Can Researchers Forecast Hurricanes Seasons a Decade in Advance?

By Andrew Moseman | November 8, 2010 12:15 pm

hurricaneearlEvery year, the coming of warmer weather in the spring brings with it the scientific parlor game of predicting how many storms the impending Atlantic hurricane season will bring. But could meteorological prognosticators soon begin to predict storms years in advance, and not just months, with some accuracy?

It is possible, a team led by Doug Smith of the U.K.’s Met Office says. In a study in Nature Geoscience, Smith essentially modeled the climate of past to see if the team’s system accurately predicted the Atlantic hurricane season.

The researchers used nine versions of its decadal prediction model to “hindcast” Atlantic hurricanes each year from 1960 to 2007. The model was set to May 1 for each of those years and then was asked how many storms would come that season. Averaging across the nine versions, the model results closely matched the changing number of hurricanes that occurred over those decades. Smith says: “We’ve found that there is some skill there.” [Science News]

Smith and colleagues then set the model to a time near the end of each year’s hurricane season—November 1—to see if it could predict the number of these storms for the following 10 years. Their model, called DePreSys for “decadal prediction system,” uses the temperature and salinity of the oceans, as well as pressure, winds, and temperature in the atmosphere as starting points. It performed well in predicting storms years down the road:

The number of Atlantic hurricanes in the study period varied from three to 15. DePreSys’s predictions, made for 10 years on from the date of the historical data, were on average within 19 per cent of the actual numbers. The team is now working on real predictions. [New Scientist]

Multi-year predictions have obvious benefits: They could allow governments and agencies to prepare for the blast of busy hurricane seasons that the models see coming down the road. The challenge for Smith’s team and others, however, is not only to hone their predictions of storm totals, but also to find out if it’s possible to predict intensity as well as frequency. The researchers’ data could also help to sort out the external factors that could be at play in hurricane frequency. They write: “In our experiments, the recent increase in tropical storm numbers was not caused by internal variability alone.”

Related Content:
80beats: Natural Disaster Report: Hurricane Threatens Haiti, Indonesian Volcano Erupts
80beats: Video: New “Disaster Lab” Simulates Hurricanes, Destroys Entire Houses
80beats: NASA’s Stunning Hurricane Pics Via Plane, Space Station & Satellite
80beats: Five Years Later, Could New Orleans Withstand Another Major Hurricane?

Image: NASA

CATEGORIZED UNDER: Environment
  • catmman

    Storm prediction does indeed have its benefits. I can see how this would be beneficial – financially to those doing the forecasting.

    Just predict a huge number of storms and beg governments for money to provide ‘data’ for protecting the populace, etc. Hell, just throw some numbers out, your bound to get a few years right. If you get a few ‘predictions’ wrong, just blame ‘unseen variables’ not the theory itself and pass it off as “Well, we were lucky. This time…” Win-win.

    As a tool for validating current predictive models and methodology, this cold be useful. But it won’t be used for that. As the article suggests, it’s already being touted in helping with global climate disruption theory.

  • VIP

    Living in a hurricane region of this world, my answer to your question is a definite no. Researchers are having great difficulties to predict hurricane activities on a day to day basis. If they could improve their predictions by a few days, it would be a great improvement. The assumption that anything beyond that time frame can be predicted is a joke.

  • Mike Saunders

    Dr. Jeff Masters often writes about how Hurricane Season predictions in January have almost no skill, so I’m pretty skeptical about this. It is quite easy to predict the past!

  • Brian Too

    It sounds like they are only making predictions in aggregate. Not that “a Category 3 storm will land here on such-and-such a day 10 years from now”. More like, “we think X major storms will materialize in the Carribean 10 years from now”.

    Correct?

NEW ON DISCOVER
OPEN
CITIZEN SCIENCE
ADVERTISEMENT

Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

80beats

80beats is DISCOVER's news aggregator, weaving together the choicest tidbits from the best articles covering the day's most compelling topics.
ADVERTISEMENT

See More

ADVERTISEMENT
Collapse bottom bar
+

Login to your Account

X
E-mail address:
Password:
Remember me
Forgot your password?
No problem. Click here to have it e-mailed to you.

Not Registered Yet?

Register now for FREE. Registration only takes a few minutes to complete. Register now »