--- - - - - - - - - - - - - - - - - 1
Not to be cited without prior reference to the authors
;;;.
~
1996 ICES Annual Science Conference
Reykjavik, Iceland
C M 1996/0:8
Theme Session on the North Atlantic
Components of Global Programs
ATLANTIC WATER TRANSPORT FROM LONG TERM CURRENT MEASUREMENTS
IN THE SVIN0Y SECTION
by
•
••
Kjell Arild Orvik and Martin Mork,
Geophysieal Institute, University ofBergen, Allegaten 70, N-5007 Bergen, Norway
ABSTRACT
The Svin~y sectionjust to the north ofthe Faroe-Shetland Channel cuts through the core ofthe
Atlantic inflow to the Norwegian Sea. Recent investigations show that the different branches of
the Atlantic inflow merges through confluence in the shelfbreak area off Svin~y, making this site
very suitable for monitoring the Atlantic inflow. This study deals with transport investigations of
the inflow based on long-term current measurements from 9 eurrent meters on 3 mooring lines
during April 1995 to August 1996. To achieve a significant estimate of the transport during
typical winter eonditions, an extensive program was performed in March-April 1996, with 24
eurrent meters on 5 mooring lines. A similar extensive program is in progress for the period
September-Oetober 1996, for typical summer conditions. Our fmdings show that the Atlantic
inflow oecurs as a 30 km wide, nearly barotropic current trapped over the steepest slope between
depths of2oo m and 900 m. Current records in the core ofthe inflow show a very stable alongisobatic flow towards north-east with yearly mean of 40 ern/s, standard deviation of 18 ern/s and
maximum of 100 crn/s. Transport estimates based on a linear regression model show an inflow
with strang variabilities between 12 Sv and 0 Sv and a fluctuating time scale ranging from a few
days to two months. The annual mean transport is estimated to 5.3 Sv with a standard deviation
of 1.9 Sv. The two months moving averaged transport estimates have a standard deviation 0.6
Sv, indieating small variabilities. This figure eorresponds with the time series which show small
seasonal variations in contrast to the common accepted annual cycle with winter maximum and
summer minimum.
.
.,I
1
1. Introduction.
•
....
;; ..f'
,During the last decades, mucn attention has been drawn to the inflow of Atl:intic water .:
to the Norwegiari Sea, an issue still under debate. This is beeause variabilities of this inflow of
\varm and saline water are assumed to be an important faetor fOf elimate, eeology and biological
production in Northern areas. To monitor this Atlantie inflow, a eurrerit measuring program was
initiated for the Svin~y section (Fig 1) in April 1995 for a 3 years period, as apart ofthe Mare
Cognitum program. The SvinPy seetion runs north-westwards from the NOI'\vegian eoast at 62
deg north and euts through the eore ofthe Atlantie inflow just to the north ofthe Faroe-Shetland
ChanneI. The objeetive is to survey this inflow by estimriting the volume transport on the base of
long term eurrent measurements.
,
Hydrographie observations in the Faroe-Shetland Channel und the Svinpy seetiori show
water properties of the Atlantic inflow as a warm and saline, wedge-shaped, baroelinie eurrent.
Based on hydrographie observations and sparse eurrent ri1easurements volume transport estimates
of the inflow have then been worked out by many scientists.,The most relevant estimates are
summarized by Gould (1985) and Hopkins (1991) and ranges from 2.0-8.0 Sv (Sverdrup).
Generally, there is little informatiori on the seasonal variability of the transport, but it has been a
eommon .eomprehension of seasonality with maximum transport in winter and minimum in
summer. Monthly rriean ealeulations by Blindheim (1993) for instance, gave u\vinter maximum
of7.9 Sv in January arid a summer rillnimum of2.9 in September, with a annual mean in 1990 of
5.5 Sv.
\Vorthirigton (1970) estimrited volume transports based on mass balance eonsiderations
and ended up with a total Atlantie iriflow of 9 Sv. The most extensive transport estirriates based
on year long direct current measurements on the continental slope northwest of ShetIarid were
made by Gould (1985). Based on 4 moorings sited between 400 in arid 1100 m, thc caIculations
show annual a\'erage of7.5 Sv. He also indieated seasonality with maximum in winter. Reeent
investigations have used satellite aItirrietry to obtnin estimates ofthe inflow. Pistek and Johnson
(1992) and Samuel et
(1994 ), worked out mcail transports of 2.9 Sv/2.7 Sv, and they both
showastrang annual cycle with summer to wiriter variations of about 100 pereent.
Recent investigations by Poulain et al. (1996) show that the major part of the Atiantic
inflO\V over the Seotland-Greenland ridge is trapped by the bottom topography at ctepths of the
order of sill-depth i.e. 500-700 m . Thc inflow over thc Faroe-Island ridge for example, turns eastwards and follows aIong the isobaths over the steep slope to the north and ,cast of Faroe Islands,
theri turris riorthward after a eyclonie turn in the Faroe-Shetland Channel and merges with the
Faroe-Shetland inflow through eonfluence in the Svin~y area. Just to the north of the Svinoy
seetion, the inflow splits into a minOf coastal und a major shelf edge branch flowing northward
aIong the Norwegian shelf edge as the Norwegirin Current, before spreading into the Barents Sea,
the Arctie Ocean and Norwegian Sea. The drift-experiments also show ihat thc slope area in the
Svinoy seetion catches the major part of outflow from the North sea. Drifters deployed in the
Norwegian trench indicate a flow pattern trapped by bottoin topography along the 200-300 in
isobaths. Because of the shallow More Plateau with depths Iess than 200 m, the mriin flow is
forced offshore towards riorthwest arid presumably passes through the slope area in the Svinoy
seetion. There seems to be small northward transports over the More Plateau, makirig this area
suitable as spawriing area. The mergmg through eonfluence of the different branches o~ the
Atlantie inflow in the Svin~y slope area, make this site a key position for monitoring the inflow.
Transport estimates presented in this study are bri.sed on direct eurrent measurements from
9 eurrent meters on 3 mooring lines (S 1, S2 and S3) in the Svmoy sectiori from April 1995 to
September 1996, sited in thc steep eontinental slope area bctwecn depths of 480 mund 1000 m.
an
aI.
...
2
The monitoring program was extended with 2 moorings (EI and E2) in between S2 and S3 for ":
a one month in March-April 1996, for typical winter conditions. Fig 3 shows the 5 mooring sites
with the 24 current meters between depths of 480-1000. A similar extensive program is in
progress for September-Gctober 1996, for typical summer conditions. Service operations on 6
months intervals with RIV "Häkon Mosby" have been combined with ship board ADC?
observations and standard CTD transects.
Based on 24 current records, the monthly volume transport will be estim1ted. And in order
to reveal applicability of the long term measurements in transport investigations, euch current
meter will be correlated towards estimated transport. Based on the best suitable current record,
a linear regression model will then be worked out to obtain relation between volume transport and
long-term current measurements.
2. Measurements und data processing
The measurements were made using Aanderaa RCM-4n current meters on subsurface
mooring lines. For the long term measurements, in each position, the instruments were deployed
at standard depths of 100 m, 300 m and 5 m or 20 m above sea bed. Pressure sensors on
uppermost meters measured vertical movements of the moorings. During the extensive period
each mooring were equiped with extra current meters at 500 m, 700 m and 5 m and 20 m above
sea bed. The data were hourly sampled and afterwards fIltered by 25 hours and 168 hours (1
week) moving average fIlters. Additionally the long-term records have been fIltered by using
moving averages over 2 weeks, 4 weeks(1 month) and 8 weeks (2 inonths) period. For the
extensive period, monthly mean velocities have been calculated for each of the 24 records und
presented as contour plot for the velocity perpendicular to the section in Fig 4.
CTD observations are presented as standard sections for temperature, salinity and sigma-t
(Fig 2). Bottom tracked ADep records with extension down to bottom depth of 600-700 m are
processed for stick-plot und contour-plot presentation. They are used as basis for extrapolations
of the lateral current profile between the shallowest mooring SI at 480 m depth and the shelf edge
ut 200 m, but will not be presented here.
3. Transport calculations und correlution analysis.
The 24 current meter records during March-April 1996 are used to calculate the volume
transport across the Svin~y section by assigning each current meter to a cross section area
surrounding it (Fig 3 ). The transports are presented cis 168 hours (one week) moving average
time series in Fig 5. For control, the monthly mean transport is also calculated based on the
contoured velocity field in Fig. 4. In order to apply our long term current measurements in
transport investigation, a linear regression model is worked out to obtain a relation between
transport as dependent variable and current observations from one current meter as independent
variable. To deduct the optimal current meter position for monitoring the inflow, euch of the 24
data series are correlated towards calculated transport. The highest correlation coefficients were
figured out for the current meters at 300 m depth on mooring SI at 500 m depth und mooring S2
at 700 m depth, respectively. The figures were 0.66 and 0.87 for 25 hours moving average data;
0.88 und 0.92 for 168 hours moving average data. Based on the fact that we have a complete time
series over 17 months for mooring SI and lack in data recovery for mooring S2, the current
meter SI at 300 m depth has bCen selected for long term transport estim1tes. Also transport
estimates based on current meter S2 at 300 m will be presented for comparison.
•
3
•
•
4. Results
;:...
The hYdrographici observations aIorig the Svinoy section in March 1996 ( Fig 2), exhibits. ~
the warm arid saline Atlantic inflow as a wide wedge-shaped baroclinic current extending -,
westward from the 200 m deep shelf edge. Maximum salinity is observed over the slope at 1000
m depth and may be interPreted as thc core of the inflow. However, the inonthly mean eurrent
field in Fig 4 reveals the iriflow as a 30 km wide current ,with nearly barotropic vertical structure.
Tbe narrow eurrent oceurs as a topographically trapped current over the steepest slope between
depths of 200 m and 900 m. The topographie trapping is also documented in ~he vertical
integrated transport in Fig 4, exhibiting a maximum transport arid thus the eore of the inflow at
about 600 m depth. There seems to be little coincidence between the baroclinic field und the flow
field arid it is a paradox that the area of maximum salinity, the so-ealled core of the inflow, is
located in the area of insignifieant currents. Most likely the saliri..ity rriaximum is a manifestation
of the distribution of mixing intensity.
For the moorings SI and S2, the time series for each current meter record show a very
stable a1orig-iSobath flow towards riorth-east through the water eolumn., with monthly means of
about 40 ern/s, staridard deviation of 18 cInls arid inaximum ()f 100 ern/s. In the site of mooring
EI at 900 m depth, the flow is moderate and variable with insignificant monthly mean currents
in accordance with Fig 4. Further westward in area of moorings E2 and SI, the records show
variable currents with a moderate mean flow towards west. The monthly mean transport based
on Fig 5 turns out as 4.9 Sv with a staridard deviation of 1.1 Sv. The calculated transport bnsed
.on the monthly mean contour field in Fig 4 is 4.6 Sv, i.e in good aecordance.
For long tenn transport investigations, tiIre series from the linear regression model for the
penod 22 April 1995 to 31 August 1996 is shown in Fig 6. The estimates are presented as hour
menn values, 25 hours moving average values and i week, 2 weeks, 4 weeks (1 morith) and 8
weeks (2 months) moving average values. Based on the 25 hours moving average transport
estirnates, the long tenn time series reveals an inflow with strong variabilities, with a fluctuation
time scale ranging from a few days to months. Maximum estimated transport is 12 Sv and
. minimum 0 Sv. The annual mean transport over one }'ear cycle is 5.3 Sv with a standard deviation
of 1.9 Sv. Comparlng the 2 wecks arid 4 weeks moving average time series, they show significant
periodicity of orie and two months, where thc tr:irisport ranges between 3 Sv and 8 Sv. However,
eonsidering the 2 months rrioVirig average time series, the transport show no systematie seasonal
signal. To some extent the time series indicate a winter maximum in December-January of about
6 Sv and a minimum in April-May of 4 Sv, ci period lci10wn for stable northerly winds in western
NorwaY. The transport for the period June-November is very stable about 5 Sv and disrupt the
common accepted comprehension of an annual c}'C1e with summer to winter variations of 50
percent ofthe means; inferred from hydrography (Bliridheirn, 1993), satellite altimetry (Pistee and
Johrison, 1992) and direct measurements (Gould; 1985). Thc feature of the estimated inflow
showing insignificant seasonality corresponds weIl to DickSon et al.,( 1990). His successful
eurrent records of the overflow from Nordic Seas show fluetuations of a few days period with
little seasonal and inter-annua1 variations. He estimated the arlnual-mean overflow to 5.6 Sv. Thc
transport estirnates based on current meter S2 at 300 m in Fig 7 show mean tr:irisports of the same
order (abOut 5 Sv) as Fig 6, for the period November 1995-June 1996. The standard deviations
indicate less variabilities in the transport. Concerning the 2 months moving average time series
,the standard deviation is omy 0.2 Sv. This very stable pattern of about 5 Sv , confums our
fmdings of small seasonal variabilities of the Atlantic inflow.
4
5. Concluding remarlcs
...
The rnerging ofthe different branches ofthe Atlantic inflow through a eonfluenee proeess
in the Svinpy slope area makes this site very suitable for monitoring the Atlantie inflow to the
Norwegian Sea. It is a paradox that while hydrographie observations in the SvinPy section reveal
the Atlantic inf]ow as a about 300 km wide and wedge-shape baroclinic current while our current
measurements show that the inflow occurs as a 30 km wide current with nearly barotropic
structure trapped over the steepest slope between 200 m and 900 m depth. It is also a paradox
that transport estimates based on radically different schemes, achieve more or less comparable
values, for example Blindheims (1993) geostrophic calculations and our fmdings which both turn
out with a annual mean just above 5 Sv.
Our transport estimates based on a monthly extensive program with 24 current meters and
long tenn estimates by a linear regression model, reveal that the inflow show strong variabilities
with fluctuation time scale ranging from a few days to 2 months, with maximum of 12 Sv and
minimum elose to 0 Sv. However the 2 month moving average mtered series show very small
seasonal variabilities eompared with common comprehension of strong seasonality of the inflow.
•
This inflow pattern is also confirmed by our estimates based on current meter S2 (300 m), which
shows a stable transport of about 5 Sv over the period November-April.
Acknowledgment
The Svinpy program is run as apart ofthe joint Norwegian research program Mare Cognitum and
is funded by the Norwegian Research Council. 0ystein Skagseth is now engaged in the program
and we thank him for excellent computer work. Also thanks to the technical staff at our institute
and the crew on board RIV"Häkon Mosby". Their positiveness and engagement have been an
important factor for our suceessful current measurements.
6. References
Blindheim, J., 1993. Seasonal variations in the Atlantie Inflow to the Nordic Seas. lees Stat.
Meeting 1993, C. M. 1993/C:39
Dickson, R. R., E. M. Gmitrowicz and A. J. \Vatson, 1990. Deep-\Vater renewal in the northern
North Atlantic. Nature, Vo1344, 26 April 1990
Gould, \V. J., J. Loynes and 1. Backhaus, 1985. Seasonality in slope current transport N. \V. of
Shetland. ICES Stat. Meeting 1985, C. M. 1985/C:7
Hopkins, T. SS, 1991. The GIN sea - A synthesis of its physical oceanography and literature
review 1972-1985. Earth- Science review, 30 (3-4):175-318.
Pistek, P. and D. R. Johnson, 1992. Transport ofthe Norwegian Atlantic Current as determined
from satellite altemetry. Geophys. Res. Letter, 10 (13): 1379-1382.
Poulain, P.-M., A. \Varn-Varnas and P. P. Niiler, 1996. Near-surface circulation of the Nordic
seas as measured by Lagrangian drifters. Journal of Geophysical Research, Vol. 101, No. C8, pp
",
18237-18258, August 15, 1996
Samuel, P., J. A. Johannessen and O. M. Johannessen, 1994. A Study on the Inflow of Atlantic
••.
·'
5
Samtiel, P., J. A. Johannessen and O. M. Joharinessen, 1994. AStudy on the Inflow of Atlaniie ..:
I
\Vater to the GIN Sea using GEOSAT Altimeter Data. Nansen Centennial Volume, 1994.
..
\Vorthington, L. V., 1970. The Norwegian Sea as a mediterranean basin. Deep Sea Res., 17:7784, 1970
•
.~,
Fig 1. Topographie ehart ofthe investigation area
The Svinpy seetion, mooring site and Mdre Plateau are iridicated on the figure.
.
2
1 St.
3
5
4
6
7
8 9 1011
12
13
14
o
Depth
M
500
TEMPERATURE
1000
0 km
50
2
1 St.
150
100
4
3
6
5
200
7
8 91011
12
13
14
0
•
Depth
M
------r
!SolO
!S.Gt
34.87
34.97
500 .
~
5
--F"
- .. ....
~
1000
Okm
SO
200
150
100
Fig 2. Hydrographie transect. the Svin~y section. March 1996.
S3
x=IZ.6
-100
-200
-300
E2
S2
SI
, x=l~.4
I
I z=200
1 x=l!.
I
I z=200
EI
I x=19.~
1
I x=13.8*
1
z=2oo
•
1
I
I
- - - -,- - - - - ..J - - - - - - - - - - - - - I
I x=13.8
I x=13.4 I x=18.2
I
,
*
I' *
I
*,
I
*
*
I
I z=200
I z=200
I z=190
~oo ----r----------r---~--I x=13.8
I x=13.4 l
.§. -500 .
*
I
*
I
*
*-5
I
I z=200
1 z=192.5 ~
g- -600 - - - -1- - - - - T - - - - -t - - -~~.iii
Cl
-700
-800
-900
-1000
-110~0
I
x=13.8
*
I
*
I
I z=190
I
1'=1'3:'& - - - - - T - - - - - -F160 .,
I
I
*
I
-I
)l=
z=Q5
'.
80
70
60
50
40
30
20
10
Distance[kml
* Indicate )ocation ofcurrent meters
'.
x [km]
zero]
Fig 3. Mooring sites in the slope of the Svin~y section.
,".
........11'*
~
Svinoysnitt-Aanderaa March-April 1996 . Transport
= 4.6 Sv
.
90
80
70
60
50
40
30
Distance[km]
o
20
E
~ 0.1
!2
•
t::
2-
(J)
§ 0.05
.=
~':"0-~8=0----:=----="60:-=~....J50~--4.L..0--3..L.0--2....L.0--..J.1 0--.-1
0
Distance[km]
Fig 4. Contour plot of montWy mean velocity field and vertical integrated transport,
March-April 1996
s12-300m
10
s22-300m
y -1.611e+OS·x + -9.486e+OS; r··2=O.n
10
8
t:
8-
4
I-
2
...
ro
8
6
c:
U)
t:
0
c-
U)
...I-c:
ro
6
4
2
0
0
-2
-2
-4
y = 1.01Se+OS·x + 2.135e+06; r··2=0.847
70
80
90
-4
100
Julian day
70
80
90
100
Julian day
Fig 5. One week moving average time series ofcalculated volume transports (fullline) and linear
regression estirnates based oei current meter 51-300 m and 52-300m ( dotted line), March-April
1996.
.
8
.;
...
OI-l....L&:-..4J:.L..:i:-I.-..!:-:'~-I.-:&::-:L.-..I:...I.....l...Jr.....L--L...l...JL...:L..:L.J.....JL....L....l-.J.....JL:l..U-..!.lW-!.....L:...I-J-!.....l:llJ......j-L,;..I.-.l.-l-!.....L.J.....I-!....L.;l-J
1020 30 10 20 30 9 19 29 9 19 29 8 1828 7 17 27 7 17 27 6 16 26 6 16 26 5 15 25 4 14 24 5 15 25 4 14 24 4 14 24 3 1323 3 13 23 2 12 22 1 11
ocr
X10'
MAY JUN JUL
AUG SEP
NOV DEC JAN
FEB MAR APR MAY JUN JUL
AUG
15 ,.....,-,.--r-r...,.-,.....,-,.--r-r...,.-,.....,-,.-r-r-r-,.-,..-,-,.--r-r-r-,.-r-.r--T--r-r-r-,.-.....,~
.....""T"'-r-,.-,..-,~-r-r....-,.--r-1r-r-,
............................~'-L .......-'--'-............................~..............-'--'-........-'-~.......J-.,;~-I.....L-..L-l-I-J..-'-..l-.l.-JL-L-I..-L...I-.L-JL.....I....J
oL....l~-'--'-
1020 30 10 20 30 9 1929 9 1929 8 1828 7 1727 7 17 27 6 1626 6 1626 5 1525 4 1424 5 1525 4 1424 4 1424 3 1323 3 1323 2 1222 1 11
"0'
10 r-r~
9
~
k
MAY
JUN
JUL
AUG
SEP
ocr
NOV
-r-r-,.-,..-,~-r-r-r-,.-r-.~-r""T"'-r-,.-,..-,~
.
DEC
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
............-r-,.-.....,~ .....-r-r-,.-.....,~.....""T"'....--r-,..-,~.....-,-....-,.-,..-,r-r......
8
7
6
J:
•
.. 3
,:: 2
1
.
o'-L
-'--'-
'-L--'--'--'-
-'-
'--'---'--'--'-
-'-........'--'-........-'-~'--'---'--'--'- ........--.L........-'--'-'-L-'-...J
1020 30 1020 30 9 1929 9 1929 8 1828 7 1727 7 17 27 6 1626 6 1626 5 1525 4 1424 5 1525 4 1424 4 1424 3 1323 3 1323 2 1222 1 11
.'0'
~
k
MAY
JUN
JUL
AUG
SEP
ocr
NOV
DEC
JAN
FEB
MAR
APR
MAY
10 1""'"T-r-r-r--r--r-1I""'"T-r-r""T"'-r-r-I""'"T-r--r--r--r--r-1I""'"T~--r-""T"'-r--r-,.....,~--r-""T"'....-r-I""'"T-r-r-r--r--r-1I""'"T-r
9
8
.
JUN
JUL
AUG
.....""T"'-r-,.-,..-,-.,...,
7
6
I~
2
1
oL..J'--'--'--'-.........L-'-''--'---'--'-........L.-l-JL-L-'--'-........L-l-J'--'---'--'-..........L.-l-JL-L--'--'-........L.-l-JL-L-'--'-........L.-l-J'--J.--'-..................L.-.l.-L....L...J
1020 30 10 20 30 9 19 29 9 19 29 8 18 28 7 17 27 7 17 27 6 16 26 6 16 26 5 15 25 4 14 24 5 1525 4 14 24 4 14 24 3 1323 3 13 23 2 12 22 1 11
.'0'
~
k
MAY
JUN
JUL
AUG
SEP
ocr
NOV
DEC
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
10 r-rr-r-r-r-r-,.-,..-,r-r-r-r-r--r-,..-,r-r--r-,-....--r-,..-,r-r--r-r-r--r--r-1I""'"T--r-r-r-,.-r-.r-r--r-,--r-,.-,..-,r-r-r--r--r--r--r-I""'"T-'
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mev Avg 672 "<lon
9
· . • • .. . . . . . . . • . . • . • • • • . . • . • . . . • • • • . . • • • . . . . • . . • . . • • • • • • . • . . . • • • . . • • . . • • • . . • . • . • • . . . . • • • Avg Tra • 5 178e+06
8
. . . . . . . . . • . . . . . . . . . . . . . . . . . . . . . . STD· 8 306e~
7
6
j~
2
I
oL..J.............-'-.......L-L--'--'-..I-...........J............l-.l.-JL-L-I..-'-..L-L..J....J.....J....l-J....J-'-...........L-..L-L..J..........1-...L-.L-J...J..........-'-~'-L .......-'-...L-J....J...J....J
102030 1020 30 9 1929 9 1929 8 1828 7 17 27 7 1727 6 1626 6 1626 5 1525 4 14 24 5 1525 4 1424 4 1424 3 1323 3 1323 2 1222 1 11
.'0'
MAY
JUN
JUL
AUG
SEP
ocr
NOV
DEC
JAN' FEB
MAR
APR
MAY
JUN
JUL
AUG
10 r-rr-r-r-r-r-,.-,..-,-,.--r-r-r--r-,..-,r-r-r-,--r--r-,..-,r-r-r-r-r--r-r-.r-r--r-r.......-r-,..-,r-r.....-,--r-,.-,..-,r-r--r-r....--r--r-1r-r-,
· . .. .. . . .. .. . . .•. . . . . . . . .•. .. .•. .. . . . .•. . .•. . . . . . . .. . . .. . . .. .•. .. . .. .. . . ..
. . , . . . . . Mev Avg 1344 "o~
9
· .. • .. . . .. .
Avg Tra· 520_06 c
8
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
'STD ··61Sge~
~ 7
~ 6
.. . .. .. ..
'-'--......-...:...:.....;...:.;
..~.,.. ..
..
..
I; :.'::.'::::::::::::::::::::::::::::::::::::::.::::::::::::::::::::::::::-.-.-.-.-.-.-.-."
.
,:: 2
1
oL...-L-I.-.J.....J-J....I.-L-L..J....J.....J-J.....L.J....L..J....J.....J-J....I.-L-L..J....J.....J-J....I.-J....L..J.....L-J-J....I.-J....L..J.....L-J-!....I.-L-L..J....J.....J--:L-..I.-.l.....l.-.L-J
102030 10 20 30 9 1929 9 1929 8 1828 7 1727 7 17 27 6 1626 6 1626 5 1525 4 1424 5 1525 4 1424 4 1424 3 1323 3 1323 2 1222 1 11
MAY
JUN
JUL
AUG
SEP
ocr
NOV
DEC
JAN
FEB
MAR
APR
MAY
Fig 6. Time series of estimated long-term volume transports for S 1-300 m,
22 April 1995-31 August 1996
JUN
JUL
AUG
•
r
I
•
.
'
9
.,O'
15
~ 10
.§.
J
~"
5
0
27
7
27
17
16
6
26
6
16
NOV
.10'
26
5
15
25
4
JAN
DEC
14
24
15
FEB
25
4
MAR
14
24
4
APR
14
24
13
MAY
23
3
13
JUN
15
°21..7-:-7-"'17:--::27~-6~-1:':6-26~....J6-..J1L..6 -26L.....J.5-..I15--2L...5-J..4-..L14-.J24-..JL.-1J..5-25J--14--1L...4-2J...-.14-.J14-.J21..4-3l--1.L3-.J23-.J3L-...J13
•
~
k
NOV
DEC
JAN
FES
MAR
APR
MAY
JUN
.10'
10 r--"T""--y-.,---r-r-"T""--y-.--r-.,r---r---y-.,.--r-r--"T""--y-.,.--r-r---r---r-...--r-,...--r---r-...--,
9
. Mov AY<J 166 "OU'"
8
.
. ... AYQ Tra • 4.ene+06 c
7
. . . . . . STD - 9.064e+Q5
6
I:
3
2
~
1
..
Ol.--...L--.L_.L..-....L27
7
17
27
~
:s.
_.L..-....L--.L_L-...L--.L_.l....--L......L_.l....--L......L_.l....-..l..--I_.l-..l..--I_.l-..l..--I_.l-..l..--I
6
16
26
6
NOV
.10'
10
9
..
. ....•.........................................
............••............
16
26
5
DEC
15
25
4
14
JAN
24
5
FEB
15
25
4
MAR
14
24
4
APR
14
24
3
MAY
13
23
3
13
JUN
r--,...-,..--r-,---r-.,---r-.,---,--.-......,r---r-,.--,.--,.--r-.,---r--,-"""'--,--r---r-,.--,...-,..---r-,.--..,
. . . . • • • . . .
. • • • • • . • • • . • . • • • • • • • • .
•. •••• ••. •.. .
8
7
6
· .....•. '
• • • • • • • • • . •
• •
~y
Avg 336 hours
• . • • • • • • . • • • . • . . . • • . • • • . • • . . . . • • • . . • . • . • • • . • • • . • • • • • • • • • • • • • • • • • •• 1<Y<J Tra - 4.901.+06 c
•............. "
..............•..
.
STO '-1.187.+05
j:
2
1
oL.---'---JL-.......--JL-.......--'_-'-....L_.L--.L._.L--.L._J.--'-_.............--JL-.......--J_.......- - ' _ - ' - - - ' _...........L_.L--.L._J.-....J
14 24
16 26
27
7
17 27
6
16 26
6
4
4
14
24
4
14 24
3
3
15 25
15' 25
13 23
13
5
•
~
k
10
9
8
7
NOV
.10'
5
~
4
~
3
JAN
MAR
FEB
MAY
APR
JUN
Mov AY<J 672 ~OU'"
1<Y<J Tra - 4.ge+06 .......
·
•
• • • . • . . • . . • . . . . . • . . . . . . • . . • . . . . . . . . • . . • • • • . • . • . . . . • • • . . . . . . . • • • . • • • . . . • . . • . • . . • • • • • . • • • . STD '-4891.+05
6
~
DEC
r---,...-,..-...,.-oy--r-.,---r-.,---r-..,........,r---r-,......-,...-,..-...,.-,.--r-.,---,--.-......,r---,--r---,...-,.--,...-,..-.....
..
..
....... ~-----~:-~..
2
1
Ol.--...l---L......JL-..L-........---I._.L--...L.......J...--I_..L-........---I._.L--...l-............I_..L--L..---I._.L--...L...............I_..L--L..--l._L......I
27
7
17 27
6
16 26
16 26
5
15 25
4
14 24
5
15 25
4
14 24
4
14 24
3
13 23
3
13
NOV
DEC
JAN
FES
MAR
APR
MAY
JUN
"0.....
• •••••••••••••••...••••••....••.•••••••••••••••••.•••••••••••••••••••••.•...••.••••••.•• MovAY<J 1344
• •••••.••••••.••.••.•••.......••••••.•••.•••.••.•.•.•••.•••••••.••••••.•..••••••••.•••• '1<Y<J Tra - 5 0148+06
-
- -- - - - - - - -.-.. --:-~-- -
· . . • . . . . . • . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . STO .- 2.114e+Q5
-
::::::::::::: :17
27
6
16
NOV
..
---.~
..-..;..
26
6
16
DEC
26
..
5
15
JAN
25
4
14
FES
24
5
15
MAR
25
4
14
APR
.~ :::::::::
:.:..:...:::: ..:..:..:"
.;..;..:..
24
4
14
24
MAY
Fig 7. Time series of estimated long-term volume transports for 52-300 m,
November 1995- June 1996
3
13
JUN
23
3
13