Climate scientists can predict hurricane strength and trajectories very well a few days out. They can anticipate hurricanes fairly well a few weeks before landfall. Sometimes they can even tell that a storm is likely to form the winter before hurricane season starts, although the measurements are rougher and they may not be able to tell if the hurricane will reach land.

But when it comes to predicting what the longer-term future holds in store where hurricanes are concerned, climatologists, like the rest of us, are a little out to sea. They cannot accurately predict whether the seeming increase in severe storms will continue. 

“There’s different ranges at which we can make different sorts of predictions,” says Adam Sobel, an atmospheric scientist at Columbia University. “Generally speaking, the longer the timescale, the less specific we can get.”

For example, we’re not necessarily out of the woods yet this season. “We are just at the peak of the season. There’s Jose out there now,” Sobel says. “I wouldn't be surprised if there are more [upcoming hurricanes], but I don’t know if they’ll hit anyone.”

Weather and climate prediction is a “rapidly improving area of science,” according to Sobel. “I think what we do predict well is the track and the intensity in terms of the peak wind,” says Frank Marks, director of the Hurricane Research Division at the National Oceanic and Atmospheric Administration (NOAA). “We’re good, but we’re not good enough.”

In May, NOAA published accurate predictions that this year’s hurricane season would be above normal. Luckily, knowing Irma’s general trajectory four days out was enough time for most people to decide whether they needed to evacuate. However, we won’t soon be able to tell what kinds of storms we should anticipate in the coming years.

Sobel notes that some in the insurance industry try to make projections of “elevated risk periods” of five to 10 years out, but he isn’t convinced those projections are very accurate.

In the time it takes to plan a vacation to Florida, hurricane predictions could easily change—although Marks says that he wouldn’t plan a cruise to the Gulf of Mexico during hurricane season, at least not without travel insurance. But that doesn’t mean you shouldn’t ever visit areas in the Atlantic that could be affected. Devastating hurricanes like Harvey and Irma are still relatively rare, and, in a worst-case scenario, you can cancel travel plans.

However, it’s a different story if you want to make a long-term investment in low-lying coastal areas. With sea levels rising and unpredictable storms, it wouldn’t be wise to build a house in coastal Florida, Sobel warns. People should make long-term decisions about infrastructure, building and investments while considering climate change, natural disasters, and unpredictable weather events. Marks adds that it’s unwise to build “pretty much anywhere on the coastline that is hurricane-prone, such as the Gulf of Mexico.”

“If we were rational, we would be taking into account the risk of hurricanes as well as other natural disasters,” Sobel says. “We should take into account these risks more as we plan and develop and inhabit these vulnerable places in our coastline and planet.”

“Whether or not we have more [hurricanes in upcoming years] doesn’t matter, because we have them,” Marks warns.