THIS PAPER NOT TO BE CITED WITHOUT PRIOR REFERENCE TO THE AUTHORS
International Couneil for the
Exploration of the Sea
GM 1986/C:5
Hydrography Committee
CURRENT METER OBSERVATIONS IN THE NORTH CHANNEL
AND THEIR RELATIONSHIP TO LOCAL WIND
C. A. Hudson and J. A. Duranee
Ministry of Agrieulture, Fisheries and Food
Direetorate of Fisheries Research
Fisheries Laboratory
Lowestoft, Suffolk NR33 OHT, England
ABSTRACT
Aseries of eurrent meter observations taken in spring 1985 was
•
used, with a simplified eross-seetion, to estimate the transport through
the North Channel of the Irish Sea.
Wind data from two loeal stations
were available over the same period, allowing a study of the relationship
between wind and transport to be made.
Similar relationships between
wind and individual eurrent meter reeords were also examined.
Estimates
were made of the direetion of the wind eomponent whieh gave the best
visual correlation with transport and flow.
The time lag between wind
and wind indueed effeets was also considered.
,
,
RESUME
L'on s'est servi d'une section simplifiee des mesures relevees au
•
eourantometre au printernps de l'annee 1985 pour estimer le transport par
le eanal septentrional de la mer d'Irlande.
L'on disposait en outre des
releves du vent l deux stations de la region sur la
a permis
d'et~dier
la
rel~tion
m~me
periode, ce qui
entre le vent et le transport.
tions analogues entre les releves du vent et les releves
courantometres ont aussi ete examinees.
a
Des rela-
chacun des
L'on a estime la direction de la
composante eolienne qui donnait lu meilleure correlation visuelle avec le
transport et le debit.
Le decalage temporel entre le vent et les effets
du vent a par ailleurs,
ete
considere.
INTRODUCTION
Previous studies of the dynamics of the Irish Sea and adjoining
waters have used a variety of methods to estimate the transport and
eurrents through the North Channel.
1
Continuity methods using tracers such as salt and caesium 137
produced results which have been summarised by McKay and Baxter (1985).
Transport estimates vary from 4.7 x 104 m3 s- 1 to 5.8 x 104 m3 s- 1 , although
earlier work by Bowden (1950) using continuity of salt and taking account
of longitudinal mixing gives an estimate of 2.6 x 104 m3 s- 1 , and a model
of caesium 137 mixing by Prandle (1984) estimated
6.0
x
3
transport of
1Q4 m3 s -1.
Early use of Lagrangian current measuring methods, using surface
drifters, indicated a southward flow along the Irish Coast but did not
allow any transport calculations to be made.
Eulerian current measurements in the area have been restricted by
the hostility of the environment and there have been no previous current
measurements from the deep part of the channel where depths in excess of
200 m and strong tides make the successful mooring of current meters
exceptionally difficult.
Howarth (1982) deployed aseries of moorings in the Malin Shelf Sea
and North Channel.
•
However, only three of these lay in the channel.
At the same time Howarth recorded the potential of the Port Patrick
to Donaghadee telegraph cable.
Using Faraday's law these measurements
provide another method of estimating through channel flow.
The method
had been used previously by Bowden and Hughes (1961), Prandle and
Harrison (1975) and by Prandle (1976) and (1979).
All these authors con-
sidered the possibility of a relationship between the cable potential and
Iocal wind.
Howarth extended study of such a relationship to his current
meter data.
He observed continuous southward flow along the Irish Coast
and near the Mull of Galloway but flow of variable direction at his
central mooring, the direction of the flow depending on the direction of
the mean wind stress.
As long as the winds were not weak, his correla-
tion between longitudinal wind stress and cable voltage was good with
values'of up to 0.85.
I
I'
I
In the present study currents have been measured directly using a
se ries of moorings across the channel (Figure 1).
to estimate the overall transport.
These have been used
Local wind data has been used to
study the wind/transport or wind/current relationship.
CURRENT METER DATA
In 1985 the MAFF, Fisheries Laboratory, Lowestoft, deployed aseries
of current meter moorings in the North Channel at the positions shown in
Figure 1.
The deployment, from 21 March to 12 June was divided into two
2
•
distinct parts by the mooring servicing around 25 April.
seven
moorin~s
were deployed
comprisin~
Initially,
fifteen current meters.
Of
these, only one (AB) failed to return any data, whilst two others (FB and
GM) returned curtailed records.
After servicing only five moorings were
redeployed and of the eleven meters, two returned no data and a third
returned only seventeen days data, compared with thirty-two days and
forty-five days from other meters.
Coverage, especially on the eastern
side, was, therefore much better for the first half of the deployment and
only figures from this data set were used to estimate transport.
Current meter data was available in component form at ten-minute
intervals but, initially only filtered daily means were used to calculate
the transport.
Data was presented in components in 330° and 60°
directions (that is parallel with and perpendicular to the axis of the
•
channel).
CALCULATION OF TRANSPORT
Transport was calculated across a section running 60° to 240°, using
the 330° current component.
A cross section of the channel was obtained
from an echo sounder transect by the MAFF Research Vessel CLIONE.
Although this may not have been very accurate, any inaccuracies apart
from those at the extreme edges were considered to be insignificant
compared with the uncertainties introduced when extrapolating the current
measurements over the area of the blocks into which the section was
divided.
Figure 2 shows the division of the cross section into blocks in
such a way that each block. except for those at the edges of the channel.
contained one current meter position.
The names by which the blocks are
referred to are also given in the figure.
In most cases the first letter
in the name refers to the mooring name and the second to the current
meter position (Top, Middle or Bottom) on the mooring.
Transport for each block was calculated by multiplying the estimated
area of the block by tlle 330° current component from the relevant meter.
For blocks not containing meters and for those where the meter returned
incomplete records, values from the nearest appropriate mooring were
used.
Daily means were calculated for the whole area (Figure 3).
The
means over the twenty five-days of the deployment were calculated for
each block and plotted (Figure 4).
In this figure the height of each bar
Is proportional to the current and the area of the bar proportional to
the transport.
3
WIND DATA
Local wind data was available at three-hourly intervals from two
stations, Orlock Point and West Fruegh (Figure 1).
speed (in knots) and direction.
Data was presented as
Daily means were calculated and the com-
ponents every ten degrees were derived.
Wind values at the two stations
correlated weIl, as might be expected with speeds at Orlock Point
generally higher (Figure 5).
In this paper only the West Fruegh data, a
stick plot of which is shown in Figure 6, are used.
RESULTS AND DISCUSSION
Figure 7 shows progressive vector diagrams (PVD) for the current
meter data after a Gaussian filter has been applied.
The variation in
scales should be noted when making intercomparisons.
Apart from the
westernmost mooring (I) flow, especially in the upper layer is generally
to the north north west, that is outward along the axis of the channel.
In the case of I, flow in the upper layer is to the south south east and
that in the lower layer to the west.
•
At the end of the period southward
flow can be seen in the upper layer as far east a mooring G.
Figure'
8(a and b) shows the variation of upper layer flow with time for
both deployments, and the north/south component of wind (note that the
wind data was only available up to 31 May).
After the service no records
were returned for the top meters of the G and E moorings.
curves in Figure
The dotted
8(a and b) show records for the middle and bottom meter
from G and E respectively.
clear from this figure.
The paucity of data after the service is
Again, the increased 'inflow' at the end of the
first period is visible.
This widening of the inflow may go some way towards explaining
observations of changes in the salinity field of the North Channel.
Figure 9 shows aseries of salinity profiles collected by MAFF, 5MBA and
DAFS between November 1984 and August 1985.
Of these, three (e, f & g)
from MAFF and DAFS cruises coincide with the period of the current meter
deployment.
The area of maximum salinity at the centre in Figure 9 (e
& f, when southward flow was restricted to the I mooring, is in the
cent re of the channel but by the end of the period in Figure 9(g) when
thc inflow has broadened the high salinity water has been displaced to
the east.
Previous studies of the wind/transport relationship used anequation
of the form
U(t+ßt)
= AW2 (t)cos(9(t)-a)+U o
4
•
where U(t), the residual flow (or in some cases the potential of the
telegraph cable) lags the wind stress W2 (t) by a time ~t with wind blowing from 0(t).
The wind 1s most effective at producing residual flow
when blowing from angle a.
A is a coefficient and U is the residual
o
flow (or potential) which does not depend on wind forcing.
Figure 10 shows plots of daily mean transport and daily mean wind
component at intervals of 30°.
Although it 1s difficult to see, from
these plots, a "best" wind direction a, the correlation with wind from
150° appears slightly better over the majority of the period.
The
notable exception is at the time of minimum transport on 31 March.
It would be expected that there would be some time lag between wind
stress being applied and change in current and hence in transport.
vious work gives values of
•
~t
Pre-
of the order of hours or less; e.g. Bowden
and Hughes (1961) 2 hours, and Prandle and Harrison (1975) a few minutes.
However, the latter do point out that the resolution in time of the wind
data (6 hours in their case) does not allow accurate determination of the
lag.
The daily mean plots of Figure 10 do not allow estimates of lag of
the order of hours.
However, the possibility of a longer time lag,
allowing for the cumulative effect of wind on transport has been
considered.
This mayaiso be equivalent to looking at the effects of
winds further away from the area, which may take days to act.
shows plots of the 150
to four days.
0
Figure 11
wind component and transport for time lags of one
The four-day lag gives reasonable correlat10n for most of
the period.
Over the twenty-five-day period the mean transport was estimated to
be 8.5 x 104m3 s-l. This is larger than the estimates made by tracer and
modelling methods but agrees well with Howarth (1982) value of
8 x 10 4 m3 s-l outflmT for his moorings A, Band C. when the wind was
southerly. However~Tfowarth also quoted a net inflow transport of
2.2 x 104m3 s-l when the wind was northerly and explained this by
pointing out the high correlation between the current at his centre
mooring Band the wind.
There is no equivalent station in the present
study as Howarth's station B was further north than all our stations, but
this reversal in flow is seen in moorings Hand C as discussed
previously.
The fact that Howarth is extrapolating his southward flow
over a far wider area than in our case, explains the southward transport.
His easternmost station, again further north than the present ones,
showed southward flow over the 45 d of his record which is not seen at
station E in our study.
5
The minimum daily mean transport in the March-April period was
0.6 x 10 4 m3 s-1 on 31 March and the maximum of 3.65 x 10 5 m3 s-1 occurred
the previous day.
Correlation between wind and current at individual meters was also
studied.
The inclusion of the wind data in Figure 8 allows a more
detailed study of wind current correlation.
This figure shows that cer-
tain 'events appear all the way across the chanilel'.
For instance, the
.southward maximum atH and I has corresponding decreases in northward
flow at the other moorings on 31 March and there is a northward maximum
.at I, Hand F on 24th May.
However the southward maximum at E with
corresponding minima in northward flow at Fand A seems to lead the
equivalent minimum at G, C, Hand I by a day.
(This may be due to the
averaging of data.)
Some of these phenomena have corresponding wind changes.
For
instance, there is a southward maximum on 11 April and northerly wind for
aperiod from the 18 April when the flow is at it's broadest.
CONCLUSIONS
Aseries of current meters has provided good coverage,of the North
Channel and has allowed an estimate to be made of the daily mean transport.
Graphs have shown that, although correlation with the wind is not
necessarily as clear as suggested in previous work, there is certainly
some wind/transport relationship.
Howarth's conclusion that there is
better correlation in the "middle" of the channel than at the edges was
not confirmed, although his "middle" station was in 150 m of water and
not in the deepest part of the channel (compare the present station G in
nearly 280 m).
Other differences between the results in this paper and
his mayaiso be due to the different coverage of the region and the
number of near-bed observations included in our transport estimates.
In order to get better estimates of "best" wind angle (a) and time
lag (Ät), calculations of mean transport every three hours would need to
be made.
These could then be used with the three-hourly wind data.
In this paper no account has been taken of other current forcing
sources such as pressure gradients.
Tide gauge da ta will shortly be
available from lOS (Bidston) from Portpatrick and Belfast.
With this
additional information allowing for estimation of these contributions to
the currents, our remarkably good data set might be analysed to reach
further conclusions on the flow pattern and wind/transport relationship
in the area.
6
•
---------
\
ACKNOWLEDGEMENTS
The authors wish to thank David
El1~tt
(SMBA) and Tony Martin (DAFS)
for supplying the salinity profiles shown in Figure 9.
REFERENCES
BOWDEN. K. F•• 1950.
Processes affecting the salinity of the Irish Sea.
Monthly Not. Roy. Astr.
Soc •• Geophys. Suppl ••
BOWDEN. K. F. and HUGHES. P•• 1961.
~(2):
63-90.
The flow of water through the Irish
Geophys. J. R. Astr. Soc •• 5:
Sea and its relation to wind.
265-291.
HOWARTH. M. J •• 1982.
Non-tidal flow in the North Channel of the lrish
Sea. pp. 205-241.
•
In: J. C. J. Nihoul (ed.).
semi-enciosed shelf seas.
Hydrodynamics of
Elsevier. Amsterdam. Oxford and New
York.
McKAY. W. A. and BAXTER. M. S•• 1985.
Water transport from the
North-east Irish Sea to Western Scottish coastal waters: further
observations from time-trend matching of SellaEield radiocaesium.
Estuar. Cstl Shelf Sei ••
PRANDLE, D., 1976.
lrish Sea.
PRANDLE, D•• 1979.
~
(4): 471-480.
Wind indueed flow through the North Channel of the
Geophys. J. R. Astr. Soe ••
~:
437-442.
Reeordings oE potential difference across the
Portpatriek-Donaghadee submarine eable (1977/78).
Rep. lnst.
Oeeanogr. Sei •• (83). 1-9. (unpublished manuseript).
PRANDLE. D•• 1984.
•
A modelling study of mixing of 13/Cs in the seas of
the European continental sheif.
Phil. Trans. Roy. Soc. Lond ••
A310: 407-436.
PRANDLE. D. and HARRISON. A. J •• 1975.
Reeording of potential difference
across the Portpatrick-Donaghadee submarine cable.
Oeeanogr. Sei •• (21), 1-14.
Rep.
(unpublished manuseript).
7
Ins~~
I
I
I
Orlock
v
II
I
Jl
l.b~=========================================================================-=:
..
-l
Figure
1. Current meter mooring positions anrl wind da ta station
posit ions.
e
west
0
,
"-
I
40
H C
I
" " "-
G
A
E
F
WE
'"
--
east
EaT
WT
IT
+
+
HT +
CT
AT
FT
ET
+
+
+
EaE /
/
I
/
/
+
EaB
GT
80
IB
120
+
160
HB
AB
200
240
280
_------'--------'--'----'---""---''--------'---J
L - I
Figure
2. Bottom profile of North Channel with blocks used in transport
calculations.
5
,........
4-
It)
+
0
.....
I.&J
x
3
2
1
CI)
"
.."
0
~
'-' -1
....
0:::
-2
0
Q..
CI)
Z
~
....
-3
-4-
•
0:::
-5
27
MARCH
1
Figure
6
DATE
11
16
APRIL
3. Daily mean transport values.
Key
11 Top
meter
~ Middle meter
o
Bottom meter
•
WE
Figure
W
He
GAF
E
Ea
EaE
4. Twenty-five day mean transport for individual blocks.
40 . . . . . - - - - - - - - - - - - - - - - - - - - - - - , 20
10
30
WEST F
~
(f)
"'~
'-'
20
0
10
-10
ORLOCK
C
LaJ
LaJ
a..
0
-20
-10
--30
(f)
- 20
L..J.-I-L-..I-...L.....I-.L..J.....L...........L...L-I-L-L..&-..I~...&.._L~
. ............&._a......L._a...._lI.__"_..........- - ' - - '
27
1
MARCH
Figure
6
DATE
11
APRIL
16
21
5. Wind data (3 hourly) for Orlock Point and West Fruegh.
26
-40
~
....
ce
0_
%0)
.......
E
~u
.... -
~
0
cn
10
5
0
-5
-10
20
MAR
25
30
85
1
APR
5
10
85
Figure
6. Stick plot of West Fruegh Wind data.
15
20
'"°1 Top- .....
!
i
,
1
r~
iN11
i
.
16'
I
I
,
•
--~~-~--~~~I
16'
.
~"~---l
~
'"
"
70
\
\
16'
"
"
\
\
"
\
20
16
,
•
16
"
10
"kllo••" tr••"
10
,
" "
16'
"
\.
].
"
"
.110... t,. ..
..
::1
16'
".
.,
'"
,
.l_____
o
20
10
l!
&0
10
tOD
UD
140
110
klloHt,.1I
..
Middle
"
10
.,
"
"
"
10
•o _. --- _---r--.-.. . .
.
10
<'0
"r--~~---~-~--t
Bottom
"
..
i
"
I
,
16
[
"1
I
I
!
',!---:0:,,-
&J
JO
.... Io.,t,. ••
-~,,- ~f
,"-,
j
,
.
.~-_._--10
I!
1o:110Htr"
H
Figure
20
1.
"
\
•• j
I
.Lo
I
I
\.
10
~O
\
I
\
_
• -..... --------.----~_- .
10
'0
'0
60
!
j.t
'C
kllo-.tru
c
7. Progressive vector diagrams for current meters in the North
Channel.
lO
\
-~
'0
51
10
1lI1o.. trt •
7D
'0
.0
l
tOD
--------------
'00
.i
... t·--~--~-~--~--~--t
,.. t·---~--~--~--~-----t
Top
"'1
".
q
---
I
\
."
'"
\
"
'"
"
\
\
,.
'"
"
,.
"'\
"
\
',f---,,-,--."'.---',-.--,,"---"',,"-,--,+,:
't,---=.,::---""";-----:"'-,---,,,,,,-,
---:!2'JOf.
IIIJ ... t,...
IIlloNt,.••
Mlddle
.. t-.--~-~-~--~-~-~----t
Bollom
"
"
'\
"
A
Figure 7.
"
"
"
Uloutr ••
F
..
"
"
'-----., ,
"j
,,1
I
""
"
i
I
'1
E
,L,
,
/
\
\
I
.
~-- _._~-_.....-~._~_.
"
" "
11 110•• t,. ..
"
"
!
~
I
~E
"
E
Continued.
I'
Deployment 6
Key
Top meter
20
Middle meter - - - E
F
10
A
$'
.-.
(J)
......
E
"'0
(J)
(J)
10
0-
G
Cf)
c:
......
10
()
t:
(J)
::J
C
1S"
10
H
•
.., 151----------------------]
",
*
c-
5
U)
",
c:
~ -5
--15
_.
27
Figure
March
6
April
16
26
8. One hund red and fifty degree components of current for the top
current meters and northerly component oE wind.
Deployment 7
10
s
1Iil,
I
~,.I
",
\
E
,... -- ~
,--_......
'-..-_
j\
I" ....
20
15
10
5
-
•
.....CIJ
E
15
"'0
Q)
Q)
Co
CIJ
10
5
'~I
....
"" /
" i
/
'"
\
r,
'y '
,
1...-4'
+00
c::
\
......
Q)
G
• ::::l
U
5
10
5
15
--,
25.
f /l
,§
'C
Cl>
Cl>
C.
'"c:
't:l
~
-15
6
Figure 8.
Continued
May
16
26
0
34.3
0
33.8
20
40
80
100
80
100
120
140
160
200
S(x10 3 )
180
200
0
5Dec 1984
Challeng er
31 Jan
-1 Feb 1985
Challenger
5MBA
5MBA
5 km
S%o.
24-25' March
1985 CHane
...........
220
DAFS
5MBA
a
b
c
300
MAFF
e
d
0
20
40
60
80
100
120
140
160
sex 10 3 ,)
180
200
0
220
5 km
l-t-1
8%0
S%o
25 April 1985
Cliane
5 May 1985
Challenger
S%.
11 June 1985
CHane
16 Aug 1985
Challenger
DAFS
MAFF
f
5MBA
MAFF
h
9
34.27 >S
34.4 > S > 34.2
S > 34.4
Figllre
9. Sali.nity profiles November 1984 to August 1985.
5MBA
j
-1
~--~-
..
-~--~~-
~
2
..... WIND
4 --TRANS
8
4
~ 31-----'
'.
&.-----,..-------------,
..... WIND
&,------,--------------,
&,-----,...------------,
4
on
WIND D1RECTlON SO
8
......
~
~
3
oll
ltl
i
~ -1
VI
~~
~
~
c
11.
~
C
~ -2
111
Z -3
-4 Z
~
-& L2LO...............25.Lo.~~JO
.......~.....J ...................8L.........~1.L.4.................1..L8....J
-~ L2LO............~~~~JO~~........4~~..J8L.... ......~IL4...............1.L.8...J
MTE
MARCH
APRIL
&
5
......
111
"""-
'"
~
WIND D1RECTlDN JO
4
8
3
4
lC
......
~
Q(
C
0
0
CL -2
VI
~
~
11.
C
-4 Z
,
i
-3
-8
-4
-5
UJ
UJ
VI
-1
0
z
........
:::E
~
oll
+
101
......
111
Vi'
2
20
25
IoIARCH
3D
4
MTE
14
8
APRIl
18
~
-8
-8
loWlCH
0
I-~-~..,..L--:...--l.....:--~~_+~-.,....:+_rl
0
;;,.
MTE
-2
-4
WIND D1RECTlON 80
8
......
VI
:::E
C
0
0
~
-4
..
111
11.
20
25
MARCH
30
4
MTE
14
8
C
-4 Z
i
-8
"
-5
UJ
UJ
VI
~ -1
~
........
~
-2
8
14
18
APRIL
&
..... WIND
- - TRANS
oll
-3
4
MTE
4
z
~
i
-8
-5 20
3
~
-3
-4
APRIL
~
+
101
)(
......
~
11.
VI
V I "
i
-4
8
~
~ 0 f--,~__,.-'--';.-~-~~....:...._':_+._+~-~___l 0
+
WIND D1RECTlDN 120
4 --TRANS
......
111
;;,.
..... WIND
4 - - TRANS
8
WIND D1RECTlDN 150
3
4
~
~
oll
+
101
lC
......
~
C
0
0
18
APRIL
0
-4 Z
'.
i
-8
-4
-5
W
W
11.
VI
-1
0 -2
CL
111
Z -3
~
~
......
VI
""":::E
20
25
lWlCH
30
4
MTE
8
14
18
APRIL
Figure 10. Transport and wind components every 30°.
(
..
o DAY
•
•
~
f
1/1
!
4
2
LAG
2 DAY LAG
"""
~
...
0
0
2
.....
Q
....
....
A-
1/1
~
~
-10
0
0
.....
Q
....
....
A-
'"
10
-4
20
Il4TE
"""
4
2
0
0
}
.....
Q
....
....
A-
'"
-2
~
f
'"
!
"""
~
4
f
~
0
1.
-10
30
IWICH
10
""
"""
~
4
2
0
2
.....
Q
....
....
A-
0
'"
""
-2
-4 20
30
IWICH
Il4TE
•
18
APllI.
-10
-4
20
30
IWICH
Il4TE
•
N'lllL
,.
e
'"
",'
-4 20
APllI.
2
.....
....A-
0
3 DAY LAG
•
•
•
•
30
1 DAY LAG
10
~
-2
IWICH
•
!
......
2
IWICH
'"
lii'
4
-2
-4 20
~
•
10
•
-2
f
.. DAY LAG
•
10
-10
Figure 11. Transport and 150 0 wind component with lags of one to four
days.
Il4TE
•
APllI.
.
,
-10