Why Are Measurements of Winds from Space Needed?

It is not hard to convince even the casual observer of weather that the measurement of winds is important for the prediction of future weather. Simply put, the weather systems that can be so disruptive of human activities are transported by the wind. Furthermore, within these systems the wind itself is a very important characteristic. Climate researchers use fields of wind data on computers to "transport" substances like water vapor, chemical constituents, and aerosols in performing research into the behavior of the climate system.

Presently, most of the wind "data" for both climate studies and weather prediction come from two sources. The first, and direct, source is from weather balloons. Twice each day at the same time, weather stations around the world launch balloons that measure several characteristics of the atmosphere, such as temperature, humidity, pressure, and winds (both speed and direction). Winds are measured by tracking the motion of the balloon as it rises through the atmosphere and is moved by the wind. This gives a fairly good measure of the wind as a function of height in the atmosphere at that locality, but the problem is that there are vast areas of the world (the oceans) where none of these measurements are taken. The second source for wind "data" is the computer models themselves, which will be discussed further below.

Satellites play an important role in providing atmospheric data over the oceans. Temperature and humidity data have been obtained by satellites for about 30 years or so, and the quality of that data is improving with each new generation of instruments. Global computer models accept that satellite data to put together detailed snapshots of the state of the atmosphere and to make weather forecasts. However, the models have to make up their own winds over vast regions of the globe.

Now, "making up" the winds by the computer models is not as bad as it sounds. The models use the wind data obtained over land, as well as the satellite temperature and humidity data over the globe, as information on which to base their "best guess" of the winds over the oceans. Those winds are mathematically consistent (i.e. consistent with physics as we know it) with what is actually observed in the atmosphere. However, the winds cannot be totally determined in that manner for two reasons: (1) we do not have perfect knowledge of all of the other atmospheric variables and (2) the mathematics of the flow are so complex that winds in fast-changing conditions cannot be inferred from the other variables. The atmosphere is very complex. As one examines it with a finer and finer microscope (observing system), one realizes that there are processes going on that were impossible to see with the previous observation.

The end result of all these considerations is that we would get much better quality snapshots of the atmosphere (and, therefore, better weather forecasts) if we could get more information. The computer models can be used to study which information would help the most. We can do things like create a digitized global atmosphere with one computer model, take simulated "observations" from that atmosphere, and use another computer model to test which observations would be most valuable. In almost all of these simulations, it has been found that obtaining global wind data would help most of all. This is not surprising to those who study the atmosphere - after all, in regions where there is no wind data, it seems obvious that any (accurate) wind data would make a big difference. (On the other hand, putting up a new temperature sensor, for example, may make only a small difference to what we already have.)

A word needs to be said for two methods of measuring winds from satellites that we already utilize. First, a measure of the winds can be made by watching the motions of clouds and other features in successive images from geostationary satellites. In regions where there are no wind measurements, this can be very valuable. (To see more on this topic, see Winds Derived From GOES Images). However, there are limitations to this approach which preclude it from making the same impact on weather forecasting that measurements of wind profiles (i.e. winds at many heights) would make. For one, the winds are obtained only at the height of the clouds or whatever other feature you are using. (Furthermore, the assignment of height of that wind may not be as accurate as needed.) Secondly, the spatial detail of the wind field may not be as good as you would like.

Another way to measure winds from satellite that has already been done is over the ocean surface. There have been satellites which have observed the ocean wave height and orientation and thus been able to estimate the wind field at the sea surface. However, it is the three-dimensional structure of the wind field that is needed to describe the atmospheric state. Wind measurements are needed throughout the depth of at least the weather layer (the troposphere, from the ground up to around 5 - 10 miles high, depending on latitude and other factors).

paul.meyer@msfc.nasa.gov

Responsible Official: Dr. James E. Arnold (jim.arnold@msfc.nasa.gov)
Page Curator: Diane Samuelson (diane.samuelson@msfc.nasa.gov)