                CTD CALIBRATION: CRUISE BON ENTENTE 1/86                   
                  (Sampling and Calibration Procedures)                    
                                                                           
A  Neil  Brown Instruments Mark III  conductivity-temperature  depth  (CTD)
probe  was used to obtain continuous profiles of temperature,  conductivity
and  pressure at each station.  A General Oceanics 12-bottle rosette  water
sampler  was  mounted  above the CTD.  The CTD was fitted with  a  6000  db
pressure sensor, the fast response thermistor was disabled and the platinum
resistance thermometer was used without the time constant being applied  to
correct  for  the  time lag.  The time constant is so small (235  ms)  that
errors induced by not applying a recursive filter are well within the error
range of the CTD.                                                          
                                                                           
The CTD probe was calibrated for pressure,  temperature and conductivity by
John  Smithers of the Institute of Oceanographic Sciences,  Wormley.  Water
samples  were  taken from the rosette Niskin bottles at selected depths  to
provide  in-situ calibration data.  Protected reversing  thermometers  were
mounted  on one rosette bottle.  The Niskin bottles were tripped during the
up-cast,  and the CTD data to be compared with the samples were recorded at
the  actual  depth at which each bottle was fired.  Salinity  samples  were
drawn from selected bottles, and were analysed on a Guildline 8400A Autosal
salinometer.  The Guildline Autosal determines water sample salinity with a
precision  of  +/-  0.002  and  an accuracy of  +/-  0.003.  The  reversing
thermometers  have an accuracy of +/- 0.02 deg C and were  corrected  using
calibrations supplied by IOS.                                              
                                                                           
A  few  of the in-situ calibration differences were much  larger  than  the
others,  as much as several standard deviations.  These usually occurred in
regions  of strong vertical gradients and were eliminated from the  in-situ
calibration  data set.  The remaining temperature differences  between  the
calibrated  temperature  readings and the corrected  reversing  thermometer
were within or close to the stated accuracy of the reversing  thermometers,
so  no additional temperature correction was applied.  The CTD pressure  at
the  surface  was noted and was always within the stated  accuracy  of  the
pressure  sensor (+/- 6.0 dbar),  so the only pressure calibrations applied
were those supplied by IOS.  The calibrations for pressure and  temperature
of the CTD supplied by IOS were:                                           
                                                                           
   T(cor) = (T(ctd) * 1.00763) + 2.34379e-2                                
   P(cor) = (P(ctd) * 1.00053325) - 5.6                                    
                                                                           
The  sample bottle conductivities were calculated using the  sample  bottle
salinity  values with the CTD temperature and pressure values.  A value  of
42.914  mmho  cm-1 for the conductivity of standard seawater at  15  deg  C
(Culkin  and  Smith  1980) was used to convert the measured  sample  bottle
conductivity  ratios  to conductivity.  Sample bottle  conductivities  were
compared to the CTD conductivities. The average and standard deviation were
outside  the  stated  accuracy of the CTD and so a linear  calibration  was
determined from salinity bottle conductivities (see Fig. 1).               
