All the raw and processed data are in the files called adcp_Y1.zip and adcp_Y2.zip. These files contain:

- cfg: Structure containing all the configuration of the ADCP.
- adcp: Structure containing all the data recorded by the ADCP. 
- map: Structure containing

 cfg and  adcp were obtained by reading the RD files into matlab using the routines originally written by R. Pawlowicz (rich@eos.ubc.ca):

[adcp,cfg]=rdradcp(`binary_file',1,-1);


 map is the gridded data, where each bin was mapped to its real depth, taking into account the vertical motion of the mooring. This requires a knowledge of the actual depth of the ADCP. The next section addresses this problem.  Time is still the original time of each ensemble. Some quality control is also done when mapping on the regular depth grid: 

abs(adcp.error_vel)<=0.5 
abs(adcp.vert_vel)<=2 
abs(adcp.east_vel)<=4
abs(adcp.north_vel)<=4

 
Pressure time series for the ADCPs

In order to map the velocities of the ADCP into a regular depth-time grid, we need to know the depth of the sphere. Unfortunately, none of the ADCPs have pressure recorder (although they have temperature). Only at KESS 7, we have the PMEL TP which was right next to the NB ADCP.

The other moorings are a little bit more tricky. Fortunately, most of the remaining moorings we had broadband ADCPs (or Narrowband), which have a signal that reached the surface (in general). By fitting the echo of the surface, we can find relatively precisely the range to the surface . Averaging the ranges from the 4 beams, we get a reasonably good estimate of the depth of the ADCP.

The results of this fit are added to the variable adcp (e.g.  adcp.sfc1 are the fit parameters for beam 1). The depth of the instrument (calculated from a conditional average of the first parameter of the fit) is saved as adcp.depth: Time series of the depth (in m) of the ADCP transducers. 

At times when the currents are really strong, the surface can get out of the range of the ADCP. In those instances, we used the MMP pressure data to patch the time series. 

Knowing the depth of the instrument, we can map the ADCP data onto a regular depth grid.