Before slamming into islands in Grenada this month, Hurricane Beryl strengthened from a tropical cyclone to a major hurricane in less than a day. The storm 鈥� the strongest on record so early in the season 鈥� killed several people and caused widespread damage on the islands. Over the next week it spun over other parts of the Caribbean, Latin America and the US Gulf Coast, where more than 300,000 homes and businesses remain without power amid a heat wave.
But Beryl鈥檚 consequences might have been worse were it not for new hurricane forecasts able to predict such 鈥渞apid intensification鈥� events more reliably. 鈥淸The forecast] was indicating this thing was really going to go,鈥� says at the US National Oceanic and Atmospheric Administration (NOAA), which has been working for years to improve its rapid intensification forecasts.
Rapid intensification events 鈥� defined as an increase in wind speed of about 55 kilometres per hour or more within a span of 24 hours 鈥� can make hurricanes more dangerous by leaving people with little time to prepare, and are becoming more likely due to climate change. But rapid intensification is notoriously challenging to forecast.
Advertisement
鈥淲e conceptually understand what contributes to rapid intensification,鈥� says at the University at Albany in New York. 鈥淧redicting and forecasting rapid intensification is a completely different question.鈥�
As recently as 2015, forecasts of rapid intensification were completely unreliable, and recently improved methods are still far from perfect. But this year, NOAA is using new forecasting models that could be the most reliable yet.
One 鈥� called SDCON-RII 鈥� all other models when tested on the past five Atlantic hurricane seasons. This stems from higher resolution modelling, more accurate physics and incorporating more observations from aircraft, satellites and other sensors to ensure models start with the correct initial conditions, says at NOAA.
In Beryl鈥檚 case, these models performed 鈥減retty well鈥�, says DeMaria. Of the storm鈥檚 nine separate rapid intensification events, NOAA鈥檚 forecasts accurately predicted four of them at least 24 hours in advance, whereas past forecasts might have predicted just one, he says. However, the forecast has limitations 鈥� it didn鈥檛 accurately predict the storm鈥檚 peak intensity and didn鈥檛 capture more than half of the events. 鈥淭here鈥檚 still potential for big misses,鈥� says DeMaria.
The main drivers of rapid intensification are lots of heat in the ocean, high humidity and a lack of wind shear, which weakens storms. Under those conditions, a hurricane can draw more air to its centre and efficiently gather energy from the ocean, strengthening its vortex somewhat like a spinning figure skater drawing their arms inward to spin faster, says Tang.
When conditions are ripe, as they were with Beryl, Tang says forecasting is easier. Things are more challenging when conditions aren鈥檛 clearly aligned for rapid intensification. Subtle factors such as eddies at the surface of the ocean or thunderstorms around the eye of the storm can set off rapid intensification. But these are difficult to measure and the physics behind their influence on storms isn鈥檛 fully understood, says Tang.
Where physics falters, statistics can help. But forecasts of rapid intensification based on statistical or have long been hindered by a lack of data on these relatively rare events, says at Pacific Northwest National Laboratory in Washington state.
Now, thanks to boosts in computing power, data collection and physical modelling, forecasts have steadily advanced, with reliability increasing between 2015 and 2020. A model NOAA introduced in 2022 called the (HAFS), which was designed to improve forecasts of intensity and the path of storms, accurately predicted the rapid intensification of 2022鈥檚 Hurricane Ian.
However, the following year, HAFS, along with all other models, failed to forecast Hurricane Otis鈥檚 rapid intensification before it hit Mexico, making clear there is still a lot of work to be done.
But researchers are optimistic that forecasting rapid intensification is a tractable problem. And as ocean temperatures rise with climate change, forecasters may have more to work with. 鈥淭here have been a lot more hurricanes,鈥� says DeMaria. 鈥淭he dataset is bigger.鈥�
