Adaptive Epistemologies and Neo-Wilds — Chapter 04
Draft
Adaptive Epistemologies and Neo-Wilds
Chapter 04
Ecology of Practice
This chapter maps the ecology of practice within which my research is situated and through which it has been produced. Rather than treating the subsequent project chapters as the work of a single, isolated author, I frame them as outcomes of an extended set of communities of practice that span landscape architecture, architecture, engineering, ecology, and environmental science (Lave and Wenger 1991; Wenger 1998). The chapter traces how collaborations with design academics, professional offices, scientists, engineers, and public institutions have provided not only sites and problems, but also methods, tools, and languages that condition what kinds of questions I can ask and what kinds of futures I propose. In doing so, it positions my practice-based PhD as emerging from a hybrid field in which dredging logistics, geomorphic modeling, responsive sensing, territorial design, and adaptive infrastructure are continuously negotiated. This ecology of practice is both empirical and methodological as it documents the networks through which the work travels and it establishes the transdisciplinary ground on which the later chapters that are focused on specific projects and key concepts should be read.
Tracings
When I describe my work as an \"ecology of practice,\" I extend the language of communities of practice into the relational and multi-scalar realm of territorial design. A community of practice, in Lave and Wenger\'s sense, is not just a group of people with similar expertise, it is a configuration of mutual engagement, a joint enterprise, and a shared repertoire of concepts, tools, and stories that evolve through doing things together (Lave and Wenger 1991; Wenger 1998, 2000).
My own ecology of practice spans multiple overlapping communities from design academics experimenting with sediment logistics and technological landscapes to professional offices testing speculative ideas against real projects to scientists and engineers building physical and computational models and to institutions whose regulations and infrastructures quietly frame what can be designed. These are not separate worlds, they are patches in a single, shifting mosaic, an anthropogenic biome of disciplinary cultures, to borrow Erle Ellis\' term for hybrid socio-ecological systems (Ellis and Ramankutty 2008; Ellis 2011).
Writing this ecology autoethnographically means that I do not pretend to stand outside it. I am implicated as collaborator, co-author, consultant, and critic. Autoethnography, understood as the systematic analysis of personal experience to understand cultural experience (Ellis, Adams, and Bochner 2011; Adams, Holman Jones, and Ellis 2015), provides one of the methodological threads of this chapter and my own trajectory, anxieties, and enthusiasms become data about how landscape knowledge is made.
What follows is not a neutral survey but a narrative map of how particular collaborations and projects have shaped my methods and concepts, and how, in turn, those methods and concepts participate in broader disciplinary shifts.
Dredge, Bennetts Creek — USACE, 2013
US Army Corps of Engineers. Public domain.
The Dredge Research Collaborative: Sediment as Medium and Method
One of the most formative communities of practice in this ecology is the Dredge Research Collaborative (DRC), specifically Rob Holmes, Brett Milligan, Sean Burkholder, Brian Davis, and Justine Holzman, among a shifting cast of designers, artists, and advocates. Together, they have spent more than a decade treating dredging not as an invisible engineering routine but as a cultural, political, and designable infrastructure.
The DRC\'s recent book Silt Sand Slurry: Dredging, Sediment, and the Worlds We Are Making consolidates this work, presenting dredging as a \"visually rich investigation into where, why, and how sediment is central to the future of America\'s coasts,\" and as an \"unseen infrastructure\" that \"shapes and enables modern life.\" The book is less a monograph than a collective atlas of fieldwork, mappings, and speculative projects, ports, disposal sites, eroding barrier islands, and vaguely defined \"placement areas\" that quietly organize coastal futures.
Within this collaborative, my role has primarily existed as an invited participant to co-produce within that shared repertoire. My methods working with DRC are research-through-design, developing workshops that explore nascent, responsive infrastructural practices, and staging public events as a primary means of inquiry (Zimmerman, Forlizzi, and Evenson 2007; Lenzholzer, Duchhart, and Koh 2013).
Autoethnographically, the DRC\'s workshops and field investigations produced a specific methodological conversion. Before this relationship, I understood dredging as background logistics, something that happened to rivers and harbors to keep them functional. The DRC\'s sustained attention to \"placement areas,\" the sites where dredged material is deposited, often vaguely defined in regulatory documents, often on the urban edge, often in the process of becoming something that no one has yet named made visible a design problem that the academic literature had not framed as one.
The DRC showed me where to look. The geomorphology table research at REAL, and later at UVA, can be understood as an attempt to develop instruments for what I had learned to see at those sites and the choreography of sediment as a medium of territorial formation, not a waste product of navigation maintenance.
Technological Genealogies: Richard Hindle and Patent Landscapes
In parallel with my sediment work is my ongoing engagement with Richard Hindle\'s research into patents and landscape technologies. Hindle mines patent archives to reconstruct the histories of infrastructural devices such as green roofs, irrigation systems, coastal defense, and retaining structures, often revealing how technological imaginaries and environmental control fantasies are encoded in intellectual property. His contributions show how patents form a spatial archive of environmental design, including unbuilt or forgotten artifacts.
Methodologically, Hindle\'s work offers a form of design-inflected historiography. Before speculating about new sensing infrastructures or hydro-mechanical interventions, I have learned to look back and trace how similar systems have been framed, which problems they claimed to solve, and what hidden assumptions they carried about nature and control. This historical work complements my field and modeling work, grounding my speculative projects in a longer genealogy of environmental technologies.
In autoethnographic terms, learning from Hindle shifted how I enter technical conversations. I am less likely to accept a new technology at face value, instead, I instinctively ask what is the historical lineage of this device? What failure modes and fantasies are already baked into it?
Owens Lake, California — Historic USGS
USGS. Public domain.
Territorial Interfaces: Alexander Robinson and Owens Lake
Alexander Robinson\'s work at the Landscape Morphologies Lab and the Office of Outdoor Research extends my focus from sediment logistics to broader territorial interfaces. His book The Spoils of Dust: Reinventing the Lake That Made Los Angeles examines Owens Lake, once California\'s third-largest inland water body, drained to serve Los Angeles, as a hybrid of infrastructural experiment, environmental justice site, and public spectacle (Robinson 2018).
In Robinson\'s work, the lakebed becomes a testing ground for collaborations between dust control engineers, ecologists, landscape architects, and nearby communities. The work weaves technical drawings and speculative propositions into an atlas that is simultaneously environmental history, design research, and political commentary.
Our conversations and joint events have impressed on me the role of interfaces such as physical test plots, monitoring systems, and mapping conventions as mediators between expertise and public perception. A map or model is not just a communication tool, it elucidates what counts as visible, negotiable, or inevitable. This sensibility directly shapes the way I approach territorial design and visualization in my own projects.
Computational Communities
The ACADIA community (Association for Computer Aided Design in Architecture) forms another crucial node in this ecology. The community has highlighted a distributed practice concerned less with digital form as a style and more with computational processes, environmental performance, and multi-scalar infrastructures.
Within this milieu, I have engaged with Jason Kelly Johnson and Nataly Gattengo through Future Cities Lab, who experiment with responsive installations and constructed prototypes that couple sensing, actuation, and urban atmospheres. Kathy Velikov\'s work on responsive envelopes and territorial architectures underscores the importance of thinking across scales and media, from building skins to regional ecologies. Dana Cupková\'s focus on material computation, waste cycles, and technologies of care foregrounds the social, thermodynamic, and climatic consequences of digital design.
The books produced within this ecology of practice deserve a moment\'s attention as knowledge contributions rather than merely community artifacts. Digital Drawing for Landscape Architecture (2010, with Wes Michaels) addressed a specific crisis, the profession had adopted digital tools wholesale from adjacent disciplines, but the representational conventions those tools encoded such as vector precision, object clarity, architectural rendering conventions were inadequate to landscape\'s constitutive concerns with indeterminacy, atmosphere, and temporal change. The ASLA Award of Excellence jury recognized the contribution, praising the book as a contemporary and accessible treatment of digital drawing for the discipline. What that recognition named was a representational void the book had filled. You cannot design what you cannot represent and the book changed what landscape architects could see.
Codify: Parametric and Computational Design in Landscape Architecture (2018, edited with Adam Mekies) made a stronger argument, that landscape architecture is not merely adopting computational methods from architecture and engineering but is uniquely positioned to define an emerging domain where ecology, urbanization, and technology converge. Richard Hindle\'s response, \"Codify convincingly argues that Landscape Architecture is uniquely positioned to define this sector of technology, and in the process redefine itself\" names the disciplinary ambition. These publications are not merely the outputs of this ecology of practice, they are interventions within it, reshaping what questions the community can ask.
In ACADIA and related venues, prototypes, scripts, and installations are explicitly framed as research-through-design artifacts, producing designed objects that embody and generate theoretical insights (Zimmerman, Forlizzi, and Evenson 2007; Zimmerman, Stolterman, and Forlizzi 2010). My own autoethnographic record of conference papers, workshop notes, and moments of failure captures the discomfort and excitement of exposing unfinished tools and ideas in public, and watching them be adopted, modified, or critiqued by others.
Publishing Communities and Collaborative Authorship
A distinct thread within this ecology of practice runs through the books and essays produced with collaborators whose intellectual contributions have been formative rather than merely additive. These are not publications that gathered existing knowledge but relationships through which new knowledge was made.
Wes Michaels and I arrived at Louisiana State University in the same year, hired as assistant professors, having come through the same post-professional MLA cohort at the Harvard Graduate School of Design. The collaboration that produced Digital Drawing for Landscape Architecture (2010) was built on complementary strengths, his grounding in design practice, my orientation toward computation and technology, and on the shared conviction that the profession needed representational tools calibrated to its own material concerns, not borrowed wholesale from adjacent disciplines. The book is now in its second edition and remains a textbook across landscape architecture programs. That continuity matters: it is evidence that the representational void the collaboration identified was real, and that the methods developed to fill it have proven transmissible across generations of practitioners.
Justine Holzman arrived as a student, became a research assistant, and became a co-author. The arc of that relationship spans nearly a decade and its intellectual contribution to this dissertation is specific and significant. Her framing of the Modify chapter in Responsive Landscapes (2016) opened a conceptual space that I had circled without naming: that actively altering an environment in real-time constitutes a knowledge-producing act. Modification as epistemology: the landscape changed, and the change was the data. That insight runs as an undercurrent through everything that follows in this dissertation.
Together, Holzman and I organized the Adaptive Devices workshop at DredgeFest Louisiana (2014), where we tested the modify-as-knowledge intuition before the language existed to name it. The workshop staged responsive sediment infrastructures at tabletop scale before the argument had been formalized, before we knew what we were demonstrating. The KERB essay "After Modify" (2015), co-authored with Holzman, was where the formalization happened: the workshop's intuitions became articulated claims. And her independent chapter in Codify (2018), published in the volume Adam Mekies and I edited, extended the argument further, hinting at the landscape itself as a model, a claim the dissertation later develops more fully. This is what a community of practice looks like when it is generative rather than merely collegial: an idea moves through a workshop, into a co-authored essay, and returns transformed in an independent voice.
Adam Mekies represents a different but equally sustained collaboration. Our relationship begins at Sherwood Design Engineers, where theoretical frameworks developed in academic contexts are tested against engineering constraints and institutional realities of professional practice. But it extends into the intellectual register through Codify, which we co-edited and opened together. The Dredgefest encounter and the Codify essay are two moments in a longer conversation about what it means to think computationally about landscape rather than merely use computation in landscape work.
Student Communities
No account of this ecology of practice is complete without acknowledging the communities of students whose work has been generative rather than merely assistive. Three registers of relationship deserve recognition, each producing a different kind of knowledge contribution.
Research assistants, working at the edges of funded projects and laboratory investigations, have extended the capacity to prototype, test, and document. Their contributions are embedded in the technical infrastructure of the work, in the sensing systems, the software scripts, the image archives, often without appearing in citations or credits.
Students in studios, seminars, and thesis work have pushed ideas into territory the laboratory could not anticipate. The student projects cited throughout this dissertation were not illustrations of pre-existing frameworks but inquiries that opened new directions. Studio work operates through a productive asymmetry: the instructor sets conditions, but the outcomes exceed the conditions, and the excess is where the knowledge is.
PhD advisees have constituted the most reciprocal relationships, collaborations in the fullest sense, where advising and being advised are not cleanly separable. Marantha Dawkins, whose research on climate and landscape culminated in a co-authored paper in Prospectives (the Bartlett BPro online journal), brought a sustained focus on environmental justice and the political dimensions of responsive systems that sharpened the dissertation's own justice arguments. Xun Liu, co-collaborator on the UVA geomorphology lab research and the Venice Biennale's Indeterminate Futures (2021), has been a consistent intellectual interlocutor in the transition from responsivity to autonomy, and co-author with Zihao Zhang and myself on a chapter in the Routledge Handbook on Artificial Intelligence in Architecture. Zihao Zhang, co-author on that same chapter and on the essay "Cultivated Wildness: Technodiversity and Wildness in Machines" published in Landscape Architecture Frontiers (2021), is the collaborator with whom the concept of the Third Intelligence was first formally framed, a concept that now anchors Chapter 11. That the dissertation's central claim about non-human agency in design emerged through collaborative inquiry with students-become-peers is not incidental. It enacts the argument.
Theoretical Interlocutors
Beyond the Waldheim and Raxworthy framing that closes the design-academic world, the ecology of practice that produced this dissertation has been shaped by a wider set of theoretical interlocutors whose ideas and relationships have been equally formative.
Christophe Girot's work sits at the intersection of terrain perception and digital computation. His more recent work with point clouds and digital terrain at ETH Zürich argues that new sensing technologies can recover attentiveness to the specificity of place that homogenized ecological planning methods have suppressed. The Codify foreword he contributed names this directly: coding, at its best, is a return to the need to inscribe essential meaning in our daily lives on the ground we tread upon, a counter to the flattening of landscape into two-dimensional scientific abstraction. Girot's insistence that computational methods must remain answerable to terrain's material and cultural specificity has been a persistent check on the more totalizing ambitions of responsive systems thinking.
Elizabeth Meyer's work on aesthetics and performance in landscape architecture grounds this dissertation's understanding of what design is for. Her argument, developed across "The Expanded Field of Landscape Architecture," "Sustaining Beauty," and related essays, is that experiential and ecological dimensions of landscape are not separable. Aesthetic experience is not decoration applied to ecological performance but the medium through which ecological value becomes legible and meaningful to those who inhabit a place. This insistence on the constitutive role of experience runs beneath this dissertation's concern for cultivated wildness and multi-species co-authorship: if the landscape's autonomy is invisible or inert to human perception, it cannot function as a site of ethical or epistemological encounter.
Robert Pietrusko and I overlapped at Harvard during the years of the REAL lab's development. His argument, that data does not neutrally record the world but produces the categories through which the world becomes known and designable, provided theoretical grounding for the sensing work at REAL and UVA. Together we developed a proposal for an MDeS program at Harvard focused on Data Ecologies, the design of data acquisition, processing, visualization, and application as a coherent disciplinary practice. The proposal was not realized, but the conversation it required crystallized the argument that the design of sensing infrastructure is a form of territorial design, a claim that runs through the technogeographies work in Chapter 7.
Nicholas de Monchaux and I were both fellows at the American Academy in Rome, and it was there that he articulated something that reoriented how I understood the dissertation's central claim: that the landscape is the medium. Not the model of the landscape, not a representation of it, not a surrogate, but the landscape itself as the medium through which knowledge is produced and tested. His Local Code, which uses parametric tools to locate distributed urban remnants and design them as collective ecological infrastructure, enacts this principle at urban scale, treating the city's actual terrain as the computational and design medium rather than abstracting it into a model.
Theoretical Bookends
Two theorists quietly bookend this design-academic world with Charles Waldheim and Julian Raxworthy.
Waldheim\'s essay \"Strategies of Indeterminacy in Recent Landscape Practice,\" written in the context of Downsview Park, articulates a mode of distanced authorship in which landscape architects design frameworks and processes rather than fixed forms (Waldheim 2001). This concept gives a name to design with sediment logistics, AI-mediated management, or responsive infrastructures and I am setting up conditions, rules, and feedback loops that will enact, over time, a design that no one fully controls.
Raxworthy\'s book Overgrown: Practices Between Landscape Architecture and Gardening insists on gardening, continuous, embodied maintenance, as central, not peripheral, to landscape architectural practice (Raxworthy 2018). He argues that design cannot be separated from the temporal, vegetal, and labor-intensive realities of care. This insight haunts my collaborations and when I work with geomorphological models, view a sediment test plot, or explore a heavily engineered coastal project, I see them through Raxworthy\'s lens as complex gardens requiring long-term tending. The technologies we deploy should not replace gardening practices, instead they are part of an expanded repertoire of wetware and reflexive stewardship. And this is the place where this dissertation departs from Raxworthy toward its own theoretical territory, defining the cultivant, as developed in Chapter 11, is my extension of the viridic, living matter is a growing medium, but also the ongoing relationship between designed intention and biological agency, enacted through cycles of maintenance, constituting a form of communication that accumulates knowledge over time (Raxworthy 2018; see Chapter 11).
beach renourishment dredge virginia beach 2013
Design Practice Tension
Process-Driven Urban and Territorial Design
In professional practice, a sustained collaboration has been with Stoss, led by Chris Reed in Boston. Stoss\'s body of work is often cited in discussions of landscape urbanism and process-based design and their projects emphasize phasing, open-ended frameworks, and infrastructures that invite adaptation over time.
My role with Stoss has been as an embedded researcher and consultant. I have participated in early concept phases, where we introduce responsive strategies and sensing methods that must survive contact with engineers, contractors, and permitting agencies.
This is a form of research through practice and iterative theorization where projects become testbeds for ideas about responsive landscapes, infrastructural ecologies, and adaptive urbanism. These projects weave inquiry, strategy, and design in ways that produce knowledge about how certain strategies have agency in the world (Deming and Swaffield 2011; Swaffield and Deming 2011).
Territorial Application of Theory
More recently, my collaboration with Sherwood Design Engineers and in particular with Adam Mekies in their New York office, has become a crucial part of this ecology of practice. Sherwood describes itself as an engineering firm working \"at the forefront of ecological infrastructure and urban systems design,\" specializing in climate-responsive planning, smart-city infrastructure, and performance-driven modeling. Mekies, an associate principal at Sherwood and my co-author on Codify, is a landscape architect whose work focuses on the role of computation and construction in environmental and ecological design.
With Sherwood, the scale and stakes of my speculative ideas intensify. Projects range from large-scale territorial landscapes that address coastal protection, regional infrastructures, and urban districts, to focused studies of specific technologies. Questions I have explored in academic and design contexts regarding sediment logistics, AI-mediated management, or infrastructural ecologies are translated into deliverables that must survive engineering review, public scrutiny, and financial analysis.
The work with Sherwood has allowed me to interface with the building of multi-scalar models that integrate hydrology, energy, and land use with spatial design scenarios focused on adaptation. It has also developed computational workflows in the form of scripts, parametric models, and data pipelines that allow engineers and designers to test variants under different environmental scenarios. In some of the most enlightening aspects of this collaboration, scenario planning engages governmental agencies and stakeholders by using speculations and models to facilitate discourse around infrastructural futures.
In these projects Sherwood\'s team of civil engineers, computational designers, and landscape architects puts my theoretical developments under stress. An idea like \"cultivated wildness\" must be translated into performance metrics and management regimes and \"wetware\" becomes technical specifications justified by ecological performance.
The NEOM consultation (2022–25) is the sharpest instance of this theoretical stress-testing. Hydrological modeling using GeoHECRAS across multiple storm frequencies revealed that conventional channelization of the wadis surrounding The Line would require infrastructure widths exceeding 200 meters with extensive hardened concrete at velocities that made ecological function impossible. The finding was not a design preference but an engineering limit. The alternative was a hybrid hydrological approach that engaged managed complexity rather than minimized it, routing water through reconceived wadi systems as holding areas, recharging aquifers, sustaining mangroves through fluctuating isohaline zones an outcome that was not theory imposed on the project but application demanded by its constraints.
This is the methodological contribution of professional collaboration that academic practice cannot replicate. The concept of \"managed complexity\" does not mean the same thing in a research seminar and in a GeoHECRAS output showing 200-meter infrastructure widths. Working with Sherwood sharpened the argument by forcing it to survive contact with the conditions it claimed to address. The friction was productive precisely because it was uncomfortable and it required translating \"cultivated wildness\" into a language that engineers could evaluate, which in turn required clarifying what the concept actually claimed.
Autoethnographically, my experience with Sherwood oscillates between exhilaration and discomfort. It is heartening to see speculative concepts about sediment as infrastructure, AI as co-designer, or computational wildness circulate in engineering drawings and policy briefs. It is also uncomfortable to see how quickly those concepts can be simplified, instrumentalized, or resisted. But this friction is methodologically productive as it forces a refinement of language, clarifies assumptions, and confronts the institutional realities that theoretical frameworks must navigate.
The work with Sherwood Design Engineers and Stoss is where theory meets productive tension and where ideas formed in research are either flattened by cost and risk or adapted and strengthened.
Co-Authoring Ecologies
AI-Mediated \"Wildness\"
My collaborations with Erle Ellis, a landscape ecologist known for his work on anthropogenic biomes, move my practice into new disciplinary territory. Ellis\'s classification of the terrestrial biosphere into human-modified biomes makes visible the extent to which ecological patterns are already shaped by long-term human land use (Ellis and Ramankutty 2008; Ellis 2011).
In joint projects and writing, with environmental historian Laura Jane Martin and technologist David Klein, we ask what it would mean to design AI-mediated systems that manage these anthropogenic biomes toward forms of perceived wildness that are orchestrating disturbance, succession, and access patterns in ways that privilege ecological complexity and more-than-human claims, while remaining legible to human governance. These are speculative exercises that test how an \"agent\" might govern land management decisions according to different objectives.
The Designing Autonomy paper, published in Trends in Ecology & Evolution, a leading journal in ecological science, not a landscape architecture venue, is itself an artifact of this ecology of practice. That publication required the Ellis collaboration, the Martin collaboration, and the willingness to place landscape architecture\'s claims in a scientific journal that hadn\'t encountered them before. One of the dissertation\'s arguments is that landscape architecture is uniquely positioned to define an emerging domain at the intersection of ecology, technology, and territorial governance and the fact that this claim could be published and received in Trends in Ecology & Evolution is evidence that the ecology of practice that produced it extends beyond landscape architecture\'s own disciplinary boundaries.
The projects are instances of research through design operating at the scale of intractable planetary issues and design scenarios and models are not only visualizations but narratives to interrogate how \"wildness,\" control, and equity are encoded in machine intelligence.
Rivers in the Laboratory
With civil engineer Clint Wilson at Louisiana State University, my research has evolved through the friction that emerges when speculation and engineering come into contact. Our work overlapped within the Louisiana State University Coastal Sustainability Studio, where a range of projects from design competitions to historical preservation assessments allowed me to test beginning intuitions about modeling alongside one of the leading scholars in civil engineering.
From Wilson\'s standpoint, physical models are aimed at understanding and predicting geomorphic behavior under different flow regimes and engineered configurations. From my point of view, they are generators of design heuristics through tangible testing of river cross-sections, levee alignments, cut-off channels, and sediment diversions, metaphorically sketching with water and sand.
The Sedimachine experiments at LSU (2012) crystallized this distinction. When the Microsoft Kinect depth sensor failed to resolve the depositional layer as the sediment was too thin for the instrument\'s resolution to capture meaningful topographic variation, Wilson and I drew different conclusions. From a hydraulic engineering standpoint, the failure indicated a technical limitation requiring either better equipment or a different experimental setup. From my standpoint, it indicated something else in that the plexiglass substrate and controlled flow conditions were producing phenomena at a spatial scale that exceeded the instrument\'s resolution, which meant that the sediment choreography must rely on the expression of the outcome (morphological results) as pattern rather than particle. The failure redirected subsequent research toward the EmRiver geomorphology table and toward ultrasonic range-finding and image analysis as complementary sensing modes.
This interpretive divergence is not a miscommunication between disciplines, it is how communities of practice with different purposes produce different knowledge from shared material. The ecology of practice depends on this divergence remaining in productive tension rather than resolving toward either purely scientific or purely design purposes. The geomorphology table research at REAL and UVA has maintained this tension deliberately, the table is not a hydraulic engineering instrument, but it draws on hydraulic engineering knowledge to generate landforms that are read as design material.
The methodology aligns with research through designing, except that designing is delegated to physical processes (Lenzholzer, Duchhart, and Koh 2013). The models produce time-based traces in the form of photos, videos, and point clouds that are aligned with scientific data to produce design material. Reading them together, Wilson is looking for predictive patterns and scaling laws while I am looking for emergent landforms that might host futures for settlement, habitat, or public access.
Geomorphology and Design Labs
With geomorphologist Ajay Limaye at the University of Virginia, I co-developed research that pairs geomorphic modeling with design experimentation. Limaye\'s work uses numerical and physical models to understand channel networks, delta morphodynamics, and river planforms.
During our work together, the collaboration aimed to build a stable workflow where design students engage directly with geomorphic models, modifying boundary conditions, exploring interventions, and interpreting outputs as design prompts. This led to the creation of physical models and computational tools that unpack assumptions about settlement patterns, infrastructure, and ecological restoration in terms of sediment budgets, flow regimes, and morphological thresholds.
The collaboration was, in effect, a designed community of practice, with student research assistants at both undergraduate and graduate levels and where geomorphologists, landscape architects, and students learn to wield cross-disciplinary tools and methods. There is an insistence in practice-based methods that landscape research should operate across the case study, lab experiment, and design research (Deming and Swaffield 2011), and this collaboration is a clear instantiation of that insistence.
Boundary Encounters
My engagement with institutions like the U.S. Army Corps of Engineers, The Nature Conservancy, and the Environmental Protection Agency is more episodic but no less formative. These agencies manage and regulate many of the landscapes where my work has relevance.
Methodologically, work with these institutions involves:
Policy and protocol analysis through the reading of dredge management plans, environmental impact statements, and design guidelines as design and governance documents rather than neutral technical texts.
Participatory and consultative practices within workshops, stakeholder meetings, and site visits where community concerns, engineering constraints, and design ambitions collide.
Boundary work, producing conceptual and visual \"boundary objects\" in maps, scenarios, and diagrams that can be understood across professional cultures.
These engagements echo the argument that participatory landscape ecology is essential if landscape planning is to be accepted and implemented, because it must weave scientific, professional, and local knowledge together (Luz 2000). In my own experience, these are also emotionally charged spaces where power, frustration, and institutional memory are palpable. Autoethnography allows me to acknowledge that my sense of caution, hope, or skepticism in these rooms is part of the research and it shapes the questions I ask and the compromises I am willing to entertain.
In terms of communities of practice, these institutions might be best understood as boundary partners rather than core communities. They are places where the repertoires developed with the DRC, ACADIA, Stoss, Sherwood, Ellis, Wilson, and Limaye are tested against formal governance systems. Wenger\'s later work on \"landscapes of practice\" is helpful here as my ecology of practice is one strand within a larger landscape of overlapping, sometimes misaligned communities that nonetheless share concerns around coasts, rivers, and urban systems (Wenger-Trayner and Wenger-Trayner 2014).
Situating the Ecology of Practice
In this section I step back, placing these collaborations in a wider disciplinary context.
From Control to Adaptive Infrastructures
Classical civil engineering, especially in hydrology, has been shaped by an ethos of control producing levees designed to historical flood stages, sediment removed to maintain navigation channels, rivers straightened to maximize conveyance. My collaborations with Wilson, Limaye, Sherwood, and the Corps participate in and push against a gradual shift toward adaptive and process-based infrastructures.
Urban ecological frameworks emphasize that the dynamics of infrastructures, ecological processes, and socio-economics must be linked conceptually and practically (Pickett et al. 2001, 2011). My ecology of practice attempts to answer this prompt in situated ways, through physical models that are read both as experiments and design sketches, AI tools judged both by their predictive accuracy and by the imaginary governance they encode, and territorial propositions evaluated both on hydraulic metrics and on socio-ecological criteria.
Landscape as Infrastructure, Garden, and System
Within landscape architecture, this ecology of practice touches and extends several key threads:
Landscape urbanism and strategies of indeterminacy (Waldheim 2001; see Theoretical Bookends above), a focus on processes, frameworks, and open-ended programs, manifests in adaptive sediment regimes, experimental river models, and responsive coastal systems.
Deming and Swaffield\'s articulation of landscape architectural research legitimates this practice as research and the combination of case studies, design experiments, and collaborative projects is not an ad hoc mash-up but a recognized mixed-methods strategy (Deming and Swaffield 2011; Swaffield and Deming 2011).
From Objects to Fields to Ecologies
In architecture and digital design, my work aligns with Stan Allen\'s concept of field conditions, spatial systems organized by local relationships and gradients rather than singular objects (Allen 1999, 1999/2013). The computational and sensing infrastructures I work with are literally fields as arrays of sensors, grids of diversions, mosaics of patches, and aggregations of robotic infrastructures.
ACADIA sits at the center of this shift, and my contributions are part of a larger movement in which architects, landscape architects, and engineers use research through design as a method to engage with climate, energy, and ecological questions. The articulation of research through design in HCI provides a methodological bridge that offers criteria for rigor (relevance, transferability, theoretical contribution) that I adapt to landscape contexts (Zimmerman, Forlizzi, and Evenson 2007; Zimmerman, Stolterman, and Forlizzi 2010).
Ecologically, my work is grounded in two frameworks. First, anthropogenic biomes frame nearly all of the landscapes I work in-from dredged estuaries to responsive infrastructures as hybrid socio-ecological systems (Ellis and Ramankutty 2008; Ellis 2011). Second, patch dynamics conceptualizes landscapes as mosaics of patches at different stages of disturbance and succession (Pickett and White 1985; Pickett et al. 2001).
The sediment management, the floodplains that are nudged in adaptive trajectories, and the sensor networks are all patchy in this sense. Physical and computational models in collaboration with engineers, as well as territorial plans with Sherwood, work on these mosaics by changing disturbance regimes, creating new patch types, or altering connectivity. My ecology of practice is, then, an autoethnography of working inside patchy anthropogenic biomes, using design to steer trajectories rather than to reset baselines.
Environmental Governance and Infrastructural Ecologies
In subsequent chapters, I build on this ecological and infrastructural grounding to develop key concepts of intelligence, technogeographies, adaptive epistemologies, cyborg ecologies, and wetware, showing how the projects emerging from this ecology of practice contribute to reimagining landscape architecture\'s role in a rapidly changing world.
##
If the practice is the research instrument, then what exactly has the practice done? What is the accumulated body of work, and how does each project build on the one before it? What were the instruments, the provocations, the failures that redirected the inquiry? And what theory has the practice been enacting all along?