RV Pelagia Cruise Report:

Cruise 64PE122, Project TripleB, 

WHP repeat area AR12

H.M. van Aken
Chief Scientist



Bay of
   Biscay
      Boundary
NIOZ, Texel, 199



Table of contents

nr.	Chapter	page
1	Cruise Narrative	5
1.1  	Highlights	5
1.2  	Cruise Summary Information	5
1.3  	List of Principal Investigators	9
1.4  	Scientific Programme and Methods	10
1.5  	Major Problems Encountered during the Cruise	16
1.6  	List of Cruise Participants	17
2	Underway Measurements	18
2.1  	Navigation	18
2.2  	Echo Sounding	18
2.3  	Thermo-Salinograph Measurements	18
2.4  	Meteorological data	18
3	Hydrographic Measurements -Descriptions, Techniques, and Calibrations	19
3.1  	Rosette Sampler and Sampler Bottles	19
3.2  	Pressure Measurements	19
3.3  	Temperature Measurements	19
3.4  	Salinity Measurements	20
3.5  	Oxygen Measurements	20
3.6  	Nutrient Measurements	21
3.7  	CTD Data Collection and Processing	22
3.8  	Micro-structure profiler measurements	23
4	Acknowledgements	23
	Appendix A (cruise summary file)	25
	Appendix B (mooring information file)	35








The research reported here was funded by the Foundation 
for Geological, Oceanographic and Atmospheric Research 
(GOA), subsidiary of the Netherlands Organization for 
Scientific Research (NWO

1 Cruise Narrative

1.1	Highlights

a:	WOCE Repeat Area AR12, RV Pelagia cruise PE122 in the Bay of Biscay

b:	Expedition Designation (EXPOCODE): 64PE122

c:	Chief Scientist:	Dr. Hendrik M. van Aken
				Netherlands Institute for Sea Research (NIOZ)
				P.O. Box 59
				1790AB Den Burg/Texel
				The Netherlands
				Telephone:	31(0)222-369416
				Telefax:	31(0)222-319674
				e-mail:	aken@nioz.nl

d:	Ship:  RV Pelagia,	Call Sign:  PGRQ
					length 66 m.
					beam 12.8 m
					draft 4 m
					maximum speed 12.5 knots

e:	Ports of Call:	Vigo (Spain) to Vigo (Spain)

f:	Cruise dates:	August 18, 1998 to September 2, 1998

1.2	Cruise Summary Information

Summary

Early before noon of 18 August RV Pelagia left the quay in the Spanish port of Vigo, and headed 
for the Bay of Biscay. We arrived at the Biscay continental slope near Gijn in the evening of 19 
August, and spent the night carrying out an echo sounder survey over the slope. The following day 
moorings BB22 to BB24 were recovered. Mooring BB21, deployed the previous year, was broken due 
to corrosion, and the upper part with two current meters was salvaged by a fisherman from Aviles, and 
subsequently brought to Vigo. After recovery of the moorings course was set to mooring BB20 near 
the Meriadzek Plateau. Underway two CTD casts were taken for test purposes, as well as one micro-
structure profiler (MSP) cast, while two ARGOS surface drifters were deployed On 22 August, after a 
nightly echo sounder survey, moorings BB17 to BB 20 were recovered. After recovery of the last 
mooring course was set to the Armorican Shelf, south-west of Brest. From there a section (A) with 
CTD and MSP stations was carried out towards the Spanish continental shelf near Gijn. On this 
section three more ARGOS drifters were deployed. Section A was finished on 26 August, and course 
was set to start the survey of section B, about 60 mile further eastwards. This section was surveyed 
from 26 to 29 August with CTD as well as with MSP casts. The survey of the final section C started at 
29 August. After midnight in the morning of 31 August course was set to Vigo, while during that day 
the analyses of the last water samples from the rosette system and the thermo-salinograph were 
finished. The ship arrived in the port of Vigo in the morning of  1 September. During that day shifting 
of laboratory containers was started, and the transport back to Texel of the buoyancy spheres of the 
moorings and of a container with all current meters and computers was prepared, while the last 
preliminary data processing was finished. Early in the morning of 2 September the scientific crew 
debarked for Texel.

Cruise Track

The cruise was carried out in the Bay of Biscay. The cruise track is shown in figure 1.


Figure 1.	Cruise track of RV Pelagia cruise 64PE122

Number of Hydrographic Stations

A total of 44 CTD casts was recorded. On 42 of these casts, water samples were taken for the 
determinations of salinity and dissolved oxygen and nutrients. Two water samplers in the rosette 
system were fitted with high accuracy reversing electronic pressure sensors. The positions of the 
hydrographic stations are indicated in figure 2.

At the hydrographic stations the SBE9/11+ CTD was lowered with a speed of about 1 m/s. Due  to 
the use of a bottom indicator switch we were able to sample to within quite a short distance from the 
bottom (5 m).


Figure 2.	Distribution of hydrographic stations and sections.

Hydrographic Sampling

During the up-cast of each CTD/rosette station water up to 25 samples were taken at regular depth 
intervals. The samples were analysed for salinity, oxygen and nutrients. For test purposes also 
Dissolved Inorganic Carbon (DIC) was determined from the samples collected on section A. These 
latter data will not be reported further, because of their still experimental status and questionable 
quality. 
The vertical distribution of the sampling locations is indicated in figure 3.

Micro-structure profiler casts

Along sections A, B, and C 20 micro-structure profiler (MSP) casts were carried out after the CTD 
cast of the stations in order to estimate turbulent intensity and dissipation. The tethered free falling 
probe was lowered close to the bottom, or to 1000 m, whichever was the shallowest. The distribution 
of the microstructure profiler casts is indicated in figure 4.


Figure 3.	Vertical distribution of the water samples versus station number.

Moorings

The moorings BB17, BB18, BB19, BB20, BB22, BB23, and BB24, deployed in 1997, were 
recovered without problems. Two of the NBA current meters suffered from malfunctioning of the 
compass, while one Aanderaa RCM 8 current meter suffered from a stalled rotor. The only Aanderaa 
RCM 9 acoustic current meter had problems with the digitizing of the velocity components, just in the 
winter period. Possibly this was due to the lack of acoustic reflectors in this season.

After recovery of the current meters, their data were read and copied to the computer network of 
RV Pelagia. At Texel the data were processed, and corrected for the magnetic variation and clock 
errors. From the processed data a low-pass data set, sub-sampled every 12 hours, as well as a high-pass 
data set, sub-sampled every hour have been produced.

Detailed information on the moorings is given in the list in appendix B. The station and cast 
numbers in this list refer to the station and cast numbers end positions referred to in the *.SUM file 
(Appendix A). In Fig. 5 the mooring positions are indicated with crosses, and BB labels.


Figure 4	Distribution of stations with micro-structure profiler casts.


ARGOS drifters

During the cruise five ARGOS drifters were deployed. The drifters used were standard spherical 
WOCE/TOGA mixed layer drifters (diameter 30 cm), fitted with a holey sock drogue at 15 m. The 
drogues had a length of 7 m, and a diameter of 1 m. The ARGOS ptt numbers of the drifters are 16118 
to 16122. In figure 5 the deployment positions are indicated with tilted squares and italic labels.

*.SUM file

A hard copy of the *.SUM file describing all stations is added in the appendix A.

1.3 List of Principal Investigators
	
Name	Responsibility	Affiliation

Dr. H.M. van Aken 	Ocean hydrography, ARGOS drifters.	NIOZ/Texel
Drs. C. Veth		micro-structure measurements.		NIOZ/Texel
Drs. J. Ligtenberg	Slope currents,				NIOZ/Texel
				current measurements


Figure 5.	Positions of the recovered moorings (crosses, BB labels) and the deployed ARGOS drifters 
(tilted squares and italic labels).

1.4 Scientific Programme and Methods

The goal of the research carried out during the cruise was to establish the structure, course 
and transport of the eastern boundary current in the Bay of Biscay, as well as the hydrographic 
structure of the Bay of Biscay, as it is affected by the eastern boundary current. For this purpose a 
hydrographic survey has been carried out in the Bay of Biscay, and five ARGOS surface drifters have 
been deployed. Seven long term current meter moorings, deployed in 1997, were recovered. The 
hydrographic survey covers part of the WOCE Hydrographic Research Programme repeat area AR12, 
and complements the hydrographic surveys & mooring and drifter deployments, carried out in 1995, 
1996 and 1997 in the Bay of Biscay in the TripleB programme.

The CTD-rosette frame was fitted with weights in order to secure a fast enough falling rate. This 
package was lowered with a velocity of about 1 m/s, except in the lowest 100 m, where the veering 
velocity was reduced. Measurements during the down-cast went on to within 5 m from the bottom, 
until the bottom switch indicated the proximity of the bottom. During the up-cast water samples where 
taken at prescribed depths, when the CTD winch was stopped. After each cast the CTD/rosette frame 
was placed on deck. Subsequently water samples were drawn for the determination of dissolved 
oxygen, nutrients, and salinity, and the readings of the reversing electronic pressure sensors were 
recorded.


Preliminary Results

After the cruise all CTD data have been processed and calibrated at NIOZ, Texel, as described in 
chapters 2 and 3. A short, preliminary overview of the results is given here. The vertical distributions 
of potential temperature (THETA) and salinity (CTDSAL) versus pressure (Figure 6) show much 
variability between 500 and 1500 dbar because of the presence of the warm and saline core of 
Mediterranean Sea Water (MSW) at about 1000 dbar. At a pressure of approximately 1800 dbar a 
number of stations have a salinity minimum caused by the presence of a core of low salinity Labrador 
Sea Water (LSW). Between the Eastern north Atlantic Central Water (ENACW) in the permanent 
thermocline and the MSW core a subsurface salinity minimum is observed at all deep stations. In the 
upper few tens of dbars of the water columns a wide variety of salinities is encountered with values 
above as well as below the salinity of the top of the underlying water.


Figure 6. Profiles of (a) potential temperature (THETA) and (b) salinity (CTDSAL) versus pressure 
(CTDPRS)

The _-S diagram (Figure 7a) confirms the presence of ENACW at potential density anomalies _ in 
the approximate range 27.05 < _(_,S) < 27.25 kg/m3, whereas MSW is found in the density range 
27.25 < _(_,S) < 27.75 kg/m3. LSW with _ _ 3.6_C and S _ 35.0 is found at a potential density 
anomaly of about 27.8 kg/m3. The vertical distribution of potential vorticity (Figure 7b) shows that 
potential vorticity minima (the Mode Water of the previous winter?) are found near the 
_(_,s) = 27.10 kg/m3 isopycnal near the top of the permanent thermocline. At the density levels of  the 
LSW and below the potential vorticity reaches even lower values. At the lowest levels the water 
approaches a potential temperature _ _ 2.0_C and a salinity S _ 34.90, characteristic for Lower Deep 
Water (LDW).


Figure 7. _-S diagram (a) and plot of baroclinic potential vorticty versus potential density anomaly (b) 
for all CTD casts. The dashed lines in (a) depict lines of constant potential density anomaly.


|Figure 8. Plots of dissolved oxygen (a, OXYGEN) and dissolved phosphate (b, PHSPHT) versus 
potential temperature (THETA).

The distribution of the non-conservative parameters also reflect the known water mass structure. The 
plot of dissolved oxygen versus potential temperature (Figure 8a) shows an oxygen minimum 
coincident with the MSW core near _ = 10_C, a maximum near the LSW core with _ = 3.6 to 4.0_C, 
and a near bottom minimum at LDW temperatures (_ = 2.0_C). At temperatures characteristic of the 
top of the permanent thermocline and the Mode Water (12 to 13_C) the O2-_ line shows a shift 
towards relatively lower oxygen concentrations in the seasonal thermocline. The distribution of the 
nutrients in temperature space (Figures 8b and 9) shows a similar structure with inflexion points at the 
levels of the MSW and LSW cores, and near bottom maxima connected with the LDW. At the highest 
temperatures, close to the sea surface, the nutrient concentrations approach zero because of the nutrient 
use during primary production.


Figure 9. Plots of dissolved nitrate (a, NITRAT) and dissolved silica (b, SILCAT) versus potential 
temperature (THETA).

The dissolved nutrients show a more or less one to one relation (Figure 10). This relation is 
however not linear over the whole range of concentration values, but parts of the nutrient relations can 
be approximated by linear relations with a constant empirical stochiometric ratio. This applies to the 
water column above the MSW core, where _P:_N:_Si = 1:16.5:5.5, and the water column below the 
LSW core, where _P:_N:_Si = 1:13:90. Most probably these ratios are not solely determined by the 
ratios in which the nutrients are consumed or produced in fixed ratios in biochemical processes (and 
dissolution for silica), but also by mixing of the different water masses with their different 
characteristic in situ and pre-formed nutrient concentrations. Especially in the deep water mass the 
very high dissolved silica concentration of the LDW core due to the contribution of Antarctic Bottom 
Water to this near bottom water mass is responsible for the low _P:_Si ratio.


Figure 10. Plots of the concentrations of dissolved nitrate (a, NITRAT) and silica (b, SILCAT) versus 
the concentration of dissolved phosphate (PHSPHT). The straight lines indicate linear 
dependencies according to the empirical stochiometric ratios given in the figures.



The Sea Surface Temperature (SST) from the continuously recording thermo-salinograph (Figure 
11) shows temperatures of over 22_C close to the Spanish coast near 5_W. As in 1997 relatively low 
temperatures (SST < 18_C) were encountered near the continental slope south-west of Brittanny. The 
lowest surface temperatures (SST < 16_C) were observed over the continental slope near western 
Galicia where upwelling took place. The Sea Surface Salinity (SSS) in the central Bay of Biscay 
shows values of over 35.8, while near the continental slope and shelf lower salinities have been 
observed (Figure 12). Also the sea surface oxygen concentration was observed with the continuously 
recording measurement system. From these data the Apparent Oxygen Utilization (AOU) was 
determined. The AOU values of the surface water were nearly everywhere negative, of the order of 5 
to 10 _mol/kg. (Figure 13). This reflects a slight over-saturation of the surface water, probably due to 
primary production in the surface layers. In the upwelling area off western Galicia AOU values in the 
surface water of over 40 _mol/kg were observed.


Figure 11. Horizontal distribution of the Sea Surface Temperature (SST) measured with the 
AQUAFLOW thermosalinograph system. The thick line indicates the 200 m isobath which coincides 
with the position of the upper continental slope. The dots show the hourly positions of RV Pelagia.


Figure 12. The horizontal distribution of the Sea Surface Salinity (SSS) as measured with the 
thermosalinograph system. The thick line indicates the 200 dbar isobath, and the dots the hourly 
positions.


Figure 13. Horizontal distribution of the Sea Surface Apparent Oxygen Utilization (AOU) derived 
from the oxygen concentration as measured with the AQUAFLOW thermosalinograph system. 
The thick line indicates the position of the 200 m isobath, and the dots the hourly positions.


1.5 Major Problems Encountered during the Cruise

With the recovery of the current meter moorings it appeared that pit corrosion had seriously 
affected the stainless steel tow bars of some NBA current meters, as well as the frame of the ADCP. 
With all affected tow bars it appeared that the PVC isolator, mounted to prevent direct contact between 
tow-bar and swivel bolt was damaged, due to wear. This caused direct contact between the tow bar and 
the stainless steel bolt in the tow bar, suggesting an electrochemical cause of the corrosion. Corrosion 
of the tow bar at the side of the swivel was also the cause of the partial loss of mooring BB21. The 
future use of different types of stainless steel and/or insulators in moorings should be re-considered.

The rotating tow link between the winch cable and the CTD twice gave an electrical shortcut. Both 
times it was caused by sea water entering the link due to leakage. By fast repair as well as the use of a 
spare tow link the problems were solved easily. The use of another type of tubing in the tow link 
should be considered.

Although the throughput in oxygen samples per day with the spectro-photometer is considerably 
larger than with the automated Winkler titration used during previous cruises, a first comparison of 
duplicate samples suggests that the spectro-photometer is considerably less precise. But the RMS of 
the difference between the duplicate (O(1.0 _mol/dm3)) is considerably less than the oxygen signal in 
the water column (O(90 _mol/dm3)). A further analysis, with regard of the applicability of the spectro-
photometer and other new methods for the high precision determination of oxygen has to be made at 
NIOZ.

Some minor problems turned up with the computer network and the recording of underway 
sampling. The automated backup of the underway data used backup tapes in a much faster rate than 
intended. The cause of this feature should be found and repaired. Halfway the cruise the signal from 
the air temperature sensor disappeared. The cause could not be found, and is considered to be located 
in the meteorological interfaces, supplied by KNMI. Documentation for repair of the interface was not 
available, and should be obtained. The fluorescence channel of the underway monitoring system only 
gave erroneous white-noise-like data as it did in 1997. Maintenance of this sensor should be 
reconsidered. The echo sounder could not always digitize the depth because of blurred echoes over 
steep and deep slopes. However the new software used for the recording of underway sampling then 
always recorded the previous value instead of an error value. The new software should be reconsidered 
in co-operation with scientific users of the data.

1.6	Lists of Cruise Participants

Scientific crew

person		responsibility						Institute

H.M. van Aken	Chief Scientist, ARGOS drifters, Data management	NIOZ
D. van As		Salinity determination						IMAU
K. Bakker		Nutrient & DIC determination					NIOZ
M. Bakker		Mooring technology & maintenance				NIOZ
R.L. Groenewegen	Acoustic releases, Electronics				NIOZ
M. Hiehle		Salinity determination						NIOZ
M.T.J. Hillebrand	Current meters, Hydro watch					NIOZ
R.X. de Koster	Data management, Hydro watch					NIOZ
J. Ligtenberg	Data processing current meters, Hydro watch		NIOZ
Y. Muilwijk		Hydro watch								IMAU
S. Ober		ADCP, Hydro watch							NIOZ
M. Smit		Oxygen determination						UT
A. van Veldhoven	Oxygen determination						IMAU
C. Veth		Current Meters, Hydro Watch					NIOZ
L. Wuis		Mooring Technology & maintenance				NIOZ

NIOZ:	Netherlands Institute for Sea Research, Texel
IMAU:	Institute for Marine and Atmospheric Research, Utrecht University
UT:	Faculty of Civil Engineering and Management, Twente University


Ships crew

J. J. Jongedijk	captain
M.D. van Duijn	first mate
H.A.M. Douma	second mate
J. Pieterse		first engineer
J. Seepma		second engineer
H. de Vries		cook
P.W. Grisnicht	sailor AB
P.-W. Saalmink	sailor AB
C.T. Stevens	sailor AB
E.W. Weuring	sailor AB


2	Underway Measurements

2.1 Navigation

Differential GPS receiver for the determination of the position. The data from the receiver were 
recorded every  ten seconds in the underway data logging system. After removal of a few spikes these 
data were sub-sampled every minute.

2.2 Echo Sounding

The 3.5 kHz echo sounder was used on board to determine the water depth. The uncorrected depths 
from this echo sounder were recorded in the underway data logging system. Over the steepest parts of 
the continental slope the depth digitizer of the echo sounder was occasionally not able to find a reliable 
depth. 
Preceding the recovery of the current meter moorings one 2 or 3 lines were surveyed to determine 
the mooring sites, complementary to the echo sounder surveys of the mooring sites carried out in 1997. 
During these surveys the echo sounder data were also recorded on paper chart in order to allow hand 
digitizing over the parts of the slope where the automatic digitizing failed.

2.3 Thermo-Salinograph Measurements

The Sea Surface Temperature, Salinity, and dissolved Oxygen concentration were measured 
continuously with an AQUAFLOW thermo-salinograph system with the water intake at a depth of 
about 3 m. For the calibration of the salinity sensor and the oxygen sensor, water samples were taken 
three times per day.

2.4 Meteorological data

Air temperature and humidity, relative wind velocity and direction as well as air pressure were 
measured and recorded by the underway logging system. During part of the cruise the air temperature 
recording failed.


3	Hydrographic measurements - Descriptions, Techniques, and 
Calibrations

3.1 Rosette Sampler and Sampler Bottles

A 25 position rosette sampler was used, fitted with 5 and 10 litre NOEX sampler bottles. A multi-
valve system, developed at NIOZ, allowed closing the sampler bottles by computer command from the 
CTD operator. The general behaviour of the samplers was good. Only a few samples are considered to 
be suspect because of sampler failure. No errors in the functioning of the rosette sampler itself could 
be detected, except a failure of the electric motor. This could be solved easily by installing a spare 
motor.

3.2 Pressure Measurements

On sampler bottles 2 and 7 thermometer racks were mounted, fitted with SIS high accuracy 
reversing electronic pressure sensors. Sampler 2 contained two such pressure sensors, sampler 7 only 
one. Before the cruise these sensors were calibrated at the national calibration facility of the Van 
Swinden Laboratory of the Netherlands Measuring Institute (NMI). On deck, prior to the CTD the 
sensors automatically recorded the air pressure and used this air pressure for further correction of the 
sea pressure. These pressure values have been reported as REVPRS. The duplicate values of REVPRS 
from sampler 2 indicated a RMS (Root Mean Square) of the difference between both SIS pressure 
sensors of 0.7 dbar. Comparison of the REVPRS values with the pressure, measured with the CTD 
(CTDPRS) revealed a pressure dependent pressure difference (2.5 dbar/5000 dbar). This is probably 
due to a slight difference between the NMI calibration facility and the pressure calibration used by the 
manufacturer of the SBE CTD. Since the specifications of the NMI facility are considered to be better, 
it was decided to correct CTDPRS with a linear correction. After applying this calibration to CTDTMP 
the RMS of the difference REVPRS-CTDPRS amounts to 0.6 dbar (73 samples).

3.3 Temperature Measurements

Mounted on the CTD-rack was a high precision SBE35 reference temperature sensor, which 
recorded the temperature every time a sampler was closed. The zero point of this temperature sensor 
has been calibrated regularly by means of an H2O triple point cell. Comparison of the primary CTD 
temperature sensor with the SBE35 sensor indicate that the temperatures, measured with the CTD is 
very close to the reference temperature. The temperature measured with the SBE35 sensor have been 
reported as the parameter REVTMP.

The temperatures, measured with the CTD during sampling were first corrected for its known 
pressure dependence (0.7 mK/5000 dbar). These corrected values (CTDTMP) were then compared 
with REVTMP. the scatter of the difference between both temperature showed a strong decrease with 
increasing pressure, probably due to the decreasing errors caused by the vertical temperature gradient. 
The mean difference between REVTMP-CTDTMP for all samples taken at levels below the 3000 dbar 
surface (117 samples) amounted to 0.2 mK, with a RMS of the difference of only 0.8 mK. It was 
decided not to apply any further temperature correction for CTDTMP.

3.4 Salinity Measurements

Water was drawn from the samplers into a 0.5 litre glass sample bottle for the salinity 
determination after 3 times rinsing. The sample bottles had a massive rubber stopper as well as a screw 
lid. Salinity of water samples (SALNTY) was determined on board by means of an Guildline Autosal 
8400A salinometer. The salinometer was used in a laboratory container, fitted with an air conditioning 
system. This kept the surrounding air temperature constant within 1C. The readings of the instrument 
were performed by computer, giving the average and statistics of 10 consecutive readings. For each 
sample 3 salinity determinations were carried out. The OSI standard water used was from batch P133 
with a K15 ratio of 0.99986 (S=34.995), prepared at 11 November 1997.

From each deep CTD/rosette cast an extra duplicate sample was drawn. Salinity determinations 
from the duplicate samples obtained from independent runs were used to determine the reproducibility 
of the salinity determination. The RMS difference between the SALNTY duplicate samples amounted 
to 0.0006.

SALNTY was compared with the salinity reading from the CTD (CTDSAL). The mean difference 
SALNTY-CTDSAL amounted to -0.0005. It was decided to apply an offset correction to the salinities, 
measured with the CTD. The scatter of the difference between SALNTY and the corrected CTDSAL 
showed a clear pressure dependence, probably because of the downward decrease of the vertical 
salinity gradient. For all samples taken below the level of the 3000 dbar surface the RMS difference 
SALNTY- CTDSAL amounted to 0.0010, with a mean difference of 0.0000 (66 samples).

3.5 Oxygen Measurements

For the oxygen determination water samples were drawn in volume calibrated 120 ml pyrex glass 
bottles. Before drawing the sample each bottle was flushed with at least 3 times its volume. When the 
samples were drawn the temperature of the sample was determined. The determination of the 
volumetric dissolved oxygen concentration of water samples was carried out by means of a spectro-
photometer Winkler technique, recently developed at NIOZ [see Su-Chen Pai et al., Marine Chemistry 
41 (1993), 343-351]. Before and after the cruise the spectro-photometer were inter-calibrated with a 
automatic end point determination Winkler method. The stock solution of KJO3 used in the analysis 
was prepared and calibrated in the laboratory by using gravimetric methods. The stock solutions were 
stored at low temperature (~4_C). 
The inter-calibration with the automatic end point calibration revealed a systematic difference with 
the spectro-photometer values of 0.6 _mol/dm3. The oxygen data were corrected for this difference, 
and a sea water blank of also 0.7 _mol/dm3, determined during earlier TripleB cruises, was subtracted 
consequently. At each cast duplicate samples were taken from the deepest and shallowest sampler, and 
occasionally from a sampler at an intermediate level. The RMS of the differences between the 
duplicate samples amounts to 0.77 _mol/dm3.

From the volumetric oxygen concentration in _mol/dm3 the densimetric oxygen concentration in 
_mol/kg (OXYGEN) was determined by dividing by the sample density at sample temperature and 
salinity. 

3.6 Nutrient Measurements

From all sampler bottles samples were drawn for the determination of the nutrients silica, nitrite, 
nitrate and phosphate. The samples were collected in polyethylene sample bottles after three times 
rinsing. The samples were stored dark and cool at 4_C. All samples were analysed for the nutrients 
silicate, phosphate, nitrate and nitrite within 10 hours with an autoanalyzer based on colorimetry. The 
lab container was equipped with a Technicon TRAACS 800 autoanalyzer. The samples, taken from the 
refrigerator, were directly pored in open polyethylene vials (6ml) and put in the auto sampler-trays. A 
maximum of 60 samples in each run was analysed.

The different nutrients were measured colorimetrical as described by Grashoff (1983);
_   Silicate reacts with ammoniummolybdate to a yellow complex, after reduction with ascorbic acid 
the obtained blue silica-molybdenum complex was measured at 800nm (oxalic acid was used to 
prevent formation of the blue phosphate-molybdenum).

_   Phosphate reacts with ammoniummolybdate at pH 1.0, and potassiumantimonyltartrate was used 
as an inhibitor. The yellow phosphate-molybdenum complex was reduced by ascorbic acid to blue 
and measured at 880nm.

_   Nitrate was mixed with a buffer imidazole at pH 7.5 and reduced by a copperized-cadmium coil 
(efficiency> 98%) to nitrite, and measured as nitrite (see nitrite). The reduction-efficiency of the 
cadmium-column was measured in each run.

_   Nitrite was diazotated with sulphanilamide and naftylethylenediamine to a pink coloured 
complex and measured at 550nm.

_   The difference of the last two measurements gave the nitrate content
Calibration standards were prepared by diluting stock solutions of the different nutrients in the 
same nutrient depleted surface ocean water as used for the baseline water. The standards were kept 
dark and cool in the same refrigerator as the samples. Standards were prepared fresh every two days. 
Each run of the system had a correlation coefficient for the standards off at least 0.9998. The samples 
were measured from the surface to the bottom to get the smallest possible carry-over-effects. In every 
run a mixed control nutrient standard containing silicate, phosphate and nitrate in a constant and well 
known ratio, a so-called nutrient-cocktail, was measured, as well as control standards, sterilized in an 
autoclave or gamma radiation. These standards  were used as a guide to check the performance of the 
analysis and the gain factor of the autoanalyzer channels. The reduction-efficiency of the cadmium-
column in the nitrate lane was measured in each run.

The autoanalyzer determined the volumetric concentration (_mol/dm3) at a temperature of 20_C. In 
order to obtain the densimetric concentration in _mol/kg the volumetric concentrations were divided 
by the density of sea water at 20_C, sample salinity, and zero sea pressure.

Duplicate measurements carried out on the deepest sample from each cast gave RMS values of the 
differences of  0.19 _mol/kg, 0.01 _mol/kg, 0.16 _mol/kg, and 0.03 _mol/kg for respectively 
dissolved silica (SILCAT), nitrite (NITRIT), nitrate (NITRAT), and phosphate (PHSPHT). Possible 
variations in the gain factor of the different channels of the autoanalyzer were determined by means of 
the nutrient cocktail (phosphate), the autoclave sterilized standard (nitrate), and the gamma ray 
sterilized standard (silica). A gain factor correction, applied to the duplicate samples, resulted in a 
reduction of the RMS of the duplicate differences to respectively 0.14 _mol/kg (SILCAT), 
0.13 _mol/kg (NITRAT), and 0.02 _mol/kg (PHSPHT). Thereupon the determined gain factors were 
applied to all SILCAT, NITRAT and PHSPHT values.

3.7 CTD Data Collection and Processing

The SBE 9/11+  CTD was fitted with temperature sensor SN1219 and conductivity sensor SN1046. 
For the data collection SEASAVE software, version 4.218, supplied by SBE, was used. The CTD data 
were recorded with a frequency of 24 data cycles per second. On-line a correction was applied for the 
sampling time difference due to the forced flushing through a tube system between temperature and 
salinity sensor. After each CTD cast the data were copied to a hard disk of the ship's computer 
network, and a daily back-up copy was made on tape. Back on Texel these data have been downloaded 
into the NIOZ computer network, via the network connection in the harbour. Separate copies of the 
back up were taken directly from Vigo to Texel.

The up-cast data files were sub-sampled to produce files with CTD data corresponding to each 
water sample, taken with the rosette sampler. After the determination of the final calibration of the 
CTD system these values were corrected accordingly.

After the cruise the raw down-cast CTD data were processed with the SEASOFT software supplied 
by SBE. A correction was applied for the temperature change between the temperature and 
conductivity sensor due to heat exchange with the flushing tube and conductivity sensor, and for 
different response times of both sensors. The correction factors for these corrections were determined 
empirically from CTD stations over the continental shelf. At these stations the vertical temperature 
gradient in the seasonal thermocline were largest, and so were the resulting salinity spikes. The 
correction settings, determined for the 1997 TripleB data appeared to be adequate for the corrections. 
Time series of mean values of the readings were produced for 0.5 s time intervals (bins). This is, given 
the typical veering velocity of about 1 m/s equivalent to an approximate pressure interval of 0.5 dbar. 
Consecutively the parameter values in physical units were determined using the final calibration 
constants.

It appeared that the SEASOFT software did not remove all spikes in the data record. Therefore the 
remaining spikes were removed by applying a median filter over 5 consecutive 0.5 s time bins. 
Hereafter the time series was filtered with a running mean over 5 time bins. Finally the time series 
were interpolated on equidistant 1 dbar pressure intervals, only using data without pressure overlap 
due to varying vertical motion of the CTD probe. Since no pressure bin averaging was applied, the 
NUMBER of OBS. in the *.CTD files was set by default to 12, the typical number of individual data 
point used to obtain the time series 0.5 s time bins which were used for the interpolation at equidistant 
pressure levels.

3.8	Micro-structure profiler measurements.

The micro-structure profiler used is a model FLY II manufactured by Sy-Tech Research Ltd. This 
instrument is tethered to a kevlar cable with electrical connector. A line puller supplies enough cable to 
allow a free fall of the instrument with a velocity of approximately 1 m/s. The FLY contains two 
perpendicular shear probes which are sampled with 280 Hz. Additionally the pressure, temperature, 
conductivity, and two tilt components are sampled with lower frequency (20 Hz). The instrument was 
lowered from the stern, while the ship had a speed of about 0.5 knots, propelled by the bow thruster. 
The maximum operational depth of the FLY II amounts to 1000 m.


4 Acknowledgements

The research reported here was funded by the Foundation for Geological, Oceanographic and 
Atmospheric Research (GOA), subsidiary of the Netherlands Organisation for Scientific Research 
(NWO), under contract number 750.197.01
I thank the ships crew and the personnel of the supporting technical departments of NIOZ for their 
professional support and active participation in the preparation and execution of the TripleB 
programme, especially for the cruise reported here. The contributions of the colleagues from the NIOZ 
department if Physical Oceanography is highly acknowledged, as is the active participation of the 
students from Utrecht and Twente.
