A.    Cruise Narrative:  S05 (GREAT AUSTRALIAN BIGHT)


A.1.  Highlights
                       WHP Cruise Summary Information

            WOCE line designation  S05
Expedition designation (ExpoCode)  09FA1094
      Chief scientist/affiliation  Matthias Tomczak/FIAMS *
                             Ship  RV Franklin
                     Cruise dates  1994 NOV.12 to 1994.DEC.05
                    Ports of call  Fremantle - Port Lincoln
               Number of stations  69
                                               33° 51.32' S
            Geographic boundaries  119° 59.6' E            132° 20.18' E
                                                48° 3.9' S
     Floats and drifters deployed  0
   Moorings deployed or recovered  0
                          Authors  Neil White

                          * Professor Matthias Tomczak
            Flinders Institute for Atmospheric and Marine Sciences,
                      Flinders University  -  GPO Box 2100
                         Adelaide  SA  5001  Australia
61-8-201-2298 (phone)  61-8-201-3573 (fax)  m.tomczak@flinders.edu.au (email)



A.2.  RESEARCH SUMMARY

Abstract: 

This dataset contains the processed Hydrology (HYD) data collected on Franklin 
voyage FR 10/94. The voyage took place in the Great Australian Bight and the 
Southern Ocean during 12 November - 5 December 1994. This dataset has been 
processed and is archived within the CSIRO Marine Research Data Centre in 
Hobart. Additional information regarding this dataset is contained in the cruise 
report for this voyage and/or the data processing report (as available). 
Franklin on-voyage hydrology data are typically obtained from water samples 
collected in Niskin bottles at various depths during each CTD cast. Parameters 
measured normally comprise temperature, salinity, dissolved oxygen, phosphate, 
nitrate, silicate and nitrite.


Scientific Programme

To determine the movement of Antarctic Intermediate Water and Indian Central 
Water in the Great Australian Bight and possible exchange between the Pacific 
and Indian Oceans in the depth range of these water masses.

To investigate the flow of Antarctic Bottom Water through the Australian-
Antarctic Discordance.

Cruise objectives

To complete three sections extending meridionally to 48°S along 120°42'E, 
zonally along 48°S to 132°E and meridionally along 132°E to form a closed box 
with the Australian south coast, with CTD/Rosette coverage from the surface 
to the bottom.



Narrative

RN Franklin left Victoria Quay, Fremantle, as planned at 6 a.m. on Saturday 
12 November 1994. A summer high pressure system was moving into the Bight 
from the west, so winds were moderate, reducing to nearly calm conditions 
over the next few days. The vessel therefore made good speed, and we arrived 
at the first station at 2 a.m. on Monday 14/11. A pod of humpback whales was 
spotted on the way.

Problems and goals not achieved:
Most of Monday was spent running out the wire on the drum and spool it back 
on under tension. A first attempt at a location with 3000 m depth had to be 
abandoned because the wire rubbed on the A-frame when trailed behind the 
ship. The wire was then run out in 4000 m of water, with an angle of 45 - 50° 
pointing away from the ship. When it was brought up it became evident that a 
substantial length at the end was badly damaged and had to be taken off the 
drum. This left about 5650 m of usable CTD wire, just enough to extend 
nearly all stations to the ocean floor as planned, missing the bottom by 
about 100 m at two stations.

A bottle test station on which all bottles were triggered in the salinity 
minimum at about 900 m depth to check for possible leaks was completed on 
late Monday. The first section (along 120°42'E) was started in good weather.

A weak front passed the region during Wednesday 16/11, followed by a large 
high pressure system. Progress with the section was therefore excellent over 
the next couple of days. The weather held out until Sunday 20/11, with winds 
of 10 - 20 knots from the southeast gradually turning southwesterly. By that 
time we had reached 47°S and passed the Subantarctic Front.

Other incidents of note:
The best aspect of the Southern Ocean is without doubt its wildlife. The ship 
is always accompanied by birds of all kind, from the size of a swallow to 
albatrosses nearly the size of a pelican. Seals pop up at CTD stations to 
check on our work, and on Sunday the ship was circled at close range by a pod 
of about 30 pilot whales. At the Subantarctic Front the water was full of 
long narrow creatures that looked like sea snakes but were fragile like 
salps. Patches of kelp were seen drifting by just to the north of the front. 
Squid, garfish and on one occasion a shark were also seen. The tropics are 
barren in comparison; during the voyage to Colombo earlier this year there 
were days when not a single bird could be seen.

Winds on Monday 2 1 /11 were westerly at 25 - 35 knots, causing the ship to 
roll heavily on its way to the last two stations of section 1. The section 
was completed without loss of a single station in the early hours of Tuesday 
22/11.

Section 2 was dominated by a low pressure system that remained nearly 
stationary for several days. To everyone's surprise winds were strong 
northerlies, making the ship's eastward course much more awkward than 
anticipated. One CTD station had to be cancelled during the passage of a 
front which saw the thermometer drop to below 2°C. This was followed by fair 
weather in the centre of the low, where the barometer dropped to below 972 
hPa. Station work proceeded well during this period, until increasing swell 
forced cancellation of a station despite moderate winds.

The pressure system eventually began moving east on Thursday 24/11, causing 
the wind to increase and turn to westerly. This lead to further cancellations.
The night of Friday 25/11 and all of Saturday 26/11 was spent waiting for the 
winds and the swell to abate. The decision was then taken to terminate section 2
at 129°30'E and begin section 3 along a course of 35°, to meet the original 
track near 45°S.

Section 3 was begun on Sunday 27/11 and proceeded well in fair weather. To 
make up for lost time stations in the central part (the South Australian 
Basin) were spaced out from 30 to 37.5 mile distance. Work on the section 
continued in constantly improving weather conditions. The last station was 
completed on the early hours of Sunday 4/12. The passage to Port Lincoln was 
made in brilliant sunshine and calm seas. The ship arrived in Port Lincoln on 
the morning of Monday 5/12 as planned.

69 CTD stations were completed. The willing and expert assistance of the 
ship's officers and crew and of the CSIRO-ORV personnel made this cruise a 
great success.

Cruise track
The final cruise track with all station positions is shown in Figure 1.


Results

As an example of the data set, Figure 2 shows the meridional section of 
potential temperature along 120°E (section 1). The Subtropical Front (STF) 
was crossed between 39°30'S and 40°S (stations 14 and 15). As usual, the 
front is better seen in the salinity, and its narrowness is better 
appreciated from the continuous record of sea surface temperature and 
salinity obtained from the thermosalinogaph (not shown).

The Subantarctic Front (SAAF) was crossed between 45°S and 46°S (stations 25 
- 27), though its surface expression was only reached towards the southern 
end of the section. Its position was evident from the geostrophic current, 
which showed a broad band of strong eastward currents between 45°S and 
47°30'S reaching to 1000 m depth and below, with maximum velocities reaching 
0.25 m s-1 at the surface associated with the front. In comparison, the 
geostrophic current associated with the Subtropical Front was weak, just 
exceeding 0.05 m s -1 and counteracted by westward flow on its southern side. 
Another region of deep reaching strong currents was seen close to the 
continental slope (stations 3 - 8). This is a region of intense eddy activity 
associated with current shear between the eastward flowing Leeuwin Current 
and the offshore circulation.

The permanent thermocline was dominated by a large volume of Subantarctic 
Mode Water with a temperature of 9 - 10°C. Water with these properties was 
found at the surface in a small region north of the Subantarctic Front.

Antarctic Intermediate Water was seen as a salinity minimum near the 1000 m 
level. The gradual erosion of the salinity minimum towards the north was 
interrupted by lenses of low salinity, suggesting that the AAIW circulation 
may not be uniform in space and time. This was supported by the patchy oxygen 
distribution at AAIW level.

Deep Water was seen near 2500 m in the south as a salinity maximum with 
highest salinity in the south, gradually decaying and sinking to 3000 m in 
the north. Its salinity and oxygen distribution both evolved slowly from 
south to north, so movement at Deep Water level is likely to be more uniform 
in space and time.

The presence of Bottom Water below 4500 m was indicated by potential 
temperatures below 0.5°C, about 0.2°C warmer than potential temperatures 
found in the AustralianAntarctic Discordance on section 2.

Figure 3 shows the zonal section of potential temperature along 48°S (section 
2). The SAAF was close to this latitude and appeared to oscillate between a 
more northward and more southward location. The crossings of the front 
dominated the section; contours were lifted upward as the front moved south 
and downward as it moved north. The frontal movement was seen in the 
geostrophic current, which showed current reversals through nearly the entire 
water column. Maximum surface velocities of .25 m s -1 northward and .35 m 
s -1 southward were in good agreement with the ship's acoustic Doppler 
current profiler (ADCP) which recorded currents up to 1.5 m s -1 to the 
north-east (60°) and about 1 m s-1 to the south-east (130°).

The GEBCO topography (Figure 4) identifies several fractures in the South 
Indian Ridge near 120°40'E and 128°E. None of these is shown as allowing 
passage of water below the 4000 m level. Section 2 is to the north of the 
sills and therefore shows depths greater than the sill depths, but it should 
capture all Bottom Water flowing across the sills. Figure 3 shows that no 
Bottom Water got through the major fracture near 120°40'E (station 32). The 
strongest indication for northward flow of Bottom Water was seen in a 
fracture near 123°40'E (station 37) where water colder than 0.4°C was found 
below 3800 m; near the bottom the potential temperature fell below OYC. Water 
with similar potential temperature was also seen to come from a fracture near 
127°E (station 39), but in an apparently smaller amount.

The Subantarctic Front was to the south of section 2 when section 3 was 
commenced, so section 3 did not show a crossing of the SAAF. The Subtropical 
Front was crossed between 37°S and 38°S (stations 61 and 62), although 
surface water with properties derived from the frontal zone was seen as far 
south as 43°S (station 52). The location of the STF as far north as 37°30'S 
is surprising, considering that it passes to the south of Tasmania and thus 
has to shift southward by some 600 kin over a zonal distance of only 1100 km 
and that over the 900 km between sections 1 and 2 it shifted northward by 
some 200 km.

Subantarctic Mode Water was seen with a substantial volume in the 8 - 9°C 
range, ie colder than in section 1. In the temperature range 9 - 10°C it had 
much more surface contact than in section 1.

The distributions of Deep Water and Bottom Water resembled those seen in 
section I closely. Bottom Water temperatures were again about 0.2°C higher 
than those found near the Discordance in section 2.

Personnel

              Ship's crew
              -------------------------------------
              Ian Sneddon       Master
              Richard Dougal    Mate
              Ian Menzies       Second Mate
              Maxwell Cameron   Chief Engineer
              Lindsay Cale      Second Engineer
              Donald Roberts    Electrical Engineer
              Jannick Hansen    Bosun
              Ronald Carr       AB
              Ron Kelleher      AB
              John McNabe       AB
              Phillip French    Greaser
              Reg Purcell       Chief Steward
              Gary Hall         Chief Cook
              Melvin Dall       Second Cook

              Scientific party
              --------------------------------------------
              Matthias Tomczak  FIAMS      Chief Scientist
              Colin Andrew      FIAMS
              Jarrad Exelby     FIAMS
              Michael Herzfeld  FIAMS
              Michael Schodlok  FIAMS
              Peter Strutton    FIAMS
              Neil White        CSIRO-ORV  Cruise Manager
              Erik Madsen       CSIRO-ORV
              Ron Plaschke      CSIRO-ORV
              Mark Rayner       CSIRO-ORV
              Dave Terhell      CSIRO-ORV



CTD Processing Notes Fr 10/94 
(Neil White)

This cruise is part of the Australian contribution to WOCE and consisted of 
three deep sections in the southern Indian Ocean.

The cruise was carried out with CTD underwater unit number 2. There are 69 
stations numbered 1-69.

The data is of reasonably good quality, but problems with salinometers and 
sample bottles have caused the salinity data to be less good than it ought to 
have been.

Of the three Yeokal salinometers on board (including one which had just returned 
from service) not one was properly functional. The most usable of the three 
salinometers had a broken thermistor, so sample temperatures had to be taken 
manually using a DSRT! While this apparently gave good results, there were a 
number of stations for which the salinities appeared to be anomalous - the 
salinities for a complete station would be different from the adjacent (and very 
nearly equivalent) stations. Uncalibrated CTD data for these groups of stations 
suggested that there were no changes in water properties, so data from stations 
8, 21, 35, 43, 61 and 65 were rejected altogether.

The sampling bottles used were designed and constructed by CSIRO. They were 
first trialled in mid-1993 and some of the major problems fixed, but there are 
still problems. One bottle in particular produced clearly anomalous readings for 
about 40% of its samples! All data for the two worst bottles were rejected, but 
it is clear from the CTD calibration process that there were still a large 
number of bad samples.

These two factors combined to produce a salinity calibration that is not as good 
as I would like.

There are a small gaps in a few stations. Some of these were introduced when 
salinity spikes causing density inversions in the 2 decibar averages were 
removed. There are four gaps of 2 decibars or so, however, which have no 
plausible explanation other than loss of data for a second or so when the data 
was logged. This is being investigated. These gaps occur in regions of low 
gradients, so should not cause any problems. The largest gap of 10 decibars 
occurs near the bottom of station 50 where some sediment or other matter fouled 
the conductivity cell.

Pressure calibration

Constants from the last laboratory calibration were used but a new offset term 
was calculated for each station from the pressure of the first Ôin waterÕ data 
records. These offsets range from 4.1 to 5.8 decibars.

Temperature calibration

Temperature calibration constants from the last laboratory calibration were 
used.

Conductivity calibration

Many groupings were tried and a large number of samples had to be thrown out 
because of the salinometer and sample bottle problems. The best calibration 
seems to come from using the groups: 1-9, 10-23, 24-46, 47-48, 49-62, 63-69.

The standard deviation of the salinity residuals is .0036 psu.



Notes on ADCP data for Fr 10/94

1 Features of this voyage

GPS "SA" degradation (see section 2) was in force during this voyage, and GPS 
coverage was almost 100%. A RACAL Differential GPS system was being used on a 
trial basis. It provided data approximately 60% of the time, and this data was 
much more accurate than non-differential GPS. It was often virtually free of 
"SA" effects. Obviously, where possible the Differential GPS has been used in 
the final product.

There was no discernable heading dependant gyro error. However, there was an 
apparent variation in the alignment error throughout the voyage which may have 
arisen from a cor-responding error in the gyro data. A time-varying correction 
has been applied, ranging from +0.2° to -0.6°.

Anomalies were detected at times which were characterised by a change in the 
current pro-file shape when the ship's velocity changed. These occurred in 
varying degrees and were associated with poor conditions, especially large 
following seas. Such events may have occurred before, but have never been 
detected. The only hypothesis for these events is that an acoustic interference 
is set up which biases the frequency shift toward lower values. As it appears to 
decay with depth, one would suppose it involves a passive use of the ping 
itself. The two most extreme cases were cut out of the dataset. Also, to reduce 
the likelihood of more subtle events corrupting the dataset, screening was 
tightened (especially RMS error velocity and percent-good).


1.1 Profiles integrated

    Bottom track corrected, no reference layer averaging in final integration:
    172 20 minute profiles (11% of voyage covered).
    GPS corrected
    RACAL differential GPS profiles are identified by a 'cnav' value of " R". 
     Ashtech non-differ-ential GPS used is denoted by 'cnav' values of "P" and 
     "D". See Data Format Guide for details.
    1529 20 minute profiles, of which 1039 were used solely Differential GPS 
     (95% of voyage covered). Use the non-differential with care, if at all, for 
     period when SA was active.
    510 60 minute profiles, of which 346 used solely Differential GPS (95% 
     coverage).


2   GPS data degraded by SA (Selective Availability)

The US Department of Defence, who operates the GPS satellites, has introduced 
deliberate complex errors into GPS data. It is generally considered that these 
errors cannot be removed without extra equipment and post processing (and even 
then cannot be achieved with deep ocean work.)

The characteristics of SA errors are probably changed from time to time, however 
they usu-ally seem to be across quite a wide time spectrum. Of most concern for 
ADCP data are the errors of order 50 cm/s over 5 to 10 minute periods. There 
also appears to be a smaller and lower frequency component, the worst case so 
far observed had a residual error of 6 cm/s after averaging an hour's data.

The implications for ADCP data are: 

    individual GPS corrected ensembles (3 minute or less) often have errors of 
     around .5 m/s. 
    The existence of such errors prohibits the use of some quality control 
     measures, espe-cially of course dv/dt. 
    20 minute integrated profiles will usually have little extra error, maybe 1    
     or 2 cm/s. How-ever, at times low frequency components of SA may cause   
     larger errors, up to 10 or 20 cm/s. 
    60 minute profiles will rarely have more than 1 or 2 cm/s extra error. 
    Incomplete 20 minute profiles (low 'icover' percentage) are less reliable 
     because they are probably incomplete due to a break in GPS coverage, and 
     data adjacent gaps is usually of poorer quality. Also, the SA errors are 
     less likely to have been removed by averaging. 
    Bottom track, Transit and shear data are, of course, unaffected by this. 
     When using GPS to get ship's position, these errors are negligible (200m or 
     300m at most).


3   Calculating the Bin Depth

    The depth to the centre of bin j, in metres, is approximately: 

    depth(j) = draught + (plen + blen)/2 + delay + blen*(j-1) + blen/10

where 
   draught - 4 m 
   blen - bin length 
   plen - pulse length 
   delay - delay after transmit (also known as DTFB - Depth To First Bin).

The depth bins are generated by the instrument using the assumption of a sound 
speed of 1475 m/s. The above approximation can therefore be refined by 
correcting for the approxi-mate real sound speed, that is, by multiplying the 
above-derived depth by (estimated_real_sound_speed)/1475. This sound speed 
estimate would be made by esti-mating the mean temperature, salinity and depth 
for the main study area.


4  Calibration

ADCP water profile vectors are calibrated by being rotated through an angle a 
and multi-plied by scaling factor 1+b. The rotational calibration primarily 
corrects for misalignment of the transducer with respect to the ship, of the 
ship with respect to the gyro compass, and the error in the gyro compass. The 
scaling multiplier primarily corrects biases arising from the profiler itself. 
Both of these calibrations make a large difference to the resultant cur-rents, 
particularly because they are both applied to the usually large ship-relative 
cur-rents. For example, a scaling multiplier of .01 applied when the water 
velocity with respect to the ship is 6 m/s alters the measured absolute currents 
by 6 cm/s. Calibration is particu-larly difficult when the coefficients change 
with time, as appears to be the case on this voyage.

Results for this voyage:
                              1 + beta ~= 1.0135

Overall mean a ~ -0.2. However, time varying a effectively applied as a gyro 
cor-rection, ranging from -0.6 to +0.2.


5   Data Quality

The data provided should not be taken as absolutely true and accurate. There are 
many sources of error, some of which are very hard to quantify. Often the 
largest error is that of determining the ship's actual velocity.

Accuracy of water velocity relative to the ship

The theoretical approximate short-term velocity error for our 150 kHz ADCP is:

         sigma = (pulse length X square root of pings per average) - 1

For a 3 minute ensemble with say 170 pings, using 8m pulse, this gives a 
theoretical error of 1 cm/s for each value (that is, independantly for each 
bin).

For 20 minute profiles, with say 1150 pings averaged, the error in measuring the 
velocity of the water relative to the ship is probably reduced to the long term 
systematic bias. Of this bias, RDI says

"Bias is typically of the order of 0.5 - 1.0 cm/s. This bias depends on a 
variety of factors including temperature, mean current speed, signal/noise 
ratio, beam geometry errors, etc. It is not yet possi-ble to measure ADCP bias 
and to calibrate or remove it in post-processing."

As well as that, there are the transducer alignment and gyro-compass errors, 
which proba-bly have a residual effect after calibrating of roughly:

0.4 cm/s per m/s of ship speed, due to say 0.4 uncertainty in alignment angle
0.3 cm/s per m/s of ship speed, due to say 0.003 uncertainty in scaling factor

This gives us say 0.5 cm/s error per m/s of ship speed, or 3 cm/s at 12 knots.

Other sources of bias might be the real-time and post-processing data screening, 
and depth-dependant bias.

GPS profiles

In the presence of SA, errors are larger and even very large errors cannot be 
removed by dv/dt screening (because this would bias the long term average - 
there is reason to assume that given a long enough period the SA error is close 
to zero).

Bottom track profiles

Firstly note that errors arising from transducer alignment and gyro limitations 
will sub-stantially cancel out. Normally, the accuracy of screened bottom track 
data appears to be of the same order of accuracy as non-SA GPS, that is, about 2 
-3cm/s for a 20 minute profile.



References : Franklin National Facility (1994). Research Summary, Cruise FR 
    10/94. Miscellaneous Publication. CSIRO Division of Oceanography [for] 
    Franklin National Facility, 8 pp. 



WHPO Data Processing Notes

Date      Contact      Data Type       Data Status Summary
--------  -----------  --------------  --------------------------------------
05/21/97  Lebedeva     CTD             Data Update  Unit Conversions Provided
          1. WOCE section name.
          
          These sections were added to the WOCE section network on the 
          initiative of M. Tomczak. They form a closed box from the 
          Australian south coast along 120E to 48S, then along 48S to 132E, 
          then along 132E to the Australian south coast. At the moment, 
          these sections have no WOCE name/number, but I submitted the data 
          files toWOCE as WOCE section I11 on the suggestion of Dr. Tomczak.
          
          2. Unit conversion.
          
          Nitrate, silicate, phosphate, nitrite and bottle oxygen data have 
          been converted from UMOL/L to UMOL/KG using the density calculated 
          from P=0, T=17 C (lab temperature, according to Dave Terhell 
          (CSIRO)) and bottle salinity. If salinity was missing for a 
          particular bottle, the interpolated salinity was used for density 
          calculations.
          
          CTD dissolved oxygen data have been converted from UMOL/L to 
          UMOL/KG using the density calculated from in-situ P,T and S.
          
          To convert the data expressed in micromole/liter units to 
          micromole/kilogram units I divided the former data by the density 
          as expressed in kilograms/liter (same as gm/cm3).  
          
          3. Additional parameters in the __.CTD files.
          
          SDTEMP - Standard deviation of good temperature samples in the 
                   interval
          SDCOND - Standard deviation of good conductivity samples in the 
                   interval
                    
08/25/97  Tomczak      CTD/BTL         Submitted for DQE
                    
06/01/99  Tomczak      CTD/BTL         Website Updated; Data Public
          Please remove the encryption from the data set obtained by 
          R/V Franklin during cruise FR10/94.
                    
02/29/00  Bartolacci   CTD/BTL/SUM     Website Updated; Expocode changed
          I have changed all occurrances of expodec 09FR10/94 in all a05 
          files (except doc) to 09FA1094 and updated all tables.
                    
06/21/01  Uribe        CTD/BTL         Website Updated with New Exchange File 
          Updated CTD and bottle exchange files were put online.
                    
08/21/01  Uribe        Line designation changed from I11 to S05
          Exchange File Updated:
          It was determined by Jerry and Danie that the sumfile should be 
          modified from i11 to s05 to fit format in bottle and ctd files. 
          Bottle was re-run through exchange, new file is online.
                    
12/26/01  Uribe        CTD             Exchange File Generated, OnLine
          CTD has been converted to exchange using the new code and put 
          online.
                    
01/03/02  Hajrasuliha  CTD             Internal "DQE" completed
          generated *check.txt file.  
                    
          
12/17/02  Kappa        DOC             Cruise Report Updated/OnLine
          Added:
            Cruise Summary Information
              Abstract
              Geographic Boundaries
              Cruise Narrative
              Scientific Program
              Cruise Objective
              Results
              Scientific Party and Ship Crew lists
            ADCP Information
              References
            WHPO Data Processing Notes
            PDF file with all the above plus:
              Cruise Track and various figures 
              Map of Station Locations
              Internal (PDF) links


