The evolution of the modern dam conflict on the Snake River, USA

ABSTRACT

As dams age and values shift, communities face dam removal decisions that involve navigating complex social, economic and ecological interactions. Sometimes, this results in decades-long conflicts, such as that over the removal of the four lower Snake River dams (LSRD) in Washington State, USA – the focus of this study. We apply a broad analytical framework to understand how key factors in this conflict perpetuate it. We find that western science and economic expertise are politicized, and that while venues and geographies for stakeholder engagement are expanding, these shifts are not powerful enough to interrupt the cycle of litigation sustaining the LSRD debate.

KEYWORDS:

• Dams
• dam removal
• salmon
• conflict
• politicization
• Snake River

Introduction

A 2021 United Nations (UN) report describes the ageing of tens of thousands of large dams as ‘an emerging global development issue’ (Perera et al., 2021, p. 4). Today, communities worldwide are pondering repair or removal costs, debating dam effectiveness, and weighing human safety and environmental health concerns. These decisions associated with managing dams are not without conflict, as demonstrated by notable dam removal deliberations over the Klamath basin dams in the USA, the Sélune River dams in France, and the Mullaperiyar dam in India.

In the USA today, dam removal outpaces dam building (Moran et al., 2018). Dams are considered for removal for a variety of reasons. For example, in the New England region, numerous dams built in the 19th and early 20th centuries have become unsafe as they age, or they have outlived their original, intended uses and no longer provide economic benefits (Fox et al., 2016). Dam removal has also become a tool for achieving ecological restoration (Hart et al., 2002). In the Pacific Northwest (PNW), for example, where some dams endanger critical habitats or impede species’ life cycles, the jeopardy of endangered species in particular has played a critical role in identifying dams for removal (Dorning, 2018; Magilligan et al., 2017).

As dams age and societal values evolve, it becomes important to develop approaches for understanding the role of dams in our water future (Ho et al., 2017), including dam removal and the accompanying conflicts and processes. Furthermore, while thousands of dams face demolition (largely in developed nations, and particularly in North America and Europe; Liuyong et al., 2019; Moran et al., 2018), simultaneously, thousands of dams are being planned and constructed worldwide (dominantly in developing countries; Moran et al., 2018). This is because hydropower is considered an essential tool for catalysing the global energy transition (International Renewable Energy Agency (IRENA), 2020), and because dams have the potential to be a tool for development, when executed sustainably (World Commission on Dams, 2000). A better understanding of the causes, mechanisms and implications associated with dam removal conflicts is therefore important for more cooperative dam end-of-life processes, for both existing and future dams, supporting the sustainable management of water infrastructure.

This study analyses the modern conflict over the management of the four lower Snake River dams (LSRD) in the Pacific Northwest of the USA. We integrate across the existing scholarship to discern and focus on three key factors that researchers propose shape modern dam conflicts by either perpetuating the conflict or helping to resolve it. These include the roles of western science (hereafter ‘science’), economic analysis and stakeholder interactions. The LSRD provide a valuable case to study for two reasons. First, the conflict is both ongoing and long-lived, stretching over three decades. This allows us to identify factors that are salient to the conflict, regardless of whether or not they lead to a particular outcome. Second, nearly all the dams removed in the USA so far have been small (< 15 m high) and privately owned, and the large dams that have been the subject of dam removal conflicts have mostly been subject to Federal Energy Regulatory Commission (FERC) relicensing. The FERC issues 30–50 year-long licences to privately owned (and some publicly owned), non-federal hydropower projects. Since hydroelectric dam construction began in the mid- to late-20th century, and since the initial licensing period lasts 50 years, the USA is in the midst of numerous relicensing processes taking place under FERC (Chaffin & Gosnell, 2017), highlighting the relevance of that venue and process in present day dam removal conflicts. However, the LSRD are large, federally owned dams, not subject to FERC licensing, and therefore may lend different lessons on dam removal conflicts.

Frameworks for studying dam removal conflicts

Dam removal conflicts are studied from a social perspective in an effort to reveal the characteristics of conflicts and share lessons learned, with the normative aim that an understanding of what happened may lessen the negative impacts of such conflicts elsewhere (e.g., Albertson, 2019; Grabowski et al., 2017; Sneddon et al., 2017a). Many studies also aim to elucidate the socio-ecological dynamics of dam removal conflicts, aiming to influence future processes to be inclusive of not only biological and physical considerations of river restoration, but also the social, cultural and political dimensions (Fox et al., 2016; Gosnell & Kelly, 2010; Sneddon et al., 2017a).

Researchers embrace political ecology, ecosystem services, actor–network theory, critical physical geography, and science and technology studies, among others, for studying dam conflicts in the USA. A common theme is to focus on the nature of dam removal conflicts by seeking to understand the social, cultural, political and environmental variables that cause them, perpetuate them and lead to resolution. A rich body of case-based scholarship exemplifies such approaches (e.g., Albertson, 2019; Barraud, 2017; Brewitt, 2019; Fox et al., 2016; Gabrowski et al., 2017). Researchers caution against finding patterns and seeking generalizations across cases of environmental conflict because conflicts are contingent on historical–geographical conditions and institutional networks of power and influence, which vary between regions, making it difficult to compare dam removals (Magilligan et al., 2017; Sneddon et al., 2017a). With this in mind, we turn to the two most well-studied regions characterized by dam controversies in the USA: New England and the Pacific Northwest.

Fox et al. (2016) studied dam removal conflicts in New England and revealed how micropolitics, cultural dynamics and divergent interpretations of nature are all key parts of ecological restoration, finding no single, key variable that explains or resolves these dam removal conflicts. They uncovered that resistance to dam removal was not due to a lack of knowledge, but rather that local residents valued aspects of their environment differently than the scientists or environmentalists seeking dam removal. In their study of how expert knowledge is disseminated, reconfigured, and contested within regulatory and cultural settings also in New England dam conflicts, Sneddon et al. (2017b) found that environmental advocacy groups and state agencies view dam removals as a way to rehabilitate river systems, and use science-based arguments to support removal, whereas opponents highlight their place-based knowledge and question motives of expert knowledge. Magilligan et al. (2017) found that where removals occurred in New England, conflicts were arbitrated by paying attention to local historical–geographical conditions and by brokering effective compromises between dam owners and various local stakeholders. In contrast, where conflict endured or removal efforts failed, dam removal proposals were part of broader projects initiated by external stakeholders without genuine input or buy-in from local residents.

In the Pacific Northwest, FERC relicensing processes have played a key role in providing a window of opportunity for dam removal. This is because the Federal Power Act that lends FERC its relicensing responsibilities stipulates that hydropower licences must be in the public interest, giving equal consideration to both power development and environmental considerations, potentially opening a space every 30–50 years for those in dammed watersheds to renegotiate societal values, which may ultimately lead to dam decommissioning and removal, rather than relicensing (Chaffin & Gosnell, 2017). Many dams in the region are considered for removal because of their impacts on salmon. Most recently, after decades of conflict, agreements have been signed to remove the Klamath dams, which may be the USA’s largest dam removal project to date. Gosnell and Kelly (2010) found that a robust legal framework and both leadership and science provided by non-governmental organizations (NGOs) were key in reaching an agreement to remove the dams. In this case, science and economics shifted in favour of dam removal, the federal government’s commitment to negotiate an agreement supporting tribal sovereignty played a key role, and local leadership moved the debate beyond a cycle of litigation (Chaffin & Gosnell, 2017). Brewitt (2019) considered the roles of political frames, advocacy coalitions and venue shopping in the Elwha, Savage Rapids and Marmot dam removal controversies. Via a comparison of the three case studies, he hypothesizes that the most important frames in dam removal politics are cultural and emotional, that stakeholders must form mega-coalitions to win political backing and power, and that venues inclusive of all stakeholders allow for conflicts to be resolved. Figure 1 reports key factors impacting conflict and cooperation in dam removal conflicts as identified in the scholarship. Colour-coding represents an interpretation by the authors of categories in which the key factors can be grouped.

Figure 1. Summary of the key factors found by researchers to be relevant to dam removal controversies in the USA.

The case of the four lower Snake River dams

A brief history of the LSRD conflict

By the early 20th century, hydropower had become a key instrument for economic development across the American West (Ho et al., 2017). The largest share of hydropower in the region is derived from the Columbia River Basin, where there are 31 hydroelectric projects, of which a subset of 14 federal projects are managed together by the US Army Corps of Engineers (Corps) and the US Bureau of Reclamation as a coordinated system, which includes the four LSRD (specifically owned and operated by the Corps). The four LSRD – Ice Harbour, Lower Monumental, Little Goose and Lower Granite – were built between 1962 and 1975 by the Corps in south-eastern Washington, where they span the Snake River between Lewiston, Idaho and its confluence with the Columbia River (Figure 2). The main purpose of the dams is for navigation, irrigation and hydroelectric power generation. With an estimated ratio of 15 cents of benefit for every US$1 of cost, Congress authorized the LSRD as ‘make-work’ projects for soldiers returning from the Second World War (Blumm, 2021).

Figure 2. Locations of the four lower Snake River dams (LSRD) within the Columbia River basin.

Declining populations of anadromous fish in the Snake River were documented concurrent with the building of the LSRD, and beginning in 1991, salmon species were listed as endangered under the federal Endangered Species Act (ESA). There are now 13 ESA-protected fish in the Columbia River Basin, and in 2005 southern resident killer whales (hereafter ‘orcas’) were also listed as endangered, as decreasing salmon populations, their primary prey, is one of multiple threats to their continued survival (National Oceanic and Atmospheric Administration (NOAA), 2005). Severe declines in salmon populations triggered calls for removal of the LSRD. Salmon is economically, spiritually and culturally important in the region, especially for the many Native American tribes of the Columbia River Basin. The federal Bonneville Power Administration (BPA) markets the power generated by the LSRD and is required to pay for environmental mitigation to compensate for the dams’ environmental impacts. Although the approximately US$17 billion BPA-funded programme is the largest fish and wildlife restoration programme in the USA, the fish remain in peril today.

The conflict over whether or not to remove the dams has stretched across three decades and involves numerous, diverse stakeholders. We define stakeholders broadly, as any relevant person, group or organization with an interest in the issue either because they will be impacted by it or because they have influence on, knowledge of, or experience with the issue (European Commission, 2003). Key stakeholders involved include environmental organizations, recreational and fishing interests, farmers, local residents, elected officials, state and federal agencies, federal courts, tribes, the transportation industry, industry associations, utilities, consultants, and scientists. The Corps is the owner and operator of the LSRD and together with the US Bureau of Reclamation and BPA oversee the Columbia River System Operations National Environmental Policy Act (NEPA) process. The NOAA Fisheries office (formerly the National Marine Fisheries Service – NMFS) is also an important stakeholder, as the producer of the biological opinions (BiOps), which direct federal agencies’ actions regarding managing the dams to avoid jeopardizing ESA-listed species.

Salmon is at the heart of this environmental, economic and social conflict. The conflict is complex because not only does disagreement abound over how to recover salmon, but also stakeholders define the problem differently. Put generally, proponents of dam removal, including the tribes and conservation NGOs, advocate for breaching the LSRD because they link the dams directly to the demise of salmon (and orca) populations. There are also calls for dam breaching to return traditional sites and fisheries to Columbia Basin tribes. Other proponents of dam breaching calculate that a free-flowing river would enhance the local economic environment through commercial fishing and outdoor recreation, and would reduce taxpayer subsidies to corporate irrigators and barging companies. In contrast, opponents of dam breaching, such as hydropower, agriculture and transportation interests, discount these links and attest to the economic value associated with the dams. Some dam breaching opponents contend that the dams are not the major stressor of the fish, and advocate for other mitigation measures instead of dam removal, and others yet see the LSRD as key to helping mitigate climate change via hydroelectricity.

Analytical framework and methods

In this paper we investigate key factors that explain the persistence of the LSRD removal conflict by drawing from previous studies and new data. We see conflict as disagreement over the course of action to be taken (Mostert, 1998) due to factual disagreement, conflicting goals and relational aspects (Dorcy & Riek, 1987, cited in Mostert, 1998). We also recognize conflict as context-specific, multicausal and multidimensional, and resulting from institutional, socio-economic, and resource and environmental factors (Haider, 2014, cited in Zeitoun et al., 2020). We draw from earlier scholarship that recognizes a multitude of factors salient to dam removal conflicts (Figure 1) to focus on the science–economy–stakeholder nexus with our analytical framework (Figure 3).

Figure 3. Our analytical framework for analysing dam removal conflicts.

In drawing from earlier scholarship, we find a number of factors salient to dam removal conflicts identified in the literature (Figure 1). We recognize three categories that encompass many of these factors and use them as a broad framework for better understanding the modern Snake River dam conflict, which include: the role of science, the role of economics and the role of stakeholder interactions (Figure 3). We chose these three categories for our analysis because they best encompass the most common factors salient to the USA dam removal cases that have been previously analysed in the literature. First, the biophysical impacts of dams are salient to dam removal debates, hence the prominence of ‘scientific understandings’ as key factors in ensuing conflict and cooperation dynamics. In this framework, as applied to the LSRD case, we refer to this category broadly as ‘science’. For our case study, we have further refined this category to ‘western science’ due to the nature of the data available to us, although we recognize that traditional ecological knowledge and other ways of knowing are also important, while outside of the scope of this work. This category encompasses knowledge about species, the environment, technology, etc. Our second factor is the role of economics and refers to the monetary and non-monetary calculations of the benefits and costs of dams that factor into many dam removal debates. Lastly, stakeholder interactions include venues and practices of stakeholder engagement that are important to shaping dam removal debates. This category encompasses an understanding of who the stakeholders are and how and where they interact around the prospect of dam removal. Other factors that researchers have found to be influential on the debate include politics, culture and legal frameworks. We see these factors as part of the broader landscape that influences all three of the major categories we have chosen to analyse, and therefore they are not separate categories on their own.

We follow a multi-method approach to examine the factors that characterize the dam removal debate in the lower Snake River (Johnson et al., 2007). Both primary and secondary sources of data were used, including peer-reviewed literature, newspaper articles, consultant and government reports, publicly available letters, and court cases. In addition, we conducted 10 targeted interviews with experts and stakeholders of the LSRD conflict to better understand different perspectives and to test our assumptions (see the supplemental data online). Interviewees included scientists, consultants, university researchers, association representatives, government representatives and NGOs, representing both dam removal and dam retention preferences. Interviews were conducted virtually through Zoom or over the phone and lasted from 30 minutes to two hours. Interviewees were informed of the research objectives and gave their consent to be interviewed with recognition that their identity would be concealed.

The nature of the modern LSRD conflict

The role of science

Scientific knowledge is one type of knowledge that features prominently in dam removal conflicts. Western scientific knowledge, produced by university scientists, federal agencies, independent institutions, non-profit organizations and consultants, dominates. In the case of the LSRD debate, ecological and biological scientific information are particularly valued because anadromous fish are at the heart of the conflict. Science is key in the debate, as stakeholders use scientific framing to define the policy problem, to describe solutions, to support their positions and to undermine the positions of other actors (Hall & White, 2008).

Western scientific knowledge is also particularly salient in the LSRD debate because it is used by the federal agencies to inform dam management strategies, which must not jeopardize ESA-listed species. The science of the BiOps has frequently been a contentious subject since they were first challenged in court in the early 1990s. Referring to a more recent NOAA Fisheries BiOp (2014), a federal court ruled, ‘NOAA Fisheries’ analysis does not apply the best available science, overlooks important aspects of the problem, and fails properly to analyse the effects of climate change […]’ (National Wildlife Federation (NWF) et al. v. National Marine Fisheries Service (NMFS) et al., 3:01-cv-00640-SI, 2016, p. 15). From the perspective of one stakeholder familiar with the litigation, ‘the science has been very central to the litigation’, and ‘by and large the court has tended to credit more the science from the CSS [Comparative Survival Study by the Fish Passage Center] and that kind of independent [science], than the agency science’ (interview 8).

There exists a dominant scientific narrative that the LSRD are largely responsible for the declining salmon population over the last 60 years, and that breaching the four LSRD is the best option for protecting and restoring salmon and steelhead (salmonids). This conclusion has broad support among university scientists, the Fish Passage Center, and representatives of involved federal and state agencies (US Army Corps of Engineers et al., 2020; Filardo et al., 2021), and some interviewed stakeholders in opposition to the dams argue that scientific consensus has been reached (interviews 3, 6, 8 and 9). The most recent Columbia River Basin dams environmental impact statement (EIS; a report mandated by the National Environmental Policy Act of 1969 to assess the potential impacts of actions, in this case the management of the dams), authored by the three federal agencies overseeing the dams, concludes that breaching the four LSRD would have ‘the highest potential benefits for Snake River salmon and steelhead’, although the agencies do not choose dam breaching as a management strategy going forward (US Army Corps of Engineers et al., 2020, p. 29). Groups of scientists have co-authored letters to assure decision-makers of such scientific consensus, leveraging the authority of science (Yearley, 1996, cited in Hall & White, 2008). Most recently, following the release of an EIS in 2020, 10 prominent scientists penned a letter to the governors of Washington, Oregon, Montana and Idaho, urging them to support dam removal, stating that, ‘we base our opinion on our deep expertise in the science of salmon and steelhead conservation including decades of collaborative research that has withstood rigorous scientific review’ (Williams et al., 2021, para. 7). One month later, another letter, ‘authored and signed by 68 of the nation’s and region’s best qualified salmon and fisheries experts’, was sent to government officials and policymakers, summarizing, ‘the scientific consensus on key aspects of salmon recovery’ (Filardo et al., 2021, para. 1).

Yet, other stakeholders disagree, such as organizations representing electric utilities and agricultural interests, who argue that there is no such scientific basis. As one stakeholder put it, ‘because so many people believe that they [the dams] are [the limiting factor], that an entire wing of science essentially has been established to try to prove that they’re the limiting factor’ (interview 5). Stakeholders supporting the LSRD point to recent peer-reviewed work, such as a study funded by Kintama Research Services and the BPA, finding that Chinook salmon survival is likely driven by broad oceanic factors (Welch et al., 2020), a study that has been critiqued as technically flawed by the Fish Passage Center, a research centre funded by BPA that produces data on fish passage for fishery agencies and tribes (DeHart, 2020).

Although the science linking the dams to salmon decline is supported by a majority of scientists, it remains contested by stakeholders against dam removal. A 2019 written exchange in the Idaho Statesman newspaper between a representative of the Northwest River Partners (NRP), a non-profit organization representing community-owned electric utilities formed in response to the 2004 court challenge to the NOAA Fisheries BiOp, and a representative of Trout Unlimited (TU), a non-profit environmental organization with a mission to conserve fisheries and their watersheds, illustrates both the authority and contestation of science in the conflict. The NRP official questions the impact that the LSRD have had on salmon, stating that ‘there is actually very little debate in the region about the fact that salmon populations were already largely decimated due to the commercial harvest that occurred well before Bonneville dam was completed in 1938’, and that ‘there is, however, a wealth of science that has consistently shown a direct tie between ocean temperatures and the number of returning adult salmon. The higher the temperature, the worse the runs will be’ (Miller, 2019). The TU official responded by criticizing the science cited by NRP, rebutting that they ‘spliced together several pieces of information – some accurate, some not – that ultimately didn’t support his conclusion’. The TU official directly links the dams with salmon decline, stating that the author ‘says he supports science-based decision-making, so I ask him to join me in looking at the overwhelming scientific evidence that we need a free-flowing lower Snake River’ (Neville, 2019).

The role of economics

A cost–benefit analysis (CBA) is often applied to evaluate the socio-economic and environmental trade-offs of building, rehabilitating or removing dams in the USA (e.g., Edward’s Dam in Maine; Gowan et al., 2006). Although CBA is a common economic tool used for deciding between policy options, it is difficult to assure all costs and benefits are accurately accounted for, as many are difficult to measure, such as the value of preserving a culturally important species (Whitelaw & MacMullan, 2002).

CBAs for the LSRD have been conducted by both the federal dam management agencies and consultants, often contracted by non-profits, and are frequently contentious. The federal management agencies have conducted two CBAs as part of environmental impact assessments they undertook as required by NEPA. First, in 2002, the Corps published their EIS, in response to the listing of the Snake River salmonids under the ESA, which included a CBA of various management scenarios for the four LSRD, including one option to breach the dams, which was not favoured (Walla Walla District Corps of Engineers, 2002). The 2002 EIS CBA has been criticized in peer-reviewed journals, newspapers and reports. Consultants often review the federal CBAs. For example, Whitelaw and MacMullan (2002) of ECONorthwest criticized the Corps for not considering environmental quality, regional economic development and social impacts when measuring the costs and benefits of different dam management strategies. Similarly, the National Research Council concluded that the Corps’ approach ignored important impacts, was out of date and did not reflect current thinking on the role of water resources in local, regional or national economies (NRC, 1999, cited in Whitelaw & MacMullan, 2002). Furthermore, in 2015, Waddell, a retired Corps employee, and Linwood re-evaluated the 2002 Corps CBA, concluding that the costs of keeping the dams were underestimated. Their analysis was refuted that same year by another former Corps economist in a report published by the Pacific Northwest Waterways Association, which represents navigation interests (Wagner, 2015).

In a successful challenge of the 2014 BiOp and for non-compliance with NEPA, the federal dam management agencies were ordered by the court to complete a new EIS by 2021 (completed in 2020). Accompanying and following these litigious events were additional commentaries and analyses of the federal CBAs, and new CBAs by non-profit groups and consultancies, who also critiqued each other’s analyses. Some of these analyses focused on just one aspect of the issue, such as the costs and benefits of the barging allowed by the LSRD. The report by consultancy RME commissioned by Save Our Wild Salmon, for example, showed that the costs of barging outweighed the benefits. Conversely, the 2020 study by consultants FCS Group commissioned by the Pacific Northwest Waterways Association concluded that losing barging would result in severe economic damage and increase carbon emissions. Others have been more encompassing, such as the 2016 report by the non-profit Earth Economics, which reviewed the 2002 federal CBA, calculating instead that the dams’ costs outweigh their benefits, and that a free-flowing river would have significantly higher benefits than costs (Mojica et al., 2016). Focusing on the energy mix, a 2018 study commissioned by the Northwest Energy Coalition found that it would be feasible and affordable to replace the LSRD hydropower with a balanced portfolio of clean energy resources, considering adequacy and reliability, cost, and emissions. The Portland-based economics firm ECONorthwest published a CBA in 2019, concluding that dam removal is the most beneficial option when non-use values are considered in the calculations. In refute, US Congressional Representatives Cathy McMorris Rogers (Republican from Spokane, Washington) and Dan Newhouse (Republican from Yakima, Washington) characterized the ECONorthwest report as, ‘another example of Seattle-based interests seeking to disrupt our way of life in Central and Eastern Washington’ (McMorris Rogers, 2019).

Most recently in 2020, in compliance with NEPA and Judge Simon’s 2016 orders to evaluate how to ensure that the management of the dam system is not likely to jeopardize endangered or threatened species (NWF et al. v. NMFS et al., 3:01-cv-00640-SI, 2016), the Columbia River System Operation (CRSO) EIS was published, which included the second CBA conducted by the federal agencies that manage the dams. The delivery date for this EIS was pushed forward by a year by the Presidential Memorandum on Promoting the Reliable Supply and Delivery of Water in the West, an order that was seen as favourable by those advocating for keeping the dams and as a political manoeuvre by those who advocate for dam removal according to newspapers (Mehaffey, 2018). The CRSO chose an operating option that includes maintaining the four LSRD and making management and structural changes to improve the passage of and conditions for impacted ESA-listed species. In the EIS, the federal agencies responsible for the dams analysed the potential effects of removing the dams, concluding that dam removal would have significant, adverse economic impacts (Moir et al., 2020). Like its predecessor, this 2020 EIS CBA also has come under scrutiny. The state of Oregon and also Earthjustice, on behalf of 11 environmental and fishing organizations, filed a notice of their intent to sue the federal government over the planned CRSO. The Save Our Wild Salmon Coalition (2019) claims that the 2020 EIS ignores the public’s value of restored salmon and orca populations, as the EIS does not consider non-use values in its analysis.

The role of stakeholder interactions

Dam removal conflicts play out in both collaborative and litigious decision-making venues, with stakeholders tending to strategically choose venues in which they can best promote their cause (Brewitt, 2019; Pralle, 2006). The courts have been a prominent venue for dam removal debates, especially when federal regulatory processes are involved, such as FERC relicensing, ESA compliance, and National Resource Damage Assessment processes (Lenhart, 2003). A key characteristic of the conflict over the LSRD is that for nearly three decades the conflict has largely played out in federal court (e.g., Blumm & DeRoy, 2019; Mulligan & Upton, 2016). Litigants have included environmental and fishing organizations, state and federal agencies, Native American tribes, electric utilities and river-dependent industries.

Since NOAA Fisheries began issuing BiOps in 1992 for the Columbia River Basin salmon and steelhead, every BiOp issued has been challenged in court. The first 1992 BiOps were challenged unsuccessfully in court by power users. The next BiOp, published in 1993, was challenged by Idaho’s Department of Fish and Game and was found to be arbitrary and capricious because it relied on improper data to calculate the baseline number of protected salmonids in the Columbia Basin (Mulligan & Upton, 2016). The 1995 BiOp was also challenged twice in court, first by a group of plaintiffs led by American Rivers, and subsequently by a group of hydroelectric power users, but both challenges were unsuccessful. The 2000 BiOp was challenged by the National Wildlife Federation because it did not comply with the ESA, marking the beginning of a cycle of litigation around ESA compliance that continues to this day.

In 2008, the Columbia River Inter-Tribal Fish Commission, along with tribes from the region, the states of Idaho, Montana, and Washington, and the federal dam management agencies signed the Columbia Basin Fish Accords (BPA et al., 2008), securing approximately US$900 million for salmon restoration projects in the basin, funded by the BPA. They also secured the tribes’ commitment to no longer pursue litigation for fish protection, to support the 2008 BiOp in any future litigation, and to refrain from advocating for dam removal. The tribes that agreed to the Fish Accords had previously been allied with the plaintiffs challenging the BiOps in court. Environmental groups, fishing interests, the Nez Perce Tribe, and the state of Oregon were not parties to the Fish Accords. Although the Fish Accords agreements have since been extended, the cycle of litigation continues. Recently, a broad coalition of environmental and fishing organizations represented by Earthjustice and the Earthrise Law Center have again filed a supplemental complaint against the federal dam management agencies over violations of the ESA and NEPA, challenging the 2020 CRSO EIS, Record of Decision, and recent BiOp (Westney, 2021).

Beyond the courtroom as the dominant formal venue for the LSRD conflict, over the past decade, there have been several attempts to foster dialogue by expanding venues for stakeholder engagement. One notable example is the Columbia Basin Partnership (CBP), a task force established by NOAA that convened from 2017 to 2020. This partnership included the Columbia Basin tribes; fishing, agriculture, conservation, river transportation, port and hydropower interests; and the states of Idaho, Montana, Washington and Oregon. A task force member representing TU characterized the group’s work this way:

the simple way I would put it is that as long as there is a salmon problem there is a people problem. We need to bring people together to coordinate actions and strategies across the basin – collaboration and cross-sector problem solving. We laid a good foundation for that to happen moving forward. (Harrison, 2020)

The CBP was instrumental in not only bringing together all stakeholders for conversation, but also in facilitating work to define a common problem and set a goal for salmon recovery in the Columbia River Basin (Columbia Basin Partnership Task Force, 2019, 2020). Following from recommendations of this task force, Oregon, Washington, Idaho, and Montana committed to undertaking a collaborative, public process together and in collaboration with the region’s tribes, federal agencies, and other stakeholders to rebuild the basin’s salmon and steelhead stocks.

In February 2020, stakeholders who have long been at odds with one another in the LSRD conflict authored a joint letter to the governors of the four states involved, urging them to use the then-upcoming CRSO EIS as an opportunity to collaboratively develop a plan to recover the fish, fulfil tribal needs, and strengthen electricity and agricultural services (Hagan et al., 2020). The signatories called for the EIS to be used to ‘have a different dialogue’ and stated that they are ‘committed to working together’. The intent to collaborate and expand new venues for engagement can also be recognized at the national scale. In October 2020, representatives from the US hydropower industry and environmental conservation associations issued a Joint Statement of Collaboration on US Hydropower, which was the outcome of a two-and-a-half-year discussion facilitated by Stanford University’s Uncommon Dialogue process. Institutions involved in the LSRD conflict are signatories, such as American Rivers and the National Hydropower Association, of which the BPA is a member.

Although the decline in salmon remains the central issue in the LSRD conflict, stakeholders have also begun to raise concerns over the dams’ indirect impacts on orcas, seals, sea lions, lampreys, and birds, shifting the geography of the conflict by re-defining the issue. In particular, over the last decade, stakeholders have made visible the correlation between declining salmon numbers and declining orca numbers (interview 3). This issue expansion can perhaps best be marked by the 2015 founding of the Orca Salmon Alliance, which united activists from the local to the international scale, including NGOs, researchers and community action groups in a mission to prevent the extinction of orcas by recovering the salmon populations on which they depend. The benefits from the linkage are mutual: the orcas have become a symbol, or even a ‘mascot’ (interview 5), for those advocating for dam removal, and the salmon issue has become an amplifier for the plight of the orca. The executive director of the Nez Perce Tribe, which advocates for dam removal, said, ‘to have the orca in peril, I hate to see it. But it shakes up this cause even more, it makes people think. You have to connect it to the cycle of life, you see all the species connected’ (Mapes, 2018). In contrast, stakeholders from Tri-Cities, Washington wrote:

after decades of using other arguments in their attempts to justify removing the four lower Snake River dams, some environmental groups have latched on to the plight of the orcas […] these groups are using the orcas to play on the public’s sympathies and promote their real goal: removal of the four lower Snake River dams. (Adrian et al., 2019, p. 2)

Additionally, the expansion of the issue has led to an increase in stakeholders, particularly from the orca community (interview 3). There is evidence for this in the letters sent to policymakers (see section titled 'The role of science'). In contrast to previous letters, largely from experts in fish science, a 2018 letter to Washington Governor Jay Inslee was signed by scientists who pointed out that ‘we do not specialize in fisheries biology’ (Giles et al., 2018, p. 1). New voices have shifted the geography and the scale of the debate, from salmon and their freshwater habitats in eastern Washington to orcas feeding in the marine waters of western Washington.

Discussion

In this paper we analyse the current character of the LSRD conflict in terms of the three categories identified in the analytical framework. Using a mixed-methods approach, we gathered evidence around how western scientific knowledge is being used, how economic analyses are being developed and used in the debate, and the ways in which stakeholders are interacting in this conflict. In this section we discuss our findings around these three factors, including how the three factors influenced each other, what impact they had on the debate, and our observations about the importance of not only what science, economics or venues are being used, but importantly, how. We then use our findings to refine the three-factor framework presented above, so that it can be best suited for use understanding other dam removal conflicts.

Interrogating key factors shaping the conflict

Over the three decades that the conflict has ensued, new science and economic analyses have been produced, yet the production of this new knowledge has not brought resolution to the debate. Rather, we find that in the LSRD conflict, stakeholders selectively use or critique scientific findings and economic expertise to support and justify their preferred status for the dams–removal or maintenance, supporting earlier observations of the conflict (Lackey, 1999). In other words, stakeholders politicize scientific and economic knowledge. As understood here, politicization is when stakeholders rely on the inherent uncertainty of science or economics to cast doubt on the existence of consensus, typically in the pursuit of a particular agenda (Bolsen & Druckman, 2015). A lack of trust in those who produce new knowledge (and therefore in the underlying assumptions) and privileging particular data or methodologies (Hall & White, 2008) are also behaviours that lead to the politicization of the relevant science and economics in this case.

In the case of the science around salmon, the uncertainty associated with it, largely due to chosen methodologies and the multiplicity of ecologic causes and effects, means that scientific studies can be seen as contradictory. Scientific uncertainty, even if minor, is harnessed by stakeholders to support their point of view regarding the dams. As stated by a scientist, some supporters of their findings appreciate the work because of the robust science behind it, but they are also pleased that they get ‘an answer that they like’ (interview 2). One long-time stakeholder in the region, representing a non-profit organization, reflected, ‘I think that there’s enough gap in the science to allow people to come to the conclusion that they already want and that’s for whatever you believe’ (interview 5).

However, conflict around the science arises not just because there are seemingly conflicting scientific conclusions or differing objective interpretations of scientific knowledge. Stakeholders interviewed from the region frequently spoke of a ‘belief’ in certain science related to the LSRD debate, suggesting that stakeholders may be driven to choose science that aligns with their values, rather than being driven to choose a management option based on scientific facts. As perceived by an interviewee leading a non-profit, ‘because people do really want to help salmon, they say ok, well we know it’s the dams, and so they start with what they believe in their hearts and then they try to build a scientific structure around that’ (interview 5). Finding science that supports a preferred world view is common in this conflict, since scientific evidence carries weight and authority in this context (see also Dorning, 2018). Another stakeholder, representing interests supporting navigation, energy, and trade in the Pacific Northwest, shared that, ‘here salmon recovery is, like so many other things in our lives, everybody has their own particular worldview […] and then the science is there to back you up’ (interview 7).

Similarly, CBAs are often met with criticism and are selectively used and critiqued by stakeholders to support a dam management preference. We observe that stakeholders capitalize on the uncertainty in the data, due to differing assumptions about the right future energy mix and current costs of dam management and energy. There is also contestation around the calculations and inclusion of non-use values in the CBA methodologies. Some long-time stakeholders interviewed recognize new shifts in the debate as renewable energy sources, such as wind and solar power, have increased in both affordability and availability, which may ultimately affect CBA calculations (interviews 4 and 8).

The construction and use of diverse, and sometimes opposing, sources of knowledge by stakeholders in conflicts over dam removal has been recognized in other studies in settings across the USA (e.g., Dorning, 2018; Fox et al., 2016; Sneddon et al., 2017b). McKean and Johnson (2019) refer to the phenomenon of ‘post-normal science’ in the LSRD conflict, where science or facts are not certain, nor value-free. Such value-laden knowledge impacts the use of science and the calculation of CBAs. Although stakeholders value independent, ‘good’ science as a key to resolving differences, neutralities seemed to ‘be in the eye of the beholder – some entities are perceived to be neutral sources of good science to some interviewees while distinctly not to others’ (Oregon Solutions & William D. Ruckelshaus Center, 2017, p. 21). While salmon recovery is at the heart of the conflict, how to manage the dams is far from just a scientific or technical decision, given how it is embedded in a complex socio-ecological system.

Despite this contestation, we observe expanding venues for engagement, as stakeholders have been engaging outside of the courts through alliances, task forces, and partnerships. Researchers in other dam removal cases in the Pacific Northwest highlight the importance of collaborative processes and the need for a forum that allows for inclusive negotiations (Gosnell & Kelly, 2010), the creation of which has proven to be important for protecting ecosystems in hydropower relicensing processes (Gowan et al., 2006). In our interviews about the LSRD, we found some stakeholders hopeful that these new venues can help resolve the impasse and increase dialogue about the debate (interviews 3, 9 and 10), while other experts noted a lack of open-mindedness within these forums (interviews 5 and 8). Despite these collaborative processes underway, litigation continues (Miller, 2020). While we do not see evidence that the expansion of these new venues has helped to lessen the conflict yet, it may be too early to tell.

Additionally, the shifting geography (from east to west) and expanding scales (from salmon to orca) we uncover have not served to alter the status quo, but rather, have further contributed to changing the conflict dynamics over the past decade. The forging of the link between declining salmon populations and declining orca populations has both brought additional stakeholders to the debate and broader attention to the conflict (interviews 3, 6 and 10). In a region fraught with strong local politics in the form of East–West tensions, connecting the dams in eastern Washington with orcas in western Washington is seemingly serving to amplify divisions. From the perspective of a federal employee, ‘it’s disingenuous to simplify the orca situation by saying if you breach the Snake River dams you’re going to save the orcas. Honestly, from a scientific standpoint it’s complete BS, but it’s a nice marketing tool’ (interview 9). Yet those supporting dam removal seem pleased that orcas linked to the dams are beloved in places like Seattle, where there is political clout, as this connection has pushed politicians who were historically cautious about the dam removal debate to agree that they are a problem (interview 8). However, a stakeholder representing a non-profit said that such political power is considered by some in eastern Washington as a problem of ‘taxation without representation’ (interview 5). The orca–salmon link that has emerged over the last decade highlights an ‘us’ versus ‘them’ dimension present in the local politics. Local control framing was recognized by Hall and White (2008) as a prominent tool used by stakeholders testifying in Congressional hearings for developing PNW salmon policy in the Pacific Northwest between 1998 and 2000. Fox et al. (2016, p. 102) emphasize that it is important to understand such ‘ordinary environmental politics’ that explain people’s daily engagements with their landscape, and not just the large-scale environmental politics characterized by environmental legislation, social movements, and campaigns.

A framework for approaching understanding dam removal controversies

Our analytical framework – the science–economy–stakeholder nexus – adopted here helps reveal underlying dynamics associated with dam removal conflicts. This framework focuses on three broad factors–the role of science, economic analysis, and stakeholder interactions–drawn from earlier research on dam conflicts, which allow for rich contextualization and analysis. While these factors have been previously recognized as central to such conflicts by other researchers, this case study is the first time they have been collectively analysed in the context of the modern LSRD conflict. Based on our analysis using this framework in the context of the LSRD conflict, we propose that the interactions between factors are equally important to understanding such conflicts. For example, stakeholder actions and interactions may facilitate the politicization of science and economics. Or, if we consider the role of science, while some degree of uncertainty is not unusual, when coupled with a lack of trust of those who produce scientific knowledge, and with the other factors salient to this debate, such as politics, values and worldviews, stakeholders on both sides of the debate harness the scientific uncertainties and contradictions to use science strategically to support their preferences for or against dam removal, fuelling this politicization. Future research should pay close attention to the interactions between these three main categories of factors, and consider how political and cultural dynamics, specific to each dam removal case, influences the meaning of these factors in a conflict.

Based on our research, we came to see how both science and economic calculations were forms of knowledge that were both used in similar ways by stakeholders, often to support values and political beliefs. We also recognize that Native American tribes have ecologic knowledge and values systems that were not considered in this analysis, yet which are central to the debate. Therefore, we can propose combining and broadening the two categories of science and economics in the original framework, under the umbrella of ‘knowledges’. Knowledges, whether biophysical, economic or spiritual in nature, are key factors in dam removal debates, impacted by framing, and political and cultural dynamics. As dam removal, and dam building, accelerates worldwide, this framework can be useful for understanding past and current conflicts, and perhaps for anticipating future issues, before conflicts stretch on for decades (Figure 4).

Figure 4. Revised analytical framework for dam removal conflicts.

Conclusions and pathways forward

We ask: How might the LSRD removal conflict be resolved? In our study of the LSRD conflict, we found several potential pathways to help resolve the conflict. First, several stakeholders spoke to us about the potential and hope that western science can still help shift the debate (interviews 3, 5, 6, 8 and 10). Although stakeholders agree that the science is politicized, they also perceive its important impact on the perceptions of powerful decision-makers (e.g., Congressperson Simpson and Governor Jay Inslee) and on the dialogue surrounding the conflict (e.g., science about the orca–salmon connection has shifted some public opinion in support of dam removal). Stakeholders suggest that it is also critical to consider the under-recognized role of traditional knowledge as a form of science influencing decision-making about the LSRD (Oregon Solutions & William D. Ruckelshaus Center, 2017). Researchers have long emphasized the importance of traditional knowledge in water management decision-making (e.g., Leonard et al., 2013; McGregor, 2012). Second, in terms of economics, most stakeholders we engaged (interviews 1, 4, 5 and 8) agree that the economics need to shift in favour of dam removal for the status quo to be altered. The cost of maintaining the LSRD is expected to rise because of the increasing availability of affordable, clean energy, which may mean BPA’s current customers do not renew their contracts expiring in 2028 (interviews 4 and 8), and because litigation may continue to require ever-more expensive mitigation measures to be adopted at the dams to rehabilitate fish populations. Third, some stakeholders see the new collaborative venues as potential pathways towards conflict resolution if these forums can sufficiently and appropriately provide a place for environmental justice to be realized and for marginalized voices to be heard, particularly those of tribal members (interviews 3, 5 and 7–10). Finally, the majority of interviewed stakeholders see leadership and legislation as key for resolving the conflict (interviews 1, 3, 6 and 8–10). Some see hope with Idaho Republican Congressperson Mike Simpson’s recent US$33.5 billion proposal to remove the four LSRD and provide resources to help the various impacted sectors adapt, especially in light of the environmental justice priorities of the Biden administration (interviews 6, 8 and 10).

Acknowledgements

We thank Robert Varady for leading the Water International mentor programme. We also give special thanks to the Water International editor, Surina Esterhuyse, to Gabriel Eckstein, past President of IWRA, to Yuanyuan Li, President of IWRA, and to two anonymous reviewers. We also are grateful to the individuals who shared their time and expertise during interviews.

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplementary material

Supplementary data for this article can be accessed at https://doi.org/10.1080/02508060.2022.2090147.