People Start 85% of Wildfires. That Doesn’t Debunk Climate Change — and the Arson Story Is Mostly Made Up.

ISSUE #046 · MYTHS & MISINFORMATION · AUGUST 2026

In June 2023 the smoke from Quebec’s forests drifted far enough south that the sky over New York turned the colour of a cheap orange lolly, and Manhattan — briefly, absurdly — had the worst air quality of any city on Earth. People stood on rooftops filming it. And underneath the footage, in the replies, a second fire was already burning: it’s arson. Not the climate. Arsonists. Somebody with a match and an agenda, and the authorities won’t admit it.

You’ve seen the claim. Maybe some version of it even sounded about right. It turns up like clockwork now — every bad fire season, same sentence — and it’s persuasive for a simple reason. It hands you a villain and quietly lets you off the hook about the climate, all in one breath. Two birds.

Trouble is, it’s answering two completely different questions as though they were one. Pull them apart and the whole thing comes to bits in your hands.

Here’s the part that irritates one side of the argument straight away: the “it’s mostly humans” bit is true. In the United States, people are responsible for something like 85 per cent of wildfires. Not lightning. Us.

A dropped cigarette. A debris burn that jumps its line. A trailer chain throwing sparks off the tarmac. Fireworks — the Fourth of July is, year in and year out, the single biggest day for wildfire ignitions in America. Power lines arcing in the wind. Campfires nobody quite bothered to drown. The lightning that starts the rest is real, and out in the empty stretches of the Mountain West it burns most of the acreage — but by sheer count of fires, the spark is nearly always a person.

So if your gut says a lot of these fires have a human behind them, your gut is right. Hold onto that. It’s about to get turned against the next claim.

This is where the viral version cheats. “Humans cause 85 per cent of wildfires” quietly becomes “arsonists cause most wildfires” — and those are not remotely the same sentence.

Arson is one slice of the human pie, and not the biggest one. Take the fires with a known cause and deliberate ignition — arson and incendiarism — comes out at roughly 18 to 21 per cent. Debris burning, equipment and vehicles account for more. Most human-caused fires are carelessness, not malice: some idiot with a burn barrel, not a man driving out at night to torch a hillside.

And here’s the fact almost nobody shouting about arson seems to know. It’s going down. Arson-caused wildfires peaked from the late 1990s to about 2006, at somewhere between 14,000 and 20,000 recorded fires a year. It’s now more like 4,300 to 5,300. A long slide downward. Nobody’s entirely sure why — some of it may be the same collapse in smoking that shows up everywhere else in the data.

Now the bit the whole argument keeps tripping over.

“Who started the fire?” and “Why did the fire burn down a town?” are different questions. The first is about a spark. The second is about everything the spark lands in.

Climate change doesn’t strike the match. What it does is lay the fuel — hotter air, longer droughts, drier vegetation, more days a year when one careless spark can run. Scientists have a slightly clinical term for this, vapour pressure deficit, which is basically how thirsty the air is. Look at large, human-lit fires in the western US and the number of “flammable days” — days dry enough for a fire to take off — comes out on average 93 per cent higher than for lightning fires, and climbing faster. About 81 per cent of that increase traces straight back to greenhouse-gas emissions.

Put it plainly. A bloke drops a cigarette on a cool damp morning and it goes out. That same bloke drops that same cigarette at the end of a record-hot summer, into grass that hasn’t seen rain in three months, and it takes a county. Same spark. Same idiot. Completely different fire. The arsonist and the climate were never competing explanations. One lit it. The other decided how far it got.

There’s a smarter counter doing the rounds, and it deserves a straight answer rather than a dodge. It runs: fine, but there aren’t actually more wildfires than there used to be — so how bad can this really be?

The first half of that is true. The number of fires has been remarkably flat. For about forty years the US has averaged somewhere near 70,000 wildfires a year, with no obvious upward trend — by some measures fire incidence was actually a touch lower from the mid-80s to 2013 than in the couple of decades before. So if someone tells you “there are loads more fires now,” they’re wrong, and you can say so.

But watch the swap — it’s the same trick as before, just running the other way. The thing that’s changed isn’t how many fires start. It’s how much they burn. In the 1990s wildfires burned about 3 million acres a year in the US. Since 2000 the average is around 7 million. More than double. Of the ten worst years for acreage on record, every single one has come since 2004. And it’s the giants doing it: on Forest Service land between 1970 and 2002, roughly 1 per cent of fires accounted for 97.5 per cent of the area burned. Same number of fires — far bigger ones — in a season that keeps starting earlier and ending later.

Once you know to look for it, the arson-not-climate claim has a rhythm. It arrives with the smoke, it’s louder than the correction, and it gets stranger as it goes.

Canada’s record-breaking 2023 season — 185,000 square kilometres burned, the worst in the country’s history, most of it lightning-lit — filled the timelines with claims that arsonists, not the climate, were behind it. Then it mutated, the way these things do. Left-wing extremists. The government torching its own forests. Space lasers, drones, weather control, state-directed arson — all of it circulating widely while people were genuinely trying to find evacuation routes. Three years earlier, during the 2020 US fires, the villain of choice had been Antifa and Black Lives Matter instead. Same structure, different local enemy: take a heatwave and a drought, and pin them on whoever you already disliked.

And every time, when investigators go through the fires one by one, the arson figures being shared online turn out to be inflated or simply invented.

If you’ve read this far expecting me to land on “so it’s all climate and the arson lot are idiots” — no. That’s the other half of the same mistake, and this newsletter doesn’t do tidy.

Arson is real. Some fires genuinely are set on purpose; some places and some seasons have a serious deliberate-ignition problem; and waving that away to protect a climate argument is its own kind of dishonesty. But “it’s all just the climate” quietly ignores two things that matter a great deal — the ignition source, which is overwhelmingly us and largely preventable, and a century of fuel piling up in forests that used to burn lightly and often, before we got very good at stamping out every small fire. Manage the land badly and you get bigger fires whatever the thermometer says.

So both tribes are half right, and both are answering the wrong question. One is so busy naming a culprit it forgets to ask why the fire didn’t just fizzle. The other is so keen to talk about the climate it waves off the person holding the match.

A fire needs two things: a spark, and something for the spark to eat. The argument online only ever picks one of them — and then acts like it’s won.

People start about 85 per cent of wildfires. That number is real, and it doesn’t debunk a thing.

Who lit it and why it grew are different questions — and almost everyone online is loudly answering the wrong one.

National Interagency Fire Center (NIFC) — wildfire investigation and cause statistics. Source for arson as a share of human-caused fires, and for the agency’s own caveat that pre-1960 acreage data should not be compared with modern figures.

Headwaters Economics — “Lessons from 30 years of wildfire ignition data” (2024). Source for the ~85% human-caused figure and the long national decline in arson-caused fires.

US EPA, Climate Change Indicators: Wildfires; and US Forest Service data via NIFC. Source for the flat fire count over four decades and the roughly doubled acreage burned since the 1980s–90s.

Pourmohamad et al., “Inference of Wildfire Causes…,” Earth’s Future (2025). Source for arson at ~18% of known-cause western US fires and its declining trend.

“Exacerbating risk in human-ignited large fires…” (PNAS). Source for the 93% higher count of flammable days for human-lit large fires and the 81% greenhouse-gas attribution.

Oregon State University Extension, “Fire FAQs — have the size and severity of forest wildfires increased?” Source for megafires (1% of fires, 97.5% of area) and longer, earlier fire seasons.

Climate Action Against Disinformation, “Flame Wars,” with reporting by The Narwhal and Wildfire Today. Source for the arson-not-climate disinformation pattern across Australia (2019–20), the US (2020) and Canada (2023).

Carbon Brief factcheck on US wildfires, citing NIFC. Source for the record 2023 Canadian season and NIFC’s position on the unreliability of pre-1960 data.

ONE MORE THING

You can read our MANY newsletters for free at our website:

Read our archive for free at OneWildWorld.com →