39/^ 



SURFACE AND OTHER EFFECTS 



velocity. This plume (Plate XI) is similar to the one at 26 foot depth, 

 but smaller because of the energy lost in the first contraction. 



The relation between the time of appearance and initial upward 

 velocity of the vertical plume is shown in Fig. 10.1 In this figure, 

 estimated time at which the plume leaves the position of the surface 

 before the explosion, and its estimated relative velocity at this time, 

 are plotted against depth. Also plotted are the times, from bubble 

 pulse period measurements, at which the bubble reaches or would reach 

 minimum size at each depth. The intersection of these curves with the 

 plume-time curve is seen to correspond closely with the maximum up- 



2.5 



2.0 



Fig. 10. 



20 30 40 50 60 70 80 90 



CHARGE DEPTH ( FT) 



Initial velocity and time of formation of vertical plumes by 300 

 pound TNT charges fired at various depths. 



ward velocity, and confirm the explanation, originally proposed by 

 Butterworth, of plumes as a result of venting of the explosion products. 



At increasingly great depths, the same cycle might be expected to 

 repeat itself but on a much smaller scale because of greater energy 

 losses. After more than two cycles, however, the bubble has an insig- 

 nificant fraction of its original energy and loses its identity into a mass 

 of turbulent water, which appears at the surface a considerable time 

 after the explosion. 



At very shallow depths, the bubble vents while exi)anding rapidly 

 and a vertical plume appears almost immediately, which is ver}^ narrow 

 and rises to great heights. If the charge is just below the surface, the 

 exact depth makes little difference, and for 1,000 pounds of TNT for 

 example the maximum height is of the order of 1,150 feet. The height 

 and nature of the plumes at other depths depend in a straightforward 

 but rather complicated way on the depth and charge weight, which can 



