Читать книгу Weather For Dummies - John D. Cox - Страница 17

The weather forecasting process begins with the need to know what the atmosphere is doing now. This information is the data that describes what computer modelers call the “initial conditions” that are the starting points for their powerful forecast software. You might think of it as a nowcast. This information has to be as absolutely accurate and as detailed as possible. Forecast modelers know that the slightest error in their description of the current state of the weather can fairly quickly lead to large errors in the forecast. If the data describing initial conditions is a little wrong, the forecast can be very wrong. (That’s what “The Butterfly Effect” later in this chapter is all about.)

Weather forecasting is such a complicated and difficult business because it is trying to predict the behavior of a system that has many features that are all changeable, all the time. Observations of every sort need to be gathered. How warm or how cold is it? Which way is the wind blowing? Is it cloudy, or is it clear? Details about all these features and more are gathered from every source available.

From the ground, these measurements come from human weather observers and from instruments on automatic weather stations. From the air, they come from weather balloons and airplanes and satellites orbiting the planet in space. From the sea, they come from ships and from instruments on anchored moorings and drifting buoys, as well as satellites. (See Chapter 16 for more about these measurement tools.)

Every day, 24 hours a day, data from thousands and thousands of weather observations around the world is streaming electronically into National Weather Service computers and other major weather centers around the world. With these millions and millions of bits of data, the computers are constantly updating and refining their highly detailed descriptions of the current state of the weather — the nowcast.