Windswept_ The Story of Wind and Weather - Marq de Villiers [73]
A few African-born Caribbean systems even track back across the Atlantic after catching the midlatitude westerlies. Some reach Britain—London's 1703 storm, during which Queen Anne was escorted to a wine cellar under the Palace of St. James, and some windmills were destroyed through friction-caused fire because they were rotating so fast, is an example of what was probably an African, then a Caribbean storm in its track, before it turned back across the ocean. It had very likely completed its extratropical transition; it was no longer a true hurricane, but it would have coincided with another low to cause a weather bomb—an explosive pressure change defined as a drop of 24 millibars in twenty-four hours with a central pressure below 1,000 millibars, a state that can cause massively high winds very like hurricanes (bombs happen every year in New England and Atlantic Canada, usually more than once). Norwegian explorer Fridtjof Nansen experienced the remnants of a tropical storm in 1888 on the Greenland ice cap. The 1900 Galveston storm was still severe as it passed over Europe and disappeared into Siberia, where no records were kept of its passing. Some of these storms travel more than six thousand miles before they expire, causing havoc in an area several hundred miles wide.
Each workstation in each hurricane center is linked to the datastream coming in from quite literally hundreds of sources—weather buoys, remote sensing stations, aircraft and air traffic control weather data, reports from ships at sea, ham radio operators from Canada through Bermuda all the way across to the Cape Verde Islands, satellite and QuikScat data, Doppler and SAR radar outputs, and many others, including conclusions reached by hurricane and typhoon centers as diverse as those in Honolulu, Tokyo, Dartmouth, London, and Paris. There is even an acoustic model of a hurricane, containing data recorded from the apartment balcony of Dennis Jones, a scientist with the Canadian Department of Defence, and it's an eerie thing to listen to—you can hear the gusts breaking trees, for example, and you can "see" the gusts through their accelerated acoustic signature.30 And each workstation has access through the broadest of bands to the various prediction models that have been put together so painstakingly for so many years.
These models, whose keepers are the Tropical Prediction Center and the National Center for Environmental Prediction, range from simple statistical tables to sophisticated three-dimensional equation simulations. They are of two kinds, track models and intensity models, designed to answer the two key questions: Where is the storm going? How strong will it be? These are the two most difficult—yet critical—things a forecaster must decide. Where is landfall to be? How rapidly will it intensify—or weaken? You know, because you have learned it to be true, that the maximum wind speeds are approximately proportional to the square root of the difference between the central pressure and the surrounding pressure, and the models will show you that. But now you need to know the speed of change, and no formula will yield that up. How to account for the explosive deepening that some storms undergo? Surface sea temperatures? The surrounding ridges and troughs? Something analogous to the intersection of wave trains that can cause rogue waves at sea? A jet stream crossing a storm's path at a critical angle? Before satellites and the models crunched on supercomputers, forecasters had relied entirely on data from ships, and to a lesser extent from planes, and matched those against known historic data; but their forward projections past forty-eight hours were frequently wrong by three hundred miles or more.31
Peter Bowyer, of the Hurricane Centre in Dartmouth, explains some of the difficulties: "A tropical storm is a disembodied unit in a much larger flow of air. Think of it as a cork in a river. The river meanders to and fro, and the cork will go pretty much wherever it is taken, wherever the river goes. In our latitudes"—he's talking