Why It’s So Hard to Tell When a Hurricane Will Blow Up

Hurricane Michael, which underwent rapid intensification.
Hurricane Michael, which underwent rapid intensification.
Image: Colorado State University

If you’ve paid attention to the past two Atlantic hurricane seasons, you’ve almost certainly seen the term rapid intensification thrown about. Hurricane Michael is the most recent example, stirring from a Category 1 to strong Category 4 storm in 48 hours.


It’s the latest in a murderer’s row of hurricanes that have undergone rapid-intensification, a technical term for when sustained winds increase at least 35 mph over a 24-hour period. Harvey, Maria, and Jose all did it last year, as did Florence this year, and there have also similarly explosive storms in other basins. There’s evidence the strongest storms intensify faster than they used to.

All of this is causing scientists to take a closer look at the processes underlying rapid intensification. While we have a general understanding of what can cause storms to blow up, predicting exactly which ones will is a developing area of science that’s increasingly relevant in the warming world.

The basic ingredients for a cyclone to undergo rapid intensification are calm upper levels winds, and warm water to fuel the storm. That said, things are never that simple.

Hurricanes form in the complex, chaotic mess of interactions between the ocean and atmosphere. There are a million games of telephone going on both between the sea and storm above it, within the storm itself, and across the atmosphere. All those conversations have some role in dictating whether a given storm will blow up or not, and it’s up to meteorologists to piece it all together.

That’s where weather models come in. Models underlie our forecasts, providing snapshots of what might happen. They’re also inherently imperfect, and getting a handle on some of the finer grained details—like thunderstorms swirling around the eye of the storm—is hard because it takes a lot of computing power.

Kristen Corbosiero, a hurricane researcher at the University of Albany, told Earther that models tend to simulate the environment around a storm at lower resolution than the storm itself. “[T]his is generally OK because the large-scale features of the atmosphere that interact with storms are better sampled by observations from satellites,” she wrote in an email.


But even capturing those big picture features of the atmosphere doesn’t always help us predict a storm’s imminent ascension. Corbosiero pointed to challenges with getting a handle on Michael’s rapid rise in strength owing to a dip in the jet stream to the north of the storm. Those interactions are notoriously hard for some models to get a handle on.

For all the challenges models face, there’s another layer that makes forecasting rapid intensification especially hard: humans. Specifically, meteorologists. The models they rely on have their quirks and biases and meteorologists have the unenviable task of looking at them and observations and making a judgment call about how to weight all that in their final forecast.


“You’ll have certain models that forecast a lot of storms to intensify rapidly,” Matt Lanza, a a Houston-based meteorologist and managing editor of Space City Weather, told Earther in a Twitter direct message. “Many of them never do. But they will ‘perform best’ with the storms that do intensify because they (those models) are biased to the strong side. So the biggest challenge I find as an operational forecaster is ‘when do I actually decide to believe this model?’”

Ahead of Michael, the National Hurricane Center forecast, rapid intensification but undercooked the storm’s 155-mph fury in its early forecasts. Other meteorologists were banging the drum that the storm could reach Category 4 status owing to how they weighed some of the differences in models.


Point is, there’s an art and science to forecasting rapid intensification. Climate change will only up the stakes, as warming oceans that can provide more fuel for hurricanes and a warming atmosphere is primed to hold more water. Multiple avenues of research suggest we could face more rapid intensifying hurricanes, which makes getting better at predicting explosive storms is of utmost importance.

Corbosiero pointed to “some really cool observational techniques” in their early stages that could help improve our understanding of rapid intensification. They include looking at lightning in the storm’s eyewall and bands of rains circling it, as well as the ring of shallow clouds swirling around the eye.


“This will be a decadal type problem where incremental improvement will likely continue as we learn model biases and establish a larger database of cases to research and understand so we can translate lessons learned from them to operations,” Lanza said.

In other words, there may also not be any singular aha! moment. Which is par for the course in science.


Managing editor, Earther


Dense non aqueous phase liquid

When talking about solving a bunch of unknowns you need an equal bunch of equations and they have to be solved simultaneously...

I like this from Brian Kahn

There are a million games of telephone going on both between the sea and storm above it, within the storm itself, and across the atmosphere.

I don’t like this, as an example of many:

An Implicitly Balanced Hurricane Model with Physics-Based Preconditioning

In this study the Jacobian-free Newton–Krylov (JFNK) approach will be used to solve an implicitly balanced version of a physical model consisting of the Navier–Stokes equations with additional equations representing cloud processes. As discussed in Knoll et al. (2003), implicitly balanced refers to function evaluations occurring at the same instant in time, thus implying for the Crank–Nicolson solution procedure that all forcing terms in the physical model are computed at time level n + 1/2. In this paper, time splitting will be defined as not being implicitly balanced or a solution procedure employing function evaluations at differing time levels. Currently, most nonhydrostatic models used to simulate hurricanes (Davis and Bosart 2002) employ time splitting in their construction and thus are not implicitly balanced. A key purpose of the hurricane simulations to be presented in this paper will be to demonstrate that, by employing a time-stepping procedure that is implicitly balanced, highly time accurate solutions can be obtained more efficiently than from a time-split solution procedure.

That’s out of my league. My head really hurts - and I only got through the intro. Hey, I made it past the abstract so give me some credit.

Brian’s description was actually brilliant. Brilliant brilliant. Not the British brilliant.