KNORR 147 LEG V CRUISE SUMMARY: LABRADOR SEA CONVECTION EXPERIMENT

Robert S. Pickart
Woods Hole Oceanographic Institution

Peter Guest
Naval Postgraduate School

Fred Dobson
Bedford Institute of Oceanography

Karl Bumke
Institut fur Meereskunde, Kiel

The initial field phase of the ONR-sponsored Labrador Sea Deep Convection
Experiment was successfully carried out on R/V KNORR from 2
February-20 March, 1997. The science party consisted of 19
participants from 7 institutions, comprising numerous collaborations.
This report provides an overview of the different components of the
cruise, and is divided into four sections corresponding to the four
major programs on the ship.


Despite initial pessimism regarding the ability of a research vessel
to operate in wintertime conditions in the Labrador Sea, and despite
predictions of the collapse of deep convection in the region, the
cruise was a remarkable success. The KNORR and her crew proved
resilient to the harsh conditions encountered, and throughout our 34
working days in the region we were only hove-to about 32 hours.
1 To give an idea of the difficult working conditions, consider
the following statistics for the time period over which we occupied
stations, 7 Feb--12 Mar. The mean air temperature was xxC (on only one
day--our last working day--did it rise above 0C, and during our
closest approach to the ice-edge it was -17C). The average wind speed
was xx knots (twice during the cruise we recovered the CTD package in
>50 knot winds). We had only two genuinely sunny days during the five
week period, and it snowed constantly, often leading to near white-out
conditions (which was one of the reasons for the reduced steaming speed). 
One favorable factor during the cruise was that, despite the frequent
passage of storms, the associated swell tended to dampen remarkably
fast; this was probably the biggest reason for the limited time
spent hove-to.


One of the main concerns operationally during the cruise was the
tendency for icing to occur under high sea-spray conditions. The crew
regularly pounded ice off of the decks and bulwarks (during one such
session it was estimated that 20 tons of ice was knocked off the
ship). It was a constant challenge to keep the CTD staging area free of
ice and snow, and freezing of the blocks and air-tugger lines was an
on-going problem. We also experienced difficulty with the float and
buoy deployments because of the high sea-state and slippery
conditions on deck. Despite these daily challenges, however, the crew
worked diligently to keep the science program operational in some
capacity, nearly full-time. Regarding the coordination of the science,
because of the extent and diversity of the different measurements 
it was a constant challenge to keep the individual components
functioning in collaborative fashion. Throughout the
cruise the communication between the various groups was excellent,
which was a strong reason for the overall success of the experiment. 
Our daily science meetings were invaluable towards the
coordination of activities and overall planning.  


It turns out that despite a mild December and early January, the latter
part of the winter of 1996-1997 was quite robust. In fact, the
atmospheric forcing during this period was strong enough to overcome
the mild start, and erode through the particularly fresh surface layer
of the Labrador Sea giving rise to convection down to 1500m.  Thus,
not only did we observe deep convection, but we did so under ``classic"
wintertime conditions in the Labrador Sea.  We owe this success to the
very capable KNORR and her captain, and to the tremendous efforts
of the crew and science party. Both the atmospheric and oceanic data
sets collected are the first of their kind in this region, and
will undoubtedly lead to improvements in our understanding of
convection in the Labrador Sea.

Hydrography and Floats

Robert S. Pickart
Woods Hole Oceanographic Institution
Woods Hole, MA  02543

The hydrographic station map 

(Figure 1) shows the end-result of our dual
strategy to obtain basin-wide coverage, while at the same reacting to
what we saw. 

All stations were occupied to the bottom, and roughly half of
them included sampling of CFCs (Figure 1b).2 Both the along-basin
section and the southern cross-basin section were repeats of the fall
HUDSON cruise, the latter being the first wintertime occupation of the
WOCE AR7W line. Note that there are a total of 5
boundary crossings. On the western side we were limited by the
proximity of the ice-pack, hence these sections do not extend
onto the shelf. As the ice-edge was approached the ship would encounter
bands of ice--including ``bergy bits" and ``growlers"--making it
impossible to proceed further onshore. On our last crossing the captain
made a special effort to maneuver the ship as far onshore as possible, and as
a result we reached the 700m isobath (within the core of the
Labrador Current). On the eastern side, particularly on the second
crossing, our biggest concern was icebergs. Despite having to divert our
cruise track a couple of times we occupied two highly resolved
sections across the West Greenland Current and Irminger Water. 

During the course of the cruise we dropped over 140 XBTs (Figure2).
The purpose of the XBT program was two-fold.  Firstly, in the interior
Labrador Sea XBTs were regularly dropped between stations in order to
increase spatial resolution and also alert us of any deep
mixed-layers.  Secondly, near the boundaries (on 3 of the 5 crossings)
a high-resolution XBT section was done prior to the CTD
work. The reason for this was to determine the boundary thermal
structure so as to optimally place the CTD stations (thus avoiding
aliasing). Both aspects of the XBT program
proved crucial. Throughout the interior, the XBT profiles unambiguously
provided the mixed-layer depth in near-real time (while steaming). This
helped shape our CTD strategy, and directly led to our observing 
deep convection. On the boundaries (particularly the eastern boundary
which is remarkably steep) the XBT information saved us valuable time
and resources with regard to the CTD effort, and provided important
small scale information to compliment the hydrography (Figure3).
It is worth noting that the cold air temperature apparently affected
the performance of the XBTs. The failure rate was high for 
both the T-7 (800m) and T-5 (1800m) probes when conditions were extreme
(say colder than 10 degreesC). After some tests we surmised that this
was most likely due to the coating on the wire becoming damaged due to
the cold. Interestingly, most of our T-7's were
on the order of 10 years old, and their failure rate was higher 
than for the (brand new) T-5's. 

While our original CTD station plan was to occupy a third complete
cross-basin section, we decided instead that it would be more fruitful
to re-occupy section 2 (see Figure1a). The reasoning was as follows.
First of all, it was evident that the atmospheric forcing we were
experiencing was more robust than anticipated beforehand. Thus,
sampling the center of the gyre late in the cruise increased our
chances of actually witnessing deep convection. Secondly, the rate of
mixed-layer deepening observed throughout the cruise was surprisingly
rapid (and in disagreement with simple 1-D mixed-layer model
predictions done onboard). It was therefore felt that a re-occupation
to see the temporal evolution (while documenting the atmospheric
forcing) would be enlightening. Our decision was a success on both
counts (Figure4): not only did we observe the deepest convection of
the experiment, but the two occupations, separated by roughly 10days,
were strikingly different.  Note, for instance, the remarkably short
spatial scales during the re-occupation. We believe that active
convection was taking place during this time period (or perhaps shortly
before), and often the up-cast CTD profile would differ significantly
from the downcast! (For example, Figure4 is contoured with the up-cast
profile of station 117; using instead the downcast profile removes the
two large warm ``intrusions" near 700m and 1200m.) A careful analysis
will be necessary to understand the observed evolution and the short
scales involved. It should be remembered that we have a complete
lowered acoustic doppler data set, as well as underway correlation
SONAR measurements, both of which provide direct velocity information.

During the cruise we performed two detailed ``to-yo" CTD surveys
(Figure1a). The first to-yo (the northern of the two) 
was early in the cruise. We noticed an abnormally sharp property jump
below the mixed layer in each of the CTD variables (confirmed by an XBT
profile), and decided to map out its lateral variation. Presently we
are still unsure as to the significance of this feature 
and its origin. The second to-yo was
carried out near the end of the cruise (during the re-occupation of
section 2). We used XBTs beforehand to ensure that at least part of this to-yo
would sample deep convection. The to-yo took 36 hours to complete and
consisted of five lines roughly comprising a 12km box (Figure5). Each line
contained 3 cycles extending from the surface to 2000m. During the
to-yo we observed the deepest mixed-layer of the experiment (> 1500m) and  
apparently sampled the collapse of a convective feature via lateral
intrusion/entrainment (see Figure6 for a vertical and lateral view). It
is envisioned that such high resolution information will help interpret
some of the moored and Lagrangian data from the other components
of the experiment. 

Finally, the cruise served as a platform for the deployment of various
drifters and floats (Table1). No less than 7 types of Lagrangian
instruments were deployed during the course of the 34 days. Upon
reaching the central portion of the gyre, near the beginning of the
cruise, a detailed array of RAFOS, VCM-PALACE, and Deep Lagrangian
Floats were set in the midst of a detailed XBT/CTD survey (Figure6).
Throughout the rest of the cruise we deployed VCM-PALACEs (northwest
portion of the domain), NSF-PALACE floats (eastern boundary), WOTAN and
BAROMETER drifters (interior Labrador Sea), 
IFM-PALACE floats (western
boundary), as well as several more RAFOS floats at selected locations. All
floats (and the majority of the drifters) were deployed at CTD
stations. It should be noted that conditions were too rough to deploy
the floats in the standard fashion off the fantail. Instead a
procedure was developed using the starboard hydro-boom in conjunction
with lines and a pivot hook. This proved quite effective and enabled
several float deployments in extreme winds and high sea-state.

-----------------------------------------------------------------------

1
It should be noted that, excluding these periods, the average
steaming speed of the ship was just below 8 knots; this is >3 knots
slower than under normal conditions, which did of course have an
impact.

2
Selected stations also included Tritium/Helium, oxygen isotope, and carbon dioxide 
measurements, supported by independent funding.
