needs to be developed and tested in a public format.

This work has been undertaken by the Council over the last year. We have discussed sustainable forest management with some of Alberta's leading forest companies as well as other provincial government. The Council met with Dr. Jack Ward Thomas, former Chief of the U.S. Forest Service and has had the document reviewed by the following people: Bob Andrews, Director, Wildlife Alberta Environmental Protection; Dr. Gordon Baskerville, Professor of Forestry, UBC; Dr. Tom Beckley, Canadian Forest Service; Ms. Lea Bill, Community Health, Sisksika Nation; Dr. Clark Binkley, Dean of Forestry, UBC; Dr. Bill Fuller, Professor Emeritus, UofA; Dr. Daryll Hebert; Alberta Pacific Industries; Mr. George Weyerhaeuser, Mr. Bruce MacMillan, Dr. Luigi Morgantini, Weyerhaeuser Canada; Mr. David Neave, Executive Director, Wildlife Habitat Canada; Dr. Brad Stelfox, Forem Consulting; Dr. Jack Ward Thomas, Boone & Crockett Wildlife Professor; Dr. Terry Veeman, Professor Rural Economy, UofA.

In the view of the Council, this protocol, should it prove feasible, will provide the province of Alberta with a science-based timber supply protocol that is consistent with sustainable forest management.

D. REFERENCES

Alberta Forest Conservation Strategy 1997. Information Centre, Alberta Environmental Protection, Edmonton, Alberta.

Alien, T.F.H. and T.B. Starr 1982. Hierarchy: perspectives for ecological complexity. University of Chicago Press, Chicago, ILL 310 pp.

Carpenter, S.R., T.M. Frost, D. Heisey, and T.K. Kratz 1989. Randomized intervention analysis and the interpretation of whole-ecosystem experiments. Ecology 70:1142-1152.

CCFM 1995. Defining Sustainable Forest Management: A Canadian Approach to Criteria and Indicators and Natural Resources Canada, Ottawa, Ontario.

Haila, Y., I.K. Hanski, J. Niemela, P. Punttila, S. Raivo and H. Tukia 1994. Forestry and the boreal fauna: matching management with natural forest dynamics. Ann. Zool. Fennici 31:187-202. Helsinki, 31 January 1994.

Haila, Y. 1994. Preserving ecological diversity in boreal forest: ecological background, research and management. Ann. Zool. Fennici 31: 203-217. Helsinki, 31 January 1994.

Holling, C.S. 1992. Cross-scale morphology, geometry, and dynamics of ecosystems. Ecol. Monogr. 62: 447-502.

Keeney, R. and H. Raiffa 1976. Decisions with Multiple Objectives: Preferences and Value Tradeoffs. John Wiley. New York.

Leiss, W. And C. Chociolko 1994. Risk and Responsibility. McGill-Queen's University Press, Montreal and Kingston.

Lee, K.N. 1993. Compass and Gyroscope. Island Press Washington, D.C.

Likens, G.E. 1985. An experimental approach for the study of ecosystems. Journal of Ecology 73: 381- 396.

McDaniels, T.L. 1996. The Structured Value Referendum: Eliciting Preferences for Environmental Policy Alternatives. Journal of Policy Analysis and Management. 15:227-251.

Mitchell, R.C. and R.T. Carson 1989. Using Surveys to Value Public Goods. Resources for the Future Press. Washington, D.C.

Murphy, P.J. 1985. Methods for Evaluating the Effects of Forest Fire Management in Alberta. PhD Thesis. University of British Columbia.

Murphy, P. 1996. Consensus-Building and Sustainable Forest Management in Canada. Paper presented at the Intergovernmental Workshop on Sustainable Forestry and Land Use: The Process of Consensus Building, Stockholm, Sweden.

Noss, R.F. 1990. Indicators for monitoring biodiversity: a hierarchical approach. Cons. Biol. 4:355-364.

Schindler, D.W. 1987. Detecting ecosystem response to anthropogenic stress. Canadian Journal of Fisheries and Aquatic Sciences 44 (Supplement):6-25.

Sedjo R. 1995 Local timber production and global trade: the environmental implications of forestry trade, p. 49-67, in Adamowicz, W.L., P.C. Boxall, M.K. Luchert, W.E. Phillips and W.A. White. (Eds). Forestry, Economics and the Environment. CAB International, Oxon, U.K. 286 pp.

Von Winterfeldt, D. and W. Edwards 1986. Decision Analysis and Behavioral Research. Cambridge University Press, Cambridge.

Walters C.J. and C.S. i-lolling 1990. Large-scale management experiments and learning by doing Ecology 71:2060-2068.

Footnotes

(1)The current definition as adopted does not include "for the benefit of all living things, locally, provincially, nationally and globally" from the AFCS and CCFM definition because what may be to the benefit of one living thing may be to the detriment of another.

"Long term health of forest ecosystems" as described in the CCFM definition puts too human a focus on forest ecosystems. "The maintenance of ecological integrity of the forest ecosystem" focuses more appropriately on the processes, flows, structures and range of variation of forest ecosystems.

"Environmental opportunities" as stated in the AFCS and CCFM definition is unclear to the Council. Ecosystem services, that provide outputs such as clear water and air, are clearly beneficial to present and future generations.