Rewilding the Plate: The Land Argument the Regenerative Story Can't Answer

Sciencegrass-fed beef Hart Mountain regenerative agriculture rewilding

The regenerative movement’s answer to every environmental critique is “it’s not the cow, it’s the how.” So let’s take that seriously, steelman the strongest version, then test it against the arithmetic, the peer-reviewed literature, and a 270,000-acre experiment in the Oregon desert.

The environmental defense of animal agriculture has evolved. Few people still argue that factory farms are good for the planet. The sophisticated position now is the regenerative one: it’s not the cow, it’s the how. Managed correctly, the story goes, moved in dense herds that mimic ancient bison, resting the land between rotations, cattle heal soil, sequester carbon, restore grasslands, and make beef part of the climate solution. It’s an appealing story, told by charismatic ranchers on beautiful land, and it deserves what every claim in this project gets: a fair hearing, followed by arithmetic.

So let’s do the arithmetic first, then hear from the land itself.

The arithmetic nobody disputes

How much land does animal agriculture use?

Start with the single most important set of numbers in food and environment, from the largest meta-analysis of global food systems ever conducted, Joseph Poore and Thomas Nemecek’s review of 38,700 farms across 119 countries, published in Science [1]. Agriculture occupies about half of all habitable land on Earth. Of that agricultural land, roughly 83%, pasture plus the cropland growing animal feed, serves meat and dairy production. And what does that 83% of the land deliver? About 18% of humanity’s calories and 37% of its protein [1, 2]. (That 83% is Poore and Nemecek’s own accounting, and the figure used throughout this series; Our World in Data expresses the same underlying data as “more than three-quarters.”) It takes on the order of 100 times more land to produce a calorie or a gram of protein from beef or lamb than from plants [2]. Run that comparison on a single plot and the gap is just as stark, two acres of good Carolina pasture against two acres of crops is its own post: Two Acres: What Beef Produces vs. What Plants Can.

Why is eating animals so much less efficient than eating plants?

That land gap is not an accident of how we farm; it is a consequence of physics, and it cannot be optimized away. Plants, working with soil microbes, are the base producers of the entire food supply: through photosynthesis they turn sunlight into the sugars, and with their microbial partners the amino acids, vitamins, and even the B12-producing bacteria, on which every animal ultimately depends. Animals do not create these nutrients. They concentrate and repackage what plants and microbes already made, and they charge a steep toll to do it. The second law of thermodynamics guarantees the toll exists: no energy transfer is ever free, and some fraction always disperses as unusable heat. Ecologists reduced the cost to a rule of thumb, roughly ten percent of the energy at one level survives into the next [47]. A cow eats the plant and spends most of those calories simply being a cow, respiring, walking, holding body temperature, growing hide and bone and organs nobody will eat. Measured directly, beef returns about 3% of the calories fed to cattle and about 3% of the protein, against a 7 to 8% average across all animal categories, making beef by far the least efficient [48].

Two points of precision, because this is the load-bearing claim of the whole land debate. That ten percent is an average, not a constant: Lindeman, who established it in 1942, measured efficiencies anywhere from a fraction of a percent up into the high thirties, and modern feed-conversion numbers vary the same way by species and system [47]. And the second law does not fix the exact size of the loss; most of it is respiration, movement, and inedible tissue rather than entropy in the abstract. What the second law does guarantee is the part that matters here: the transfer can never be free. Adaptive multi-paddock grazing does not repeal it. Holistic management does not repeal it. Better genetics do not repeal it. You can breed a faster-growing cow, but you cannot repeal the energy loss between trophic levels any more than you can build a perpetual-motion machine. The debate can only ever be about how large the tax is, never whether it is levied, and the 83% of farmland and the roughly hundredfold land penalty are simply that tax, scaled to continents.

How much land would we save if everyone went vegan?

Now run the scenario the regenerative advocates least want to discuss. If the world ate fully plant-based, total agricultural land use would fall from about 4.1 billion hectares to about 1 billion, a 75% reduction, an area roughly the size of North America and Brazil combined [2]. Every hectare of grazing land, nearly three-quarters of agriculture’s entire footprint, comes back. And cropland shrinks too, because the enormous acreage currently growing feed crops disappears faster than human-food cropland grows to replace the calories [2]. That is the fact behind the claim you may have heard: go vegan and you return essentially all of the world’s pasture to the wild, plus a meaningful slice of its cropland.

Isn’t most grazing land unsuitable for crops anyway?

Here the regenerative side raises its favorite objection: most grazing land is marginal, so you can’t grow crops there! Correct, about two-thirds of grazing land isn’t suitable for cropping [3]. Its companion claim, that 86% of livestock feed is inedible to humans, comes from an FAO analysis whose remaining figures rarely get quoted: the same study reports livestock eating about a third of global cereal production and consuming roughly 3 kg of human-edible feed per kg of meat produced [3]. The full breakdown of that statistic is its own post, The 86% Meme, Debunked. But notice what the marginal-land objection assumes: that the only alternative use of land is farming it. The freed land doesn’t need to grow crops. It needs to grow ecosystems.

What a billion freed hectares is actually worth

How much carbon could rewilded farmland absorb?

In 2021, researchers at NYU, Harvard, Colorado State, and Oregon State published in Nature Sustainability what they called the “carbon opportunity cost” of animal-sourced food: the carbon that the land under pasture and feed crops would pull from the sky if allowed to return to native vegetation [4]. The numbers are staggering. A global shift to plant-based diets by 2050 would allow the regrowing forests, grasslands, and wetlands to sequester an estimated 547 billion tonnes of CO2, roughly the past sixteen years of global fossil fuel emissions, with ecosystem soils potentially adding a couple hundred billion tonnes more [4]. Even the partial shift recommended by the EAT-Lancet Commission would sequester about 332 billion tonnes, nine years of fossil emissions [4]. Put another way: dietary change alone could recapture 99–163% of the entire remaining carbon budget for holding warming to 1.5°C, an amount of carbon removal that mainstream climate models otherwise assign to speculative machine-capture technology that barely exists [4]. And that’s before counting the biodiversity dividend: agricultural expansion is the leading driver of deforestation and extinction worldwide, with beef the single largest commodity driver of tropical forest loss, a debt this post settles in its own section below [2, 5]. Remember one of that carbon paper’s co-authors, William Ripple of Oregon State. He’ll reappear shortly, standing in a creek bed with a camera.

So the regenerative claim isn’t competing against factory farming. It’s competing against that, against returning a North-America-plus-Brazil-sized landmass to forests, wetlands, and wild grasslands. Keep that bar in mind as we examine the claim on its merits.

The rainforest the regenerative story would rather not mention

Is Amazon deforestation actually caused by beef?

Steelman it first: the regenerative advocate will say the Amazon is a governance failure, not a beef failure, that well-managed ranching need not fell a single standing tree, and that Brazil’s own satellites now show deforestation falling. Both halves of that are true. Under President Lula, Brazil’s INPE recorded Legal Amazon clearing of 5,796 square kilometers for the year to July 2025, down 11% and the third-lowest reading since monitoring began in 1988, while alert data for the year to July 2026 fell another 36% to roughly 2,874 square kilometers, the lowest since 2013 [37]. Clearing has fallen about half since 2022. Good. Now do the arithmetic on what is actually taking the forest down.

How many acres of rainforest are cleared for cattle each day?

Between 90% and 99% of tropical deforestation is driven, directly or indirectly, by agriculture, and the single largest commodity is beef [38]. Pendrill and colleagues attribute about 41% of tropical deforestation to pasture expansion for cattle, roughly 2.1 million hectares a year, an area about half the size of the Netherlands, or some 5,750 hectares, about 14,200 acres, every single day [39]. In Brazil, cattle ranching drives on the order of 72% of deforestation, and cattle pasture occupies roughly 70 to 80% of already-cleared Amazon land [40]. Soy is the usual scapegoat, but 77% of the world’s soy is fed to animals, not people, so most “soy deforestation” is, at one remove, meat deforestation too [41].

Is animal agriculture responsible for 91% of Amazon deforestation?

You will sometimes see the figure “91% of Amazon destruction” pinned on animal agriculture. Resist it, even though it favors this post’s argument. It traces through the film Cowspiracy to a misreading of a 2004 World Bank report by Sérgio Margulis, who attributed most, not 91%, of clearing to ranching [42]. The honest number, roughly 70 to 80% cattle pasture, is damning enough without inflation, and using it is precisely how this argument keeps its credibility. The same discipline applies to the daily-acreage figure: the dramatic “14,400 acres a day” sometimes quoted is not a separate, larger measurement but the same 2.1-million-hectare annual total run through a different unit conversion. One number, honestly stated, beats two that invite a fact-check.

What percentage of mammals on Earth are livestock?

And the biological cost is not abstract. By biomass, livestock now make up about 60% of all mammals on Earth, humans 36%, and wild mammals just 4%; all the world’s remaining wild land mammals together weigh about 20 million tonnes, roughly three kilograms per living person [43]. The frontier has moved beyond the Amazon, too, into the Cerrado savanna and the Gran Chaco, converted for soy and cattle at rates that rarely reach the headlines. Agriculture is the leading driver of terrestrial extinction, and the EU’s own Deforestation Regulation, covering cattle, soy, palm, cocoa, coffee, rubber, and timber, has now slipped to December 2026 [44]. The forest is not being lost to feed people. It is being lost to feed livestock.

Steelmanning the regenerative story, then weighing it

Does regenerative grazing actually sequester carbon?

Here is the claim at its strongest. Holistic and adaptive multi-paddock grazing, its advocates say, stimulates grass growth, drives roots deeper, builds soil carbon, and, per the most famous versions of the argument, can offset the cattle’s own emissions or even go carbon-negative, while reversing desertification. One celebrated farm’s sponsored life-cycle analysis reported net-negative beef. A famous TED talk claimed planned grazing could green the deserts. These are testable claims, and to their credit, serious scientists have tested them.

The most comprehensive evaluation is Grazed and Confused?, a two-year assessment by an international team led from Oxford’s Food Climate Research Network, dissecting every version of the grazing-as-climate-solution argument against the available science [6]. Its findings: grazing can boost soil carbon only under specific local conditions; the effect is small, time-limited (soils saturate within decades), and reversible (stop the practice, lose the carbon); and at the global maximum, roughly 0.3 to 0.8 billion tonnes of CO2 per year: it would offset only 20–60% of the emissions from the grazing animals themselves, never mind the rest of a livestock sector responsible for about 12% of all human-caused greenhouse emissions, around 6.2 billion tonnes of CO2-equivalent a year, of which cattle alone are 62% [6, 36]. The lead author’s conclusion was blunt: grass-fed livestock are not a climate solution; grazing animals are net contributors to the problem, and eating less meat, of all types, is what helps. A 2023 analysis in Nature Communications sharpened the point: for grassland carbon sequestration to genuinely offset current ruminant emissions, regional soil carbon stocks would have to grow by 25% to 2,000%, which is to say that it is not feasible [7].

Can grass-fed beef feed everyone?

Then there’s the scaling problem, which is really a land problem, which brings us back to the arithmetic. Grass-finished beef is slower and land-hungrier than feedlot beef. When researchers modeled converting all US beef to grass-fed, they found it would require a cattle population roughly 30% larger, and even then, the country’s grasslands couldn’t support current beef consumption, forcing either much less beef or the conversion of more wild and crop land to pasture [8]. Even the regenerative movement’s showcase: the sponsored life-cycle analysis of its most famous farm, conceded the operation required roughly 2.5 times the land of conventional production per unit of beef, with sequestration that plateaus as soils approach saturation [9]. And the experimental literature on rotational grazing itself, synthesized across decades of rangeland trials, finds no consistent superiority over continuous grazing on plant or animal production measures: the perception, the researchers concluded, has outrun the evidence [10].

If cows only ate grass and food scraps, how much meat could we have?

Now take the movement’s own premise seriously, seriously enough to model it. The regenerative case rests on upcycling: animals earn their place by converting grass and crop leftovers, things humans can’t eat, into food we can. Fine. What happens if livestock are restricted to only that? Researchers led by Hannah van Zanten at Wageningen ran exactly this scenario, and their answer is the most quietly devastating number in the debate. Livestock fed solely on leftover streams and grassland could supply roughly 9–23 grams of animal protein per person per day, against the 50–60 grams of total protein an adult needs; ruminants fed solely on existing grassland come to about 7 grams [30]. Set that beside current global supply of about 27 grams of terrestrial animal protein per person per day [30], and beside the far larger figures in wealthy countries, where Europe averages 51 grams. The upcycling world contains a fraction of today’s meat, and rich-country consumption has the furthest to fall. This is the deepest irony in the regenerative argument: it borrows the moral glow of a model whose own arithmetic forbids the scale of production it’s used to defend. Advocates invoke the leftover-eating cow to justify the steak; the leftover-eating cow’s math says the steak mostly has to go.

Is grass-fed beef better for the climate than feedlot beef?

The scale problem shows up in the actual production numbers too. Grass-fed beef supplies roughly one gram of protein per person per day globally, ruminants as a whole contribute about 13, which is to say that a rounding error in the world’s protein supply is being deployed to defend a global industry [6]. And the carbon accounting, when done carefully, doesn’t rescue it either: a 2025 analysis in PNAS found US grass-fed beef is about as carbon-intensive as industrial beef and roughly ten times more emissions-intensive per unit of protein than common protein-dense alternatives, even after crediting the soil sequestration its defenders cite [31]. The same paper ran the counterfactual this post keeps returning to: cropland reallocated from beef production to food grown directly for people delivers three to seven times more protein per kilogram of CO2 emitted, per hectare occupied, and per kilogram of reactive nitrogen used [31].

Notice the shape of this dispute, because it mirrors the health debate exactly. The regenerative claims live in TED talks, brand-sponsored analyses, and gorgeous documentary footage. The rebuttals live in multi-year systematic assessments, Nature-family journals, and rangeland-science syntheses. We built a ladder for exactly this situation. And even granting regenerative grazing its best case, modestly less-bad beef on modestly healthier soil, less bad is not the alternative on offer. The alternative is the land coming back entirely. Which is not a hypothetical. It has a name, an acreage, and thirty years of data.

The 270,608-acre control group

What happened when Hart Mountain removed all its cattle?

In the sagebrush steppe of southeastern Oregon sits Hart Mountain National Antelope Refuge: the subject of the documentary Rewilding a Mountain, and the closest thing the American West has to a controlled experiment on the regenerative question [11]. The refuge was created in 1936 in a desperate bid to save the pronghorn, an animal built by a million years of evolution to outrun extinct American cheetahs but not the rifles and fences of settlement: from an estimated thirty to thirty-five million animals before European settlement, the continental population had collapsed to roughly thirteen thousand by the 1920s [11, 33]. The land itself, meanwhile, had been continuously grazed by cattle since the late 1800s, straight through the refuge’s creation and for a half-century after it.

By the late 1980s, the contradiction was impossible to ignore. Riparian zones: the green ribbons along streams that occupy under 5% of the arid West’s landscape but support the majority of its wildlife species, a standard finding across Great Basin riparian ecology, were functionally dead: bare, caving banks, compacted soil that one scientist compared to a worn-out sponge that won’t hold water, and almost no willow or young aspen anywhere [11]. When refuge manager Barry Reiswig, himself from a livestock family, asked the world’s leading pronghorn authority which grazing system would best serve the antelope, the answer was three words: there isn’t one [11]. A multi-year environmental impact study led by riparian ecologist Boone Kauffman and avian ecologist David Dobkin reached the conclusion the data forced: grazing was incompatible with the refuge’s legal mandate. So in 1990–91, Hart Mountain became one of the first, and among the largest, western refuges to remove livestock entirely, buying out ranchers’ permits and beginning two decades of fence removal across 270,608 acres [11, 12]. No reseeding campaigns, no engineered stream repairs at scale. Mostly, the management plan was subtraction: take the cattle off and watch.

How fast does land recover after livestock are removed?

What happened next is now peer-reviewed. Oregon State’s Jonathan Batchelor and William Ripple, yes, the same Ripple who later co-authored the global carbon-opportunity-cost paper, compared 64 matched photographs of refuge streams taken before cattle removal in 1990 against retakes 23 years later [12]. Bare soil declined by 90%. Exposed, eroding channel declined by 63%. Willow cover increased 388% and rushes 389%; channel widths narrowed at 64% of sites and eroding banks healed at 73%; grasses, sedges, and forbs all gained [12]. The authors’ conclusion was that removing cattle produced dramatic riparian recovery even in a semi-arid landscape without active restoration, passive restoration, nature doing the work for free [12]. The tree rings tell the same story: after grazing ended in 1990, aspen recruitment rose by more than an order of magnitude, over 1,000%, filling the “missing middle” age class left by a century of cattle eating every young shoot [34]. The birds followed the plants: repeated transect surveys found overall bird abundance in the riparian aspen up about a third, with ground- and understory-nesting species specifically more than doubling, a 133% increase, even as many of those same species were declining regionally [13, 14]. A 24-year follow-up found the gains held, with three times as many species colonizing as dropping out, though cavity-nesting birds declined locally, for want of the old, rot-softened trees a century of grazing never let grow, an honest reminder that recovery can’t instantly undo everything [35]. Springs that longtime staff remembered as bone-dry now audibly run [11]. And the pronghorn, after an initial dip that critics gleefully mistook for failure, climbed from summer counts in the low hundreds in the 1970s to recent estimates of 2,000–3,500 as the last fences came down [11, 12].

Do cattle really mimic bison on native grassland?

Two lessons from Hart Mountain cut straight through the regenerative story. The first is ecological specificity. The “cattle mimic bison” argument borrows its plausibility from the Great Plains, where sod-forming grasses co-evolved with massive migrating herds. Great Basin bunchgrasses did not: they evolved with no such grazing pressure, so cattle eat them before they seed, native cover collapses, and invasive cheatgrass and desertification move in, which is precisely what the surrounding grazed lands show while the ungrazed refuge regains species found nowhere off it [11]. Regenerative grazing’s premise is, at very best, regional; marketed as universal, it is simply false. The second lesson is the documentary’s own closing verdict: millions of acres across the West are managed today the way Hart Mountain was before removal, and reaching these ecological milestones through any active management that kept cattle on the land would likely not have been possible [11]. Passive restoration worked. The regenerative movement promises that the right kind of grazing can heal land; Hart Mountain demonstrates what the land does when asked a simpler question, what if we just stopped? It’s worth adding that no rancher was ruined to find out: permits were purchased, not confiscated, a working template for a just transition, one buyout and one fence at a time.

The wildlife that doesn’t get to come back

Hart Mountain shows what returns when livestock leave. There’s a grimmer corollary: on the land where livestock stay, wildlife doesn’t merely get crowded out, a federal program kills it.

How many wild animals does the US government kill for livestock?

Wildlife Services, an arm of the USDA, exists largely to remove animals that agricultural producers consider threats. Its own reported figures are substantial: 375,045 native animals killed in 2023, including 68,562 coyotes, 24,603 beavers, 2,122 foxes, 469 bobcats, 430 black bears, 305 gray wolves, and 235 mountain lions [26]. Prior years run similar, over 400,000 native animals in 2021 [27]. Add invasive-species removals and the annual program total climbs into the millions. The methods include aerial gunning from low-flying aircraft, neck snares, leghold traps, den gassing, and M-44 sodium cyanide devices, and by the agency’s own accounting they kill thousands of unintended animals a year, golden and bald eagles, grizzlies, bighorn sheep, family pets, and, with real irony, livestock-protection dogs [26, 27].

Do predators actually kill much livestock?

Two facts make this hard to defend on its own terms. First, predators account for a small share of livestock losses overall, disease, weather, and birthing complications kill far more, so the ecological price is being paid for a modest slice of a modest problem [28]. Second, an NPR investigation using records obtained under FOIA found roughly 11,000 animals killed at locations with no recorded livestock damage at all; at one Montana site the program killed 318 coyotes with no documented coyote predation there, and across three years it killed 71 wolves at five locations where wolves had harmed 61 animals out of a state herd of roughly 2.5 million, about 6% of Montana’s wolves for predation on 0.002% of its livestock [29]. This is not the exception the program’s defenders describe; it’s the routine.

Set that beside the arithmetic from the top of this post. Livestock occupy about 83% of agricultural land, and on a meaningful share of that land, native carnivores and ecosystem engineers like beavers are killed at public expense to keep the arrangement intact. Beavers deserve their own footnote here, because Hart Mountain explains why: they build the wetlands, raise water tables, and restore exactly the riparian function the refuge regained, meaning the program is killing, by the tens of thousands, the animals that would do the restoration work for free. The wildlife-versus-livestock conflict isn’t a hypothetical tradeoff at the margins. It’s a line item in the federal budget, and dietary demand is what funds it. In the nine-column body count of the food system, field deaths, slaughter, wild capture, predator control, disease culls, on-farm mortality, pesticide kills, dead zones, and habitat loss: this is the column almost nobody tallies; The Crop Deaths Argument, Counted Honestly counts them all.

The fertility objection: “you can’t farm crops without cows”

Can you farm crops without cows for manure?

One more argument deserves the steelman treatment, because it sounds like ecology and travels fast: without livestock, farming would depend entirely on synthetic agrochemicals, manure is the natural fertility loop, grazing enriches the land, and plowing for crops poisons and kills everything anyway. Every clause of that inverts the actual accounting.

Where does nitrogen in fertilizer actually come from?

Start with what a cow contributes to soil: nothing she didn’t eat. Cattle manufacture no nitrogen, phosphorus, or potassium, manure is plant nutrients passed through an animal, minus substantial losses to air and water along the way. So the fertility question is really: where did the plants’ nitrogen come from? In today’s system, overwhelmingly from the Haber-Bosch process, synthetic fertilizer that already feeds roughly half of humanity [21] and that saturates the very feed crops whose nutrients end up in the manure pile. The livestock system doesn’t shield agriculture from agrochemicals; it is the agrochemical system, with the cow inserted as a leaky middleman. Readers of the health post will recognize the shape of this argument: it’s B12 again: the industry feeds its animals the supplement, then points to the animal as the source. Nitrogen’s actual natural source is bacteria, rhizobia on legume roots, fixing it from the air, which is precisely what the clover in a “self-sufficient” pasture is doing beneath the cattle’s feet.

Does animal agriculture cause more pesticide use?

The pesticide charge inverts the same way. Roughly two-thirds of America’s crop calories are fed to livestock rather than people [22], which means the pesticide- and herbicide-intensive corn–soy monoculture exists chiefly in service of animal agriculture. Federal regulation documents this in its own numbers: glyphosate tolerances on animal feed commodities run 1,500 to 4,000 times those permitted on the fruits and vegetables sold to people, a direct readout of where chemical intensity concentrates [32]. Shrink the livestock and you shrink the monoculture, its chemistry, and the harvested acreage, along with the field-animal deaths that acreage entails, since routing crops through animals multiplies the acres a food system must harvest, not the other way around: the full body count runs the same direction. The full version of that argument, including where the residues end up inside the animal, is its own post, The Pesticide Argument the Carnivore Movement Can’t Answer. Nor is manure the benign input of the story: its runoff helps feed the dead zone we met earlier, its nitrates seep into rural groundwater, and produce-contamination outbreaks have repeatedly pointed investigators toward neighboring cattle operations.

What is veganic farming and does it work?

Can farms actually run without any animal inputs, no manure, no blood meal, no bone meal, no grazing? They already do, with decades of receipts. The Stockfree Organic standard, developed with the Vegan Organic Network, certifies exactly this, and its flagship is Tolhurst Organic in England: the first certified stockfree farm in the world, growing on the order of 120 tonnes of vegetables a year on about 20 acres of low-grade land, its fertility built entirely from green-manure rotations, undersown cover crops, and woodchip compost from the farm’s own hedges and coppice, sustained now for more than three decades [23, 24]. The quiet irony is that mainstream organic standards already concede the principle: they specify that primary fertility must come from fertility-building crops, with manure permitted only as an adjunct [23]. Veganic growing isn’t a hypothetical awaiting invention; it’s the fertility-building-crop rule taken to its logical conclusion, with the bonus, as the stockfree growers note, of removing manure-borne pathogen pathways from vegetable production entirely [23].

Does grazing improve or degrade soil?

As for grazing “enriching” the land: the IPCC lists overgrazing among the leading human causes of soil degradation and desertification worldwide [25], and Hart Mountain’s soils told the same story in miniature, hardpan you could barely dent after a century of cattle, sponge-soft and water-holding within two decades of their removal [11, 12]. The land did not need the animals. It needed the absence of them.

The rest of the footprint, briefly

What else does animal agriculture damage?

Land and carbon are the headline, but the same subtraction pays down every other column. Routing crops through animals is why feed-crop fertilizer and manure runoff feed the Gulf of Mexico’s recurring dead zone [15]. In 2023, more than six million kilograms of medically important antibiotics were sold for use in US food animals, 56% of all antibiotics sold for livestock, with swine and cattle taking the largest shares, a primary engine of antibiotic resistance [16]. About three-quarters of emerging infectious diseases are zoonotic: the 2009 pandemic flu emerged from swine, and H5N1 bird flu, first confirmed in US dairy cattle in March 2024, had spread to more than a thousand herds across some seventeen states by early 2026, with a second independent spillover into Nevada cattle in 2025, and dense confinement operations are purpose-built incubators [17]. The response to those outbreaks is itself one of the largest body counts in the food system, and it goes almost entirely uncounted: more than 185 million birds have been killed in US flocks since February 2022, on top of the 50.5 million lost in the 2014–15 outbreak [45]; and African swine fever is estimated to have killed or forced the culling of some 143 million pigs in China across 2018–19, a figure that blends disease deaths, official culling, and panicked liquidation by farmers dumping herds ahead of the virus [46]. The waste has nowhere to go: industrial hog operations store feces and urine in open-air lagoons that emit ammonia and hydrogen sulfide over surrounding communities and have flooded into waterways during hurricanes; since 2013, the largest US pork producer, Smithfield, has been owned by China’s WH Group, meaning much of the pork is exported while American communities keep the lagoons [18]. And all of it exists to process more than 80 billion land animals through slaughter every year [19], which is the point where the environmental column meets the ethical one. As the health post noted, veganism was never a diet; it is the commitment to avoid animal exploitation as far as possible and practicable [20]. The remarkable thing is how completely the ledgers converge: the same act that spares the animals frees the land, and the same plate that compresses human morbidity un-compresses the wild.

The bottom line

What is the single biggest thing you can do for the planet?

Every serious climate solution humanity is pursuing requires building something, turbines, panels, grids, carbon-capture machinery. Rewilding through diet is the rare one that runs on subtraction. Stop routing 83% of agricultural land through animals to produce 18% of calories, and a landmass roughly the size of the US, China, the EU, and Australia combined begins pulling sixteen years’ worth of fossil carbon out of the sky while the willows, the warblers, and the pronghorn move back in, and the coyotes, wolves, and beavers stop being shot to keep them out [1, 2, 4, 11, 12, 26, 29]. The regenerative story asks us to admire a slightly better way of occupying that land. Hart Mountain, soil soft as a marshmallow where hardpan used to be, streams audible where there was silence, shows what happens when we give it back. On the refuge, they tell a story about the day after a fence came down: a dozen pronghorn approached the old line, hesitated where the wire had always been, then burst through open country as the crew cheered from the rimrock [11]. That, scaled to a planet, is what’s actually on the table.

References

  1. Poore J, Nemecek T. Reducing food’s environmental impacts through producers and consumers. Science. 2018;360(6392):987–992. Link
  2. Ritchie H. If the world adopted a plant-based diet, we would reduce global agricultural land use from 4 to 1 billion hectares. Our World in Data; 2021. Link
  3. Mottet A, de Haan C, Falcucci A, Tempio G, Opio C, Gerber P. Livestock: on our plates or eating at our table? A new analysis of the feed/food debate. Global Food Security. 2017;14:1–8.
  4. Hayek MN, Harwatt H, Ripple WJ, Mueller ND. The carbon opportunity cost of animal-sourced food production on land. Nature Sustainability. 2021;4:21–24. Link
  5. Ritchie H, Roser M. Drivers of deforestation. Our World in Data. Link
  6. Garnett T, Godde C, Muller A, et al. Grazed and Confused? Ruminating on cattle, grazing systems, methane, nitrous oxide, the soil carbon sequestration question, and what it all means for greenhouse gas emissions. Food Climate Research Network, University of Oxford; 2017. Link
  7. Wang Y, de Boer IJM, Persson UM, et al. Risk to rely on soil carbon sequestration to offset global ruminant emissions. Nature Communications. 2023;14:7625. Link
  8. Hayek MN, Garrett RD. Nationwide shift to grass-fed beef requires larger cattle population. Environmental Research Letters. 2018;13:084005.
  9. Rowntree JE, Stanley PL, Maciel ICF, et al. Ecosystem impacts and productive capacity of a multi-species pastured livestock system. Frontiers in Sustainable Food Systems. 2020;4:544984.
  10. Briske DD, Derner JD, Brown JR, et al. Rotational grazing on rangelands: reconciliation of perception and experimental evidence. Rangeland Ecology & Management. 2008;61(1):3–17.
  11. Rewilding a Mountain (documentary film on Hart Mountain National Antelope Refuge).
  12. Batchelor JL, Ripple WJ, Wilson TM, Painter LE. Restoration of riparian areas following the removal of cattle in the northwestern Great Basin. Environmental Management. 2015;55(4):930–942. Link
  13. Dobkin DS, Rich AC, Pyle WH. Habitat and avifaunal recovery from livestock grazing in a riparian meadow system of the northwestern Great Basin. Conservation Biology. 1998;12(1):209–221.
  14. Earnst SL, Dobkin DS, Ballard JA. Changes in avian and plant communities of aspen woodlands over 12 years after livestock removal in the northwestern Great Basin. Conservation Biology. 2012;26(5):862–872. Link
  15. National Oceanic and Atmospheric Administration. Gulf of Mexico hypoxic zone (annual monitoring); nutrient runoff from agricultural sources in the Mississippi basin.
  16. US Food and Drug Administration. Summary report on antimicrobials sold or distributed for use in food-producing animals (annual).
  17. Taylor LH, Latham SM, Woolhouse MEJ. Risk factors for human disease emergence. Philosophical Transactions of the Royal Society B. 2001;356(1411):983–989.
  18. Reuters. Shuanghui (WH Group) completes acquisition of Smithfield Foods. September 2013.
  19. FAO. FAOSTAT: livestock primary, annual global slaughter statistics.
  20. The Vegan Society. Definition of veganism (established 1944).
  21. Erisman JW, Sutton MA, Galloway J, Klimont Z, Winiwarter W. How a century of ammonia synthesis changed the world. Nature Geoscience. 2008;1:636–639.
  22. Cassidy ES, West PC, Gerber JS, Foley JA. Redefining agricultural yields: from tonnes to people nourished per hectare. Environmental Research Letters. 2013;8:034015.
  23. Tolhurst Organic; Stockfree Farming. What is Stockfree Organic? Farm profile and standards (tolhurstorganic.co.uk; stockfreefarming.org).
  24. Soil Association. Stockfree organic farming: abundance without animals? March 2023.
  25. IPCC. Climate Change and Land: An IPCC Special Report. 2019. Link
  26. USDA APHIS Wildlife Services. Program Data Reports, FY2023 (as compiled and reported by the Center for Biological Diversity, April 2024).
  27. USDA APHIS Wildlife Services. Program Data Reports, FY2021; WildEarth Guardians analysis of FY2021–2022 data.
  28. USDA National Agricultural Statistics Service. Cattle and Sheep Death Loss reports (predation as a share of total livestock losses).
  29. NPR Investigations. How a federal program kills wildlife for private interests. October 10, 2024. Link
  30. van Zanten HHE, Herrero M, van Hal O, et al. Defining a land boundary for sustainable livestock consumption. Global Change Biology. 2018;24(9):4185–4194. Link
  31. Eshel G, Flamholz AI, Shepon A, Milo R. US grass-fed beef is as carbon intensive as industrial beef and ≈10-fold more intensive than common protein-dense alternatives. PNAS. 2025;122(12):e2404329122. Link
  32. US Environmental Protection Agency. Glyphosate tolerances, 40 CFR § 180.364 (commodity-by-commodity residue limits for human food vs. animal feed).
  33. O’Gara BW, Yoakum JD. Pronghorn: Ecology and Management. Wildlife Management Institute / University Press of Colorado; 2004 (historic continental population estimates).
  34. Beschta RL, Painter LE, Levi T, Ripple WJ. Long-term livestock grazing alters aspen age structure in the northwestern Great Basin. Forest Ecology and Management. 2015;329:30–36. Link
  35. Poessel SA, Hagar JC, Haggerty PK, Katzner TE. Removal of cattle grazing correlates with increases in vegetation productivity and in abundance of imperiled breeding birds. Biological Conservation. 2020;241:108378. Link
  36. FAO. Pathways towards lower emissions: A global assessment of the greenhouse gas emissions and mitigation options from livestock agrifood systems. Rome; 2023. (Livestock ≈6.2 Gt CO2-eq, ~12% of anthropogenic emissions; cattle 62% of terrestrial animal-source-food emissions. Earlier estimate: Gerber PJ et al., Tackling Climate Change through Livestock, FAO 2013, 14.5%.)
  37. Instituto Nacional de Pesquisas Espaciais (INPE). PRODES and DETER Legal Amazon deforestation data, 2024–2026 (via gov.br/Secom; Mongabay reporting). Link
  38. Pendrill F, Gardner TA, Meyfroidt P, et al. Disentangling the numbers behind agriculture-driven tropical deforestation. Science. 2022;377(6611):eabm9267. Link
  39. Pendrill F, Persson UM, Godar J, et al. Agricultural and forestry trade drives large share of tropical deforestation emissions. Global Environmental Change. 2019;56:1–10; and Ritchie H, Drivers of Deforestation, Our World in Data (beef ≈41% of tropical deforestation, ≈2.1 Mha/yr). Link
  40. Ritchie H, Roser M. Drivers of Deforestation and Soy. Our World in Data; Greenpeace/Mongabay (cattle ranching ≈72% of Brazilian deforestation; ~70–80% of cleared Amazon land in pasture). Link
  41. Ritchie H. Soy. Our World in Data (77% of global soy used as animal feed; 7% direct human food). Link
  42. Margulis S. Causes of Deforestation of the Brazilian Amazon. World Bank Working Paper No. 22; 2004 (cited to document, and preempt, the inflated “91%” advocacy figure popularized by the film Cowspiracy).
  43. Bar-On YM, Phillips R, Milo R. The biomass distribution on Earth. PNAS. 2018;115(25):6506–6511; and Greenspoon L, Krieger E, Sender R, et al. The global biomass of wild mammals. PNAS. 2023;120(10):e2204892120. Link
  44. Regulation (EU) 2025/2650 amending the EU Deforestation Regulation (application dates 30 December 2026 / 30 June 2027).
  45. USDA Animal and Plant Health Inspection Service. Highly Pathogenic Avian Influenza: commercial and backyard flock detections and depopulation totals, February 2022–present; and HPAI 2014–2015 outbreak summary. Link
  46. FAO and USDA Economic Research Service. African swine fever in China: herd losses and culling, 2018–2019.
  47. Lindeman RL. The trophic-dynamic aspect of ecology. Ecology. 1942;23(4):399-417. Link
  48. Shepon A, Eshel G, Noor E, Milo R. Energy and protein feed-to-food conversion efficiencies in the US and potential food security gains from dietary changes. Environmental Research Letters. 2016;11:105002. Link
From Barren Gully to Lush Valley 1 768x432

Leave a Review

Footer preview