Find the weather in any city, state, ZIP code or country
|
|
|
|
Weather or not? Most predictions fail to materialise
Fort Collins, Colorado _ Last week, daily forecasts told weather worry-warts for five straight evenings that heavy snow was likely on the next afternoon, and five afternoons passed, sunny and rather pleasant winter days. And on the sixth day, it snowed heavily. And why does the forecast of crippling weather in Smalltown, Middle USA, matter in Big Mango, Thailand, where snow is never expected and predictions for tomorrow's weather pretty well always amount to this: ''Look out the window. See that weather out there? Tomorrow, expect more of the same.''? By ALAN DAWSON Saturday January 19, 2008 found at bangkokpost.com It matters because this pleasant, cowboy-populated town on the Front Range of the Rocky Mountains _ more than 2,500 kilometres from the Atlantic Ocean _ is home of the Atlantic Ocean hurricane forecasting station for Earth. The forecasts, however, have been just about as reliable as the snow warnings of last week, and it seems polite now to wonder whether science really can predict what's happening with the weather. If you keep predicting snowfall during the Rocky Mountain weather, one day you are going to be right. It doesn't take a cynic to wonder if the attempts to predict the number and ferocity of hurricanes are any more credible and reliable. It is a small step from there to scepticism over forecasts of climate change 10 and 50 years down the road. Located on the campus of Colorado State University, the Tropical Meteorology Project each December issues a forecast of hurricanes (storms originating in the northern Atlantic Ocean) for the coming year. Last month, as he did last year, project chief Professor William Gray explained why he got his forecasts so horribly wrong. But then, unlike the boy who cried wolf, Mr Gray and his office got a lot of publicity and money to carry on for a third year of crying ''Wolf!'' yet again. ''We believe that the Atlantic basin is still in an active hurricane cycle,'' said Mr Gray, as he forecast that there will be seven hurricanes this year, beginning around June _ three of them major storms likely to bring serious damage to the East and Gulf coasts of the United States. The first obvious question seems to be why anyone should believe a twice-failed prognosticator. Last year, he predicted nine hurricanes; six occurred. The previous year, he ventured there would be nine hurricanes, and there were five. Far worse _ which means, of course ''far better'' _ no disastrous hurricane struck the US mainland, even though this is the main point of interest about the forecasts publicised in the United States. Loyal assistant Phil Klotzbach, who compiles and writes the entertaining prognosis in early December and the explanations of failure in April, claimed at this year's small media gathering that all this is a credible exercise. In seven of the past nine years, he said, the team has correctly forecast whether the intensity of a hurricane season would be above or below average. Of course, this is about what nine tosses of a coin would get. A mathematician would do a better job analysing the project's detailed forecasts and projections, including what Mr Gray and his staff call ''US landfall strike probability,'' which makes specific predictions. In 2008, they say, there is a 60% chance that a major (Category 3-to-5) storm will strike the US East or Gulf coast. The average chance of that happening over the past 100 years, they state, was 52%. But last year, Mr Gray and his team crawled out on the same branch, forecasting there was a 64% chance of a major hurricane strike on the US. For 2006, they rated the chance at 81%. Both times, nothing much happened. The 81% forecast in 2005 was the prognostication that put the Colorado laboratory on the media map. It came right after the superstorm Katrina hit and devastated New Orleans. The sensationalist press, always welcoming new disaster, leapt on the scientific predictions of the Gray team. And when the prophecy failed, the team was bewildered. ''This is the fewest named storms to form in the Atlantic since 1997,'' the explanation began. ''This is only the 11th year since 1945 that no hurricanes have made United States landfall.'' There is recurring discussion of vertical wind shears and a bottom line: ''We attribute a large portion of this forecast over-prediction to a late-developing El Nino and increased mid-level dryness in the tropical Atlantic.'' And so on. The Colorado-based hurricane centre far from the centres of any hurricanes is not the only place where forecasts are made. Last week, Chirapol Sintunawa of the Association for the Development of Environmental Quality predicted that the Pacific Ocean's La Nina effect will cause so much rain this year that tourists will cancel their Bangkok trips. The Asian Development Bank came up with $900,000 so that Thailand and the region can spend a year looking for ways to slow climate change by cutting greenhouse gas emissions. Mark Twain held that ''Everyone is talking about the weather but nobody's doing anything about it.'' That may be more true today than when Twain was skewing the pundits. |
Weather predictions or not? Most fail to materialise BBC weather UK New Barkfly and other insect species arrives in UK due to hot weather. more BBC weather Weather TAGS
: Weather books :
Weather
Underground >> |
|
What Next ?
|
