power measurements was within 1.5 db and that of frequency measure- 

 ments was within 2.5Mc/s. The instrument precision is considerably 

 better than ±0.5 db and far better than the operator's judgment. 



Observation Technique 



The signal strength measurements were made from the maximum range 

 of detection to the nearest practical approach (usually the upper range of 

 the signal generator). After an iceberg target was selected, the ship 

 steamed away from the berg on a course whose reciprocal would take it 

 back to the target on a bearing determined by the berg aspect desired 

 during the actual test-run. During this period the overall performance of 

 the radar system was measured and a synthetic signal generated on the 

 "A" scan (oscilloscope or "M" scope). When the target could no longer 

 be discerned, the ship proceeded on the same course for a few minutes 

 and then turned to approach the target at reduced speed (usually 6 knots). 

 The first instant the target was painted on the PPI scope the range was 

 recorded as the maximum range of detection and the radar antenna 

 stopped and held on the target by manual training as necessary. For 

 several reasons the target echo was constantly varying in amplitude 

 and at times the scintillations were over 10 decibels in magnitude. The 

 recorded values of reflected power are not instantaneous readings but 

 rather the maximum signal strength observed at any time during the half 

 minute interval ending with the time and range for which the reading 

 was recorded. This method was inaugurated in 1945 and it is believed 

 that a reasonable record of echo strength versus range was thus made. 

 Continuous surveillance of the radar performance was made by test set 

 measurements and ringing time checks. Either before or immediately 

 after the test runs a photograph, sketch, physical dimensions and descrip- 

 tion of the berg surface were obtained. This data was used to obtain the 

 cross-sectional area of the target presented to the radar during the test 

 runs and to interpret the importance of morphology on signal strength. 

 Although there was little calving noted, there was considerable change in 

 aspect during some of the runs. Some icebergs, reaching a point of in- 

 stability, rocked to and fro through an appreciable angle while others 

 rotated. The indeterminate error of aspect change was eliminated when 

 grounded bergs were studied. 



The photographs of the PPI scope were taken with a Polaroid Land 

 camera. Transparency film was used and the exposure time was 11 sec- 

 onds. It is believed this longer exposure time on the PPI scope results in 

 a picture which better depicts what the radar operator actually sees than 

 the synchronized short exposures usually used. 



Preparation of Results 



As an investigation of this sort involves many different approaches and 

 as many different sciences some mention of the data handling methods is 



54 



