Structured Decision-Making Incorporates Stakeholder Values into Management Decisions Thereby Fulfilling Moral and Legal Obligations to Conserve Species

Earth is likely in the midst of a sixth mass extinction event and the current rate of extinction is higher than that seen in the fossil record (Jablonski 1991; Pimm et al. 1995; Barnosky et al. 2011; Dirzo et al. 2014). As of December 2018, >1,400 species of animals in the United States are listed as threatened or endangered under the US Endangered Species Act (ESA 1973, as amended; https://ecos.fws.gov/ecp/) and habitat destruction is the leading cause of population decline (Pimm and Raven 2000). Once a species is listed the U.S. Fish and Wildlife Service (USFWS) or National Marine Fisheries Service is legally obliged by Section 4(f)(1) of the ESA to develop a recovery plan that incorporates site-specific management actions and objective, measurable criteria that when met will result in recovery of the species. A recovery plan serves as a guide for management actions to be taken by federal, state, or other entities to help reverse the threats to species and ensure long-term survival (i.e., recovery) and conservation. Since the signing of the ESA in 1973, <2% of listed species have been recovered and 11 species have been classified as extinct (https://ecos.fws.gov/ecp/). Actions to date have produced limited success in recovering species, but there is little doubt the ESA has helped some rare species continue to persist (Male and Bean 2005).

Biological and ecological data have historically been the foundation upon which recovery actions are based. However, as the understanding of biology and ecology has grown considerably, recovery success has not followed suit as suggested by the relatively low number of species recovered. Within the scientific community, there is an emerging thought that integrating the best available information from the natural sciences with social sciences will move conservation and recovery toward greater success (Bennett et al. 2017b). Social sciences can be classified as classic (e.g., psychology, anthropology, political science, history) or applied (e.g., law, education, communication). When social sciences are integrated with philosophy, arts, and humanities they provide an understanding of conservation policy, culture, ideologies, and individual and organizational values (Bennett et al. 2017a).

Perspectives toward fish and wildlife differ as a function of goals and objectives, value systems, and the knowledge base of stakeholders and are driven by resource constraints, agency roles, and special interest groups (Clark 2005). In general, these perspectives fall under two categories—domination versus mutualism (Teel and Manfredo 2009) or consumptive versus appreciative (Fulton et al. 1996). Domination or consumptive uses of wildlife are value systems that suggest wildlife may be conserved for their benefit to humans and prioritize human well-being over wildlife. A mutualism or appreciative perspective is one that views wildlife as more equal to humans and deserving of greater rights (Teel and Manfredo 2009). It is often difficult, if not impossible, to reach consensus on management decisions with these types of conflicting perspectives and differing goals.

To best work within a diversity of people's perspectives, one must understand their values. At the individual level, values are instilled at a young age and are partially dictated by ancestry and experiences (Manfredo et al. 2017). Changing deep-rooted value systems is unlikely and life-altering events are often required for individuals to reflect on and change their value systems. Values are adaptive and it is more likely that values will change on a generational scale (Fulton et al. 1996) as humans continually adapt to their surroundings (Manfredo et al. 2017). Conservation actions work within the borders of existing values and adapt as baseline values shifts (Manfredo et al. 2017).

Here, we promote the use of structured decision-making as a formal process of integrating scientific information with stakeholder values during conservation planning and decision-making processes for ESA-listed species. Our goal is to link natural and social science with humanities to create transparent and enduring conservation and recovery programs that consider limited resources, stakeholder objectives, and a myriad of obligations (e.g., legal, societal, moral) and by doing so, increase the likelihood of successful conservation programs. We provide an example of this approach with Bull Trout Salvelinus confluentus, a cold water salmonid that was listed as threatened in 1999 under the ESA (USFWS 1999). However, this approach can also be considered for other species listed under the ESA. Our objectives are to 1) clearly identify the obligation to recover species and restore habitats; 2) identify potential management actions that will support conservation by maintaining the evolutionary potential or adaptive capacity of a species; 3) promote structured decision-making with adaptive management as the appropriate process to guide species conservation decisions, primarily because of its ability to incorporate diverse stakeholder values; and 4) provide an example of how structured decision-making can be beneficial to conserve species and honor various obligations from the perspective of three philosophical frameworks: consequentialism, deontology, and virtue theory. Consequentialism judges the rightness of an action or policy based on the outcome of the act or failing to act. Deontological theory suggests there is an obligation or duty to act regardless of the outcome. Virtue theory dictates that decisions be made on the basis of good moral character (Proctor 1998; Moore and Nelson 2010). We propose that structured decision-making can integrate diverse stakeholder value systems with biological and scientific knowledge, and in doing so, can transparently incorporate normative beliefs to promote species conservation.

The first challenge to recovering a species is defining what exactly a species is. The ESA defines a species as “any subspecies of fish or wildlife or plants, and any distinct population segment of any species of vertebrate fish or wildlife which interbreeds when mature.” For a Pacific salmonid species to be considered a distinct population segment under the ESA and be afforded conservation and recovery efforts it will represent an “evolutionarily significant unit (ESU) of the biological species (Waples 1991, Waples 1995)”. An evolutionary significant unit is 1) substantially reproductively isolated from other conspecific population units, and 2) represents an important component in the evolutionary legacy of the species. This interpretation of a distinct population segment as applied to Pacific salmonids was adopted by the National Marine Fisheries Service and the USFWS in 1996 (USFWS 1996) and suggests the value of a species lies in its evolutionary legacy.

Restoring the ecological processes and function of habitat that supports natural reproducing populations can take years or decades to accomplish. Small or rapidly declining populations may require intervention to conserve their evolutionary legacy if the threat of extinction is too great for natural recovery to occur. Assisted or controlled propagation is recognized as a strategy to support species recovery (USFWS and NMFS 2000; Seddon et al. 2007). Recovery can be supported through research, reintroduction, or by providing a sanctuary for species and a reserve for genetic diversity until habitat is restored. However, these types of strategies are uncertain and focused research can be conducted on captive populations of imperiled species to understand their biological requirements and behavior. This information can support recovery of wild populations and inform the efficacy of rearing animals in captivity before population declines dictate that more extreme conservation measures (e.g., assisted propagation, assisted colonization) are the only management option (Figure 1).

The USFWS ranks listed species by degree of threat, recovery potential, genetic distinctiveness, and conflict with construction, development projects, or other forms of economic activity (USFWS and NMFS 2000). However, the USFWS has been criticized for how species are prioritized for recovery and the subsequent allocation of funding to those efforts (Simon et al. 1995; Restani and Marzluff 2002; USGAO 2005; Schwartz 2008; Gerber 2016). The situation is made more complex by a variety of stakeholders, who are affected by management actions and decisions, and have differing and sometimes conflicting objectives and value systems. For example, the Columbia River Basin was dammed in the mid–20th century to provide hydropower, flood control, agricultural, and economic benefits to an ever expanding human population. Dam operation, combined with hatchery practices, overharvest and habitat degradation resulted in threatened or endangered listings of 13 distinct populations of salmon Oncorhynchus spp. and Steelhead O. mykiss, Bull Trout, and Kootenai River White Sturgeon Acipenser transmontanus in the Columbia River Basin (NOAA 2008). Eight of the 10 highest ranked priority species for recovery in the United States are located in the Columbia River Basin (USFWS 2014). The stakeholders in the basin span two countries and four States, and include federal, state, and tribal agencies; power companies; barge operators; farmers; irrigators; commercial and sport fisherman; recreational boaters; landowners; and the public at large. Balancing the rights and objectives of all stakeholders with species recovery and habitat restoration has proved to be difficult.

Scientists and agencies have alternative options for ranking imperiled species and prioritizing conservation actions given limited resources (Arponen 2012; Macdonald and Willis 2013). Weitzman (1998) proposes a method to determine conservation priorities that balances financial costs with four components that represent resource benefits: distinctiveness of a species, utility or value of a species, increase in species persistence, and cost to increase species persistence. Joseph et al. (2009) took this method beyond a cost and benefit analysis to determine the probability of success (e.g., recovery) of a specific management action. However, endangered species management often requires a succession of management actions or decisions to be made one after another. Ideally, after a decision is made and implemented, the outcome or value to conservation is measured through a monitoring program and used to inform subsequent decisions in an iterative process that continuously improves management outcomes. This process is called adaptive management, which is a special case of structured decision-making (Williams et al. 2009; Gregory et al. 2012; Conroy and Peterson 2013). A similar process was adopted by the USFWS in 2008 (i.e., Strategic Habitat Conservation; USFWS 2008) and is recommended for implementation in Recovery Plans (e.g., USFWS 2013, 2015).

One benefit of the structured decision-making approach is that the framework transparently incorporates values and stakeholder objectives in the decision process. As scientists, we cannot prescribe values to determine which diminishing habitats receive support to provide a high likelihood of recovery. We can collect the necessary data to develop biologically sound hypotheses and scientifically defensible findings to identify the causes of decline and life-history requirements of the species, and suggest alternative options to restore habitat and recover species. What would be helpful is a formal structure that can incorporate scientific information with societal values and obligations for consideration in guiding decisions and providing accountability. An understanding and incorporation of societal values in the conservation process will improve the communication, cooperation, and consensus of all stakeholders, resulting in transparent programs that are enduring (Pister 1997). Conservation is driven by the realized value of species and habitat and is not a discrete action, it is a continuous dynamic action; and therefore, to be successful, conservation will have to endure on a long time scale.

Authors, scientists, and philosophers such as Emerson, Thoreau, Muir, Pinchot, and Leopold made the connection between conservation and ethics long ago (Callicott 1991). Sandler (2010) suggests the value of a species lies in its instrumental, ecological, and intrinsic value. Instrumental value is value placed on a species as a function of its ability to benefit humans, ecological value is value based on a species' ability to benefit the ecosystem in which it lives, and intrinsic value is the value of a species in its own right. Values can be social, political, personal, or biological and founded based on cultural significance, economics, evolutionary potential, or ecosystem services (Joseph et al. 2009). All species have value. This is a primary tenet of conservation. Regardless of the value, or combination of values, one places on a species, they are making a value judgment and some level of conservation follows.

Societal obligations become legal obligations when society agrees it is a requirement to enforce and uphold the obligation (Johnson 1975). The ESA is the legal foundation on which society's moral obligation to conserve species is built. The word conservation is strewn throughout, and the ideas of concern and value are directly addressed in the findings of Congress, which state that “(1) various species of fish, wildlife, and plants in the United States have been rendered extinct as a consequence of economic growth and development untempered by adequate concern and conservation; (2) other species of fish, wildlife, and plants have been so depleted in numbers that they are in danger of or threatened with extinction; (3) these species of fish, wildlife, and plants are of esthetic, ecological, educational, historical, recreational, and scientific value to the Nation and its people.” This concept of adequate concern implies there is some level (i.e., adequate) of worry, anxiety, fear, or feeling (i.e., concern) associated with the loss of value that would result from the extinction or threat of extinction of a species. According to the ESA, species “are of esthetic, ecological, educational, historical, recreational, and scientific value to the Nation and its people” (ESA 1973). The ESA is more than simply an act or policy, it is a commentary on societal obligation to conserve species.

The ESA also addresses a societal obligation to conserve “the ecosystems upon which endangered species and threatened species depend” (ESA 1973). Prior to the ESA, Congress acknowledged the societal obligation to conserve habitat with the passing of the Clean Air Act (1970) and the Clean Water Act (1972). The primary purpose of these Acts is to regulate harmful emissions to benefit public health and public welfare. These Acts placed value on clean, quality habitat by recognizing that emissions into the atmosphere and waterways also affected fish, wildlife, and ecosystem health. Promoting public health and welfare, and conserving species and their habitats are not mutually exclusive; on the contrary, they are inseparable and mutually beneficial.

Obligations to conserve species is also driven by legal requirements. The obligation to conserve species and their habitats in the Columbia River Basin is further substantiated by the Boldt, Belloni, and Martinez Decisions (see Blumm and Steadman 2009 for review of these cases). Similarly, Native American tribes have various legally granted rights. The Tribes and the U.S. Government signed a series of treaties in the Pacific Northwest in the 1850s that resulted in the Tribes ceding their lands to the Government in exchange for the right to hunt and fish in their usual and accustomed places. Lawsuits were brought against the States of Oregon and Washington in the 1960s and 1970s by Northwest Tribal members, who claimed the states were interfering with their treaty protected rights to harvest fish. Judges Boldt and Belloni found Washington State and Oregon State, respectively, had a legal obligation to honor the treaty rights and sovereignty of Tribes in that they were entitled to an equal share of the harvest with nontreaty fishers. It was ruled that harvest regulations should be the least restrictive regulations consistent with ensuring conservation of the species. A series of court decisions took the Boldt and Belloni Decisions a step further. Most recently the Martinez Decision found the Tribes' right to take fish implied the States have a legal obligation to protect habitats that support fish populations. This is one example of the complexity between interacting legal obligations that society tries to balance; honoring a duty to conserve species and their habitats and, honoring the duty to the rights of all stakeholders (e.g., the Tribes sovereignty).

Congressional acts have also imposed values on fish and habitat conservation. The Pacific Northwest Electric Power Planning and Conservation Act of 1980 placed equal value on fish and habitat conservation along with other priorities for which the Columbia River hydropower system is operated (Blumm and Johnson 1980). In addition to assuring “the Pacific Northwest of an adequate, efficient, economical, and reliable power supply,” the act provides for “the participation and consultation of the Pacific Northwest States, local governments, consumers, customers, users of the Columbia River System (including Federal and State fish and wildlife agencies and appropriate Indian tribes), and the public at large within the region to protect, mitigate and enhance the fish and wildlife, including related spawning grounds and habitat, of the Columbia River and its tributaries, particularly anadromous fish which are of significant importance to the social and economic well-being of the Pacific Northwest and the Nation and which are dependent on suitable environmental conditions substantially obtainable from the management and operation of Federal Columbia River Power System and other power generating facilities on the Columbia River and its tributaries.” This statement reflects the moral obligation to recover and conserve fish, wildlife, and their habitats while balancing potentially conflicting objectives (e.g., power generation) and can be done while considering input and support from multiple stakeholders (e.g., public, states, tribes).

Society's obligation to recover Bull Trout was acknowledged in 1999 when the species was listed as threatened in the coterminous United States. Bull Trout require cold water for spawning and early rearing (Buchanan and Gregory 1997), and if climate change predictions (IPCC 2007; Mote and Salathé 2010) are accurate and human-caused habitat degradation and fragmentation continue to limit connectivity, then populations are expected to decline even further (Rieman et al. 1997). Bull Trout have instrumental value as indicators of healthy aquatic habitats. They require clean, cold water, in highly connected and complex aquatic systems; therefore, healthy Bull Trout populations are an indication that the regulatory and legal obligations defined by the Clean Water Act are being fulfilled. These same habitat requirements are indicative of the ecological value of the species; juvenile salmon and steelhead populations, some of which are listed as threatened or endangered, require similar habitat conditions. Bull Trout possess intrinsic value defined as value for its own sake, commonly referred to as existence value; therefore, their extinction would be a loss to the world.

The Pacific Northwest warmed 0.8°C in the 20th century, and this trend is expected to continue and magnify (IPCC 2007; Mote and Salathé 2010). Rieman et al. (2007) evaluated Bull Trout response to a range of predicted climate warming scenarios for the Columbia River Basin and estimated a loss ranging from 18 to 92% of suitable natal habitat. The range of this estimate suggests much uncertainty about the effects of climate change on Bull Trout populations. It is thought warming trends will restrict populations to small isolated patches in headwater streams, further limiting population connectivity (Rieman et al. 2007) and increasing extinction risk (Rieman and McIntyre 1995). Large populations residing in less degraded habitat will increase in importance as source populations for recolonization and reintroduction efforts (Dunham and Rieman 1999).

Resource managers have options to counter the harmful effects of climate change on Bull Trout populations. Isaak et al. (2010) suggested by minimizing disturbances in riparian habitat (i.e., grazing, road building, and timber harvest) stream temperatures can be buffered from additional warming. Unoccupied habitat can be made available for recolonization or reintroduction (Dunham and Rieman 1999) by removing or reengineering manmade structures that limit connectivity. Given that Bull Trout exhibit multiple life-history strategies, resident populations can supply individuals to migratory populations and vice versa (Rieman and McIntyre 1993). Connectivity between suitable habitats promotes gene flow and persistence of the population by supporting complex life-history strategies. Combining multiple isolated small populations that result from a warming climate may preserve and promote metapopulation structure (Dunham and Rieman 1999) through gene flow between small populations, which is thought to be important for the conservation of a population's evolutionary potential. Reclaiming, or reconnecting historical and existing Bull Trout habitat is complex and difficult, and can require years or decades to evaluate which restoration actions were successful. If some small populations continue a rapid decline, in the interim managers could consider instituting an assisted propagation program to preserve the evolutionary potential of these declining populations.

The Final Bull Trout Recovery Plan (USFWS 2015) suggests that assisted propagation can play an important role in recovering certain core-area populations. Assisted propagation is a human-mediated increase in animal abundance or distribution for the purpose of species conservation and recovery. For example, reintroduction programs could be implemented in situations where rehabilitating degraded habitat or reclaiming and reconnecting existing suitable habitat is not feasible or does not result in natural recolonization. Introducing a species outside its historical range is referred to as conservation introduction; and while this strategy can have unforeseen ecological, social, or economic impacts, it could be a feasible strategy to increase abundance or distribution for certain species and in certain situations (IUCN/SSC 2013). These recovery strategies have the potential to preserve the evolutionary potential of the species by conserving “important genetic resources in nature, thus allowing the dynamic process of evolution to continue largely unaffected by human factors” (Waples 1995). Animals can be held in captivity in the most extreme scenarios for multiple generations until human population growth and climate change are realized and restoration actions coupled with societal life-style changes and technological advances are able to reestablish functioning natural habitats. Regardless of the strategy employed, recovery will be accomplished when the long-term persistence of self-sustaining populations, distributed across the species' native range and representing the genetic and life-history diversity of the species, has been achieved (USFWS 2015).

Captive propagation to support reintroduction actions in conjunction with habitat restoration may be necessary in some areas (USFWS 2015). Managers have three reintroduction strategies that are considered when developing a conservation plan for Bull Trout (Shivley et al. 2007); translocation (or transplantation), captive rearing, and artificial production (or captive breeding). The IUCN/SSC (2013) defines a translocation as a “human-mediated movement of living organisms from one area, with release in another.” They use this definition within the broad concept of conservation and it can include moving organisms from wild or captive origins. Translocation as a reintroduction strategy has a slightly different definition for our purposes. A translocation strategy involves collecting wild animals and, after a short holding period or immediately following collection, transporting those animals into the recipient habitat. A captive-rearing strategy involves collecting wild animals and rearing them in captivity for an extended period of time until they are released into the recipient habitat to reproduce naturally. This strategy typically is used to increase the survival of life stages that exhibit high rates of mortality in the wild (e.g., embryos, fry, or juvenile fish). An artificial production strategy is the most extreme of the three reintroduction strategies. It differs from captive rearing in that wild fish are brought into captivity and used to develop a captive broodstock. This broodstock is artificially spawned and the resulting progeny reared in captivity to maturity and spawned to maintain the population in captivity, or they are released into the wild at any given life stage as part of a reintroduction program (Hard et al. 1992). Each strategy comes with its own list of advantages and disadvantages, and selecting the best approach to benefit Bull Trout conservation will depend on the level of uncertainty, the amount of risk and financial cost managers are willing to incur, and the probability of achieving the desired outcome (e.g., Bull Trout recovery or local population restoration).

Translocation, captive rearing, and artificial production of Bull Trout have all been attempted in the past. Creston National Fish Hatchery (Kalispell, Montana) has been successful at artificially producing Bull Trout from wild parents; however, individuals were never released into the wild. The resulting progeny were raised to adulthood and artificially spawned. The rate of fertilization (70.5%) from the captive broodstock was lower than that of embryos from wild parents (97.4% and 96.6% [Fredenberg et al. 1995; Fredenberg 1998]). Artificial production was used as a reintroduction strategy for the McCloud River, California. The progeny of wild parents exhibited poor survival in captivity, resulting in small numbers of fish (n = 270 fingerlings) available for reintroduction. This, coupled with a decline in the donor stock from the Klamath Basin, led to the termination of the McCloud River reintroduction in 1995 after 5 years of monitoring (Rode 1990; Buchanan et al. 1997). Translocations are ongoing in the Clackamas River, Oregon, and monitoring efforts have documented spawning behavior by the out-planted adults (Barry et al. 2014). It is too early to determine the level of success of the Clackamas River reintroduction program, but observations of spawning behavior is a positive finding. Translocation of Bull Trout fry from the McKenzie River, Oregon, to the Upper Willamette River, Oregon, took place from 1997 to 2013, with a small number (∼20) of adults spawning (Ziller and Taylor 2000; UWBTWG 2010; Soorae 2011). In an effort to increase survival to adulthood of translocated fry, a captive-rearing component to the Willamette River reintroduction began in 2007 (N. Zymonas, Oregon Department of Fish and Wildlife, personal communication); monitoring is ongoing.

The benefits of a translocation reintroduction strategy include minimizing the effects of captivity on fish behavior and increasing genetic and ecological diversity. This is possible because of the minimal time the fish spend in a hatchery environment (Flagg and Nash 1999; Brown and Day 2002; Fraser 2008; Naish et al. 2008). Also, translocation is the least expensive of the three strategies. For these reasons it is the first option among many managers and has produced some positive outcomes. Disadvantages to translocation include higher risk to donor populations relative to other strategies because of the number of individuals estimated to be required to create a self-sustaining population (Shivley et al. 2007) and local adaptation of donor populations that could negatively influence their performance in the recipient system. Direct transfers of wild animals will not always allow for adequate health screening, which can increase the potential for disease transmission to resident fish populations in the receiving habitat.

The primary benefit of captive rearing is removing the natural causes of mortality on younger, smaller individuals and rearing them to a larger size in a protected environment. This is intended to increase the probability of survival to reproduction after being released into the wild. Captive rearing allows for disease testing of the population being released, ensuring pathogens are not transferred between basins. Relative to translocation, captive rearing requires fewer individuals to be removed from the donor population (Shively et al. 2007). There are also negative aspects to captive rearing. Individuals can develop behavioral, physiological, and other phenotypic deficiencies, and consequently genetic fitness can be compromised (Flagg and Nash 1999; Brown and Day 2002; Fraser 2008; Naish et al. 2008). Bull Trout are piscivorous by nature and cannibalism has been documented in wild (Beauchamp and Van Tassell 2001; Lowery and Beauchamp 2015) and cultured (W.R. Brignon, personal observation) populations. Cannibalism occurring in captive populations will result in fewer individuals available for reintroduction. Captive-rearing programs can take a conservation hatchery approach (Flagg and Nash 1999) to best prepare captive individuals for release back into the wild. In general, the costs and benefits (i.e., consequences) of a captive-rearing strategy fall between those of translocation and artificial production.

The overarching benefit of an artificial production program is in producing a large number of individuals for reintroduction while at the same time minimizing the negative impacts on donor populations (Shivley et al. 2007). Additional benefits include the ability to screen fish for pathogens and rear fish to target sizes for release. Artificial production also has its downsides. There is a suite of recent literature summarizing the effects to salmonids of being artificially produced and reared in captivity (Flagg and Nash 1999; Brown and Day 2002; Fraser 2008; Naish et al. 2008). For example, stress (Dickens et al. 2010), behavioral differences (Berejikian et al. 2001), morphological differences (Taylor 1986) and loss of genetic diversity (Fraser 2008) are inevitable in artificial production programs. This reintroduction strategy requires more financial commitment than other strategies because of the cost of animal husbandry and rearing facilities. Typically, managers opt to use artificial production as a reintroduction strategy only when the natural population abundance is very low, all other options have been exhausted, and the risk of extinction or extirpation of local populations is outweighs by the risks associated with artificial production (Figure 1; Hard et al. 1992). Considered the most extreme of the strategies, artificial production can be used to hold animals in captivity for extended periods of time, or even generations, to miminze risk of extirpation or extinction and increase the likelihood of maintaining the genetic component of a population and, therefore, the evolutionary potential of the population.

Captive rearing and artificial production have the potential to recover Bull Trout in a shorter time frame than translocation. This is because individuals taken into a captive environment can survive at a higher rate than individuals in the wild, thereby resulting in more individuals for reintroduction and subsequent reproduction relative to a translocation strategy. However, the benefit to the species can be weighted with potential negative effects cause by domestication (Fraser 2008). Fish may survive at a higher rate in captivity but they show poorer survival upon release into the wild; this results in fewer spawning adults, which is a function of overly simplified rearing environments, handling stress, and release strategy (Minckley 1995; Flagg and Nash 1999; Dickens et al. 2010). Currently, the effects of captivity on Bull Trout are relatively unknown and not well-documented.

It would be helpful for managers to understand the benefits, limitations, and overall utility of all reintroduction and recovery strategies before population decline limits the available options (Figure 1). Any assisted propagation program to support species recovery can incorporate conservation hatchery techniques (Flagg and Nash 1999; Brown and Day 2002). Hatcheries are undergoing a paradigm shift (Pearsons 2010; Paquet et al. 2011). Historically, salmon hatcheries were developed to produce as many fish as possible for release and eventual harvest with less concern for individuals after release or the environment in which they were released, as long as adults returned to be harvested. More recently, hatcheries are being viewed as a tool that can be used to honor duties to mitigation, recreational fisheries, and conservation and recovery programs. A stronger commitment to conservation principles (Meffe and Carroll 1997) is being integrated into hatchery programs where captive-rearing environments are being designed to more closely mimic the natural environment and, therefore, better prepare these individuals for life postrelease (Maynard et al. 1995; Brignon et al. 2017). Habitat protection, restoration, and reclamation are becoming ongoing management strategies regardless of whether, or if, a reintroduction strategy is implemented.

Structured decision-making (McGuire 1986; Williams et al. 2009; Conroy and Peterson 2013) is an analytical framework that can integrate the best available science with stakeholder values to evaluate tradeoffs between alternative management actions. This approach incorporates institutional knowledge, scientific research, and stakeholder values to provide transparency, improved communication, and a direct connection between management decisions and objectives, and is intended to result in more efficient use of limited resources. Once the decision problem is defined, decisions are deconstructed into the three basic components: 1) explicit, quantifiable objectives that represent stakeholder values; 2) mathematical and conceptual models used to predict the effect of management decisions on the outcome of objectives; and 3) management alternatives or actions. Objectives can be either means or fundamental. Fundamental objectives represent what stakeholders care about most (i.e., core values). Means objectives are those that support the fulfillment of fundamental objectives (Conroy and Peterson 2013). These three basic components of a structured decision-making are then linked in a conceptual model that is developed collaboratively by all stakeholders (Conroy and Peterson 2013). These models are used to communicate the global belief in how complex ecological systems function and interact, explicitly and transparently identify what is important and valuable to the stakeholders (e.g., minimizing costs, improving habitat quality, maintaining water supply, increasing abundance), and the critical factors necessary for survival and recovery of a species (e.g., environmental parameters, water temperature, rearing habitat). Decision models can be parameterized with existing empirical and theoretical data to inform management decisions through a better understanding of predicted outcomes. Uncertainty can be incorporated into the model using probability distributions to evaluate the tradeoffs and consequences of different decisions, and the components with the greatest uncertainty and influence on the outcome are areas where future experimentation, research, monitoring, and funds can be focused, thereby maximizing the benefits from limited resources. However, because of the complexity of ecological systems these models are oversimplified representations of nature.

Structured decision-making is the foundation of developing a formal adaptive-management plan. First, models are explicit representations of system dynamics and transparently convey the understanding of the decision context to all stakeholders; scientists, land users, conservationists, and decisions makers alike. Second, these models can be continuously updated and refined as monitoring programs generate new information and a better understanding of the cause and effect of management decisions, resulting in a better understanding of system dynamics. Most natural-resource management decisions are not discrete; they are recurrent decisions made throughout space and time. As new information is generated and incorporated into the analysis, the optimal decisions, represented by alternative hypotheses, are adjusted to reach the most desirable outcome consistent with the goals of conservation and given the values of stakeholders. This later iterative process is called adaptive management. Decision processes that do not incorporate these three components—recurring decisions, alternative hypotheses, and monitoring—are not considered formal adaptive-management processes (Conroy and Peterson 2013). Recognizing the short- and long-term benefits of this approach, the USFWS has recently adopted the formal process of adaptive management to manage species and their habitats into the future (i.e., Strategic Habitat Conservation; USFWS 2008). In addition, the use of structured decision-making is explicitly recognized in the Final Bull Trout Recovery Plan as a tool to adaptively manage recovery actions, as managers gain knowledge of their effectiveness and impacts of future climates (USFWS 2015).

There are many examples of decision models in the literature that include a definition of the decision context, influence diagrams, parameterization with expert opinion and empirical data, and sensitivity analysis (Peterson and Evans 2003; Dunham et al. 2011; Peterson et al. 2013; Brignon et al. 2018). For the purposes of this essay, we simply present an influence diagram (Figure 2) of a hypothetical Bull Trout management scenario to convey how stakeholder values can be transparently incorporated and communicated in the decision-making process. In this example, the stakeholder group consists of state, federal, and tribal resource managers, land owners, and irrigators in a defined geographic area of interest. Collectively the stakeholders define the decision context: to maximize the value of decisions regarding the use of an artificial production program for Bull Trout and to modify in-stream barriers to promote the migratory life history of Bull Trout while maintaining water withdrawals from the watershed for agriculture.

The overall value of a decision outcome can be a combination of metrics, and the influence diagram in Figure 2 displays three fundamental values identified by the stakeholders—management costs represent economic value; probability of persistence represents environmental, ecological, or biodiversity value; and maintaining irrigation withdrawals represents cultural, socio-economic, or human-use based values. The probability of persistence component represents the evolutionary potential of the population. This is the characteristic of a species the ESA requires society to conserve to recover a species and promote long-term persistence of a population in a dynamic world. This represents the “esthetic, ecological, educational, historical, recreational, and scientific value” of a species identified in the ESA. Additional values are represented by the other components of the model that represent the belief in system dynamics and are used to predict the effect of management decisions (i.e., alternate hypothesis) on the value of the decision outcome. For example, stakeholders value suitable stream temperatures, spawning and juvenile rearing habitats, and natural recolonization. These components carry a suite of societal values (e.g., ecological, historical, scientific), but their utility (i.e., use value) to the decision process is in their ability to effect change in the fundamental values.

Ethical frameworks provide a means to weigh the rightness of an action. We chose to focus on consequentialism, deontology, and virtue theory to make the argument that structured decision-making is a just method to guide natural resource decisions in the future and will honor society's obligation to recovery species and their habitats. Consequentialism judges the rightness of an action or policy based on the outcome of the act or failing to act. Deontological theory suggests there is an obligation or duty to act regardless of the outcome. Virtue theory dictates that decisions are made on the basis of good moral character (Proctor 1998; Moore and Nelson 2010).

Historically, natural resource managers would default to consequentialism when making decisions regarding species and habitat recovery. This is a function of recovery plans requiring the setting of goals and objectives, or preferred outcomes (i.e., consequences), and setting actions in motion that are anticipated to accomplish those goals. In essence this is the scientific method, in which a question is asked, hypotheses are derived, an experiment is designed and conducted to test the hypotheses, and results or consequences are then reported and communicated. Consequentialism is the foundation of the scientific method; and as long as managers are expected to make decisions based on the best available science, consequentialism will continue to be a primary ethical framework in determining natural resource decisions into the future.

The most difficult part of applying the consequentialist theory is determining how to weigh the best outcome and for whom. How does one weigh the value of Bull Trout conservation for the various stakeholders, habitats, and interrelated species? Are short-term benefits of an action more important than long-term benefits, or vice versa? Is Bull Trout conservation valuable in itself or is the value increased with the function these fish provide humans and the ecosystem? Or, do the benefits gained in modifying Bull Trout habitat for human use (e.g., agriculture, hydropower, road building) out-weigh those values of maintaining healthy habitat for Bull Trout alone? A consequentialist will want to understand the predicted outcomes of competing conservation actions before making an informed decision to balance the benefits to stakeholders, species, and habitat conservation. Species recovery is a long-term dynamic process with recurring decisions; and, as monitoring generates new information, subsequent decisions can be adjusted based on the new information to increase the likelihood of preferred outcomes (i.e., adaptive management; Williams et al. 2009; Conroy and Peterson 2013). Thus, species recovery planning considers the end-result not just on species, but human values as well; therefore, using consequentialism is often the driving justification.

Deontological theory suggests one acts justly when a decision fulfills one's duty in relation to legal or social responsibilities, justice or equality for all groups, the rights of stakeholders, and one's role as a decision-maker or natural resource manager, regardless of the consequences. A just environmental ethic requires society to honor duties to sentient life, organic life, endangered species, and ecosystems. The ESA and binding court decisions (Blumm and Steadman 2009) define the legal duty to endangered species and ecosystems. The Final Bull Trout Recovery Plan (USFWS 2015) suggests there is a duty to evaluate potential recovery actions, their effectiveness, and impacts of future climates to promote species recovery. It is inequitable to reap the benefits of actions that limit Bull Trout persistence while other stakeholders such as tribes, fisherman, ecosystems, and Bull Trout populations bear the burdens. All stakeholders, human and otherwise, have the right to experience healthy aquatic ecosystems that support natural populations of multiple species. The application of structured decision-making in an adaptive management context provides the transparency to the public and stakeholders that helps fulfill a duty to justice or equality for all stakeholders.

Deontological considerations also extend to ecological management consideration. What biologists manage is not a singular species, but rather a complex mosaic of animals and the habitats in which they interact (Soulé et al. 2005). In evaluating all possible strategies for Bull Trout recovery, managers will fulfill a duty to honor the rights of multiple species and their habitats. Climate change scientists predict suitable Bull Trout habitat will be marginalized and in some areas cease to exist in the near future (Rieman et al. 2007). This does not alleviate the duty and obligation to attempt Bull Trout recovery. On the contrary, it increases the responsibility to act. Bull Trout fulfill an important ecological niche; they require cold, clean water and complex, connected habitats, and primarily feed on juvenile salmon, steelhead, and trout. Therefore, sustainable Bull Trout populations are indicators of high-quality habitat and abundant forage fish, which in some instances are threatened and endangered themselves. The obligation to attempt Bull Trout recovery is inevitably linked to a greater duty of maintaining healthy ecosystems and a sustainable planet. Structured decision-making honors society's duties to stakeholders because stakeholder values and objectives are used to build the model framework, evaluate tradeoffs, and identify the best management alternative. In our example of using structured decision-making to promote Bull Trout recovery, duties to the evolutionary potential of species, habitats, captive animals, land users, and funding sources are honored (Figure 2). Duties to stakeholders, species, habitats, and the ESA will be honored with the development of an enduring recovery program that evaluates potential recovery actions (e.g., habitat restoration, reintroduction strategies; Figure 2), allows conservation decisions to proceed despite uncertainty, and reduces uncertainty through learning with focused monitoring programs, ultimately resulting in adaptive management (e.g., Strategic Habitat Conservation).

Virtue theory differs from consequentialism and deontology in that the focus is placed on the agent rather than the action (Frasz 1993). Decisions are made by an agent (e.g., person, agency, or society) on the basis of good moral character. People honor in practice what they believe in principle. This allows one to weigh considerations of motivation and character. In fisheries management, decisions would be right if they are fair, open-minded, scientifically sound, impartial, cautious in the face of uncertainty, consistent between beliefs and practices, and transparent. How one responds to natural resources through emotion and action depicts their environmental virtue (Sandler 2013).

As a society, we began a path of habitat alteration and degradation when settling the Pacific Northwest, resulting in the current state of natural resources. Resources were thought boundless and impervious to overexploitation. The environmental virtue was defined by self-survival, financial benefit, and advancing “the American Dream.” For example, the forefathers built the first hatcheries in the Pacific Northwest in a virtuous attempt to provide fish as well as power, navigation, agricultural activities, and flood control for developing comminutes. However, as time progresses so does the evolution of society's virtue. Currently, scientific findings suggest conventional hatchery practices produced fish less fit for survival in wild habitats, and their behaviors and large numbers can negatively affect wild stocks (Brown and Day 2002; Kostow 2009). These data, combined with shifting value systems, have steered hatchery practices in a variety of directions meant to replicate the rearing conditions experienced in the wild and minimize the impacts of hatchery fish on wild species. This has been termed the era of “hatchery reform” (Paquet et al. 2011). Hatchery reform is a virtuous attempt at refining harvest and fish rearing (i.e., hatchery) practices to meet both sustainable harvest and conservation goals.

A virtuous manager will acknowledge uncertainty in the recovery decision processes. Uncertainty can be environmental, statistical, economic, or structural (i.e., cause and effect relationships in the model) and can result in wide population and habitat responses after management decisions are made and implemented. Uncertainty can be made transparent to the stakeholders and incorporated directly into structured decision-making as statistical distributions or alternative models of system processes (Conroy and Peterson 2013). Using the example presented in this essay (Figure 2), the efficacy of conservation hatchery practices to produce Bull Trout for recovery actions and the feasibility of restoring suitable habitat processes is uncertain. The wide range of climate change predictions (IPCC 2007) and their future impact to existing suitable Bull Trout habitat (Rieman et al. 2007; Wenger et al. 2013) is uncertain. A virtuous Bull Trout recovery program requires acknowledgement of these uncertainties, and decisions regarding recovery actions to be made cautiously. This approach is explicitly embraced in the Final Bull Trout Recovery Plan by recognizing the importance of transparency in decision-making, and a way to evaluate the tradeoffs of different management actions the face of uncertainty and future climates (USFWS 2015).

We acknowledge consequentialism is the foundation and focus of structured decision-making and adaptive management in natural resources. However, other ethical considerations (e.g., virtue theory, deontological ethics) are not completely removed from the decision-making process. Rather they are essential components that can be acknowledged, communicated, incorporated, and advocated by resource managers. Transparency is paramount in the decision-making process and is the conduit through which virtue and duty will be communicated to the public and stakeholders and realized by all.

As climate change is realized and societal values change (Adger et al. 2009; O'Brien and Wolf 2010), the decisions related to improving connectivity for species can become more complex. Habitat conditions (e.g., irrigation diversion, dams, thermal barriers, low flow barriers) that limit connectivity for Bull Trout populations are typically found lower in river basins near human population centers, and the moral obligation to provide water for municipal and agriculture purposes could increase in conflict with the obligation to conserve multiple species and habitats. A vital component to Bull Trout recovery will likely revolve around decisions associated with restoring connectivity and balancing societal obligations to other species as climate change progresses. Recovering species and habitats in the Columbia River Basin is a complex process that can balance the needs of multiple species and stakeholders, as well as other uses of the resource (e.g., clean water for human consumption, hydropower generation, and agriculture, commercial, sport, and tribal harvest). This can be done while simultaneously and continuously balancing progress of improving habitat conditions and responsible recovery strategies for imperiled species.

The ethics of Bull Trout management and restoration were first described by Pister (1997). In closing Pister states that “ethically sound programs inevitably are biologically sound and enduring programs.” A biologically sound management and restoration program for Bull Trout would consider the complex habitat requirements of a species that exhibits a diverse life history (Schaller et al. 2014). Habitat restoration will also allow for unrestricted expression of complete life histories that are essential for population persistence. Individuals exhibit an anadromous, adfluvial, fluvial, or resident life-history strategy (McPhail and Baxter 1996; Brenkman and Corbett 2005), requiring conservation of a wide array of essential habitats (i.e., spawning, juvenile rearing, and adult rearing) in conjunction with connecting migratory corridors between these habitats. Habitat restoration will span substantial spatial dimensions from headwater streams to large lakes, reservoirs, or rivers, and span a diverse group of stakeholders' influence, including federal, state, and tribal agencies, landowners, and the general public.

The time scale of restoration and recovery actions is important to consider. An enduring management and restoration program will balance long-term improvements in habitat condition concurrent with maintaining a population's evolutionary potential. It took the better part of the 20th century for anthropogenic habitat alteration (e.g., modern agriculture, urbanization, industrial pollution) to occur, and restoration of habitats following these types of alterations is on a similar time scale as some natural disasters (Dobson et al. 1997). In addition, maintaining evolutionary potential in small populations could require an assisted propagation program to be instituted with a focus on minimizing the effects of domestication while providing individuals for reintroduction programs. Decisions regarding assisted propagation will differ depending on scientific knowledge and uncertainty about the status of donor population, suitable recipient habitat, and the potential to connect to and support a metapopulation. In populations at high risk of extinction all remaining individuals can be provided refuge in captive environments (e.g., California condor Gymnogyps californianus, black-footed ferret Mustela nigripe, and Snake River Sockeye Salmon O. nerka).

Natural sciences offer a biological and physical basis for estimating the outcomes of decisions. An understanding of humanities is helpful in providing context for the myriad of societal obligations. This is a world of limited resources: natural, financial, spatial, and temporal. Determining the most valuable, just, and efficient manner in which to utilize these resources, and for whose benefit, is complicated and further confounded by uncertainty in the natural and social sciences and the disconnect between the fields. Which value system will be used to determine where limited resources are spent, and on which species? If we consume certain resources to recover one species, then those resources are not available for the recovery of another species. If genetic fitness of the population in a reintroduction program is not the same as the genetic fitness of the historical population in the recipient habitat, does it affect the evolutionary potential of the distinct population segment? Do the potential for species recovery and a duty to the species with which we have been entrusted alleviate these types of concerns? Science and technology only provide the technical, empirical, and theoretical foundation for actions. We believe the matrix of societal obligations can be integrated in the process of species recovery and made transparent through the formal process of structured decision-making to honor moral, legal, and societal obligations and duties.

Please note: The Journal of Fish and Wildlife Management is not responsible for the content or functionality of any supplemental material. Queries should be directed to the corresponding author for the article.

Reference S1. Barry PM, Hudson JM, Williamson JD, Koski ML, Clements SP. 2014. Clackamas River bull trout reintroduction project. 2013 Annual Report. Salem: Oregon Department of Fish and Wildlife; and Vancouver, Washington: U.S. Fish and Wildlife Service.

Found at DOI: https://doi.org/10.3996/062017-JFWM-051.S1 (1.69 MB PDF).

Reference S2. Blumm MC, Steadman JG. 2009. Indian treaty fishing rights and habitat protection: the Martinez Decision supplies a resounding judicial reaffirmation. Natural Resources Journal 49:653–706.

Found at DOI: https://doi.org/10.3996/062017-JFWM-051.S2 (352 KB PDF).

Reference S3. Buchanan DV, Hanson ML, Hooton RM. 1997. Status of Oregon's Bull Trout: distribution, life history, limiting factors, management considerations, and status. Portland, Oregon: Technical Report to Bonneville Power Administration, Project 199505400.

Found at DOI: https://doi.org/10.3996/062017-JFWM-051.S3 (21.29 MB PDF).

Reference S4. Flagg TA, Nash CF, editors. 1999. A conceptual framework for conservation hatchery strategies for Pacific salmonids. Seattle: National Marine Fisheries Service, Northwest Fisheries Science Center. U.S. Department of Commerce, National Oceanic and Atmospheric Administration technical memorandum NMFS-NWFSC-38.

Found at DOI: https://doi.org/10.3996/062017-JFWM-051.S4 (436 KB PDF).

Reference S5. Fredenberg W. 1998. Experimental bull trout hatchery. Progress report two: experimental broodstock development, 1995–1997. Kalispell, Montana: U.S. Fish and Wildlife Service, Creston Fish and Wildlife Center.

Found at DOI: https://doi.org/10.3996/062017-JFWM-051.S5 (874 KB PDF).

Reference S6. Fredenberg W, Dwyer P, Barrows R. 1995. Experimental bull trout hatchery. Progress report, 1993–1994. Kalispell, Montana: U.S. Fish and Wildlife Service, Creston Fish and Wildlife Center.

Found at DOI: https://doi.org/10.3996/062017-JFWM-051.S6 (1.39 MB PDF).

Reference S7. Hard JJ, Jones RP, Delarm MR, Waples RS. 1992. Pacific salmon and artificial propagation under the Endangered Species Act. U.S. Department of Commerce. National Oceanic and Atmospheric Administration Technical Memorandum NMFS-NWFSC-2. Seattle: National Marine Fisheries Service, Northwest Fisheries Science Center.

Found at DOI: https://doi.org/10.3996/062017-JFWM-051.S7 (104 KB PDF); also available at http://www.krisweb.com/biblio/gen_nmfs_hardetal_1992_tm2.pdf.

Reference S8. McPhail JD, Baxter JS. 1996. A review of bull trout (Salvelinus confluentus) life-history and habitat use in relation to compensation and improvement opportunities. British Columbia Ministry of the Environment - Fisheries Branch. Fisheries Management Report No. 104.

Found at DOI: https://doi.org/10.3996/062017-JFWM-051.S8 (469 KB PDF); also available at http://ww.krisweb.com/biblio/kootenai_bcfish_mcphailetal_1996_bull.pdf.

Reference S9. [NOAA] National Oceanic Atmospheric Association. 2008. Executive summary of the FCRPS 2008 biological opinion. National Oceanic Atmospheric Association.

Found at DOI: https://doi.org/10.3996/062017-JFWM-051.S9 (1.11 MB PDF); also available at http://www.westcoast.fisheries.noaa.gov/publications/hydropower/fcrps/2008fcrps_execsummary.pdf.

Reference S10. Pister EP. 1997. Ethics of bull trout restoration and management. Pages 15–20 in MacKay WC, Brewin MK, Motina M, editors. Friends of the Bull Trout conference proceedings. Calgary, Alberta, Canada: Bull Trout Task Force, Trout Unlimited.

Found at DOI: https://doi.org/10.3996/062017-JFWM-051.S10 (73 KB PDF).

Reference S11. Rieman BE, McIntyre JD. 1993. Demographic and habitat requirements for conservation of bull trout. Ogden, Utah: U.S. Forest Service, Intermountain Research Station General Technical Report INT-302.

Found at DOI: https://doi.org/10.3996/062017-JFWM-051.S11 (2.82 MB PDF).

Reference S12. Rode M. 1990. Bull trout, Salvelinus confluentus Suckley, in the McCloud River: status and recovery recommendations. Inland Fisheries administrative report no. 90-15. Redding: California Department of Fish & Game.

Found at DOI: https://doi.org/10.3996/062017-JFWM-051.S12 (8.46 MB PDF).

Reference S13. Schaller HA, Budy P, Newlon C, Haeseker SL, Harris JE, Barrows M Gallion D, Koch RC, Bowerman T, Conner M, Al-Chokhachy R, Skalicky J, Anglin D. 2014. Walla Walla River bull trout ten year retrospective analysis and implications for recovery planning. Vancouver, Washington: U.S. Fish and Wildlife Service, Columbia River Fisheries Program Office.

Found at DOI: https://doi.org/10.3996/062017-JFWM-051.S13 (17.47 MB PDF); also available at https://www.fws.gov/columbiariver/publications/10YearWallaWallaBullTroutSynthesis_FINAL_9_30_14.pdf.

Reference S14. Shively D, Allen C, Alsbury T, Bergamini B, Goehring B, Horning T, Strobel B. 2007. Clackamas River bull trout reintroduction feasibility assessment. Sandy, Oregon: U.S. Department of Agriculture Forest Service, Mt. Hood National Forest; Portland: U.S. Fish and Wildlife Service, Oregon State Office; and Clackamas: Oregon Department of Fish and Wildlife, North Willamette Region.

Found at DOI: https://doi.org/10.3996/062017-JFWM-051.S14 (3.29 MB PDF).

Reference S15. [UWBTWG] Upper Willamette Bull Trout Working Group. 2010. Upper Willamette Basin bull trout action plan 2010, Oregon, USA. Upper Willamette Bull Trout Working Group.

Found at DOI: https://doi.org/10.3996/062017-JFWM-051.S15 (170 KB PDF); also available at https://nrimp.dfw.state.or.us/CRL/Reports/MWBTRP/UWBT_ActionPlan_2010.pdf.

Reference S16. [USFWS] U.S. Fish and Wildlife Service. 2008. Strategic habitat conservation handbook: a guide to implementing the technical elements of strategic habitat conservation. Washington, D.C.: U.S. Fish and Wildlife Service.

Found at DOI: https://doi.org/10.3996/062017-JFWM-051.S16 (2.39 MB PDF); also available at https://www.fws.gov/landscape-conservation/pdf/SHCHandbook.pdf.

Reference S17. [USFWS] U.S. Fish and Wildlife Service. 2013. Recovery plan for the black-footed ferret (Mustela nigripes). Denver: U.S. Fish and Wildlife Service.

Found at DOI: https://doi.org/10.3996/062017-JFWM-051.S17 (1.28 MB PDF).

Reference S18. [USFWS] U.S. Fish and Wildlife Service. 2014. Federal and state endangered and threatened species expenditures. Fiscal year 2012. Washington, D.C.: U.S. Fish and Wildlife Service.

Found at DOI: https://doi.org/10.3996/062017-JFWM-051.18 (2.45 MB PDF).

Reference S19. [USFWS] U.S. Fish and Wildlife Service. 2015. Recovery plan for the coterminous United States population of bull trout (Salvelinus confluentus). Portland, Oregon: U.S. Fish and Wildlife Service.

Found at DOI: https://doi.org/10.3996/062017-JFWM-051.S19 (8.72 MB PDF).

Reference S20. [USGAO] U.S. Government Accountability Office. 2005. Fish and Wildlife Service generally focuses recovery funding on high priority species, but needs to periodically assess its funding decisions. USGAO Report No. GAO-05-211. Washington, D.C.: U.S. Government Accountability Office.

Found at DOI: https://doi.org/10.3996/062017-JFWM-051.S20 (1.28 MB PDF).

Reference S21. Williams BK, Szaro RC, Shapiro CD. 2009. Adaptive management: the U.S. Department of the Interior technical guide. Adaptive Management Working Group, Washington, D.C.: U.S. Department of the Interior.

Found at DOI: https://doi.org/10.3996/062017-JFWM-051.S21 (38.19 MB PDF).

We would like to acknowledge Kathleen Dean Moore, Douglas E. Olson, James T. Peterson, Jason B. Dunham, Mark Bagdovitz, Don Campton, an anonymous Associate Editor, and three anonymous reviewers, for their thoughtful discussions and comments of previous drafts of this manuscript.

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