Fewer Hurricanes Expected
But conditions still indicate and above normal season.


T he hurricane research team at Colorado State University (CSU) has lowered by two its earlier prediction on the number of hurricanes to expect this year, but still says the 2006 Atlantic hurricane season will be above normal and that the probability of a landfalling hurricane is about 40 percent above the long-term average.

    Information the CSU team has obtained through July indicates this season will be more active than the 1950-2000 average (140 percent of average), now forecasting seven hurricanes, three intense storms (Category 3, 4 or 5), 15 named storms and 35 hurricane days—each figure above the average.

    CSU’s early August forecast is based on a newly devised extended range forecast procedure, which utilizes 57 years of past global reanalysis data. The revised forecast, released August 3 reduces the forecast from its early December 2005, early April 2006 and late May 2006 predictions due to small changes in June-July atmospheric and oceanic data that indicate conditions are less favorable for tropical cyclone development. The CSU team had predicted nine hurricanes, five intense storms and 17 named storms (see HP, Summer 2006, page 8).

LANDFALL PROBABILITIES
    Landfall probabilities for the East Coast of the United States are quite high this year, the team says. The odds of a major hurricane making landfall along the East Coast are more than twice the climatological average this year.

    For the Gulf Coast, landfall probabilities are slightly below average as steering current predictors for the Gulf Coast are mixed. However, it should be noted that Gulf Coast landfall probabilities are near the climatological averages, so coastal residents should prepare for a 26 percent probability of a landfalling major hurricane along the Gulf Coast, the team says.

    Landfall probabilities are based upon the team’s expectation for another active season as well as analysis of new steering current predictors for the East Coast and Gulf Coast.

ACTIVE AUGUST, SEPTEMBER
    That there have been two named storms and no hurricanes as of August 1 does not affect the team’s forecast. In fact, independent statistical predictors for August 2006 indicate well above-average activity and the CSU team is calling for considerable activity during the month. Predictor values for September 2006 are mixed, so the forecast is calling for slightly above-average activity for that month.

    The development of tropical cyclone activity predictions for individual months is a relatively new aspect of CSU’s climate research. There are often monthly periods within active and inactive Atlantic hurricane seasons that do not conform to the overall season. For example, 1961 was an active hurricane season, but there was no tropical cyclone activity during August. By contrast, the inactive season of 1941 had only six named storms, but four of them developed during September.

    The CSU team reduced its forecast from earlier predictions because of changes in the atmosphere and ocean during June and July. There have been no large changes in any particular predictor, the team says, but there has been a combination of changes that indicate this season is no longer likely to be as active as earlier predictions indicated.

    Physical features that have become less favorable for an active hurricane season include:
•    An increase in sea level pressure in the tropical Atlantic that indicates increased stability.
•    Stronger trade winds in the tropical Atlantic, which increases evaporation and cools Atlantic sea surface temperatures.
•    A decrease in tropical Atlantic sea surface temperatures. Cooler sea surface temperatures provide less heat (i.e., less fuel) for developing tropical cyclones.