486 



VIGOUIIEUX AND HERSEY 



[chap. 12 



From these simple cases we pass to the more general one when the tempera- 

 ture, which almost always affects velocity to a greater degree than salinity or 

 depth, is fairly uniform over the first 50 to 1000 fathoms, depending on locality 

 and season, then decreases with depth perhaps by some 10 to 20°C in the next 

 200 fathoms, after which it decreases only very slightly with depth right to 

 the bottom. The region of rapid change is called the "thermocline". There 

 velocity and temperature profiles have approximately the same shape (Fig. 4) 



VELOCITY 



RANGE 



SURFACE 



Fig. 4. Sound rays from directional source above a thermocline. 



and rays emitted by a source near the surface are bent downwards and eventu- 

 ally reach the bottom where reflection occurs as in the first example, although 

 those rays emitted nearly horizontally may keep close to the surface if the 

 velocity gradient above the thermocline is due to pressure only. 



Finally, if beyond a certain depth temperature is constant or decreases very 

 little, as often happens in deep water, there is a depth after which the velocity, 

 which at first decreased with depth because of decreasing temperature, starts 

 increasing, since the effect of pressure, or depth (formula (8)), eventually pre- 

 dominates. In this case the rays may be bent upwards and become horizontal 

 again before they reach the bottom. Indeed, rays emitted in directions near to 

 horizontal from a source at a depth where the velocity is a minimum are 

 continually returned to that same depth, since they are convex upwards above 

 it and concave upwards below. As these rays reach neither the surface nor the 

 bottom there is no loss by reflection and almost no loss by scattering, which 

 occurs mostly at those two boundaries. The geometrical spread in this case is 

 cylindrical and the loss due almost solely to absorption. As this loss is low at 

 low frequencies, a disturbance rich in low frequency — for instance, the explosion 

 of a mine — can be detected thousands of miles away by a detector at the same 

 depth, usually a few hundred fathoms. 



In 1952, U.S. Navy Sofar stations at Point Sur and Point Arena, California, 

 gave remarkable records of the eruption of the Myojin submarine volcano, 

 8600 km away (Dietz and Sheehy, 1954), about 200 nautical miles south of 

 Tokyo. In March, 1960, shots (200 and 300 lb of TNT) fired at the axis of the 



