148 



SHALLOW-WATER TRANSMISSION 



2000 

 RANGE IN YARDS 



4000 



Figure 5. Transmission over SAND for different hydrophone depths. 



We have already noted, from consideration of the 

 Rio values, that for SAND-AND-MUD bottoms the 

 agreement between WHOI and UCDWR is very 

 poor. This discrepancy is confirmed by the median 

 and quartile transmission curves of those runs over 

 SAND-AND-MUD which were carried out with a 

 shallow hydrophone in the presence of NAN pattern 

 (strong downward refraction), shown in Figure 4. In 

 this case, the discrepancy is undoubtedly real and 

 not caused by different calibration methods; for not 

 only are the transmission anomahes at a given range 

 different, but the WHOI median curve has a much 

 steeper slope. The slope of the median UCDWR is 

 roughly between 8 and 10 db per 1,000 yd, while the 

 slope of the WHOI median curve is about 18 db per 

 1,000 yd. 



No other comparisons were made between WHOI 

 and UCDWR transmission data because most of the 

 WHOI samples were too small for such comparisons. 

 All the median curves to be discussed later are based 

 exclusively on UCDWR runs. 



To examine the effect of hydrophone depth over 

 a well-reflecting bottom, median curves over SAND 

 were determined for the three classes of hydrophone 

 depth without regard to refraction pattern. In Fig- 

 ure 5, the three resulting curves are superimposed on 



each other, identified as s (shallow), i (intermediate), 

 and d (deep). Table 1 shows that the bulk of these 

 runs were carried out in the presence of downward 

 refraction, with about one-fourth of the BT patterns 

 showing a shallow mixed layer above the thermocline. 

 In Figure 5 there are no significant differences 

 between the three curves. 



The quartile curves have not been reproduced, but 

 they are all fairly similar, deviating from the median 

 curve by about 5 db at 3,000 yd. The transmission 

 anomaly over SAND can be represented fairly well 

 by a straight line passing through zero at zero range 

 and having a slope of 5 + 2 db per kyd. This 

 numerical estimate is also good for the median and 

 quartile curves shown in Figure 3, which do not in- 

 clude the SHALLOW MIKE cases forming part of 

 the sample used in constructing Figure 5. 



Figure 6 shows median curves, for shallow and for 

 deep hydrophones, of all runs obtained over ROCK 

 bottoms. It will be noted that the transmission over 

 ROCK is not quite so good as over SAND, the aver- 

 age slope for ROCK being 6 db per kyd. The quartiles 

 deviate from the median, in this case also, by roughly 

 2 db per kyd. 



Figure 7 shows the median and quartile curves for 

 all runs obtained by UCDWR over STONY bottoms. 



