Читать книгу Weather For Dummies - John D. Cox - Страница 13

Before the advent of electronic computers in the 1950s, weather forecasting was a much more difficult and, frankly, less accurate service. In those days, weather forecasting developed a reputation for inaccuracy that still haunts the business. Some early television forecasters even used puppets and other light-hearted distractions to ham up their presentations. The idea was that predictions were not expected to be taken too seriously.

Many forecasters in those days used a method that relied on maps of other days’ weather. They would draw a map locating areas of high and low pressure, warm and cold air masses, and moving storm fronts. Then they would page through big collections of earlier maps in search of patterns that most closely resembled the weather they saw before them. This was standard procedure: The map with the best fit would serve as an analog of what to expect from the current day’s pattern. A certain amount of personal experience and reputation and even “hunches” went into it. After decades of forecasting weather, this was the state of the art. Sometimes this analog method produced a fairly accurate prediction of the day ahead, and sometimes it didn’t.

Looking back, it is easy to see now that early forecasters were attempting to do something that just wasn’t humanly possible. The problem they were trying to solve was too big for their science. For one thing, their data was too sparse, their picture of the atmosphere was too sketchy. Even if they had more data, they didn’t have a way to manipulate the information fast enough to stay ahead of the passing weather.