"Le secret d'ennuyer," re marked Voltaire, "est de tout dire" - "The trick of being boring is to say everything." Let me risk that epithet this morning by rambling on a little more on satellites - specifically about precisely what they tell us.
It is sometimes tempting to assume that if the forecaster has a bird's-eye view of the weather pattern, all his or her problems have been solved, but this is not so.
A satellite picture provides the meteorologist with a great deal of additional information about what the weather pattern looks like now; when it comes to the future, however, the satellite maintains a discreet silence - and the future, after all, is what weather forecasting is about.
So what information about the present does the satellite provide? Its most obvious pro duct is a simple photograph, a snapshot of the world taken from many hundreds of miles above us. This is of great assistance to the forecaster in pin-pointing the exact position of depressions and fronts, particularly in areas where the conventional ground-based weather observations are sparse - as they are, for instance, over the oceans.
There are more sophisticated uses to which satellite pictures can be put. In the case of rainfall, for example, the brightness and the texture of cloud patterns can be used to estimate rainfall amounts over a large area with reasonable accuracy. Temperatures, on the other hand, can be obtained from infra-red satellite pictures - images produced by detecting infra-red radiation rather than the visible light of conventional pictures.
On such images, the brightness at any particular spot is proportional to the temperature of the object in view, so if the height of any cloud in the picture is known, for example, the air temperature at that level can be estimated.
Winds, too, at the upper levels of the atmosphere can be assessed. These are estimated by comparing two satellite pictures taken, say, 30 minutes apart; the distance travelled in that time by a number of distinctive features of the cloud pattern is observed and if it is assumed that the cloud moves at the speed of the wind - of ten a reasonable assumption - then the wind speed at that level can be calculated.
All these data are extracted by automated methods from the images provided by the satellites and are particularly useful over those large areas of the world where conventional ground-based observations are few and far between.
They are used as the raw material for the mathematical models of the atmosphere from which the daily weather forecasts by computer are produced.