Predicting when small fires turn into big ones

Jerome Fox, a postdoctoral fellow at Harvard’s Whitesides Research Group, is studying the movement of fire.
Jerome Fox, a postdoctoral fellow at Harvard’s Whitesides Research Group, is studying the movement of fire.

It’s fun to encounter phenomena that all-knowing scientists can’t explain. Like, how do you predict the movement of a fire? It’s a basic question that’s proven to be more complicated than experiments can really account for.

But now comes Jerome Fox with the makings of an answer. Fox is a postdoctoral fellow at Harvard’s Whitesides Research Group with an interest in systems that amble along and then suddenly leap into an entirely new state, often without any obvious warning. These include financial markets, which often fly just before they crash, and ecosystems, which can seem fine one day and collapse the next. They also include forest fires.

“A fire is [the] very example of the unpredictable event,” says Fox. “It’s hard to predict where it will spread and how quickly.” This is especially true, he continues, when it comes to anticipating when a slow-moving, containable fire will erupt into something far more treacherous.


To try and learn how to predict these sudden leaps in intensity, Fox closely observed a series of controlled fires in his lab at Harvard, the results of which were published in PNAS Feb. 24. In particular, he was looking for evidence of what complex systems researchers call “critical slowing down” — the somewhat counterintuitive idea that something like a fire becomes more vulnerable to being jolted into a dramatically new phase, as it becomes more sluggish (which is another way of saying it reverts more slowly to its previous state after being affected by disturbances).

Get Arguable in your inbox:
Jeff Jacoby on everything from politics to pet peeves to the passions of the day.
Thank you for signing up! Sign up for more newsletters here

In the lab experiments, Fox set fire to one end of a highly flammable nitrocellulose strip. The middle of the strip contained a protruding fold, which was meant to “knock” the advancing fire as Fox observed it through a high-speed camera. This bump was likely to transition small flames that moved slowly along the strip into large flames that moved rapidly along the strip. Under conditions where such transitions were more likely, the small, slowly moving flames displayed symptoms of sluggishness: they flickered more sluggishly from side to side, and recovered more slowly from random disturbances like small movements of air, and irregularities in the surface supporting the strip. This is similar to the way that a spinning top behaves: When it’s spinning fast, it holds its position, and is less likely to be altered by, say, a pebble in its path; but when it’s slowing down, it wobbles violently, and even the slightest knock can cause it to transition into a whole new state (i.e. fall down).

Evidence that sluggishness is a harbinger of dramatic change is found in people with epilepsy — electrical fluctuations in the brain slow down just before a seizure hits — and it’s also implicit in the notion of “the calm before the storm.” Fox imagines this research could help to refine the way the US Forest Service models forest fires, and might even provide some visual cues that firefighters can use to anticipate trouble in the field.

Kevin Hartnett is a writer in South Carolina. He can be reached at