308 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1956 



Figure 4. — Cross section through a hypothetical hurricane showing a highly schematized 

 model of part of the vertical circulation. 



HURRICANE FORECASTING 



The forecast problem may be subdivided into six units: 



1. Forecasting inception. 



2. Forecasting the movement of the hurricane. 



3. Forecasting changes of intensity. 



4. Forecasting the rainfall resulting from the hurricane. 



5. Forecasting floods associated with hurricanes. 



6. Forecasting the height of the storm surge. 



1. INCEPTION OF HURRICANES 



During the past century, tropical meteorologists thought tropical 

 storms developed when there was excessive heating some place in 

 the Tropics. The heated air should rise and cause clouds with great 

 vertical development and showers. The explanation was given that 

 the more intense the heating, the heavier became the showers until 

 finally a storm formed. Close examination of the data, however, 

 revealed that hurricanes formed in areas where there was very little 

 change in the sea-surface temperatures for great distances. The 

 theory that hurricanes were formed by intense heat merely asso- 

 ciated the shower activity with the storm formation and offered no 

 definite mechanism for organizing the showers into a circular pattern 

 and furnished no means for predicting whether the tropical storm 

 would form. Since showers occur over the tropical oceans during 

 much of the year and since only a very small percentage of even 

 the areas of heavy showers ever develop into tropical cyclones, these 

 are serious limitations. Among the first mechanisms suggested for 

 concentrating and intensifying the disturbed areas into tropical 

 storms was the Norwegian model of an unstable frontal wave. In 

 this model a front is the boundary surface between two masses of 

 air that come from different source regions and that have different 



