From Watts Up With That?
Guest Post by Willis Eschenbach (@WEschenbach on X, my personal blog is here)
We live in an odd age when a billionaire can look at the planetary mess, point at cows, and announce that they’re a danger to the climate, while an entire continent quietly demonstrates that the real trouble came in the form of a virus.
Bill Gates likes to say that about 6% of global emissions are from cows, and that we should either “fix the cows” so they stop doing that, or “make beef without the cow”. Catchy line. The Serengeti, however, has a different script, and it starts not with cows, but with rinderpest.
Roll back to East Africa before the late 1800s. The Serengeti is running on its factory settings: huge migratory herds of wildebeest and buffalo sweeping back and forth with the rains, shaving the grass down as they go. Grazers are in charge; grass fuel stays modest; fires happen, but they’re patchy and relatively small; woodlands hang on as scattered trees and clumps that can survive the occasional, not too intense burn.
In that state, the place is not some methane-free Eden—those herds are belching merrily away—but it’s a functioning savanna where herbivores, fire, and trees have worked out a long-term compromise.
Then we improve it.
Import Indian cattle, import the rinderpest virus along with them, and suddenly the system’s main fuel controllers—wildebeest and their friends—hit a wall. The rinderpest chews through the ungulates, slashing wildebeest numbers to a fraction of their former glory and hammering buffalo as well.
Nobody has changed the rainfall, the soils, or the grass species. They’ve just imposed a new, top-down mortality factor that doesn’t care about carrying capacity or migration. The Serengeti shifts from “limited by food” to “limited by rinderpest.”
Once the big grazers are gone, the grass celebrates. It grows tall, cures, and lies there as a continuous carpet of fine fuel. What used to be eaten is now waiting for a match. In the rinderpest era, fires become more frequent, burn larger areas, and bite harder into seedlings and saplings that might otherwise have grown into trees.
Holdo, Holt, Fryxell, and colleagues reconstructed this period and concluded that the disease-depressed Serengeti was not just more flammable; it likely acted as a net carbon source, as repeated burning and reduced woody cover drained carbon from biomass and soils. If you’d flown over then, you might have called it a “natural fire-maintained grassland.” In reality, it was a savanna on crutches, being held open by an imported virus.
This is the part the climate spreadsheets never see. Our modern global tables show “livestock: ~12–14% of emissions, cattle ~ two-thirds of that,” and from this we get the sermon: cows are a climate threat, cows are 6%, cows must be fixed or replaced. Gates leans hard on that framing in interviews and in his climate book—livestock methane as a stand-alone villain, synthetic beef as the enlightened alternative.
But the Serengeti’s experience says that when you take out the large herbivores—the wild analogue of our cattle—you don’t automatically get a climate win; you can get more fire, less wood, and less stored carbon.
Now fast forward again to the mid-20th century, when the rinderpest story goes into reverse. Veterinarians roll out mass vaccination campaigns for cattle across East Africa, cutting the virus off at its preferred reservoir. As rinderpest fades from the livestock, it stops spilling over into wildebeest calves. Calf survival rises, adult mortality drops, and suddenly the main brake on herd size isn’t an exotic disease anymore—it’s how much grass the system can grow. Wildebeest numbers respond like a coiled spring: from a few hundred thousand under disease pressure up to over a million animals within a couple of decades. Buffalo and other grazers increase, too. The four-legged fuel management crew is back.
What happens next is the part that nobody who talks only about “6% from cows” seems to grapple with.
- More grazers means heavier grazing.
- Heavier grazing means less tall, continuous grass.
- Less fuel means fewer and smaller fires.
Holdo and co-authors found a tight inverse link: as wildebeest biomass went up after rinderpest eradication, burned area went down. Fire didn’t vanish, but its dominance did. With the fire regime dialed back, tree seedlings and shrubs suddenly get more windows of opportunity to grow past the flame-licking stage. Over time, woody cover expands, vegetation structure becomes more complex, and the system stores more carbon above and below ground.
Quantitatively, when they stitched the demography, fire data, and vegetation together, their conclusion was striking:
- Under rinderpest, the Serengeti’s savanna woodland mosaic behaved like a carbon source.
- With the virus gone and herbivores abundant, it flipped into a net carbon sink, sequestering on the order of a million tonnes of carbon per year in woody biomass over tens of thousands of square kilometers.
All that happened while the total methane output of the herbivores must have increased, because there were simply more of them. In other words: more burps, fewer fires, more trees, more carbon stored.
Meanwhile, in the global climate discourse, we get the flattening: cows are 6% of emissions, cars are 7%, therefore cows are “almost as bad as cars,” therefore rich countries should “move to 100% synthetic beef,” and we should “fix the cows” with methane-inhibiting feed or genetics.
It gets worse. Jeroen Remmers, a representative/advocate from the TAPP Coalition (True Animal Protein Price Coalition, which pushes for policies like meat/dairy taxes to account for “environmental costs”), recently spoke at the COP29 United Nations Climate Change Conference in Baku, Azerbaijan.
His claim? “Eating meat and dairy is causing 20% of all greenhouse gas emissions. … So the producers of meat and dairy should pay for the damage they cause.”
I guess 6% wasn’t scary enough, so he’s jacked it to 20%. That’s not science, that’s doing ecosystem surgery with a pie chart.
Yes, the FAO is clear that livestock produce methane and nitrous oxide. But it’s just as clear that a big chunk of those emissions is tied up with land use change, feed production, and manure handling—things that depend heavily on how and where animals are kept. The Serengeti case adds another dimension: change the herbivore–fire–vegetation feedback, and you can change the whole carbon behavior of a landscape without ever “fixing” a single burp.
The real insanity isn’t that people want to cut emissions. It’s that we’ve let the discussion be dominated by single-number thinking.
On one side, a savanna that went from balanced to overburned to carbon-storing, depending entirely on whether a virus was present in cattle, bison, and wildebeest.
On the other side, a global narrative where cows are framed as free-floating climate bombs and the proper response is to phase them out in favor of steel and fermenter “solutions.” One story is about how tightly life, fire, and carbon are coupled in real ecosystems. The other is about rearranging categories on a slide deck.
So yes, cattle have emissions. So do computers, cargo ships, and jet-fuel-powered climate summits.
The Serengeti reminds us that the big question isn’t “Do herbivores emit?”—they do—and it isn’t even “What percent of a global total do they represent?”
The big question is “What happens to the land, the fires, and the trees if they’re there—or if they’re not?”
When rinderpest arrived, the Serengeti lost its major grazers and likely lost carbon. When rinderpest left, the grazers came back, methane went up, fires went down, woodlands thickened, and carbon storage increased. That’s not a morality tale about cows. It’s a cautionary tale about thinking you can fix the climate by attacking one component of a system you haven’t really bothered to understand.
My very best to everyone on a gorgeous sunny spring afternoon,
w.
PS—My analysis, entitled “Computational implementation and empirical validation of a Constructal climate model”, has been published in the journal Biosystems and is available here. My thanks to Drs. Adrian Bejan and Umit Gunes, as well as to Anthony Watts, for encouraging me to publish my work.
PPS—Yeah, I know you’ve heard it before, but when you comment, please QUOTE the exact words you are discussing. It prevents endless misunderstandings.
You must be logged in to post a comment.