Charles Darwin's last book — The Formation of Vegetable Mould, through the Action of Worms (1881) — was not about evolution. It was about how earthworms reshape landscapes. Darwin spent 40 years measuring how much soil earthworms moved, burying objects, plowing fields from below, and creating the topsoil structure that makes agriculture possible. He estimated that earthworms in English farmland moved 18 tons of soil per acre per year — turning the entire top layer every few decades. He placed stones on his lawn and measured how far they sank over the years. Methodical. Obsessive. Revolutionary.
What Darwin described without naming it is the founding experiment of ecogeomorphology — the study of how organisms modify landforms, and how landforms in turn constrain organisms. The field wasn't formalized until the 1980s, when researchers began systematically studying biogeomorphic processes: bioerosion (organisms breaking down rock and substrate), bioprotection (organisms shielding surfaces from erosion), bioconstruction (organisms building structures that accumulate sediment), and bioturbation (organisms mixing sediment).
The UC Davis Center for Watershed Sciences course (Dr. Carson Jeffres, Dr. Sarah Yarnell) applied these principles to California, Oregon, and Washington watershed systems — the rivers, estuaries, and floodplains of the Pacific states. The course integrates fluvial geomorphology, riparian ecology, and watershed hydrology into an applied science for conservation and restoration. The same principles that govern beaver dams in the Sierra Nevada govern mangrove accretion along the Mozambique Channel coast and salt marsh expansion on Puget Sound.
A river is a sediment transport system. Water flows downhill; it carries sediment; it deposits sediment where its energy decreases. The fundamental equation is Lane's balance: the relationship between discharge (Q), slope (S), sediment load (Qs), and grain size (D). When you change one variable, the river adjusts the others to re-establish equilibrium. This is why dams cause downstream erosion (sediment trapped in reservoir; river below has excess energy, erodes its bed), and why removing dams causes temporary downstream sediment pulses.
The three river morphology types that matter most for ecogeomorphology: braided (high energy, high sediment, wide shallow channels with shifting bars), meandering (lower gradient, single-thread channel with point bars and cut banks, lateral migration over decades), and straight/confined (canyon rivers, bedrock control). Pacific Northwest rivers tend to be hybrid systems — steep gradients, episodic disturbance from landslides and debris flows, and strong biotic modification by large woody debris and salmon.
Stream order (Strahler classification) provides a framework: headwater streams (1st–2nd order) have steep gradients and control fine sediment delivery; mid-order streams (3rd–5th) have floodplain development and strong riparian interaction; large rivers (6th+ order) have broad alluvial valleys and complex channel networks. The Coquille, Umpqua, Klamath, Trinity, Sacramento — all PNW/California river systems — follow this gradient from mountain headwaters to estuary. Understanding stream order tells you what kind of ecological interventions work where.
Ecogeomorphologists distinguish between geomorphically-dominated phases and biogeomorphically-dominated phases in landscape evolution. The concept of the "biogeomorphic window" — developed by Stallins (2006) and Corenblit et al. — describes the conditions under which biological processes can control landform development rather than being controlled by it. Outside the window (extreme disturbance, high energy, fast geomorphic processes), organisms are passengers. Inside the window (moderate disturbance, stable enough for colonization), organisms become drivers.
On a gravel bar in a braided river, the biogeomorphic window opens when flood disturbance decreases enough for pioneer plants (willows, cottonwoods, alder) to colonize. Their roots stabilize sediment; their stems trap more sediment; the bar builds; the window widens; more diverse species establish; the bar becomes an island, then a floodplain terrace. Remove the pioneer species — by herbivory, drought, or human intervention — and the window closes; the geomorphic processes reclaim the bar. This succession-disturbance feedback is central to understanding riparian restoration.
Complex systems themes in ecogeomorphology (from Stallins 2006): multiple causality (multiple processes operating simultaneously), ecological memory (past states encoded in current form), ecological topology (spatial variation in species-process relationships), and ecosystem engineering. These themes parallel complex systems concepts in GIS and spatial analysis — non-linear feedbacks, threshold behavior, path dependency, scale dependence. Ecogeomorphology is inherently spatial and inherently complex.
Before European contact, 60–400 million beavers occupied North American watersheds from the Arctic to the Rio Grande. They built an estimated 15 million dams, creating millions of acres of wetland, slowing rivers, raising water tables, and storing more water in the western US than all current reservoirs combined. Beaver trapping for the fur trade (1600–1850) essentially dewatered the West — removing the organism that had been the primary hydrological engineer for 3 million years. The arid, erosion-prone landscapes Europeans documented as "natural" were already profoundly altered landscapes.
A beaver dam does several things simultaneously: raises local water table (critical in the dry West, where riparian vegetation depends on shallow groundwater); creates pond habitat; traps sediment (reducing downstream turbidity); slows stream velocity (reducing peak flood flows); and creates fire refugia (wet areas that don't burn in wildland fire). The riparian zone behind a beaver dam is among the most productive ecosystems in temperate North America — the "oasis effect." Moose browse willows; moose create more willows (through defecation and trampling); willows feed beavers; beavers create more ponds; ponds grow more willows. A positive feedback loop.
Beaver dam analog (BDA) restoration — building structures that mimic beaver dam effects using posts and natural materials — is now a mainstream low-cost restoration technique in the intermountain West. Projects on degraded stream reaches in the Methow Valley (WA), Elko County (NV), and dozens of Oregon sites show rapid channel narrowing, water table recovery, and riparian vegetation recovery within 2–5 years. The question is always: can you reintroduce actual beavers? Where willows are available, they are far more effective than human-built structures.
In old-growth Pacific Northwest forests, fallen trees enter streams and restructure the entire channel. Large woody debris (LWD) — logs greater than 10cm diameter and 1m length — creates pools by deflecting flow and scouring channel beds, traps gravel spawning habitat for salmon, creates hydraulic complexity (alternating fast and slow water), retains leaf litter and organic matter (the base of the stream food web), and dissipates flood energy. A single large log can create a pool system that persists for decades and concentrates salmon spawning activity for its entire lifespan.
Historical logging in the PNW removed both the riparian trees that would have fallen into streams and actively "cleaned" streams of existing LWD under the mistaken belief that it impeded fish migration. The result was simplified, "channelized" streams with less hydraulic complexity, degraded spawning habitat, and reduced invertebrate diversity. The science reversing this — showing that LWD is not an obstacle but a fundamental structural component of PNW river ecosystems — took 30 years to filter from research to policy. LWD placement is now a standard restoration technique.
The nutrient connection: Pacific salmon return marine nitrogen, phosphorus, and carbon to freshwater and riparian systems. A spawned-out salmon carcass is a marine nutrient subsidy — bears, eagles, and ravens drag carcasses into the forest; the nitrogen from Pacific fish has been found in tree rings 500 meters from the stream. Remove salmon (via dam blockage, habitat degradation), and the riparian forest becomes nitrogen-limited. The fish feed the trees. The trees fall into the river. The logs create habitat for the fish. The feedback runs in both directions.
Mangroves are the premier coastal ecosystem engineers in tropical and subtropical systems. Their prop roots reduce wave energy, trap fine sediment, and build peat — raising the land surface as sea level rises. A healthy mangrove system can accrete vertically at 2–5mm per year, keeping pace with current rates of sea level rise. Mangrove deforestation (for shrimp ponds, coastal development) exposes the underlying peat to oxidation — releasing centuries of stored carbon and causing the coastline to subside, creating a doubly bad outcome: loss of coastal protection and significant carbon emissions.
Salt marshes perform the same engineering in temperate systems — grasses (Spartina, Salicornia) trap fine sediment and build peat, dissipating wave energy and providing nursery habitat for commercially important fish and invertebrates. Puget Sound's tidal flats and estuaries are salt marsh-dominated — or were, before 70% of the historical salt marsh fringe was diked, filled, or developed. Restoration of Puget Sound tidal marshes is a major conservation priority partly because of their role in juvenile Chinook salmon rearing — the same salmon that orcas depend on.
Oyster reefs — once the dominant habitat structure in many temperate estuaries — function as both bioconstructors and bioprotectors. Live oyster reefs dissipate wave energy (bioprotection), accumulate shell hash that builds reef elevation (bioconstruction), and provide habitat for hundreds of associated species. Chesapeake Bay's oyster population is 1% of its historical level — the loss removed an organism that could filter the entire bay's water volume in a week. The turbidity that followed drove the decline of SAV (submerged aquatic vegetation), which collapsed the habitat base for blue crabs and juvenile fish. One species removal cascading through the entire system.
Tree roots are the primary stabilizing mechanism on steep slopes in humid mountain environments. Root reinforcement — the additional cohesion provided by roots crossing potential failure planes — can increase slope stability by 2–30 kPa, enough to prevent shallow landslides on slopes that would otherwise fail. The critical zone is the root penetration depth: roots must cross the slip surface, typically 0.5–2m depth for shallow translational slides. Larger, deeper-rooted species (Douglas fir, Sitka spruce) provide far more stability than shallow-rooted species (red alder, many grasses).
Forest harvesting creates a temporal window of high landslide risk. Root reinforcement declines over 3–15 years after harvest as roots decay — this lag effect means peak landslide risk occurs years after logging, not immediately. The 1996 storm events in western Oregon and Washington produced exceptional landslide activity concentrated in areas clearcut 5–10 years earlier, documented in multiple GIS-based spatial analyses. Salvage logging after wildfire has the same effect: removing fire-killed trees removes root anchoring before new roots establish, creating a window of elevated mass movement hazard.
Debris flows — hyperconcentrated mixtures of water, sediment, and wood that move at velocities of 5–20 m/s — are the primary sediment transport mechanism in steep mountain channels. They begin as shallow landslides, entrain water and channel sediment, and travel long distances before depositing at channel junctions or valley floors. In the Oregon Coast Range, debris flows occur at rates of 1–10 per km² per century. They are the dominant force shaping hollow topography (concave headwater valleys) in PNW mountains. They also deliver large woody debris to stream channels — making landslides a key link between the terrestrial and aquatic systems.
Pacific Northwest rivers are disturbance-driven systems. Unlike the stable, meandering rivers of the Midwest, PNW rivers experience episodic pulses of sediment and wood from landslides, debris flows, and wildfire that fundamentally restructure channel morphology on decadal timescales. The disturbance cycle: wildfire kills forest → storm delivers landslides and debris flows → debris flows deliver LWD and sediment to channels → channels aggrade (fill with sediment), braid, and shift → pioneer vegetation colonizes bars → forest recovers → fire → repeat. Salmon evolved in this system — they require the channel complexity created by disturbance, not the simplified, stable channels that result when disturbance is suppressed.
The Rogue, Umpqua, Willamette, Deschutes, McKenzie, Sandy, Clackamas, Cowlitz, Nisqually, Carbon, White, Puyallup — these rivers all drain the Cascades into Puget Sound or the Oregon/Washington coast. Each has a distinct character based on geology (basaltic vs. andesitic Cascades, different erosion rates), disturbance history, and degree of human modification (dams, levees, gravel mining, riparian clearing). The Puyallup River drains Mt. Rainier — where lahar risk (volcanic debris flows) adds a dimension of ecogeomorphic complexity found nowhere else in the contiguous US.
Winter steelhead, Chinook, coho, and chum salmon all require different channel conditions at different life stages. Chinook spawn in large-gravel riffles in mainstem rivers; coho use off-channel alcoves and small tributaries; steelhead use high-gradient headwater streams. The distribution of these species across a watershed is a spatial expression of channel morphology — which is itself a spatial expression of geology, disturbance history, and ecogeomorphic processes. Mapping salmon distribution is, in a sense, mapping ecogeomorphic outcomes.
The removal of Elwha Dam and Glines Canyon Dam on the Elwha River (2011–2014) was the largest dam removal in US history at the time — and the most intensively studied ecogeomorphic experiment ever conducted. Researchers predicted what would happen to the 21 million cubic meters of sediment stored behind both dams, how rapidly the river would re-establish its channel, and how fast salmon would recolonize. The results exceeded predictions on almost every metric.
Within months of removal, the Elwha River cut through the reservoir deltas, mobilizing decades of fine sediment downstream. The river re-established a braided channel through former reservoir beds within a year. Cottonwood and willow established on exposed bars within two growing seasons. Coho and Chinook recolonized 70km of previously blocked river within 18 months of removal. The Lower Elwha Klallam Tribe, whose fishing culture had been severed for a century by the illegal dams, harvested the first salmon run in 2012. The river "remembered" its pre-dam form — the geomorphic template was encoded in the valley, waiting for the constraint to be removed.
The Klamath River dams (Oregon-California) are currently being removed in the largest dam removal project ever undertaken — four dams on 400km of river, with 57,000 acres of tribal fishing grounds and critically depleted Chinook and coho runs at stake. The Karuk and Yurok tribes — whose salmon-based cultures were decimated by the dams — were central advocates for removal over 20 years of legal and political battles. The ecogeomorphic recovery of the Klamath will be monitored for decades.
Puget Sound is a glacially carved fjord system — scoured by the Puget Lobe of the Cordilleran Ice Sheet to depths of 300m, then drowned by sea level rise as the ice melted. The resulting complex of basins, sills, and inlets creates extraordinary tidal variation and habitat complexity. The rivers entering Puget Sound — Nisqually, Puyallup, White, Duwamish, Snohomish, Skagit, Stillaguamish — deliver sediment and fresh water that build deltas, sustain eelgrass beds, and support the juvenile salmon rearing habitat that orca populations depend on.
The orca (Southern Resident Killer Whale) population of Puget Sound is critically endangered — 73 individuals remain. Their near-exclusive prey is Chinook salmon. Chinook salmon require: (1) clean cold rivers with spawning gravel — threatened by dam blockage, warming, and fine sediment from roads and development; (2) floodplain and off-channel rearing habitat — lost to levees and tidegate installation; (3) estuarine juvenile rearing habitat — lost to 70% destruction of Puget Sound salt marsh. The orca-salmon-river-estuary-marsh system is a connected ecogeomorphic chain. Protecting orcas means protecting rivers, estuaries, and the geomorphic processes that maintain them.
Nisqually Delta — at the south end of Puget Sound, below the mountain you can see from Lakewood on clear days — underwent the largest tidal wetland restoration on the US West Coast in 2009. Removing 8km of dikes restored 762 acres of tidal marsh and flats. Monitoring since then shows rapid sediment accretion, vegetation establishment, and use by juvenile salmon within 2 years. The restoration is a textbook ecogeomorphic recovery — and it's visible from your commute.
Global mean sea level has risen 20cm since 1900 and is rising at 3.6mm per year — accelerating. By 2100, IPCC projections suggest 0.5–1m of additional rise under moderate emissions, potentially 2m under high emissions. Coastal ecosystems — salt marshes, mangroves, tidal flats — can theoretically keep pace with sea level rise through vertical accretion (sediment trapping and organic matter accumulation). The critical question is whether accretion rates can match or exceed sea level rise rates.
Research from Puget Sound tidal marshes (Morris et al., Thom et al.) shows that healthy marshes with adequate sediment supply can accrete at rates matching or exceeding current sea level rise. But "coastal squeeze" — marshes caught between rising sea levels on the seaward side and human development (roads, levees, development) on the landward side — prevents marsh migration. A marsh that cannot migrate landward will drown in place. The Nisqually, Skagit, and Snohomish deltas all face coastal squeeze to varying degrees.
In southern African coastal systems, climate-driven changes to storm frequency, sediment supply, and sea level are restructuring coastal geomorphology. The Orange River (Namibia/South Africa border) delivers sediment to the Atlantic coast — the same material that built the Namib's barrier islands and dune fields. Reduced river flows from increased irrigation and drought upstream have reduced sediment supply to the coast, contributing to barrier island erosion and shoreline retreat. Land and sea are connected through sediment budgets that climate change is rewriting — and the Namib coast is a clear example of that connection.
Ecogeomorphology is inherently spatial — the feedbacks between organisms and landforms operate at specific locations, across gradients, through connected networks. GIS provides the analytical framework for studying these relationships at watershed and landscape scales. The key datasets: LiDAR-derived digital elevation models (slope, aspect, curvature, flow accumulation, drainage networks); vegetation data (NLCD, Landsat time series, aerial photo classification); hydrological data (USGS stream gages, NHDPlus); and disturbance records (fire perimeters, landslide inventories, flood recurrence).
Slope stability modeling — predicting landslide probability across a landscape — integrates topography, soil type, root reinforcement estimates, and pore pressure models into spatially distributed hazard maps. SHALSTAB (Shallow Landsliding Stability Model) and similar models are applied in the Oregon Coast Range to prioritize areas for road decommissioning, timber harvest restrictions, and emergency response. The model is geomorphic (slope, contributing area, soil cohesion) but the best calibration datasets come from decades of landslide mapping correlated with forest management history — the ecogeomorphic signal in the spatial record.
The Tobler connection: ecogeomorphic processes are spatially autocorrelated — "near things are more related than distant things." A beaver dam's effects are strongest immediately downstream and diminish with distance. Root reinforcement is strongest at the tree and declines with distance from the root mass. Debris flows affect channels directly below initiation sites. Every ecogeomorphic process has a spatial signature — a pattern of influence that decreases with distance and is bounded by topographic barriers. Mapping those signatures is GIS. Understanding the processes that create them is ecogeomorphology.