SHORT-RANGE PROPAGATION IN DEEP WATER 



193 



in the orientation of the hydrophone sometimes have 

 a surprisingly large effect on the form of the recorded 

 pressure-time curve. Tiny quantities of gas occluded 

 on the face of the hydrophone or included in water- 

 proofing or insulating materials can slow up the 

 response to a steep-fronted pulse, and make the be- 

 havior of the hydrophone nonlinear. Natural reso- 

 nances in the hydrophone can be .shock-excited by a 

 steep-fronted pulse, causing spurious wiggles in the 

 pressure-time curve, and in some cases making the 

 pulse appear to last many times longer than it 

 actually does. At short ranges, where relatively in- 

 sensitive hydrophones may be used, emf's due to the 

 impact of the pressure wave on the connecting cable 

 may give spurious signals. With long cables, im- 

 pedance matching and dielectric losses may have to 

 be considered. These and many other points are dis- 

 cussed at length in other reports.^"* 



In the following sections we shall first consider 

 propagation of explosive pulses through the water 

 alone, and later, in Section 9.4, shall take up pulses 

 reflected from or transmitted through the bottom. 



9.2 SHORT-RANGE PROPAGATION IN 

 DEEP WATER 



9.2.1 Attenuation and Change in 

 Form of the Pulse 



As the earlier chapters of this volume have shown, 

 the most important single factor affecting the shape 

 and strength of a sound pulse of given frequency 

 traveling through the ocean is the variation of the 

 velocity of sound from point to point, due chiefly 

 to temperature gradients but produced also to some 

 extent by pressure and salinity gradients. Since to a 

 first approximation the velocity of sound is a func- 

 tion simply of the depth and to this approximation 

 can be calculated from bathythermograph records, 

 it will be convenient to separate, as far as possible, 

 those features of explosive sound propagation which 

 are due to this variation of velocity with depth from 

 those features which are due to other properties of 

 sea water and which would be encountered even 

 when the bathythermograph record indicates no ap- 

 preciable refraction. In this section we shall consider 

 the latter features, recognizing, however, that unde- 

 tected small-scale fluctuations in the velocity of 

 sound may possibly be an important factor in ac- 

 counting for them. 



One of the most interesting features to be found in 



the measurements of explosive pulses at ranges from 

 30 to 2,000 yd is that with increasing range there is 

 an increase in the time required for the pressure in 

 the initial pulse to rise to its peak value. At shorter 

 ranges, it will be remembered, this initial pulse is a 

 shock wave and its time of rise is less than the re- 

 solving time of any measuring apparatus which has 

 been used (see Section 8.3). Unfortunately, the meas- 

 urements which have been made of the time of rise 

 are not sufficiently detailed to establish the cause of 

 this variation with range. Table 1 summarizes the 

 experimental information to date; this information 

 was taken from two NDRC reports.''* In this table 

 "time of rise" is defined as the interval between the 

 first measurable increase of pressure and the maxi- 

 mum of the pressure-time curve. "Resolving time" is 

 defined as the value of time of rise which the system 

 would record for an instantaneous rise in pressure in 

 the water. For the first set of observations this time 

 was measured directly from records of shots at close 

 range; for the other two sets it was merely estimated 

 from acoustical and electrical characteristics of the 

 hydrophone and circuit. 



The data given in Table 1 have been chosen to 

 exclude any cases where the hydrophone was in or 

 near the shadow zone predicted from bathythermo- 

 graph data. They therefore presumably represent an 

 effect which occurs in the absence of large-scale re- 

 fraction, although it is not impossible that through 

 inaccuracy of the computed ray diagrams some of the 

 shots at the longer ranges may have been close 

 enough to the shadow zone to increase the time of 

 rise by virtue of the shadow-zone effect discussed 

 in Section 9.2.3 and shown in Figure 9. The 

 deep-water data of reference 8, which are given in 

 the table, are plotted in Figure 9 of that section, 

 for comparison with similar time-of-rise data taken 

 in the shadow zone. It is worth noting that in these 

 experiments no marked dependence of time of rise 

 on the depth of the explosion was found for those 

 cases where the hydrophone was not in or near the 

 shadow zone; a slight increase in time of rise with 

 increasing depth was observed, but this was not 

 significantly greater than the experimental error. 

 Slightly more than half of the observations of refer- 

 ence 8 fell within ±2 ;isec of the means given in 

 Table 1. 



For the deep-water shots off San Diego, the varia- 

 tion of apparent time of rise with range is approxi- 

 mately what one would expect if the resolving time of 

 the apparatus were actually about 10 /usee and if the 



