life after warming

The Return of the Urban Firestorm

What happened in Colorado was something much scarier than a wildfire.

An aerial view of one of the Boulder County neighborhoods that burned to the ground on Thursday. Photo: Hart Van Denburg/AP/Shutterstock
An aerial view of one of the Boulder County neighborhoods that burned to the ground on Thursday. Photo: Hart Van Denburg/AP/Shutterstock

On Thursday afternoon, in the space of a few hours just a day before the new year, 100-plus-mph winds carried the most destructive fire in Colorado history through the suburban sprawl of greater Denver, destroying much of the towns of Louisville and Superior and forcing tens of thousands to flee, including many who had entered shopping malls from sunny skies just a few minutes before. As many as 1,000 homes were destroyed. Two people currently remain missing; if the death count stays at zero, Colorado governor Jared Polis said Friday, it would be “a New Year’s miracle.”

By the standards of the megafires and gigafires of the last few years, the Marshall Fire was quite small — 6,000 acres, all told, once it was finally, poetically, brought to an end by snowfall on New Year’s Eve. But following the driest and second-warmest fall in 150 years, the devastation was harrowing out of proportion to its scale, since, unlike most wildfire, it was not in wildland or forest but was — as the climate scientist Daniel Swain, who lives in Boulder, put it — an “urban firestorm.”

That may sound like hyperbole, but on Friday the governor echoed the language: “It wasn’t a wildfire in the forest, it was a suburban and urban fire. The Costco we all shop at, the Target we buy our kids’ clothes at — all damaged.” In fact, though the fire did not begin there, it quickly jumped to a strip of big-box stores and their parking lots — to most Americans perhaps the very picture of an inflammable Anthropocene.* But as Swain told me on Friday when we spoke by phone, “Fire finds a way.” The way, typically, is wind; during the Marshall Fire, it carried flames and embers at hurricane-force speed for eight straight hours, consuming “football-field lengths of land in seconds.”

I wanted to start with what feels different about this fire. For a lot of Americans, the Camp Fire in Paradise, California, in 2018 may be the most horrifying recent memory — 18,000 structures burned, 85 deaths. But while Paradise was a relatively dense community, it was in what’s called the “wildland-urban interface” — essentially, houses in the woods. What we saw Thursday was so different — subdivisions, tract homes, fire just tearing through suburban environments we’ve been taught to think of as safe in every way.
There are lots of people calling this a Colorado forest fire. It was absolutely not a forest fire, by any stretch of the imagination. It arguably started as a wildfire, mainly as a brush and grass fire, with some woodland near the point of origin. But it was a wildfire that became an urban conflagration pretty quickly — just a matter of hours into the event. And that sets it apart in a lot of ways from some of the Western megafires we’ve been talking about in recent years, that burned hundred of thousands of acres, and sometimes burn hundreds or even thousands of homes, but they do so over a longer period of time.

And almost incidentally.
Whereas this was really burning through that urban environment.

The final footprint of this fire is probably gonna be about 6,000 acres, which is not small. It’s a large fire, but in the context of the kinds of fires we’ve been talking about in the wide-open spaces of the West, 6,000 acres is tiny. That’s a blip on the radar, if it occurred in a remote area these days. But this did not occur in a remote area.

The videos of the aftermath — I feel like I’m driving down a generic suburban street, except each of the houses is gone or burned all the way down to its bones.
This is all literally very close to home. We had gone to run errands at Costco, the one that was evacuated. And between where we live in south Boulder and the shopping center in Superior is Marshall Road. It’s actually very pretty. It has a mix of sparse homes and a pretty brushy and partly wooded area. It’s a gorgeous road to drive or bike along. But one of the things I’ve always thought ever since we moved here, as somebody who studies wildfires and climate change, every time we’re on that road, I thought, man, this aligns perfectly with these down-slope winds. If there’s ever a fire in this canyon, when the winds come up, it’s just going to blow through the canyon and beyond. And that’s unfortunately exactly what happened Thursday. We missed it by about five or ten minutes. We came through that canyon. We actually saw power lines down. The utility crew looked like they were already on scene. I thought, I hope they got a handle on that, because at that point, the winds are starting to blow at 70, 80 miles an hour and escalating. I thought, I hope that gets dealt with awfully fast, or there’s going to be a big problem. And then about 15 minutes later, we could see the smoke.

But actually the initial context is in that sparsely developed wildland-urban interface (WUI) that you mentioned, where there’s isolated or scattered homes. That is where the fire started. But with 90-to-100-mph wind gusts, within a half hour, it moved into a very different regime.

I think the next thing it hit was the shopping center, where there is a Costco and a Target and a bunch of other cookie-cutter suburban kind of stores. Superior is not one of those towns in the woodlands, in the WUI. It’s a sprawly suburb with a lot of tract homes. Shopping plazas with grotesque, large parking lots. The fire burned a lot of the structures in this context, which was kind shocking.

Yeah. How was that possible? I think of like a Costco in a parking lot as being about as nonflammable an environment as you can imagine. The parking lot is a huge expanse of concrete. The store itself is not exactly made out of wood. How does a fire actually power through that space rather than getting stopped by it?
You know, everybody in every physical science discipline has a similar quote within their field. In hydrology, it’s “water finds a way,” right? In Jurassic Park, it’s “life finds a way.” In the fire community, it’s “fire finds a way.” And really all you need is wind.

I don’t know if you saw the video from the Costco. I think it was taken like 10 minutes after we left. It looks like the apocalypse. And what you see in the video is everybody’s confused because it was windy when they came in, but it was also sunny. And then they go outside, and it’s pitch black except for the red glow.

And there is this sea of ash and embers, literally a sea flying around. And there’s other videos of just the landscaping in the parking lot: the brown grass, the trees they planted in the medians, they’re all just igniting from this ember storm.

And these big-box stores — they’re structures, they’re bad for a whole lot of other reasons, but you would think that the cinder-block frame would be fairly fire resistant. And it probably is, but under constant assault — they call it an ember attack. If you have a 90-mph sea of embers just constantly hitting the building, eventually it’s gonna find the gaps. It’s gonna find the air vents. It’s gonna find the gap between ceiling and the roof. And it finds its way in and those embers get a hold. And then the fire’s inside the building.

There’s a lot of flammable stuff in there.
A lot of the homes that burn in these kinds of fires don’t burn from the outside in, they actually burn from the inside out, because embers have blown inside and started a house fire that then burns from the inside house.

The Element Hotel in Superior, Colorado, which burned to the ground in the Marshall Fire. Photo: RJ Sangosti/MediaNews Group/The Denver Post via Getty Images

This fire — the first structures it hit you’d expect to be maximally resistant. But not only did those places burn, but then the fire kept going. It hopscotches from the median trees to the big-box stores, to the multistory hotel under construction to the Tesla dealership, and then it jumped the six-lane freeway, and then it landed on more homes and more homes.

And then at that point it effectively became an urban conflagration. It wasn’t really a wildfire at all anymore. It was burning, mainly, structures.

Homes, mostly.
The vegetation in between them was kind of just acting as a wick. But it wasn’t really the main thing that was burning. That was what happened in Coffey Park, in Santa Rosa.

That was in 2017.
That was definitely a wildfire. Initially, the ones that moved into the city limits, it was just burning house to house. And then there, too, it burned big-box stores.

Or thinking back to the Oakland Hills fire in the 1990s — that was only like a 2,500-acre fire, but it was massively devastating because it burned mainly through neighborhoods. There were some pockets of parkland, but it burned thousands of structures and killed dozens of people. I don’t think this is quite on that magnitude in terms of structures, at least. And I don’t think we know anything about casualties yet. But in other ways it was kind of similar.

On Twitter you called this fire in Colorado an “urban firestorm.”
This was something that we didn’t really have to think about for most of the 20th century, in relatively wealthy nations, with well-developed modern firefighting. Really the only major 20th-century urban fires in hose settings that occurred were the intentional ones during World War II, the fire bombings of Dresden and Japan. We just didn’t really have to think about that much, especially happening accidentally, in the latter half of the 20th century.

But now we’re seeing it again in the early part of the 21st century. And it’s due to a constellation of factors. Part of it is how that wildland-urban interface has greatly expanded, in some cases pretty recklessly — not only has it expanded, but it’s expanded in ways that make a lot of these structures in these neighborhoods quite vulnerable. It’s also because of climate.

Right, of course. All 20 of the largest wildfires in recorded Colorado history have taken place in the last 20 years.
But the fact that a fire can essentially become self-sustaining outside the wildland, in an urban environment — it’s a reemergence of an old problem. In centuries past, whole cities would burn down due to urban fires. You know, the Great Chicago Fire …

There’s actually a lot of cities in history — in some cases, they burned more than once.

But the reemergence of this phenomenon suggests two things. One, it suggests that the risks of the physical hazard of fire is increasing, which we know it is, for a range of reasons, including climate change and the historic policies of fire suppression and the lack of prescribed fire. But it’s also because we’ve sort of lost a lot of the knowledge and behavioral lessons of how to deal with fire in urban areas.

Fire consumes a home in a suburban neighborhood in Superior, Colorado, on December 30. Photo: Helen H. Richardson/MediaNews Group/The Denver Post via Getty Images

But some of these conditions make those lessons harder, don’t they? When wind is 80 or 90 or 100 mph, there’s not much firefighting you can do — which is something I think most Americans still don’t appreciate about these fires.
I mean, the winds were so strong Thursday that some of the flames that I was seeing were actually self-extinguishing for brief periods, because it was like blowing out a candle. Of course the net effect was extremely rapid spread, but during the most extreme gusts, when it was literally 90 miles an hour, the fire would lay down, you couldn’t see flames, it was just smoldering. And then there’d be a low in between gusts and it would flare up and then jump. And then the next gust would then bring those burning embers and throw them, you know, hundreds of yards in advance. I think that’s not uncommon in these really high wind conditions — the highly nonlinear spread. The spread isn’t continuous flames from horizon to horizon; the spread is these ember storms. And that’s why a lot of these homes catch. They might be pretty resilient to a grass fire just burning up to the edge of the brick wall. But if you have this sea or shower of embers coming down from above or blowing horizontally …

A lot of people have been yelling at me on Twitter. Why didn’t they put water on it? Which I think is just an incredible mismatch of scale. I think that there are a lot of people who don’t live in the West who still think that wildfires are, like, the size of someone’s backyard or something.

And even this one — as you said earlier, it’s considerably smaller than some of the legendary recent fires, but by any human standard it’s still enormous.
Yeah. I mean, it’s still miles on a side, over the course of a couple hours. But even a really small fire — imagine trying to put out a campfire-size fire if the winds are 90 miles an hour and there’s flammable stuff around, I mean, that’s not an easy task. If the winds are 90 miles an hour, the fire can be a mile away in minutes.

If you really want to stop a fire, you have to sort of address the underlying conditions rather than chasing the flames once they start.
There’s some language I’ve seen other folks using that I actually really thought was good and appropriate in this context, which is that events like this are climate enabled and weather driven. The climate-change signal is very strong, but it’s mainly in the preconditions. And if you try and do a climate-attribution study on Thursday’s weather, I don’t think you’re really going to find anything, but that would be also missing the point, because that’s not really where the climate signal would be coming from anyways.

People really like to simplify fire — wildfire in particular. They make sweeping claims about why it happened, what the risk factors were, what the context was. And I think that’s really problematic because there really is a lot of complexity baked into these things, as you well know.

There’s the geographic and ecological context — this fire was initially brush and grass fire, not a forest fire. And then there was this big urban component to it as well. But then there’s the weather and climate context. The weather context is pretty obvious. The winds were insane Thursday. I was going outside to do some photographic documentation and analysis, and I was wearing ballistics glasses like you’d wear in the metal shop in case something comes loose, because there were pebbles flying through the air at 90 miles per hour.

This is before the fire and the smoke, just to deal with the wind?
And during the fire, there was debris flying through the air at 80 or 90 miles an hour. Just to give a sense of how extreme the winds were.

I mean, there’s quite a bit of wind damage just, just from the winds themselves. And there’s lots of trees down and windows broken. We have chunks of the neighbor’s roof in our yard. The winds themselves were quite extreme. And, you know, the Front Range — Boulder, Fort Collins, these cities that sit right on the foothills, they are in that down-slope wind corridor. So it’s not like this is a place where you wouldn’t expect to see extreme down-slope wind. This is precisely where you’d expect to see those kinds of extreme winds. It’s also the time of year when you would tend to expect to see them — they’re most often strongest in winter. I think these were some of the most extreme such winds in the past decade or so, but they weren’t unprecedented.

But what was unprecedented were the antecedent conditions leading up to the winds. So the winds themselves, yes. But usually they don’t occur when conditions are as incredibly tinder-dry as they are right now. I mean, the autumn and the early winter this year did not feel like autumn or early winter. There were a lot of days in the 60s and 70s in Boulder, overnight lows above freezing. And keep in mind Boulder’s over 5,000 feet in elevation.

So that was extraordinary. And the summer before it was also really hot. So the lead-up to this was record warmth — and I’m not talking about a couple days before, I mean like the whole half year before. A lot of cumulative extreme warmth, and then also cumulative extreme dryness. This is one of the top-five driest such six-month periods on record, which notably followed a really wet period last spring, and so there was a tremendously snowy and wet spring, which led to a ton of grass and brush growth. And then we had record high temperatures and record dry conditions that dried everything out. And then we didn’t get any of that autumn precipitation. Normally Boulder would see at least a couple feet of snow by this point in the season. As of Thursday, we’d had about an inch cumulatively for the season. Essentially nothing. And, obviously, had there been snow on the ground, this wouldn’t have happened. And even if there hadn’t been snow on the ground, but there had been more precipitation and lower temperatures in recent months, there might have still been a fire, given these extreme winds, but it very plausibly would’ve had a very different outcome. It would have behaved differently.

And that’s sort of the key — that’s the crux of the wildfire-climate connection. It’s not like climate change is causing things to spontaneously ignite, but what it is doing is changing the character of wildfires. It’s expanding the window of what’s possible. It’s expanding wildfire season. It’s increasing the upper end of how intense fires can become — how hot they burn, how fast they move. And the faster moving and more intense they are, the more dangerous they are to us humans, the more, the more, you know, the more destructive they tend to be in terms of structures and homes lost. So it’s not that it would’ve been impossible for there to be a fire during an extreme wind event in December in this part of the world, because the winter here is the drier season — it’s not like California where the winter’s the wet season. But on the other hand, the record warmth and dryness leading up to this period definitely played a role in how dry things were. And we’ve kind of been seeing this time and time again.

What about the winds? You hear people talking about climate intensifying wind patterns, but there doesn’t seem to be much research yet to back that up.
There isn’t a lot of evidence of it yet. But there also hasn’t been a lot of study of it yet. So it’s more of an absence of evidence versus evidence of absence situation.

But even if we assume for a moment that the winds themselves don’t change at all with climate change, these antecedent conditions changing really matters a lot. Even if the winds don’t change but the fire season and the magnitude of the vegetation dryness increases, that matters a lot because of the sequencing.

That’s something I think that’s becoming increasingly clear in a bunch of different climate extremist perspective: it’s not just the incremental increases of this or that aspect. It’s that the natural events that we’re used to experiencing are somewhat dependent on some particular sequence of events. You know, the spring arrives and things warm up, it gets wetter and when the summer arrives, it gets hotter. The winter arrives, the snow comes, it gets colder. But baked into those seasonal transitions are specific types of weather that don’t actually occur all year round. So in California, for example, you get these strong offshore down-slope winds, mainly in autumn or early winter. You don’t get them in summer, the hottest time of year. That’s why autumn is peak fire season in California. It’s not because it’s the hottest season, it’s because the winds are most prevalent. And if you extend fire season by drying things out and warming things up later into the autumn, then you get the same season winds …

But instead of them coming during “fall” conditions, they’re effectively coming during “summer” conditions.
Because now you have summerlike dryness conditions all the way into autumn where you didn’t before. And that gives you a multiplicative increase in risk, just from that seasonal shift — more than you’d expect just from the increase in vegetation and dryness alone.

You’ve talked about these climate shifts, but there’s also the human contribution, the way we choose to live in areas which we know have at least some risk. Sometimes when we see a really devastating wildfire in the California forest burning through some homes, it can be tempting to think, Well, we just shouldn’t build there. But when the homes themselves are the fuel, and the fire isn’t burning primarily through forest, it raises a different set of questions.
I think that for a lot of those reasons, a lot of people here are shocked. I mean, people who have lived here for a long time, they’ve seen extreme winds, they’ve seen fires, but they’ve never seen the confluence of the extreme winds in a fire that just burns right into the highly populated suburbs and destroys a thousand homes. And I think that there’s a certain level of disbelief that some of the places that burned did in fact. I mean, I don’t think anybody would have been shocked had a fire burned hundreds of homes in the foothills above Boulder. I mean, everyone would’ve been horrified, but no one would’ve been shocked. Everyone knows it’s obviously at very high fire risk, those houses nestled in the woods. It wouldn’t come as a surprise to any of the people who live there.

But I think this probably did come genuinely as a surprise to a lot of the people who lived in the places that burned Thursday, in the sense that they were not living in that wildland landscape and yet these whole neighborhoods burned anyway.

The other twist is that Boulder is such a global hub for atmospheric science. And since Boulder is now so expensive, a lot of the scientists can’t live in it, so they live in the towns and cities immediately surrounding it. There were probably a lot of atmospheric- and earth-science people who were directly affected by this fire, much more than would’ve been the case had this occurred almost anywhere else in the world. So there’s going to be a lot of, I think, introspection. This has already kind of happened with scientists in California, but the expertise is more geographically dispersed there.

It’s interesting to hear you make the contrast with California. This is gross regional stereotyping, but the culture of California seems in certain ways to acknowledge and even celebrate the brutality of the landscape of the state. Whereas the cartoon of Colorado is Patagonia vests and hiking and everyone feeling they live in harmony with nature.
I don’t know if I necessarily agree that Coloradans live more in harmony with nature.

I don’t mean in reality, exactly, but in the mythology of the state.
Maybe there’s some of that in the mythology. But I think that’s more of a mountain Colorado mythology. Boulder’s kind of weird because it kind of straddles that line — it’s right up against the foothills, but it’s technically on the plains. And most of the people in Colorado live on the plains. The Denver Metro area is technically on the western fringe of the plains. And so there’s an interesting cultural divide with the mountain people and the plains people. Because out on the plains, as far as you can see in the eastern Denver suburbs, it’s suburban sprawl and hydraulic fracturing wells.

And it’s interesting that this happened right at the boundary, this fire literally started precisely at that geographic and cultural boundary, right at the base of the foothills and spread into this other world. This was a small fire that spanned a pretty wide and pretty unusual cross section of geography — physical and cultural and otherwise. But it’s also so recent. It’s still less than 24 hours ago that this started, so I think a lot of us haven’t processed it yet.

How’re you feeling?
I mean, I’m doing all right. We’re physically okay. It’s just kind of … another thing. It’s been a long list of disasters in 2021. And a lot of them have been distant. This one was not distant. And one thing we’ve been reflecting on is, had this exact fire footprint occurred maybe half a mile or a mile farther north — and there’s no reason it couldn’t have — it probably would have taken out a good portion of south Boulder, where we live. It’s just a matter of luck for us that this occurred where it did.

But I think a lot of people underestimate risk. I’m someone who’s always partly, because of my job, hyperfocused on risk. But we’ve been on that road so many times; it’s a pretty road, it’s a gorgeous landscape. So it’s obvious why people like to live there. But what are the risks that entails? And people say, “Oh, well, don’t live in a fire zone.” Okay, well, what about the flood zones? You know, what about tornado alley? What about sea-level rise? What about, you know, in New York City, people dying in their basement apartments because of the flash floods in the summer.

So when people say, “Where do we go?” And I get this question all the time now: “Where should we go? What’s the safest place?” they ask. And I have absolutely no idea how to answer that question other than I definitely wouldn’t live along the coast anymore for obvious reasons. Other than that, I don’t really know how to answer that question. And even the immediate coast issue is not easy if you already live there.

It’s a question that frustrates me because people expect there to be an answer.

They expect there has to be a way to eliminate risk, that there has to be an exit from climate risk.
They expect me, a climate scientist, to say, “Go to Nebraska or move to Canada.” I don’t know what they expect, maybe it’s New Zealand. But there is no good answer to that question.

The obvious line is that climate change is a global problem. But a more specific bit is that we keep getting surprised. I mean, I honestly don’t think any climate scientist would have honestly predicted that in 2021, the glacial valleys of British Columbia would see Death Valley–like temperatures. I mean, I’m still completely blown away by the fact that it was 120 degrees in British Columbia this summer. That’s just one example.

And I’ve heard people give British Columbia as the answer to the “where do we go” question. And those people probably weren’t thinking about fires, which have always been a problem there. But this year was an eye-opener nevertheless. It wasn’t just the heat dome, but the fires that followed, and then the mudslides and floods that followed that. Now they’re dealing with record low temperatures.
And I don’t think there’s anything special about British Columbia. That was the shocking event this year, but who knows what it’ll be next year. And this fire in Colorado is shocking given the context. There were plenty of fires in California this summer that did shocking things. The flooding in different parts of the world — New York and Western Europe and China and other places. It was also shocking. And there’s been lots of things that were so far outside of the historical envelope that they really gave people pause. And the point isn’t that these are risk hot spots, necessarily, it’s that this just happens to be where we observe these kinds of really extreme extremes recently. But, you know, next year it will probably be a different set of cities and regions where we see them. And in the decades to come, it’ll be other places.

It’s not about finding an escape from risk but choosing what kind of risk you’re comfortable with.
And that’s sort of how I try to answer the question. What worries you the most? I mean, if you have the luxury and the flexibility to actually choose where you’re going to live on this basis anyway, then that already presupposes certain things about what your status in global society is. That in itself helps you make that decision in certain ways. But some people are really freaked out by earthquakes because you just don’t know they’re coming. If you’re lucky and you have the smartphone app, maybe you’ll get ten seconds of warning. But if you’re in L.A. for the big one, you’re not gonna have an earthquake watch when you get a couple days to prepare, it’s just pretty much just gonna happen. You’re gonna have to deal with the consequences.

You’re never going to be in a place where there’s a hurricane that sneaks up on you and suddenly hits land. Fortunately, those sorts of things don’t happen.

You can see the storm coming.
Maybe if you don’t have the resources to do anything about it or leave, it doesn’t matter that you have great warning, but if you have those resources, a hurricane is never gonna take you by surprise. I think fires have sort of transcended that, though. It used to be the case that everyone could sort of assume you had time to figure out what you were going to do — that you could leave if you needed to leave. And I think we’ve seen some recent examples where that’s not always the case. Even the things people thought were out-runnable or predictable, or would come with meaningful warning — that’s not always the case.

As we talked about the last time we spoke, the smoke is a real issue there too. You can have a house that you know is pretty safe from fire risk, but that doesn’t mean that you’re insulated from toxic smoke.
That’s a really good point. The co-occurrence of really bad ozone days and extreme particulates days — which is mostly from smoke because most of the other sources have decreased — is actually increasing in the West pretty dramatically in the last couple of decades. And that’s pretty concerning from a public-health perspective. You can’t escape it, no matter where you live; it’s just everywhere. It even made it to the East Coast this year.

I think about the story from the Dust Bowl — it’s told in the Timothy Egan book The Worst Hard Time. There was a guy who was the head of the soil-conservation service at the time. There was this massive soil erosion because of agricultural practices, and the Dust Bowl was expanding and it was going to get to the point where it was just gonna turn the central part of the country into a permanent desert. And the soil scientist realized this. And I don’t know whether this was really the catalyst for what followed, but it did actually happen. He traveled to D.C. and he was giving congressional testimony on how bad things were and what needed to be done to fix it from a land-management policy perspective. And as he was about to enter the chamber, the sky got really dark outside. He realized it was a dust storm that had made it all the way to Washington from the Great Plains, the Dust Bowl region. And he dramatically opened the shutters and said, “Look outside, this is what it’s come to, it’s come for you here.”

Part of me was thinking this past summer, when the sky got red across New York and D.C., that it was sort of a similar moment. But I’m not sure that there was any equivalent character. And there certainly is not an equivalent Congress.

This article was updated to more clearly state that the fire did not begin in a shopping center, but moved quickly there after ignition.

More From This Series

See All
The Return of the Urban Firestorm