Hurricane Otis turned into a monster overnight. Forecasters don't know why

The normally bustling beach resort town of Acapulco, Mexico, looks like a scene out of a disaster movie: palm trees stripped of its fronds; windows of hotels and homes shattered; twisted metal signs strewn across streets. All wreckage wrought after Category 5 Hurricane Otis slammed into the region on Wednesday just after midnight local time.

In the end, 27 people lost their lives.

The hurricane took everyone by surprise, including forecasters. The lingering question of why remains.

Otis began as a tropical depression (as all hurricanes do) on Sunday afternoon before transitioning into a tropical storm hours later. 

The U.S. National Hurricane Center (NHC), a world leader in tropical storm forecasting, was closely monitoring it. 

By Monday evening, it forecast tropical storm conditions and "possible" hurricane conditions beginning late Tuesday. 

But that's not what happened. 

Instead Otis went from winds of 100 km/h to 268 km/h in 24 hours from Tuesday to Wednesday. By Wednesday evening, the NHC called it a "nightmare scenario." 

This type of strengthening is referred to as "rapid intensification." While these types of hurricanes are becoming more common, it was Otis's magnitude that was so shocking.

To put it into perspective, the NHC defines rapid intensification as storm winds that increase by 55 km/h in a 24-hour period. Otis more than doubled that.

"We were expecting maybe it could be a strong tropical storm or hurricane. Some of the models were kind of getting to that, but… even when it started to intensify as much as it did, I was kind of thinking maybe it would peak out at a Cat 3," said Andy Hazelton, an assistant scientist at the University of Miami's Cooperative Institute for Marine & Atmospheric Studies who specializes in hurricanes.

"But it just kept going all the way up to landfall, which is sort of a worst-case scenario, especially when you're in that populated of an area."

Weather forecasters don't rely on just one model, but look at many different ones. The problem with Hurricane Otis was that not one model predicted this rapid intensification, especially to that degree.

So the question is: Why not?