SOFAR PROPAGATION OF WIDE-BAND SIGNALS 

 TO LONG RANGES 



R. P. Porter 



Woods Hole Oceanographic Institution 

 Woods Hole, Massachusetts 



A fundamental shortcoming is evident in low-frequency 

 acoustics investigations being carried out today. Since 

 operational surveillance systems must cope with extremely 

 narrow-band sources, nearly all transmission assessment 

 programs are based on continuous-tone analytical models. 

 These models do not agree in detail with shot data — 

 nonetheless, wide-band sources are usually preferred for 

 measurement of long-range transmission in a SOFAR channel 

 because it is then possible to sort out the arrival 

 structure of the signal. Thus, the propagation along 

 individual paths can be assessed. Understanding of the 

 relationship between continuous-tone loss models and 

 wide-band data is necessary if we are to have confidence 

 in our transmission assessment programs. 



In this paper we review the basic phenomena encountered 

 in propagating wide-band signals to long range. We show 

 that there is a well defined arrival structure in fre- 

 quency as well as time. Some theoretical ideas are pre- 

 sented that allow efficient estimation of transmission 

 loss. We argue that wide-band loss estimates need not 

 taJte longer to carry out than narrow-band estimates. Some 

 loss calculations are presented for the Mediterranean Sea 

 that show reasonable agreement with measurements. Finally, 

 we suggest future experiments that will further develop- 

 ment of wide-band models and should reveal presently 

 unobserved phenomena. 



INTRODUCTION 



Detonated charges or other wide-band signals are the usual means 

 employed to sort out long-range sound transmission for the simple reason 

 that they provide the only practical way to evaluate sound transmission 



633 



