Clark and Yarnall 



Such experiments are always subject to a dual interpretation. They 

 may be viewed as a study of the medium, wherein the wave is used as 

 a research tool or probe. In this case, a propagation model is given, 

 and the properties of the medium are inferred from the results of the 

 experiment. In the second case, a model for the medium is given, and 

 the experiment becomes a study of wave propagation under prescribed 

 conditions. 



This paper is a preliminary attempt to interpret the Straits 

 of Florida cw data from the case I viewpoint: the acoustic wave is 

 regarded as a probe for environmental study. The nature of the ex- 

 periment forces this attempt to be qualitative and speculative in 

 many aspects. The propagation path is both acoustically and hydro- 

 dynamically complex. At 420 Hz the wave length to water depth rela- 

 tionships do not clearly recommend either the shallow or the deep 

 water acoustical theoretical approach. In addition, the path tran- 

 sects the Florida Current in a region of high current velocity which 

 may give rise to unusual conditions of instability in the water col- 

 umn. The dual nature of the experiment will be evident in what 

 follows, and the need for a flexible approach to the analysis will be 

 very obvious . 



THE ACOUSTICAL MEASUREMENTS 



For the purpose of data analysis, the techniques of the 

 environmental measurements will require little further comment. The 

 acoustical measurements, however, need elaboration. Figure 3 repre- 

 sents the design of the cw experiments, reduced to essential func- 

 tional elements. The functions enclosed in dashed lines are perform- 

 ed by the phase coherent demodulator,! a key item in the cw exper- 

 iments. It should be understood that a direct cable connection from 

 the Miami laboratory to the H43 hydrophone near Bimini would make the 

 use of matched precision oscillator "B" unnecessary. This secondary 

 phase reference is a matter of logistics only. In the interpretation, 

 the signals of H3 and H43 are assumed to be referred to the same 

 phase standard. The system has been designed to have a relative 

 phase drift of less than 360° per year at 420 Hz. A serious effort 

 has not yet been made to obtain absolute _in situ calibrations for H3 

 and H43 ; consequently, all amplitude measurements are relative. A 

 calibration for phase sense has been established, as indicated in 

 Figure 4a. The strip chart phase display extends only over 360°. 

 For measurements which exceed this range, the total phase excursions 

 are determined by manually accumulating multiples of 360° on a se- 

 parate graph called the expanded display. This is shown in Figure 

 4b. A zero value of phase is assumed in each experiment at the time 

 the equipment is put into operation. To clarify a potentially con- 

 fusing feature of subsequent illustrations, it should be pointed out 

 that the signal employed for cw testing is more accurately described 

 as "interrupted cw" . A basic cycle of 45 seconds of signal on, fol- 

 lowed by 15 seconds of signal off, was most frequently used. The 

 resulting display takes the form illustrated in Figure 4c. This 



310 



