Embedded Files
Meteorologists are working long and hard to try to improve the ability to forecast the detailed distribution and intensity of precipitation. They're even trying to figure out ways to incorporate radar data into numerical models so that they can be used in computer forecasts. However, it's still extremely difficult to actually forecast the evolution of precipitation systems. The most useful method is still good old-fashioned extrapolation.
Extrapolation means taking what has happened in the recent past and assuming it will continue to happen in the future. The simplest application of this technique would be to use radar to determine the motion of a precipitation system and to use that motion to infer when the precipitation will reach a given area. For this method to work, you have to be able to assume that the precipitation will not suddenly dissipate. If it's an individual thunderstorm, you'd better expect that the thunderstorm will dissipate within a couple of hours. If it's a squall line, the precipitation will probably be longer-lived.
When extrapolating for more than an hour or two, you have to pay attention to the "envelope". By this I mean that in most precipitation systems, individual precipitation elements are constantly forming, moving, and dissipating while the system itself lives on. A good simple example of this is lake-effect snow. Lake-effect snow takes the form of bands of precipitation extending from a large lake to the downwind shore and partly over land. Each band is composed of individual convective elements. The convective elements form over the lake, move to shore, and dissipate. But new convective elements are constantly taking their place, and if you want to know how the lake-effect snow is evolving, you'd better follow the motion of the overall region of snow, which is generally nearly stationary.
Another example is precipitation north of a low center. Typically, broad bands of precipitation move northward and dissipate, only to be replaced by the next band from the south. So if you're watching a band moving north, and estimate that it will reach your location in six hours, you'd better think again because that particular band is likely to dissipate before it reaches you. Of course, if the storm system itself is moving north, perhaps you'll get nailed by the following band of precipitation, or maybe the one after that.
Perhaps you have heard of the "College Station effect", that severe thunderstorms nail other locations but always seem to dissipate before they hit College Station. Well, everywhere I've been has had an "effect" like that, and it's a consequence of the short life of thunderstorms: if you see a thunderstorm out there, it's probably going to die soon no matter where you are.
Report abuse