C.M.1993/C:S9
Hydrography COßuuittee
International Councll for the
Exploration of the Sea
Greenland Sea Deep Waters:
Areport on ·1993 winter and spring. ~ruises
by RVs POLARSTERN and VALDIVIA
B. Rudels,
1. Meincke, H. Friedrich and K. Schulze
Institut für Meereskunde der Universität Hamburg
Troplowitzstr. 7, D-22529 Hamburg, Gerrnany
Introdtlction
The outflow of deep water from the Canadian BaSiri, which mal' be identified in Fram Strait
(Rudels, 1986), was thought to bypass the Greenland Sea and either enter the Icelandic Sca
across 'ar thc Norwegian Sea along thc Jan Maycn Fracture Zone (Aagard ct al., 1991;
Rudels,' 1986).
A third decp water component 'supplied by th~ Arctic Occan is the lighter Upper P~lar Dcep
Water. It fills the Arctic Ocean between the Atlantic Layer and the sill depth '. of thc
1
,
, '
I'
I,
I,
I'
.
!
,.
',
,
I,
'
,
Lomonosov Ridge. It is forrned in both basins of the Arctic O~ean and it lies in the same
dcnsity range as the Arctic Intcnncdhitc Watcr forrned!inthc Grccnland Sea. In conträst to.
the latter, it is identified by vertlc:u grndients stable in both heat and salto
,
The distinct charaeteristlcs of the Arctic Ocean and the Greenland Sea dcep' watcrs make
them easily recognizable and we shall examine the inOuence and the importance of the Arctic
Ocean waters for ihe ventilation of the' deep Greenland, ,Sea. Thework is basrid on
obserVation in the Greenlarid Sea collected in late.winter and spring 1993 from the vessels
RV Polarstern (24/3 - 18/4) and RV Valdivia (15/5- 17/6). The station positions are shown
in figure 1. Tbc winter convcction had cCaScd arid advcctive proccsscs domimitcd the rriixing.
On the Polarstern cruisc a NBIS Mark llIb CTD was used while a SBE-11 CTD was
employed on the Valdivia cruise. Both instruments worked properly. The conduciivÜy was
calibrated by ,vater sampies analyzed on board using Guildiine Autosal. The accuracy in
' !
.
p.s.ti. is 0.003.
I .
a
Deep ,vnter renewnl in
{he GrcenJnncl Sen.
j
Recent efforts to study deep convection in the Greenla~d Sea (Meiricke, 1991) have shown
that the corivection operates iritermittently at rriost and tracer studies iridicate that no deep
convectiori has occurred in the Greenland Sea betwcenl 1982 and,1988 (Rhein, 1991); This
shiftS the balance in the ventilation of the Greerila.rid Sea towards astate, where the deep
water contrihütlons of the Arctic Ocean become more prominent. A comparison between 0-S
curves taken in, the central Greenland Sea from RV \Valdivia wiih those obtained from
RV Hudson in 1982 confirms this. Figure 2 displays ;0-S curves for two stations in the
central Greenland Sea laken on the Htidson and Valdivia cruises respeetively.
'
,
'
I
'
..
The most notabie difference is the ,varrrier und more' saline deep temperature
maximum
J " ; ' . ' .., , •
, '.
observed 10.1993. The sahmty, 1S In fact so h1gh that a second shallower sahmty max1mum
is prescnt at aboiii 1000 m. This riuciirrii1l11 is seen at most stations in the western and in the
southem Grcenlarid Sea although' Ü is absent at station i8, riearest to station 12 towards the
cast. Themaximum is tao dccp und too derise to be crcatcd by an inOux of AiIantic Watcr,
either from the Norwegian Atlantic Current in the east or from the Atlantic Return Current
in the west. Since. the maximum is strongest to the ,;,rest arid to the sOlith (compare the
salinity sections, givcn in figure 3 for the Valdivia cruise) it is probably the signal of
Cariadian Brisin beer Water, CBD\V, penelrating from 'the Greenland coritinental slope arid
from the, Jan Mayen Fracttire Zone to\vurds the up\veiling regime in the ceriter of the
cyclonic Gre~nland Sea gyre.
"
.
• _.
.,.
J
.'
. ' ,
l'
~'"
'.
l,
~
•
"
"
I
~,'
"
"
;
';
. '..
, .' ,
'",'
I'
,.'
~
. ,
" , . ,
'..
li
"
.
The'. " 0-S ctirves •from
stations
in .
the.south,'
Close io!
the fraeture
zone (flgure' ., 4 shows
•\
.,.
',
' 4,
•
,
,.,
•.
2 examples), exhIbIt a slope perpendicular to the isopycnals above the upper saliriity
maximum. This shows that also the upper prirt of the' deep boundarY current exiiing the
Arctie Ocean tllrough Fram Strait might detach from the' continental slope and peneti-ate into
the deep central basin albeit alorig the Jari Mayen Fracture Zone.
' 'I
I
, ,
•
,>
'
,:.
"
" ,
,
,,;
j
,',
,','
,
,
",I"
•
"
'"
!
,
In the wmter 1992/93, the convectlOn was limited to thc upper 1000 m (compare figure 5)
and only water in the range of Arctie Intermcdiate Water, AI\V, was forrried. The sti-ength of
the convection earl be assessed from the thickness of the associatcd cold, iow salinity lense.
In thc ccntral gyrc the lense is elose to 1000 rri thick but the ihick11ess diminishcs towards
the pcfiphery of lhe gyrc (cciinparc figure 3). The intera~tion beiwecn the cold lense arid lhe '
I
, I
, I
2
I
I
I
I
I,
periphery occurs through isopycrial interleaving and it is stfo~gest t~wards th~ west and
south. Especially prominent is thc intruding shallow salinity and temperature maximum of the
Atlantic Return Current~ which is present on most of the western stations (2 cxamplcs are
displayed in figure 6).
. '
.
Thus, thc shallower convcction occurring hi recent ye:irs riiust be responsiblc fOf the
differericcs between the observations from 1982 and from 1993. Norrnally the convecting
waier would be denser and the CBDW would be entrained into arid mixed downward by the
convectirig plumes. The upper salinity maximum would theri be rerriove~ and the temperature
maxirrium would be reduced, heat and saIt bcirig added to ihe bYpa.ssing convectirigwater.
The spreading of newly formed AI\V from thc center of the gyre' will rCs~lt in a
compensating inOow of CBDW; which - because of the shallow convection - is easily
deiected by its saIinity and temperature maximum. Inversions arc ~~en in thc profiles and in
8-S curves which indicate intrusive mixirig witholder Greenland Sea Deep Water, GSD\V.
The deeper penetration towards the center thus also appcars io be isopycriic.· .
The corivectivc arid the advective contrÜmtions to the deep waters of the Greenland Sea thus
alternate in strCrigth and importaricc. At times the.local conveciion is even stronger than in
1982 and capable to remove riot only ihe upper salinity maximum but to penetrate the deeper
maximum also a.nd renew the bottom water. Most of the time~ however, the ouillow from the
Arciic Occan domiriates and rapidly reforrrls the temperature and sa1inity struetu're of the'
deep waters.
.
This simple picture is~omp1icated by the observation thai the freon coritent of GSBDW
above 2000 m has increa.sed as compared to 1988 (Rhein, pers. comm.). This is at odds with
a perictration of low freon CBD\V futo the central Greenland Sea. Ii ca.n only be understood,
if ihe convection eventS down to 2200 m reported from ihe winter 1988/89 were more
extensive than previously believed (Meincke, 1991). That would provide the necessary freon
irijectiori and the high freon content of the convecting siirface water would more thalt
compensate for the inriux or' "olderi, CBDW. The possibility ihat the freori content has
increaSed by turbulent diffusiori from above appears less 1ikely, since this would' mean that
the Greenland Sen..
no corivcctiori was needed to ventilate the deeper layers
or
The most c~nspicuous effect of the reduced convection in the Greenland Sea is the overall
decre,ase iri density of almost the entire water column as compan~d to 1982. The ,vinter
1992/93 was exiremely stormy and the lower densities found 3t all levels in 1993 in
comparison to 1982 suggest thai ihe doming of isopycnals in the cenu-al Greenland Sen
(Helland-Hansen and Nanseri, 1909) might to a large exterii be causcd not by the wind but
by the convection itseIf. When the convection is reduced, will slump down, spread out of the
gyre lind induce a compensating inOow at higher levels. If this inOow occurs along
isopycnals it will~ because of the relaxation of the domirig, lead to a reduction iri the density
of thc central water column. This in turn weakens thc cyclonic circulation and the arctic deep
waters willless be pressed against the continerital slope and instead of bypassing rather eriter
further increase the temperature and sa1inity in a sense which
thc Greenland Sea. This
reduces the density of the water column in the central gyre.
will
3
I
•
The higher densities, fmirid also at
levels ,..lafter penods. .of, act1ve conveciion,
. intennediate
.." ,
suggestthat a eonvcetive build up of denser watcr is nceded in the cntire water colurnn
before the loeal convection can reach the bottom~ This could imply that thc obSerVed trend
of inereasing tcmperature and salinitY,and dcereasing density in the,bottorn layers, caused by
turbulent vertieal inixirig of iritruding Europeari Basin Deep Water EBDW with; the
Greenland Sea Bottom \Vater;does .,riot indicate anl im~irierit d,eep, c~ri.ve:tion event
penetrating through and renewing the bottorn water. A massIve convcct1ve SItuatIOn appears
neeessary gradually involvirig almost the entire water column. This requires a strong surfaee
forcing and appropdate eonditions of the upper wa'ter eolumn, whatever those right
preeonditions might bc..
,1
'
..
:.
,-,
,
-,
.
I
I
.
.
Acknowledgrnent: This ~ork was supported by Soncterforschungsbereich :318:
Klimarelevante Prozesse im System Ozeim..Atmosphfu-e-Kfyosphfu-e. We also appreciate the
.'
.readily aetive support by eaptains und crews of the p~ciiJatirig vessCIs.·
,I
Refererices: '
; .
.
..
I
;. ",'
I,
.' .
i .
!...
. .,
Aagard K~, J.B. Swift and E.C. Carrnack - 1985: Thermohaline eirculation in the Aretic
Mediterrancan Seas. J.Gcophys.Res.,90,4833-4846
Aagard K., E. Fahrbach, J. l\leiriCke rind J.H. Swift • 1991: Saline outDow from' the
Afctie Ocean:
ItS eoritribution to the .dcep waters of the'I Greenland, Norwegian and Ieeland
' . '
'
Seas. J.Geophys.Res.,96,20433-20441
! . ' ;
Heinze C., P. Schlösser, K.P. Kolterrilanri and J. l\tleincke • 1990: A tracer study of the
deep watcr renewal in the European pol~ seas. Deep Sea Res.,37,1425-1453
HelIand-Hansen B. and F. Nansen • 1909: Thc Norwegiari sea. ItS PhysicalOceanography~
based upon the norwegian researches 1900-1904. Kristiania det Mallingske Bogtrykkcri.
l\leincke J.• 1991: Variability of convective eonditions in the Greenland Sea. '
ICES,VariabilitySyrnposium Mariehamm, paper 2
i
'. '
Rhein 1\-1. - 1991: Ventilation 'rates of the Greenläridl rind Norwegian scäs' dedvcd from
distributions of the chloroDuoroinethanes Fl1 und F12. Deep Seii Res.,38,485-503
'
Rudels. B. ~ 1986: On thc 8-S structure in the Northem SenS: Implieations for the decp
circulation. PolarRes.,4,133-159
..
. ,' .
I
Srnethie 'V.I\I., II.G. ÖsUund and II.II. Loosli - 1986: Ventilation of the dcep Greenland
and Norwegian Seas: evidcnce from krypton-85, tritium',carbon-14 and argon-39.
,
"
j, ' .
,
.
' .. '
Deep Sea Res.,33,675-7.o3
Smethie 'V.M., D.\V•. Chipman, J.II. Swift. arid K.P. KoJtermann - 1988:
Chloronuoromethanes in the Ärctic Mediterranean SenS: Evidence for formation
bottom
water in the Eurasiari. Basin and deep-water exchange through
Fram
Strait.
4
•
Deep Sea Res.,35,347-369
'l
.
I
.
.
"
.-'
,
. '
~
I ,
'. .
'..
,"
.
p
,
oe
I
I.
I
I'
4
.
r - - - - - - - - - - - - - - - - - - - - - - - - -...--- ...
Polarstern ARK 9/1
-10'
O'
23.3 - 17.4. 1993
35
t
3~Cl~I--'--'
~5
3.7
6~ 6'l,.el
e~
6.7
.3B
-10'
O'
b)
VALDrvrA 136
18.5.- 8.6.1993
O'
Fig. 1: Slalion posilions. a) RV POLARSTERN, b) RV VALDIVIA
.. --
-_._.~_.
-_._---- --
.-- ...-
. _ ..
._._w.·~
·~·
__ ·_·_·
HUDSON 5TA. NO. 58 CTD CAST 1
n)
2
~
0
-1
34.70
34.80
34.90
S,\UNITY (0/00)
35.00
i
: 3.I1ßßO
- .....
b)
-z.ßmm - I
!
I
I
J4.7R80
- - - - - - - ..__._-----._.._.. - -
Fig. 2: G-S dingrams. a) RV HUDSON, station 58 (75N;OGW;March 19R2)
h) RV VALDIVJA, station 12 (74.5N;OGW;May 1993)
--
---
.......
---------------------
Fig. 3: RV VALDIVIA cruise 136, salinity sections. a) zonal scction at 75N
b) meridional scction, central Greenland Sea
•
._---------------------------'
I
/
...
"
o'""
3
"l:l
o
"l
~
~
1=
"l
o
..,
P-
o
(l:l
...
(")
,/;
./.
;
.
27 '
I 10::;;1
.,;./
/'
: ,/~;
b) :
I
i
:
;
98
I I I ·1
/
I
I I ( I
. ·~:;.;;4--1--/f,L:- j-/i
P
!
/~
:/.
./.,
A
27 '
96
I I I~
jZ(}'
;
e
B
:7/
-··~'·····i················-·-l~1-······-j···-~~-······.' - / ZR. IR
i
//:
!
/<1,/!
; /"
- :;A~·········-~~t······V·-·-!--·/(-·
/;/·~l,/·J-·/i /1
~ J// i
~:~<:: . L.
L.L
L.
L.J
L . l...
~
,/
i"
.
:
/
/1 i
.
:
I
Fig. 4: E>-S diagrams, RV VALDIVIA cruise 136. a) station 16 (72.5N;08W;May 1993)
b) station 20 (72.25N;14.3W;May 1993)
•
.............. :·
····
:
...... .. ~
·····..·····..····r··.
:.
.....
. ... .
.
.!
.;
·
.
·.. ···1··. ···..·····.. ···r.
~
_
.
:
u •••••••••.••
.
.
............................. ············..····;······..········1·····..··..•··..·;·····
~
.
!
.
···•..•..···..··•·..;···..··········· ··············1·····
~
s~linity, rSS~7ß
~
:
rrSUJ
.
.
:
3~.B~e{j
Fig. 5: T,S-profiles, RV VALDIVIA cruise 136, station 8 (74N;04.3W;May 1993)
..
-----------_._-_._-----------,
a)
·t·.. · · . · · . ·.. t········ .-
..:.
.
..
:
.
';,.
T
.
• ••••.••••• ~
j
;-....
\"
. •. ~.,
'[
.
~ .•••• ,_ ••••••.•• -l •.••••••.
!
i
..
Si
T
I
I
!
J
----------------------_._-----:Y
b)
~
(
~-~
:.. J
_
. .r
.....
..•..
...•.1'
. •.•... ...~
................................................................,
;.
-,
~.
;.
.
- .
.
Fig. 6: T,S-profiles, RV VALDIVIA cruise 136. a) station 12 (74.5N;06W;May 1993)
b) station 13 (74N;06.5W;May 1993)
•