Rapp. Comm. int. Mer Médit., 36,2001
64
Seasonal signal in the ?ows through the Strait
From October 1995 to May 1998 a mooring array of recording cur-
rentmeters was deployed in the eastern part of the Strait of Gibraltar
within CANIGO Project. The observed velocities were used to pro-
duce time series of the slowly-varying in?ow (Q
1
), out?ow (Q
2
), and
the depth of the interface (
?
2
), defined as the surface of null along-
strait velocity.Time series were fitted to a sinusoidal model of annual
and semiannual periodicity in order to investigate the existence of sea-
sonal signals [1]. Table 1 shows the annual parameters, the cycle in
which the present work concentrates.
Table 1. Annual parameters of the model. All phases are
referred do the year-day 1.
Variable
A
a
?
a
Q
1,a
(Sv)
0.101±0.038
225º±21º
Q
2,a
(Sv)
0.028±0.025
201º±51º
Q
0,a
(Sv)
0.077±0.044
234º±33º
?
2,a
(m)
4.557±3.269
61º±42º
(??
)
a
(kg/m
3
)0.29
7±3.269
238º
±42º
The net barotropic signal and the Mediterranean Sea level cycle
The simplest model to balance the net barotropic signal in the
Mediterranean Sea is Q
0
,
a
=
d
V
M
E
D
/dt= A
M
E
D
d
?
/dt (V
M
E
D
is the vo
l-
ume of the Mediterranean Sea, A
M
E
D
= 2.5 10
1
2
m
2
its area, and
?
t
h
e
sea level), and ignores the eva
p
o
r
a
t
ive seasonal cycle (
E
-
P
)
a
. Ta
k
i
n
g
Q
0
,
a
from Table 1, this equation would imply an unrealistic annual sig-
nal in
?
of 15±9 cm with maximum value in Nove
m
b e r.
TOPEX/POSEIDON altimetry data provide evidence of a seasonal
cycle in the Mediterranean mean sea level of 8 to 10 cm of amplitude
with maximum value in October [2]. This annual signal is quite ev
i
d
e
n
t
in longer time series of altimetry data (P.
Y. LeTraon, personal commu-
nication). How
eve
r, part of this signal is due to steric effects. In [1] and
using the MEDATLAS data set, these anomalies (averaged over the
whole Mediterranean) are computed to give an amplitude of 5.5±0.4
cm and a phase of 253±12º for the thermal contribution and non-sig-
n
i
ficantly different from zero for the haline part (0.11±0.24 cm).
Taking into account the thermal steric anomaly, the eff
e
c
t
ive mass
variation within the sea is
?
m
=
?
o
b
s
-
?
s
t
r
, where
?
o
b
s
is the observe
d
sea level signal mentioned in [2] and
?
s
t
r
the steric contribution. T
h
e
RHS of the aforementioned balance equation should be 0.02 Sv at 230º,
far from the value of Q
0
,
a
reported in Table 1. The correct mass balance
must include the climatological (E-P) forcing, Q
0
,
a
-
(
E
-
P
)
a
=
A
M
E
D
d
?
m
/
d
t, where (
E
-
P
)
a
is the annual signal in the net evaporation. Using
the previous value for the RHS and Q
0
,
a
of Table 1, we obtain 0.06 Sv
at 237º for (
E
-
P
)
a
.
The seasonal net evaporative cycle is difficult to measure. [3] sug-
gests a cycle of around 6 cm/month (0.06 Sv) that peaks in August
(215º), while [4] indicates a seasonal signal of 7.1cm/month (0.07 Sv)
at 181º using the World Survey of Climatology.The agreement of these
amplitudes with the amplitude deduced from the mass imbalance is
good. Phases will probably agree within the confidence intervals of the
aforementioned values of (E-P)
a
.The estimate of Q
0,a
from our obser-
vations is consistent with the requested mass balance within the
Mediterranean.
Driving force for the net barotropic ?ow annual signal
Table 1 includes the seasonal signal of the density contrast between
Atlantic and Mediterranean waters given in [4]. The phase agreement
between(
??
)
a
andQ
0,a
suggests a cause-effect relationship. A simple
unidimensional two-layer model for the exchange [1] shows that under
maximal exchange, a positive net barotropic signal, for instance, is
achieved by means of an increase of half this signal in Q
1
and a simi-
lar diminution in Q
2
.Table 1 shows that this is not the case since both
Q
1,a
andQ
2,a
oscillate near-in-phase with (
??
)
a
.The seasonal signal
inQ
0
=Q
1
-Q
2
arises because the amplitude of Q
1,a
is greater than the
amplitude of Q
2,a
,that is, the in?ow is more sensitive to
??
variations
than the out?ow.The model shows that this "asymmetry" is only com-
patible with submaximal exchange. The observed composite Froude
number at the eastern section during CANIGO was clearly subcritical
[1], which implies submaximal exchange [5] according to our reason-
ing.
The hydraulic theory says that if the exchange is submaximal, the
conditions in the adjacent reservoirs are able to in?uence the exchange
[5]. For instance, well-developed anticyclonic gyres in the Western
Alboran Basin may favour the in?ow and hamper the out?ow by sink-
ing the interface near the eastern entrance of the Strait (the opposite if
the gyre reduces), thus linking the seasonal signals in the exchange
with the hydrographic features in the Western Alboran Sea. This pos-
sibility is explored in [1] to put forward that the (
??)
a
signal favours
the presence of greater (smaller) and more (less) stable anticyclonic
gyres during summer (winter), a fact easily observable in SST
imagery.This seasonal pattern would induce a phase of the interface
depth like that in Table 1.
Acknowledgements
This work was supported by the European Commission through the
Canary Azores Gibraltar Observations, "CANIGO", Project (MAS3-
PL95-0443). Partial support by the Spanish National Programme of
Marine Science and Technology (MAR95-1950-C02-01) is also
acknowledged.
References
1. Garcia Lafuente, J., Delgado, J., Vargas, J.M., Vargas, M. and Plaza, F.
2001. Low frequency variability of the exchanged ?ows through the
Strait of Gibraltar during CANIGO. Deep Sea Res.(in press)
2.Ayoub, N., LeTraon, P.Y., De Mey, P. 1998. A description of the
Mediterranean surface variable circulation from combined ERS-1 and
TOPEX/POSEIDON altimetric data. J. Mar. Systems, 18: 3-40.
3. Carter, D.B. 1956. The water balance of the Mediterranean and Black
Seas.Publ. Climatol., 9: 127-174.
4. Bormans, M., Garrett, C. and Thompson, K.R. 1986. Surface variabili-
ty of the surface in?ow through the Strait of Gibraltar.Oceanol. Acta, 9:
403-414.
5. Farmer, D.M. and Armi, L. 1986. Maximal two-layer exchange over a
sill and through the combination of a sill and contraction with barotropic
?ow.J. Fluid. Mech., 164: 53-76.
SEASONALVARIABILITY OF THE EXCHANGE THROUGH THE STRAIT OF GIBRALTAR
AND THE MEDITERRANEAN MEAN SEA LEVEL
Jesús García Lafuente
1
*, Juan M. Vargas
1
, Javier Delgado
1
,Tarek Sarhan
2
, Francisco Plaza
1
, Manuel Vargas
2
1
Departamento de Física Aplicada II, Universidad de Málaga, Málaga, Spain - glafuente@ctima.uma.es
2
Laboratorio Oceanográfico de Málaga, Instituto Español de Oceanografía, Spain
Abstract
Currentmeter observations at the eastern section of the Strait of Gibraltar were used to determine the seasonal signal in the exchanged
?ows. The computed signal of the net barotropic ?ow is compatible with the water mass balance within the Mediterranean sea at a sea-
sonal scale. The enhanced response of the in?ow to the annual cycle of the density contrast between Mediterranean and surface Atlantic
waters appears to be the cause of this cycle. This enhanced response is possible if the exchange is submaximal, in which case seasonally-
induced variations of the size of the Western Alboran Gyre can become an effective mechanism to induce the observed seasonal signals.
Keywords: Alboran Sea, Strait of Gibraltar, sea level.
