Viewing page 56 of 79

This transcription has been completed. Contact us with corrections.

the Carnot cycle, which assumes constant sea surface temperature. It is difficult to test this idea because of the paucity of oceanographic observations in and around tropical storms. 
     [[bold]]Hurricane Motion[[/bold]]
Forecasting the movement of hurricanes is obviously a matter of great consequence. For a time, it was thought that hurricanes simply move with the air flow in which they are embedded. More  recent research suggests that tropical cyclones are capable of altering the flow in which they reside; that is, they are capable of moving themselves.
 
The early theories of self-propagation simplified the vortex as one which did not change character with height. That is, the hurricane was assumed to have the same winds at all altitudes of the atmosphere. Even these models showed that hurricanes should steer themselves by creating eddies on certain sides of the storms, known as "beta gyres". These arise owing to the curvature of the earth's surface. But these simple models do not always correctly predict hurricane motion. 
We know from models and from inferences about upper atmospheric winds made using satellite photographs of clouds that hurricanes actually spin the opposite direction very high in the atmosphere. Although these winds are weaker, they cover a much larger area than do the lower, stronger winds. This oppositely spinning flow also creates eddies, but these tend to move the hurricane in a direction different from that predicted by the simple models. Also, vertical wind shear in the atmosphere may "push" the counter-rotating gyre away from its normal position atop the lower vortex, causing the two vortices to move each other. Unfortunately, very little is understood about this process because there are almost no good observations of the upper regions of hurricanes, well above the operating altitudes of most research aircraft. Worse, it is almost impossible to make a good forecast of hurricane motion using sophisticated computer models if there are no data to start the models with. 
    [[bold]]Fundamental Questions[[/bold]]
1. Why are hurricanes so rare, given that their energy source is present most of the time? Why do most tropical disturbance die without becoming cyclones?
2. Why don't most hurricanes that do manage to develop ever reach the intensity that nature permits, even though simulated storms always do?
3. What conditions of the atmosphere are necessary to transform an innocuous tropical disturbance into a violent storm? What do we need to measure in practice to see if these conditions are present so that hurricane formation can be predicted?
4. How do hurricanes influence their own motion? What role does the counter-rotating circulation in the upper atmosphere play in this?
5. What do we need to observe to predict hurricane motion and intensity as accurately as possible? Clearly, the present system of observations is woefully inadequate.

[[bold]]The Potential of Pilotless Aircraft in Understanding and Predicting Hurricanes[[/bold]]

Consider the current state of meteorological observations. Over North America, Europe and Asia the daily weather forecast hinges crucially on observations made twice daily by weather balloons launched simultaneously from stations based on solid ground. Although primitive, these observations provide detailed information about temperature, moisture and winds through most of the depth of the atmosphere. Besides these, there are numerous surface observations made from land-based stations, buoys and ships, and some valuable data near the tropopause collected by commercial aircraft. These data are also essential to basic research in atmosphere science.

Virtually none of these data are of any use to the hurricane forecaster or researcher, simply because hurricanes spend most of their lives well removed from these sources. This data void is not usually

              [[bold]]I-10[[/bold]]