Nine of the storms will strengthen into hurricanes, with winds of at least 74 mph (119 kph), the team said, reaffirming an early prediction made in December and updated to include current trends like the La Nina weather phenomenon, cool Pacific waters and an abnormally warm Atlantic.

The Colorado State forecasters said five of the hurricanes were likely to be major storms, reaching at least Category 3 on the five-step Saffir-Simpson scale of hurricane intensity, and boasting winds of at least 111 mph (178 kph). Storms of Category 3 strength and above cause the most destruction.

But they also said there were likely to be fewer major storms making landfall in the United States compared to 2005, when virtually every hurricane record was broken, and also 2004, when Florida was bashed by four consecutive hurricanes.

"Even though we expect to see the current active period of Atlantic major hurricane activity to continue for another 15-20 years, it is statistically unlikely that the coming 2006-2007 hurricane seasons, or the seasons that follow, will have the number of major hurricane US landfall events as we have seen in 2004-2005," Gray said in a statement.

HARD TO PREDICT

Gray's predictions are valued by companies but their accuracy can be difficult to gauge because they are revised regularly as a season progresses.

The Colorado State team, for example, initially predicted 13 storms for 2005 and raised the forecast in May last year to 15. It wasn't until Aug. 5 -- almost halfway through the season-- that Gray increased the prediction to 20 storms. In the event, 2005 saw a record 27 named storms, of which 15 became hurricanes.

Last year was the costliest and most destructive season ever, with US$80 billion in damages from Katrina alone.

Hurricane Rita hammered Texas and Hurricane Wilma became the most intense Atlantic hurricane observed before slamming into the Mexican resort of Cancun and then curving back over South Florida, where it caused US$10 billion in damage.

Hurricane Stan, meanwhile, killed up to 2,000 people across Central America.

The long-term statistical average is for around 10 named storms per season, of which six become hurricanes.

Gray is a leading skeptic about human-induced global warming and believes that heightened Atlantic hurricane activity since 1995 reflects a natural variation in water temperatures and atmospheric conditions that may last for up to 20 more years.

Climatologists have found few indications that global warming could be linked to the increasing number of Atlantic storms.

But there is a growing body of evidence to suggest that warmer sea surface temperatures are already causing hurricanes in the Atlantic, and typhoons and cyclones in the Pacific, to become more powerful.

Gray and CSU team will update their forecast on May 31, Aug. 3, Sept. 1 and Oct. 3.