The ecobiopolitics of environmental mitigation: Remaking fish habitat through the Savannah Harbor Expansion Project

Abstract

Environmental mitigation has become a catch-all term for efforts to avoid, minimize or compensate for the adverse impacts of development. Through an analysis of the expensive and complex plan developed to mitigate the anticipated impacts of deepening Savannah Harbor, I develop an ecobiopolitical approach to mitigation. Environmental mitigation is triage, involving difficult choices about which entities are worthy of concern and, thus, candidates for intervention – and, by extension, which are not. It involves decisions about which among the chosen deserve strict protection and which merit looser forms of care. As these processes move to center stage in twenty-first-century governance and politics, it has become important to understand what kinds of environments mitigation generates. What survives? What dies? What flourishes? This article focuses on initiatives designed to maintain minimally suitable conditions for non-human life. Insomuch as the object of habitat mitigation is the animal milieu, rather than the body or population, it can be understood as a form of ecobiopolitics. By contrasting the projected fates of three fish in the post-mitigation ecology of the Savannah River, I argue that the ecobiopolitics of habitat mitigation can be conceptualized at four registers. The first, comparity, highlights the value-laden processes through which some entities become candidates for mitigation and others do not. The second, hierarchy, underscores how candidates for mitigation are ranked in ways that shape the interventions pursued. The third, nonfungibility, foregrounds how problems of commensuration are negotiated in mitigation practice. The fourth, overflow, emphasizes how mitigation aimed at one entity can lead to other ecological changes.

Keywords

environmental mitigation commensuration infrastructure political ecology water

The New Savannah River Bluff Lock and Dam near Augusta, Georgia is the future site of a major infrastructure project: a hundred million dollar fish passage designed to allow endangered shortnose sturgeon to access historical spawning habitat that has been inaccessible since the complex was built in the 1930s. Both the impetus and funding for this project came from a much larger infrastructure project underway at the Port of Savannah, some 170 miles down the Savannah River. The Savannah Harbor Expansion Project (SHEP) will improve port access for a new generation of massive oceangoing ships by deepening shipping channels in the harbor from 42 to 47 feet. How are these seemingly distinct projects at distant points along the river connected? What, in other words, does the upriver movement of sturgeon have to do with harbor deepening?

The fish passage is one of many costly mitigation projects administered by the US Army Corps of Engineers (the Corps) to offset the environmental impacts expected to accompany the SHEP. Dredging deeper shipping channels was predicted to increase salinity and reduce dissolved oxygen in the lower Savannah River. Modelers predicted that those changes would, in turn, have significant implications for some plant and animal communities. Among the most consequential was the potential loss of hundreds of acres of sturgeon habitat in the river’s estuary. Before the dredging could proceed, the Corps was, in its role as the implementing agency, required to complete an environmental impact statement and develop a corresponding mitigation plan. This is standard procedure for infrastructure projects in the United States and much of the world. But what made the SHEP remarkable was the complexity and cost of the mitigation. Over half of the total project budget, estimated at nearly one billion dollars, was dedicated to mitigation features: re-routing river flow to minimize the impacts of rising salinity on fish habitat and wetlands; installing an oxygen injection system to maintain a dissolved oxygen level in the river sufficient for aquatic life; purchasing thousands of acres of land to compensate for degraded wetlands; and building an upriver fish passage to replace sturgeon habitat lost in the estuary (Figure 1).

Figure 1.

The Savannah Harbor Expansion Project’s navigation and environmental mitigation measures. Approximately half of the budget was spent on mitigation.

Environmental mitigation has become a catch-all term for efforts to minimize or compensate for the adverse impacts of development, with an emphasis on maintaining ecological resources, services or functions. The term is used to describe responses to a range of environmental concerns defined at different spatial scales, from actions to minimize the local impacts of infrastructure projects to global efforts to reduce the risks associated with climate change by cutting emissions and planting forests. Environmental mitigation scholarship encompasses a range of conversations that are too varied to summarize in this article. What is salient here are the broad orientations that researchers adopt toward mitigation as an approach. For heuristic purposes, I identify three: The techno-reformist orientation is concerned with whether mitigation policies and practices meet their stated goals and how to improve effectiveness (Bull et al., 2013; Fischer and Newell, 2008; McKenney and Kiesecker, 2010). The alternatives orientation emphasizes moving beyond a fundamentally flawed status quo by creating different frameworks (Jenkins, 2016; Moreno-Mateos et al., 2015). The critical orientation, most common in STS, anthropology and geography, interrogates mitigation in its conception and/or practice, often focusing on linkages with processes of accumulation, dispossession and political struggle (Bumpus and Liverman, 2008; Li, 2013; Little, 2012, 2013; Robertson, 2004, 2006). These are all important conversations, but, as mitigation moves to center stage in twenty-first-century governance and politics, an additional set of questions becomes urgent.

What kinds of environments does mitigation generate? What survives, dies and flourishes therein? The SHEP is illustrative. Project documents present its extensive mitigation features – the fish passage, oxygen injection system, river flow re-routing, etc. – as a set of expensive and technically complex interventions designed to neutralize the most significant environmental impacts associated with deepening the harbor to improve port access. Framed in this way, the key questions seem obvious: Will these efforts prove effective? Can they offset the damage? In this article, I approach mitigation somewhat differently, conceptualizing it as a parallel infrastructure project – an additional layer of engineering that alters environments and ecological relationships in ways that reflect particular rationalities, priorities and assumptions.

Environmental mitigation is triage. In a project like the SHEP, state bureaucrats, scientists and project personnel develop plans to minimize or offset impacts within significant institutional, financial and time constraints. This messy but critically important process involves difficult choices about which species, landscapes and other entities are worthy of concern and, thus, candidates for mitigation – and, by extension, which are not. It involves decisions about which among the chosen deserve strict protection and which merit looser forms of care. Because mitigation involves decisions – some conscious, others unrecognized – about making live and letting die (Rabinow and Rose, 2006), I argue that it can be understood as a form of ecobiopolitics (Olson, 2010).

Biopolitics (Foucault, 2008) can be glossed as a modern political rationality that takes the health and well-being of the human population as an object of government. This is a productive entry point for understanding how existing techniques, institutional structures and technologies are recombined in ways specific to a time, a society, a sector (Collier, 2009) – or, perhaps, a fish. If biopolitics is always mutating (Rabinow and Rose, 2006: 215), environmental mitigation can be associated with its ‘ecological’ mutations in the 1960s and 1970s. This period saw a shift in concern from the ‘biological foundations of politics’ to ‘life processes’ (Lemke, 2011: 23). In the United States, landmark environmental laws like the National Environmental Protection Act of 1969 and Endangered Species Act of 1973 required government agencies to determine the potential impacts of their activities on living beings and their habitats.

As Olson (2010) emphasizes, the ascent of ecobiopolitics marks a shift in which milieus, rather than bodies and populations, become the object of truth claims and power relations. The ecobiopolitics concept has been used to explore the politics of environmental health in spaces constructed for and inhabited by humans (Saxton, 2015; Singer and Hodge, 2016), but it is also useful for analysing the managerial rationalities and practices aimed at maintaining habitable space for non-human life. In animal and multispecies studies, biopolitics has often been used to examine how disciplinary power works on individual animal bodies or through management at the scale of the breed or species (Lorimer, 2017: 35). Adopting such an approach would obscure how the US environmental mitigation bureaucracy operates with regard to fish. The SHEP mitigation plan, for example, was not focused on managing the sturgeon population, per se, but maintaining its milieu, quantified as the acreage of ‘acceptable’ or ‘suitable’ sturgeon habitat in the river.

Life support may be the critical project of the Anthropocene.¹ Seen in this light, habitat mitigation is emblematic of a broader class of interventions designed to maintain minimally suitable conditions for existence. Drawing on project documents and interviews with actors involved in the SHEP mitigation process, I contrast the projected fates of three estuary-dependent fish – shortnose sturgeon, striped bass, and Southern flounder – in a deeper, post-mitigation Savannah River. As we will see, the sturgeon is expected to persist in the wild after mitigation; the bass to become dependent on off-site hatcheries; and the flounder to flourish in a changed estuary. Through an analysis of the SHEP case, I argue that the ecobiopolitics of habitat mitigation can be conceptualized at four registers. The first, comparity, highlights the value-laden processes through which some entities become candidates for mitigation and others do not. The second, hierarchy, underscores how candidates for mitigation are ranked in ways that shape the interventions pursued. The third, nonfungibility, foregrounds how problems of commensuration are negotiated in mitigation practice. The fourth, overflow, emphasizes how mitigation measures aimed at one entity – say, preserving the habitat of a particular fish – can lead to additional ecological changes. In order to make sense of this story of mitigation in situ, we must first understand the rationality that underpins it.

Compensatory mitigation, commensuration and nonfungibility

Mitigation is codified differently across national legal frameworks (Rundcrantz and Skarback, 2003), but a shared ethos animates many governmental and market-based efforts. The ‘mitigation sequence’ or ‘mitigation hierarchy’ (McKenney and Kiesecker, 2010) is an ethical and procedural standard in US, European and Australian law advocating a three-step process: first, seek to avoid adverse environmental impacts by seeking alternative sites or solutions; second, design and execute projects to minimize adverse impacts; third, in cases where impacts are ‘unavoidable’ and minimization is inadequate, compensate. Compensatory mitigation refers to efforts to replace lost environmental functions, services or habitats through acts of restoration, preservation and offsetting. A last resort in theory, it is common in environmental management practice. The construction of a fish passage to offset the impacts of harbor dredging may seem strange, but it is actually an instructive example of how contemporary mitigation can move from minimizing adverse impacts to producing additional layers of environmental engineering.

Compensatory mitigation is a paradigmatic modern response to environmental damage. Operating through the mechanism of exchange, it is built on the premise that it is possible to carry out projects with significant impacts without a decline in overall environmental quality. Emblematic of the compensatory rationality is ‘no net loss’, a concept and policy goal based on the premise that impacts can be offset, or balanced against equivalent gains. This framework has been used to set mitigation targets for carbon stocks, biodiversity, ecosystem services, water quality and animal habitat (Bull et al., 2013; Maron et al., 2018; Robertson, 2000; Salzman and Ruhl, 2006). ‘No net loss’ has an obvious rhetorical appeal because it seems to balance economic development and environmental protection goals. If we accept its central premise, then a project like the SHEP, which is predicted to have significant environmental impacts, can be completed, produce the expected impacts and, through mitigation accounting, be presented as environmentally neutral. You can have your cake and eat it, too. Or can you?

With its focus on exchangeable units – of emissions, wetlands, habitats and so on – compensatory mitigation is a form of commensuration: It depends on a common metric to compare and establish equivalencies across different types of entities (Espeland and Stevens, 1998: 315). Conceptualized in such terms, compensatory mitigation is akin to other modern theories of value (Grossberg, 2010). Just as modern financial systems depend on money as a tool of commensuration, abstraction and quantification (Maurer, 2006), mitigation works through its own exchangeable units, or currencies, such as: tons of emissions, acres of wetlands and, the focus of this article, acres of animal habitat (Salzman and Ruhl, 2000). The circulation of mitigation currencies depends on the establishment and stabilization of a metrological regime (Barry, 2002) – a formal apparatus for measurement and calculation – that can standardize and legitimize exchanges (Cooper, 2015).

The US environmental mitigation regime has developed over the past century. For fisheries, some activities now described as mitigation date to the early twentieth century (Taylor, 1999: 222–230). The Federal Power Act of 1920 laid out mitigation responsibilities for impacts to fish and wildlife habitat associated with infrastructure projects. The Mitchell Act, passed in 1938, instructed the Corps of Engineers and Bureau of Reclamation to work with fisheries agencies to reduce the impacts of dams through various technological fixes that would not impede river development, such as fish passages, irrigation screens, fish relocation and hatchery production, particularly for salmon management in the Pacific Northwest. The emphasis on defining and modeling minimal thresholds of environmental suitability – on full display in the SHEP – became prominent during the postwar decades, particularly through the Clean Air Act and Clean Water Act, which emphasized ‘acceptable levels’ of pollution (Liboiron, 2016: 8-9). As we will see, the impacts of the SHEP were anticipated, presented and contested through environmental models that predicted changes in water levels, dissolved oxygen and salinity. Those parameters were, in turn, used to predict impacts to key ‘environmental resources’ like changes in acreage of ‘suitable’ fish habitat.

If the underlying premise of compensatory mitigation is that units of carbon, pollution, wetlands or habitat can be similar enough that exchanging one unit (the loss) for another (the replacement) allows projects with recognized impacts to be characterized as environmentally neutral, then it becomes important to ask: How similar is similar enough? How different is too different? How would we go about making such a distinction? Legal scholars Salzman and Ruhl (2000) call this problem ‘currency adequacy’, a concept that focuses attention on which social or ecological values a given mitigation currency captures and the degree to which those values are aligned with larger environmental goals. Because many objects of environmental concern occupy a position between the universal and the place-specific (Choy, 2011), compensatory mitigation introduces problems of nonfungibility; some costs and benefits will be externalized in any exchange. Seen in this way, compensation is not simply an exchange of one entity for another, but a mechanism through which ecological value is framed and operationalized. Which – and whose – values are used to determine adequacy within a given mitigation regime? How are those values ranked? What can be externalized in a compensatory exchange? What cannot? To raise such questions around the construction of fungibility in environmental mitigation is not simply a critique. These are entry points for analysing the ecobiopolitics of mitigation policy and practice.

Carbon by the ton, fish by the acre: Does matter matter in environmental mitigation?

Social scientists have demonstrated that the material specificities of resources – water, timber, soil, fish, oil and so on – can enable, constrain and disrupt how political-economic processes unfold (Bakker and Bridge, 2006; Richardson and Weszkalnys, 2014). Just as the physical characteristics of water make it difficult to privatize (Bakker, 2004) and the heterogeneity of Douglas Fir forests troubles industrial processes (Prudham, 2005), the objects of mitigation concern vary in the degree to which they ‘cooperate’ with compensatory exchanges. The claim that compensatory mitigation does what it is supposed to do – that a loss can be offset through replacement or substitution – are strongest when the resource or ecosystem service of concern can be articulated is terms of an inherent property; this reductive potential makes it more conducive to alienation and exchange.

One object, carbon – or carbon-dioxide equivalent – has played an outsized role in mitigation theory, policy and practice due to its association with climate change. Carbon, like water (Helmreich, 2011), is a particular type of theory machine. Thinking through its properties, which are distinct from those of other objects of concern like wetlands and habitats, can lead to new insights and new reifications. Scholars have rightly critiqued ‘elemental reductionism’ in climate change research (Bridge, 2010: 821), particularly in work on carbon markets (Dalsgaard, 2013; Goodman and Boyd, 2011; Lohmann, 2005). Indeed, carbon is ‘a fictitious molecular category through which several different greenhouse gases are made commensurable’ (Bridge, 2010: 822). STS scholarship has shown how carbon, which assumes myriad forms, is stabilized as a sociotechnical object so that it can be measured, assigned market value and exchanged as a commodity (Bumpus and Liverman, 2008; Cooper, 2015; Lohmann, 2005; Lovell and Mackenzie, 2011; MacKenzie, 2009). In practice, carbon is multiple (Günel, 2016), and materializes in many ways (Bumpus, 2011; Lovell and Liverman, 2010). And yet, as a fictitious elemental category (Bridge, 2010), its reductivity – and, by extension, its alienability – make it more cooperative with compensatory mitigation exchanges than, say, freshwater wetlands or fish habitat. One way to conceptualize this distinction is to focus on the problems of nonfungibility that a given exchange enacted through some currency presents in terms of larger environmental goals (Salzman and Ruhl, 2000).

Nonfungibilities of space are qualitative differences among entities related to location and the scale of environmental management goals. The contrast between carbon offsets and animal habitat exchanges is instructive here. Insomuch as the problem of climate change is defined at the scale of the Earth’s atmosphere, it follows that the mitigation – carbon offsets – might be abstracted from the specificities of place and ‘exchanged’ by the ton across space. In this respect, planting a forest on one side of the planet might reasonably mitigate emissions on the other. However, management goals related to biodiversity, wetlands and animal habitat tend to be specific to places and ecological relationships, and, thus, less conducive to claims of substitutability at a distance. As Robertson (2004: 368) points out, ‘any definition of an ecosystem commodity expressed as a function of ecosystem dynamics will carry with it an implicit argument about the spatial limits of its commensurability’. To this point, US federal agencies prefer that mitigation for impacts to aquatic resources be ‘on-site’, meaning compensation within or adjacent to the project space. ‘Off-site’, by contrast, means replacement at a distance. In the case of shortnose sturgeon, however, compensatory mitigation is not only complicated by the spatial limits of commensurability, but by the fish’s biology, life course and tendency to move around the river according to season and life stage in ways that scientists do not fully understand. Crucially, there is a tension between the mobility of the object of mitigation concern, the sturgeon population, and the spatially fixed currency: habitat acres.

Nonfungibilities of type are another important consideration in compensatory mitigation. Here, too, entities have properties and occupy relationships that affect their interchangeability. Salzman and Ruhl (2000) observe that distinctions between tradeable allowances of pollutants – tons of sulfur, nitrogen or hydrocarbons – are relatively straightforward, ‘but as one moves into habitat trading, clear delineations of type begin to blur as the units become increasingly nonfungible. Habitats are inevitably heterogeneous, both in biophysical terms (their soil, flora and fauna, hydrology, climate) and as a result of the services they provide. Yet a simple currency of acres will never capture these differences’ (p. 629). Nonfungibilities of type are recognized in US environmental policy and federal agency practice, where ‘in-kind’ mitigation refers to the replacement of an affected unit with another of the same type (e.g., salt marsh with salt marsh) and ‘out-of-kind’ mitigation means replacement with a different type (e.g., salt marsh with freshwater marsh). In addition to on-site mitigation, US federal agencies prefer in-kind exchanges when offsetting the unavoidable impacts of dredging. As I explain below, the construction of a fish passage on the Savannah River is out-of-kind mitigation because it represents the exchange of two different types of sturgeon habitat.

It may not surprise readers of this journal that compensatory mitigation involves commensuration and introduces problems of nonfungibility, nor that the resulting exchanges can be messy in practice. But a focus on ecobiopolitics foregrounds something more. On paper, the exchange of, say, fish habitat acres makes it possible to claim ‘no net loss’ – that the anticipated impacts of an infrastructure project have been neutralized. But, in practice, this kind of exchange can transform the ecosystem that it was designed to protect, sometimes in unexpected ways. In this sense, as Mathews and Barnes (2016: 23) observe, ‘Matter becomes insistently political in environmental and natural resource futures, and we cannot, in advance, tell what kinds of politics will ensue’. With that in mind, let’s return to the Savannah Harbor Expansion Project and the work of mitigating its impacts.

The Savannah Harbor Expansion Project

The story of the fish passage at the New Savannah River Bluff Lock and Dam does not begin in Augusta, Georgia, nor even downriver at the Port of Savannah. It begins in Panama. In 2007, Panamanian voters approved the first major expansion of the Panama Canal since it opened in 1914. Over the following decade, the Canal was expanded to accommodate a new generation of megaships through the construction of a flight of larger locks and the widening and deepening of shipping channels. Even before its completion in 2016, the Canal expansion reverberated through the shipping industry and coastal communities. Port authorities, state and local governments, and firms associated with nearly every major port on the US Atlantic and Gulf coasts used the expansion to make a case for deepening their harbors to approach the new 50-foot depth of the Canal in hopes of attracting more Neo-Panamax ship traffic coming from Asia (Figure 2).

Figure 2.

In July 2016, two weeks after the opening of the Panama Canal’s expanded locks, the first Neo-Panamax vessel, the MOL Benefactor, arrived at the Port of Savannah. It was, at that time, the largest vessel to ever call on the port. The Savannah Harbor Expansion Project was intended to expand the port’s capacity to accommodate ships of this size and larger.

The Port of Savannah, operated by the Georgia Ports Authority, is located on the city of Savannah’s industrial waterfront – some 18 miles up the Savannah River from the ocean (Figure 3). Its neighbors include a paper pulp mill, sugar refinery and a liquified natural gas terminal. As the center of one of the largest logistics clusters in the United States (Rodrigue, 2020), the port has surpassed other regional industries in economic importance. It is now the country’s fourth-busiest container port and the second-busiest on the Atlantic coast (USDOT Bureau of Transportation Statistics, 2020) providing ‘access to 44% of US consumers in 2–3 days’ (Georgia Ports Authority, 2016). However, ship access to the port has long been limited by the physical geography of the Savannah River. In the eighteenth century, the shipping channel linking the port with the ocean was less than ten feet deep, and often shallower depending on river levels, shoaling and other obstructions. Mechanical dredging began in the 1790s and accelerated in the 1820s, when the Savannah District of the Corps was established and the federal government began to appropriate funds for deepening and widening the shipping channel (Barber and Gann, 1989: 29). In the twentieth century, Savannah Harbor was deepened half a dozen more times, reaching 42 feet in 1994 – some 30 feet deeper than when the city was established (Granger, 1968; Ramos, 2017). This was its navigable depth when the SHEP began.

Figure 3.

Port of Savannah terminals and Savannah River ship channel approach.

The SHEP, funded by the US federal government and Georgia Ports Authority, will deepen shipping channels from 42 to 47 feet. The project began in the late 1990s, when the Georgia Ports Authority initiated studies to deepen the harbor again to accommodate larger ships in order to increase port efficiency and competitiveness. The project received federal authorization in 1999, conditional upon the approval of an environmental impact statement and mitigation plan by the Army (through the Corps) and several resource agencies: the US Fish and Wildlife Service, National Marine Fisheries Service and Environmental Protection Agency. Inner harbor dredging did not begin until 2019. In the interim, a full two decades, the Panama Canal was expanded (2007–2016) and the port authority’s existing goals of increasing efficiency and competitiveness through harbor deepening were linked to attracting the Neo-Panamax ship traffic that would transit the larger canal (USACE, 2012b). The long project duration was an outcome, in large part, of a complex and contentious environmental impact assessment process associated with the project’s geographical and ecological setting. Because Savannah is a river port (Figure 3), harbor deepening meant dredging a long stretch of estuary (the inner harbor), raising concerns about impacts on wetlands, wildlife and urban infrastructure. Ultimately, the project was slowed by the required federal agency approvals and a lawsuit and settlement (USACE-USEPA, 2013).

Navigation dredging – excavating underwater sediment from harbors and shipping channels – is a precondition for twenty-first century transportation and, by extension, the logistics-led development strategy adopted in Savannah. To accommodate Neo-Panamax vessels and other megaships, ports, harbors and rivers are dredged to match shipping standards that are pegged to maritime chokepoints like the Panama Canal, Suez Canal and the Strait of Malacca. This is a key example of the environmental standardization efforts associated with global transportation infrastructure (Carse and Lewis, 2017). In freshwater and estuarine systems, dredging can increase salinity levels and lower dissolved oxygen levels, changes with potentially significant implications for plant and animal communities (Bray, 2008). Displaced sediment can increase water turbidity and, potentially, toxicity, if it is contaminated. Dredging can affect coral reefs, seagrasses and mammals (Erftemeijer et al., 2012; Erftemeijer and Robin Lewis, 2006; Todd et al., 2015), and even municipal water systems.

In each project, the Corps, the federal agency responsible for the development and maintenance of navigable waterways, is subject to federal and state laws and regulations that require it to mitigate anticipated environmental impacts. Among the most important laws affecting navigation dredging in the United States is the Clean Water Act, specifically Section 404, which establishes mitigation requirements for impacts to aquatic resources. In 1990, the Corps and Environmental Protection Agency wrote a memorandum of agreement establishing a three-part ‘mitigation sequence’: 1) avoidance, or selecting the ‘least-damaging’ project type and location compatible with the project purpose; 2) minimization, managing the severity of a project impact on resources; and 3) compensatory mitigation, or offsetting impact through the replacement or substitution of resources (via restoration, enhancement or preservation) when impacts are unavoidable and minimization is inadequate (USEPA-USACE, 1990).

In 1999, Georgia Ports Authority established the Stakeholder Evaluation Group, which included representatives from government, business and the public. Created in response to project opposition, the group was to develop a consensus on environmental impacts and mitigation within a context of scientific uncertainty (Covington, 2015). In practice, however, facilitators treated members of government agencies, scientists and members of the lay public differently, ‘excluding less knowledgeable stakeholders from the technical deliberations … and disregarding consensus decisions’ (Wills-Toker, 2004: 198). Georgia Ports ultimately adopted a decision-making approach that treated stakeholder input from the group as recommendations, rather than authoritative (Wills, 2001). Indeed, the most influential deliberations about environmental impacts and mitigation did not directly involve the public. The Corps organized an interagency coordination team for each of five key resources – wetlands, water quality, sediment placement, groundwater and fisheries – that included scientists and representatives of state and federal agencies (USACE, 2012b: 245–246). It was through this structure that impacts on shortnose sturgeon habitat emerged as one of the SHEP’s major mitigation concerns.

Where (and when) is sturgeon habitat?

Sturgeon are sometimes called living fossils. They have existed for around 100 million years in their modern form (Carey, 2005: 3, 4), but their family, Acipenseridae, is even older, dating back 200 to 400 million years – likely preceding the first dinosaurs. The International Union for Conservation of Nature (IUCN, 2010) has described sturgeon as ‘more critically endangered than any group of species’. Populations worldwide have been devastated by overharvesting for their flesh and eggs – caviar, a desired, expensive and often illicit commodity – but the principal threat is the loss and degradation of habitat (IUCN, 2020). The three species found in eastern North America – Gulf sturgeon, Atlantic sturgeon and shortnose sturgeon – are all protected by the US Endangered Species Act. Both Atlantic and shortnose sturgeon are found in the Savannah River.

Most sturgeon, including Atlantic sturgeon, are anadromous. Like salmon or striped bass, they spend much of their lives in the ocean and migrate into rivers to spawn. Shortnose sturgeon (Figure 4) are amphidromous, meaning they live in their birth (or natal) river, with potential migratory trips into the ocean. They migrate extensively within the river according to the season and their life stage, including movements to escape extremely warm or cool temperatures and to spawn in headwaters. Shortnose sturgeon inhabit rivers along the Atlantic coast, from southern New Brunswick, Canada to northern Florida (Shortnose Sturgeon Status Review Team, 2010). In the twentieth century, shortnose populations were devastated by overharvesting, bycatch, dams and pollution. In the United States, the species was listed as endangered and afforded protection in 1967 through the Endangered Species Preservation Act. A US Department of Interior publication from the era stated that the species was ‘in peril … gone in most of the rivers of its former range [but] probably not as yet extinct’ (USDOI, 1973: 6). The Savannah River population – now estimated to be between 1000 and 3000 – has been historically affected by all of the above threats. But in the Savannah, as in many other rivers, the principal threat to the survival of the shortnose sturgeon is habitat degradation and loss caused by upriver dams, water pollution and dredging (NMFS, 1998).

Figure 4.

Shortnose sturgeon.

Army Corps administrators convened the first meeting of the SHEP interagency coordination team for fisheries in 1999. The team, which met throughout the decade that followed, was comprised of agency professionals from the Corps and Georgia Ports (the implementing organizations), staff and scientists from federal and state resource agencies and a few representatives from conservation organizations (USACE, 2012a: Appendix N, 204). At the outset, they worked to determine which fishery resources should be the focus of a future environmental impact assessment. Participants were asked to identify the species and habitats of greatest concern, where – geographically – project impacts could be expected and at what thresholds the changes to those resources would become unacceptable. A wide range of marine and freshwater species were initially discussed, but a group of estuary-dependent fish, including shortnose sturgeon, ultimately emerged as the greatest concern for the interagency coordination team.

Shortnose sturgeon became emblematic of the SHEP’s anticipated impacts on the Savannah River because they inhabit a niche expected to be significantly impacted by the project and are protected under the Endangered Species Act. The smallest of the sturgeon species in eastern North America, they spend most of their lives in the slow-moving waters of estuaries, particularly the interface between saltwater and freshwater. Deepening the shipping channel was expected to increase salinity levels in the estuary, pushing this interface upriver and eliminating habitat.² Shortnose sturgeon are bottom feeders that eat mollusks, crustaceans and insects. Because a deeper channel might also mean lower dissolved oxygen levels at the bottom of the water column, even more critical habitat might be eliminated, particularly for young (juvenile) fish that are more sensitive to high salinity and low oxygen than adult fish.

While discussing the SHEP mitigation process with someone who participated in the interagency coordination team, I referred, in passing, to sturgeon as the ‘most important species’ for mitigation. He immediately corrected me: ‘That’s not exactly right.’ The focus of fisheries impact assessment and mitigation for a project like the SHEP, he emphasized, was not the population of a species, per se, but modeling how predicted changes in the river’s velocity, dissolved oxygen and salinity would affect that fish’s suitable habitat area. This was estimated for each species of concern by merging a hydrodynamic model (some called it the ‘hydro model’ for short) with a species-specific Habitat Suitability Index (USACE, 2012a: Appendix N, 220). The index represents the quality of a given unit of habitat using a number between 0.0 (unsuitable) to 1.0 (optimum). With this numerical input, modelers could estimate change in suitable habitat area for each species of concern for multiple ‘alternative’ versions of the proposed project – in this case, dredging to a range of different channel depths. When the models showed that an acre of habitat defined as suitable before the project became unsuitable afterwards, then it required mitigation.³

In 2002, the Corps filed a notice of intent with the Environmental Protection Agency to prepare a draft environmental impact statement for deepening Savannah Harbor. Congress had conditionally authorized a channel depth of up to 48 feet but required an incremental analysis of alternative channel depths between 42 and 48 feet. The hydro model and habitat suitability model were coupled in order to predict changes to acceptable habitat for four ‘critical and representative estuarine fish species’ – striped bass, shortnose sturgeon, Southern flounder and American shad (USACE, 2012a: 5.83–5.84) – at each depth under consideration: 42 feet (no change), 43 feet, 44 feet, 45 feet, 46 feet, 47 feet and 48 feet. Notably, most species within the fisheries purview were not candidates for detailed impact assessment and analysis.

Recall, here, that the SHEP’s congressional authorization required approval by several federal resource agencies. This put it on a trajectory toward increased mitigation. Although US federal and state resource agencies typically participate in mitigation processes through the mechanism of biological opinions, these are considered ‘consultations’ and the agency’s approval is not required. So, in a case where a dredging project might have impacts on endangered fish habitat, the implementing agency – in this case, the Corps – would approach the relevant federal agency – in this case, the National Marine Fisheries Service – for a formal consultation on a proposed environmental impact statement and mitigation plan. The agency would then assemble a group to work on a biological opinion and task it with describing the action, determining what adverse impacts it might have, and assessing the sufficiency of the mitigation plan. To do this, agency staff draw on previous consultations as precedents, focusing on the type of action (harbor deepening), species (shortnose sturgeon) and proposed mitigation. One fisheries agency professional told me this is relatively straightforward when there is good data on the status of a population, but when it does not exist (which is ‘fairly often’, she said) staff use the best available science, erring on the side of caution for the species.

In order to determine how the deepening of Savannah Harbor would affect fisheries, environmental modelers and biologists collaborated to determine how predicted changes in dissolved oxygen, salinity, velocity and temperature at channel depths between 42 and 48 feet would affect habitat. This process raised a number of place- and species-specific questions. Where are the critical habitats of shortnose sturgeon in the river? How and when are those habitats used by different age groups within the population (larvae, juveniles, adults and spawning adults)? How tolerant are fish of various ages to the environmental changes predicted to accompany dredging? Which age groups would be most affected and what would that mean at a population level? However, despite the longue durée of human-sturgeon relations, scientists still have much to learn about the family Acipenseridae and disagree about some of what they do know (Jarić et al., 2017).

I asked a marine biologist who had spent decades studying sturgeon what motivated him. What, for him, made them interesting? He paused, then said, ‘Pick any question and the answer is still relatively unknown.’ Among the biggest unknowns with regard to the sturgeon found in the Savannah River is this: Where, exactly, are the fish at different life stages (as larvae, juveniles and adults) and different seasons of the year? Or, to put it another way: Where and when is critical sturgeon habitat? In a sense, the biologist explained, the entire river is critical habitat. Juveniles like to ‘hang out’ in deep estuarine channels with low salinity – further from the ocean. Because adults can withstand saltier water, they can move further downriver. When they spawn, adults migrate to fast-moving headwaters. As this suggests, shortnose sturgeon may occupy multiple habitats up and down the Savannah River depending on the season and their life stage (Collins et al., 2000). Scientists lack the fine-grained, site-specific understanding of the fish’s life history that would be ideal for discerning the impacts of deepening in a particular part of the river.⁴ The gaps in knowledge about the fish’s milieu(s) posed challenges for habitat modeling and mitigation.

After a decade of work and rounds of comment by agencies and the public, the Corps published two key documents in 2012 – a Final General Re-Evaluation Report and Environmental Impact Statement. These documents stated that the harbor would be deepened to 47 feet, a decision that was principally rationalized in terms of maximizing economic benefits rather than avoiding or minimizing environmental impacts (USACE, 2012b: v–vi). The final environmental impact statement concluded that dredging the channel to that depth would create a saltier, less oxygenated Savannah River estuary, with significant impacts on fisheries, freshwater wetlands and water quality. It highlighted ‘unavoidable impacts to significant resources such as loss of Shortnose sturgeon and Striped bass habitat’ (USACE, 2012a: vi). This meant that the Corps was required to compensate for lost environmental functions in the manner codified in US law.

The result was a mitigation plan with an array of measures designed to minimize and offset predicted impacts (USACE, 2012b). The first was removing a tide gate and blocking off cuts in the estuary to ‘re-route’ more fresh water into the ecologically sensitive back river (Figure 1). This was mitigation for saltwater intrusion. The second was constructing an untested hundred million dollar dissolved oxygen injection system (known colloquially as ‘bubblers’) designed to raise those levels to minimum standards set by the Clean Water Act. This was mitigation for reduced dissolved oxygen levels. (Conservation organizations argued that this would amount to putting the river on ‘life support’ because the system would have to operate in perpetuity to maintain the minimum legally acceptable water quality levels.) But modelers concluded that, even with those interventions, dredging to 47 feet would lead to an ‘unacceptable’ loss of critical habitat for sturgeon and striped bass. And, crucially, administrators could not identify any more fish habitat mitigation options in the estuary – that is, appropriate on-site measures – beyond the river flow re-routing and oxygen injection already included in the models. This presented a difficult situation. What else could be done, given that impacts to fisheries were ‘unavoidable’ and the first round of mitigation measures were deemed insufficient?

The fates of three fish, or the ecobiopolitics of habitat mitigation

What kinds of environments does mitigation generate? What survives, dies and flourishes therein? The ecobiopolitics of habitat mitigation can be conceptualized at four registers. Imagine mitigation bureaucracy and practice as a grid overlaying the space of concern. One register of ecobioplitics, comparity, focuses attention on the construction of the boundaries of that grid. Which species, landscapes or environmental entities among many possibilities become candidates for mitigation in the first place? Which do not? Why? How is candidacy determined? A second register, hierarchy, relates to the topography of the grid. How are the various candidates for intervention valued in comparative legal, political or cultural terms? How does this influence the measures taken to minimize or offset predicted impacts? A third register, nonfungibility, foregrounds how problems of commensuration are negotiated in compensatory mitigation. How close is close enough to make a given act of replacement or substitution adequate? How loose is too loose? A fourth register of ecobiopolitics, overflow, underscores how mitigation interventions aimed at a particular entity can precipitate additional processes of ecological change.

Consider, as an illustration, the prospective fates of three fish – shortnose sturgeon, striped bass and Southern flounder – in the post-mitigation Savannah River according to the SHEP Final Environmental Impact Statement (USACE, 2012a). All three were identified as critical and representative species with habitats that might be impacted. For each, agency bureaucrats, scientists and modelers worked to marry habitat suitability criteria to the hydrodynamic to predict change in acres of suitable habitat at various channel depths. The impact analysis did not focus on the habitat of the species in general, but on groups categorized by season (summer versus winter) and life stage (larvae, juvenile, adult, spawning, etc.). Table 1 presents predicted changes in acceptable habitat area at 47 feet of channel depth, with and without mitigation.

Table 1.

Predicted changes in acceptable habitat area due to the SHEP with the channel dredged to 47 feet (the selected plan), by species, season and life stage (USACE, 2012a).

Predicted change in acceptable habitat due to the SHEP at 47-foot channel depth	Without mitigation	With mitigation*
Shortnose adult (Jan)	−0.8% (−32 acres)	−6.9% (−266 acres)
Shortnose adult (Aug)	−13.3% (−185 acres)	+6.5% (+89 acres)
Shortnose juvenile (Jan)	−19% (−328 acres)	−7.6% (−251 acres)
Striped bass spawning	−18.1% (−188 acres)	−13.5% (−140 acres)
Striped bass larvae	−13.8% (−78 acres)	−5.0% (−28 acres)
Southern flounder	−7.8% (−146 acres)	+57.3% (+1072 acres)

Source: USACE (2012a: 5.85–5.86)

Models with mitigation included proposed river flow re-routing and oxygen injection system.

The vast majority of fisheries ‘resources’ in the Savannah River were neither candidates for careful impact assessment nor, by extension, for mitigation. As Schinkel (2016) observes, questions of commensurability and incommensurability are comparative statements that follow from a prior evaluation that he calls comparity. He argues that we can only compare and differentiate entities once they are rendered as ‘objects for comparative scrutiny’ by being placed alongside one another (pp. 366–367). The SHEP fisheries interagency coordination team discussed red drum, red snapper, white and brown shrimp, golden and blue crabs, scallops, billfish, swordfish, tuna, sharks, Atlantic sturgeon, spotted seatrout, Atlantic silversides, and others (USACE, 2012a: Appendix N). Among many candidates considered, only four estuarine species – shortnose sturgeon, striped bass, Southern flounder and American shad – were identified as sufficiently ‘critical’ and ‘representative’ to merit detailed impact modeling and analysis. As a form of ecobiopolitics, then, habitat mitigation directs attention and resources to protecting a circumscribed set of milieus (Olson, 2010: 181), effectively externalizing others.

The interagency coordination team was tasked with providing modelers with habitat suitability inputs for the sturgeon, bass, flounder and shad, including the minimum thresholds, or ‘tolerances’, of each species in terms of key model parameters: salinity, dissolved oxygen and temperature. While all four species depend on estuarine habitat, they also migrate up and down rivers and into the ocean across life stages and seasons. With shortnose sturgeon, for example, adults can ‘tolerate’ higher salinity levels and lower dissolved oxygen levels. Given this, SHEP modelers wanted to identify minimum thresholds for suitability, or tolerances, for each category. The modelers had predictions about how dissolved oxygen, salinity and temperature would change if the channel were dredged to various depths and asked scientists to predict the tolerances of juvenile sturgeon – expected to be the most impacted age group due to their lower salinity tolerance and the fact that nursery habitat was in the vicinity of proposed dredging activities. A similar process was implemented for bass, shad and flounder, but sturgeon were prioritized.

Hierarchy is a second register of ecobiopolitics evident in the SHEP. Mitigation elevates some elements or life forms – shortnose sturgeon and freshwater tidal wetlands in this project – above others and seeks to secure them. Those entities that are candidates for mitigation are not treated as equals. Consider the prospective fate of the striped bass (Figure 5). Even with mitigation, suitable bass spawning habitat was predicted to decrease by 140 acres and larvae habitat by 28 acres (Table 1). Given that federal agencies could not identify additional compensatory mitigation options, they concluded that the only replacement for lost bass habitat was funding the expansion of an existing Georgia Department of Natural Resources stocking program by $125,000 per year. But, because the striped bass habitat wasn’t expected to recover, the Corps proposed a one-time payment of several million dollars. This was monetary compensation for impacts deemed unavoidable and unmitigable. And yet, replacing lost habitat (in acres) with stocked bass (in bodies) as if they were fungible amounted to a shift in mitigation type or ‘kind’ that effectively displaced spawning and larvae habitat from the river to off-site hatcheries.⁵ Thus, bass would become dependent on institutionalized life support, a kind of palliative care (Errington and Gewertz, 2018).

Figure 5.

Striped bass.

Due to the legal status of the shortnose sturgeon as an endangered species, the mitigation process unfolded differently. Under Section 7 of the Act, federal agencies are required to ensure that any action they authorize, fund or carry out is not likely to jeopardize the existence of an endangered species due to adverse habitat modification. If the habitat is expected to be destroyed or modified, it is the role of the biological opinion to identify ‘reasonable and prudent alternatives’ that mitigate threats to the species (Lakoff, 2016). By law, adverse impacts to endangered fish habitat cannot be offset through a stocking program like the one used for striped bass mitigation. As Lakoff (2016: 250) writes, ‘It is in the details of these alternatives – which function as regulations – that the goal of species preservation is integrated into the techno-administrative procedures of resource management.’ Thus, the comparative legal statuses of the fish had material infrastructural implications.

This brings us back to the puzzle presented in the article’s opening paragraph: Why build a hundred million dollar fish passage in Augusta, Georgia as part of a harbor deepening project 170 miles downriver in Savannah? According to the models, the mitigation measures designed to minimize impacts to sturgeon habitat in the ‘action area’ of the estuary (re-routing river flow and oxygen injection) would be inadequate if the channel were dredged to 47 feet deep (Table 1). This meant that the Corps was required to carry out additional compensatory mitigation, per the three-step mitigation sequence encoded in the regulations for this class of project (USEPA-USACE, 1990). But what? And where? The Corps asked agencies to think more expansively about habitable space for sturgeon, given the lack of options in the estuary. What were the limiting factors? The New Savannah River Bluff Lock and Dam in Augusta was one. Built in the 1930s, when the river was still used for inland commercial navigation, it restricted sturgeon access to spawning areas further upriver. Thus, the construction of a new fish passage would, in theory, add shortnose sturgeon habitat to the river by allowing the fish to access historical spawning grounds. As compensatory mitigation, it might serve as a replacement for the loss of habitat in the estuary.

The proposed sturgeon habitat exchange illustrates how problems of nonfungibility can be negotiated in mitigation practice. The fish passage raised concerns about nonfungibilities of space. As one project administrator told me, ‘This drives the Corps crazy because it is way upstream’. A federal fisheries scientist explained that such a proposal would also merit a higher level of agency scrutiny because the replacement was so far off-site. What’s more, the exchange was also out-of-kind. Models predicted that a significant portion of the sturgeon habitat to be impacted was juvenile habitat in the estuary, but the compensation was upriver adult spawning habitat. Treating habitats that are critical at different life stages and in different seasons as if they were substitutable raises concerns about nonfungibilities of type. I asked the same fisheries scientist what she would look for in assessing an off-site, out-of-kind mitigation measure like the fish passage. She said that she would want to see evidence from elsewhere that it could work – some precedent – and a robust monitoring plan to collect data before, during and after the project.⁶ It is in this kind of messy, ambiguous mitigation practice that compensatory mitigation can have ecobiopolitical implications, even for a species like the shortnose, which attracts substantial concern and investment due to its legal status.

To make the fish passage work as compensatory mitigation, critical habitats in the plural, that is, areas of the river inhabited by fish at different life stages and seasons, were collapsed into a single category of suitable, or acceptable, sturgeon habitat. As a mitigation currency, ‘acceptable habitat acres’ is useful because it is fixed, legible and, thus, conducive to exchange and claims of equivalence. But if the ultimate goal is maintaining a dynamic migratory fish population, the currency presents adequacy issues because it can externalize key biological and ecological considerations. When I asked a biologist familiar with the SHEP what he thought about the fish passage, he said, ‘That’s not really mitigation. They can spawn all they want [above the passage] and just produce larvae that go downstream and die.’ Another biologist elaborated: ‘I identify mitigation as something that is in-kind. If you eliminate downstream juvenile habitat and increase the spawning habitat, I’m not sure how it’s going to work. If there are no juveniles, there won’t be any adults.’ But what happens – or what should happen – when there are no in-kind options?

The mitigation approach that the fish passage represents is not classical Foucauldian biopolitics, with its careful attention to population composition and dynamics. It is a form of ecobiopolitics in which the milieu, sturgeon habitat, is rendered fungible to a degree that even a carefully constructed mitigation plan risks coming unmoored from the fish’s biology. Within the context of a politically and economically important project like the SHEP, the pressure to find adequate mitigation meant that, in the end, fish habitat was treated an aggregated, abstracted mitigation currency that belied the nuanced understanding of the agency professionals and scientists involved. Thus, the fish passage can be understood as a form of care for the species that may be uncaring in its ecobiopolitical implications.⁷

Our story ends with the Southern flounder. Even as the SHEP administrators went to great, if contested, lengths to mitigate potential impacts to sturgeon habitat in the lower Savannah River, their models suggested that the interplay of dredging-induced environmental changes and the mitigation measures designed in response might overflow the grid and set the estuary on a new ecological trajectory. The river might, in a sense, be more sturgeon-shaped due to the shortnose’s elevated status in the mitigation bureaucracy, but it does not follow that only sturgeon will benefit. If some forms of life, like the sturgeon, persist in the wild through mitigation and some, like the striped bass, are displaced to off-site facilities, then others, like the Southern flounder, are poised to flourish along the ecological seams of global infrastructures and supply chains (Kirksey, 2015; Swanson, 2017; Tsing et al., 2020). Thus, the SHEP’s final Environmental Impact Statement points not only to what may be lost or offset through the project, but what will be generated: an altered estuarine environment with ecological parameters that are pegged to the latest infrastructure standards of the global shipping industry (channel depth) and the environmental impact models that define minimal thresholds of suitability for aquatic life.

As the lower Savannah River becomes saltier through dredging and artificially oxygenated through the installation of an injection system, the estuary may actually become a better place for Southern flounder to live (Table 1). The oval-shaped flatfish, which is brown with white spots on the top and white on the bottom, has a distinctive appearance (Figure 6). Like other flounder, it is born with eyes on both sides of its head. As the fish ages, one eye migrates so both are on the topside. Like the sturgeon, it is a bottom-feeder, preying on shrimp, crabs and small fish. The species inhabits Atlantic coastal waters and estuaries between Virginia and Florida, as well as the Gulf Coast. It is found in the Savannah River year-round and at every life stage. Researchers conducting pre-construction monitoring for the SHEP found flounder to be among the most abundant fish in the estuary (USACE, 2015: 28). Born in the ocean, juvenile flounder migrate to lower-salinity nursery areas in estuaries and sounds, but unlike adult sturgeon and striped bass, which migrate upriver to spawn, mature flounder migrate into the ocean.

Figure 6.

Southern flounder.

Whereas salinity is a limiting factor on sturgeon habitat, the salinity tolerances of Southern flounder are, according to a Habitat Suitability Report prepared for the SHEP, ‘so wide that it can be disregarded for this species’ (USACE, 2012a: Appendix N, 257). Without mitigation, there would be adverse impacts to flounder habitat, but with the oxygen injection and flow-altering measures included in the model, the mitigation was expected to have ‘significant positive impacts’ (USACE, 2012a: 5.93), increasing suitable flounder habitat by over a thousand acres (Table 1). This led a Corps administrator to ask in an email: ‘Do we get any credit for improving Flounder habitat? ;)’ (USACE, 2012a: Appendix N, 521). He was joking, but the joke underscores the role of mitigation in bringing new ecologies into being.

The idea that humans have reorganized a river such that some forms of life flourish and others struggle echoes White’s (1995: 90) analysis of the Columbia as an organic machine. ‘If this were the old Columbia River’, he writes, ‘there would be salmon, but this is a different river. It is not the river salmon evolved in. This new river produces carp and shad. The architects of the new river … have quite consciously made a choice against the conditions that produce salmon.’ In asking whether we can protect salmon in such a transformed river, White implicitly raises thorny questions about the limits of mitigation in a place like Savannah.

Conclusion: Recalling avoidance

At the time of writing, a concerted effort is underway in the United States to dismantle the foundational environmental rules and regulations behind the mitigation analysed here, along with the institutions responsible for conducting the relevant scientific and organizational work. The existing regulatory approach has problems, but – assuming that one values non-human life and environmental quality – it is preferable to one alternative: the industry-centred approach of the past, which could return without vigilance. The Savannah River is a case in point. It is far cleaner in 2020 than it was in 1970 when a group of researchers associated with Ralph Nader arrived to study an estuary suffocated by oxygen-consuming wastes emitted from the industrial waterfront, particularly the world’s largest kraft paper mill, run by the Union Camp Corporation (Fallows, 1971). The precursor to the Clean Water Act, passed two years later, was transformative there.

The scientists and government professionals who assess environmental impacts and develop mitigation plans are tasked with an unresolvable problem. The ideal river for shipping and the ideal river for sturgeon (or bass or factories or tourists) are not the same – indeed, they can’t be. As I have emphasized, mitigation is a form of triage: minimizing and offsetting impacts to selected entities within significant institutional, financial and time constraints. The ascendant approach, compensatory mitigation, assumes that projects with significant adverse impacts can be carried out without a decline in net environmental quality through acts of substitution and replacement. For this to work, the units exchanged – tons of carbon-dioxide equivalent or fish habitat-acres – have to be deemed fungible, or fungible enough that such an intervention might conceivably neutralize what is expected to be lost. This is a conundrum for the environmentalist, who may find compensation necessary and necessarily inadequate.

Over the past few years, I have found myself mulling over a comment that a member of the Savannah environmentalist community made in passing as we discussed the ever-expanding budget for the SHEP’s baroque assembly of mitigation measures. She said, ‘If the airbag is half the cost of the car, you might want to reconsider buying the car.’ Infrastructure projects like the SHEP can seem inevitable given the powerful interests that stand behind them. This, in turn, makes their environmental impacts appear unavoidable, making compensation appear to be the only pragmatic option. Indeed, many of the scientists and conservationists I interviewed about the SHEP framed the dynamic in this way: The mitigation is messy and probably won’t actually work, but the project will proceed regardless. Given this, what positives can come out of the project? For example, can it advance key research or an organization’s conservation priorities?

How, then, might mitigation be reconceptualized? As twenty-first century environmental management embraces the rationality of exchange, substitution and replacement associated with compensatory mitigation, it is worth recalling, and perhaps reimagining, the first step of the mitigation sequence: avoidance. According to Section 404 of the US Clean Water Act, ‘Avoidance means mitigating an aquatic resource impact by selecting the least-damaging project type, spatial location and extent compatible with achieving the purpose of the project.’ One of my interlocutors, a fish biologist at a state agency, emphasized this point, ‘In our world, we’d step it [the channel depth] back to where there wouldn’t be an impact.’ If, however, the default assumption is that projects with significant political-economic backing are inevitable, possibilities for avoidance are quite narrow. We are left with: How can the expected impacts be minimized or offset? Or, how do we build a better airbag? As cars grow larger and faster and highways get busier, increasing the risk and significance of impacts, we can design ever more elaborate airbags. The airbag can even become a defining feature of the vehicle’s design. But the airbag cannot alter the relations that produce vehicle collisions in the first place.

If the airbag is half the cost of the car, you might want to reconsider buying the car. If the mitigation is half the cost of the infrastructure project, you may want to reconsider the project itself. As operationalized in the United States, compensatory mitigation is a response to a development approach shaped by an economic growth-centered ideology. A broader conceptualization of avoidance and, thus, mitigation, might move from assessing projects like the SHEP in terms of the economic and environmental implications of dredging the Savannah harbor to various depths – alternative versions of one project – to asking more fundamental questions concerning choices among multiple projects with common goals. Do we really need deep-dredged harbors with channels nearly 50 feet deep in Savannah, Charleston and Jacksonville?

Footnotes

Acknowledgements

It took me a long time to write and revise this article. Over the years, I have received incredibly helpful feedback on drafts from Jessica Barnes, Jessica Cattelino, Maron Greenleaf, Stefan Helmreich, Colin Hoag, Alex Nading and Tasha Rijke-Epstein, as well as participants in the Fluid Ecologies Conference at Stanford University and the Robert Penn Warren Center’s Science, Technology and Society Seminar at Vanderbilt University. I would also like to thank the four anonymous reviewers for their constructive criticism and useful suggestions.

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Ashley Carse

Notes

Author biography

Ashley Carse is an anthropologist of infrastructure and the environment. He is the author of Beyond the Big Ditch: Politics, Ecology, and Infrastructure at the Panama Canal (MIT, 2014). His current research focuses on the political ecology of the global shipping industry, with a focus on dredging and logistics-led development in Savannah and the southeastern United States.

References

Bakker

(2004) Uncooperative Commodity: Privatising Water in England and Wales. Oxford: Oxford University Press.

Bakker

Bridge

(2006) Material worlds? Resource geographies and the ‘matter of nature’. Progress in Human Geography 30(1): 5–27.

Barber

Gann

(1989) A History of the Savannah District 1829–1989. Savannah: US Army Corps of Engineers.

Barry

(2002) The anti-political economy. Economy and Society 31(2): 268–284.

Bray

(2008) Environmental Aspects of Dredging. Leiden: Taylor and Francis.

Bridge

(2010) Resource geographies 1: Making carbon economies, old and new. Progess in Human Geography 35(6): 820–834.

Bull

Suttle

Gordon

, et al. (2013) Biodiversity offsets in theory and practice. Oryx 47(3): 369–380.

Bumpus

(2011) The matter of carbon: Understanding the materiality of CO2 in carbon offsets. Antipode 43(3): 612–638.

Bumpus

Liverman

(2008) Accumulation by decarbonization and the the governance of carbon offsets. Economic Geography 84(2): 127–155.

10.

Carey

(2005) The Philosopher Fish: Sturgeon, Caviar and the Geography of Desire. New York: Counterpoint.

11.

Carse

Lewis

(2017) Toward a political ecology of infrastructure standards: Or, how to think about ships, waterways, sediment and communities together. Environment and Planning A 49(1): 9–28.

12.

Choy

(2011) Ecologies of Comparison: An Ethnography of Endangerment. Durham: Duke University Press.

13.

Collier

(2009) Topologies of power: Foucault’s analysis of political government beyond ‘governmentality’. Theory Culture Society 26(6): 78–108.

14.

Collins

Rogers

Smith

TIJ

, et al. (2000) Primary factors affecting sturgeon populations in the southeastern United States: Fishing mortality and degradation of essential habitats. Bulletin of Marine Science 66(3): 917–928.

15.

Cooper

(2015) Measure for measure: Commensuration, commodification and metrology in emissions markets and beyond. Environment and Planning A 47(9): 1787–1804.

16.

Covington

(2015) Estimating Errors: The Politics of Environmental Impact Assessment Along the Savannah River. PhD Thesis, University of Wisconsin-Milwaukee.

17.

Crance

(1986) Habitat suitability models and instream flow suitability curves: Shortnose sturgeon. Report no. 82(10.129). Fort Collins: US Fish and Wildlife Service.

18.

Dalsgaard

(2013) The commensurability of carbon: Making value and money of climate change. Journal of Ethnographic Theroy 3(1): 80–98.

19.

Erftemeijer

PLA

Robin Lewis

(2006) Environmental impacts of dredging on seagrasses: A review. Marine Pollution Bulletin 52(12): 1553–1572.

20.

Erftemeijer

PLA

Riegl

Hoeksema

, et al. (2012) Environmental impacts of dredging and other sediment disturbances on corals: A review. Marine Pollution Bulletin 64(9):1737–1765.

21.

Errington

Gewertz

(2018) Managing an endangered species: Palliative care for the pallid sturgeon. American Ethnologist 45(2): 186–200.

22.

Espeland

Stevens

(1998) Commensuration as social process. Annual Review of Sociology 24(1): 313–343.

23.

Fallows

(1971) The Water Lords: Ralph Nader’s Study Group Report on Industry and Environmental Crisis in Savannah, Georgia. New York: Grossman Publishers.

24.

Fischer

Newell

(2008) Environmental and technology policies for climate mitigation. Journal of Environmental Economics and Management 55(2): 142–162.

25.

Foucault

(2008) The Birth of Biopolitics: Lectures at the Collège de France, 1978–1979. New York: Picador.

26.

Georgia Ports Authority (2016) Economic development brochure. Available at: https://gaports.com/economic-development/ (accessed 20 December 2020).

27.

Goodman

Boyd

(2011) A social life for carbon? Commodification, markets and care. Geographical Journal 177(2): 102–109.

28.

Granger

(1968) Savannah Harbor: Its Origin and Development, 1733–1890. Savannah: US Army Corps of Engineers.

29.

Grossberg

(2010) Modernity and commensuration: A reading of a contemporary (economic) crisis. Cultural Studies 24(3): 295–332.

30.

Günel

(2016) What is carbon dioxide? When is carbon dioxide? PoLAR 39(1): 33–45.

31.

Helmreich

(2011) Nature-culture-seawater. American Anthropologist 113(1): 132–144.

32.

International Union for Conservation of Nature (IUCN) (2010) Sturgeon more critically endangered than any other group of species. Available at: https://www.iucn.org/content/sturgeon-more-critically-endangered-any-other-group-species (accessed 14 December 2017).

33.

International Union for Conservation of Nature (IUCN) (2020) Sturgeon. Available at: https://www.iucn.org/commissions/ssc-groups/fishes/sturgeon-specialist-group (accessed 20 December 2020).

34.

Jarić

Riepe

Gessner

(2017) Sturgeon and paddlefish life history and management: Experts’ knowledge and beliefs. Applied Icthyology 34(2): 1–14.

35.

Jenkins

(2016) From mitigation to sustainability: The evolution of incorporating environmental factors into development decisions in Australasia. International Journal of Sustainable Development and Planning 11(6): 920–929.

36.

Jensen

(2020) A flood of models: Mekong ecologies of comparison. Social Studies of Science 50(1): 76–93.

37.

Kirksey

(2015) Emergent Ecologies. Durham: Duke University Press.

38.

Lakoff

(2016) The indicator species: Tracking ecosystem collapse in arid California. Public Culture 28(2): 237–259.

39.

Lemke

(2011) Biopolitics: An Advanced Introduction. New York: New York University Press.

40.

(2013) Contested equivalences: Controversies over water and mining in Peru and Chile. In: Wagner

(ed.) The Social Life of Water. New York: Berghahn, pp. 18–35.

41.

Liboiron

(2016) Redefining pollution and action: The matter of plastics. Journal of Material Culture 21(1): 87–110.

42.

Little

(2012) Another angle on pollution experience: Toward an anthropology of the emotional ecology of risk mitigation. Ethos 40(4): 431–452.

43.

Little

(2013) Envisioning the political ecology of mitigation in a microelectronic disaster setting. Journal of Political Ecology 20: 217–237.

44.

Lohmann

(2005) Marketing and making carbon dumps: Commodification, calculation and counterfactuals in climate change mitigation. Science as Culture 14(3): 203–235.

45.

Lorimer

(2017) Probiotic environmentalities: Rewilding with wolves and worms. Theory, Culture and Society 34(4): 27–48.

46.

Lovell

Liverman

(2010) Understanding carbon offset technologies. New Political Economy 15(2): 255–273.

47.

Lovell

Mackenzie

(2011) Accounting for carbon: The role of accounting professional organisations in governing climate change. Antipode 43(3): 704–730.

48.

MacKenzie

(2009) Making things the same: Gases, emission rights and the politics of carbon markets. Accounting, Organizations and Society 34(3–4): 440–455.

49.

Maron

Brownlie

Bull

, et al. (2018) The many meanings of no net loss in environmental policy. Nature Sustainability 1(1): 19–27.

50.

Mathews

Barnes

(2016) Prognosis: Visions of environmental futures. Journal of the Royal Anthropological Institute 22: 9–26.

51.

Maurer

(2006) The anthropology of money. Annual Review of Anthropology 35: 15–36.

52.

McKenney

Kiesecker

(2010) Policy development for biodiversity offsets: A review of offset frameworks. Environmental Management 45(1): 165–176.

53.

Moreno-Mateos

Maris

Béchet

, et al. (2015) The true loss caused by biodiversity offsets. Biological Conservation 192: 552–559.

54.

National Marine Fisheries Services (NMFS) (1998) Recovery plan for the shortnose sturgeon, Acipenser brevirostrum. Report for the National Oceaninc and Atmospheric Administration. Silver Spring: NMFS.

55.

Olson

(2010) The ecobiopolitics of space biomedicine. Medical Anthropology: Cross Cultural Studies in Health and Illness 29(2): 170–193.

56.

Prudham

(2005) Knock on Wood: Nature as Commodity in Douglas-Fir Country. London: Routledge.

57.

Rabinow

Rose

(2006) Biopower today. BioSocieties 1(2): 195–217.

58.

Ramos

(2017) Resilience, path dependence and the port: The case of Savannah. Journal of Urban History. Epub ahead of print 19 April 2017. DOI: 10.1177/0096144217704183.

59.

Reinert

(2004) Decline and Recovery of Striped Bass in the Savannah River Estuary: Synthesis and Re-analysis of Historical Information and Evaluation of Restoration Potential. Athens: University of Georgia.

60.

Richardson

Weszkalnys

(2014) Introduction: Resource materialities. Anthropological Quarterly 87(1): 5–30.

61.

Robertson

(2000) No net loss: Wetland restoration and the incomplete capitalization of nature. Antipode 32(4): 463–493.

62.

Robertson

(2004) The neoliberalization of ecosystem services: Wetland mitigation banking and problems in environmental governance. Geoforum 35(3): 361–373.

63.

Robertson

(2006) The nature that capital can see: Science, state and market in the commodification of ecosystem services. Environment and Planning D: Society and Space 24(3): 367–387.

64.

Rodrigue

(2020) The Port of Savannah logistics cluster. Available at: https://porteconomicsmanagement.org/?page_id=2046 (accessed 25 January 2021).

65.

Rundcrantz

Skarback

(2003) Environmental compensation in planning: A review of five different countries with major emphasis on the German system. European Environment 13(4): 204–226.

66.

Salzman

Ruhl

(2000) Currencies and the commodification of environmental law. Stanford Law Review 53(3): 607–692.

67.

Salzman

Ruhl

(2006) No net loss: Instrument choice in wetlands protection. In: Freeman J and Kolstad CD (eds) Moving to Markets in Environmental Regulation: Lessons from Twenty Years of Experience. New York: Oxford University Press, pp.323–350.

68.

Saxton

(2015) Strawberry fields as extreme environments: The ecobiopolitics of farmworker health. Medical Anthropology 34(2): 166–183.

69.

Schinkel

(2016) Making climates comparable: Comparison in paleoclimatology. Social Studies of Science 46(3): 374–395.

70.

Shortnose Sturgeon Status Review Team (2010) Biological Assesment of Shortnose Sturgeon (Acipenser brevirostrum). Report for the National Marine Fisheries Service, Report no. AR-405, 1 November. Vero Beach: US Fish and Wildlife Service.

71.

Singer

Hodge

(2016) Ecobiopolitics and the making of native American reservation health inequities. In: Singer

(ed.) A Companion to the Anthropology of Environmental Health. West Sussex, UK: Wiley Blackwell, pp.193–215.

72.

Swanson

(2017) Methods for multispecies ethnography: Thinking with salmon otoliths and scales. Social Analysis 61(2): 81–99.

73.

Taylor

(1999) Making Salmon: An Environmental History of the Northwest Fisheries Crisis. Seattle: University of Washington Press.

74.

Todd

VLG

Todd

Gardiner

, et al. (2015) A review of impacts of marine dredging activities on marine mammals Victoria. ICES (Institute for Clinical Evaluative Sciences) Journal of Marine Science 72(2): 328–340.

75.

Tsing

Deger

Keleman

, et al. (2020) Feral Atlas: The More-Than-Human Anthropocene. Palo Alto: Stanford University Press.

76.

US Army Corps of Engineers (USACE) (2012a) Final Environmental Impact Statement: Savannah Harbor Expansion Project. Report for the Savannah Harbor Expansion. Savannah: USACE.

77.

US Army Corps of Engineers (USACE) (2012b) Final General Re-Evaluation Report: Savannah Harbor Expansion Project. Savannah: USACE.

78.

US Army Corps of Engineers (USACE) (2015) Monitoring Report: Savannah Harbor Expansion Project. Savannah, GA: USACE.

79.

US Army Corps of Engineers (USACE) (2017) SHEP placemat. Available at: https://www.sas.usace.army.mil/Portals/61/siteimages/CongressionalVisit/SHEPPlacemat.pdf (accessed 27 January 2021).

80.

US Army Corps of Engineers (USACE)-US Environmental Protection Agency (EPA) (2013) Compromise and Settlement Agreement for the Savannah Riverkeeper v. United States Army Corps of Engineers. Available at: http://caselaw.findlaw.com/us-6th-circuit/1059617.html (accessed 23 December 2020).

81.

US Department of the Interior (USDOI) (1973) Threatened Wildlife of the United States. Report no. 114. Washington, DC: Office of Endangered Species and International Activities.

82.

US Department of Transportation (USDOT) Bureau of Transportation Statistics (2020) Port Profile: Port of Savannah. Available at: https://explore.dot.gov/#/views/PortPerformance-temp-view2/ProfileDashboard?Port ID=776 (accessed 23 December 2020).

83.

US Environmental Protection Agency (USEPA)-US Army Corps of Engineers (USACE) (1990) Memorandum of Agreement Between the Environmental Protection Agency and the Department of the Army Concerning the Determination of Mitigation Under the Clean Water Act Section 404(b)(1) Guidelines. Available at: https://www.epa.gov/sites/production/files/2019-05/documents/1990_army-epa_mitigation_moa.pdf (accessed 23 December 2020).

84.

White

(1995) The Organic Machine: The Remaking of the Columbia River. New York: Hill and Wang.

85.

Wills

(2001) An Analysis of the Consensus-Based Participation Model in the Georgia Ports Authority’s Stakeholder Evaluation Group. PhD Thesis, University of Georgia.

86.

Wills-Toker

(2004) Public participation or stakeholder frustration: An analysis of consensus-based participation in the Georgia Ports Authority’s Stakeholder Evaluation Group. In: Depoe

Delicath

Elsenbeer

M-FA

(eds) Communication and Public Participation in Environmental Decision Making. Albany: SUNY Press, pp.175–199.