Processed data#
Once data is processed (for example, calibrated using compute_Sv),
the outputs are in the form of a generic xarray Dataset.
For such processed data, we currently do not follow any specific conventions,
but we retain provenance information in the dataset and also include
variables that were used in the processing.
As we move upward in the processing chain, we will continue to retain provenance information in the processed datasets. However, whether and how data variables used in the processing will be stored remain to be determined.
Range of echo samples#
The calibration operation includes the computation of a new data variable, echo_range, which is the physically meaningful range (in meters) of the echo samples, from the range_sample dimension in the raw data. Recall that range_sample contains 0-based indices of the digitized sample numbers of the received echoes. This computation requires knowledge of the sound speed, so that the sampling interval (the variable sample_interval in a Sonar/Beam_groupX group) can be converted to distance along range in meters, i.e. like below:
where \(\Delta r\) is the along-range “length” of each sample, \(\tau\) is the sample_interval, and \(c\) is sound speed. The factor 2 is due to the round-trip travel from the transmitter to the scatterer and back to the receiver.
Because sample_interval can be different for each echosounder frequency channel, the resulting echo_range arrays can be different across frequency channels even if they have the same number of digitized samples.