observed that this measurement required the most objective treatment 

 and close adherence to a standard technique. The data were taken on 

 four bearings and as has been demonstrated before the strongest echo 

 arrived from windward. A good correlation does not exist between wave 

 height estimates and the reflected power curves from the two different 

 ships which made these measurements; however, enough measurements 

 have been made to quantitatively show that sea return for wave heights 

 above 4 feet is sufficient to entirely obscure dangerous growlers. This is 

 is best illustrated by figure 39 in which the reflected power curves for a 

 growler and a 4.5-foot sea are compared. Because this figure illustrates the 

 important results of very carefully controlled measurements under ideal 

 conditions a detailed account of the observations is given: During the 

 day of 28 May 1959 the CGC Androscoggin commenced test-runs on a 

 large growler (22 by 76 ft.; 665 sq. feet). From a temperature and hu- 

 midity sounding (Wiresonde) from the surface to 290 feet, the atmospheric 

 conditions were assessed as being near standard throughout the first 200 

 feet. The sea at this time was approximately 2 feet high, 500 feet long. 

 Because a small growler (4 by 20 feet) in the vicinity was being partly ob- 

 scured by sea return, the ship remained near the 22-foot growler through- 

 out the day for anticlutter measurements. That evening the wind in- 

 creased to 20 kts and the seas to 4.5 feet high, 200 feet long. Considerable 

 difficulty was experienced in detecting the 22-foot growler and the 4-foot 

 growler was completely obscured. Twenty-eight measurements of the re- 

 flected power from the sea were made at short range intervals on four 

 bearings. The observations made in the windward direction were 10 

 decibels greater than those made in the backsides direction and 5 decibels 

 greater than those made parallel to the waves. The curve for the windward 

 measurements is presented in figure 39. The 20-knot winds and inter- 

 mittent rain preclude subnormal propagation conditions. The optimum 

 detection range for this target was 6,200 yards and although the max- 

 imum range observed earlier that day was 11,200 yards, the maximum 

 range during the 4.5-foot sea was 9,000 yards. This formation could not 

 be detected in the sea return using the available anticlutter devices. It 

 should be mentioned that although the aspect shown in figure 39 was the 

 aspect observed during the test-run earlier that day, a slight rotation 

 would present a considerably smaller physical cross section. This forma- 

 tion no doubt was rocking and rotating during the observations, but an 

 indeterminate error throughout these measurements is precluded by the 

 fact that this formation was observed continuously for 11 hours and 

 found to be of the same shape after 18 hours. These measurements 

 caused some concern as although they were supported by other less pre- 

 cise observations the fact that 4.5-foot waves reflected better than a 22- 

 foot growler whose aspect showed some target gain seemed to conflict 

 with reason. The theoretical analysis of this problem presented below has 

 reestablished the validity of these observations. 



85 



