Predicting the 2008 Atlantic Hurricane Season

Fresh from eating crow over their busted 2007 hurricane forecasts, Phil Klotzbach and William Gray are back with an even tougher endeavor: a long-range pre-season forecast (PDF) of what kind of hurricane activity we will get in the Atlantic in 2008.

According to the pair, 2008 will be a lot like 2007: slightly above average, with 13 named storms, 7 hurricanes, and 3 hurricanes of Category 3 strength or higher. 2007, for comparison, had 14 named storms, 6 hurricanes, and 2 intense ones.

The way that Klotzbach and Gray do these long range forecasts is to look at how a number of current climate parameters (or "predictors") are shaping up, and then go back in history and find out how hurricane seasons in the past have turned out when preceded by these same or similar parameters. By these lights, they think that some good analogues for 2008 would be the 1953, 1956, 1989, 1999, and 2000 hurricane seasons.

Atlantic Tropical Storms, 2000 – NOAA

We should, of course, take this with a massive dousing of salt. As is now well established, the chief factor that seems to influence activity levels for Atlantic hurricanes is whether or not there's an El Nino situation in the Pacific ocean. El Nino tends to suppress Atlantic storm activity, while its opposite, La Nina, tends to enhance it.

We're in a moderate La Nina now, but there's no telling where we'll be by the time the 2008 Atlantic hurricane season rolls around in June of next year – much less where we'll be during the peak months of August and September. As Jeff Masters puts it, commenting on the Klotzbach/Gray prediction: "Until we can forecast the evolution of El Nino more than six months in advance, December forecasts of Atlantic hurricane activity are merely interesting mental exercises that don't deserve the media attention they get."

And indeed, Klotzbach and Gray have not done very well with their December forecasts of late: They underestimated the dramatic 2005 Atlantic hurricane season and overestimated 2006. For 2007, they were actually pretty close in December – they said 14 named storms, 7 hurricanes, and 3 intense ones, when the reality, again, was 14, 6, and 2 – but the trouble here was that they upped the forecast as the season approached, rather than sticking with this initial prediction.

There is a lot of pessimism about these long range hurricane forecasts, and understandably so. Still, my view is that it takes an admirable amount of courage to be willing to put a prediction like this out there – and as Klotzbach and Gray argue, that's really the only way you learn something.

So while I wouldn't base my personal or financial decisions on this forecast, it's informative and gets dialogue started about the next hurricane year – something that simply can't happen soon enough.