Hurricane forecaster tries new technique
Hurricane forecaster tries new technique for ’08 seasonDENVER – Using a simplified forecasting technique, researcher William Gray is predicting an above-average hurricane season in the Atlantic next year, with seven hurricanes, three of them major.Gray’s team at Colorado State University had called for above-average storm seasons the past two Decembers – and both turned out to be wrong. But Gray said he believes this year’s forecast will be better.”We think we’re finally onto a scheme that will be more accurate,” he said Friday.The new forecast calls for a total of 13 named storms in the Atlantic. It says there is a slightly higher-than-average chance that at least one major hurricane will hit the United States.Gray’s predictions, now in their 25th year, are watched closely by emergency responders and others in hurricane-prone areas. But officials routinely express concerns that residents might become apathetic if predictions prove overblown. Government forecasters also predicted an above-average season for 2007.Over the long term, an average of 5.9 hurricanes form in the Atlantic each year.Gray said his team based the 2008 extended-range forecast, issued six months before the June-November season, on just three predictors. In the past, they have used as many as six.”Sometimes when you go to four, five or six, you tend to tune it too much to the past,” he said.”We’ve been issuing this extended-range forecast since 1992 and we haven’t shown much skill in them,” Gray said. “We obviously couldn’t keep using the same method if it wasn’t working in real time.”The new method relies on water temperatures in the Atlantic and Pacific and the relationship between barometric pressure and altitude over the Atlantic. Gray said the method fared well when his forecasters applied it to conditions in previous years.Gray plans updated 2008 forecasts in April, June, August, September and October.Although Gray’s April forecast was wrong in 2006 and 2007, his team has correctly predicted whether hurricane activity would be above or below average in seven of the past nine years, said Phil Klotzbach, a member of Gray’s team and the lead author of the 2008 forecast.Gerry Bell, lead seasonal hurricane forecaster for the National Oceanic and Atmospheric Administration in Camp Springs, Md., said the number of predictors isn’t a deciding factor in the accuracy of a hurricane forecast.”What matters is that they use the correct predictors,” he said. “The number matters less than the correct ones.”Bell said the primary predictor is that an active hurricane era began in 1995.Bell declined to discuss Gray’s forecasting approach, adding that differences between Gray’s method and the NOAA’s are mainly technical. But he did say that predicting next year’s season in December would be difficult.NOAA issues its first forecast in May.Gray said the active era is not likely to end for another two decades, even though the past two seasons were below average.”We’ve been very lucky the past couple years,” he said.—On the Net:Gray’s forecast: http://hurricane.atmos.colostate.edu/Forecasts/2007/dec2007/dec2007. pdf
Start a dialogue, stay on topic and be civil.
If you don't follow the rules, your comment may be deleted.