Hurricanes are moving more slowly. That makes them more dangerous.

The globe's hurricanes have seen a striking slowdown in their speed of movement across landscapes and seascapes over the past 65 years, a finding that suggests rising rainfall and storm-surge risks, according to research reported Wednesday.

A study published Wednesday in the journal Nature finds a 10 percent slowdown in storm speed between 1949 and 2016. It points directly to the example of Hurricane Harvey, whose catastrophic rains were enabled by the storm's lingering in the Houston area for such a long period.

Slower-moving storms will rain more over a given area, will batter that area longer with their winds, and will pile up more water ahead of them as they approach shorelines, said Jim Kossin, a scientist with the National Oceanic and Atmospheric Administration and the study's author.

"Every one of the hazards that we know tropical cyclones carry with them, all of them are just going to stick around longer," Kossin said. "And so that's never a good thing."

The question of hurricane speed, and whether it would change under global warming, has drawn little attention in the past in comparison with more headline-grabbing questions, such as whether storms are getting stronger overall.

But Kossin decided to investigate it, based on the expectation that climate change is already altering the general, large-scale circulation of the atmosphere, within which hurricanes are embedded and by which they are steered. "Not quite like a cork in a stream, but similar," he said.

In particular, a slowing of circulation as the polar regions warm up faster than equator ought to slow down storm tracks, as well.

"I went in with that hypothesis and looked at the data, and out popped the signal that was much bigger than anything I was expecting," Kossin said.

And when it came to the storms' travel over land, the slowdown in some cases was even more pronounced than it was over the oceans or in general. In the Atlantic region, storms moved 20 percent slower over land, the study found.

The overall magnitude of the change in storm speed – a 10-degree reduction – was also striking in light of other changes expected in hurricanes under the amount of global warming that we have seen.

For instance, it is expected that hurricanes will rain about 7 to 10 percent more per degree Celsius (1.8 degrees Fahrenheit) of warming, as the atmosphere retains more water vapor, Kossin explained. But here was a 10 percent slowdown in storm movement speed with only a half- degree Celsius (.9 degrees Fahrenheit) of warming globally over the period he studied.

These two trends ought to work in tandem to make today's storms much worse rainmakers. But Kossin thinks the slower speed of movement – which naturally adds more rainfall to any region the storm crosses – may actually be a bigger deal than the simple increase in rain overall.

"It is plausible to say that the local rainfall impacts, the impacts from slowing, are equal to and possibly greater than the impacts from increased water vapor in the atmosphere," he said.

"We've kind of hypothesized that this type of behavior may happen, this slowing down of the forward speed of the cyclones," said Colin Zarzycki, a climate scientist at the National Center for Atmospheric Research who has reviewed at Kossin's study. "This is the first, to my knowledge, study that's tried to look at the historical record to try to quantify whether that's the case."

Zarzycki did raise a few questions, though – although he says he isn't sure how much they would actually change the result.

First, he noted that over the more than 60-year period of the study, there may be natural, decades-long cycles in the climate system that could affect the steering of storms and have little or nothing to do with global warming. So it isn't clear just how much of the change that Kossin found is actually attributable to human-induced climate change.

Kossin would actually agree on that point.

"My study is pretty far from an attribution study," he said. "I'm finding something that might be considered consistent [with climate change], but, really, no idea what's contributing what to this signal. At least not yet."

Zarzycki's second point is that our means of studying hurricanes have also changed. Indeed, after around 1980, we could observe them by geostationary satellite – before that, storms in the open ocean might have been missed completely and gone unrecorded, at least if they never encountered any vessel.

That means that storms farther from land in the earlier part of the study may not have had their speeds included in the study.

But Kossin, in his paper, writes that he wouldn't expect big changes in his results due to different means of measurement, since "estimates of tropical-cyclone position should be comparatively insensitive to such changes."

Overall, while scientists will need to dissect and better understand the new findings, it's hard to mistake the implication that intense, torrential rainfall associated with hurricanes could be getting worse when they make landfall because the storms are, basically, dragging out the punishment that they deliver to the places where they strike.

"Inland flooding, freshwater flooding, is taking over as the key mortality risk now associated with these storms," Kossin said. "There's been a sea change there in terms of what's dangerous. And, unfortunately, this signal would point to more freshwater flooding."