The very first profile ensemble collected by the instrument when data collection starts has the transmitter disabled. This gives us a noise floor record for the system (i.e. an amplitude value for each bin with no acoustic energy in the system). When the data is displayed, two data sets are displayed: the Amplitude and the SNR. The Amplitude signal is the measured amplitude from the instrument (i.e. Amplitude_signal). The SNR is the amplitude signal minus the noise floor signal. Vectrino Profiler: In the Data structure in the MATLAB export, you will see Profiles_AmpBeam* (Amplitude_signal) and Profiles_SNRBeam* fields. You will also see VelocityHeader_AmpBeam* fields. The velocity header contains the noise floor (Amplitude_noise) and the SNR is simply the Profiles_AmpBeam* - VelocityHeader_AmpBeam*.

Vectrino: The noise level is stored along with the “noise correlation” in the header file (*.hdr). 

Vector: The noise level is stored along with the “noise correlation” in the Velocity Header data (*.vhd) file.

Note that the SNR is derived from the echo strength in each receiver arm. As such, it is associated with the measurements along each beam and not with the XYZ coordinates.

Noise Correlation: Typical values are below 10%, which is expected since noise is uncorrelated. As the SNR decreases the velocity correlation will approach this value. What this means is that when you discard the data with a correlation below 70%, you are screening the data with respect to the SNR as well. This is why you end up with the same SNR numbers after discarding samples with correlation below 70%.