seasonal forecast

Heaviest Weather Still to Come
CSU team sees more activity as the season progresses.

Information obtained through July 2007 continues to indicate that the 2007 Atlantic hurricane season will be more active than the average 1950-2000 season, reports the Colorado State University (CSU) forecast team. While the team has reduced its earlier April and May forecasts (see HP, Show Issue, 2007, page 6), it still anticipates eight hurricanes this season and four intense hurricanes (Category 3, 4 or 5). The average is 5.9 and 2.3, respectively.

The CSU team’s early August forecast is based on a newly devised extended range statistical forecast procedure that was developed on 40 years of past global reanalysis data. It then was tested on an additional 15 years of global data.

“We have lowered our forecast from our early April and May predictions due to slightly less favorable conditions in the tropical Atlantic,” said Phil Klotzbach, author of the report. Sea surface temperature anomalies have cooled across the tropical Atlantic in recent weeks, and several significant dust outbreaks from Africa signify a generally stable air mass over the tropical Atlantic, the report stated.

SEASON EXTENDED
This year’s hurricane season has gotten off to a fairly slow start primarily due to unfavorable thermodynamic conditions, a primary governing factor. However, the CSU team notes that there often are monthly periods within active and inactive seasons that do not conform to the overall season.

For example, this year, the team believes August will have activity at slightly above-average levels with two hurricanes predicted (1.6 is the average), but that September is likely to be more active and October and November are expected to be “very active.”

“Climatologically, September is the most active month of the hurricane season,” the CSU team reported. The team predicts four hurricanes developing in September (average is 2.4) with two reaching intense levels (average is 1.3). A combination of factors makes this development likely: low values of vertical wind shear, sea surface temperatures in the tropical Atlantic reaching their maximum values and surface pressures reaching their lowest values.

Typically, the end of the season is governed by rising values of vertical wind sheer, but with either cool or weak La Niña conditions expected this year “the end of the Atlantic basin hurricane [season] will likely be extended this year,” the team reported.

LANDFALL LIKELY
A significant focus of the CSU team’s research involves efforts to develop forecasts of the probability of a hurricane making landfall along the U.S. coastline. Whereas individual hurricane landfall events cannot be forecast months in advance, the total seasonal probability of landfall can be forecast.

This season, the team projects a 95 percent probability that one or more hurricanes will make landfall along the entire U.S. coast. For the Gulf Coast region there is a 76 percent probability of a landfall and for eastern Florida and the U.S. East Coast there is a 78 percent probability of one or more hurricanes making landfall.

For eastern Florida and the East Coast the probability of a Category 1 or 2 hurricane specifically making landfall is 60 percent and 44 percent for a Category 3, 4 or 5.

NATURALLY OCCURRING CYCLE
Arguments for the role that global warming plays in increasing the number or intensity of hurricanes have been given much attention by the media. The CSU team reports, however, that despite the global warming of the sea surface that has taken place over the last three decades, the global numbers of hurricanes and their intensities have not shown increases in recent years except in the Atlantic basin were recent hurricane increases are likely the result of naturally occurring multi-decadal variations.

“There have been similar past periods (1940s and 1950s) when the Atlantic was just as active as in recent years,” the team reports. When Atlantic basin hurricane numbers are compared over the 15-year period of 1990 to 2004 to an earlier 15-year period (1950 to 1964), there is “no difference in hurricane frequency or intensity even though the global surface temperatures were cooler [back then].”

Now in its 24th year, the CSU forecast was prepared by Phil Klotzbach, Ph.D., and William Gray, who has prepared the forecasts since 1984. A review of the team’s early August forecasts and actual observed hurricane activity from 1984 to 2006 (see graph) shows that when the team predicts an above-average or a below-average season, the season tends to be above or below average even if the exact numbers do not verify.