**  Documentation  for CTD data collected on RRS Charles Darwin  Cruise  50                                                         
(Jun  -  Jul  1990)  by the Institute  of  Oceanographic  Sciences  (Deacon                                                         
Laboratory), Godalming, Surrey, UK, under the direction of P.M. Saunders **                                                         
                                                                                                                                    
The  instrument used was a Neil Brown Systems CTD which measured  pressure,                                                         
temperature and conductivity and was fitted with a Beckman dissolved oxygen                                                         
electrode.   The  CTD  was  used  alongside  a  General  Oceanics   Rosette                                                         
Multisampler with 12 water bottles,  a 10kHz pinger,  a bottom echo-sounder                                                         
and a SeaTech 1m path transmissometer.  Lowering and retrieval rates of 0.5                                                         
to  1.5m/s were employed and the sensors were flushed with distilled  water                                                         
on  recovery.  Bottle samples and reversing thermometer  measurements  were                                                         
made  on  ascent  and sea water samples were  analysed  using  a  Guildline                                                         
Autolab Salinometer.                                                                                                                
                                                                                                                                    
PRESSURE:  The pre-calibration data for Charles Darwin 50 for pressure  was                                                         
obtained at a temperature of 20 degC on 10 May 1990 and was as follows:                                                             
                                                                                                                                    
        P = 0.099662 * PRAW - 4.28E-7 * PRAW**2 - 4.3                                                                               
                                                                                                                                    
The  calibration  was obtained for increasing pressure  and  hence  applies                                                         
strictly  to  the  downcast.  The goodness of fit was  0.8  dbars  and  the                                                         
deadweight  employed  was certified to an accuracy of 0.03  per  cent  full                                                         
scale  (i.e.  1.8  dbar at 6000 dbar).  In the ocean,  because  of  varying                                                         
temperature, the following correction is applied:                                                                                   
                                                                                                                                    
        PCOR = P - 0.39 (t1 -9)                                                                                                     
                                                                                                                                    
where  t1  is  a lagged temperature,  in degC,  constructed  from  the  CTD                                                         
temperature  data using a first order equation with a time constant of  400                                                         
seconds. This time constant and the temperature sensitivity of the pressure                                                         
offset,  0.39  degC/dbar,  were determined by  laboratory  trials.  On  the                                                         
upcast,  a  further correction is made for the hysteresis of  the  pressure                                                         
sensor  (which  can  reach  5  dbar),  again  determined  from   laboratory                                                         
measurements.                                                                                                                       
                                                                                                                                    
TEMPERATURE:  The  pre-cruise  calibration   for  the  temperature  sensor,                                                         
conducted with only partial immersion of the instrument, was found on 9 May                                                         
1990  as  T  =  0.9987 * TRAW - 0.014.  The goodness of fit of  a  6  point                                                         
calibration  between  0.7 and +25 degC,  was 0.5 milliK.  Temperatures  are                                                         
given  in  the above calibration in degC on the ITS90 scale,  which  differ                                                         
significantly  from  the  IPTS68  scale  only   at  the  high  end  of  the                                                         
environmental   temperatures   (Saunders,  1990).  For  this  cruise   data                                                         
comparisons may be made with earlier measurements without reference to  the                                                         
change in scale.                                                                                                                    
                                                                                                                                    
The mismatch between the slower time constant of the temperature sensor and                                                         
the  fast  response of the conductivity sensor,  which  leads  to  salinity                                                         
spikes,  has  been  dealt with by correcting the temperature  according  to                                                         
Procedure 1 described in Chapter 5 of the SCOR Working Group Report (Crease                                                         
et al, 1988). A time constant of 0.25 seconds is assumed.                                                                           
                                                                                                                                    
Comparisons  were made with 7 digital reversing  thermometers  manufactured                                                         
and calibrated by SIS, Kiel, Germany,  and operated on the upcast by Niskin                                                         
bottle  closure  in  the  usual manner.  The mean  difference  of  the  217                                                         
comparisons  was  -2.3 milliK with a standard error of +/-0.8  milliK.  The                                                         
individual  differences  reveal that at pressures exceeding 500  dbars  the                                                         
differences  tend  to  average  closer to -5  milliK  (with  the  reversing                                                         
thermometers showing higher temperatures on average).                                                                               
                                                                                                                                    
CONDUCTIVITY/SALINITY: On Station 1 the conductivity sensor failed and  was                                                         
replaced on retrieval,  so no CTD salinities were obtained on this station.                                                         
For  stations  2 to 21 and 22 to 57 unique but different cell factors  were                                                         
used  in  calculating  salinity  for the  CTD,  i.e.  0.999535  and  0.9955                                                         
respectively  (for  p=0,  T=15).  The cell factor was assumed to vary  with                                                         
pressure  and temperature in the manner described in the SCOR WG 51  report                                                         
(Crease  et  al,  1988) with nominal values employed  for  the  temperature                                                         
expansion and pressure contraction coefficients of the cell material.                                                               
                                                                                                                                    
From the sample salinities measured on each station values of the  apparent                                                         
salinity  difference  SCTD - Ssample were constructed,  where SCTD  is  the                                                         
estimate  taken  on  the upcast when stopped.  A time series  plot  of  the                                                         
apparent  salinity  difference revealed slow drifts and small jumps in  the                                                         
CTD conductivity sensor response.  Adjustments were made to CTD salinities,                                                         
on a station by station basis.  Despite, or perhaps because of, employing a                                                         
brand new cell,  the calibration drifted approximately 0.015 (in  practical                                                         
salinity  units)  during stations 2 to 21,  then jumped 0.03;  it  remained                                                         
stable  for  the  rest  of  the  cruise.  After  adjustment,  the  salinity                                                         
differences  were plotted as a function of pressure.  No significant  trend                                                         
was  found  and  the rms scatter around zero mean difference  for  the  490                                                         
comparisons was 0.0022.                                                                                                             
                                                                                                                                    
OXYGEN: CTD oxygen values were calculated using a standard algorithm  found                                                         
in Owens and Millard (1985). Parameters required from the CTD were selected                                                         
on the downcast at the pressure corresponding to the upcast sample.  A non-                                                         
linear  regression  between these CTD parameters and the sample oxygen  was                                                         
performed: for this purpose the data set was divided into two approximately                                                         
equal  parts.  No significant differences were found for each half  of  the                                                         
data for the coefficient of temperature dependence (alpha), the coefficient                                                         
of  pressure dependence (beta) and fraction (the mix of ambient and  oxygen                                                         
cell  temperatures).  Values  determined  were alpha  =  -0.04148,  beta  =                                                         
0.000165 and fraction = 0.512. Both oxygen current lag and bias were set to                                                         
zero,  since non-zero values did not improve the fit.  The cell factor  was                                                         
determined  as  1.467.  All of these values were used for the  initial  CTD                                                         
oxygen determinations.                                                                                                              
                                                                                                                                    
A  time  series  plot  of the apparent oxygen  difference,  OCTD  -  Osamp,                                                         
revealed  drifts and jumps.  Adjustments were made to the CTD oxygens on  a                                                         
station  by station basis.  The range of adjustments is from -0.34 to  0.17                                                         
ml/l  (i.e.  about  0.5  ml/l).  After adjustment,  the  individual  oxygen                                                         
differences were plotted versus pressure;  the rms scatter around zero mean                                                         
difference for the 490 comparisons was 0.08 ml/l (3.5 micromol/kg).                                                                 
                                                                                                                                    
TRANSMITTANCE:   Potential  transmittance,  which  takes  account  of   the                                                         
increasing  mass of clear water in the 1 metre path of the instrument  with                                                         
increasing pressure, was calculated.                                                                                                
                                                                                                                                    
DATA  PROCESSING:  Original values were averaged over an  interval  of  one                                                         
second  and  calibration  coefficients   and  correction  factors  applied.                                                         
Differences between successive values of each parameter were examined;  the                                                         
mean  difference and its standard deviation calculated and  values  greater                                                         
than  several  standard deviations from the mean difference  were  checked.                                                         
Only a limited amount of editing of the data was required. Data were sorted                                                         
on  pressure,  averaged at 2 dbars and missing  values  were  interpolated.                                                         
Derived  quantities were computed from algorithms published by Fofonoff and                                                         
Millard (1983).                                                                                                                     
                                                                                                                                    
                                                                                                                                    
References:                                                                                                                         
                                                                                                                                    
Crease, J. et. al.  1988.  The acquisition, calibration and analysis of CTD                                                         
data. UNESCO Technical Papers in Marine Science. No. 54, 96pp.                                                                      
                                                                                                                                    
Fofonoff,  N.P.  and Millard,  R.C.  1983.  Algorithms for  computation  of                                                         
fundamental  properties  of  seawater.  UNESCO Technical Papers  in  Marine                                                         
Science. No. 44, 53pp.                                                                                                              
                                                                                                                                    
Owens,  W.B.  and  Millard,  R.C.  1985.  A new algorithm  for  CTD  oxygen                                                         
calibration. Journal of Physical Oceanography, 15, 621-631.                                                                         
                                                                                                                                    
Saunders, P.M. 1990. The International Temperature Scale 1990, ITS90.  WOCE                                                         
Newsletter No. 10, p10 (Unpublished Manuscript).                                                                                    
                                                                                                                                    
Saunders, P.M., Gould, W.G., Hydes, D.J. and Brandon, M.A.  1991.  CTDO and                                                         
nutrient data from Charles Darwin Cruise 50 in the Iceland Faeroes  region.                                                         
Institute  of  Oceanographic Sciences Deacon Laboratory,  Report  No.  282,                                                         
74pp.                                                                                                                               
