How Did the Los Angeles Fires Get So Out of Control?

The fires in and around Los Angeles have already claimed dozens of lives, destroyed thousands of homes, and led to evacuation orders for hundreds of thousands of people. The economic damage is projected to be as much as a hundred and fifty billion dollars. Daniel Swain is a climate scientist at U.C.L.A. and the University of California Agriculture and Natural Resources who studies extreme weather events and their connection to climate change. He and I recently spoke by phone. Our conversation, edited for length and clarity, is below. In it, we discuss what really caused these fires to rage out of control, how he tries to communicate the role that climate change plays in natural disasters, and how the Los Angeles area became so vulnerable to catastrophe.

What makes these fires unique other than the extent of the damage they have caused?

I do think the sheer scope of the damage in terms of the number of structures lost and the economic losses is what jumps out the most. That said, there are other aspects of the over-all situation that are very different from anything we’ve seen before, too.

Let’s start with what’s not unusual, since I think that’s helpful for setting the stage. It is not unusual for strong dry winds to occur in the mountains and valleys near Los Angeles in January. This particular event was notably strong and did extend across more of the high-population and lower-valley regions than usual. But if you were to pick a time of year when you would expect to see strong dry winds, it would be now.

These are the Santa Ana winds you’re speaking of?

Yeah, these are Santa Ana-like. These were a little bit different from a traditional Santa Ana event because they were driven by a slightly different low-pressure system in a slightly different spot. That is partly why the winds are coming more from the north than from the east, and also why they were stronger than usual and reached deeper into the valleys than usual. But if you can think of it as being Santa Ana or Santa Ana-adjacent, it’s the same general idea. We only see winds that strong every five or ten years or so. So it’s notable for sure, but it’s far from unprecedented in its own right.

The preconditions, though, were drastically more unusual, bordering on unprecedented—specifically how dry conditions have been. And that is essentially quantified by looking at how much rain has or has not fallen in the Los Angeles area or San Diego area. And what we find is that this is now either the driest or second-driest start to the season on record throughout Southern California, going back a hundred years. In modern history, it has not been this dry this late in the ostensible rainy season.

That is something that really sets the stage for these fires, because had these same winds occurred following, say, an inch or two of rain so far in the season, even if that’s below average, it’s still a good soaking. If that had occurred, we wouldn’t be seeing the fires that we’re currently seeing. We wouldn’t have that explosively dry vegetation. It essentially has not rained in Los Angeles since last spring—in many areas, about a tenth of an inch or less, which is insignificant from a wildfire perspective.

On top of that, the inland parts of Southern California—the mountains, the elevated plateaus, and the desert regions—experienced their hottest summer on record. The city of Los Angeles did not, to be clear. But then, in early September, even the city of Los Angeles and really the entire Los Angeles basin did experience a record-breaking heat wave. And that was actually associated with the major wildfire outbreak at the time, if you recall, in early September. Many structures were destroyed. It remained unusually warm and hasn’t rained at all since then. That is the real anomaly here, with the winds being sort of a second-order anomaly.

Anyone can understand why a dry twig is easier to burn. But is that too simplistic a way to think about why dryness leads to fires that can get out of control? Is there another level on which it is dangerous?

There are different levels, although I think the basic intuition gets you a good part of the way there. Imagine trying to light a damp log for a campfire. It’s not going to happen. There’s this binary switch, where if vegetation is too wet, you can’t get combustion at all. There won’t be any flame. You light a match and it’ll sputter out. You can’t start your campfire.

Then there’s another tier of dryness, where the wood might be not damp, but also not particularly dry. It’s somewhere in the middle. And if you’re very good at starting a campfire, you can probably get that wood going with some preheating, right? So if you get some twigs and some grass, the wood will eventually catch because the flame below it kind of dries it out enough.

But there’s also another level of dryness where the vegetation is extremely receptive to a spark. And not only does it make it more likely to ignite in the first place but it also greatly increases the actual intensity of the subsequent combustion or the fire. And by intensity, I literally mean the amount of thermal energy that it outputs. That matters a lot because it dictates not just how intense the flames are and how much damage they can cause in situ but it also affects how quickly those flames can spread. The more intense the fire generally, the faster potential spread it can have. And also, it increases the propensity of the fire to start to generate its own localized weather conditions which can sort of self-amplify.

So there’s these self-reinforcing, vicious-cycle feedbacks. Vegetation can be too wet to burn at all, in which case the risk of wildfire approaches zero, right? And then, there’s this other condition where it’s dry enough to burn, but maybe it won’t burn very intensely or very readily. And then, there is what’s known as critically dry vegetation, which will burn essentially at the drop of a hat and burn with great intensity, facilitating all of these cascading positive feedbacks in terms of increasingly extreme and exotic fire behavior, which are of course compounded if you have hurricane-force winds acting on those critically dry fuels, as we did this week.

I have seen you mention something called “atmospheric thirstiness,” but wasn’t sure what you meant. Is that connected to what you’re talking about?

It is, and the main connection between increasing wildfire risk and climate change is actually through this same process, the aridity or the dryness of the vegetation itself, which is of course what eventually becomes fuel for wildfires. By definition, a wildfire is a fire that’s burning vegetation. And there are really two ways to dry out vegetation pretty quickly. One of them is to, on the supply side, not deliver any water. In other words, if there’s a lack of rain or snow, lack of precipitation, the soil becomes dry and the plants don’t really have water to work with in the first place.

There’s also another way, which is even if there’s plenty of precipitation, but if there’s an excess of evaporation and transpiration—evaporation being the passive part of the process and transpiration being the part of the process that happens when water passes through living plants and comes out the stomata in the leaves—there is something known as atmospheric evaporative demand. That’s the technical term for thirstiness, and it essentially represents the propensity of air to extract water or evaporate water from surfaces or from plants. Surfaces can mean bodies of water or it can be the soil. Soil usually has at least some water in it, and so therefore water that can potentially be evaporated, and will be evaporated at an increasing rate the greater the evaporative demand or thirstiness.

And the warmer the temperature, the greater the thirstiness. So with rising temperatures, we see increases in atmospheric evaporative demand, increases in the propensity of that air to draw moisture out of both living and dead plants, out of the soil, out of wooden fences in people’s back yards or brush in the freeway medium. Whatever it is, that increased atmospheric evaporative demand is going to desiccate the stuff that could potentially burn in a fire.

With climate change, temperatures are getting warmer, but at the same time we see increases in extreme temperatures in both directions. Has that been an exacerbating factor at all?

I’ve been hearing this a lot lately and I’m very curious where it comes from—the notion that we’re seeing both hotter and colder temperatures, more extreme temperatures. There’s no evidence that we’re seeing colder colds. We’re certainly seeing hotter hots, but it’s all moving up. We’re sort of on that escalator, unfortunately.

There are not a lot of places that are experiencing increasing extreme weather in the other direction?

We’re not really seeing any locations on Earth that are having more cold extremes. We do still see episodically these sorts of cold outbreaks in places that have them historically. But they actually are significantly less common in these places than they used to be, even though they still do occur.

While that’s true with temperature, it’s actually a different story when it comes to hydroclimate, or precipitation. And this gets to the other part I wanted to address, which is this notion of increasing hydroclimate whiplash. We’re not seeing increasing temperature whiplash. It’s just getting hotter and hotter and hotter. We still have cold periods occasionally, but they’re fewer and farther between, and they’re not quite as cold as they would’ve been. We’re certainly seeing more extreme heat waves and warmer temperatures on average.

When it comes to precipitation, though, it gets a lot trickier. We see an increased frequency and magnitude of swings between extremely wet and extremely dry conditions. That turns out to be exactly what happened in Southern California this year—in 2024 into 2025—and is another part of the story. The last two winters in Southern California were some of the wettest on record, following historically severe drought for the better part of the last decade.

And yet, what we see is that in some parts of Southern California, the risk of wildfire increases following wet years. Why is that? The reason is that the primary vegetation type in the places that are currently experiencing these fires is grass and brush rather than trees in a forest. And if you know anything about grasslands, you actually get a lot of extra growth of grass and brush following wet favorable conditions for growing. That’s why you water your lawn.

And so, following the wet period, there’s an overabundance of grass. In fact, in 2024, over the summer, there was twice the average amount of biomass in grasslands and scrublands in coastal Southern California due to these two wet winters in a row. But then we had periods of highly unusual to record-breaking warmth, and then followed it with the driest start to the winter on record. You can kind of see where I’m going with this.

That wet-to-dry whiplash sequence of hydroclimate actually helped facilitate the preconditions for these fires. Not only did we add more potential fuel, literally, to the fires, based on the wet conditions, but then we followed those up with some of the driest conditions on record in the very same places that had this huge accumulation of grass and brush.

Is there any reason to think that those extremely wet seasons were connected to climate change?

Actually, we do expect to see both wetter wets and drier dries. It’s twofold.

O.K. Just not colder colds.

Right. So we do expect to see an increasing likelihood of really wet winters, like the ones we just saw, and really dry winters, like the one we’re now seeing. A lot of folks will say, “You can’t have it both ways. You’ve got to pick one.” Well, guess what? The climate system doesn’t want to pick one. And the evidence shows that we actually will see more of both, leading to a situation where the average precipitation might remain roughly the same, leading to a very misleading indicator.

How do you think about attributing specific events to climate change? Without getting into discussion about this fire, per se, I’m just sort of curious about your analytical framework.

I think the challenge here is there’s rarely, if ever, a singular cause for any complex event. Often there’s a real tendency for folks to get sucked in by sort of disingenuous straw-man binaries, if you will, that either this event was caused by x, or this event was caused by y, or z—whatever your favorite pet physical process or political deficiency is in a given context.

And the reality is, it is almost always the result of the confluence of multiple interacting factors that are relevant to different degrees in different contexts, which is exactly the opposite of the answer that you’d want to give in the context of a sound bite. But it is the most honest answer and the one that helps you understand what’s going on the most clearly.

In the case of wildfires, for example, there’s no question that wildfire disasters themselves are getting worse. There is really alarming acceleration in structure losses and life loss and acres burned. But that’s the result of several different things interacting, one of which is climate change for sure. But there are other factors, including things like the expanding bull’s-eye effect, this notion that there’s more people than ever who are living in high-residential zones, more people moving into places that they didn’t use to be in, which were always at risk of fire, and therefore having more structures, and more assets, at risk.

There’s also, in some settings, questions about land management. Forest management came up again this time, even though none of these fires are primarily burning in forests, so it’s really truly a red herring in this context. But in other contexts it is a fair point to raise. And here, there are other land-management considerations, too.

When we talk about the links between climate change and wildfire, or climate change and any extreme weather event—hurricanes or severe thunderstorms or extreme heat waves—the line that you can never link a specific event to climate change is out of date, I think, even though there is still nuance there.

The question I often get from journalists is, “Did climate change cause event x?” I always turn that question immediately around to them and say, “Is there any other noun or collective noun other than climate change to which anybody could ever give an affirmative answer to that question?” Was the wildfire caused by political incompetence? Was it caused by winds? Was it caused by whatever process initiated the spark that started the conflagration?

A lot of people will say yes to each of these things, and they’re all both right and wrong because all of those things were necessary, but insufficient on their own for the event to transpire as it did. The reframing of that question that I really encourage folks to use, and partly because this turns into a scientifically answerable question, is: To what extent did climate change increase the likelihood or severity or some other quantitative measurable characteristic of an event?

This is so-called attribution science, correct?

Yeah. And in the context of California wildfire, for example, we put out a study in 2020 where we looked and said, We can’t control for all of these factors at once. So let’s do an analysis where we only look at the piece that we could actually do a controlled experiment with in the setting, which is let’s actually look at the role of climate change.

If you know that one of the greatest risk factors for catastrophic fires is extreme fire weather, which can be quantified in terms of wind and humidity and vegetation dryness and antecedent rainfall and all that kind of stuff, if you create that index and you calculate how it’s changed over time, you get a sense of how extreme wildfire-burning conditions have changed over time. You don’t get a sense of whether there’s more people in harm’s way than there used to be. You can’t determine whether or not certain ignition sources are becoming more or less common over time.

But you can isolate the effect of changes in weather and climate conditions over decades. And when we did that, we found that extreme wildfire weather conditions had doubled in California between 1980 and 2020. That’s not a small increase. And that’s the way we try to ask these questions. ♦