In this photo provided by Cathy Pasquariello, a tornado approaches Ocean City, Md. on Thursday, Sept. 15, 2011. (AP)

Can we say anything about how many tornadoes we’ll have in 2012? That’s debatable but forecasters and research groups are starting to give it a try.

Forecasting companies AccuWeather, for the first time, and Telvent have issued 2012 tornado outlooks. On the academic side, Columbia University’s International Research Institute (IRI) for Climate and Society has published a new technique for developing seasonal tornado forecasts.

What are the forecasters predicting and how does the new research fit in?

AccuWeather predicts above-normal tornado numbers in 2012, but not as many as 2011. It says warmer-than-normal water in the Gulf of Mexico will provide the moist, unstable air to energize thunderstorm development and help spin-up tornadoes.

On the other hand, AccuWeather notes La Nina is forecast to weaken. The La Nina in 2011 led to a very strong northern jet stream which fueled violent storms as it dove into the Midwest and South.

Although La Nina is starting to fade, meteorologist Jeff Johnson of Telvent thinks it will continue to exert an influence on tornado activity.

“As we head toward the spring season our expectations are that 2012 will favor above normal tornado numbers once again, primarily due to La Nina,” Johnson said in a recent blog post.

But the Columbia IRI researchers aren’t convinced La Nina forecasts are necessarily useful in predicting tornado activity.

“...such a connection really has not been demonstrated in the historical data and hasn’t been shown to provide a basis for a skillful tornado activity forecast” said Michael Tippett (in an article in USA Today), lead author of a new study proposing a new tornado forecasting method.

Tippett’s new method provides about a month’s lead time on expected tornado activity. Using historical data, he and his colleagues identified two weather variables most associated with tornadoes, rain and spin in the atmosphere, and created an index that can be put into a model to make forecasts.

When the model was tested, it showed some skill, IRI said.

[The model] was able to use the index to forecast monthly tornado activity with some success up to a month in advance. This success, especially notable in June, is the first evidence for the predictability of monthly tornado activity.

Some are skeptical about the promise of such tornado forecasts. Writing for Earth Sky, meteorologist Matt Daniels talks about all of the complex factors involved in tornado prediction:

You need a lot of features to come together to produce tornadoes such as available moisture, high instability, high wind shear, and decent helicity (spin in the atmosphere). Not only are significant patterns such as a La Nina or El Nino recognition important, but noting snowpack in the north, the warm waters in the Gulf of Mexico, and monitoring the various oscillation indexes can be vital as well. We need to understand the strength and movement of the jet streams across the United States too.

Daniels said that if forecaster misdiagnose any of the many factors, the prediction may fail. He said this has been a problem in issuing general winter outlooks, for example.

Of course, even if perfect outlooks about the number of tornadoes were possible, what is the value? Tornadoes are so localized, that simply knowing in advance that a given month or season is going to be active may not be helpful if you can’t pinpoint where they’re going to occur. This same question even comes up in the discussions about the usefulness of seasonal hurricane outlooks, even though hurricanes affect much larger areas.

But Harold Brooks, a tornado researcher at NOAA, thinks these outlooks might offer some utility.

“...if I’m a state emergency manager I might be really interested in knowing at the end of March that by the end of April we could have a big problem,” he told IRI. “You could be better prepared with generators and supplies.”