Reintroduce a predator. Watch the grassland recover. Watch the river change course. Watch the carbon return to the soil. The most powerful climate interventions may not come from technology — they may come from stepping back.
Decarbonarma — June 2026 — 12 min read
There is a concept in ecology called a trophic cascade. It describes the way that a change at the top of a food chain — the arrival or removal of a single apex predator — ripples downward through every level of an ecosystem, reshaping not just species populations but the physical landscape, the chemistry of rivers, and ultimately the amount of carbon locked into soil and vegetation.
The idea is not new to ecologists. But it is becoming urgently relevant to climate scientists, because trophic cascades turn out to be one of the most powerful carbon sequestration mechanisms on Earth — one that costs almost nothing to activate, requires no new technology, and in many cases simply involves allowing animals to do what they evolved to do.
The examples below span three continents. Each tells a version of the same story: when humans removed a predator, a cascade of damage followed that we attributed to other causes. When we return the predator — or simply restore the conditions that allow ecosystems to self-regulate — the carbon comes back.
We have been engineering the climate downward for centuries, not through our emissions alone, but through the systematic removal of the animals that held ecosystems together.
Understanding trophic cascades does not diminish the importance of reducing emissions. But it reframes what we mean by “natural climate solutions.” The conversation has focused on planting trees. The evidence increasingly suggests we should also be thinking about planting predators.
Case study 01
The sea otter, the sea urchin, and the vanishing kelp forests
Pacific Coast — British Columbia, California, Alaska
Trophic cascade — sea otter recovery
🦦
Sea otter returns
Hunted to near-extinction for fur. Recovering along BC coast
⭐
Sea urchins eaten
Urchin populations collapse. Grazing pressure on kelp removed
🌿
Kelp forest returns
Canopy density recovers. Whole marine ecosystem rebuilds
🌎
Carbon sequestered
~12x
Kelp sequesters carbon far faster than terrestrial forest per hectare
Research by Estes et al. (2011) in Science estimated that sea otter recovery along the Aleutian Islands caused kelp biomass increases equivalent to sequestering millions of tonnes of CO². A 2021 study in PLOS ONE valued the carbon sequestration service of sea otters along the California coast at US$408–704 million annually.
The Pacific coast of British Columbia was once fringed with dense kelp forests stretching for hundreds of kilometres. Kelp is among the most productive ecosystems on Earth — growing up to 60 centimetres per day, absorbing carbon dioxide at rates that dwarf those of tropical rainforest per unit area, and providing habitat for hundreds of species from lingcod to humpback whales.
When European fur traders arrived in the 18th century, sea otters were hunted almost to extinction. By 1900 the global population had collapsed from perhaps 300,000 animals to fewer than 2,000. The consequence was not immediately obvious. The kelp forests began to disappear, but the cause was attributed to water temperature, pollution, and fishing pressure. The otter connection took decades to establish.
Sea urchins eat kelp. In a healthy ecosystem, otters eat sea urchins, keeping their populations in balance. Without otters, urchin populations exploded. They grazed kelp holdfasts to the rock, creating what oceanographers call “urchin barrens” — stretches of bare seabed that had once been cathedral forests of swaying fronds.
74%
Decline in kelp forest area on parts of the Northern California coast since 2014
500km
Stretch of BC coast where sea otter reintroduction is actively recovering kelp habitat
4–8
Tonnes of CO² sequestered per hectare per year by healthy kelp forest
Where sea otters have returned — reintroduced along parts of the British Columbia coast from Alaskan populations in the 1960s and 1970s — the transformation has been striking. Kelp forests have regrown. The species that depend on them have followed. And the carbon cycle has re-engaged: kelp absorbs CO² from surface water as it grows, and when kelp fronds break off and sink to the ocean floor, they take that carbon with them into long-term storage.
The challenge on the BC coast is not ecological but political. Commercial shellfish fisheries — sea urchin, crab, abalone — compete directly with recovering otter populations. The otter is simultaneously a conservation success and an economic inconvenience. The carbon value of the kelp it protects does not appear in any fishing industry balance sheet.
The sea otter is doing something no carbon capture technology can yet replicate: turning a barren seafloor back into a forest, at no cost, using nothing but appetite.
Case study 02
The dingo fence and the desertification of Australia
South Australia — The Dingo Fence, 5,614 km
Trophic cascade — dingo exclusion effect
🐕
Dingo excluded
World longest fence (5,614 km) keeps dingoes from sheep country since 1880s
🦘
Kangaroo explosion
No apex predator. Kangaroo numbers surge to 50M+. Overgrazing intensifies
🏡
Vegetation collapse
Native grasses stripped. Topsoil exposed. Shrubland degrades to desert
🌋
Carbon released
Soil carbon lost as organic matter decomposes. Albedo change warms landscape
Research published in Proceedings of the Royal Society B by Letnic, Ritchie & Dickman (2012) compared ecosystem health on both sides of the dingo fence and found dramatically different vegetation cover, small mammal populations, and soil carbon levels. The dingo-present side consistently outperformed on every ecological measure.
The world’s longest fence runs for 5,614 kilometres across the Australian interior. Built in the 1880s to protect sheep from dingo predation, it bisects the continent and has created, entirely by accident, the world’s most striking ecological experiment. On one side of the fence: dingoes. On the other: no dingoes.
The difference is visible from space.
The dingo-free side of the fence supports dramatically higher kangaroo populations. Kangaroos are efficient grazers — in the absence of any predator to make them wary or keep them moving, they concentrate on the most palatable grasses and graze them to the root. Native grass cover declines, the topsoil that grass roots hold in place becomes exposed to wind and sun, and the slow process of desertification begins.
This is not desertification caused by climate change. It is desertification caused by the removal of a predator and the unchecked population growth of its primary prey.
50M
Kangaroos in Australia — roughly double the human population, with no natural predator in most areas
5,614km
Length of the dingo fence — longer than the distance from London to New York
30%
More vegetation cover on the dingo side of the fence in comparable semi-arid zones
Researchers who have compared matched sites on either side of the dingo fence have consistently found that the dingo-present side has more diverse vegetation, more small mammals, better soil structure, and higher soil carbon. The dingo does not just suppress kangaroo numbers — it changes kangaroo behaviour. Prey animals in the presence of predators avoid lingering in open areas, keep moving, and allow vegetation to recover between grazing events. This “landscape of fear” effect is as important as direct predation.
The implications for Australia’s carbon accounts are significant. The continent holds vast areas of degraded grassland and shrubland that have lost substantial fractions of their original soil carbon. Conservative estimates suggest that restoring healthy predator-prey dynamics across even a fraction of this area could sequester tens of millions of tonnes of CO² annually — at negative cost, since the alternative involves expensive and ineffective kangaroo culling programmes.
The dingo fence has been running the world’s longest unintentional carbon experiment for over a century. The results are unambiguous — and we are on the wrong side of the fence.
Case study 03
How wolves change rivers: Yellowstone’s trophic cascade
Yellowstone National Park, Wyoming — Wolf reintroduction 1995
Trophic cascade — wolf reintroduction 1995
🐺
Wolves return
14 wolves reintroduced from Canada. Population grows to 100+ within a decade
🦌
Elk behaviour changes
Elk avoid valleys and riverbanks. Overgrazing of willows and aspens ends
🌿
Vegetation recovers
Willows, aspens, cottonwoods return to riverbanks. Beavers follow. Rivers stabilise
🌞
Geomorphology shifts
Rivers move
Bank erosion reduced. Meanders slow. Wetland carbon stores rebuild
The Yellowstone wolf reintroduction is the most documented trophic cascade in history. William Ripple and Robert Beschta at Oregon State University tracked vegetation changes from 1995 onward. The term “geomorphic cascade” was coined to describe how wolf reintroduction physically altered the course of the Lamar and other rivers through its effects on riverbank vegetation and beaver populations.
In 1995, fourteen grey wolves were captured in Canada and released into Yellowstone National Park. They were the first wolves in the park since the last pack was eliminated in 1926. What happened next is one of the most remarkable ecological stories of the 20th century — and one of the clearest demonstrations of how a single species can reshape a landscape at a scale that no human intervention could replicate.
The park’s elk population had grown unchecked for seventy years. They grazed riparian areas — the strips of vegetation along riverbanks — intensively, stripping willows and aspens to bare twigs. Without deep-rooted vegetation, riverbanks eroded. Rivers widened and became shallower. Beaver, which depend on willows for food and dam construction, had largely disappeared.
Within a few years of the wolves’ return, something unexpected was happening. The elk were not simply being eaten in larger numbers — their behaviour was changing. They began to avoid open valleys and riverbanks, where wolves could run them down. The “landscape of fear” that the wolves created was reshaping where elk chose to spend their time.
In the areas the elk abandoned, vegetation recovered rapidly. Willows and aspens grew tall enough to shade the rivers. Beavers returned to build dams. The dams slowed river flow, raised water tables, and created wetland habitats. Songbird and beaver populations increased. And the rivers — the physical rivers themselves — began to meander differently, with more stable banks and deeper channels.
Indicator
Pre-wolves (1994)
Post-wolves (2010)
Willow height in surveyed plots
Under 50cm
Up to 250cm
Beaver colonies in park
1
12+
Riparian songbird species
Reduced
Significantly recovered
River channel stability
Eroding, widening
Narrowing, deepening
Wetland carbon storage
Depleted
Recovering
Yellowstone is one example, but wolf reintroductions across Europe — in Poland’s Bialowieza Forest, in the Italian Apennines, gradually expanding into France and Germany — are producing similar effects at different scales. Each recovery is slower and messier than the Yellowstone narrative suggests, complicated by livestock predation and human conflict. But the ecological direction is consistent.
Case study 04
The whale pump: how the largest animals fertilise the ocean
Global oceans — great whale recovery
Trophic cascade — whale nutrient cycling
🐋
Whale populations recover
Commercial whaling ended 1986. Slow recovery of great whales globally
🧬
Nutrient cycling increases
Whales defecate iron-rich nutrients at the surface. Phytoplankton blooms respond
🍏
Phytoplankton blooms
Phytoplankton fix CO². When they die they sink, sequestering carbon in deep ocean
🧳
Carbon sinks
33 tonnes
Average carbon stored per great whale carcass when it sinks to seabed (“whale fall”)
Research by Roman et al. (2014) in Frontiers in Ecology quantified the “whale pump” effect. A 2019 IMF working paper by Chami et al. estimated that restoring great whale populations to pre-whaling levels could increase carbon sequestration by 1.7 billion tonnes of CO² per year — equivalent to planting two trillion trees.
Whales are not predators in the trophic cascade sense — they do not suppress their prey to protect a lower trophic level. But they are architects of ocean carbon cycling in a way that makes them among the most important organisms on Earth for climate regulation, and one of the most overlooked.
The mechanism is indirect but powerful. Great whales feed at depth, then surface to breathe — and to defecate. Whale faeces are extraordinarily rich in iron and nitrogen: exactly the nutrients that limit phytoplankton growth in large stretches of the open ocean. Whales effectively act as a nutrient conveyor belt, bringing deep-ocean minerals to the sunlit surface layer where photosynthesis can use them. The resulting phytoplankton blooms absorb atmospheric CO⊂2; when the phytoplankton die, most sink to the ocean floor, taking that carbon with them into storage measured in centuries.
This is called the “whale pump.” It was diminished dramatically by commercial whaling, which removed an estimated 3 million great whales from the oceans over the 20th century — along with all the nutrient cycling they would have performed. Estimates of the resulting reduction in ocean carbon capture vary, but they are very large.
3M
Great whales killed by commercial whaling in the 20th century
33t
CO² sequestered per great whale in its body over a lifetime, then stored on the seabed
1.7Gt
CO²/year that could be sequestered if great whale populations returned to pre-whaling levels
The pattern behind the examples
Each case study above describes a different ecosystem, a different animal, and a different carbon pathway. But the underlying pattern is identical. A predator or keystone species was removed — by hunting, by fencing, by persecution — and an ecosystem that had been self-regulating for millions of years lost its capacity to do so. Prey populations exploded, vegetation collapsed, soils degraded, and carbon was released.
The return of the keystone species reversed the process. Not immediately, not cleanly, and not without conflict. But the direction was consistent and the carbon implications were significant.
What makes this relevant to Decarbonarma’s normal territory — heat pumps, solar panels, home batteries — is a question of scale. Every heat pump installed in a UK home avoids perhaps two to three tonnes of CO⊂2; per year. The recovery of great whale populations to pre-whaling levels could sequester 1.7 billion tonnes per year. These are not alternative strategies; they are complementary ones. But the scale difference is worth sitting with.
Trophic cascades work through behaviour change as much as population change. The “landscape of fear” that predators create is as important as direct predation in reshaping ecosystems.
The carbon value of ecosystem services is real but rarely priced. Sea otter kelp protection, dingo grassland maintenance, whale nutrient cycling — none of these appear in national carbon accounts.
Conflict with human economic activity is the primary barrier to recovery. Shellfish fisheries vs otters; sheep farming vs dingoes; cattle ranching vs wolves. These are political and economic problems, not ecological ones.
Rewilding is not a sufficient response to the climate emergency on its own. But it is a necessary one — and one that in many cases pays for itself through the ecosystem services it restores.
The cheapest carbon sequestration on Earth may simply be getting out of the way of the animals that already know how to do it.
Decarbonarma Newsletter
Get the next one in your inbox
Clear, no-nonsense analysis of UK home decarbonisation — heat pumps, solar, EVs, batteries and the grid. Occasional emails, no spam, unsubscribe anytime.
We’ll never share your email. Powered by Mailchimp.