Adaptive Management

The U.S. Department of the Interior

Technical Guide
Adaptive Management Working Group
Chair: Olivia Barton Ferriter
Director, Conservation, Partnerships & Management Policy

Lead Authors
Byron K. Williams, Robert C. Szaro, and Carl D. Shapiro

Other Contributors
Robert Adamcik, Mary Boatman, Sarah Bransom,
Jeremy Casterson, John Fay, Scott Florence,
Douglas Growitz, Caroline Hermans, Fred A. Johnson, James Kendall,
Dennis Kubly, Michael Mayer, Susan Moyer, Malka Pattison,
Randall Peterson, Laura van Riper, Michael Runge, Robert Snow,
Christine Turner, and Rich Whitley

Book Design
Vitmary Rodriguez
 U.S. Department of the Interior
For more information on the Technical Guide, please visit:
http://www.doi.gov/initiatives/AdaptiveManagement/

2009 edition—updated (first edition—2007)

Suggested citation:

Williams, B. K., R. C. Szaro, and C. D. Shapiro. 2009.

Adaptive Management: The U.S. Department of the Interior
Technical Guide. Adaptive Management Working Group, U.S.
Department of the Interior, Washington, DC.

International Standard Book Number: 978-1-4133-2478-7

Any use of trade, product, or firm names in this publication is for descriptive purposes only and does not imply endorsement by the
U.S. Government.
Preface

The Department of the Interior (DOI) Adaptive Management Working Group (AMWG) sponsored the develop-
ment of this technical guide to clearly and consistently define adaptive management and describe conditions for its
implementation. AMWG membership includes representatives from across DOI’s bureaus and offices.

A writing team of resource managers, technical experts, and other specialists worked with AMWG to address
four basic questions concerning adaptive management: (1) What is adaptive management? (2) When should it be used?
(3) How should it be implemented? (4) How can its success be recognized and measured? These questions were used
to organize both the writing effort and the structure of the guide itself, with individual chapters addressing each of the
questions.

The authors sought to describe adaptive management at an appropriate level of technical detail, while remaining
focused on its definition, operational components, and conditions in which it applies. A key challenge was to provide
sufficient detail for clarification, while limiting the length and complexity of the document.

The authors gratefully acknowledge the efforts by the many people who participated in developing this guide,
including Maryanne Bach, Donna Brewer, Tara Conrad, John Dennis, Mike Ferguson, Larry Finfer, Gary Fraser, Ken
Havran, Jacob Hoogland, Rick Lemon, Kim Magraw, Chris Pease, Steve Tyson, Ben Tuggle, John Vitello, Suzanne
Weedman, and Tom Weimer. The authors also thank the people who provided valuable comments as reviewers,
including Charles Baxter, Robert Fisher, Scott Florence, Clint Moore, Roger Sayre, and Brent Uilenberg.

i
ii
Contents

Preface............................................................................................................................................................. i
Problem-Scoping Key for Adaptive Management.......................................................................................... iv
Executive Summary......................................................................................................................................... v

CHAPTER 1: WHAT IS ADAPTIVE MANAGEMENT?.................................................................................. 1

1.1. Decision making and natural resource management......................................................................................... 2

1.2. Operational definition of adaptive management............................................................................................... 4

CHAPTER 2: WHEN SHOULD ADAPTIVE MANAGEMENT BE USED?..................................................... 9

2.1. Conditions that warrant an adaptive management approach.............................................................................. 9

2.2. Institutional context for adaptive management................................................................................................. 13
2.3. Limitations of adaptive management.............................................................................................................. 15
2.4. Benefits and challenges in using adaptive management.................................................................................... 17

CHAPTER 3: HOW SHOULD ADAPTIVE MANAGEMENT BE IMPLEMENTED?..................................... 21

3.1. Operational sequence for adaptive management............................................................................................... 21

3.2. Legal considerations when applying adaptive management............................................................................... 38

CHAPTER 4: WHEN IS ADAPTIVE MANAGEMENT SUCCESSFUL?......................................................... 49

4.1. Recognizing success in adaptive management................................................................................................. 49

4.2. Key questions for successful implementation.................................................................................................. 53

CHAPTER 5: OTHER OPERATIONAL ISSUES............................................................................................. 59

5.1. Uses of information in natural resource management....................................................................................... 59
5.2. Accounting for uncertainty in adaptive management........................................................................................ 61

5.3. The measurement of learning......................................................................................................................... 62

5.4. Learning organizations.................................................................................................................................. 62
5.5. Realistic expectations for adaptive management.............................................................................................. 63

CONCLUDING REMARKS............................................................................................................................. 65

References........................................................................................................................................................ 66
Key Terms......................................................................................................................................................... 71

iii
 Problem-Scoping Key for Adaptive Management
The following key can help in dissecting a particular management problem and determining whether adaptive
management is an appropriate approach to decision making. If the answer to any question in the key is negative, then an
approach other than adaptive management is likely to be more appropriate.

1. Is some kind of management decision to be made?

(see Sections 1.1, 2.1, 2.3, 3.1, and 5.5)
No – decision analysis and monitoring are unnecessary when no decision options exist.
Yes – go to step 2.

2. Can stakeholders be engaged?

(see Sections 1.1, 1.2, 2.1, 3.1, and 4.2)
No – without active stakeholder involvement an adaptive management process is unlikely to be effective.
Yes – go to step 3.

3. Can management objective(s) be stated explicitly?

(see Sections 1.2, 2.1, 2.2, 2.3, 3.1, 4.2 and 5.1)
No – adaptive management is not possible if objectives are not identified.
Yes – go to step 4.

4. Is decision making confounded by uncertainty about potential management impacts?

(see Sections 1.1, 1.2, 2.1, 3.1, 4.1, 4.2 and 5.2)
No – in the absence of uncertainty adaptive management is not needed.
Yes – go to step 5.

5. Can resource relationships and management impacts be represented in models?

(see Sections 1.2, 3.1, 4.2, and 5.1)
No – adaptive management cannot proceed without the predictions generated by models.
Yes – go to step 6.

6. Can monitoring be designed to inform decision making?

(see Sections 2.1, 2.3, 3.1, and 4.2)
No – in the absence of targeted monitoring it is not possible to reduce uncertainty and improve management.
Yes – go to step 7.

7. Can progress be measured in achieving management objectives?

(see Sections 1.1, 3.1, 4.1, and 4.2)
No – adaptive management is not feasible if progress in understanding and improving management is
unrecognizable.
Yes – go to step 8.

8. Can management actions be adjusted in response to what has been learned?

(see Sections 1.2, 2.1, 3.1, 4.1, 4.2, 5.3, and 5.4)
No – adaptive management is not possible without the flexibility to adjust management strategies.
Yes – go to step 9.

9. Does the whole process fit within the appropriate legal framework?
(see Sections 2.3, 2.4, 3.2, 4.1, and 4.2)
No – adaptive management should not proceed absent full compliance with the relevant laws, regulations,
and authorities.
Yes – all of the basic conditions are met, and adaptive management is appropriate for this problem.
iv
Executive Summary

T he purpose of this technical guide is to present an operational definition of adaptive management, identify the
conditions in which adaptive management should be considered, and describe the process of using adaptive manage-
ment for managing natural resources. The guide is not an exhaustive discussion of adaptive management, nor does it
include detailed specifications for individual projects. However, it should aid U.S. Department of the Interior (DOI)
managers and practitioners in determining when and how to apply adaptive management.

Adaptive management is framed within the context of structured decision making, with an emphasis on uncertainty
about resource responses to management actions and the value of reducing that uncertainty to improve management.
Though learning plays a key role in adaptive management, it is seen here as a means to an end, namely good manage-
ment, and not an end in itself. The operational definition used in the guide is adopted from the National Research
Council, which characterizes adaptive management as an iterative learning process producing improved understanding
and improved management over time:

Adaptive management [is a decision process that] promotes flexible decision making that can be
adjusted in the face of uncertainties as outcomes from management actions and other events become
better understood. Careful monitoring of these outcomes both advances scientific understanding and
helps adjust policies or operations as part of an iterative learning process. Adaptive management also
recognizes the importance of natural variability in contributing to ecological resilience and productivity.
It is not a ‘trial and error’ process, but rather emphasizes learning while doing. Adaptive management
does not represent an end in itself, but rather a means to more effective decisions and enhanced benefits.
Its true measure is in how well it helps meet environmental, social, and economic goals, increases
scientific knowledge, and reduces tensions among stakeholders.

Adaptive management as defined here involves ongoing, real-time learning and knowledge creation, both in a
substantive sense and in terms of the adaptive process itself. It is described in what follows in a series of 9 steps, as
summarized in section 4.1, involving stakeholder involvement, management objectives, management alternatives,
predictive models, monitoring plans, decision making, monitoring responses to management, assessment, and adjust-
ment to management actions. An adaptive approach actively engages stakeholders in all phases of a project over its
timeframe, facilitating mutual learning and reinforcing the commitment to learning-based management. Adaptive
management in DOI is implemented within a legal context that includes statutory authorities such as the National
Environmental Policy Act (NEPA), the Endangered Species Act, and the Federal Advisory Committee Act.

For many important problems now facing the resource management community, adaptive management holds great
promise in reducing the uncertainties that limit the effective management of natural resource systems. For many conser-
vation and management problems, utilizing management itself in an experimental context may be the only feasible way
to gain the system understanding needed to improve management.

Though it is commonly thought that an adaptive approach can produce results quickly at low cost, the opposite is
more likely to be true. An initial investment of time and effort will increase the likelihood of better decision making and
resource stewardship in the future, but patience, flexibility, and support are needed over the life of an adaptive manage-
ment project. For these reasons it is important to carefully consider the potential use of an adaptive approach, and to
engage in careful planning and evaluation when adaptive management is used.

v
Chapter 1: What is Adaptive Management? 1
Adaptive management is a systematic approach ment is much more than simply tracking and changing
for improving resource management by learning from management direction in the face of failed policies,
management outcomes (1). Its origin can be traced back and, in fact, such a tactic could actually be maladaptive
to ideas of scientific management pioneered by Frederick (14). An adaptive approach involves exploring alterna-
Taylor in the early 1900s (2,3). Various perspectives tive ways to meet management objectives, predicting
on adaptive management are rooted in parallel concepts the outcomes of alternatives based on the current state
found in business (total quality management and learning of knowledge, implementing one or more of these
organizations [4]), experimental science (hypothesis alternatives, monitoring to learn about the impacts of
testing [5]), systems theory (feedback control [6]), management actions, and then using the results to update
and industrial ecology (7). The concept has attracted knowledge and adjust management actions (15). Adaptive
attention as a means of linking learning with policy and management focuses on learning and adapting, through
implementation (8,9). Although the idea of learning from partnerships of managers, scientists, and other stake-
experience and modifying subsequent behavior in light of holders who learn together how to create and maintain
that experience has long been reported in the literature, sustainable resource systems (3).
the specific idea of adaptive management as a strategy for
natural resource management can be traced to the seminal The purpose of this technical guide is to present an
work of Holling (10), Walters (11), and Lee (12). operational definition of adaptive management, identify
the conditions in which adaptive management should be
Adaptive management as described here is considered, and describe the process of using adaptive
infrequently implemented, even though many resource management for managing natural resources. The guide is
planning documents call for it and numerous resource not an exhaustive discussion of adaptive management,
managers refer to it (13). It is thought by many that nor does it include detailed specifications for individual
merely by monitoring activities and occasionally projects. However, it should aid both U.S. Department of
changing them, one is doing adaptive management. Interior (DOI) managers and practitioners in determining
Contrary to this commonly held belief, adaptive manage- when and how to apply adaptive management.

Examples of decision making in natural resource

management include the control of water releases
from a dam, direct manipulation of plant or
animal populations through harvesting, stocking,
or transplanting, and manipulations of ecosystems
through chemical or physical changes to habitats.

1
 1.1. Decision Making and Natural The following management issues exemplify sequential
decision making in natural resources in the face of
Resource Management
uncertainty:
A context for resource management involves a deci-
• In a newly established meta-population of wolves, how
sion making environment characterized by multiple (often
many animals (if any) should be relocated periodically
competing) management objectives, constrained manage-
to maximize the probability that the meta-population
ment authorities and capabilities, dynamic ecological and
will persist over the long term?
physical systems, and uncertain responses to management
actions. Management thus involves not only predicting
• What amount and timing of water release from a dam
how ecological or physical systems are likely to respond
will maintain downstream water quality, water quantity,
to interventions, but also identifying what management
and living resources, including people and communities?
options are available, what outcomes are desired, how
much risk can be tolerated, and how best to choose among
• How can an area be managed to minimize the impacts of
a set of alternative actions. The challenge confronting
recreational use on flora and fauna?
managers is to make “good” decisions in this complex
environment, recognizing that the quality of decision
• When and how much should water levels be raised or
making in the face of uncertainty should be judged by
lowered in an impoundment to maximize abundance and
the decision making process as well as progress towards
availability of invertebrates for foraging shorebirds?
desired outcomes.
• How can plant communities in an area be managed so
A common problem in natural resources management
as to protect and sustain archeological resources in the
involves a temporal sequence of decisions, in which the
area at minimum cost?
best action at each decision point depends on the state
of the managed system. Because management actions at
• How much forest should be cut each year as part of a
each point in time can influence change in the resource
pine regeneration program to maximize old-growth pine
system from that time forward, the goal of management
for use by red-cockaded woodpeckers?
is to prescribe objective-driven strategies that account for
both the current and future impacts of decisions. A key
• How can fuel loads be decreased while minimizing
issue is how best to choose management actions, recog-
effects on forested ecosystems?
nizing that the most appropriate management strategy is
obscured by limited understanding.
• Should annual hunting-season regulations be restrictive
or moderate to maximize the longterm cumulative
Often the uncertainty about management impacts
harvest of mallards?
is expressed as disagreements among stakeholders who
have differing views about the direction and magnitude of
• How much and how often should herbicide be applied to
resource change in response to management. An adaptive
minimize the proliferation of the invasive plant hydrilla
approach explicitly articulates these viewpoints, incor-
in a group of southern lakes?
porates them into the decision making process, and uses
management itself to help identify the most appropriate
• In what order should patches of isolated bull trout
view about resource dynamics. In this way, understanding
habitat be reconnected in a network of tributaries to
of the resource can be enhanced over time, and manage-
maximize the probability of population persistence while
ment can be improved.
minimizing costs?
Examples of this kind of decision problem include
• When and where should prescribed burns be used in a
the control of water releases from a dam, direct manipula-
collection of management units to maximize the prob-
tion of plant or animal populations through harvesting,
ability that Florida scrub-jays will persist at a refuge
stocking, or transplanting, and manipulations of ecosys-
over the long term?
tems through chemical or physical changes to habitats.

2
 Management of problems like these increasingly
involves a systems approach with explicit and agreed-
upon objectives, management alternatives, and analytical
approaches that can identify the most appropriate
management strategies. Adaptive management exempli-
fies such an approach; however, its focus is not only on
making good decisions in the present, but also on gaining
experience and knowledge so that future management
decisions can be improved.

Adaptive management as an example

of structured decision making

The move toward accountability and explicitness

in natural resource management has led to a growing
need for a more structured approach to decision making.
Improved clarity about key elements in a decision making
ment as learning about system processes accumulates
process can help decision makers focus attention on
over time.
what, why, and how actions will be taken. Activities in
a structured approach to decision making include the
following:
Embracing uncertainty
• Engaging the relevant stakeholders in the decision
making process Making a sequence of good management decisions is
more difficult in the presence of uncertainty, an inherent
• Identifying the problem to be addressed and pervasive feature of managing ecological systems
(16,17). Uncertainties arise with incomplete control of
• Specifying objectives and tradeoffs that capture the management actions, errors in measurement and sampling
values of stakeholders variation, environmental variability, and an incomplete
understanding of system dynamics (see Section 5.2).
• Identifying the range of decision alternatives from These uncertainties potentially degrade management
which actions are to be selected performance and contribute to acrimony in the decision
making process.
• Specifying assumptions about resource structures
and functions Perhaps not surprisingly, managers have some-
times been reluctant to acknowledge uncertainty in
• Projecting the consequences of alternative actions environmental assessments and management strategies
(18). Often there is a perception that asserting certainty
• Identifying key uncertainties as to management impacts is more convincing, and
acknowledging uncertainty increases the likelihood that
• Measuring risk tolerance for potential consequences recommended actions will be ignored. Acknowledgement
of decisions of uncertain management outcomes is sometimes seen as
an invitation for confrontation among different interest
• Accounting for future impacts of present decisions groups, resulting in an inability to reach timely agreement
on a proposed action.
• Accounting for legal guidelines and constraints
Adaptive management forces stakeholders to confront
In the ensuing chapters it will be clear that adaptive unresolved uncertainties that can significantly influence
management is itself a structured approach to decision management performance. An adaptive approach provides
making, in that it includes the key elements listed above. a framework for making good decisions in the face of
The distinguishing features of adaptive management are critical uncertainties, and a formal process for reducing
its emphasis on sequential decision making in the face of uncertainties so that management performance can be
uncertainty and the opportunity for improved manage- improved over time.

3
 1.1 Key Points This definition gives special emphasis to uncertainty
about management impacts, iterative learning to reduce
 Resource management involves decision making in uncertainty, and improved management as a result of
an environment of multiple management objectives, learning. Key points in the definition are discussed in
constrained management authorities and capabilities, more detail below:
dynamic resource systems, and uncertain responses to
management actions. Adaptive management openly acknowledges uncertainty
about how ecological systems function and how they
 Resource management increasingly involves the respond to management actions (20,21). However, adap-
articulation of objectives and management options tive management is not a random trial-and-error process.
and the use of analytical techniques to identify Instead, it involves formulating the resource problem,
optimal management strategies. developing conceptual models based on specific assump-
tions about the structure and function of the resource
 Adaptive management is a structured approach to system, and identifying actions that might be used to
decision making that emphasizes accountability and resolve the problem. Through the monitoring of outcomes
explicitness in decision making. following management interventions, adaptive manage-
ment promotes improved understanding about which
 Adaptive management is useful when there is actions work, and why.
substantial uncertainty regarding the most appropriate
strategy for managing natural resources. Adaptive management is designed to improve under-
standing of how a system works, so as to achieve
management objectives (20,21). Models are used in
adaptive management to embed hypotheses about system
behaviors and enable managers to predict the impacts of
1.2. Operational Definition of their activities. These predictions are the basis for learn-
Adaptive Management ing later on. Once activities are implemented, the testing
of underlying model assumptions against monitoring data
For the U.S. Department of the Interior to effectively provides the foundation for learning and the improvement
implement adaptive management in a consistent and of management based on what is learned.
coherent manner across all bureaus, an operational defini-
tion is needed that will be applicable for all of DOI. The Adaptive management is about taking action pursuant
definition used in this technical guide is adopted from the to desired outcomes (21). In adaptive management, the
National Research Council (19): outcomes of decisions, assessed through followup moni-
toring, are compared against explicit predictions of those
Adaptive management [is a decision process that] outcomes (20), with the comparative results fed back
promotes flexible decision making that can be into decision making to produce more effective decision
adjusted in the face of uncertainties as outcomes making (11,22,23,24). Actual and expected results can
from management actions and other events become differ for many reasons: underlying assumptions are
better understood. Careful monitoring of these wrong, actions are poorly executed, environmental condi-
outcomes both advances scientific understanding tions have changed, monitoring is inadequate, or some
and helps adjust policies or operations as part of combination of these problems. An adaptive approach
an iterative learning process. Adaptive manage- helps isolate inadequacies in a management application,
ment also recognizes the importance of natural var allowing adjustments to be made and management to be
ability in contributing to ecological resilience and improved.
productivity. It is not a ‘trial and error’ process, but
rather emphasizes learning while doing. Adaptive Adaptive management requires the participation of
management does not represent an end in itself, stakeholders. Stakeholders include people and organiza-
but rather a means to more effective decisions and tions who use, influence, and have an interest, or “stake,”
enhanced benefits. Its true measure is in how well in a given resource (25). Stakeholders should be involved
it helps meet environmental, social, and economic early in the adaptive management cycle, to help assess
goals, increases scientific knowledge, and reduces the problem and design activities to solve it. Stakeholders
tensions among stakeholders. also can help to implement and monitor those activities,
and participate in the evaluation of results. Involvement

4
of stakeholders from the beginning increases management
effectiveness and the likelihood of achieving agreed-upon
outcomes (25).

There are many definitions in the literature on

adaptive management, but a common theme shared by
them all is that adaptive management is a learning-based
process (26). The definition used in this guide was chosen
because it emphasizes the use of learning to improve
management decisions and because it is germane to
resource management in DOI.The sequence of activities
shown in Fig. 1.1 is often used to characterize adaptive
management. Additional structure can be incorporated
into this sequence, by recognizing an embedded feedback
loop of monitoring, evaluation, and management adjust-
ments that focuses specifically on learning about the
impacts of management. Multiple iterations of this loop
may occur within each iteration of the overall cycle,
accelerating learning about ecological process within the
more comprehensive cycle that includes learning about
the adaptive process itself (through periodic problem
reassessment, design, and implementation). Learning at
both levels is discussed in more detail in Section 3.1.

Assess problem

Adjust Design

Evaluate Implement

Monitor

Figure 1.1. Diagram of the adaptive management process.

5
 Other approaches to resource management system does not itself make an application adaptive.
A great many resource systems are monitored in some
Learning from the experience of management manner, but in most cases the resulting data are not used
is certainly not a new idea, but the purposeful and systematically for learning and improvement in a context
systematic pursuit of knowledge as an explicit part of of objective-driven management.
management has rarely been practiced. The term “adap-
tive management” has been used to describe a broad array More formal approaches to decision making can
of approaches that involve learning while doing, but the be identified, depending on the amount of uncertainty
phrase is not always appropriate. For example, manage- facing managers and the capacity to influence the system
ment by trial and error is sometimes described as adaptive being managed (Fig. 1.2). In an ideal situation in which
management, but at best it is likely to be inefficient, and system controllability is high and management impacts
at worst it can retard the institutionalization of experi- are predictable, formal optimal control approaches can be
ence and learning. Nor should adaptive management be used to identify optimal management strategies. If one’s
confused with conflict resolution, which focuses on nego- ability to control the system is limited, hedging strategies
tiating tradeoffs among competing interests. Management or scenario planning can be useful, depending on how
approaches that primarily depend on expert opinion and well the effects of management can be predicted. As
advice for decision making are not by themselves adap- indicated in Fig. 1.2, adaptive management is appropriate
tive. Finally, in the absence of additional structure in a if management can strongly influence the system but
decision making process, monitoring a managed resource uncertainty about management impacts is high (27).

High Scenario Adaptive

Planning Management
Uncertainty

Optimal
Hedging Control
Low

Low High

Controllability

Figure 1.2. Approaches to decision making in a natural resource

system. The appropriate approach depends on the influence
decisions can have on system behavior and the amount of
uncertainty about management impacts (27).

6
 Adaptive management requires stated management
objectives to guide decisions about what to try, and
1.2 Key Points
explicit assumptions about expected outcomes to compare
 Adaptive management acknowledges uncertainty
against actual outcomes. It is important to know what the
about how natural resource systems function and how
available management options and alternative assump-
they respond to management actions.
tions are, in case the action that is tried does not work as
expected. The linkages among management objectives,
 Adaptive management is designed to improve under-
learning about the system, and adjusting direction based
standing of how a resource system works, so as to
on what is learned distinguish adaptive management
achieve management objectives.
from a simple trial and error process. In the chapters that
follow, we describe adaptive management formally in
 Adaptive management makes use of management
terms of objectives, management options, and models
interventions and followup monitoring to promote
that embed alternative hypotheses about management
understanding and improve subsequent decision
responses. But in essence, adaptive management will be
making.
seen to be learning by doing, and adapting based on what
is learned (28). A comparison of adaptive management
with some other approaches to natural resource manage-
ment is presented in Section 5.1.

Adaptive management requires stated management

objectives to guide decisions about what actions
to take, and explicit assumptions about expected
outcomes to compare against actual outcomes.

7
8
Chapter 2: When Should Adaptive Management be Used? 2
In this chapter we describe the conditions under management. The following discussion draws from
which adaptive management is applicable, and highlight published sources as well as the experiences of manage-
some challenges, limitations, and benefits of an adaptive ment agencies within the Department of the Interior.
approach to resource management.
There are two key conditions that are mentioned in
2.1. Conditions that Warrant an Adaptive all thoughtful analyses. First, “there must be a mandate to
take action in the face of uncertainty” (12,24). That is, the
Management Approach problem must be important enough to require action of
one sort or another. Situations without this imperative can
Not all decisions can or should be adaptive. In some
result in either delayed action as more information is
cases there is no opportunity to apply learning; in others,
acquired or action foregone altogether. Second, there
there is little uncertainty about what action to choose;
must be the institutional capacity and commitment to
and in still others, there is disagreement about objectives.
undertake and sustain an adaptive program. This condi-
But the concept of adaptive management is so intuitively
tion includes an institutional stability for long-term
appealing that the phrase has been applied indiscrimi-
measurement and evaluation of outcomes, which should
nately, with the result that many management applications
allow the early investment in an adaptive approach to
fail to achieve the improvements expected from adaptive
pay off in long-term management. Together, these two
management. In many instances, that failure may have
conditions imply that decision makers must be motivated
less to do with the approach itself than with the inappro-
and patient, that is, they must care about improving
priate contexts within which it is purported to apply (29).
management over extended time frames (12).
An important question is which decision problems are
appropriate for the application of adaptive management.
In addition to these two overarching conditions,
six more conditions can be identified directly from
There is a considerable literature that explores reasons
the meaning and context of adaptive management, as
why the practice of adaptive management has not lived up
described in the previous chapter. Adaptive management
to its promise, and extensive documentation of some of
is warranted when there are consequential decisions to
the more prominent failures. But only recently has atten-
be made, when there is an opportunity to apply learning,
tion focused proactively on those attributes of resource
when the objectives of management are clear, when the
management that make a problem amenable to adaptive
value of reducing uncertainty is high, when uncertainty
can be expressed as a set of competing, testable models,
and when a monitoring system can be put in place with a
reasonable expectation of reducing uncertainty.

A real management choice is to be made

As described in Chapter 1, adaptive management

is first and foremost an approach to the management
of natural resources and not simply an opportunity to
learn. Thus, an application of adaptive management must
involve a real choice among management alternatives that
affect resource systems. The variability among alterna-
tives must be consequential (i.e., different alternatives
produce substantively different management impacts),
and the alternatives must be ecologically, economically,
politically, and legally feasible.

9
 The genesis of alternatives should be multidisci-
decision making process to adjust management actions
plinary and participatory. They can arise from within the
in response to measured outcomes (32). This requires
management agency, from scientists or engineers working
both flexibility in the actions themselves as well as
for, with, or in opposition to the management agency,
flexibility within the management institutions to adopt
from the regulated community, or from other stake-
the change. Second, management institutions must have
holders. Some decisions are particularly difficult because
the stability to measure outcomes and use the results at
a suitable range of alternatives cannot be easily identified.
later times. Adaptive management sometimes has failed
In such cases, a collaborative approach in identifying
because institutions managing the process dissolved
alternative actions is especially useful.
before the learning could be applied (33). Third, it must
be possible to acquire understanding quickly enough
Because natural resource systems operate at multiple
to apply it to subsequent management decisions. Some
spatial and temporal scales and involve interactions
ecological processes respond very slowly to management
among many component systems, the development
(for example, forest systems). If learning can occur only
of alternative actions should account for multiscale
after observing slow response variables, many iterations
responses. One consequence of this complexity is that
of decision making may have passed before the new
several pathways may exist to achieve similar outcomes,
knowledge can be applied.
with alternative pathways differing enough in some
relevant aspects (feasibility, cost, public acceptance) to be
Ideally the response to previous management
considered as bona fide alternatives.
actions can be assessed before a decision about the next
management action is made. For example, the response of
The alternatives considered in adaptive manage-
waterfowl populations to hunting regulations in one year
ment are constrained by existing laws, regulations, and
can be assessed in time to inform the setting of hunting
policies, both substantive and procedural. A number
regulations in the following year (34). On the other hand,
of substantive laws govern natural resource decision
making (for example, Clean Water Act, Clean Air Act,
Endangered Species Act, etc.). Of the procedural laws,
the National Environmental Policy Act (NEPA) and its
implementing regulations provide considerable guidance
about developing and considering alternative management
actions. An emerging view discussed in Section 3.2 sees
the NEPA process as a powerful and potentially effective
way to embody adaptive management (30).

There is an opportunity to apply learning

A condition of adaptive management is that resource

management decisions can be revisited and modified
over time or that multiple decisions of a similar nature
can be made over time. That is, decision making needs
to be iterative over time and possibly space; otherwise,
learning cannot be applied. Many examples of adaptive
management treat a single management unit (for example,
a single river or a continental population of ducks) over
time, applying the learning derived from earlier actions
to decisions made at later times. But equally appropriate
are situations where similar management units are each Ideally the response to previous management actions
treated only once, and the learning accrued from treat- can be assessed before a decision about the next
ments of some units is used in decisions about how to
treat other units at a later time (31).
management action is made. For example, the response
of waterfowl populations to hunting regulations in one
Besides iterative decision making, several other year can be assessed in time to inform the setting of
considerations affect the opportunity to apply learning. hunting regulations in the following year.
First, perhaps obviously, the adaptation of actions must
be possible. That is, there must be flexibility in the
10
applications of adaptive management in forestry can be making. An adaptive management application should
limited by the fact that the relevant response variables target learning that will change management actions and
may not be measurable until decades after a management improve the ability to achieve management objectives.
action is taken (35).
The “value of information” refers to how much better
Clear and measurable management objectives the expected performance of a managed system would be
can be identified if uncertainty were reduced. A high value of information
means that the decision maker will potentially choose
An adaptive approach requires explicit and measur- different alternatives if the system is better understood.
able objectives. As described in the next three conditions, With improved understanding comes better decisions, so
uncertainty about how to achieve objectives is what that success in achieving objectives becomes more likely.
motivates adaptive management and drives the design The prospect of substantially improved decision making
of the monitoring system. To address this uncertainty justifies the cost of monitoring and assessment (24) in
stakeholders must agree on the objectives. Although an adaptive management. Conversely, an adaptive approach
adaptive management framework can serve to structure is not warranted if the value of information is low,
dialogue among stakeholders, adaptive management itself essentially because the potential improvement in manage-
is not designed to resolve conflicts about management ment does not justify its costs. Sometimes the tradeoffs
objectives. If the objectives are not clear and measurable, between costs and benefits can be made explicitly,
the adaptive framework is undermined. particularly in applications in which an economic value
can be ascribed to learning.
Objectives need to be measurable for two purposes:
first, so progress toward their achievement can be Provided careful thinking, analysis, and modeling
assessed; second, so performance that deviates from are undertaken prior to implementation, one of the
objectives may trigger a change in management direc- advantages of an adaptive approach is that surprises can
tion. Explicit articulation of measurable objectives helps be anticipated (29). Preparing for the unexpected means
to separate adaptive management from trial and error, fully acknowledging uncertainty, articulating the ways in
because the exploration of management options over time which assumptions might be wrong, exploring the conse-
is directed and justified by the use of objectives. quences to management of uncertainty, and having the
appropriate monitoring in place to recognize and benefit
Objectives must be relevant to the project or program from unexpected outcomes.
to which they apply. An example of a project objective
might be to increase biodiversity of amphibians by 25
percent in a local watershed. An example of a program
objective is that used by the U.S. Fish and Wildlife
Service for adaptive harvest management, namely the
maximum long-term harvest of waterfowl consistent
with population goals in the North American Waterfowl
Management Plan (36, 37). In both cases, the objectives
are measurable, relevant to the management problem, and
useful for decision making, evaluation, and learning. The
nature and use of objectives are discussed in more detail
in Section 3.1.

The value of information for decision

making is high

The fundamental motivation for adaptive manage-

ment is that the impact of management actions on
resources is uncertain, and the reduction of that uncer-
tainty will accelerate progress in meeting management
objectives over time (34). Although uncertainty can be
identified in almost any resource management problem,
its reduction does not automatically lead to better decision
11
 Uncertainty can be expressed as a set of to allow the assumptions of decision makers and stake-
testable models holders—essentially, the models that exist in the minds of
a few individuals—to remain unexpressed and untested.
A formal approach to adaptive management uses
the tools of structured decision analysis to inform and A monitoring system can be established
analyze the problem. A key step is to predict the effects to reduce uncertainty
of management actions that are relevant to the objectives.
But predictions require models, whether conceptual or Monitoring is fundamental for adaptive management,
quantitative. Adaptive management utilizes multiple as a source of data with which to test alternative models
models, each imbedding a particular hypothesis about and measure progress toward accomplishing management
how the natural resource system responds to manage- objectives. Simply put, adaptive management is not
ment. These models are tested with monitoring data to possible without effective monitoring (see Section 3.1 for
determine which model best represents system responses. a more detailed discussion of monitoring).
In this way the hypotheses underlying management
decisions can be expressed and tested. There are important details, however, that influence
whether a monitoring system will help reduce uncertainty
Models are critical in an adaptive management to any useful degree, and these should be considered
process, if only as a means to encourage managers, when evaluating whether to undertake an adaptive
scientists, and other stakeholders to think carefully program. For example, anticipated effects of alternative
about the structure and dynamics of the systems they are actions need to be substantial, because field monitoring
managing. When there is contention among stakeholders can seldom detect subtle differences. Thus, management
about how the system will respond to management, experiments must be dramatic enough to produce an
modeling forces stakeholders to express these differences observable response from the ecosystem, or they will
as alternative hypotheses, which then can be tested. The not facilitate learning (32). It is useful here to recall that
models embodying the hypotheses in question also can the statistical power to distinguish among hypotheses is
be used to identify critical monitoring variables to use for influenced by sample size and the magnitude of treatment
comparing hypotheses, and they provide a framework for effects, and these factors apply as well to an adaptive
interpreting monitoring results and evaluating alternative monitoring system. Poor monitoring precision does more
actions to best achieve management objectives. than simply slow the rate of learning; imprecise moni-
toring can produce misleading evidence that supports
Models can be qualitative and conceptual, quantita- inappropriate management. Inaccurate or imprecise
tive and highly detailed, or anywhere in between. In all monitoring can actually be counterproductive to the goals
cases, their function in an adaptive management context of management (38).
is to make predictions about how a natural system will
respond to management actions and to evaluate the Learning is accelerated when the principles of
consequences of uncertainty. A common complaint used experimental design—replication, randomization, and
to justify not undertaking an adaptive program is that the control—are used. Attention to these principles, and
data are sparse and there is too much uncertainty to build their incorporation where possible, will lead to more
models. But this is precisely where adaptive management rapid improvements in management. The amount of
is most valuable—in expressing and reducing uncertainty. temporal and/or spatial replication is an especially
The alternative to building models of system dynamics is important design feature because it determines the

12
necessary sampling intensity. All other things being when it was first initiated. Thus, stakeholders were
equal, management actions that can be replicated many already involved in harvest management, well developed
times at different locations will reduce uncertainty more models were available, a decision making process was
quickly. Replication over time is also valuable (and more in place that involved federal, state, and public interests,
common), but typically the sample size is necessarily and extensive, long-term monitoring programs were
lower, and information accrues more slowly. ongoing (39). AHM was designed from the outset to
take advantage of these preadaptations, recognizing that
A realistic assessment of the potential for monitoring considerable uncertainty still remained about the impacts
is a critical condition for adaptive management. This of harvest regulations.
assessment should include not only the power of the
monitoring system and the efficiency of its design, but Commitment of executive leadership
also the institutional resources needed to sustain the
monitoring (and analysis of the resulting data) over the Adaptive management involves an ongoing commit-
time frame required to inform management. ment of leadership and support. Soon after the initiation
of an adaptive management project, executive leadership
2.2. Institutional Context for may anticipate a reduction or elimination of stakeholder
conflict, a rapid reduction in the amount of scientific
Adaptive Management
investigation that is needed, and early declines in funding
needs. But adaptive management activities require
There are a number of factors associated with
management involvement and funding throughout the life
management problems that can encourage the use of
of the project, not just at its inception.
adaptive management. These include not only certain
characteristics of the management situation itself, but
From a financial perspective, long-term funding
also the nature and commitments of implementing
highlights the commitment of implementing organizations
organizations.
to adaptive decision making, and it promotes the planning
and implementation of the monitoring and evaluation
Using pre-existing institutional structures needed for adaptive management. Conversely, a lack of
long-term support limits progress in reducing uncertainty.
Certain characteristics of the record of management
may help to determine whether adaptive management is
The support required for an adaptive approach may
appropriate for a particular situation. One is a history of
include not only funding for monitoring and evaluation,
decision making that indicates a willingness to address
but also an investment in more inclusive and robust
the risk of unintended and/or undesirable natural resource
decision making processes. It is essential that execu-
impacts. Others include previous stakeholder involvement
tive leadership be aware of uncertain outcomes and be
in a collaborative group environment, cost sharing of
prepared to make the necessary changes as adaptive
collaborative efforts, and a demonstrated commitment
management progresses through implementation.
to evaluation and scientific rigor. The existence of these
characteristics prior to the creation of a formal program
Finally, executive leadership is needed to support an
is a strong indication that adaptive management is poten-
institutional culture and the organizational arrangements
tially useful.
that will acknowledge uncertainty and promote learning.
Adaptive management flourishes in a learning organiza-
An example of designing around pre-existing
tion that encourages experimentation, rewards risk taking,
conditions is the U.S. Fish and Wildlife Service adap-
and embraces the lessons learned from experimentation.
tive harvest management (AHM) program, which was
To successfully employ an adaptive approach in manage-
built upon a series of features that were already in place
ment, a philosophical shift from “expert” to collaborative

13
 learning will likely be necessary. Executive leadership
must play a critical role in the transition to a learning
organization and in sustaining it thereafter. These issues
are discussed in more detail in section 5.4.

Consensus on management objectives

Although technical information and scientific

understanding are required to assess tradeoffs and levels
of risk associated with different actions, the selection of
an appropriate management strategy is in essence a social
decision that requires consensus building. In order for a
management strategy to work on the ground, stakeholders
must support the project goals and its objectives.

Consensus on goals and objectives at the beginning

of an adaptive management project sets the stage for an
iterative, adaptive management cycle (40). However,
consensus must continue through the life of the project.
Consensus is sustained by ongoing collaboration, through
which the potential conflicts arising in experiential
learning can be resolved (41,42).

Consensus is promoted by collaborative frameworks

that foster mutual learning, relationship building, and the
creation of a shared understanding as the basis for agree-
ment. Collaborative structures are in essence negotiated
agreements among stakeholders, which are embraced and
sustained because the stakeholders accept the outcome
of a process they perceive to be participatory and fair
(43,44).

2.2 Key Points

 For adaptive management to be successful, executive
leadership must support needed changes to existing
institutional culture and structures.

 Stakeholders must be willing to work collaboratively

in a group environment to plan specific courses of
action.

 In order for a specific adaptive management strategy

to work on the ground, stakeholders must support the
strategy goals and objectives.

 Implementation of adaptive management can be

Glen Canyon Dam
facilitated by using pre-existing structures and
processes.
14
2.3. Limitations of • A firm commitment to funding and institutional
support for monitoring is lacking. Adaptive management
Adaptive Management
should not be employed without a clear commitment to
monitoring over the life of the project. If a commitment
Although adaptive management often can enhance an
for monitoring is in question, it may be necessary to take
agency’s ability to achieve resource objectives, there are
another approach to decision making that does not rely
situations where its application may not be appropriate.
on monitoring, such as expert systems, management
An agency considering the employment of adaptive
intuition, or non-technical understanding of the system.
management should ensure that its use is suitable for the
particular situation. Adaptive management should not
be employed if one or more of the following limitations There are irresolvable conflicts in defining
apply. explicit and measurable management
objectives or alternatives
Decision making occurs only once
If explicit and measurable management objectives
If resource management decisions cannot be revisited cannot be identified or alternatives cannot be determined,
and modified over time, then adaptive management then adaptive management is not feasible. Conflicts may
cannot be meaningfully employed. Many decisions are arise in a collaborative process in which stakeholders with
essentially irreversible in that follow-up adaptation is different interests fail to agree upon these components.
either infeasible or impossible. An example is the removal One alternative in this situation is conflict resolution.
of a dam on a large river where the decision can be Collaborative management is never easy, and agencies
made only once. Of course, such a decision may be part and stakeholder groups should not abandon the approach
of a larger decision making program, for example the until the possibility of agreement on the key components
management of a watershed that includes many dams, of adaptive management is exhausted.
where learning that follows from the removal of one dam
informs subsequent decisions about other dams. Decisions that affect resource systems and
outcomes cannot be made
Monitoring cannot provide useful information
for decision making Adaptive management should only be considered in
situations where management actions substantially influ-
A suitable monitoring strategy is a key requirement ence the outcome. In certain situations, a management
for any adaptive management approach. Data collected agency can only partially influence the resource system.
from monitoring are used to test alternative models and For example, if an agency manages a relatively small area
measure progress toward management objectives. There surrounded by private land, and the adaptive management
are several situations in which an effective monitoring project applies only to the agency-managed land, manage-
program cannot be established: ment activities on the private lands may well dominate
the effect of agency actions. In such a situation, adaptive
• The frequency of monitoring cannot keep pace management is unlikely to be useful.
with changes in the natural system. If monitoring is too
infrequent or the system changes at too rapid a pace,
monitoring data may be unrepresentative of the resource
system by the time a decision is to be made.

• A design for experimental management and Some resource decisions are essentially irreversible
monitoring cannot be developed to test hypotheses. in that follow-up adaptation is either infeasible or
If understanding of the resource system is so limited
(or management is so constrained) that designing a impossible. An example is the removal of a dam on a
meaningful experiment becomes problematic, adaptive large river where the decision can be made only once.
management may not be appropriate. This problem is
most likely to occur when the geographic scale of the
problem is extensive, replication is difficult or impossible,
or there are many potentially confounding environmental
factors that combine to influence outcomes.

15
 Risks associated with learning-based decision In fact, an adaptive management approach often can
making are too high. alleviate the level of risk through a careful articulation of
objectives, management alternatives, and other elements
It is sometimes considered inadvisable to use adaptive of the resource problem. If the levels of uncertainty and
management when the “worst case scenario” resulting risk are high, an adaptive approach that includes pessi-
from a management action would be unacceptable to mistic alternatives and very high penalties for negative
stakeholders. An example of such a situation might arise outcomes may well be the preferred approach to manage-
when management actions can lead to the extinction of ment.
extremely rare, threatened, or endangered resources. In
this particular situation one approach might be to include Each of the limitations listed above is often encoun-
management thresholds that prevent the worst-case tered in natural resource management. In addition, other
scenario from occurring. conditions can undermine adaptive management, for
example an inability to reach agreement about the key
The limits of acceptable risk can vary substantially elements of structured decision making, or a mismatch
among applications, so adaptive management should not between the rate of change in system process and the
automatically be discounted even when dealing with rare frequency of management interventions. When such limi-
or fragile resources. The relevant issue here is the value tations are encountered, decision makers should question
of information (see Section 2.1), taking into account the whether the use of adaptive management is appropriate,
risks associated with learning-based decision making. In and perhaps consider other approaches to manage public
assessing risks, it is important to analyze the risk of the lands and resources.
“do nothing” alternative, because the risk associated with
maintaining the status quo may well be as high or higher
than that of the alternatives.

2.3 Key Points

 Adaptive management is not appropriate for single-
time decision making.

 It is not appropriate if monitoring information is

unavailable to decision makers.

 It is not appropriate if there are irresolvable conflicts

about objectives or decision alternatives.

 It is not appropriate if management interventions

cannot influence system behaviors in ways that affect
management returns.

 It is not appropriate if there is not a commitment to

sustained funding for monitoring and assessment.

An example of high-risk management involves

management actions that can lead to the
extinction of extremely rare, threatened,
or endangered resources.

16
2.4. Benefits and Challenges in Adaptive management encourages long-term
using Adaptive Management collaboration among stakeholders

Adaptive management brings resource managers,

Benefits researchers, and other stakeholders together and encour-
ages long-term collaboration through the development
An adaptive approach provides flexibility and strengthening of institutional ties (45). These ties are
to act in the face of uncertainty important in maintaining the level of support needed to
successfully implement adaptive management. Through
Adaptive management helps managers address strengthened collaboration, stakeholders can be encour-
resource issues by providing the flexibility to adjust aged to remain involved over the life of an adaptive man-
management actions as additional understanding is agement project.
gained. It can help determine whether management
actions are having desired effects and whether mitigation Adaptive management promotes
measures are cost effective. The flexibility of adaptive optimal decision making with the
management to respond to changing environmental condi- information available
tions and improved understanding can result in better
decision making. Adaptive management fosters the acquisition of new
knowledge and understanding by specifying hypotheses
An adaptive management and designing management alternatives to test them
approach is learning based against field data (8, 46). The information accumulated
through this process is used to adjust strategy periodically
The concept of learning is central to adaptive on the basis of what has been learned. In this sense, adap-
management (8), with learning seen as a means to good tive management allows decision makers at each juncture
management. Learning within the context of adaptive to make the best decisions they can with the information
management derives from evaluation of previous available at that time (41).
management actions, the results of which are used to
inform subsequent actions (8). The premise of an adaptive
management approach is that the behavior of resource
systems is uncertain but management is required anyway,
and the reduction of uncertainty over time can lead to
better management.

Adaptive management specifies what

actions are to be taken and when

Adaptive management produces management strate-

gies that specify what management actions are to be
taken and how and when they should be adjusted. These
strategies are based in turn on an explicit articulation of
the management problem, what is known (and not known)
about the resource system being managed, and the
objectives of management (8). This explicitness makes
it possible for stakeholders to focus on the key attributes
involved in learning-based resource management, while
avoiding the confusion and controversy that typically
results when key management elements are not open to
discussion and negotiation.

17
 Challenges Commitment to monitoring and evaluation
over the life of the project
Institutional reluctance to change
In times of shrinking funding, managers must care-
For adaptive management to be embraced on an fully assess the cost of the monitoring and assessment
institutional level, refinements in existing approaches that inform decision making in an adaptive approach.
to natural resource management are needed (8,47). An The costs of timely monitoring and assessment over
example might involve new ways of dealing with over- extended time scales are substantial, and often appear
lapping responsibilities and authorities among agencies, to be especially high at the outset of a project when
so as to reduce or eliminate resistance to one agency’s compared with the costs of trial and error with only
adaptive management project by another agency that has incidental monitoring (8). Agencies must be willing to
regulatory oversight. make a commitment to cover the costs of monitoring
and evaluation over the life of an adaptive management
Some barriers to implementation go beyond the project; otherwise, discontinuing the monitoring effort
operational level. One such barrier is an inadequate will lead quickly to the cessation of adaptive decision
recognition that the targets of resource management making. Agencies also need to commit to a schedule for
are rapidly becoming more inclusive. For example, monitoring, analysis, and re-examination of decisions as
ecosystem management traditionally is approached by understanding accumulates. In the absence of a commit-
targeting only one or a few system attributes, failing ment of resources for timely monitoring and evaluation
to account for the broader resource context and its over the life of the project, the use of adaptive manage-
implications for resource management. A framework for ment becomes problematic.
adaptive management allows the resource problem to be
identified in a more inclusive context that includes issues Significant time lags between management
like system viability and sustainability. actions and their impacts

Another institutional barrier is a lack of the resource Time itself is a challenge in implementing adaptive
planning and design capacity that are required for adap- management (48). In many cases, the overall costs
tive management (see Section 3.1). For example, agency associated with adaptive management are tied as much to
programs often have an inadequate capacity for the the timeframe of the project as they are to its complexity.
outcome-based monitoring needed for adaptive manage- Some adaptive management plans require years of
ment. The problem here is not so much an inability monitoring in order to be able to ascertain the results of
to understand the process and procedures of adaptive an initial action. Of course, models that forecast some
management, as it is that program operations focus on future endpoint as a consequence of a decision or series of
tracking and assessment of activities and outputs rather decisions should also be able to predict resource status at
than resource outcomes. various intervals prior to that endpoint, allowing manage-
ment assessments to be performed on the predicted status
Implementation of adaptive management will require over an abbreviated interval. The problem of time lags
a shift in focus toward resource sustainability as a stra- is further complicated by the fact that individual deci-
tegic target, with resource planning and design, decision- sion makers and/or managers rarely remain in the same
based monitoring, and assumption-driven research as position over the needed timeframes (8).
central activities. In essence, adaptive management will
require refinements to the resource management business Implementing adaptive management in a
model and adjustments in the organizational and institu- complex legal environment
tional arrangements that support it. See Section 5.4 for
further discussion on organizational roles Legal issues must be weighed when deciding whether
and implications. to implement an adaptive management strategy. In many
cases, a NEPA decision process is required of federal
agencies.

18
Depending on the resource problem and the scope of
the project, requirements under other federal laws may
2.4 Key Points
also be triggered. Some laws may constrain or even
preclude the use of adaptive management (see Section  Adaptive management promotes cooperative decision
making in the face of uncertainty about the impacts of
3.2); on the other hand, legal considerations sometimes
management interventions.
can be successfully integrated with it. Indeed, the case
studies included on the enclosed CD suggest that adap-
tive management might make NEPA compliance more  Adaptive management produces management strate-
gies consisting of actions that are tied to resource
effective and efficient in some instances.
status and current understanding.
Collecting enough information
 Adaptive management brings resource managers,
to evaluate progress researchers, and other stakeholders together and
encourages long term collaboration.
The amount of data required for adaptive manage-
ment depends on the system being managed, the actions
 Resistance to institutional change and a complex legal
being implemented, the objectives of management, and environment can be impediments to adaptive
the amount of uncertainty (49,50). Project costs obviously management.
increase for applications that require more frequent
monitoring and the collection of larger amounts of data
 Agencies must be willing to commit to monitoring
during each monitoring event. and evaluation over the life of an adaptive manage-
ment project.
Projects should be assessed individually, with each
project tailored to the resource being managed, the
environmental conditions of the project area, the project
objectives, and the capabilities of the manager to imple-
ment decisions and carry out the subsequent monitoring
and assessment. A considerable amount of up-front plan-
ning may be required; however, an initial investment of
time and effort increases the likelihood of better decision
making and resource stewardship in the future.

Projects should be tailored to the resource being

managed, the environmental conditions of the project
area, the project objectives, and the capabilities of the
manager to implement decisions and carry out the
subsequent monitoring and assessment.

19
20
Chapter 3: How Should Adaptive Management be Implemented? 3
From an operational point of view, adaptive manage- Although adaptive management is described here
ment simply means learning by doing (i.e., learning by a series of steps in the set-up and iterative phases,
through management) and adapting what one does it is important to recognize that adaptive management
based on what is learned (i.e., adjusting management is a complex endeavor that includes much more than
as understanding improves). Learning contributes to simply following a sequence of steps. Properly executed,
management by providing information on which to base the process involves ongoing, real-time learning, both
management strategies, and management reinforces in a technical sense and in terms of process itself.
learning by implementing actions that are useful in Stakeholders need to be engaged at the stage of initial
investigating the resource system. A sequential applica- problem formulation and remain engaged throughout
tion of these component activities should produce both implementation. By implication, an adaptive approach to
improved understanding of resource dynamics and management improves on the traditional communications
improved resource management (51). approach in which scientists create knowledge and then
pass it on to practitioners, with other stakeholders acting
As noted earlier, the emphasis in an adaptive as passive observers (51). Instead, an adaptive approach
approach is first and foremost on resource management. actively engages parties in all phases of the project,
The value of understanding, and the monitoring and facilitating mutual learning and reinforcing the commit-
analysis that produce understanding, is inherited from ment to learning-based management.
their contributions to the objectives of resource manage-
ment. Although the focus is on learning, the ultimate goal
of the effort is smart management. Set-up phase
This chapter focuses on the implementation of Adaptive management prescribes the integration
adaptive management. The actual process of using of decision making, monitoring, and assessment into
adaptive management is discussed in terms of its key an iterative process of learning-based management. To
structural elements and the integration of these elements implement the process, certain elements must be put
into an iterative cycle of management, monitoring, and in place, and then used in a cycle of iterative decision
assessment. Also highlighted are some legal issues that making.
focus on compliance with the relevant environmental and
administrative laws.

3.1. Operational Sequence for

Adaptive Management
Implementation of adaptive management can be
described in terms of two phases: a set-up phase in
which its key components are developed, and an iterative
phase in which the components are linked together in a
sequential decision process. The set-up phase has five
structural elements, namely stakeholder involvement,
management objectives, potential management actions,
predictive models, and monitoring plans. The iterative
phase uses these elements in an ongoing cycle of learning
about system structure and function, and managing based
on what is learned.

21
 Step 1- Stakeholder involvement
Ensure stakeholder commitment to adaptively
manage the enterprise for its duration

Of particular importance in adaptive management is

that stakeholders assess the resource problem (see Fig.
1.1) and reach agreement about its scope, objectives,
and potential management actions, recognizing that
differences of opinion about system responses may exist
even when there is consensus on these issues. Clearly,
agreement about scope, objectives, and interventions is
not possible in the absence of stakeholder involvement in
establishing them. Thus, a first step in adaptive manage-
ment is to engage the appropriate stakeholders (See Case
Study 1 and BLM Collaboration Desk Guide on CD for a
discussion of stakeholder involvement).

Several activities are involved. First, stakeholders Case Study 1: For decades, there has been concern about the
must be identified and encouraged to participate. This ecological impacts of the operation of Glen Canyon Dam on
might involve personal contacts, public announcements, downstream resources, particularly the riparian areas along the
Colorado River in the Grand Canyon, Arizona. In recent years
formal consultations, or other means. Second, a process efforts have been made to evaluate and adapt management
must be implemented that solicits stakeholder input in actions for resource protection through experiments that are
the design of the adaptive management project and, in monitored for their effects in the Grand Canyon. (See included
particular, the identification of management objectives CD for additional information on this project).
and potential management actions. Depending on the
project, this may require formal or informal consultations,
legally mandated and administered procedures, or other
approaches. In any case, stakeholder involvement in age fish stocks before releasing them in a reservoir. Other
identifying key components of the project should be open projects may involve a large number of disparate stake-
and transparent. Third, stakeholders must commit to an holders, as with the seasonal release of water from a dam
agreed-upon process of reducing uncertainties and/or on a large river, or the production of timber on a large
disagreements about the effects of management. That is, regional forest, or the management of a coastal fishery.
having reached agreement on the scope of the manage- As a general rule, the number of stakeholders, and the
ment problem and its objectives and potential interven- breadth of their perspectives and values, will vary with
tions, stakeholders must then commit to an iterative the geographic, ecological, administrative, and political
process of objective-driven decision making. scale of the adaptive management project.

In general, the group of stakeholders should be broad Stakeholders should play a role in identifying
enough to express the uncertainty (perhaps through the scope of the project as well as the objectives and
disagreement) that is the focus of adaptive management. potential management actions. Within a context of legal
However, adaptive management is not prescriptive about and institutional boundaries, stakeholders help to define
who the stakeholders are, how many there are, or what the operating environment of an adaptive management
their perspectives or values are. The scale and complexity project, and they influence both decision making and
of stakeholder involvement can vary greatly among the opportunity to learn. All too frequently, a decision
projects and is influenced by the scale and complexity making process is undertaken without agreement about
of the application itself. Many adaptive management scope, objectives, and management alternatives. Without
projects involve only one or a few stakeholders, as with this agreement, any management strategy likely will
a refuge manager who is unsure how to manage water be viewed as reflecting unshared objectives and inap-
control structures on the refuge, or a farmer who is unsure propriate or unnecessary limitations on management.
how to seed and cultivate some of his farmland, or a fish The prospects for conflict increase dramatically in such a
hatchery manager who is uncertain about how long to situation.

22
 The success of an adaptive management project
requires an adequate understanding of the resource issues
Step1- Key Points
and the dedication to stay abreast of new information. The
interface between scientific investigation and stakeholder  A strong effort must be made to identify and engage
the appropriate stakeholders.
understanding becomes increasingly difficult as proposed
actions become more technical, and it is not uncommon
for both scientists and stakeholders to become impatient  All phases of the adaptive management process must
be open, transparent, and accessible to stakeholders.
and frustrated. Engagement and communication among
stakeholders is critically important if arguments on the
meaning of science are to be minimized.  Stakeholders must strive for agreement on scope,
objectives, and management alternatives for the adap-
tive management application.
Of particular concern in adaptive management is the
asymmetry between management interventions, which
often must be implemented in a relatively short amount of  Stakeholders must commit to a process for adjusting
management strategy over time, based on resource
time, and their impacts, which sometimes require years or
status and learning.
even decades to be recognized. This asymmetry imposes
special demands on stakeholders to remain engaged over
an adequate timeframe for learning to occur. Among other  Stakeholder organizations must be encouraged to
commit time and energy to adaptively manage the
things, it may be useful, and even necessary, to design
resource over the agreed-upon timeframe.
monitoring and assessment programs at different scales,
so as to build understanding incrementally while antici-
pating unexpected results that may require adaptations in  Stakeholders must commit resources for monitoring
and assessment, in addition to decision making.
the project. Otherwise, premature interruption of moni-
toring efforts, or stakeholder pressure to terminate the
application of adaptive management, could short-circuit
the expected benefits of improved decision making.

Recognizing stakeholder interests and ensuring

their involvement for the duration of the management
enterprise are requirements for learning–based manage-
ment in general. But involvement is more than passive
participation in information sharing and other stakeholder
prerogatives. Adaptive management involves the
commitment of time, resources, and active engagement
of stakeholders. These requirements often are underesti-
mated. If stakeholders are unwilling to dedicate sufficient
time and effort, group deliberations have the potential
to devolve into value-based arguments, minimizing the
positive impacts of monitoring and evaluation. Program
documents should explicitly state the responsibilities of
the stakeholders, and every effort must be made to ensure
that stakeholders will meet those responsibilities over the
life of the project.

Bemidji Crude-Oil Research Project; measuring crude-oil

thicknesses during the aquifer test.

23
 Step 2- Objectives to resource management and seek a common basis for
recognizing management success. In particular, objec-
tives should be defined cooperatively through a dialogue
Identify clear, measurable, and agreed-upon among managers, scientists, and other stakeholders.
management objectives to guide decision
making and evaluate management In the context of adaptive management, objectives
effectiveness over time must be relevant to the decision making process and
possess a number of attributes that render them useful
Objectives, resource status, and learning all influence as guides to management (52). To be useful for decision
the choice of management interventions in adaptive making and evaluation, objectives need to exhibit the
management. But objectives also play a crucial role following technical features:
in evaluating performance, reducing uncertainty, and
improving management through time. It therefore is • Specific:
important to have clear, measurable, and agreed-upon Objectives should be unambiguous, with specific
objectives at the outset, to guide decision making and metrics and specific target conditions. Specificity can
assess progress in achieving management success (See be encouraged by articulating objectives with Who,
Case Study 2 for a discussion of setting objectives). What, Why, and/or Where phrases.

The term “objective” is used here to mean some • Measurable:

desired outcome or performance measure that can be used Objectives should contain elements that can be
to guide decision making and measure success. Objectives readily measured, so as to promote the evaluation of
typically are expressed in terms of management perfor- management actions and recognize their contributions
mance over the timeframe of a project. For example, to successful management.
measures might be harvest yield, population size, water
flows, or the probability of a negative impact on resource • Achievable:
status, with an objective of maximizing accumulated Objectives should be based on the capacities of
harvest, achieving a desired population size, maintaining the natural resource system being managed and
water flow, or minimizing a probability of extinction. the political or social system within which manage-
ment occurs.
Because management objectives are used to guide
decisions in managing (and often changing) certain • Results-oriented:
aspects of a target resource through time, they should be Objectives should contain for resource endpoints
more specific than common, “broad-brush” statements and/or conditions representing their achievement.
or overall program purposes that appear in many project For example, a results-oriented habitat objective
documents. For example, generic statements such as might describe the habitat conditions expected when
“provide public access and recreational opportunities” or the objective is achieved.
“improve water quality to enhance and restore commer-
cial fishing” are purpose statements indicating why • Time-fixed:
management is to be undertaken, rather than objectives Objectives should indicate the timeframe for achieve-
that can help to guide decision making. ment, consistent with the duration of the project.
Project implementation may be in stages, but the
Objectives should address the resource issue or overall timeframe should be clear.
problem that initially motivated management, and reflect
the social, economic, and/or ecological values of stake-
holders. Underlying an adaptive approach is the recogni-
tion that stakeholders influence what is to be managed
and under what circumstances. Finding common ground
among disparate and often contentious parties is not an
easy task when there are differences in understanding
about the resource system and differences in ideas about
the desired focus and direction of management. For objec-
tives to be realistic and mutually acceptable, parties must
work toward an agreement on the purpose and approach

24
 It is often the case in adaptive management that Step 2- Key Points
there are multiple objectives for resource management.
For example, one might seek to sustain species richness  Objectives substantively influence decisions and
in a refuge, while attempting to maximize visitor use, management strategies.
maintain a harvest program for one or more species of
wildlife, and allocate resources to these activities so as  Objectives should incorporate the social, economic
to minimize costs. In such a situation it is important to and/or ecological values of stakeholders, and reflect
be able to weigh different objectives in terms of their the value of learning over time.
perceived importance, so as to facilitate the comparison
and prioritization of management alternatives.  To be useful as guides for decision making and evalu-
ation, objectives should be specific and unambiguous,
measurable with the appropriate field data, achievable
but challenging, results-oriented, and applicable over
the timeframe of the enterprise.

Case Study 2: The Trout Creek Mountain Restoration focuses

on compatibility between livestock grazing and critical habitat
for listed Lahontan cutthroat trout within the Trout Creek and
Oregon Canyon Mountains, Oregon. This time sequence shows
the dramatic changes along Cottonwood Creek, V Pasture,
looking downstream starting in 1988. Until 1989 this pasture
was grazed during the summer every year. Despite changes in
the timing and intensity of grazing, chronic trespass prevented
acceptable levels of riparian improvement. Enforcement,
changes in permittees, and fence repairs since then have
allowed 2 years of actual rest. Note the increased reproduction
in aspen by 2002. (See included CD for additional information
on this project).

25
 Step 3- Management actions
Identify a set of potential management actions
for decision making

Like any iterative decision process, decision making

in adaptive management involves the selection of an
appropriate management action at each point in time,
given the status of the resources being managed at that
time. Resource managers and stakeholders, typically
working with scientists, have the responsibility of identi-
fying the set of potential actions from which this selection
is made (See Case Study 3 for a discussion about identi-
fying management actions).

The management alternatives in an adaptive manage-

ment project constitute a key element in its operating
environment, and they can strongly influence the selection
of a management strategy (see Section 5.1). Just as the Case Study 3: The Sonoita Valley Planning Partnership is an
ad hoc, volunteer association of Federal, State, and local agen-
choices made in daily life depend on one’s available
cies, user groups, organizations, and individuals with a common
options, so too are strategy choices in an adaptive interest in the upper Cienega watershed, which includes the
management project constrained by the set of available 42,000 acres of public lands within the Las Cienegas National
options. If these options fail to span a reasonable range of Conservation Area, Arizona. The goal of the partnership is to
management activities or fail to produce recognizable and perpetuate naturally functioning ecosystems while preserving
rural grasslands for future generations. (See CD for additional
distinct patterns in system responses, adaptive manage- information on this project).
ment will be unable to produce effective and informative
management strategies. This argues for careful thinking
about the potential actions to be included in a project.

26
Case Study 4: The Five Rivers Landscape Management Project began in 1998 as an attempt to apply adaptive management at
large scales. The project was designed for 32,000 acres of productive Siuslaw National Forest land in coastal Oregon. (See CD for
additional information on this project).

Management alternatives in adaptive management different management alternatives: (1) passive manage-
often focus on a potential change in resource status or ment in which plantations are allowed to develop into old
the alteration of process rates. Examples of the former growth with no intervention other than road closures; (2)
include culling a livestock herd to maintain a population frequent light-touch thinning and road maintenance; and
at carrying capacity, stocking a lake to sustain a fishery, (3) heavy thinning followed by 30-year road closures.
or withdrawing water from a reservoir to maintain an Another example involves the impact of harvest, which
appropriate volume of water in it. Examples of the latter is likely to be easier (and less costly) to recognize with
include harvest regulations that target an acceptable a few widely spaced harvest rates rather than many that
mortality rate, alteration of nesting habitat to enhance are closely spaced. In both examples a smaller number
population reproductive rate, or control of the amount of alternatives helps to reduce implementation costs,
and timing of visitor disturbance so as to influence avian minimize problems that otherwise can arise with partial
migration patterns positively. controllability (see Section 5.2), and highlight differential
responses of the resource.
In designing an adaptive management project,
management alternatives should be included that will Because of natural variation, resource systems often
produce different responses and thereby promote learning. are extraordinarily difficult to control with management
One way to structure alternatives for this purpose actions, and “cause and effect” relations are usually
is to limit their number, and maximize differences unclear and difficult to recognize. It is important to
among them. An example is the Five Rivers Landscape include options that can help to reduce these difficul-
Management Project (See Case study 4), which used three ties, though this sometimes leads to a broader range of

27
 potential actions than otherwise would be desireable. In
any case, the options under consideration should always
be designed to achieve specific objectives.

To ensure clarity and transparency it is important to

make the management options explicit. Too often the set
of options is unstated, and simply assumed to be recog-
nized and understood by managers and other stakeholders.
Ambiguity as to the alternatives under consideration can
lead to conflict among stakeholders and the possibility of
legal challenges to the decision making process.

The identification of management options is often a

greater challenge than some anticipate. Just as different
stakeholders see the resource system differently and
identify different objectives for its management, so will
they recognize differences in the feasibility and accept-
ability of management options. Even when there is rough
agreement on their nature and extent, there still may be Because of natural variation, resource
different perspectives about the appropriate number and systems often are extraordinarily difficult to
composition of management alternatives. It is important
to take the necessary time and effort to think carefully
control with management actions, and “cause
about these issues, so as to reach agreement on a realistic and effect” relations are usually unclear and
and politically acceptable set of alternatives. difficult to recognize.

Step 3- Key Points

 Potential actions consist of activities under manage-
ment control (for example, harvest, stocking, restora-
tion).

 Alternatives typically focus on alterations of resource

status or process rates.

 The suite of available actions should be designed to

promote learning.

 The alternatives should be explicit and documented.

 Stakeholders should participate in the identification of

alternatives.

28
 Step 4- Models
Identify models that characterize different
ideas (hypotheses) about how the
system works

Models play an important role in virtually all applica-

tions of structured decision making, whether adaptive or
otherwise. In order to make smart decisions, it always is
important to compare and contrast management alterna-
tives in terms of their costs, benefits, and resource conse-
quences. Models typically express benefits and costs as
outputs of management through time. More importantly,
they allow one to forecast the impacts of management.

The term “model” is used here to mean a plausible

representation of a dynamic natural resource system (See
Case Study 5 for a discussion of model development).
Models can be as informal as a verbal description of
system dynamics (for example, a simple description of
reservoir size that is positively influenced by runoff and
negatively influenced by water release), or as formal
as a detailed mathematical expression of change (for
example, an age-structured multi-species model with
density-dependent vital rates that are affected by random
environmental changes). The models used in an adaptive
management project are not restricted to a particular
kind. In many instances only a few models are required
to capture contrasting views about the system, and these
often can be described with limited technical detail.

Case Study 5: In 1995, the U.S. Fish and Wildlife Service

implemented an approach known as adaptive harvest manage-
ment, in which managers seek to maximize sustainable harvests
against a background of various sources and degrees of
uncertainty. (See CD for additional information on this project).

29
 Although model complexity can vary widely An example that highlights many of these
depending on the ecological and management scale of points is the modeling framework used for adaptive
the application, the models used in adaptive management harvest management of waterfowl. Adaptive Harvest
generally share certain attributes (53): Management (AHM) was initiated in 1995 as a process
for setting annual regulations for the sport hunting of
• Resources are described as changing though time, so as waterfowl in North America (37). For AHM, a simple
to allow learning to occur and management to adapt to model was used to account for associations among fall
learning (Fig. 3.1). harvest, seasonal survivorship, and spring reproduction
(Fig. 3.2). Contrasting hypotheses about the impact of
• The resource system is characterized by key components harvest on annual survivorship were easily incorporated
of interest (for example, population size, resource into different versions of the model, by describing
biomass or volume, biodiversity) that are the focus of different functional relations between harvest rates and
management and the targets of monitoring. post-harvest survival. In addition, contrasting hypotheses
about the importance of density dependence in recruit-
• Resource changes often are described in terms of ment were incorporated by describing recruitment in
processes (for example, reproduction, mortality, spatial terms of spring population size. In combination, these
movement) that are thought to be directly influenced by hypotheses define four models, each with its own
management. predictions about harvest impacts and each with its own
measure of confidence that evolves over time (54). The
• Fluctuating environmental conditions are incorporated models and their measures of confidence characterize
as needed to characterize resource dynamics. structural or process uncertainty, which is reduced over
time as harvest actions are taken and post-harvest moni-
• Management impacts are described in terms of costs, toring data are used to update the confidence measures.
benefits, and influences on resource components or
processes that are highlighted in the model.

• Models are calibrated with available data and knowl-

edge, to ensure compatibility with current understanding
about resource structures and functions.
management management management
action action action

resource resource resource

system system system

Models play a key role in representing uncertainty. In

adaptive management, structural or process uncertainty is
captured in contrasting hypotheses about system structure environmental environmental environmental
conditions conditions conditions
and function, and the hypotheses are imbedded in the
suite of models used to forecast resource changes through t-1 t t+1
time

time. At any point, the available evidence will suggest

differences in the adequacy of each of these models to
represent resource dynamics. As evidence accumulates
over time, the confidence placed in each model (and its Figure 3.1. Change in resource system as influenced by
associated hypothesis) evolves, through a comparison of fluctuating environmental conditions and management actions.
model predictions against monitoring data. Management produces immediate returns (costs and/or
benefits) and longer-term changes in resource status.
To be useful, the models in a particular project must
meet certain conditions, including a requirement that
different models predict different outcomes in response
to management. However much two models may differ
in the way they describe system dynamics, if both predict
the same responses to management, then recognizing
which is more scientifically credible will be of little use in
improving management.
30
 Like the management options in an adaptive
management application, models can strongly influence Step 5- Monitoring plans
the identification (and acceptability) of management
strategies. If the models fail to incorporate meaningful Design and implement a monitoring plan to
hypotheses, or fail to produce recognizable differences track resource status and other key
in population dynamics, an adaptive approach may be resource attributes
unable to produce useful and informative strategies. This
argues for engaging managers, resource scientists, and The learning that is at the heart of adaptive manage-
other stakeholders in a thoughtful and deliberate process ment occurs through a comparison of model-based
of selecting the models to be used in an application. predictions against estimated responses based on
monitoring data. It is by means of these comparisons that
monitoring is used to understand resource dynamics, and
thus to confirm the most appropriate hypotheses about
A fall resource processes and their responses to management.
young By tracking useful measures of system response, well
designed monitoring programs facilitate evaluation and
learning in adaptive management.
breeding Ss fall
population adults 1-hy In general, monitoring provides data in adaptive
management for four key purposes:
1-ha
(i) to evaluate progress toward achieving objectives;

Sw (ii) to determine resource status, in order to identify

winter
population appropriate management actions;

(iii) to increase understanding of resource dynamics

via the comparison of predictions against survey
Figure 3.2. Conceptual model of the annual cycle of mallard
data; and
population dynamics. Model includes survival rates for spring-
summer (Ss) and fall-winter (Sw), along with harvest rates for
young (hy) and adults (ha) and age ratio (A) for reproduction/ (iv) to enhance and develop models of resource dynamics
recruitment. as needed and appropriate.

Monitoring programs should be designed from

the outset to inform decision making with data that
are relevant to the management issues in the adaptive
Step 4- Key Points management project (55,56). For example, variables such
as survival, harvest, stocking, and reproduction rates can
 Models in adaptive management should character- be estimated with properly designed monitoring efforts
ize system behaviors and responses to management and used to adaptively manage a biological population.
actions. Surveys producing such information can be invaluable
as sources of longterm data with which to develop the
 Models should incorporate different ideas (hypoth- adaptive management project (See Case Study 6 for a
eses) about how the resource system works and how discussion about developing a monitoring plan).
it responds to management.

 The suite of models should capture key uncertainties

(or disagreements) about resource processes and
management effects.

 Models must be compatible with, and calibrated to,

available data and knowledge.

31
 Monitoring in adaptive management inherits its
focus and design from the larger management context of
which it is a part. Thus, field surveys are not motivated by
scientific curiosity, nor are survey data gathered with only
a vague hope that somehow they will prove useful for
management. Instead, monitoring programs are designed
to focus on the information needed to make manage-
ment decisions and evaluate their impacts. There may
be scientific or other values in broad-scale surveillance
monitoring, and data collected in this way can sometimes
prove useful for resource conservation. But monitoring
in the context of adaptive management is much more
efficient and effective if it targets specific attributes for
the specific purposes listed above. Simply put, the value
of monitoring in adaptive management is derived from its
contribution to adaptive decision making, and monitoring
efforts should be designed with that goal in mind (57).

Step 5- Key Points

 A monitoring plan should be designed to estimate
system state and other attributes needed for decision
making and evaluation.

 The plan should promote learning through a compari-

son of estimates against model-based predictions.

 The plan should be efficient, in that it produces esti-

mates that have maximum precision for a given cost,
or minimum cost for a given level of precision.

Case Study 6: The Bully Creek Landscape Area Management

Project of the Bureau of Land Management is within the Mal-
heur Resource Area, Vale District, Oregon. The project repre-
sents ground-level resource planning for public lands. On the
top are views of Allotment #3 on a tributary of Cottonwood Creek
in 2001 (first year after change in riparian management) and
2005 (after 5 seasons of riparian management). On the bottom
are views of the Muir Spring aspen stand (with protective fenc-
ing) in 2003 (first year after treatment), and 2006, which shows
a close-up of the same site with continued aspen regeneration.
(See CD for additional information on this project).

32
 Iterative phase Step 6- Key Points
At this point in the operational sequence of adap-  At each point in time, selection of a management
tive management it is assumed that the key elements action is made from the set of possible alternatives.
are in place. Thus, the appropriate stakeholders have
been engaged in articulating the scope and nature of the  The selection of a management action is guided by
resource issue. The objectives and management alterna- objectives, which are used to evaluate alternatives
tives of the project have been identified. Forecasting and identify an action that contributes to meeting the
models that capture uncertainty (or disagreement) about objectives.
the impacts of management have been identified. And
a monitoring effort has been designed that targets the  The appropriate action depends on resource status
resource attributes needed for learning, evaluation, and and the current level of understanding about resource
decision making. The stage is now set to incorporate these dynamics.
elements into an iterative decision process that will lead to
improved understanding and management.  Management is adjusted over time as resource condi-
tions change and understanding evolves.

Step 6- Decision making

Select management actions based on
management objectives, resource conditions, Step 7- Followup monitoring
and understanding
Use monitoring to track system responses to
At each decision point in the timeframe of an adap- management actions
tive management project, an action is chosen from the set
of available management alternatives. The management Monitoring is used in adaptive management to track
objective identified in Step 2 is used to guide this selec- system behavior, and in particular to track the responses
tion, given the state of the system and the level of under- to management through time. In the context of adaptive
standing when the selection is made. The appropriate management, monitoring is seen as an ongoing activity,
action is likely to change through time, as understanding producing data after each management intervention to
evolves and the resource system responds to environ- evaluate the intervention, update the measures of model
mental conditions and management actions. That is, confidence, and prioritize management options in the next
management is adjusted both to changing resource status time period (See Case Study 7 for a discussion of follow-
and to learning. It is the influence of reduced uncertainty up monitoring).
(or learning) on decision making that makes the decision
process adaptive. Field surveys, banding programs, remote sensing, and
other monitoring approaches can generate the data needed
There are many ways to design the actual process of to estimate key indicators of resource dynamics (20). The
selecting an alternative. For example, formal optimization information thus produced is used to compare data-based
methods can be used to select from the available manage- estimates of system components against expectations, and
ment alternatives an option that best accounts for current thus to measure performance in achieving management
and future consequences (53,58,59). Alternatively, less objectives. Monitoring information also factors directly
computation-intensive search procedures can be used to into the learning process, through the comparison of
produce suboptimal (but in many cases quite acceptable) estimates against model predictions. Subsequent decision
management strategies. Finally, one sometimes can rely making reflects the increasing degree of credibility earned
on less structured approaches or common sense to iden- by the most appropriate model(s), and management
tify acceptable strategies. Irrespective of the approach, performance is adaptively improved.
decision making should be driven by the management
objectives and informed by resource status and process Monitoring data play into these assessments in
uncertainties. multiple ways. For example, estimates of system state
typically are used in the process of identifying state-

33
 specific decisions. In some cases, estimates of resource intervention can be used to compare population size
status are directly included in the objectives, and thus are before and after the initial intervention. On the other
needed to assess the expected benefits, costs, and conse- hand, an objective of maximizing harvest may not require
quences of particular decisions. Of particular importance baseline conditions prior to the start of management. In
in adaptive management is the use of estimates of status the latter case, decision making pursuant to the objective
and perhaps other attributes for comparison against model is not informed by a comparison of resource status against
predictions so as to improve understanding of resource a starting value. The point here is that the design of a
dynamics. monitoring effort, and in particular the need for initial
“baseline” information, is determined by the nature of the
In many but not all instances, it is useful to collect project and its objectives. Even when baseline informa-
data prior to initiating management. For example, if the tion is needed, its comparative value declines rapidly as
management objective is to increase the size of a previ- the project proceeds through time, essentially because
ously unperturbed population over some time period, monitoring after each intervention establishes new
a “baseline” population size prior to any management “baselines” throughout the life of the project.

Step 7- Key Points

 Monitoring typically occurs after management inter-
ventions.

 Resource status and other key indicators of impacts are

estimated with monitoring data.

 Estimates based on monitoring data are used to evalu-

ate management impacts and inform decision making
Case Study 7: The National Park Service (NPS), State of at the next decision point.
Montana, USDA Forest Service, and the USDA Animal and
Plant Health Inspection Service developed an adaptive man-  Because the amount of monitoring data increases over
agement strategy in 2000 that allows Montana to maintain its
brucellosis-free status and NPS to maintain a bison population
the course of an application, the amount of information
that fluctuates in response to ecosystem processes. (See CD for about system processes also increases.
additional information on this project).

34
 Step 8- Assessment Step 8- Key Points
Improve understanding of resource dynamics  Assessment/analysis includes parameter estimation,
comparative assessments, and prioritization of manage-
by comparing predicted and observed changes
ment alternatives.
in resource status

The information produced by monitoring folds into  Comparison of predicted and actual responses is used
to update understanding of management impacts.
assessments of decision making, performance evaluation,
and learning. For example, the comparison of model
predictions against estimates of actual responses is a  Comparison and ranking of projected outcomes
for management alternatives is used in selection of
key element of learning, with the degree of coincidence
management actions.
between predicted and observed changes used as an
indicator of model adequacy. Confidence is increased in
models that accurately predict change, and confidence  Comparison of desired and actual outcomes is used to
evaluate management effectiveness.
decreases for models that are poor predictors of change.
In this way evidence accumulates over time for the most
appropriate hypothesis about resource dynamics, and
understanding of the resource system is thereby advanced.

As important as it is, learning is not the only role

played by analysis and assessment in adaptive manage-
ment. Thus, an assessment of desired against actual
outcomes can be used to evaluate the effectiveness of
management and measure its success in attaining manage-
ment objectives. In addition, an assessment of manage-
ment alternatives as to their projected costs, benefits,
and resource impacts contributes to the selection of a
management option in the next time period.

35
 As shown in Fig. 3.3, the iterative cycle can begin

Step 9- Iteration with any of the Steps 6 to 8. However, it is useful to think

of it as starting with a management decision, which is
followed by post-decision monitoring and the subsequent
Cycle back to Step 6 assessment of monitoring data. This sequence of activities
is repeated over the course of the application, during
The gain in understanding from monitoring and which learning occurs continuously and the management
assessment in Steps 7 and 8 is used to inform the selection strategy is continuously adjusted based on what is learned.
of a management action at the next decision point. As
understanding evolves, so too does the decision making The cycle typically terminates at the end of the time-
that is influenced by improved understanding. In this way, frame with a final management action that is informed by
the iterative cycle of decision making, monitoring, and assessment of the data collected just prior to the terminal
assessment leads gradually to improved understanding time. It is also possible, if unlikely, that all uncertainty
of resource dynamics, and improved management as a about ecological structure and function can be eliminated
consequence of improved understanding. at some point, whereupon the learning-based adaptive
approach can give way to non-adaptive resource manage-
ment for the remaining time.

… management assessment management assessment …

action action

monitoring monitoring

time

Figure 3.3. Iterative cycle of adaptive management. Management actions are based on objectives, resource status, and learning.
Data from followup monitoring are used to assess impacts and update understanding. Results from assessment guide decision
making in the next time period.

36
 Step 9- Key Points In many applications of adaptive management, both
kinds of learning are of key importance. For example, it
can be as important to understand and track the social and
 The cycle of Steps 6 through 9 is iterated until the end
institutional relations that influence adaptive management
of the timeframe.
elements and stakeholder perspectives, as it is to resolve
technical issues about system structure and process (54).
 Iterations can begin at any point in the cycle; however
Although the motivation of an adaptive approach is to
a natural entry point is with decision making.
improve resource management by reducing structural
uncertainty, its success can be impeded by a failure to
 The direct linkage from assessment to management
adapt to social and institutional changes that inevitably
action in Fig. 3.3 expresses the contribution of learn-
occur over time. Because these changes can themselves
ing to decision making, by providing information on
be a result of early successes in reaching objectives, it is
which to base smart decisions.
important to recognize and if possible account for them as
decision making moves forward.
 The two-step linkage from management action to
assessment in Fig. 3.3 expresses the contribution of
The need to better understand and characterize the
management to learning, through interventions that
elements of adaptive management often becomes more
are useful in investigating the resource system.
pressing as the iterative process of adaptive management
rolls forward. Thus, stakeholder perspectives and values
can shift as the adaptive process unfolds, and previously
Technical and process learning unanticipated patterns in resource dynamics can arise
in adaptive management that require an adjustment of objectives, alternatives, and
other elements of the process. In this sense, learning about
The operational sequence described above provides the adaptive management process extends the context
a framework for implementing adaptive management, of adaptive learning to include changes in institutional
with a focus on reducing structural or process uncertainty arrangements and stakeholder values as well as changes
and thereby improving management. However, adaptive in the resource system.
management involves much more than simply following
the sequence of steps outlined here.

Just as adaptive management can be described in

terms of two phases, learning with adaptive manage-
ment can be seen to occur at two levels. Thus, adaptive
management provides an opportunity to learn not only
about ecological processes, but also about the adaptive
process itself. In Chapter 1, adaptive management was
described in terms of a cycle that included not only
monitoring, evaluation, and adjustment of management
actions, but also problem assessment and design (Fig.
1.1). The latter elements focus on problem formulation,
stakeholder involvement, identification of objectives, and
the other elements included in the setup phase of adaptive
management. In fact, it is useful to think of adaptive
management not only in terms of an ongoing sequence of
decision making, post-decision monitoring, and assess-
ment as described here, but also in terms of periodic
but less frequent recycling through the elements in the
setup phase and adjustment of these elements as needed
to account for evolving stakeholder perspectives and
institutional arrangements (Fig. 3.4). The broader context
of learning that focuses on the components of adaptive
management as well as technical uncertainty is sometimes
called “double-loop” learning (12).
37
 An emphasis on both technical and process learning 3.2. Legal Considerations when Applying
has important implications for the institutional framework
Adaptive Management
for resource management. Many managers contend
that adaptive management is simply common business
In addition to ecological and societal factors, a
practice in their organizations and that they have been
number of legal considerations influence adaptive
using adaptive management all along (51). In fact, the use
management and potentially constrain its success. The
of adaptive management in resource management almost
Federal government is often criticized for the length of
always requires a fundamental shift from the status quo.
time it takes to plan and implement a particular action,
For example, it typically is necessary to rethink the nature
and public concerns increase when an agency reconsiders
of risk aversion that characterizes decision making in
its action and begins planning and compliance activities
most Federal agencies, and to explicitly recognize uncer-
all over again. Well-designed and executed adaptive
tainty as a key attribute of natural resource management.
management strategies can help to alleviate this concern.
Without a willingness to embrace uncertainty, adaptive
management is unlikely to succeed (47).
Agency officials should invest significant effort on
legal issues at two critical stages of adaptive manage-
In addition, adaptive management requires a
ment: (1) at the time a decision is made to utilize adaptive
much more open process of decision making, in which
management for a particular project and (2) at the time
stakeholders are directly engaged and decision making
the agency seeks to adjust management decisions based
authority is shared among them. It also requires that
on the information derived from monitoring and assess-
objectives, assumptions, and the other elements of
ment. Knowing what federal laws and regulations require,
the decision making process be explicit, and therefore
and what limitations apply prior to agency decisions,
amenable to analysis and debate. Finally, it requires a
allows stakeholders to anticipate the legal requirements
strong commitment by managers to the necessary moni-
and integrate them into the adaptive management process.
toring and assessment that underlie adaptive management,
However, agency officials should recognize that some
not as marginal activities but as essential elements of the
laws and implementing regulations prescribe specific
process. It is undoubtedly true that many, perhaps most,
activities and assessments in ways that could limit or even
projects in DOI involve monitoring, and in some cases
preclude the use of adaptive management.
management actually considers the results of monitoring.
But that by itself is a long way from structured, adaptive
decision making in a learning-based environment.

Set-up phase
stakeholders
objectives
alternatives
models
monitoring

Iterative phase
decision making

monitoring

assessment

Figure 3.4. Two-phase learning in adaptive management.

Technical learning involves an iterative sequence of decision
making, monitoring, and assessment. Process and institutional
learning involves periodic reconsideration of the adaptive
management set-up elements.

38
 Application of adaptive management will vary The statutes discussed here are not intended to
depending on geographic setting, ecological complexity provide an exhaustive accounting of applicable law.
and participating stakeholders, as well as the specific However, they are some key authorities that agency
statutory provisions authorizing and sometimes personnel should consider regarding adaptive manage-
constraining actions. Some examples include: ment. Many other statutes such as the Clean Air Act, the
Clean Water Act, and the National Historic Preservation
• National Park Service Organic Act of 1916 Act may be involved in particular applications of adaptive
(applicable to the National Park Service). management.

• Reclamation Act of 1902

(applicable to the Bureau of Reclamation).

• Surface Mining Control and Reclamation Act of 1977

(applicable to the Office of Surface Mining).

• Federal Land Policy Management Act of 1976

(applicable to the Bureau of Land Management).

As a general matter, statutes of this type impose

obligations on a particular agency. In addition, a number
of statutes apply to all Federal agencies and serve to
either limit agency actions or require certain tasks before
an agency takes action. Regulations issued by other
Executive Branch agencies can also limit agency actions.

The use of adaptive management within the

Department of the Interior does not alter the legal context One of the benefits of clean water in the Cold River,
that applies to Federal agency actions. Simply put, all of a salmon stream in Massachusetts
the applicable laws, regulations, and polices continue to
apply to agency actions whether or not adaptive manage-
ment principles are used in a particular context.

Statutory and other authorities that

apply to Interior agencies

The first place to look for the authorized boundaries

of agency activity is in the statutory foundation or appro-
priations authority that supports the activity. Although
few useful generalizations can be made about statutory
requirements, the importance of identifying underlying
authorities cannot be overstated. Employees working on
an adaptive management project should understand the
authority under which they are entering into the project
and be sure to embark on it only when there is clear
statutory authorization.
Native American wickiup structures were listed in 2003,
on Colorado’s Most Endangered Places List by Colorado
Preservation, Inc. The Wickiup Project is supported by the
Bureau of Land Management, Dominquez Archaeological
Research Group and the Colorado State Historic fund.

39
 In the examples below, Departmental employees activities resulting in a Federal action covered by NEPA
are assumed to be actively working with non-Federal must comply with NEPA’s documentation and procedural
stakeholders on an adaptive management project. Most requirements. Given that many Federal actions may be
projects of adaptive management involve such interac- challenged in Federal court, non-Federal stakeholders
tions at some level of the application. should be made aware of the possibility of litigation
inherent in Federal actions.
Statutory authorities that apply
directly to adaptive management An EIS incorporating adaptive management, whether
as a “stand-alone” alternative or part of another alterna-
National Environmental Policy Act. One of the tive, needs to clearly describe how the approach would
most important statutes for an agency to consider as be implemented. This not only includes what types of
it implements adaptive management is the National actions are proposed initially, but also the results that are
Environmental Policy Act (NEPA). The primary goal of expected from monitoring and assessment, and future
this statute is to ensure that agency decision makers and actions that may be implemented based on those results.
the public recognize and account for environmental and Decision makers and the public must be able to see how
other related impacts of proposed agency actions. the adaptive management approach would be imple-
mented, including potential future actions and anticipated
Compliance with NEPA generally requires a series impacts on the environment.
of procedural steps, and certain NEPA processes involve
public participation and public review and comment. In One common challenge to making adaptive manage-
complex or controversial situations, NEPA potentially ment work in natural resource decision making is that
involves the preparation of a Notice of Intent to prepare ongoing monitoring may reveal “new, significant informa-
an Environmental Impact Statement (EIS), as well as tion” that requires an agency to prepare a Supplemental
a Draft EIS, a Final EIS, and a Record of Decision Environmental Impact Statement. This requirement
assessing environmental impacts of major Federal actions. is triggered when “[t]he agency makes changes in the
In less complex or controversial actions, environmental proposed action that are relevant to environmental
compliance often can be accomplished with a simpler concerns; or [t]here are significant new circumstances
Environmental Assessment (EA) that culminates in a or information relevant to environmental concerns and
Finding of No Significant Impact. In some cases, an bearing on the proposed action or its impacts” (40 CFR §
action can be categorically excluded from the requirement 1502.9 (c)). If management adaptations that could occur
for NEPA compliance. in light of new information are fully documented and
analyzed at the beginning of a NEPA process, the need to
The NEPA requires an EIS of proposed “major supplement NEPA documents may be reduced. Put differ-
Federal actions significantly affecting the quality of the ently, if an EIS anticipates significant information that can
human environment.” An EIS must include an analysis of arise from monitoring and assessment, the agency may
alternatives to a proposed action. The activities resulting not need to supplement the EIS when invoking manage-
from a particular adaptive management process may rise ment changes based on the newly acquired information.
to the level of a major Federal action requiring an EIS,
and in any event they likely will need to be analyzed
for NEPA compliance. Less complex or controversional
actions may be addressed by a less comprehensive EA.
Under NEPA, following the completion of an EA the
agency will either identify significant impacts (and
prepare an EIS), or prepare a Finding of No Significant
Impact. Of course, some actions can be categorically
excluded from NEPA’s documentation requirement.

As an example, a proposal resulting from collabora-

tion between a DOI agency and an environmental
nonprofit organization might call for the nonprofit to
conduct habitat improvement for a protected species. It is
possible that this undertaking could be considered a major
Federal action. In such a circumstance, the collaborative
40
An important statute for an agency to consider as it
implements adaptive management is the National
Environmental Policy Act . The primary goal of
this statute is to ensure that agency decision makers
and the public recognize and account for
environmental and other related impacts of
proposed agency actions.

41
 Endangered Species Act. The Endangered Species and effective consultation is an initial description of the
Act of 1973 (ESA) provides a broad, comprehensive range of potential adaptations and effects of those actions
approach to the conservation of threatened and endan- on listed species and their designated critical habitats.
gered species. By Congressional direction, the ESA is Re-initiation of consultation is far less likely to be needed
administered by the U.S. Fish and Wildlife Service (FWS) if the initial consultation clearly considers the action to be
and the NOAA-Fisheries (National Marine Fisheries adaptive and addresses the full range of possible adapta-
Service). In general, the FWS deals with terrestrial and tions and their associated potential effects.
fresh water species, while NOAA-Fisheries deals with
anadromous fish and marine species. As part of their In some cases involving large-scale Federal programs,
administration of the ESA, these agencies: consultation is appropriate at both a broad programmatic
level as well as the level of individual projects or actions
• “List” endangered and threatened species that may affect listed species. Careful consideration of
effects and alternatives can set the stage (for example,
• Designate critical habitat for listed species through adoption of best management practices or design
standards) for expedited consultation on later individual
• Publish plans to identify actions needed to assist in the actions.
recovery of listed species
Agencies whose actions may affect listed species
• Consult with other federal agencies whose actions may should design monitoring programs with input from FWS
affect listed species and/or NOAA-Fisheries. Learning by doing - the critical
centerpiece of adaptive management - is particularly
• Work with non-federal entities to develop and approve important in ESA situations, where cause and effect can
habitat conservation plans be particularly difficult to ascertain. New information on
listed species, or the effects of actions on listed species,
• Work cooperatively with other nations to conserve may require re-initiation of Section 7 consultation by a
listed species Federal agency, or may trigger changes in habitat protec-
tion pursuant to approved Habitat Conservation Plans
• Administer international agreements to limit trade in (HCP). Knowing and understanding the ESA and its
endangered and listed species requirements will be essential to successfully integrating
the elements of an adaptive management program with
Of particular relevance to Federal agencies engaged efficient ESA procedural compliance.
in activities that may affect listed species is Section 7 of
the ESA. Section 7 enlists agencies of the Federal govern- Integration of adaptive management principles has
ment to support species conservation and avoid actions been utilized by the FWS in the context of HCPs under
that would contribute to species extinction. Section 10 of the ESA. The FWS has developed guide-
lines regarding this aspect of HCP planning (included in
Given the importance of conserving endangered and the reference CD). Particular attention should be given
listed species, the complexity associated with protecting to the issue of which party is to be responsible for any
these imperiled species, and the impacts the ESA may required changes in mitigation and/or minimization
have on society and agency decision making, any adap- of “take” to listed species as a result of monitoring
tive management program that may affect listed species or programs. Any effort in this regard must take into consid-
critical habitat is more likely to be successful if it involves eration the No Surprises assurances that a permittee can
FWS and/or NOAA early in the process. Key to efficient receive pursuant to the regulations that implement Section
10(a)(1)(B).

42
The Endangered Species Act’s purposes include providing
a means to conserve the ecosystems upon which endangered
and threatened species depend, and providing a program
for the conservation of the species

43
 Federal Advisory Committee Act (“FACA”). Under Antideficiency Act. The Antideficiency Act contains
FACA, Department officials may not receive advice a series of controls over the use of Federal appropriated
from a group that the Department has established or that funds to ensure that Federal agencies “pay as they go.”
it uses (i.e., manages or controls) unless the Department Government officials are prohibited (without specific
complies with the provisions of FACA. The FACA authority) from making payments, or committing the
requires certain actions to set up and operate a committee United States to make payments at some future time,
or similar group to provide advice to Federal officials. unless there are available agency funds to cover the cost
FACA does not require any particular outcome regarding in full. The Antideficiency Act applies to applications of
the substance of advice on a particular matter. Rather, it adaptive management just as it does to all other Federal
establishes a number of actions and approaches to ensure activities. In essence, no agreement should be entered
balanced consideration and input. These actions include into that commits an agency to the payment of funds in
filing a charter, providing public notice of meetings in the future, in advance of available appropriations to fund
the Federal Register, and making advisory committee activities under the agreement. For example, an agree-
information publicly available. ment by an agency that commits $100,000 in grant funds
to a particular organization for each of the next 5 years
Federal officials who receive advice from non-federal will likely be improper under the Antideficiency Act,
stakeholders should be aware of FACA’s potential appli- unless there are sufficient agency funds that are available
cability. However, FACA does not apply to every situa- for the grant for more than one year.
tion in which a Departmental official receives advice, but
only to those situations in which the advice comes from Other relevant statutes and authorities
a group that the Department has established or utilized.
This means that FACA does not apply to advice received Administrative Procedure Act (APA). The APA
from individuals, even in a group setting (such as “town provides a procedure by which Federal agency actions
hall” meetings). Nor does it apply to advice received from may be challenged in court. Although there are several
preexisting groups, or groups that the Government neither ways to challenge an agency action under APA, the most
manages nor controls. It does not apply to groups that commonly employed is the claim that an agency action is
simply exchange facts or information; or groups that are “arbitrary, capricious, an abuse of discretion, or otherwise
authorized to carry out operational functions; or groups not in accordance with law.” This standard applies equally
consisting of only Federal, state, local, and tribal govern- to adaptive management projects as to other activities,
ment employees exchanging views, information, or advice so agency decisions in a particular adaptive management
on programs with shared intergovernmental responsibili- application may ultimately be reviewed by a court. To
ties. Finally, under some circumstances groups may be survive a court challenge, agency decisions must therefore
exempted from the requirements of FACA by another be rational, reasonable, and carefully articulated.
statute (such as the ESA for Recovery Implementation
Teams). For more information on FACA, employees APA challenges are usually decided on the basis
should consult FACA regulations in 41 C.F.R. Part 102-3. of the administrative record (the materials upon which
the agency officials relied in taking action). Challenges
Funding authorities to activities in adaptive management applications are
no different. Departmental employees must therefore
Annual Appropriations Acts and Funding. Agency ensure that their decisions and actions are based on, and
activities cannot be undertaken without available funds supported by, a complete and thoroughly documented
that are allocated for a particular purpose. Occasionally, administrative record. Finally, the APA may also provide
Congress may insert specific limitations in annual the basis for challenges under NEPA, the ESA, and other
appropriations acts that could affect agency activities. statutes.
For example, Congress may enact a provision stating
that no Federal funds may be spent to map natural gas Substantive Statutory Authority. Often a statute that
deposits off the coast of California or to study decommis- authorizes an activity will also contain specific limita-
sioning of the Glen Canyon Dam. In these situations, any tions. For example, a statute that authorizes the Secretary
Departmental effort to map deposits or undertake such to establish a wildlife refuge in cooperation with a State
a study would be impermissible, even if a non-Federal may also require the Secretary to ensure that State laws
stakeholder thought the activities would assist a particular apply on the refuge. Conversely, Congress may enact
adaptive management process or a non-Federal partner statutes that forbid a particular activity. For example,
donated the necessary resources and/or data for the Congress has forbidden Federal agencies from creating
efforts. business corporations.
44
 Lobbying Activities. Under 18 U.S.C. § 1913 and
related authorities, most Department employees may not
expend appropriated funds for “grass roots” lobbying that
is designed to influence a member of Congress or official
of any government (Federal, State, local, tribal) regarding
his or her position on legislation. That is, Federal
employees may not engage in lobbying. This restriction
prohibits encouraging a stakeholder to undertake lobbying
activities in support of the Department.

Freedom of Information Act. The Freedom of

Information Act (FOIA), codified in 5 U.S.C. § 552,
provides members of the general public potential access
to any information created or obtained by the Department
and under the Department’s control at the time of a FOIA
request. Department personnel should keep in mind
that documents generated during adaptive management
activities will generally be agency records that are subject
to release under the FOIA. However, the Department may
withhold documents from the public that fall within one
of nine specified FOIA exemptions. Agency personnel
will want to consider how they intend to gather, store and
publish information developed as part of the adaptive
management process, and work with their agency’s FOIA
offices when responding to requests for information.

Data Quality Act. Although relatively new, the Data

Quality Act (DQA) is increasingly affecting the work of
Interior agencies. Passed in 2001 as part of the Treasury
and General Government Appropriations Act, the DQA
directs the Office of Management and Budget to publish
guidelines applicable to Federal agencies, and provide
policy and procedural guidance to Federal agencies for
ensuring and maximizing the quality, objectivity, utility,
and integrity of information (including statistical informa-
tion) disseminated by Federal agencies. Each agency
should assess the applicability of the DQA to new or
ongoing adaptive management projects.

Governmental Integrity. Because agency decisions,

whether products of adaptive management or not, will
only stand up if they comport with the law and are
not “arbitrary and capricious,” agency officials must
constantly work to ensure that the agency’s decisions
are supportable based on the information that is avail-
able to the agency. In carrying out agency activities,
Departmental employees should always act in an
unbiased, fair, and equitable manner, so as to ensure that
the public will not have cause to question the integrity of
the Federal government. All Federal employees should
endeavor to act impartially and to avoid giving preferen-
tial treatment, or the appearance of preferential treatment,
to any private entity.

45
 Making Adaptive Management Work:
The Importance of Integrating Adaptive Assessments could help to integrate NEPA and adaptive
Management and NEPA management, especially if environmental analyses will be
needed subsequent to an initial EIS and impacts are not
NEPA serves as a pre-decisional requirement and expected to be significant. An Environmental Assessment
analytical process with the goal of ensuring that both the is generally much easier to prepare than an EIS.
public and Federal decision makers are fully aware of
the potential impacts of a discretionary Federal action. If an agency is constantly undergoing NEPA compli-
Expanding the NEPA framework to accommodate the ance and documentation, it may be because the original
iterative, data-driven process of adaptive management NEPA process failed to fully consider a broad range of
requires the integration of learning-based strategies into potential adaptive management modifications. Frequent
the existing framework of NEPA requirements and the follow-up NEPA compliance can erode the public’s
implementing regulations. Of course, compliance with perception of the integrity of the process, and impede the
other environmental statutes, regulations, and Executive ability to react to new, relevant information. NEPA is
Orders is also necessary. first and foremost a public process, and its credibility and
successful implementation is predicated on the public’s
Federal agencies are encouraged to use adaptive perception that it is transparent and involves full disclo-
management as part of the NEPA planning process, sure of the potential environmental effects of Federal
particularly in circumstances where long-term impacts are actions.
uncertain and informed decisions in the future will depend
on monitoring and assessment at that time. Any adaptive
management alternatives in an Environmental Impact
Statement can be crafted to allow the necessary flexibility
for strategy adjustment as learning advances through
monitoring and assessment. Agencies are encouraged to
build this flexibility into their management alternatives
and NEPA compliance activities. Training of NEPA
practitioners in this important environmental concept is
paramount.

Because NEPA analysis and compliance is proce-

dural and predictive, and includes the evaluation of “all
reasonable alternatives,” it is important to include a
sufficiently broad range of management options to ensure
a comprehensive environmental analysis and a “NEPA
coverage” that is adequate for whatever contingencies are
anticipated. The idea is to satisfy the NEPA requirements
for scientific analysis of potential impacts, and provide
full disclosure of those impacts. Thoughtful and carefully
structured alternatives should avoid the triggering of
additional NEPA compliance later.

Alternatively, another approach to NEPA

compliance that has proven successful for adaptive
management programs is to prepare a “programmatic”
EIS at the start, which broadly covers the likely range of
actions that may be taken under the particular adaptive
management program. Later, any NEPA compliance
needed for subsequent shifts in the management actions
as a result of the adaptive management process can then Coyote Gulch
“tier” off of the initial programmatic EIS, saving consid-
erable time and work.

Where appropriate, greater use of Environmental

46
 3.2 Key Points
 Adaptive management must be integrated with all
existing legal obligations of the agency; it is not a
replacement for environmental compliance.

 Adaptive management must comply with NEPA and

other environmental laws. Integration of adaptive
management and other legal obligations requires
thoughtful “up-front” planning, and involves an
investment of time and resources by the agency and
other stakeholders.

 Integrating environmental review procedures with

adaptive management requires consideration of the
range of adaptive actions and attendant environmental
effects that can reasonably be anticipated at the time
of the environmental review.

 The management alternatives and effects considered

in an adaptive management application must be
reviewed in light of the relevant environmental laws
and regulations, so that environmental compliance
applies not only for the initial action, but also for
adaptive redirections that may be needed in the future.

 For some adaptive management applications, it may

be appropriate to assess environmental effects of
future adaptive actions on a case-by-case basis using
streamlined environmental assessments or informal
consultations. Such an approach may serve to
strengthen the case for compliance when uncertainty
exists regarding the environmental effects of future
adaptive actions.

 Key to successful integration of NEPA and adaptive

management is a well planned and thorough up-front
consideration of the range of potential actions and
their effects, so as to ensure that future actions and
their effects are within the scope of the initial analysis
and do not require subsequent environmental analysis.

47
48
Chapter 4: When is Adaptive Management Successful? 4
Chapters 1 and 2 address the context and conditions described in the nine-step operational sequence in Chapter
for adaptive management to be applicable to resource 3. Specifically, stakeholder involvement, an effective
management, and Chapter 3 describes adaptive manage- monitoring program, and agreed-upon objectives,
ment in operational and legal terms. In these chapters management alternatives, and models must be integrated
we referred to the successful implementation of adaptive into an iterative learning cycle (see Fig. 3.4). Of course,
management without ever explicitly defining success. the decision making process must be framed in a context
Here we offer a definition, criteria, and steps to promote of applicable laws, authorities, and regulations.
successful implementation.
Based on this definition, we suggest the following
4.1. Recognizing Success four criteria for recognizing success in adaptive manage-
ment (Fig. 4.1):
in Adaptive Management
• Stakeholders are actively involved and committed
In general, the implementation of adaptive manage-
to the process.
ment is defined as successful if progress is made toward
achieving management goals through a learning-based
• Progress is made toward achieving management
(adaptive) decision process. This definition contains
objectives.
two essential elements. First, it requires progress toward
achieving objectives, a primary indicator of success
• Results from monitoring and assessment are used to
with any management strategy, whether adaptive or
adjust and improve management decisions.
not. Second, it requires learning-based management, as
• Implementation is consistent with applicable laws.

Adaptive Management Success Model

adaptive management success

stakeholder involvement progress toward achieving informative monitoring implementation consistent

and support resource objectives and assessment with applicable laws

success success success success

factors factors factors factors

Figure 4.1. Adaptive management success model, with four success criteria. Success factors for each criterion are addressed by a
series of questions that help practitioners increase the likelihood of success.

49
 These criteria integrate the structural elements and
processes described in the preceding chapters. They are Including stakeholders in the adaptive management
interrelated and interdependent, and should be viewed process reinforces stakeholder perceptions of adaptive
collectively as indicators of success. management as a legitimate process, which in turn
encourages cooperation and reduces the likelihood of
For an adaptive management project to be successful, conflict. Stakeholder involvement in problem identifica-
all four criteria must be met over the project timeframe. tion, process design, monitoring and assessment, and
For example, an adaptive management project is not other elements of the adaptive management process builds
considered fully successful if stakeholders do not see the support for the process and provides a foundation for
resource management process as legitimate. Likewise, learning-based resource management. It also provides
results from monitoring and assessments must be used to opportunities for resource managers to obtain additional
inform adjustments in management practices for it to be information about the natural system and priorities for
fully successful. its management before decisions are made. Conversely,
a lack of stakeholder involvement can by itself cause the
Stakeholders are actively involved process to fail.
and committed to the process
Progress is made toward achieving
Broad stakeholder involvement is critical for adap- management objectives
tive management success. Recall from Chapter 3 that
engaging stakeholders is one of the key steps in adaptive In structured decision processes, clear and measurable
management. Ideally, stakeholders are engaged in objectives guide decision making and serve as metrics for
every aspect of adaptive management, from the initial assessing management performance. Because adaptive
problem formulation to the identification of objectives management by design pursues management benefits as
and the design of monitoring and assessment. The fact expressed in the management objectives, progress toward
that stakeholders play a role in all aspects of the process achieving objectives is a natural criterion of success.
argues for singling out stakeholder involvement as a key
success criterion.

50
 Two points about objectives are emphasized here. Results from monitoring and assessment
First, the objectives in natural resources management are used to adjust and improve
often are multidimensional, and thus involve tradeoffs. management decisions
Improving an outcome associated with one objective may
involve tradeoffs with outcomes associated with other The accumulation of understanding and subsequent
objectives. When multiple management objectives are adaptation of management strategy depends on feeding
identified, it is important to recognize and account for monitoring and assessment results back into the decision
the relationships among them, so that potential tradeoffs making process. Monitoring and assessment efforts
can inform decision making. In this way, consistent and should be designed to ensure that key resource parameters
achievable resource management goals and thresholds can are adequately measured and appropriately focused, so
be established. as to contribute to achieving success. In Chapter 3 these
key process elements were seen to factor directly into the
Second, it is important to recognize that manage- operational sequence that defines adaptive management.
ment objectives may change. In many cases, success
in attaining objectives can be an ongoing process that Here we emphasize the importance of monitoring
involves refinement of objectives as understanding accu- resource responses and using the resulting information to
mulates and stakeholder perspectives change. That is, the assess system models, update their confidence measures,
adaptive management system is itself dynamic, including and reduce system uncertainty. Scientists, managers, and
its objectives. Adaptive management needs to include not stakeholders should collaborate in an interdisciplinary
only the cyclic evaluation of project performance, but also assessment of what is known and what is learned about
a periodic reassessment of the project objectives. the system being managed. Success in adaptive manage-
ment ultimately depends on effectively linking monitoring
and assessment to objective-driven decision making.

51
 Implementation is consistent The success model provides a framework for both
with applicable laws technical and process learning, as described in Section
3.1. The focus is on adaptation not only at the technical
It is almost axiomatic that implementation of adaptive level, through an iterative application of decision making,
management must be consistent with applicable laws to monitoring, and assessment, but also at the institutional
be considered successful. The use of adaptive manage- and process level, through periodic assessments of stake-
ment in a manner that is not consistent with applicable holder perspectives, management objectives, and other
laws will eventually lead to distractions from the project process elements. The successful application of adaptive
objectives, or to a breakdown in trust among stakeholders, management potentially involves learning and adaptation
or to a litany of official – and unofficial – sanctions. at both levels. The following section includes questions
that can be useful in implementing and evaluating the
Ensuring that an adaptive management application success of an adaptive management project.
complies with applicable laws, regulations, and policies
can be challenging because of the many legal consider-
ations and complexities that may be involved. Thoughtful,
detailed planning and constant attentiveness to the
requirements of laws, regulations, and policies are needed
to ensure the success of an adaptive management project.

This criterion should not be assessed in isolation.

In large measure, success with the previous criteria
will affect stakeholder assessments of whether an adap-
tive management project remains in compliance with
applicable legal considerations. For example, if serious
disagreements persist among stakeholders regarding the
appropriateness of monitoring protocols and results, one
or more of the stakeholders may challenge the validity of
the program’s results. This in turn can lead to withdrawal
of stakeholder support, loss of continued funding, and
even litigation.

Using the success model

A key success metric that adaptive management

shares with all management strategies is progress toward
the achievement of management objectives. As an adap-
tive management project moves forward in time, it should
be possible to compare management performance against
expectations, to get some idea of the progress made up to
that time in achieving objectives. In the absence of at least
some indication of progress over a reasonable period,
there is little justification for continuing a project in its
current form.

All four success criteria need to be met for successful

implementation of adaptive management. They should be
considered when initially developing an adaptive manage-
ment project, and reviewed periodically over the life of
the project to evaluate project implementation and assess
the likelihood of achieving objectives. In fact, the success
criteria can themselves be used adaptively, to determine
where changes are needed in order to achieve success.
Avian Influenza Sampling Project 2006

52
 Box 4.1 Adaptive Management
Operational Steps

Set-up phase

 Step 1 - Stakeholder involvement

Ensure stakeholder commitment to adaptively manage
the enterprise for its duration

 Step 2 - Objectives
Identify clear, measurable, and agreed-upon
management objectives to guide decision making
and evaluate management effectiveness over time

 Step 3 - Management actions

Identify a set of potential management actions for
decision making

 Step 4 - Models
Identify models that characterize different ideas
(hypotheses) about how the system works

 Step 5 - Monitoring plans

Design and implement a monitoring plan to track
resource status and other key resource attributes

Iterative phase

 Step 6 - Decision making

Select management actions based on management
objectives, resource conditions, and enhanced
USGS scientist is measuring various water-quality conditions in understanding
Holes Creek at Huffman Park in Kettering, Ohio.
 Step 7 - Follow-up monitoring
Use monitoring to track system responses to
management actions
4.2. Key Questions for Successful
Implementation  Step 8 - Assessment
Improve understanding of resource dynamics by
The following questions can be used to increase the comparing predicted vs. observed change in
likelihood of successful adaptive management, and to resource status
evaluate progress in achieving objectives. The questions
address each of the criteria identified in Section 4.1 and  Step 9 - Iteration
are tied to the operational steps discussed in Section Cycle back to Step 6 and, less frequently, to Step 1
3.1 and shown in Box 4.1. The questions should be
considered throughout the timeframe of an adaptive
management project, to ensure that success is continually
considered and evaluated as the project progresses. A
particular adaptive management project may not be able
to address all of the issues outlined in the following ques-
tions. However, the questions encompass an important set
of issues that should be considered, so as to increase the
likelihood of success.
53
 Step 1- Stakeholder involvement Step 2- Objectives
Two key aspects of stakeholder involvement are Objectives in adaptive management play important
emphasized here. First, participation by stakeholders roles in decision making, evaluation, and learning. These
is essential if only because stakeholder perceptions roles are enhanced by the articulation of clear, measure-
of adaptive management as a legitimate process can able, and agreed-upon objectives. Because they can
promote cooperation and reduce the likelihood of conflict. change through time as the resource system changes and
Second, including stakeholders in the process provides stakeholder values evolve, it is useful to revisit objectives
opportunities for resource managers to obtain additional periodically.
information about both the natural system and stakeholder
values and priorities before decisions are made. Involving Keeping the following questions in mind will increase
stakeholders builds support for the process and provides a the likelihood that decision making will be guided by
foundation for learning-based resource management. resource management objectives:

Key questions to consider when involving stake- • Have explicit and measurable management objectives
holders in an adaptive management project include: been identified and developed?

• Are the management objectives achievable and

• Has a systematic process been developed that facilitates sustainable?
effective participation by stakeholders?
• Have performance metrics relating to the management
• Have key stakeholders been identified? objectives been developed?

• Have agreed upon lines of communication been estab- • Has a system of monitoring and assessment relevant to
lished and is their importance to successful adaptive the management objectives been developed and imple-
management processes understood? mented so that progress in meeting the objectives can be
tracked?
• Are stakeholders committed to and involved in the adap-
tive management process including the monitoring and • Have tradeoffs among management objectives been
assessment program? considered and are they understood?

• Is the adaptive management process able to adapt to

changes in stakeholder and public viewpoints?

54
 Step 3- Management actions Step 4- Models
The set of potential management actions determines Models serve as expressions of ecological under-
the range of management flexibility for an adaptive standing, as engines for deductive inference, and as
project, and influences learning rates as well as progress articulations of resource response to management and
in achieving management objectives. Learning is environmental change. They also help bring together
promoted by a wide range of management alternatives, scientists, managers, and other stakeholders in a joint
but hampered by alternatives that differ only marginally. assessment of what is known about the system being
As with other components of an adaptive management managed, and facilitate an interdisciplinary approach to
project, the acceptable range of management options can understanding through monitoring and assessment. It is
change through time, as stakeholder perspectives and important to keep the many roles of models in mind as an
resource conditions evolve, legal requirements change, adaptive management project is implemented.
and new information becomes available.
The following questions relate to the models used in
Key questions about the set of management alterna- an adaptive management project:
tives in an adaptive management project include the
following: • Are the hypotheses underlying the strategies for resource
management expressed as testable models?
• Has a range of potential management actions
been developed? • Have explicit links between management actions and
resource dynamics been incorporated into the models?
• Have the specific tasks to implement the management
alternatives been identified? • Are the ecological/resource processes that drive resource
dynamics understood?
• Is the range of potential actions appropriate for the
timeframe under which changes are likely to occur? • Are the relevant environmental factors incorporated into
the models?
• Can the set of management alternatives be adjusted
through time if needed? • Are the models calibrated with available information?

55
 Step 5- Monitoring plans Step 6- Decision making
The use of objectives to guide decision making An adaptive management strategy identifies actions
depends on linking monitoring and assessment results at each point in time, with strategy implementation at
with the decision making process. Both monitoring and a particular time based on resource status and under-
assessment should be designed to ensure that resource standing. The strategy typically evolves through time, as
parameters are adequately measured and appropriately learning accumulates and the resource system responds to
focused on relevant performance indicators. Effective and management actions.
useful monitoring is required for the hypothesis testing
that leads to the reduction of uncertainty that is key to Key questions about iterative decision making in an
adaptive management. adaptive management project include the following:

Important questions about the monitoring plan for an • Is it clear how decisions will be made?
adaptive management project include the following:
• Are decisions at each point in time based on the current
• Will the monitoring plan support the testing of alterna- status and understanding of the resource?
tive models and measurement of progress toward accom-
plishing management objectives? • Are decisions being guided by management objectives?

• Is it clear what monitoring data need to be collected to • Are stakeholders informed and consulted before deci-
estimate the relevant resource attributes? sions are made or changed?

• Has the level of accuracy that is needed been identified?

Step 7- Follow-up monitoring
• Are commitments among managers, scientists, and other
stakeholders in place to sustain an ongoing monitoring
and assessment program? The effects of decision making are tracked with
post-decision monitoring, and the data collected are
• Will meaningful and useful data and information be used to gauge progress in meeting objectives and
available within timeframes that allow for adaptive deci- improve ecological understanding. Ideally, post-decision
sion making? monitoring data can be folded into analysis/assessment
before the next decision point, so that decision making at
that time can take advantage of updated information and
understanding.

Key questions about follow-up monitoring in an

adaptive management project include the following:

• Are the analysis needs understood?

• Is monitoring conducted on a timely basis?

• Is monitoring targeted to system attributes that are

useful for evaluation and learning?

• Are monitoring data collected and managed so that they

are available and easy to access?

• Can the monitoring data be used to update the measures

of model confidence?

56
 Step 8- Assessment • Are the targets identified in the performance metrics
likely to be achieved within the specified timeframe?

The information from monitoring is used to evaluate

Legal considerations
management, improve understanding, and guide deci-
Because of its scope and complexity, the legal
sion making. As noted in Chapter 3, these functions are
framework for an adaptive management project can be
promoted by the estimation of resource parameters and
confusing, and it is not always obvious to managers
the comparison of these estimates against model-based
and practitioners which legal considerations to focus
predictions.
on. Among other things, this suggests the usefulness of
engaging members of the DOI Solicitor’s Office early on
Key questions about analysis and assessment in an
in an adaptive management project.
adaptive management project include the following:
Key questions about legal considerations for an
• Have the expected impacts of alternative management
adaptive management project include the following:
strategies been evaluated?
• Are the applicable resource management laws and
• Is it clear how results are to be understood and
regulations understood by managers, scientists, and
interpreted?
other stakeholders?
• Have thresholds that indicate a change in management
• Have steps been taken to comply with applicable laws
been recognized?
and regulations?
• Have the action(s) to be taken when a threshold is
• Has a process been developed to focus on compliance
reached been identified?
throughout the project life?

Step 9- Iteration • Is the process in compliance with specific legal

mandates?

The adaptive cycle of decision making, post-decision

monitoring, and analysis/assessment (Fig. 3.3) leads to
improved understanding as well as improved manage-
ment. Periodic but less frequent cycling through the
components of an adaptive management application (Fig.
4.2 Key Points
3.4) allows for adjustments as stakeholder perspectives,
institutional arrangement, and resource conditions evolve.  Adaptive management allows managers to determine
systematically whether management activities are
succeeding or failing to achieve objectives.
Key questions about iterative feedback in an adaptive
management project include the following:
 An adaptive management project is recognized as
successful if (1) stakeholders are involved and com-
• Are management actions and decisions reviewed fre-
mitted to the process; (2) progress is made toward
quently based on monitoring and assessment?
achieving management objectives; (3) results from
monitoring and assessment are used to adjust manage-
• Have incentives been developed to encourage experi-
ment decisions; and (4) implementation is consistent
mentation and learning?
with applicable laws.
• Have resource management alternatives been revisited
and/or modified over time?  The implementation of adaptive management can be
facilitated by considering a series of questions related
to the success criteria and the operational steps.
• Has uncertainty related to resource dynamics and the
impacts of management actions been reduced through
learning over time?

57
 Rather than relying on anecdotal information
about resource status, managers can use
monitoring and evaluation adaptively
to make smart decisions

58
Chapter 5: Other Operational Issues 5
In the course of writing this technical guide, contribu- decisions, and anecdotal information about resource status
tors and reviewers raised a number of important issues is all that is needed to inform those decisions.
that do not fit naturally into other chapters, but neverthe-
less merit discussion. Because no document can address Management based on resource status
all the issues that might arise in adaptive management,
DOI bureaus and offices may wish to develop their own Here the focus of decision making is on achieving
planning and implementation guidelines, tailored to management objectives, with little or no recognition
specific legal and institutional contexts and focused more of uncertainty in the decision making framework.
directly on relevant authorities. Monitoring and assessment focus primarily on resource
status, rather than the understanding of ecological
5.1. Uses of Information in processes. This approach is sometimes misidentified as
adaptive management, presumably because the measures
Natural Resource Management of resource status obtained through monitoring are
considered in management actions. A great many multi-
When considering the application of adaptive
year resource applications are of this kind. However, few
management, it is important to account for both learning
of these applications specifically focus on learning about
and progress in achieving management objectives, as
the processes that control system dynamics.
well as the possible tradeoffs between them. As indicated
earlier, learning in adaptive management occurs through
the comparison of model-based predictions against
Passive adaptive management
information from monitoring. The role played by moni-
In this case uncertainty is recognized in the decision
toring, and the information produced from monitoring, is
making framework, but the focus is on the achievement
essential in adaptive management.
of management objectives, with learning as an untargeted
byproduct. Ongoing monitoring programs focus on
Several different approaches to resource manage-
resource status as well as other system attributes that are
ment can be distinguished, depending on the relative
useful for improved understanding through time, and
emphases on learning and management objectives (38).
assessment produces estimates of resource attributes that
Management approaches can range from an exclusive
are used for learning. Because decision making is not
focus on management objectives with no concern for
focused specifically on learning, the rate of learning is
information and learning, to an exclusive focus on
likely to be substantially lower than with a more proactive
learning with little regard for achieving management
approach.
objectives (60). The most extreme example of the latter is
the use of management in a rigorously designed experi-
ment, where the goal is to maximize the precision of Active adaptive management
contrasts among management treatments.
Decision making involves the active pursuit of
learning, either through experimental management that
Management in the absence of
focuses directly on learning, or quasi-experimental
systematic monitoring
management that focuses simultaneously on learning
and achievement of management objectives. Both
In this situation, decision making is loosely focused
approaches anticipate the effect of management on the
on management objectives, and is based on prior
rate of learning, and both are included under the rubric
experience, intuition, expert opinion, etc. Monitoring
of “management by experiment.” Monitoring focuses on
and assessment are not used systematically in decision
resource status as well as other system attributes needed
making, so there is little or no opportunity for learning.
to improve understanding through time, and assessment
This situation occurs more frequently than many believe.
produces estimates of resource attributes that can be used
For example, managers often feel that their understanding
for learning.
of a resource system is sufficient for them to make smart

59
 It should be emphasized that both active and passive
adaptive management utilize management interventions
in a learning process. The key distinction between the
two approaches is the degree to which decision makers
anticipate the influence of management on learning, and
the degree to which management is used proactively to
accelerate the rate of learning.

Figure 5.1 orders the management approaches

described above with respect to their emphases on
learning. Several points can be made. First, adaptive
approaches to management place a greater emphasis
on uncertainty and learning than non-adaptive decision
making. Second, non-adaptive management is oriented
solely on management objectives, whereas adaptive As a general rule, adaptive
management considers learning as well. Third, passive management is most useful when
adaptive management places a stronger emphasis on the consequences of management are
learning than non-adaptive management. As a general
rule, it makes little sense to manage adaptively to reduce uncertain, but objectives are clear
uncertainty, if uncertainty is not at issue in the manage- and the potential for management
ment problem. When uncertainty does limit effective intervention is high.
management, there often is substantial value in managing
adaptively (see Section 2.1 for additional discussion on
this point).
level of uncertainty / emphasis on learning

high

active adaptive management

passive adaptive management

management based on resource status

low

Figure 5.1. Priorities in different management approaches.

Not shown is management in the absence of monitoring, which
devalues learning and focuses only loosely on management
objectives.

60
5.2. Accounting for Uncertainty to the possible misrepresentation of management inter-
in Adaptive Management ventions and thus to an inadequate accounting of their
influence on resource behavior.
An important concern is how to represent and account
for uncertainty in applications of adaptive management Structural or process uncertainty concerns a lack of
(61,62). At a minimum, four sources of uncertainty influ- understanding (or lack of agreement) about the structure
ence the management of natural resource systems. of biological and ecological relations that drive resource
dynamics.
Environmental variation is the most prevalent source
Environmental variation, partial observability, partial
of uncertainty, and is largely uncontrollable and possibly
controllability, and structural uncertainty all limit a
unrecognized. It often has a strong influence on natural
decision maker’s ability to make informed management
resource systems, through such factors as random vari-
decisions (Fig. 5.2). Special emphasis is given in adap-
ability in climate.
tive management applications to structural or process
uncertainty. However, the other forms uncertainty
Partial observability refers to uncertainty about
also can be incorporated in an adaptive management
resource status. An obvious expression of partial observ-
project, depending on their importance. For example, a
ability is the sampling variation that arises in resource
typical approach to environmental variation is to include
monitoring.
environmental conditions in the resource models in an
adaptive management project (Fig. 5.2), with probabilities
Partial controllability expresses the difference assigned to different values of the relevant environmental
between the actions targeted by decision makers and the variables. In this way model behaviors will reflect
actions that are actually implemented. This uncertainty environmental variation, as will the projected responses
typically arises when indirect means (for example, to management actions. Environmental variation there-
regulations) are used to implement a targeted action (for fore ramifies through the decision making process, as
example, setting a harvest or stocking rate), and it leads projected responses to management guide the selection of
management actions.

management management management

action action action
partial structural
control uncertainty

... resource
system
resource
system
resource
system
...
environmental partial
variation observability

environmental environmental environmental

conditions conditions conditions

time
t-1 t t+1

Figure 5.2. Uncertainty sources in natural resource management. Partial control limits the influence of management actions.
Environmental variation affects resource system status and dynamics. Partial observability limits the recognition of system status.
Structural uncertainty limits the ability to characterize system change.

61
 5.3. The Measurement of Learning recognition of uncertainty as a key attribute of natural
resource management. Indeed, adaptive management is
Much has been said about learning in the preceding not feasible unless the relevant management institutions
chapters, but questions remain about how learning have the capability and willingness to embrace uncer-
actually is achieved and recognized. In Chapter 3, tainty (47). Among other things, embracing uncertainty
uncertainty was described in terms of different hypotheses means recognizing different views of a managed system,
about how a resource system responds to management as well as a direct involvement of stakeholders who
actions, along with models imbedding these hypotheses have different perspectives, and a commitment to shared
and their associated measures of confidence. As evidence decision making that allows uncertainty to be reduced.
accumulates through monitoring, confidence grows
in the models (and their associated hypotheses) that At issue here is an organizational structure and
accurately predict responses to management, and confi- context that can promote and facilitate an adaptive
dence declines for models that are poor predictors. It is approach to resource management. Attributes of a
through the sequential comparison of predictions against learning organization include the following:
monitoring data that the adequacy of a hypothesis about
biological and ecological processes is gradually revealed. • acknowledgement that the world is uncertain and that
A comparison of hypothesis-based predictions against failure to predict outcomes accurately is common.
evidence is an essential feature of scientific investigation,
and a key reason why adaptive management is described • recognition of the importance of training people in
as “science-based.” group interactions and collaboration.

Questions remain about possible mechanisms for • positive reinforcement and rewards for experimentation
updating the confidence in a particular hypothesis and learning.
Generically, at each point in time one can use a measure
of the difference between the response predicted by a • recognition that surprises and even crises can be
model and the response estimated with monitoring data. opportunities for learning (65).
A small difference indicates a good fit for the model, and
a large difference indicates a poor fit. These differences
can be calculated for each model after each post-decision
monitoring event, and used to update confidence levels of
the models through time. Depending on the desired rigor,
an updating protocol can be fairly simple or technically
complicated (63).

5.4. Learning Organizations

Learning in adaptive management derives from
management actions, and is used in turn to inform
subsequent actions. But many important issues about
how best to facilitate learning are framed in terms of
adaptive institutional arrangements, structures, and
processes—features that are often lacking in traditional
management (64). Despite frequent assertions about the
use of adaptive management and the depiction of learning
as a key element in applications, there has been limited
progress in making adjustments to promote learning
institutionally (48).

The notion of technical and process learning bears

directly on the concept of learning-based organization,
the institutional framework for adaptive management. At
the heart of both an adaptive management project and the
learning-based organization that supports it is the explicit
62
 In fact, many observers think that the major chal- learning can be accelerated by the use of active adaptive
lenges facing adaptive management are fundamentally management, which utilizes management interventions
institutional (8). Institutions are built on major premises proactively for the purpose of learning. Learning rates
and long-held beliefs that are deeply imbedded in are maximized when interventions are imposed in an
educational systems, policies, and norms of professional experimental context that includes randomization, replica-
behavior (66). Yet Senge (4) argues that a learning tion and experimental control. Under these circumstances,
organization is “… where people continually expand their contributions to resource objectives are temporarily
capacity to create the results they truly desire, where new postponed so that understanding can be attained as
and expansive patterns of thinking are nurtured, where quickly as possible (68). Even here, however, environ-
collective aspiration is set free, and where people are mental variation, partial controllability, partial observ-
continually learning to see the whole together.” There ability, and the magnitude of structural uncertainty can
is a natural tension between the tendency of large, long- slow the accumulation of knowledge, and thus impede the
standing organizations to maintain a strong institutional improvement of management.
framework for thinking and decision making, and the
need in adaptive management for an open, collabora- One caveat about adaptation and the rate of learning
tive approach that recognizes alternative perspectives, should be mentioned here. We have described adaptive
embraces uncertainty, and utilizes participative decision management in terms of a cycle of decision making,
making (67). monitoring, and assessment that aims at reducing
structural uncertainty (Fig. 3.3). Uncertainty is reduced
Structuring a learning-based adaptive organization gradually in adaptive management, through the sequential
can be handicapped by a pervasive belief that adaptive evaluation of hypotheses and accretion of knowledge
management does not constitute a significant departure about them. However, the accretion of knowledge is
from the past, but is only a process of adjusting over time clearly undermined if the resource system changes more
(51). One consequence is that little attention is given rapidly than the rate of learning about it. Even if system
to the institutional barriers to its implementation, and structure and processes remain relatively stable, the iden-
little effort is expended on the redesign of organizational tification of strategies to achieve management objectives
structures and processes to accommodate an adaptive is undermined if the objectives change more rapidly than
style of management. At a minimum, it is necessary adaptive management can learn how to achieve them.
to rethink the notions of risk and risk aversion, and to The point here is that for adaptive management to be
promote conditions that encourage, reward, and sustain effective, the need to reassess and possibly change set-up
learning by individuals. phase components (stakeholders, objectives, alternatives,
models, monitoring) should be less frequent than the
5.5. Realistic Expectations for iterative cycle of technical learning (decision making,
monitoring, and assessment) (Fig. 3.4). Otherwise,
Adaptive Management learning cannot keep pace with changes in the structure of
the resource system and changing stakeholder values and
Adaptive management is designed to produce
perspectives.
gradual improvements in management through a
stage-wise process that promotes incremental learning.
However, there is nothing prescriptive in the notion of
adaptive management about the length of time required
to see substantive improvements in understanding
and management. In some cases it may be possible to
recognize improvement in only one or a few cycles of
the adaptive cycle (Fig. 3.3). In others, learning occurs
much less rapidly. Several conditions can influence the
rate of learning, including the size and complexity of the
resource system, the number and extent of management
alternatives, and the sources and magnitudes of uncer-
tainty.

Of interest here is the influence of the management

approach itself on rates of learning. As mentioned above,
smallmouth bass

63
 Key Points
 Active and passive approaches to adaptive manage-
ment can be distinguished from other management
approaches based on their treatment of uncertainty
and emphasis on learning.

 Multiple sources of uncertainty can influence resource

systems and alter the capacity to manage them.

 Learning is advanced by the sequential comparison of

model predictions against monitoring data, whereby
confidence in an underlying hypothesis is based on
the relative accuracy of model predictions.

 The practice of adaptive management flourishes in a

learning organization that is open to surprise, accom-
modates risk, and encourages and rewards learning.

 Learning in adaptive management proceeds most

rapidly when pursuit of resource objectives is tempo-
rarily postponed so that management interventions are
implemented according to an experimental design.

64
Concluding Remarks

A daptive management can be applicable to local resource projects as well as large-scale conservation programs,
though the legal constraints on an adaptive approach may differ across scales. But the basic framework presented in this
technical guide, involving an iterative process of management, monitoring, and evaluation, applies in either case. The
key issues in deciding to use adaptive management are whether there is substantial uncertainty about the impacts on
management, and whether the reduction of that uncertainty can be expected to improve management.

For many important problems, adaptive management holds great promise in reducing the uncertainties that limit
the effective management of natural resource systems. In many cases, utilizing management itself in an experimental
context may be the only feasible way to gain the system understanding needed to improve management. In concept,
adaptive management is neither conceptually complex nor operationally intricate. However, it requires refinements of
the business models of DOI agencies, to more fully reflect system sustainability and resilience, more explicitly account
for uncertainty, and more fully incorporate conservation planning, decision-based monitoring, and evaluation.

A realistic assessment of its challenges suggests that adaptive management is likely to be neither short-term nor
inexpensive, and a considerable amount of up-front planning may be required. Stakeholders and implementing organiza-
tions must commit to providing the necessary resources for monitoring and assessment over the time required to make
progress in achieving project objectives. In particular, an initial investment of time and effort will increase the likeli-
hood of better decision making and resource stewardship in the future. The need for patience and flexibility in adaptive
management highlights the importance of carefully considering the potential use of an adaptive approach, and careful
planning and evaluation when adaptive management is used.

65
 References
1. Sexton, W.T., A. Malk, R.C. Szaro, and N. Johnson (editors). 1999. Ecological Stewardship: A Common Reference
for Ecosystem Management, Volume 3: Values, Social Dimensions, Economic Dimensions, Information Tools. Elsevier
Science, Oxford, UK.

2. Haber, S. 1964. Efficiency and Uplift: Scientific Management in the Progressive Era, 1890–1920. University of
Chicago Press. Chicago, IL.

3. Bormann, B.T., D.C. Lee, A.R. Kiester, D.E. Busch, J.R. Martin, and R.W. Haynes. 2006a. Adaptive Management
and Regional Monitoring. Chapter 10 in: R.W. Haynes, B.T. Bormann, and J.R. Martin (eds.). Northwest Forest
Plan—the First Ten Years (1994-2003): Synthesis of Monitoring and Research Results. PNW GTR 651, USDA Forest
Service, Pacific Northwest Research Station, Portland, OR.

4. Senge, P.M. 1990. The Fifth Discipline: The Art and Practice of the Learning Organization. Currency Doubleday,
New York.

5. Kuhn, T. 1996. The Structure of Scientific Revolutions, 3rd Edition. University of Chicago Press, Chicago, IL.

6. Ashworth, M.J. 1982. Feedback Design of Systems with Significant Uncertainty. Research Studies Press, Chichester,
UK.

7. Allenby, B.R. and D.J. Richards. 1994. The Greening of Industrial Ecosystems. National Academy Press,
Washington, DC.

8. Stankey, G.H., R.N. Clark, and B.T. Bormann. 2005. Adaptive management of natural resources: theory, concepts,
and management institutions. PNW-GTR-654, USDA Forest Service, Pacific Northwest Research Station, Portland, OR.

9. Szaro, R.C. 1996. Biodiversity in Managed Landscapes: Principles, Practice, and Policy. Pages 727-770 in: R.C.
Szaro and D.W. Johnston (eds.). Biodiversity in Managed Landscapes. Oxford University Press, New York.

10. Holling, C.S. 1978. Adaptive Environmental Assessment and Management. John Wiley and Sons, New York, NY.

11. Walters, C.J. 1986. Adaptive Management of Renewable Resources. Blackburn Press, Caldwell, NJ.

12. Lee, K.N. 1993. Compass and Gyroscope: Integrating Science and Politics for the Environment. Island Press,
Washington, DC.

13. Elliott, G., M. Chase, G. Geupel, and E. Cohen. 2004. Developing and Implementing an Adaptive Conservation
Strategy: A Guide for Improving Adaptive Management and Sharing the Learning Among Conservation Practitioners.
PRBO Conservation Science, CA.

14. MacDonald, G. B., J. Fraser, and P. Gray (editors). 1999. Adaptive Management Forum: Linking Management and
Science to Achieve Ecological Sustainability. Ontario Ministry of Natural Resources, Peterborough, Ontario, Canada.

15. Murray, C. and D.R. Marmorek. 2004. Adaptive management: A science-based approach to managing ecosystems
in the face of uncertainty. In: N.W.P. Munro, T.B. Herman, K. Beazley and P. Dearden (eds.). Making Ecosystem-based
Management Work: Proceedings of the Fifth International Conference on Science and Management of Protected Areas,
Victoria, BC, May, 2003. Science and Management of Protected Areas Association, Wolfville, Nova Scotia. Available
online at: http://www.essa.com/downloads/AM_paper_Fifth_International_SAMPAA_Conference.pdf

66
References
16. Szaro, R.C., C.E. Peterson, and K. von Gadow. 2006. Operational experiments for sustainably managing forests.
Allgemeine Forst und Jagdzeitung 177(6/7):98-103.

17. Szaro, R.C., J.A. Sayer, D. Sheil, L. Snook, A. Gillison, G. Applegate, J. Poulsen, and R. Nasi. 1999c. Biodiversity
Conservation in Production Forests. Center for International Forestry Research, Jakarta, Indonesia and International
Union for Forestry Research Organizations, Vienna, Austria for the World Bank and the Secretariat of the Global
Convention for the Conservation of Biological Diversity.

18. Harwood, J. and K. Stokes. 2003. Coping with uncertainty in ecological advice: lessons from fisheries. Trends in
Ecology and Evolution 18:617-622.

19. National Research Council. 2004. Adaptive Management for Water Resources Planning, The National Academies
Press. Washington, DC.

20. Atkinson, A.J., P.C. Trenham, R.N. Fisher, S.A. Hathaway, B.S. Johnson, S.G. Torres, and Y.C. Moore. 2004.
Designing monitoring programs in an adaptive management context for regional multiple species conservation plans.
U.S. Geological Survey Technical Report, USGS Western Ecological Research Center, Sacramento, CA.

21. Salafsky, N., R. Margoluis, and K. Redford. 2001. Adaptive Management: A Tool for Conservation Practitioners.
Biodiversity Support Program, Washington, DC. Available on www.worldwildlife.org/bsp/.
.
22. Noss, R.F., M.A. O’Connell, and D.D. Murphy. 1997. The Science of Conservation Planning: Habitat Conservation
under the Endangered Species Act. Island Press, Washington, DC.

23. Nyberg J.B. 1998. Statistics and the practice of adaptive management. Pages 1-8 in: V. Sit. and B. Taylor (eds.).
Statistical Methods for Adaptive Management Studies. Land Management Handbook No. 42, British Columbia
Ministry of Forest Research, Victoria, BC.

24. Wilhere, G.F. 2002. Adaptive management in habitat conservation plans. Conservation Biology 16:20-29.

25. USAID. 2002. Biodiversity Conservation Program Design and Management: A Guide for USAID Staff. United
States Agency for International Development, Office of Environment and Natural Resources, Bureau of Economic
Growth, Agriculture, and Trade, Washington, DC.

26. Bormann, B.T., R.W. Haynes, and J. R. Martin. 2006. Adaptive management of forest ecosystems: some rubber hits
the road? Bioscience (in press).

27. Peterson, G.D., G.S. Cumming, and S.R. Carpenter. 2003. Scenario planning: a tool for conservation in an uncertain
world. Conservation Biology 17:358-366.

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

29. Gregory R, D. Ohlson, and J. Arvai. 2006. Deconstructing adaptive management: criteria for applications to environ-
mental management. Ecological Applications 16:2411-2425.

30. Boling, E.A. 2005. Environmental management systems and NEPA: a framework for productive harmony.
Environmental Law Reporter 35:10022-10031.

67
 References

31. Williams, B.K., J.D. Nichols, and M.J. Conroy. 2002. Analysis and Management of Animal Populations. Academic
Press, San Diego, CA.

32. Doremus, H. 2001. Adaptive management, the Endangered Species Act, and the institutional challenges of “new
age” environmental protection. Washburn Law Journal 41:50-89.

33. Stankey, G.H. 2003. Adaptive management at the regional scale: breakthrough innovation or mission Impossible?
A report on an American Experience. Pages 159-177 in B.P. Wilson and A. Curtis (eds.). Agriculture for the Australian
Environment, Proceedings of the 2002 Fenner Conference on the Environment. Johnstone Centre, Charles Stuart
University, Albury, Australia.

34. Johnson, F.A., C.T. Moore, W.L. Kendall, J.A. Dubovsky, D.F. Caithamer, J.T. Kelley, Jr., and B.K. Williams. 1997.
Uncertainty and the management of mallard harvests. Journal of Wildlife Management 61:202-216.

35. Moore C.T., W.T. Plummer, and M.J. Conroy. 2005. Forest management under uncertainty for multiple bird popula-
tion objectives. USDA Forest Service Gen. Tech. Rep. PSW-GTR-191:373-380.

36. North American Waterfowl Management Plan, Plan Committee. 2004. North American Waterfowl Management Plan
2004. Strategic Guidance: Strengthening the Biological Foundation. Canadian Wildlife Service, U.S. Fish and Wildlife
Service, Secretaria de Medio Ambiente y Recursos Naturales.

37. Williams, B.K., and F.A. Johnson. 1995. Adaptive management and the regulation of waterfowl harvests. Wildlife
Society Bulletin 23:430-436.

38. Williams, B.K. 1997. Approaches to the management of waterfowl under uncertainty. Wildlife Society Bulletin
25:714-720.

39. Nichols, J.D. 2000. Evolution of harvest management for North American waterfowl: Selective pressures and pread-
aptations for adaptive harvest management. Transactions, North American Wildlife and Natural Resources Conference
65:65-77.

40. Rogers K.H. and H. Biggs. 1999. Integrating indicators, end points and value systems in the strategic management
of the Kruger National Park. Freshwater Biol. 41: 439–51.

41. Lee, K.N. 1999. Appraising adaptive management. Conservation Ecology 3(2): 3. [online] URL: http://www.
consecol.org/Journal/vol3/iss2/art3/

42. Holling, C. S. 1999. Introduction to the special feature: just complex enough for understanding; just simple enough
for communication. Conservation Ecology 3(2): 1. [online] URL: http://www.consecol.org/vol3/iss2/art1/

43. Knopp, T.B. and E.S. Caldbeck. 1990. The role of participatory democracy in forest management. Journal of
Forestry 88(5):13-18.

44. Lauber, T.B. and B. Knuth. 1997. Fairness in moose management decision making: the citizen’s perspective.
Wildlife Society Bulletin 25(4):776-787.

45. Ringold, P.R., J. Alegria, R.L. Czaplewski, B.S. Mulder, T. Tolle, and K. Burnett. 1996. Adaptive monitoring design
for ecosystem management. Ecological Applications 6(3):745–747.
68
References

46. Lee, K.N. and J. Lawrence. 1986. Adaptive management: learning from the Columbia River basin fish and wildlife
program. Environmental Law 16: 431–460.

47. Gunderson, H.H., C.S. Holling, and S.S. Light (editors). 1995. Barriers and Bridges to the Renewal of Ecosystems
and Institutions. Columbia University Press, New York.

48. Stankey, G.H., B.T. Bormann, C. Ryan, B. Shindler, V. Sturtevant, R.N. Clark, and C. Philpot. 2003. Adaptive
management and the northwest forest plan: rhetoric and reality. Journal of Forestry 101(1):40-46.

49. Szaro, R.C., P. Angelstam, and D. Sheil. 2005. Information needs for ecosystem forestry. Pages 101-114 in: J.A.
Sayer and S. Maginnis (eds.). Ecosystem Approach to Sustainable Forest Management. IUCN/Earthscan, London.

50. Szaro, R.C., J. Berc, S. Cameron, S. Cordle, M. Crosby, L.Martin, D. Norton, R. O’Malley, and G. Ruark. 1998. The
ecosystem approach: science and information management issues. Landscape and Urban Planning 40 (1-3): 89-102.

51. Stankey, G.H. and R.N. Clark. 2006. Chapter 7: Adaptive management: facing up to the challenges. Pages 121-161
in: R.W. Haynes, B.T. Bormann, and J.R. Martin (eds.). Northwest Forest Plan—the First Ten Years (1994-2003):
Synthesis of Monitoring and Research Results. PNW GTR 651. USDA Forest Service, Pacific Northwest Research
Station, Portland, OR.

52. Adamcik, R.S., E.S. Bellantoni, D.C. DeLong, Jr., J.H. Schomaker, D.B. Hamilton, M.K. Lauban, and R.L.
Schroeder. 2004. Writing Refuge Management Goals and Objectives: A Handbook. U.S. Fish and Wildlife Service,
Washington, DC.

53. Williams, B.K. 1996b. Adaptive optimization of renewable natural resources: solution algorithms and a computer
program. Ecological Modelling 93:101-111.

54. Williams, B.K. 2006. Adaptive harvest management: Where we are, how we got here, and what we have learned
thus far. Transactions, North American Wildlife and Natural Resources Conference 71:259-274.

55. Szaro, R.C., A. Cooperrider, C.S. Correll, J. Lint, and A. Malk. 1999a. Data collection, management and inventory.
Pages 231-239 in: N., Johnson, A. Malk, R.C. Szaro, and W.T. Sexton (eds.). Ecological Stewardship: A Common
Reference for Ecosystem Management, Volume 1: Key Findings. Elsevier Science, Oxford, UK.

56. Szaro, R.C., D. Maddox, T. Tolle, and M. McBurney. 1999b. Monitoring and evaluation. Pages 223-230 in Johnson,
N., A. Malk, R.C. Szaro and W.T. Sexton (eds.). Ecological Stewardship: A Common Reference for Ecosystem
Management, Volume 1: Key Findings. Elsevier Science, Oxford, UK.

57. Nichols, J.D. and B.K. Williams. 2006. Monitoring for conservation. Trends in Ecology and Evolution
21(12):668-673.

58. Williams, B.K. 1989. Review of dynamic optimization methods in renewable natural resource management. Natural
Resources Modeling 3:137-216.

59. Williams, B.K. 1996a. Adaptive optimization and the harvest of biological populations. Mathematical Biosciences
136:1-20.

69
 References

60. Bormann, B.T., D.C. Lee, A.R. Kiester, T.A. Spies, R.W. Haynes, G.H. Reeves, and M.G. Raphael. 2006b.
Synthesis—interpreting the Northwest Forest Plan as more than the sum of its parts. Chapter 3 in: R.W. Haynes, B.T.
Bormann, and J.R. Martin (eds.). Northwest Forest Plan—the First Ten Years (1994-2003): Synthesis of Monitoring and
Research Results. PNW GTR 651, USDA Forest Service, Pacific Northwest Research Station, Portland, OR.

61. Bormann, B.T. and A.R. Kiester. 2004. Options Forestry: Acting on Uncertainty. Journal of Forestry 102: 22-27.

62. Moore, C.T. and M.J. Conroy. 2006. Optimal regeneration planning for old-growth forest: addressing scientific
uncertainty in endangered species recovery through adaptive management. Forest Science 52(2):155-172.

63. Williams, B.K. 2001. Uncertainty, learning, and optimization in wildlife management. Environmental and
Ecological Statistics 8:269-288.

64. Grumbine, R.E. 1997. Reflections on “What is ecosystem management?” Conservation Biology 11(1): 41–47.

65. Michael, D.N. 1995. Barriers and bridges to learning in a turbulent human ecology. Pages 461–488 in: L.H.
Gunderson, C.S. Holling, and S.S. Light (eds.). Barriers and Bridges to the Renewal of Ecosystems and Institutions.
Columbia University Press, New York.

66. Miller, A. 1999. Environmental Problem Solving: Psychosocial Barriers to Adaptive Change. Springer-Verlag, New
York.

67. Gunderson, L. 1999. Stepping back: assessing for understanding in complex regional systems. Pages 27–40 in:
K.N. Johnson, F. Swanson, M. Herring, and S. Greene (eds.). Bioregional Assessments: Science at the Crossroads of
Management and Policy. Island Press, Washington, DC.

68. Walters, C. and R. Green. 1997. Valuation of experimental management options for ecological systems. Journal of
Wildlife Management 61:987-1006.

70
Key Terms
Estimation

The aggregation of field data into measures of resource attributes. Examples include means, variances, and correla-
tion coefficients computed with sample data. Multiple estimators are always available for any resource attribute, and the
choice of which particular estimator to use is based on statistical features such as bias and precision.

Experimentation

The imposition of treatments on subjects or experimental units for the explicit purpose of learning about treatment
effects by observing outcomes. Ideally experimentation involves random allocation of treatments to experimental units,
replication of treatments, and the use of controls for comparative purposes.

Experimental management

The use of management interventions for the purpose of understanding the effects of management. Interventions
are used as experimental treatments, ideally (but infrequently) in the context of randomization, replication, and experi-
mental control.

Hypothesis

A suggested but unconfirmed assertion or explanation of observed patterns. Hypotheses can take many forms,
for example, a hypothesized magnitude of a resource attribute or a mathematical relationship between attributes.
Hypotheses are tested by comparison against field data.

Management by experiment

An approach to management that recognizes management interventions as experiments, by means of which under-
standing can be enhanced as management proceeds through time.

Management action

An action affecting a managed system, taken as a result of a management decision. In the context of natural
resources, management actions typically influence the status of resources or the processes that control resource
dynamics.

Management alternative

A potential management action. In sequential management, a management action is selected at each point in time
from an identified set of management alternatives. The set of management alternatives constrains and influences the
choice of a management strategy.

Management decision

A decision to take a management action. In adaptive management, decision making typically is driven by manage-
ment objectives, with active stakeholder involvement. Adaptive decision making takes into account both the current
status of resources and the level of understanding about them.

71
 Key Terms

Management option

Used interchangeably with management alternative.

Management strategy

A prescription of management actions pursuant to management objectives. In the context of adaptive management,
a management strategy describes time-specific management actions to be taken, conditional on current resource status
and the level of understanding about resource dynamics. Management strategies often are expressed in terms of resource
thresholds, on either side of which a different action is to be taken.

Model

Any representation, whether verbal, diagrammatic, or mathematical, of an object or phenomenon. Natural resource
models typically characterize resource systems in terms of their status and change through time. Models imbed hypoth-
eses about resource structures and functions, and they generate predictions about the effects of management actions.

Objective

A desired outcome or performance measure that expresses stakeholder values and serves to guide natural resource
decision making and evaluation of success.

Stakeholders

Individuals and organizations (e.g., managers, scientists, private citizens, nongovernmental organizations) with a
vested interest in a shared enterprise. Interests can include an expectation of received benefit, a perceived threat, a prior
investment of time and/or resources, or values shared with others associated with the enterprise. Active engagement of
stakeholders promotes the successful implementation of adaptive management.

Threshold

The limiting value of a resource attribute that triggers a change in management actions. Management strategies
often include thresholds, such that one action is specified for resource values less than the threshold and a different
action is specified for larger resource values.

72