


                         Cruise Report for Knorr 154-2 

I. Cruise Narrative

I.1 Highlights

   Ship:             Knorr 154-2
   WOCE designation: AR24
   Dates:            5 October - 19 November 1997
   Ports:            Azores - Iceland - Woods Hole
__________________________________________________________________
               
I.2. Cruise Summary

163 CTD stations (30 stations were NOT full water column due to time)
 43 PALACE floats
 41 RAFOS floats
    LADCP
    Rosette Salts and Oxygens
    Carbon sampling (Dissolved organic, Dissolved inorganic, Alkalinity)

__________________________________________________________________________

I.3. List of Principal Investigators

Ms. Ruth G. Curry, WHOI, CTD/hydrography
Dr. Michael S. McCartney, WHOI, CTD/hydrography
Dr. Eric Firing, Univ. of Hawaii, ADCP/LADCP
Dr. W. Brechner Owens, WHOI, PALACE floats
Dr. Amy Bower, WHOI, RAFOS floats
Dr. Phil Richardson, WHOI, RAFOS floats
Dr. Catherine Goyet, WHOI, Carbon Chemistry

__________________________________________________________________________
I.4  List of Cruise Participants

Master:  A.D. Colburn III
Chief Scientist:  Ruth Curry, WHOI
Science Crew:
 Marshall Swartz, WHOI, Watch Leader
 Shelley Ugsted, WHOI, Watch Leader
 Gwyneth Hufford, WHOI, Watch Leader
 Sarah Zimmermann, WHOI, CTD data processor
 Alex Nimmo-Smith, UK, Watchstander
 Tania Casal, Portugal, Watchstander
 Helder Martins, Portugal, Watchstander	
 Heather Hunt, WHOI, Watchstander 
 Mike McCartney, WHOI, Watchstander
 Marti Jeglinski, WHOI, Watchstander
 George Knapp, WHOI, Oxygens
 Pete Landry, WHOI, Salinity
 Jules Hummon, U Hawaii, LADCP
 Kathy Donoghue, U Hawaii, LADCP
 Greg Eischeid, WHOI, CO2 tech 
 Erin Sweeney , WHOI, CO2 tech
 Rick Healy, WHOI, CO2 tech 
 Cindy Moore, U Fla, CO2 tech 
 Greg Packard, WHOI, SSSG tech 
__________________________________________________________________________
I.5. Preliminary results 

 163 casts were made using a NBIS Mark III CTD measuring pressure, temperature,
conductivity, oxygen current, and oxygen temperature.  For each cast, water 
samples were collected at discrete intervals and analyzed for salinity and 
dissolved oxygen -- primarily for the purpose of calibrating the CTD sensors. 
Carbon chemistry measurements were also made for total organic carbon, total 
inorganic carbon, and alkalinity.  Lowered Acoustic Doppler Current Profile 
(LADCP) measurements were acquired on each cast. Underway measurements include 
surface temperature and salinity, plus shipboard ADCP which profiled velocity 
in the upper 400 meters along track.
 
    

CTD casts
________

sta 1-27  Azores to Charlie Gibbs Fracture Zone (CGFZ)
sta 28-31 Southern trough CGFZ
sta 31-33 E-W section including break in ridge separating N/S channels
sta 33-37 Northern trough CGFZ
sta 38-62 section along crest of Reykjanes Ridge to Iceland
sta 63-65 1500m casts where PALACE floats were deployed (Iceland-Greenland)
sta 66-109 Angmagssalik - Ireland section
sta 110-115 eastern boundary, Porcupine Bank
sta 116-120 eastern boundary, Porcupine Bank
sta 121-127 eastern boundary, Porcupine Bank
sta 128-162 Goban Spur to Terceira, Azores

Most casts were full water column, EXCEPT the following stations
where shorter casts were done in order to keep the 30 nm station spacing
in the allotted cruise time:
  SHORT casts: Sta 14,16,18,20,23,25,63,64,65,113,114,115,120,121,122,
               123,124,135,137,139,141,143,145,147,149,151,153,155,157,
               159

Problem stations:
  Station 21 has bad CTD oxygen. 
  Stations 21 to 40 have troubled CTD oxygen, some stations worse than
  others. 

Equipment/sensor changes:
  New oxygen sensors put on for stations 902, 3, and 21.
  CTD 8's oxygen assembly and sensor put on for station 41.

__________________________________________________________________________

II. Finalized Description of Measurements.


II.1  	CTD MEASUREMENTS
	by Marshall Swartz and Sarah Zimmermann

II.1.a	MAJOR DIFFICULTIES 
There were problems with the CTD oxygen data for the first 40
stations. Three sensors were tried, and although most of the data was
successfully fit to the bottle data, it was neccessary to fit almost
every station individually.  A new assembly, with a sensor only used
once at the start of the cruise for a test station, was used
beginning with station 41 and worked well for the remainder of the
cruise.  The results were more stable allowing for larger station groups
for fitting to bottle oxygen.  Stations that have not been successfully
fit or have offsets, up to .3ml/l,  over part of the station are
stations 20, 21, 25, and 30 through 40.  


II.1.b	EQUIPMENT CONFIGURATION
A WHOI-modified EG&G Mk-III CTDs was used throughout the cruise (CTD 
#9).  It was provided with an oxygen current and temperature
channel, a platinum temperature probe, and a 3 cm conductivity cell. CTD9
was modified at WHOI to install a thermally-isolated
titanium pressure transucer, with a separately digitized pressure
temperature channel (Toole, Bond, Millard, 1991)(Toole et. al., 1993).
Pressure, temperature and conductivity calibrations were performed at
WHOI prior to the cruise. Calibrations will also be performed upon 
return to WHOI.

A Sensormedics oxygen sensor was installed for the test station prior
to station 1.  Not having worked well for three stations it was
replaced before station 3. The sensor was replaced again before station
21. Over the next 20 stations the quality deteriated, and it was
decided to change out the whole assembly, replacing it before station
41 with the assembly and sensor that were used for the test station of
the backup CTD. 

A redundant 400ms platinum thermometer (OTM Ocean Temperature Module
#1326 produced by Falmouth Scientific Intruments) was connected to the
CTD for the first four stations.  It added noise to the CTD data so was
removed for the remainder of the cruise. 

Two FSI 24- position Sure Fire Water Samplers were provided for the 
cruise  The pylon was controlled through a
dedicated personal computer from which commands were issued and pylon
responses displayed and logged. 

An FSI Integrated CTD, ICTD#1338, recording data internally, was used
to collect backup data for selected stations (48, 103, 104, and 136).
ICTD 1338 data was downloaded directly to a computer, demodulated using
FSI software.  


Two rosette frames were provided for the cruise.  Both frames held 24
4 liter bottles.   Only one was used during the cruise.  The bottles
had been produced at WHOI. A Lowered Acoustic Doppler Current Profilers (LADCP),
from University of Hawaii was mounted on the frame.  Besides the CTD, the
LADCP and the occasional FSI backup CTD, a 12-khz pinger was secured on
the frame.    


II.1.c	AQUISITION AND PROCESSING METHODS
Data from CTD 9 was aquired at 23.8 hz and with a temperature lag of 
150 ms. The temperature lag was checked  by comparing density reversals 
in theta salininity (TS) plots (Giles and McDonald, 1986).  It was found 
that 150 ms showed the least amount of looping or density reversals.

The CTD data was acquired by a FSI DT 1050WS deckunit providing
demodulated data to two personal computers running EG&G version 5.2 CTD
acquisition software (EG&G, Oceansoft acquisition manual, 1990), one
providing graphical data to screen and plotter, and the other a running
listing output.  Bottom approach was controlled by following the pinger
direct and bottom return signals on the ship-provided PDR trace.

After each station, the CTD data was forwarded to another set of
personal computers running both EG&G CTD post-processing software and
custom-built software from WHOI (Millard and Yang, 1993).  The data was
first-differenced, lag corrected, pressure sorted and centered into 2
decibar bins for final data quality control and analysis, including
fitting to water sample salinity and oxygen results.  This data was
then forwarded to the PI for analysis, to compare to historical
and water sample data.


II.1.d	SUMMARY OF LABORATORY CTD CALIBRATIONS 
The pressure, temperature, and conductivity sensors were calibrated
by Marshall Swartz at the WHOI's Calibration Laboratory.

PRESSURE CALIBRATION
Method/Calibration Standards
The CTD pressure transducer was calibrated in a temperature
controlled bath against the WHOI Ruska dead Weight Tester (DWT) as
descibed by Millard and Yang (1993).  The pressure temperature S1 and
S2 terms were calculated from the results of the two temperature
pressure calibrations.   D0, the pressure temperature dynamic term, was
determined from a temperature dunk test in the laboratory and also from
pressure analyses at sea. 
The pre-cruise calibration, performed Sept 19 to 24, 1997, consisted of
pressure calibrations at two temperatures, 2.0C and 30.4C. 
  
                         BIAS         SLOPE        QUADRATIC
  CTD9
  pre-cruise  2.0C   -.128314E+02 0.999371E-01 0.150538E-09 
  STANDARD DEVIATION =   0.462988E+00

  pre-cruise  30.4C  -.808852E+01 0.999495E-01 0.563499E-10 
  STANDARD DEVIATION =   0.330181E+00


  PRESSURE TEMPERATURE
  CTD9        		 S1       S2       T0    	D1
  pre-cruise       2.3969e-007  0.1583     2.0      -290.15
  adjusted at sea	-	-	-	    -380.0
  

TEMPERATURE CALIBRATION
Method/Calibration Standards
The CTD's temperature, pressure temperature and oxygen temperature
sensors were calibrated in the large temperature controlled bath
against the F18/SPRT4070.  The CTD is totaly immersed in the salt water
bath for a full calibration of seven points from 30 to 1 deg C.
The pre-cruise temperature calibration was performed Sept 19 to 24,1997.

  PLATINUM THERMOMETER
               BIAS             SLOPE                 QUADRATIC
  CTD 9  
  pre-cruise    -.179154E+01 0.496266E-03 0.483534E-11 
  STANDARD DEVIATION =   0.276514E-03


  PRESSURE TEMPERATURE
  CTD9    	BIAS    	SLOPE
  pre-cruise  0.374454E+02 -.918590E-02 
  STANDARD DEVIATION =   0.704427E-01


  OXYGEN TEMPERATURE
               BIAS             SLOPE                 QUADRATIC
  CTD 9  
  pre-cruise  0.140770E+00 0.122493E+00 0.522138E-05 
  (Used for stations 1 to 40)
  STANDARD DEVIATION =   0.151450E+00


CONDUCTIVITY CALIBRATION
Method/Calibration Standards
Bottle salinities were drawn during the temperature calibration, four
bottles at each temperature.  The salinties were then converted to
conductivity and compared to the values read by the CTD at the 
each temperatures point (Millard and Yang, 1993). 
       
                BIAS             SLOPE           
 CTD9
 pre-cruise   -.152394E-01 0.996828E-03 
  STANDARD DEVIATION =   0.119487E-02




II.1.e	SUMMARY OF AT-SEA CTD CALIBRATIONS
PRESSURE CALIBRATION
The pressure bias of CTD9 at the sea surface was monitored at the
beginning of each station to make sure there was no signifigant drift
in the calibration. The pressure bias was found looking at
the calculated pressure prior to the CTD entering the water. The bias
varied from 0 to 1db over the 163 stations.  The bias did not vary
enough to warrant a change in the pressure calibration.


CONDUCTIVITY CALIBRATION
Basic fitting procedure
The CTD conductivity sensor data was fit to the water sample
conductivity as described in Millard and Yang 1993. All the stations
were fit as one large group, and divided into sections where 
there was a noticible shift in the sensor. These groups were fit for 
both slope and bias. 

Data Quality
The overall drift of the sensor during the cruise was .006 psu fresh.
Calibrated, the overall standard deviation of the CTD and water sample
differences was .0105. Reducing the number of observations from 3343 to
3337 by removing the outliers greater than +/- .1 psu the
standard deviation dropped to .0055 with a mean of -9.78e-05.  Looking
at the data deeper than 1000db, and removing the 6 outliers less than
+/- .1 but greater than +/- .02 leaves a standard deviation of .0019.

After being fit to the water sample salinities, the CTD data below
3000db still trends saltier than water samples for a maximum offset of
.001psu at 5000db.  If this needs to be corrected, changing beta from
1.5e-8 to .75e-8, and then refitting the groups' slope should work.


OXYGEN CALIBRATION
Basic Fitting procedure
The CTD oxygen sensor variables were fit to water sample oxygen data to
determine the six parameters of the oxygen algorithm (Millard and Yang, 
1993). As with conductivity, the stations were fit after being broken
into groups where there was a shift in the response of the sensor.

Data Quality
Before being fit, the difference between water sample and CTD oxygen
varied greatly from -2 to 2 ml/l.  Calibrated, the overall standard
deviation of the CTD and water sample differences was .157.  Removing
differences greater than .5 ml/l, reducing the number of observations
from 3351 to 3287 leaves a standard deviation of .089 and a mean of
-.0026.  Looking at the deep data, below 1000db, and removing the
outliers greater than +/- .2 ml/l leaves a standard deviation of .0428.

As mentioned earlier, there were problems with  the CTD oxygen data for
the first 40 stations. Three sensors were tried, and although most of
the data was successfully fit to the bottle data, it was neccessary to
fit almost every station individually.  A new assembly, with a sensor
only used once at the start of the cruise for a test station, was used
beginning with station 41 and worked well for the remainder of the
cruise.  The results were more stable, allowing larger station groups
to be used in the fits.  Stations that have not been successfully
fit or have partial offsets, up to .3ml/l, are stations 20, 21, 25, and
30 through 40.  Quite a few stations required spikes or sections to be
removed using interpolation. The stations requiring interpolation and
the pressure bounds interpolated over are listed further down. 



II.1.f	OTHER NOTEABLE DATA ACQUISTION/PROCESSING ISSUES
The CTD hit bottom on station 102. There was no resulting shift in the
conductivity calibration.  The backup CTD, ICTD 1338, was put on for
stations 103 and 104 for comparison.  The ealier station 48 needs to be
compared to determine if there was any shift between the CTDs before
and after station 102.  

At-sea logs were kept for both CTD data acquistion and processing. They
include anything of note regarding each station: equipment changes,
instrument behavior, equipment or operational problems, how data noise
was delt with, where it occured.  KN54ACQ.DOC contains the data
acquistion descriptions. KN54PROC.DOC contains the data processing
descriptions and pressure bounds of bad data removed through
interpolation.  



__________________________________________________________________________

II.2  	Salinity and Dissolved Oxygen Measurements for Knorr 154/leg 2 
	by George Knapp

II.2.a 	Summary
Water samples were collected from virtually every bottle during this cruise
for the determination of salinity and dissolved oxygen.  The primary purpose
of these measurements is to accurately calibrate the sensors on the CTD.

II.2.b 	Salinity 
Water was collected in 8 ounce glass bottles.  The bottles are
rinsed twice, and then filled to the neck.  After the samples reached the
lab temperature of 22 degrees , they were analyzed for salinity using a
Guildline Autosal Model 8400B (WHOI #11) salinometer.  The salinometer was
standardized once a day using IAPSO Standard Seawater Batch P-132 (dated
09-APR-97).  The Autosal worked flawlessly and showed virtually no drift for
the entire cruise.  Conductivity readings are logged automatically to a
computer, salinity is calculated and merged with the CTD data, and finally
used to update the CTD calibrations.  Accuracy of salinity measurements are
+/- 0.002.


II.2.c	Dissolved Oxygen 
Measurements are made using a modified Winkler technique
similar to that described by Strickland and Parsons (1972).  Each seawater
sample is collected in a 150 ml brown glass Tincture bottle.  When reagents
are added to this sample, iodine is liberated which is proportional to the
dissolved oxygen in the sample.  A carefully measured aliquot is collected
from the prepared oxygen sample and titrated for total iodine content.
Titration is automated, using a PC controller and a Metrohm Model 665
Dosimat buret.  The titration endpoint is determined amperometrically using
a dual plate platinum electrode, with a resolution better than 0.001 ml.
Accuracy is about 0.02 ml/l, with a standard deviation of replicate samples
of 0.005.  This technique is described more thoroughly by Knapp et al (1990).
Calculated oxygen is merged with the CTD data, and used to update the CTD
calibrations. 

Standardization of the thiosulphate titrant was performed daily.  The
titration apparatus worked flawlessly, and no unusual problems were noted.


__________________________________________________________________________

II.3   Shipboard ADCP   
      contributed by  Dr. Jules Hummon, University of Hawaii

Upper ocean current measurements were made throughout the cruise using the
hull-mounted acoustic Doppler current profiler (ADCP) system that is
permanently installed on the R/V Knorr.  The system includes five components:

1) an incoherent (narrow bandwidth, uncoded pulse) 4-beam Doppler sonar
operating at 153 kHz (model VM-150 made by RD Instruments), mounted with
beams pointing 30 degrees from the vertical and 45 degrees azimuth from the
the keel;

2) the ship's main gyro compass, continuously providing ship's heading
measurements to the ADCP via a 1:1 synchro;

3) a Global Positioning System (GPS) attitude sensor (Ashtech model 3DF),
which uses a 4-antenna array to provide interferometric measurements of
ship's pitch, roll, and heading;

4) a GPS navigation receiver (Trimble Tasman) providing position fixes using
both GPS frequency bands (L1 and L2) and the P and Y codes (military
"Precision Positioning Service", or PPS);

5) an IBM-compatible personal computer running the Data Acquisition Software
(DAS) version 2.48 from RD Instruments, augmented by Firing's software
interrupt handler ("user exit") program "ue4", C. Flagg's user exit "agcave",
and Flagg's TSR watchdog timer program.

The ADCP was configured for 16-m pulse length, 8-m processing bin, and a 4-m
blanking interval (all distances being projections on the vertical and based
on a nominal sound speed of 1470 m/s).  The transducer depth was 5 m; 60
velocity measurements were made at 8-m intervals starting 21 m below the
surface.  About 240 pings were sent in each 5-minute averaging interval.  For
each ping, velocities relative to the transducer were rotated to a
geographical coordinate system using the gyro compass heading, but assuming
pitch and roll to be zero.  The single-ping velocities were then
vector-averaged over the 5-minute ensemble.  The ensemble-averaging was done
separately for the vertical average from bins 2 through 10 and for the
deviation of each bin from this vertical subset; the two parts were then
added back together and stored.  The conversion from Doppler shift to
velocity was done using soundspeed calculated from the temperature measured
by a sensor in the transducer, assuming a constant salinity of 35 psu.  When
a velocity estimate in one of the four beams was missing, velocity was
calculated from the remaining three beams.

In regions of shallow water, the ADCP was configured to track the bottom with
one bottom-tracking ping for each water-tracking ping.  This was effective to
depths of 600 m or more.  From the time the ship left Woods Hole to the last
station of the present cruise, approximately 60 hours of underway bottom
tracking data were collected.  This is significant for the calibration
calculations discussed below.

The user exit program integrated the GPS position and attitude information
into the ADCP data stream.  Position fixes were recorded at the start and end
of each ADCP averaging interval (5-minute ensemble).  Attitude from the 3DF
was sampled at each ping and edited within each ensemble.  The mean, standard
deviation, minimum, and maximum values of pitch, roll, and compass heading
error were calculated and recorded.  The compass error is the quantity of
primary interest: for each ping, the compass reading used by the ADCP was
subtracted from the most recent 3DF heading (updated once per second), and
this difference was taken as the time-variable compass error plus some
constant misalignment of the 3DF antenna array.  The 3DF attitude information
was not used for the real-time vector-averaging of velocity because it is not
quite reliable enough; dropouts and outliers do occur.

Velocity, position, and attitude measurements were post-processed using the
University of Hawaii CODAS software package, generally as described by
Firing in WHP Office Report WHPO 91-1, WOCE report 68/91.  The essential
modification since then is the rotation of the velocity measurements
relative to the ship to correct for the gyro compass error as measured by
the 3DF.  After this correction, and a small but varying sound speed
correction, standard water and bottom tracking calibration methods (Joyce,
1989; Pollard and Read, 1989) yield two constants: a velocity scale factor,
and a horizontal angular offset between the transducer and the 3DF antenna
array.  The angular offset is particularly important; an error of 0.1 degree
leads to a cross-track bias of 1 cm/s for a ship speed of 11 kts.  Bottom
track data were primarily obtained leaving from and returning to Woods Hole,
off the coast of Iceland and Greenland, and off the coast of Ireland.  Water
track calibration calculations based on the entire cruise (all
stations--water track calibration requires ship accelerations, such as stops
for stations).  Bottom and water track final calibration values indicated an
angle offset of -.51 degrees. Closer inspection of all available calibration
information indicates that the "constant" factors are measurably not
constant.  The angle offset factor may vary within a range of up to plus or
minus 0.2 degrees.  A possible cause is under investigation; it is not clear
whether it will be possible to reduce this uncertainty in the present or
future data sets.

The quality of the shipboard ADCP data set from this cruise is reasonably
good.  No instrument problems were detected and there was an abundance of
acoustic targets on the entire cruise track.  There were no known compass
failures and no long dropouts of 3DF data.  During the first two-thirds of
the cruise, data were typically obtained to 400-450m.  During the last third
of the cruise, bad weather reduced the penetration to 350m in places.
The transit from the Azores back to Woods Hole had sufficiently bad weather
to cause data loss to as shallow as 250m in places.  

The upper ocean velocity field during the cruise is summarized in a map of
shipboard ADCP velocity vectors averaged from 100 to 300 m (adcp Figure);
vertical shear was weak on most of the cruise track, so this layer average
is representative.  Many characteristics of the velocity field in this
figure are similar to the previous two cruises.  The overall impression is
of weak currents--usually under 50 cm/s, and mostly in the form of
ubiquitous small-scale squirts and eddies.  The East Greenland Current
stands out as a narrow jet flowing southwestward along the Greenland
coast. The eddy field was relatively strong in the Rockall Trough and in the
Iceland and Irminger basins on the section from Greenland to Ireland.  One
deep barotropic eddy found in the Iceland basin was present in each of the
three cruises.  Currents were moderately weak and quite varied between the
Azores and Ireland.



__________________________________________________________________________

II.4   Lowered ADCP
      contributed by  Dr. Jules Hummon, University of Hawaii

To measure velocity throughout the water column at each station, a
self-contained ADCP was mounted on the rosette; this is referred to as the
lowered ADCP (LADCP).  The LADCP includes a magnetic compass and a tilt
sensor, so the velocity profiles can be rotated into the local east-north-up
coordinate system. Because the motion of the rosette over the ground is not
measured, the LADCP measurements of current relative to the instrument cannot
be used directly to infer the current over the ground.  Instead, the
single-ping velocity profiles are differentiated vertically to remove the
package motion (which changes only slightly between the time a ping is
transmitted and the time the backscattered return is received).  The vertical
shear estimates from all pings are then interpolated and averaged on a single
uniform depth grid covering the whole water column.  This full-depth shear
profile is integrated vertically to yield a velocity profile with an unknown
constant of integration; and the constant is calculated from the known
displacement of the instrument between beginning and end of the cast,
together with the shape of the relative velocity profile and the measured
current past the instrument as a function of time during the cast.  The
method is explained in detail by Fischer and Visbeck (1993).

The instrument used on this cruise was a new 150-kHz coded-pulse
("Broadband") profiler made by RD Instruments (a specially modified Phase-III
DR-BBADCP), with four beams angled 30 degrees from the vertical. 

All 161 profiles were made with the following instrument parameters:
blanking interval, pulse length, and processing bin length were all set to
16 m (projected on the vertical).  Sixteen depth bins were recorded.  Pings
were transmitted alternately at 1 and 1.5 or 1.6 second intervals.  Data
from each ping was recorded individually, with no averaging.  Ambiguity
resolution mode 1 (no automatic resolution) was used, with an ambiguity
intervals of 3 m/s, 3.5 m/s or 4 m/s--the smallest value was used when
weather was exceptionally calm.  Medium bandwidth was selected.  Three-beam
velocity solutions were not used, and solutions with an error velocity
exceeding 15 cm/s were rejected.  Bin-mapping based on tilt was selected.

Immediately after each station the data were dumped from the LADCP to a PC
via a serial line (RS-422), and transferred to a Sun workstation for
archiving and processing.  The profile was processed using the University of
Hawaii system, a mixture of C, Matlab, and Perl programs.  Velocity and
shear data are automatically edited based on several criteria including
correlation magnitude (typically 70-count minimum), error velocity (10 cm/s
maximum), deviation of vertical velocity in a given bin from its vertical
average (5 cm/s maximum), and deviation of individual shear estimates from a
mean shear profile (3.5 standard deviations).  Additional editing is done on
the upcast: the top two depth bins are rejected if the current, profiler
vertical velocity, and profiler orientation are such that one beam may be
intersecting the profiler's wake.  Depth bins subject to contamination from
the sidelobe return from the bottom, or from the return of the previous ping
from the bottom, are also automatically rejected.  Critical to this part of
the editing is accurate knowledge of the depth of the bottom and the depth
of the profiler.  Therefore we have an automated routine for matching the
time series of vertical velocity measured by the LADCP with the time series
of vertical velocity calculated from the CTD pressure record, and then
assigning the corresponding CTD-derived depths to the LADCP.  With these
instrument depths in the LADCP database, another program scans the LADCP
backscatter amplitude profiles in the near-bottom region; the LADCP depth
plus the vertical range to the amplitude maximum is the bottom depth.

Accurate position fixes at the start and end of the LADCP profile are
essential to the calculation of absolute velocities.  We log the PPS GPS
fixes at the full 1 Hz sampling rate.  The processing software accesses these
files and extracts the subsets needed for each profile.  Magnetic variation
is needed to calculate true direction from the compass readings; we calculate
the variation from a standard model of the earth's magnetic field.  

An unfortunate problem appeared to develop in the LADCP after about 50
casts: the LADCP velocity data became increasingly unrealistic both in
physical terms and in comparison with the shipboard ADCP data.  It was
determined that the problem lay in the magnetic heading recorded by the
instrument.  The compass used by this instrument was a TCM2 magnetic flux
gate compass.  This compass is strapped down to the chassis and measures
three orthogonal components of the magnetic field: under ideal conditions,
the measured magnetic field vector would be the earth's magnetic field.  The
LADCP uses the horizontal component of this vector to transform the
velocities from instrument coordinates to geographic coordinates.  It is
this horizontal component that we determined to be in error.  A total of 74
casts were taken with the first compass.  A spare TCM2 compass was mounted
during station 76.  This compass ultimately appeared to have the same
problem and was swapped out between casts 127 and 128. A total of 50 casts
were taken with the second compass. The new components (provided by Terry
Joyce) included a KVH compass, compass circuitry, tilt sensors, and
transducer driver boards.  The KVH compass differs from the TCM2 compass
because it is fluid-gimbaled and measures only the horizontal components of
the magnetic field.  The data from the last compass system (36 casts)
appeared to be reasonable.

The LADCP compass headings were corrected by finding the heading-dependent
error which minimized the magnitude of the vector difference between a given
LADCP profile and the overlapping region of shipboard data during that cast.
The heading-dependent correction was modeled as a sinusoidal function of
measured heading. The following table shows the improvement in the comparison
between LADCP and shipboard adcp values after the correction was applied:

============================================================================
Comparison of (LADCP-shipboard adcp) values for all stations with each of 
the three compasses.                                                   
                         MEAN ladcp-adcp values (cm/s)      
                               u       v    mag
---------------------------------------------------
(TCM2 compass #1)  (n=74)                          
orig                        -10.80  -3.17  12.85   
rotated                      -0.21  -0.67   2.76   
---------------------------------------------------
(TCM2 compass #2)  (n=50)                          
orig                         -8.11  -2.74  10.48   
rotated                      -0.42  -0.19   2.80   
---------------------------------------------------
(KVH compass)      (n=36)                          
orig                         -0.46  -1.65   3.59   
rotated                      -0.07  -0.57   3.15   
---------------------------------------------------

summary (orig)     (n=160)   -7.63  -2.70  10.03   
summary (rotated)  (n=160)   -0.24  -0.50   2.86

============================================================================

The correction in LADCP heading brought the magnitude of the vector
difference between LADCP and shipboard ADCP down to 2.9 cm/s from 10cm/s
(dominated by errors from the first two compasses).  For comparison, the
average magnitude of the vector difference between LADCP and shipboard ADCP
velocities from the cruise 6 months previous (which had an extremely good
LADCP-adcp comparison) was 2.6 cm/s.  The mean differences in u and v were
reduced from -7.6 cm/s and -2.7 cm/s (original u and v) to -0.2 cm/s and
-0.5 cm/s (for rotated u and v).  Again, for comparison, the mean
differences in u and v for data collected 6 months earlier were 0.04 cm/s
and 0.03 cm/s.  These corrections have clearly done a relatively good job of
correcting for the failing TCM2 compasses.

A map of corrected LADCP current vectors averaged over the full depth
range of the profile (ladcp Figure) shows some characteristics of the
currents as observed on this cruise.  As in the shipboard ADCP data,
the East Greenland Current stands out as a prominent feature amid the
welter of eddies.  The barotropic component of the eddy field is
weakest on the Ireland-Azores section and strongest on the
Greenland-Ireland section, where vertically averaged velocities of 10
cm/s or more are common.   The eddy in the Iceland Basin is captured by
the LADCP data.  In general  the eddy field is not well resolved by the
station spacing; the velocity profiles typically change radically from
one station to the next. 

_________________________________________________________________________________

II.5 Expendable Current Profilers


A total of 19 expendable current profilers (XCP; Sanford et al., 1982,
1993) were deployed during this cruise.  Each probe was launched while
the CTD was descending, typically passing the CTD at about 1000 m.
Data from 18 XCPs (at stations 17, 21, and 23-38) were good.
The XCP measures currents relative to an unknown depth-independent
constant, with 2-m vertical resolution and 0.5 cm/s rms uncertainty.
Profiles extend from the surface to 1600 m, but approximately the top
250 m is contaminated by the electromagnetic signal of the ship, and a
region of about 200 m extent near the CTD was contaminated by the
electric field of the CTD on some stations.

_________________________________________________________________________________
II.6   Carbonate Chemistry
       contribute by Greg Eischeid/WHOI, Chipboard Leader of CO2 team
       Principal Investigator: Catherine Goyet/WHOI
       Crew Members: Healy/Sweeney/WHOI, Moore/RSMAS

 A number of instruments were used aboard the R/V Knorr to measure carbonate
 chemistry continuously in surface sea water and discretely in the water column. Underway measurements
include continuous analysis of total inorganic carbon, partial pressure of CO2, and chlorophyll
concentration. Discrete samples werealso taken from the underway sea water 
supply for total alkalinity and total organic carbon analysis. These discrete 
samples were taken on the average every 2 hours when transitting and at every 
station position for an average spacing of 30 nautical miles.
Discrete samples were taken from CTD casts and analyzed for total inorganic
carbon, total alkalinity, and total organic carbon. A total of 44 stations were sampled, 472 samples for
total organic carbon and 786 samples each for total inorganic carbon and total 
alkalinity. All total inorganic carbon and alkalinity analysis were performed 
at sea, approximately 75 percent of total organic carbon analysis will be done
post-cruise in the labs at WHOI.

_________________________________________________________________________________
II.7 RAFOS floats

  A RAFOS float is a freely drifting subsurface acoustic float which determines its
  position from moored sound sources and measures temperature and pressure along its 
  trajectory.  After a year the float surfaces and telemeters its accumulated data 
  to a satellite.  These floats were ballasted to drift a the 27.5 sigma-t density
  surface.
  
  These and other RAFOS float data are archived and accessible via the website: 
          http://wfdac.whoi.edu
  
  'KN154RAFOS_flttable.txt' gives an overview of all the floats 
launched on KN154, November-December 1997.
There are two files per float, the rfb-file contains the raw float data,
the rfc-file contains the final edited temperature, pressure, track data, and velocities.

ACCE KN154 RAFOS Float Summary (PIs: Bower, Richardson):
Float ID, Target Density, LAUNCH {Date Time(GMT) Lat  Lon}, ...	   
      SURFACE {Date Time(GMT) ARGOS Fix-time(GMT) Lat Lon}, Status Code 
__________________________________________________________________________
w321 27.5 971104 0257 52.251N 16.407W no show
w325 27.5 971021 2119 54.902N 34.996W 990810 0300 0716 54.994N 48.767W 66?
w341 27.5 971017 1520 46.154N 30.967W 991017 0300 0615 48.785N 26.871W 00?
w342 27.5 971019 0924 50.595N 33.489W 991019 0300 0407 53.325N 18.240W 00
w343 27.5 971018 1412 48.512N 32.285W 991018 0300 0600 54.915N 18.756W 00
w344 27.5 971103 0026 53.157N 14.999W 991102 0300 0450 58.661N 30.413W 00
w345 27.5 971023 0128 58.150N 31.900W no show
w346 27.5 971018 0922 48.037N 32.012W 991018 0300 0421 49.627N 12.739W 00
w347 27.5 971019 1203 50.832N 33.611W 991019 0300 0548 60.785N 27.380W 00
w348 27.5 971018 0058 47.097N 31.501W 991017 0300 0609 62.555N 22.046W 00
w350 27.5 971103 2314 52.245N 15.703W 991103 0300 0617 62.014N 25.806W 00
w351 27.5 971106 0330 48.715N 12.347W 980414 0300 0906 48.820N 11.730W 80
w352 27.5 971106 0531 48.585N 12.434W 991105 0300 0418 45.672N  6.568W 00
w353 27.5 971103 2009 52.252N 15.318W no show
w354 27.5 971104 1515 50.998N 16.014W 991104 0300 2018 46.894N  9.019W 00
w355 27.5 971017 2036 46.615N 31.202W 991017 0300 0749 60.193N 28.391W 00
w357 27.5 971106 1223 48.175N 12.987W 991105 0300 1011 50.184N 11.715W 00?
w358 27.5 971106 0831 48.404N 12.632W 991105 0300 0705 47.787N 16.894W 00
w360 27.5 971020 1603 52.538N 35.338W 991019 0300 0728 58.392N 23.913W 00
w361 27.5 971104 2201 51.222N 15.258W 991104 0300 0609 54.826N 10.737W 00
w362 27.5 971103 0556 53.167N 15.834W 991102 0300 1116 57.004N 17.829W 00
w363 27.5 971103 0837 53.174N 16.245W 991102 0300 0950 62.078N 23.396W 00
w364 27.5 971104 1109 50.865N 16.525W 991104 0300 1034 50.296N 21.474W 00
w389 27.5 971019 0758 50.364N 33.344W 991018 0300 0558 61.235N 33.690W 00
w400 27.5 971019 0421 50.129N 33.217W 991018 0300 0600 54.647N 27.527W 00
w413 27.5 971018 2222 49.453N 32.805W 991018 0300 0704 60.667N 21.995W 00
w414 27.5 971018 0550 47.594N 31.752W 991018 0300 0830 57.549N 20.564W 00
w415 27.5 971020 0505 52.304N 34.170W 991019 0300 0731 57.390N 39.197W 00
w416 27.5 971018 1847 48.998N 32.545W 991018 0300 0557 62.750N 22.499W 00
w418 27.5 971018 2336 49.660N 32.939W 990616 0300 0556 47.939N 36.222W 66
w419 27.5 971105 0000 51.266N 15.061W 991104 0300 1033 46.347N 18.179W 00?
w420 27.5 971019 0306 49.913N 33.078W 991018 0300 0741 52.544N 33.255W 00
w421 27.5 971022 0715 56.110N 34.249W 991021 0300 0847 63.016N 53.656W 00
w422 27.5 971021 1328 54.001N 35.502W 991018 0300 0846 60.979N 17.608W 00
w423 27.5 971017 1118 45.687N 30.747W 991017 0241 0613 42.752N 24.555W 00 
w424 27.5 971017 0548 45.216N 30.473W 991016 0300 0624 46.200N 24.811W 00? 
w425 27.5 971022 1741 57.326N 32.958W 991022 0300 1016 63.771N 55.439W 00
w426 27.5 971023 1235 59.389N 30.262W 991023 0300 0649 43.242N 52.165W 00
w427 27.5 971021 0507 52.867N 35.384W 991020 0300 0536 64.467N 30.141W 00
w428 27.5 971019 1658 51.309N 33.888W 991019 0300 0642 57.755N 21.639W 00
w429 27.5 971019 2136 51.781N 34.144W 991019 0300 0549 55.248N 34.145W 00
________________________________________________________________________________
Status codes at end of float mission:  0, 00: normal mission,  66: low battery,  
80: over pressure,  83: lost weight.  
If '?', then first message not received, and status code is assumed.

________________________________________________________________________________


II.8 PALACE floats



________________________________________________________________________________
III.  References

Fischer, J., and M. Visbeck, 1993.  Deep velocity profiling with
     self-contained ADCPs.  J. Atmos. Oceanic Technol., 10, 764-773.

Giles, Alan B. and Trevor J. McDonald. 1986. Two methods for the reduction
     of Slalinity Spiking of CTDs. Deep Sea Research. vol 33, no 9.
     1253-1274.

Joyce, T. M., 1989.  On in situ "calibration" of shipboard
     ADCPs.  J. Atmos. Oceanic Technol., 6, 169-172.

Knapp, G.P., M. Stalcup and R.J. Stanley (1990), Automated Oxygen
     Titration and Salinity Determination,.  WHOI Technical Report, WHOI-90-35,
     25 pp.

Magnum, B.W. and G.T. Furukawa.1990. Guidelines for Realizing the 
     International Temperature Scale of 1990 (ITS-90). Nist Technical 
     Notes 1265.

Millard, R. C. and K. Yang.1993. CTD Calibration and Processing Methods
     used at Woods Hole Oceanographic Insitute. Technical Report No. 93-44,
     96 pages.

Oceansoft MKIII/SCTD Aquisition Software Manual. 1990. P/N Manual 10239.
     EG&G Marine Instruments.

Owens, Brechner W. and Robert C. Millard, Jr.1985. A New Algorithm for 
     CTD Oxygen Calibrations. J. Phys. Oc. vol 15.621-631. 

Pollard, R., and J. Read, 1989.  A method for calibrating
     shipmounted Acoustic Doppler Profilers, and the
     limitations of gyro compasses. J. Atmos. Oceanic Technol.,
     6, 859-865

Strickland, J.D.H. and T.R. Parsons (1972)  The Practical Handbook of
     Seawater Analysis.  Bulletin 167, Fisheries Research Board of Canada, 310        pp.

