Applying Adaptive Management in British Columbia's Forests

J. B. Nyberg and B. Taylor

British Columbia Ministry of Forests,
Victoria, British Columbia, Canada

Reference: Nyberg, J.B. and B. S. Taylor. 1995. Applying adaptive management in British Columbia's forests.

In Proceedings of the FAO/ECE/ILO International Forestry Seminar, Prince George, BC, September 9-15, 1995, pp. 239-245. Canadian Forest Service.

Abstract
Adaptive management offers forest managers the potential to learn rapidly from the results of operational policies and practices as they are being implemented, and thus to keep pace with the rapidly changing demands of industrial and public clients. Despite published examples of the successes of adaptive management in other fields, it has seldom been applied rigorously to forestry issues. The BC Forest Service is now developing a province-wide initiative that aims to establish adaptive management as a standard approach to situations in which the optimal policies or practices are uncertain.

We define adaptive management as a formal process for continually improving management by learning from the outcomes of operational plans. Critical steps in the process include:

1. acknowledging uncertainty about what policy or practice is "best" for the particular management issue,
2. thoughtfully selecting the policies or practices to be applied,
3. carefully implementing the plan of action,
4. monitoring key response indicators,
5. analyzing the outcome in light of the original objectives, and
6. incorporating the results into future decisions.

So-called "active" adaptive management, in which alternative policies or practices are compared through operational experiments, is the most powerful variation of the concept.

Adaptive management can be an important supplement to forest research programs, especially where demands for change do not allow the luxury of intensive, process-level research before starting widespread implementation of new approaches. Current issues in British Columbia that have been identified as high priorities for testing through adaptive management include harvesting techniques and silvicultural systems that provide alternatives to conventional clear-cutting, methods for protecting riparian habitat and streams, landscape and stand-scale practices for maintaining biological diversity and sensitive wildlife values, and watershed restoration techniques. A few projects which demonstrate the application of adaptive management to wildlife, timber, and vegetation competition issues are already underway in the province.

Introduction
Resource managers traditionally rely on two sources for guidance: personal experience gained through trial and error by themselves and others who work or have worked in the field, and research results from scientific studies. These sources of knowledge may be more or less adequate to support successful management programs when social and environmental conditions are stable, but they often fall short when new objectives or field conditions arise. In such circumstances, managers faced with difficult decisions must push beyond their base of reliable knowledge to make interpretations, extrapolations, or even guesses as to what the best policies or practices may be. Often they call on scientists for help, but the required research may be years in coming or may not be possible at all due to restricted budgets or other limitations. In the interim, with no other options available but the usually unacceptable one of no action, managers must proceed on an often tortuous path of trial and error.

The process of adaptive management (Walters 1986) promises a more efficient means of gaining critical knowledge for resource management. During the first day's sessions at this seminar, Clark Binkley and Winifred Kessler both called for adaptive management to be applied in forestry; and Professor Heinimann called for codes of forest practice that embody a cycle of continuous improvement or "self- learning", a concept very similar to that of adaptive management. It is clearly an idea that has wide acceptance. As Dr. Binkley correctly pointed out, however, there are few examples of the successful application of adaptive management in forestry, at least at any scale higher than that of the forest stand, anywhere in the world.

Recently the British Columbia Forest Service (Ministry of Forests) initiated a review of the principles and potential of adaptive management, with the intent of applying the approach widely throughout the province's public forests should the review suggest this is feasible. In this paper, we report some of the observations and conclusions we have drawn during the early stages of the Forest Service initiative. We define and describe adaptive management in the context of forest management, evaluate its advantages and the impediments to its use, and discuss its application to several questions of forest policy and practice in British Columbia. We conclude that adaptive management is an approach that would be valuable wherever forestry is practised.

Definition and Characteristics of Adaptive Management
Since the concept was first developed and described during the 1970's by members of the Institute of Animal Resource Ecology at the University of British Columbia (e.g., Holling 1978, Walters and Hilborn 1978), the term adaptive management has been used, and sometimes misused, in a variety of ways. In simple terms it is a strategy for generating reliable information--that is, for learning--from resource management actions. It recognises that many policies and practices are essentially experiments, from which resource managers can and should derive information that will allow them to manage more effectively in future. Management is viewed as a continuous learning cycle.

While the concept of adaptive management is simple and easily grasped by most people (hence its popularity in the literature if not in practice), it should not be taken as merely a different means of describing the rather haphazard way that most people learn from their mistakes and successes. True adaptive management requires more planning, more careful measurements and analysis, and more comfort with change than managers typically bring to their daily responsibilities.

As we have found no concise and operationally meaningful description of adaptive management in the literature, we have developed the following working definition for use in communicating with Forest Service staff:

Adaptive management is a formal process for continually improving management policies and practices, by learning from the outcomes of operational programs. In all cases it requires:

1. acknowledging uncertainty about what policy or practice is the "best" for the particular management issue being addressed,
2. thoughtfully selecting the policies or practices to be applied,
3. carefully implementing the plan of action,
4. monitoring key response indicators,
5. analyzing the outcome in light of the original objectives, and
6. incorporating the results into future decisions.

For further information on the steps in adaptive management, readers should refer to the literature cited, especially Holling (1978), Walters (1986), Walters and Holling (1990), Lee (1993), and Taylor et al. (in prep.).

One issue that needs to be clarified is the distinction between scientific research and adaptive management. They share some common elements (e.g., hypothesis-testing, experimental design, data collection and analysis) and there are no clean boundaries between the two. In general terms, adaptive management usually differs from research in aspects such as who leads the projects (managers vs. scientists), the scale of application (routine operations in the field vs. precise experimentation in laboratories or small plots), the rigor of the design (adaptive management projects are often less tightly controlled and replicated than research projects), and the intensity of measurements (monitoring only of key response indicators vs. many researchers measuring many variables each).

Benefits of Adaptive Management
Most contentious issues surrounding forest management are ecologically, socially, and economically complex. In addition, many natural events are unpredictable, and understanding of forest ecosystems is often limited by scarce research and short histories of experience with management relative to rotation lengths. Together these factors result in substantial uncertainty about the outcomes of forestry decisions. Changes in social values and goals further increase uncertainty and contribute to controversy. In British Columbia and elsewhere in the world, many cases can be found of complex issues for which the best policy is uncertain. Local examples include issues such as how to restore watersheds damaged by past logging and road-building, how to apply partial cutting systems in cold mountain forests, and how to maintain biological diversity in managed forest landscapes.

Adaptive management offers many benefits in such situations. The explicit recognition of uncertainty that must begin the process can, if publicly communicated, serve to build trust and respect among potential critics and stakeholder groups. Many people today prefer to know that managers are not sure of the eventual results of their actions but are taking explicit measures to improve their knowledge base, instead of being told that "We know what we are doing and everything is under control--trust us". Careful analysis of policy options, ecological relations, management criteria, and interactions among these factors should be part of option selection and project design. This analysis can clarify what really needs to be known to guide decision-making and what must be measured, for how long, to determine success or failure. The operational scale of treatments under adaptive management can reveal interactions and emergent effects that would not be detected in the models, laboratories, or small experimental installations that typify most research studies.

Furthermore, managers can usually be more confident when they extrapolate results to new areas if those results come from widespread operational trials rather than small, isolated research installations. When managers commit to managing adaptively they commit to monitoring the results, over long periods of time in many cases; to analysing the resulting data; and to documenting the results. This increases the likelihood that the project leader's insights will be shared and communicated widely. These steps are sometimes neglected by busy field staff who conduct ad hoc trials but do not follow up because they are impatient to get on with the next job.

Most importantly, adaptive management allows managers to proceed systematically and responsibly in the face of inadequate information. When it is not possible to wait for the results of detailed research before acting, there is no other approach that can so efficiently and powerfully fill gaps in understanding.

Why Now, Why Here?
The concept of adaptive management is now more than 20 years old. It has been thoroughly described in the literature, promoted locally and internationally by those who conceived it, and embraced in principle by many individuals and agencies from Alaska to Australia. In British Columbia, we are home to several of the leading experts on the subject, and to a company (ESSA, Environmental and Social Systems Analysts) that has a world-wide reputation for providing expert services in adaptive environmental assessment. Yet to date, in British Columbia as elsewhere, it has not been implemented in a comprehensive way by any public or private resource agency.

This lack of effective implementation suggests that there are serious obstacles to adaptive management; we outline some of those obstacles in the next section. But it also seems to be an idea whose time has now come, at least in the Pacific Northwest region of North America, as both the US Forest Service (Bormann et al. 1994) and the BC Forest Service are embarking on adaptive management initiatives. What makes it so timely in 1995?

The answer for both the US and BC seems to lie in public demands for change in the ways forests are managed. In a few years (far too few for research or field experience to provide the required answers) these jurisdictions have experienced revolutionary changes in goals, objectives, and procedures that have many managers asking "How do I deal with all this?". Protection of water, soils, old-growth forests, ecosystems, wilderness, and biological diversity are suddenly a major part of everyone's job. In the US, this is most clearly illustrated by the President's Forest Plan for the Pacific Northwest, the culmination of work by various spotted owl and ecosystem management committees. In British Columbia the Protected Areas Strategy, the work of the Commission on Resources and Environment, and the Forest Practices Code have brought a whole new look to the business of forest resource protection and management. In both jurisdictions, management agencies have realised that both the new demands and the new measures with which the agencies responded to them have outstripped the reliable knowledge base, so that long- term results are uncertain.

This is clearly illustrated by certain aspects of the Forest Practices Code, the major catalyst for the new interest in adaptive management in British Columbia. Because the Code embodies so many new measures, especially for protection of non-timber values such as watersheds, riparian zones, and biological diversity, there is a crucial need to evaluate its effectiveness, to revise any ineffective components that may be identified, and to consider whether there are alternative approaches to achieving the Code's objectives. Scientific research efforts are expanding significantly, especially through the province's new Forest Renewal Plan, but there is no hope that all the questions and uncertainties about the Code can be answered by new research in a few short years. Adaptive management, if applied widely and in concert with appropriate scientific research, is now recognised as a potential solution to this information deficit.

Challenges to Implementation
If adaptive management is so powerful and so widely applicable, why has it not been more commonly used in the past? The answer to this question reveals some of the major obstacles that must be overcome if we in British Columbia are to succeed where other jurisdictions have failed. Only a brief overview is provided here; those interested in more details should consult Taylor et al. (in prep.).

Any agency adopting an adaptive management philosophy must be prepared to acknowledge publicly that it is uncertain about the results of at least some of its actions. This is often a difficult admission, especially for people who have been delegated great authority and power in the past. The agency must also explicitly allow for results that critics may subsequently call "mistakes". Some of the options in any set being tested will unavoidably prove less successful than others, but these "mistakes" are essential to learning. And, to reward the accomplishments of adaptive managers through career advancement, agencies must develop measures for performance evaluation that recognize risk-taking and allow for variation in program outputs as long as effective learning is promoted.

Because adaptive management requires more careful planning, implementation, and documentation than is often required for routine operations, more staff and money will be required to implement it widely. The attention and funding required for effective monitoring programs, field layout, data storage, and data analysis are often in short supply in agencies hard pressed to keep up with day-to-day tasks. These extra costs may, however, be much lower than would be incurred in conducting scientific research on the same area. Also, the expected value of the new knowledge derived will often outweigh the costs (e.g., McAllister and Peterman 1992).

Experimental or active adaptive management requires careful design, including replication of treatments and experimental controls. Additional training in scientific methods may thus be required for resource managers. When each replicate is at the scale of a watershed (e.g., a drainage basin of 100 km2 or more) it may be difficult to find sufficient similar experimental units. New statistical techniques may be required to make sense of the data from so-called "incomplete" designs and large-scale experiments. Managers must commit to following through on the selected treatments, often over a period of a decade or more, even when some clients demand that their special short-term interests should be satisfied instead.

These and other potential roadblocks to adaptive management must be removed before an initiative such as ours can succeed. It will make sense to try if the costs and effort required to manage adaptively are balanced by the increased efficiency and effectiveness of management that can be achieved, by the value of the new opportunities that may be revealed, and by reducing the risk and cost that would result from continuing to manage with unproved techniques.

Examples and Opportunities Because of the impetus for change that has been created, first by public criticism of past practices in British Columbia and now by the implementation of our Forest Practices Code, the climate has never been better for adaptive management. Many managers are already experimenting with new approaches, though usually not as thoroughly as adaptive management principles would require. There are, however, a few cases in which adaptive management is the stated goal of the work and the approach is more in keeping with the process described earlier, as in the first two examples below.

Watershed Restoration Program This operational program will spend tens of millions of dollars in each of the next several years to rehabilitate watersheds damaged by past logging and road-building practices. To evaluate its effectiveness in achieving the stated objectives of improving water quality and fish habitat, the program's proponents have designed a three-level monitoring program that employs a nested design and proposes deliberate experimentation at the watershed scale in many areas of the province. For background, see Keeley and Walters (1994).

Grizzly Bear Habitat and Silvicultural Practices in Coastal Valley

This project compares various gap sizes and thinning regimes in valley-bottom plantations on the

mainland coast of British Columbia. The goal is to determine which regimes will produce high-

quality timber while maintaining, throughout the rotation, understory vegetation that provides crucial forage for grizzly bears. The various regimes are being applied by resource managers to a variety of sites over a period of several years, and monitoring of results is underway (McLennan and Johnson 1993; T. Hamilton, personal communication).

Of several issues that have been queried or criticized during the development of the Forest Practices Code, the following are prime candidates for exploration through adaptive management.

Size and Management of Riparian Zones
Regulations under the Forest Practices Code prescribe minimum widths for no-harvest reserves of timber adjacent to various classes of streams (e.g., 20 m of reserved forest on each side of all streams 1.5-5 m wide, if those streams bear fish or supply licensed water users). Further restrictions on timber harvesting are applied to other areas, and again the width of these areas varies by stream class (e.g., a further 20 m of "management zone" is required adjacent to the reserve zone described previously). The sizes of these zones, their success in protecting fish habitat and water quality, and the harvesting rates and methods that can safely be applied in the management zone have all been questioned. To evaluate the effectiveness and e fficiency of these measures, a widely distributed series of field trials would be required so as to encompass the variability in site, vegetation, and climatic conditions that occurs in the province.

Landscape- and Stand-Scale Measures for Biodiversity Conservation
The BC Code contains provisions for reserves of older forest in many watersheds, so habitat can be maintained for organisms dependent on old-growth forest conditions. The recommended amounts, sizes, and patterns of these old-forest reserves were determined from ecological theory and from studies and experience in other areas of the world, but have not yet been proven effective in British Columbia. The specific recommendations for these reserves, as well as provisions for maintaining connections ("corridors") of old forest between the larger reserves, have been criticized. At the stand scale, retention of "wildlife trees" is recommended to provide habitat for cavity-nesting birds and other animals, but the appropriate amounts and distributions of these trees must be verified. All of these aspects of biodiversity conservation could be explored using adaptive management.

Adaptive management could also be applied to topics that are not environmental. For example, it could be just as valuable in improving recreation site management, or in assessing the harvesting efficiency of various types of logging machinery. Garner and Visher (1988, cited in Halbert 1993) describe a radically different example, in which Minneapolis police compared alternative approaches for effectiveness in reducing domestic violence.

Conclusions
We expect that the potential benefits of adaptive management will generate a significant effort by government resource agencies to apply the approach widely in British Columbia. This will come slowly, in part because successful implementation requires changes in individual perspectives of managers and in the culture of the organizations they work for. The pace of adoption of adaptive management will also be restricted by the huge demands already placed on field staff to implement the Forest Practices Code, apply new planning procedures, and implement the Forest Renewal Plan.

In the meantime, education and training of managers and stakeholders both within and outside government will be needed. Partnerships between government, industry, and other interests must be built. Scientific research will continue to be required to complement adaptive management, and will be particularly important for elucidating some of the functional aspects of forest ecosystems that may not be revealed by less intensive techniques.

It is unlikely that the next decade will bring a better time than the present to adopt adaptive management in British Columbia. The same may well be true in many other forestry jurisdictions around the world. We look forward to sharing the results of the British Columbian experience as our initiative proceeds.

References

Bormann, B.T., P.G. Cunningham, M.H. Brookes, V.W. Manning, and M.W. Collopy. 1994. Adaptive ecosystem management in the Pacific Northwest. USDA For. Serv. Gen. Tech. Rep. PNW-GTR-341. Portland, OR. p. 22

Garner, J.H. and C.A. Visher. 1988. Policy experiments come of age. Nation. Inst. Justice Rep. 211.

Halbert, C.L. How adaptive is adaptive management? Implementing adaptive management in Washington State and British Columbia. Rev. Fish. Sci. 1: 261-283.

Holling, C.S., ed. 1978. Adaptive Environmental Assessment and Management. John Wiley and Sons, Toronto. p. 377

Keeley, E.R. and C.J. Walters. 1994. The British Columbia Watershed Restoration Program: Summary of Experimental Design, Monitoring, and Restoration Techniques Workshop. Watershed Restor. Manage. Rep. 1. Univ. Brit. Col., Vancouver

Lee, K.N.1993. Compass and gyroscope: integrating science and politics for the environment. Island Press, Washington, DC. p. 243

McAllister, M.K. and R.M. Peterman. 1992. Decision analysis of a large-scale fishing experiment designed to test for a genetic effect of size-selective fishing on British Columbia pink salmon (Oncorhyncus gorbuscha). Can. J. Fish. Aqua. Sci. 49: 1305-1314.

McLennan, D.S. and T. Johnson. 1993. An adaptive management approach for integrating grizzly bear habitat requirements and silvicultural practices in coastal British Columbia: working plan. Brit. Col. Wildl. Branch, Victoria, BC.

Taylor, B., L.L. Kremsater, and R.M. Ellis. In preparation. Adaptive management of British Columbia's forests: discussion paper. Brit. Col. For. Serv., Victoria, BC.

Walters, C.J. 1986. Adaptive Management of Renewable Resources. Macmillan, New York. p. 374

Walters, C.J. and R. Hilborn. 1978. Ecological optimization and adaptive management. Ann. Rev. Ecol. Sys. 9:157-188.

Walters, C.J. and C.S. Holling. 1990. Large-scale experiments and learning by doing. Ecology 71: 2060-2068.

        Back to the top