Diapositive 1

1
THE NORTH ATLANTIC AND MEDITERRANEAN MERCATOR
MODEL
Romain BOURDALLE-BADIE, Yann DRILLET, Laure
SIEFRIDT, Christian LE PROVOST
Model OPA-8.1 (1998) Buffer zones 9ºN
Nordic Seas Gulf of Cadiz Horizontal
resolution 5 to 7 km (1288 x 1022 pts) Vertical
resolution 43 levels Bathymetry Smith and
Sandwell (1997) Forcing ECMWF daily averaged
analysis, 1998 to 2000 Computational cost 4 x
7.2 Gb 1 year
(on VPP5000) 180 h
Introduction The North Atlantic and
Mediterranean Mercator system is a primitive
equation model assimilating real-time altimeter
and in situ observations by optimal
interpolation. It will be run in
quasi-operational mode in 2002 to provide weekly
analyses and 14-day forecasts. The model has been
run without assimilation for several years. An
11-year spin-up simulation has been previously
evaluated (cf Mercator Newsletter n5). This
poster presents some recent results of a 1998 to
2000 inter-annual simulation (PAM-20) without
assimilation.
Contact PAM_at_cerfacs.fr
The Gulf Stream trajectory
The pathway of the Gulf Stream in the PAM-20
simulation is relatively correct compared to the
observations. After a light overshoot at Cap
Hatteras, the trajectory is close to the
data. The same diagnostic has been applied on the
synthetic mean sea surface height computed from
TP/ERS and in-situ data, on a POP (1/10
horizontal resolution in in z-level vertical
coordinates) and on a MICOM (1/12 horizontal
resolution in in isopycnal vertical coordinates).
The pathway in the MICOM simulation is perfect
while it is too much south in the POP simulation.
Meddies representation
A restoring zone has been introduced in the Gulf
of Cadiz. It successfully restores the
Mediterranean waters at the right depth without
impacting too much on the model variability.
However, the meddy formation still occurs at
800m, at the maximum of potential vorticity. The
bathymetry primarily controls the trajectories of
the meddies, while the model circulation advects
them too much westward. A deepening is observed
during their movement that may be attributed to a
new buoyancy equilibrium reached with getting
away from the restoring zone. The simulated
Meddies present realistic salinity and speed
profiles. Transport in the eddy is evaluated at 3
to 4 Sv and maximum speed at 20cm/s from 5-day
averaged fields.
Location of the Gulf Stream detected from the
maximum gradient in the mean Sea Surface Height
courtesy M Brémond (LEGOS) MSSH M-H Rio (CLS),
POP (Smith al, 2000), MICOM (Chassignet al,
2001)
Bathymetry isovalues and trajectory of a meddy
superposed on a snapshot of salinity at 868m.
Transports in the Caribbean sea
Sea Surface Height over the fourth year of the
PAM-20 simulation
The circulation around the Caribbean islands is
in good agreement with the observations although
the current intensity appears quite weak in the
Florida Strait (27.2 Sv versus 33 Sv for cable
data) and strong in Windward Passage. Eddies in
the Gulf of Mexico are generated with a realistic
frequency (1-2 per years).
Vertical section of salinity field through a
meddie

• Conclusion
• A realistic general circulation
• A satisfying mesoscale activity
• Improvement of the Gulf Stream separation
• Some problems subsist in the vertical dynamic
  (deep convection too deep,
• unsatisfactory sinking of dense water, )
• Perspectives
• Operate the PAM system in quasi operational mode
• Open the southern boundary
• Investigate the advection scheme the lateral
  boundary condition
• Improve the sinking of Mediterranean waters by
  implementing
• - partial steps
• - a Bottom Boundary Layer (Beckman Döscher,
  1997)
• Improve the deep convection with a K-Profile
  Parameterisation (Large
• et al, 1998) together with a Gent et Mac
  Williams (1990) parameterisation

Three-year averaged transports accross the
straits superposed on the barotropic
streamfunction over the third year.

• References
• Madec G., Delecluse P., Imbard M., Lévy C.,
  1998 OPA8.1 ocean general circulation model
  reference manual, Notes du pôle de
  modélisation IPSL, 11.
• Smith W., Sandwell D., 1997 Global sea floor
  topography from satellite altimetry and ship
  depth soundings, Science, 277.
• Reynaud T., Legrand P., Mercier H., Barnier B.,
  1998 A new analysis of hydrographic data in the
  Atlantic and its application to an inverse
  modelling study, International WOCE Newsletters,
  32.
• Chassignet E.P. and Garraffo Z.D., 2001
  Viscosity parameterization and the Gulf Stream
  separation. In "From Stirring to Mixing in a
  Stratified Ocean". Proceedings 'Aha Huliko'a
  Hawaiian Winter Workshop. U. of Hawaii. January
  15-19, 2001. P. Muller and D. Henderson, Eds.,
  37-41.
• Smith R. D., Maltrud M.E., Bryan F.O. and Hecht
  M.W., 2000 Numerical Simulation of the North
  Atlantic Ocean at 1/10º. J. Phys. Oceanogr.,30,
  1532-1561.
• Beckmann A. and Dörscher R., 1997 A method for
  improved representation of dense water spreading
  over topography in geopotential-coordinate
  models. J. Phys. Oceanogr., 27, 581-591.
• Large W., Chassignet E.P., Verron J., 1998
  Modeling and parameterization ocean planetary
  boundary Layers, Ocean Modeling and
  Parameterization, Kluwer Academic Publishers,
  81-120