On the subject of the reanalysis of the SNA "Emeraude" sonar tapes, it would be interesting to know what data recovery technique has been used. I suspect that they have mixed a very low level 37.5kHz tone into the noise which when combined with the pinger 37.5kHz tone will provide (37.5 +/- 37.5) = 75kHz or 0. The level of the tone is critical for the mixing and the summing to take place.
We should assume that there are very efficient digital methods available these days (as autocorrelation), that play in the frequency domain and generate a waterfall diagram of frequencies acquired.