                  CTD CALIBRATION: DISCOVERY CRUISE 164                    
                                                                           
The  instrument used was a Neil Brown Instrument Systems (NBIS) CTD,  which
measured pressure, temperature and conductivity. The CTD was used alongside
a General Oceanics Rosette Multisampler with 12 Niskin bottles and a 10 kHz
pinger.  Lowering  and retrieval rates of 0.5 to 1.0 m/s were employed  and
the sensors were flushed with distilled water on recovery.  Bottle  samples
and  reversing  thermometer measurements were made on descent and  the  sea
water  samples  were  analysed   using  a  Guildline  Autolab  Salinometer.
Reversing thermometers were calibrated before and after the cruise.        
                                                                           
All  casts  were  full depth,  most proceeding to within 10 -  30m  of  the
bottom.                                                                    
                                                                           
Data  were  initially logged at 16 hz from the NBIS CTD onto an  NBIS  deck
unit, which had been enhanced by IOS with an ASCII interface,  allowing raw
data  to  be written to a 1600 bpi 0.5 inch magnetic tape.  Data  are  then
passed  to a microcomputer which edited and averaged the raw data to 1  hz,
using algorithms developed by Pollard et al. (1986).  All SMP messages from
all  instruments  were  archived  using a Plessey 68000 and  a  UNIX  based
Plessey  68000  was available for further processing.  CTD data  were  then
analysed  using  editing,  utility and display  programs  (Pollard,  1983).
Further details of the processing can be found in Pollard et al. (1987).   
                                                                           
a)  Pressure:  Raw  pressure R(raw) was calibrated to dbar  P(cal)  by  the
formula                                                                    
                                                                           
   P(cal) = 0.1 * P(raw) - 12.0                                            
                                                                           
The  pressure  calibration was checked by comparing the CTD  depth  at  the
bottom  of  each  cast with the value derived from the IOS  Precision  Echo
Sounder.  In  total,  35 values fell within a Gaussian  distribution  which
tailed off at 15m, and these values yielded                                
                                                                           
   D(pes) - D(ctd) = 0.4 +/- 6.3                                           
                                                                           
From  this  it was deduced that pressures were therefore  probably  correct
within a few metres.                                                       
                                                                           
b)  Temperature:  CTD temperatures were calibrated at sea  using  the  most
recent laboratory calibration                                              
                                                                           
   T(C) = T(raw) * 0.0005 * 0.9990317 + 0.0258                             
                                                                           
which  reinforced  the  long-term   stability   of  the  platinum  resistor
calibration.  Reversing  thermometers  were deployed at two depths  on  all
casts, to check for gross errors and, on the evidence gained, no change was
made to the CTD temperature calibration.                                   
                                                                           
c)  Salinity:  An overall correction of 0.050 given by all 165  values  for
casts  11404-58,  with a standard deviation of 0.005,  was  applied.  Casts
11401-03   were   individually  corrected  by  0.027,   0.036   and   0.045
respectively.  After  cast  11458,  a new conductivity  sensor  had  to  be
installed,  and calibration points available for casts 11459-61 were  found
to  be  unreliable.  Therefore,  a correction of -0.019 was applied to  all
three casts.                                                               
                                                                           
d)  Oxygen: Offsets between up- and down-casts were minimised by  replacing
lagged  temperature  by  unlagged CTD temperature.  It was found  that  the
calibration  drifted enough that casts had to be fitted individually.  Most
CTD  values  were  believed to be within 0.1 ml/l of their true  value,  or
certainly 0.2 ml/l.                                                        
                                                                           
References:                                                                
Pollard, R.T. 1983. PSTAR shipboard data processing programs.  Institute of
Oceanographic Sciences, Marine Physics Group. (Unpublished manuscript)     
                                                                           
Pollard, R.T., Holford, D., Ellis, S., Read, J.F.  and Smithers,  J.  1986.
CTD data from the Northeast Atlantic Ocean 37deg - 47deg N, 10deg - 16deg W
collected  on  RRS Discovery Cruise 145 in late winter 1984.  Institute  of
Oceanographic sciences, Report No. 223.                                    
                                                                           
Pollard, R.T.,  Read,  J.F.,  Smithers,  J.  1987.  CTD sections across the
Southwest Indian Ocean and Antarctic Circumpolar Current in Southern Summer
1986/7. Institute of Oceanographic sciences, Report No. 243.               
