The world is in motion. Twice a year, on the West Coast of North America, millions of salmon fight their way upriver, returning to the shady bend in a small stream where they were born — or the spot below the fish hatchery where they were dumped into the river. Millions of humans make the opposite journey, leaving home in search of economic or environmental security. Biomes grow and shrink. The ocean expands. The Pacific Plate grinds and slips against the North American Plate, traveling northwest a few inches every year.
Environmental change is constant, although it is not always perceptible or predictable. Landscape designers grapple daily with this problem, and many now focus their practice on designing for adaptation to change. In Projective Ecologies, Chris Reed and Nina-Marie Lister trace the ecological turn in the biological and earth sciences as it reverberates across the humanities and design fields. Landscape architects and planners have followed ecologists “toward a more organic model of open-endedness, flexibility, resilience and adaptation, and away from a mechanistic model of stability and control.” 1
Qualitatively different landscapes can and do manifest upon a single geographic terrain.
And yet, most of us — designers included — imagine migration as the movement of isolated things (fish, birds, people) against a fixed background (the Klamath River, Pacific Flyway, U.S.-Mexico border). We know that environmental conditions are always changing, but we allow ourselves the fiction of background stability. When we limit our thinking in this way, our political and design responses are circumscribed. (Allot water rights. Designate a wildlife refuge. Build a wall.) Not surprisingly, they often fail.
I would like to assert a more relational definition of migration as patterned movement across space and time. We can then look beyond the movement of individuals or species to the migration of landscapes. A landscape migrates when its unique assembly of components — the materials, entities, and actors that define it — shifts such that, over time, a new assembly forms. Qualitatively different landscapes can and do manifest upon a single geographic terrain.
We are now well into a geologic era — the Anthropocene — characterized by the acceleration of environmental change. Landscapes are changing faster (moving farther) than ever before. Understanding this phenomenon as landscape migration can help us observe these processes and effects. The material, ecological, and social frameworks that structure the landscape — its infrastructures — are the primary mechanisms by which landscape migration occurs and the avenues of intervention for designers and planners who wish to influence trajectories of environmental change.
Let’s start with those salmon.
The Klamath River and its tributaries drain a watershed of 16,000 square miles in California and Oregon which has been intensely contested by farmers, fishers, energy producers, environmentalists, and indigenous peoples. Distinguished by its large, shallow lakes and extensive wetlands, the Klamath Basin was once known as the “Everglades of the West.” Because the lakes and marshes were located near the headwaters rather than the mouth, it is sometimes called an upside down river. 2 From the late 19th to mid 20th century, much of the upper basin, on the eastern flank of the Cascade mountain range, was converted to ranching and agricultural use. 3
Before the river was dammed and the lakes and wetlands drained, the Klamath Basin supported the third largest salmon migration on the West Coast. Some say it will yet again. In 2010, Interior Secretary Ken Salazar and governors of the two states brokered agreements among more than 50 groups to remove four dams on the Klamath and “restore” favorable conditions for migratory fish. 4 If approved by Congress (which has delayed voting on a bill), the dam removal project will be the largest in U.S. history.
The dividing line between the lower and upper Klamath is Iron Gate Dam: a 175-foot earthen wall, 190 miles from the ocean, that blocks migratory fish from traveling farther upriver. Just below the dam is the Iron Gate Fish Hatchery, legally required to manufacture the migration that it otherwise obstructs. Twice a year, Chinook and coho salmon and steelhead return here, to the laboratory where they were spawned. They are collected, counted, euthanized, and cut into pieces — work performed physically by hand on a disassembly line. Then their carcasses are loaded onto refrigerated trailers and shipped to a processing plant in Bellingham, Washington, where they are filleted, frozen, and distributed to charitable and non-profit organizations serving disadvantaged communities from Grants Pass, Oregon, to Sacramento, California. (This year, 99,000 pounds of fillets were distributed.) Back at the hatchery, sperm and eggs extracted from the salmon are stirred together, and the fertilized eggs are incubated in a large array of trays in a nondescript cinder-block building. Then the tiny fish are reared in a series of raceway pools until they are large enough to be released into the river. They swim to the ocean, where they will grow to adulthood before returning to complete the cycle. 5
Above the dam is a novel ecosystem: a 54,000 acre-foot capacity reservoir supporting exotic species of yellow perch and largemouth bass. Here the still, deep water is often an opaque emerald green, thanks to dense blooms of cyanobacteria that thrive in the warm conditions. Signs warn visitors of the health risks of ingestion. If Iron Gate Dam is removed, this landscape will change again, as a nascent river flows freely through a valley resembling an empty bathtub, its barren, muddy slopes subject to vegetative succession. The obsolete hatchery will be left to ruin. The upper reaches of the Klamath will be reconstituted as a new assembly of sediments, nutrients, fish, and rushing water — a landscape attempting to migrate to an earlier state, when wild salmon thrived.
Farther upriver, past five more dams, is the Klamath Project, one of the first large-scale irrigation projects in the United States. In 1902, the Newlands Reclamation Act authorized the federal government to construct “irrigation works for the storage, diversion and development of waters” in arid lands across 16 western states. At the time, reclamation had a specific purpose: to facilitate homesteading in regions where lack of rainfall was the limiting factor. 6 Water flowing to the oceans in Western rivers was perceived as a wasted resource that could be put to more productive use, and wetlands were seen as spawning grounds for disease. 7 Within five years, the Reclamation Service (later the Bureau of Reclamation) began dozens of projects across the West, “making lakes at some places and drying them up at others … at some places turning rivers out of their beds and at others bringing underground waters to the surface.” 8 It was an unprecedented experiment in infrastructural development whose works still shape landscapes throughout the West. At the Klamath Project, the Bureau drained many lakes and marshes — including two large shallow lakes, Lower Klamath and Tule, which covered nearly 200,000 acres — and built a system of dams and reservoirs to store water which could be distributed for irrigation.
Lower Klamath Lake was extensively surveyed in 1903 and again in 1905 to assess its potential as a reclamation project. Remarkable differences between those surveys reveal an extremely dynamic environment that stretches our understanding of the term “lake.” High water in the Klamath River during winter and spring would overflow into the lake through the Klamath Straits, and during dry summers the water would reverse flow back into the river. Drought and climate variability could push these cycles into longer, multi-year intervals. Over time, these migrations produced a shifting mosaic of landscapes ranging from deep open water to shallow wetlands. Like a massive detention basin, the extensive complex of wetlands absorbed much of the Klamath’s high nutrient load. A rich collection of migratory fauna and flora evolved according to these patterns. 9
As the U.S. Geological Survey and the Bureau of Reclamation assessed the landscape, the Audubon Society sent two men to conduct their own proprietary survey. William Finley and Herman Bohlman trudged into the vibrant wetlands of Lower Klamath Lake with a large format camera to document birdlife. Among the “jungles” of an “endless area of floating tule islands,” they encountered “the most extensive breeding ground in the west for all kinds of inland water birds.” 10 Little understood at the time, these lakes were one of the most important stopovers along the migration route of the Pacific Flyway.
The evocative photographs of the Audubon expedition sparked a national conservation effort, which led Theodore Roosevelt to amend his own legislation with executive order 924, creating the Klamath Lake Reservation, “a preserve and breeding ground for native birds.” 11 Later renamed the Lower Klamath National Wildlife Refuge, it was the first wildlife refuge in the United States, established on lands already ceded to the Bureau of Reclamation. It soon became a popular spectacle:
Literally clouds of birds of many species darkened the sky; the thunder of their wings was like the roar of distant surf, and their voices drowned out all other sounds. These vast summer flocks were greatly increased each fall when the legions of ducks, geese, swans, and cranes from northern nesting grounds stopped on their journey to their winter homes in the valleys of California, and again in spring on their return northward. …
Visitors from the city of Klamath Falls [boated] down the Klamath River and directly into Lower Klamath Lake to see the vast colonies of western and eared grebes or the pelicans, herons, gulls, and terns that nested there by the tens of thousands. 12
But in 1917, water was diverted for the irrigation project, and the effect was severe:
A gate was built across the channel leading from Klamath River into Lower Klamath Lake, thus preventing the water from flowing into the lake. Within 4 years this vast waterfowl paradise had dried up. Peat fires started, which in many places burned to a depth of 6 feet or more, leaving nothing but a vast alkaline, ashy desert from which clouds of choking dust arose, often obscuring the sun. 13
The fiery dust bowl on Lower Klamath Lake was one of many unforeseen consequences of reclamation. Draining the lake induced an aggregate migration of water, vegetation, geese, peat soils, and vast clouds of ash that reordered the landscape. But the Bureau never reversed its actions; it simply recalibrated its infrastructure with more engineering. After that entropic mishap, the Bureau bored a 6,000-foot tunnel through Sheepy Ridge, which divided Tule and Lower Klamath lakes. Water pumped through the tunnel connected the previously separate watersheds.
Thus, the Lower Klamath Refuge became part of the Bureau’s infrastructure, both as a sump for irrigation runoff from the new agricultural lands, and as a client for water deliveries to irrigate the refuge and create bird habitat. But here’s the rub: because it was created after the reclamation project, the refuge is dubiously considered last in time, which means, legally, last in line for water. The refuge has no senior water rights; indeed, no guarantee that in dry years it will receive any water at all. 14 Managers have had to figure out how to design wildlife habitat under “extreme constraints and compromises” in a landscape transformed for agricultural production. 15
Refuge biologists are effectively designing with water to induce migratory processes.
Like most U.S. wildlife refuges, Lower Klamath is managed as a collection of units — leveed parcels of land that are flood-irrigated by a network of canals running between them. By manipulating each unit’s water levels, refuge biologists have learned to create a wide variety of habitats for water birds. As stated in the management plan: “The natural timing and duration of many of the forces that historically shaped the marsh no longer occur. Wetland managers must now manipulate these forces (fire, flooding, and drainage) and other tools to affect wetland succession on the refuge, thereby providing for a variety of vegetative communities and their associated wildlife species.” 16
Through these experiments, the biologists found that wetland landscapes require change and disturbance to remain productive. Fluctuation in water levels facilitates the regeneration of aquatic plants, which in turn renews habitat features — exactly how Lower Klamath Lake performed prior to the reclamation project. Landscape infrastructure such as pumps, canals, gates, and levees gives managers the tools to recreate those conditions. They are effectively designing with water to induce migratory processes. Today, Lower Klamath is one of the most biologically productive refuges of its kind, which is remarkable, considering how thoroughly plumbed and heavily managed it is. 17
In the 1990s, refuge biologists took their experiments one step further by hybridizing the habitat scheme with agricultural production. 18 Inspired by the longstanding practice of crop rotation, they designed an innovative system of wetland rotation. Prototyped by David Mauser in the 1990s, the “walking wetlands” program began with the experimental flooding of Tule Lake’s agricultural lease lands after fall harvest, creating immediate habitat benefits for eagles and other birds. The units were flooded for one to three years (as the wetlands progressed through various successional stages) before they were returned to farming. The results were surprising: reduced soil pathogens, enhanced soil fertility and tilth, reduced need for farming inputs and fertilizers, and higher quantity and quality of agricultural yields. 19 In fact, the ephemeral wetlands improved soils to such a degree that high-end organic farming became possible. Klamath farmers, historically at odds with the refuge, realized the value of the program and sought to extend it onto their own lands through a crop sharing program.
Now the Klamath Basin hosts a migratory network of walking wetlands: a shifting mosaic of landscapes passing through different material states.
Now the Klamath Basin hosts a migratory network of walking wetlands — a shifting mosaic of landscapes passing through different material states — that comprises public and private land. The management techniques are being replicated in other agricultural regions throughout the West. 20 The biologists’ sophisticated design is especially notable for having been achieved within a confounding cultural and political framework of zoned, parceled, engineered land. They had no choice but to learn how to farm wildlife by appropriating the Bureau of Reclamation infrastructures, and they did it through 1:1 scaled experimental trials, followed by extensive monitoring and engagement with the landscape’s performative responses. 21 Rather than restoring the landscape to a former “natural” state, refuge managers found ways to reclaim reclamation by retooling the very system that had eradicated the lakes and wetlands.
Across the country, the Fish and Wildlife Service manages 70 fish hatcheries and more than 560 wildlife refuges. Their interventions on the Klamath River illustrate how dramatically landscapes can be changed by the infrastructures embedded within them. Landscape assemblies are elastic, dynamic formations, but they are also grounded within geographic and socio-political contexts.
Even the meaning of “reclamation” is contextually bound. Since the Middle Ages, the Dutch have been refining the world’s most advanced reclamation practice, carving out over 50 percent of their national land area from the North Sea. This constructed landschap, whose geometry is defined by polders, dikes, and open fields, is tied to a cultural gestalt that regards landscapes as fully moldable. Over centuries of embodied practice — which includes deadly failures as well as successes — Dutch land-making has become more sophisticated through experimentation. Here the tendency is to work with landscape’s migratory propensities, rather than to wall them off. 22
In 1995, the landscape architecture firm West 8 produced a speculative design for Buckthorn City, a real estate development that would accommodate urban expansion in the Rotterdam-Hague region. The firm proposed to reclaim a portion of the North Sea coast by depositing large quantities of sand just offshore. This infrastructural base would migrate freely, pushed by wind and water, until it achieved an architecturally desirable form. At that “decisive moment,” the sand would be heavily seeded with buckthorn to stabilize the formation. 23 Over time, forests, grasslands, and creeks would manifest, and a city of approximately 8,400 acres would be grafted into that prefab ecology. 24
Along the coast, there sits a 20 million cubic meter mound of sand, in the shape of a giant hook.
A decade later, West 8 came back with an even more ambitious proposal for Happy Isles, a massive offshore archipelago manufactured by industrial dredging. The design echoed island-building projects in Dubai, but here the goal was climate resilience. The plan promised nothing less than an entire reordering of the North Sea coastal zone, thickening and diversifying it by engineering an inland sea and barrier island typology. 25 West 8’s design functioned not only as a material proposal but also as a political instrument for questioning the government’s response to climate change and rising seas.
Along the coast where Buckthorn City was proposed, there now sits a 20 million cubic meter mound of sand, in the shape of a giant hook, deposited in 2011 by the Dutch Ministry of Infrastructure and the Environment. Here we find dunes, seals, kite boarders, and a 40-meter tower with time lapse cameras monitoring all the action. The Zandmotor, or Sand Engine, is a prototypical infrastructure designed to replenish the coast over the next 20 years. The coastline of the Netherlands moves and erodes (as most do), and for decades the Dutch have reinforced the coast every five years with deposits of imported sand. The Sand Engine attempts to spring-load this process by placing a larger deposit in one location. The form and placement of this massive earthwork is based on a statistical model that predicts where the sand will go, accounting for wave fields, flow velocity fields, sediment transport, predicted erosion of adjacent coastlines, and accelerated sea level rise. Theoretically, environmental forces will do the work once performed by bulldozers and backhoes. The government is monitoring the migration of sand and the waning of the landform to see if the computational design strategy plays out as intended.
Migrations of the Mississippi River
How we represent and model landscapes influences how we perceive and engage with them. Designers such as James Corner, Alan Berger, Anuradha Mathur and Dilip da Cunha, Jane Wolff, and Bradley Cantrell have repeatedly demonstrated this point. In particular, Mathur and da Cunha have advanced a critical inquiry of cartographic conventions and received habits of landscape representation. Their work — a self-described activist practice — emphasizes the ways that conventional mapmaking and planning suppresses movement, variability, and flux. They problematize the drawing or projection of fixed boundaries onto inherently dynamic landscapes, showing “how these divisions and lines can harden in the landscape, in civic administration, and indeed in the design disciplines.” By interrogating these conventions, they open up “possibilities and material practices that have been marginalized, or are not even on the table.” 26 In Mississippi Floods, they pair Harold Fisk’s mappings of historic meander courses of the Mississippi River (1944) with the Army Corps of Engineers’ Design Flood Diagram, juxtaposing radically different takes on space, time, migration, and landscape agency. 27 Through their layered diagrams of meanders, flows, banks, and beds, Mathur and da Cunha illustrate just how extensively the Lower Mississippi was altered by multiple infrastructures.
River deltas are extraordinary migratory landscapes where nearly everything is in constant motion at varying time scales, from daily tides to thousand-year intervals in river avulsions. The re-engineering of the Mississippi River was a pivotal and transformative event in its history, but infrastructural constraints do not arrest time. Landscapes never stop moving, no matter how hard we try to fix them in place. Rather, flows and processes within them are subject to distortion, to variable slowing and acceleration. Landscapes respond to constraint by moving differently, often arriving at surprising and undesirable manifestations. Mississippi Floods leaves us with a lingering question: what is the new migratory syndrome of the river and its watershed?
Dams throughout the Mississippi Basin (which encompasses 40 percent of the contiguous United States) have reduced the volume of sediment traveling downriver by more than half. What’s more, the concrete armoring of the river’s banks speeds up its flow, so that much of the sediment it does carry is swept into the Gulf of Mexico during high flows. Largely as a result of this altered sedimentary budget, 28 the Mississippi Delta is disappearing at the average rate of one football field every hour. In less than a century, we have inadvertently reversed the trajectory of the world’s third largest delta. In response to this crisis, a regional recovery effort has been organized under the 2012 Louisiana Coastal Master Plan. 29 The plan’s core strategy is to create sediment diversions within the delta: engineered openings in the levee network through which sediments will be redirected to create new coastal land masses. The emerging discipline of restoration sedimentology 30 relies on an explicitly migratory strategy, using sediment as an infrastructural base which can be colonized by other parts of the ensemble, from plants and animals to recreational boaters.
Just as we saw in the Klamath River examples, the infrastructure of the Mississippi River Delta is being retrofitted (or de-engineered) in an attempt to persuade landscape elements to move and behave as they did in an earlier time. In effect, designers hope to induce a return migration of the landscape.
In an urban context, we can trace landscape migration by focusing on shifting economic and demographic patterns. Alan Berger’s Drosscape describes material cycles of urban agglomeration and disassembly. 31 In Detroit and other shrinking cities, we witness the emigration of the human population and the emergence of feral urban lands. 32 Like many migrations, the Rust Belt exodus is “driven by the transitory availability and shifting location of resources.” 33 De-industrialization is not the disappearance of industry, but rather the migration of industrial production elsewhere. Detroit’s transformation is part of a broad, complex set of economic processes, yet not so complex that we fail to see the spatial patterns.
Here we must remember that not every migration is one of return. The most promising design and planning strategies for shrinking cities work opportunistically and projectively with emergent conditions, rather than trying to counter or reverse the trajectories of change. 34
Toward a Theory of Landscape Migration
The act of migration is inseparable from the material and spatial dynamics in which a living thing or group of living things interact; rendering migration as “a complicated, challenging and diverse phenomenon involving changing statuses and multiple geographic trajectories.”
— Michael Samers, Migration (2010) 35
At a planetary scale, we know that landforms migrate through plate tectonics. There is no significant debate about this phenomenon. Yet, surprisingly, we’ve only understood this for about 50 years. 36 In that brief span, we’ve become adept at visualizing, monitoring, and anticipating earthquakes, volcanic eruptions, and other geologic events. The San Andreas Fault in California is a tangible artifact of these processes: a trench bordered on both sides by jagged hills where the North American Plate rubs against the Pacific. Aerial views reinforce the fact that material in motion creates qualitative difference. Landscape migration is both spatial and qualitative.
As designers, planners, and land managers, we lack a concept of migration which would focus our attention on the dynamic moving whole.
If we accept the idea that the entire surface of the earth is migratory, then why not landscapes in particular? A landscape — as a scene, landschap, ecosystem, and socio-political territory — is a material assembly of moving entities, a dynamic medium which changes in quality and structure through the aggregate movements or actions of the things that constitute it. Unfortunately, as designers, planners, and land managers, we lack a concept of migration which would focus our attention on the dynamic moving whole. How can we effectively apply the concept of migration — again, patterned movement across space and time — to landscapes? What models can we build upon? Briefly, I offer three strands of theory and practice that substantiate landscape migration as a real, material phenomenon.
Ecology provides a fundamentally relational view of landscapes, focusing on the generative interactions between things and their environment. Early ecological paradigms, which tended to privilege stability and linear progressions to sustained climax states, have been superseded by models that emphasize entropy, disturbance mechanisms, indeterminacy, and non-linear dynamics. 37 The shift to integrative ecologies that acknowledge human agency has been largely welcomed in landscape architecture. Rather than being a prescriptive straitjacket, the new ecological thinking has opened avenues for creative practice that focus on mechanisms of process and change, wherein spatial reality is characterized by “movement and passage” and “propulsive life unfolding in time.” 38 In James Corner’s formulation, landscape is not terra firma but “terra fluxus”; spatial form is a “provisional state of matter, on its way to becoming something else.” 39 Reed and Lister survey the expanding repertoire (or cacophony) of ecological theory that has infiltrated design practice, and they argue for a more rigorous application of ecological concepts across the material, social, spatial, and political dimensions of landscape. 40
The scientific concept of resilience also resonates in design discourse. Landscape resilience — the degree to which a landscape can absorb and adapt to change without becoming something qualitatively different — is perhaps best expressed by the theory of adaptive change advanced by Lance Gunderson and C.S. Holling in Panarchy: Understanding Transformations in Human and Natural Systems. Panarchy theory demonstrates landscape’s intrinsic mobilities and elasticities, asserting that there is not one but multiple possible semi-stable states for any given landscape, arrived at through diverse processes of “creative destruction and renewal.” 41 The Panarchy diagram — which has been widely circulated in contemporary design research — depicts four entropic stages: exploitation, conservation, release, and reorganization. One example is forest fire suppression in the American West over the past century. We now understand that fire is an ecologically beneficial and regenerative force that periodically renews the landscape. Conventional practices of fire suppression disrupt this process, leading to massive accumulations of woody matter, which burn catastrophically in huge releases.
The Panarchy model is especially useful for focusing attention on the thresholds of “stability domains” 42 at which the assembly and performance of a landscape changes. Resilience theory instilled new rigor into practices of adaptive management, as it was one of the first constructs to demonstrate the need for landscapes to move and fluctuate over time. Designing for stasis is ineffective, if not impossible, as doing so tends to produce unintended migrations. Landscapes are always on the move, so the real question is: Where are they going and to what effect?
Ecologists, archaeologists, geographers, urbanists, and philosophers all comfortably speak of assemblages, whether they are talking about remains of buried civilizations, environmental associations of species, or the socio-material assembly of cities. 43 That landscape architects have not embraced assemblage theory is remarkable, given that assemblage — literally, the relational assembly of disparate and heterogeneous things — is the very definition of landscape. In geography and the social sciences, assemblage theory has been more effective than ecological models in analyzing dynamic spatial and political structures.
Manuel De Landa’s distinction between relations of interiority and exteriority is essential here. Although we often conceive landscape as a “body,” this metaphor is deeply problematic when accounting for environmental change. A body requires “relations of interiority,” that is, relationships between parts — lungs, heart, spinal cord, etc. — that are largely fixed and fully dependent upon one another. If we extract one item from the body, the assembly ceases to function. Generally, landscapes do not behave like this; they are far more elastic and persistent. Assemblages are characterized by “relations of exteriority” that are not predetermined or strictly defined; these relations are loose and “contingently obligatory.” Being a component of an assemblage may “exercise a part’s physical capacities but is not an exhaustive or totalized property of it.” 44
We can observe periods of stability as well as critical moments of change when a new assemblage forms.
A landscape’s capacity for qualitative and material change far exceeds that of a body. Things flow in and out of landscapes interchangeably, and their movements indeterminately affect the qualities of the whole. As such, landscapes are vibrant material ensembles in which agencies are fluidly and unevenly distributed within a spatial medium. 45 The semi-autonomy of parts does not negate a deeply relational (i.e. ecological) conception of reality; it merely obliges us to investigate more specifically the contingent and dynamic nature of those relationships. Examining landscapes along a historical trajectory, we can observe periods of stability as well as critical moments of change or bifurcation when the assemblage is ruptured and a new assemblage forms. Here De Landa applies the Deleuzian term deterritorialization. We might also speak of landscape as a multiplicity, inherently plural, as theorist Rod Barnett does in explaining landscape emergence. 46
These models emphasize a more democratic distribution of processes of becoming, and thus they make it difficult to locate agency within a diverse field of actors. But this is not a flaw in the models; indeed, it puts them closer to reality. Panarchy theory was developed as a tool to better assess how, where, and when to intervene in landscape performance, by observing patterns and trajectories of movement and then actively tinkering with those trajectories. Although highly inclusive of human and non-human vectors of change, Panarchy holds that “the great complexity, diversity, and opportunity in complex regional systems emerge from a handful of critical variables and processes that operate over distinctly different scales in space and time.” 47 In other words, the parts of a landscape are not equal in their ability to maintain or destabilize the whole. Water, for example, exercises some of the most potent interactive capacities of all materials, and is thus critical to landscape structure. Understanding how landscapes are assembled is closely aligned with what we do as environmental designers and land managers. It also leads us to the critical and political question of choice. Which parts and processes of landscape do we choose to engage, and which courses of action do we take?
Infrastructure is where specific landscape assemblies are chosen and articulated, and it is here that design can play its most influential role. 48 J.B. Jackson famously defined landscape as “a composition of [hu]man-made or [hu]man-modified spaces to serve as infrastructure or background for our collective existence.” Infrastructure is not just engineered hardware but software and substrate. 49 Bruno Latour later expanded Jackson’s collective to include non-humans, an idea that resonates within political ecology, speculative realism, and posthumanites. 50 Together, they point toward a broader definition of infrastructure as the critical components that structure landscape’s form and its performance. Pierre Bélanger takes this idea to its logical conclusion, landscape as infrastructure, seeking to “redefine the conventional meaning of modern infrastructure by amplifying the biophysical landscape that it has historically suppressed.” 51
As the manifestation of a design choice, infrastructure is contextual and political.
As the manifestation of a design choice, infrastructure is contextual and political, rather than universal or given. 52 In the Klamath Basin examples, we see how infrastructure is enacted, performed, and contested, rather than simply made. We must be skeptical of all apolitical infrastructural design schemes, which do not actually exist. Instead, we must ask: What is privileged in infrastructure’s assembly, and what is left out? Who does the choosing, and who or what benefits? Anthropologist Ashley Carse investigated the question of why the Panama Canal watershed is richly forested and found that government policies encouraged forests that would retain water and supply the canal’s locks, at the expense of other land uses. He observed, “As a landscape becomes infrastructure for one system of production, rather than another, a different group of environmental services (purposefully selected from a multiplicity of possibilities) becomes relevant.” 53 In the Klamath Basin, the Bureau of Reclamation and the Fish and Wildlife Service embodied oppositional notions of landscape infrastructure; the Bureau had far more power and funding and so dictated the terms. The design of infrastructure articulates — and, for a time, fixes — a particular material and socio-ecological construction of the landscape. It thus determines how the landscape will move and behave in the future.
Climate Change as Accelerated Migration
The Anthropocene presents us with a quandary. The more influence we exert upon environments and the atmosphere, the more the earth surprises us by accelerating away from known states. At the moment when the human role in ecological crisis is clear, it becomes impossible to hold the environment or nature at a distance, stranger as it is. 54 In the Anthropocene, we are coauthoring reality even as it becomes more novel and foreign. 55
One of the Anthropocene’s defining characteristics is the shifting around of materials and processes at exponentially accelerated rates. 56 Within this great acceleration, migratory landscapes can be seen all around us. Biologically, we have shuffled the locations of species throughout the world, creating diasporas of naturalized, exotic, and invasive entities, while also engendering the sixth great extinction. 57 We have both impoverished and enriched landscape’s biological structure, rendering most ecological assemblages as radically different. 58 And these altered migration patterns extend far beyond biological entities; they include the very mineral and elemental substrate of landscapes. 59 Since the invention of ammonia fertilizer in 1910, humans have been extracting nitrogen from the atmosphere and circulating it through environments, where it is absorbed by our bodies, so that now fully half the nitrogen in our organs and muscles can be attributed to processes of industrial migration. 60
Which brings us to the most important example of all. Draw in a breath. The air you breathe today contains nearly 40 percent more carbon dioxide molecules than 18th century air. 61 The primary cause of global warming is the migration of massive quantities of carbon, formerly stored underground, into the atmosphere. This is the largest change we have instigated yet.
In fact, we can think of human-induced climate change as accelerated landscape migration. The movement of carbon induces a cascade of attendant motions, leading to rapid landscape change at scales we can perceive and experience. Every decade, biomes migrate approximately 3.8 miles toward the earth’s poles; spring events, such as the flowering of fruit trees, occur 2.5 days earlier. Water melting near the poles redistributes itself throughout the oceans. Rising seas and powerful storms transform and relocate coastlines.
As we begin to understand climate change as a migratory phenomenon, we must emphasize that the migration of landscapes over the next century will not occur in a unified, determinate manner, but rather through mutable and contingent assemblies. As a report for the 2013 National Climate Assessment states:
Range shifts will result in new community assemblages, new associations among species, and promote interactions among species that have not existed in the past. Changes in the spatial distribution and seasonal timing of flora and fauna within marine, aquatic, and terrestrial environments can result in trophic mismatches and asynchronies. Novel species assemblages can also substantially alter ecosystem structure and function and the distribution of ecosystem services. 62
Landscapes will be reassembled with unknown and potentially disastrous effects. Under these regimens of accelerated change, even the notion of a species loses meaning, as observed in the emergence of vigorous hybrid fauna spawned by habitat shifts, such as the coyote-wolf, the lynx-bobcat, and the polar-grizzly bear. 63
How we engage these accelerated landscape migrations is the preeminent challenge of our time.
How we engage these accelerated landscape migrations is the preeminent challenge of our time. Design responses vary but generally fall into two broad categories: those that attempt to adapt to and engage with rapidly changing landscapes (resilience), and those that seek to counter or reverse processes of global warming through deliberate, large-scale intervention (geoengineering).
Among the former, we have examples such as the recent competition Rebuild by Design, which sponsored interdisciplinary design schemes to enhance resilience of the U.S. North Atlantic seaboard after Hurricane Sandy. Similarly, InfraNet Lab’s project Next North produced speculative interventions for Canada’s arctic and tundra territories, where rates of landscape migration precipitated by global warming are occurring 2 to 4 times faster than in the rest of the world. 64 These interventions locate changes occurring within regional landscapes and seek to find new tactical modes to dwell within them.
Geoengineering flips the wager and attempts to alter the very trajectories of climate change at a global scale. In this category, we have carbon markets and forestry protocols, as well as more radical techniques such as dumping vast quantities of iron or lime into the ocean to increase marine photosynthesis. Carbon sequestration is landscape migration on the most abstract, deliberate, and grandest scale — a global return migration of elemental material. Other geoengineering schemes, such as solar radiation management techniques, like cloud whitening or stratospheric aerosol spraying, seek to minimize symptoms of global warming rather than address its root causes. 65
Sequestering carbon in forest biomass is a relatively safe terrestrial infrastructure, but many geoengineering schemes read like avant-garde terraforming scenarios. The scale at which they try to alter landscapes and atmosphere renders them susceptible to unknown feedback mechanisms, which could easily lead to unintended, catastrophic changes. Paleo-climatologists have identified multiple historic events during which the earth’s climate radically changed within decades, substantiating this concern. But as we progress further into the climate crisis with no viable solutions, geoengineering becomes ever more plausible. 66
This is the dilemma posed by the Anthropocene more generally: whether to adapt to radically altered landscapes, or to try to reverse them. As Emma Marris puts it: “What is interesting about climate change is that it pits two common [conservation] assumptions against each other: the pristineness myth and the myth of a correct baseline for each area.” 67 Inevitably, solutions will tear through both mythologies, as exemplified by British Columbia’s new experimental forestry practices. During the 60 to 80 year lifespan of a tree, the climate of British Columbia is expected to rise by 3 to 4 degrees. Responding to these projections, the Ministry of Forests created the Assisted Migration Adaptation Trail, planting seeds from 15 different tree species at 48 reforestation sites from northern California to the southern Yukon. The trees were planted outside of their current ranges, and their growth and health are being closely monitored to identify varieties most adaptable to current and future climates. Conservation translocation is the nascent term for this intentional movement of organisms from one site to another, in order “to yield a measurable conservation benefit at the levels of a population, species or ecosystem.” 68 Like geoengineering, these hybrid migration tactics 69 can incite a broad assembly of unintended changes and are thus controversial.
Here the sublime comes to us as the experience of a lack of stability or unified form.
As designers, planners, and land managers, we have no choice but to engage with global warming and accelerated landscape change. Understanding landscapes as dynamic and mobile assemblies can provide us with better access to the real within this zeitgeist. The concept of landscape migration might be applied to advance new modes of design practice and research, and to provoke additional forms of geographic description. Over the past two decades, architects and landscape architects have shown renewed interest in working with landscape’s “dynamism,” “mobility,” “process,” and “flexibility,” as variably discussed and pursued (to varying success) in our “process discourse.” 70 But the added value of migration theory is that we can deal with landscape dynamics as they unfold, rather than as they are theoretically recovered, since change and movement are nearly everywhere augmented via our agency, both deliberate and inadvertent. The same upgrade in thinking we required half a century ago to understand a process like plate tectonics is now needed to properly engage with the intensified movement of landscapes.
By investigating migratory processes, we might better understand where a landscape has been and where it might go, as we articulate trajectories of change and design infrastructures that make or realign these movement patterns. When we physically encounter a landscape, what we see and sense is just a snapshot in a string of historic and future transformations, all unfolding at varied cadences. To gain real traction, we need to know how the assembly has arrived at its present state — one of many possibilities, with us, no doubt, having played some part in it — in order to speculate upon its potential futures. 71 If we move toward a notion of landscape as a plurality of forms, then “belief in a fixed criterion of [evolutionary] optimality disappears,” which enables a broader spectrum of choice and possibilities. It also allows for “real historical processes to reassert themselves once more.” 72 Through such an approach we might begin to dissect and reassemble climate change, as well as the many wicked landscape problems we currently face, including efforts to refashion our many “crisis” regions, such as the Klamath Basin, the California and Mississippi Deltas, and the Florida Everglades.
This is the landscape medium in which we will design.
The experience of landscape migration, like global warming, is an uncanny one. In the great acceleration, the sublime — that feeling of being thrilled, disoriented, and overwhelmed — is not contained in any fixed scene of landscape grandeur but in quicksilver movements and metamorphic capacities. Picture James Belog’s time lapse sequences of the Arctic’s ice melting faster and faster, caught up in autocatalytic processes triggered by a warming climate. Imagine a time lapse of Louisiana’s coastal marshes, where more than half the terrain has literally dissolved in ten years. Think of the multitude of professionals and citizen scientists who are now voluntarily uploading thousands upon thousands of their embodied observations into the National Phenology Network database. This collective diary of earthly observations, from the timing of plum tree blossoms in a backyard to the arrival of tropical migratory birds, is slowly building a picture of systemic environmental asynchronicities. Here the sublime comes to us as the experience of a lack of stability or unified form. New patterns of flux are all around us, and everything is more easily seen as part of the novel and emergent construction that it is. Within the accelerated reshuffling of just about everything, environmental baselines lose traction within an overriding condition of migration. This is the landscape medium in which we will design.