MARTIN/PERRONE : GEOGRAPHICAL VARIATION OF AMBIENT NOISE IN THE OCEAN 

 FOR THE FREQUENCY RANGE FROM 1 HERTZ TO 5 KILOHERTZ 



I think the answer is that anything that is published in the 

 literature that the investigator was able to get by a bunch of editors 

 is assumed to be valid. (Laughter) 



Dr. Ross: I hope the proceedings will indicate laughter. 

 (Laughter) 



Ms. E. A. Christian (Naval Surface Weapons Center) : Appreciative 

 laughter. 



Mr. Martin: 1 meant it to. 



Dr. S. W. Marshall (Naval Research Laboratory): I'd like to 

 respond to your query. I think any time you adopt rules of thumb for 

 diddling with data, you are treading on thin ice, and I submit that 

 the proper way to represent a long time series of data if you're 

 interested in a number is to present the entire cumulative 

 probability distribution function and let someone select whatever 

 percentile he wants. 



Mr. B. M. Buck (Polar Research Laboratory, Inc.): I'm up to the 

 north again. (Laughter) I don't want to belabor the obvious, but 

 it's on the same question that was just posed. I always felt pretty 

 comfortable in the Arctic making 1/3-octave noise measurements be- 

 cause there are no ships. But it seems to me that to measure 1/3- 

 octaves as a function of depth is more an exercise in propagation 

 than it is in noise. 



It seems to me that rather than trying to view with aircraft or 

 whatever means the presence of nearby ships, it seems that better 

 technique would be to make your measurements with narrowband, do a 

 narrowband spectrum analysis and rule out the obvious lines and 

 report the general noise. 



836 



