Internal Tides in the WeddellScotia Confluence Region, Antarctica - PowerPoint PPT Presentation

Title: Internal Tides in the WeddellScotia Confluence Region, Antarctica


1
Internal Tides in the Weddell-Scotia Confluence
Region, Antarctica Susan L. Howard, Laurence
Padman, and Robin D. Muench
Introduction Recent observations, backed up by
3-D model simulations of tides, highlight the
role of internal tide generation over ridges as a
source of velocity variability and ocean mixing.
Most research is on low- and mid-latitude ridges.
Here, we study this process in a high-latitude
environment where stratification is significantly
weaker than elsewhere. We focus on the
Weddell-Scotia Confluence along the South Scotia
Ridge (SSR), which was the focus of the Deep
Ocean Ventilation Through Antarctic Intermediate
Layers (DOVETAIL) experiment Muench and Hellmer,
2002. Water mass mixing and air-sea interaction
in the northern Weddell Sea and along the SSR
influence the properties of the dense water
escaping from the Weddell Sea into the World
Ocean. We propose that internal tides generated
at the SSR could explain some of the observed
velocity shear and mixing during the DOVETAIL
field program Muench et al., 2002.
For more information, contact Susan
Howard Earth Space Research http//www.esr.org h
oward_at_esr.org (206) 726-0501 x15
Energy Conversion and Depth-Dependence of
Baroclinic Velocity and Displacements

• Conclusions
• Internal tides are generated at the South Scotia
  Ridge and propagate south into the Powell Basin
  and northern Weddell Sea, and north into the
  Scotia Sea. Energy flux is a factor of 3 lower
  than we get using the Morozov 1995 ridge
  generation model.
• Vertical displacements due to baroclinic waves
  can exceed 100 m peak-to-peak.
• The region in which surface tidal currents are
  significant is much more extensive around the
  ridge than is shown by purely barotropic models.
• Tidal shear, strain and divergence acting on ice
  are much larger when baroclinicity is included,
  relative to barotropic-only models.
• The model supports the patchiness of mixing
  observed in the Powell Basin during DOVETAIL
  Muench et al., 2002.

Transect of velocity
Model density profile
Energy Flux from the barotropic to the baroclinic
tide (M2 only)
Transect of vertical displacement
68oS
58oS
Weddell Powell
Scotia Sea Basin
Sea
SSR
Weddell Powell
Scotia Sea Basin
Sea
SSR

• Our Hypotheses
• Internal tides are generated through interaction
  between barotropic tidal currents and the
  irregular and steep seafloor topography of the
  SSR.
• These internal tides are sufficiently energetic
  to affect winter sea ice properties through
  shear, strain, and divergence.
• Locally high mixing in the main pycnocline might
  result from internal wave shear.

Figures show energy flux magnitude (color scale)
and direction (arrows). Most generation occurs
across the SSR north of Powell Basin. Little
generation occurs at the continental slopes
surrounding the South Orkney Plateau and
Antarctic Peninsula, supporting the prevailing
view that ridges rather than continental slopes
contribute most internal tide energy. Maximum
fluxes are 200 W m-2, and the ridge average is a
factor of 3 smaller than Morozov 1995
predictions.
South-to-North slice of the baroclinic component
of northward velocity (v) for the transect shown
in the figure to the left. Well away from the
ridge system, wavelengths agree with ray tracing
Muench et al., 2002. Near the ridge system,
currents are complicated as multiple generation
sites contribute to the modeled velocities.
Little energy escapes into the northern Weddell
Sea.
Profile of s? used to initialize model runs. Most
stratification is near the surface, between 100
and 300 m. Stratification at the depth of the
ridges (gt1000 m) is weak.
Vertical displacement of isopycnals due to M2
tides. Maximum values exceed 50 m, including in
the strong stratification over the crest of the
SSR.

• Future Work
• The results reported here are preliminary. To
  fully understand baroclinic tides in this region,
  we need to make new runs with
• Updated (Smith and Sandwell) bathymetry
• Realistic, varying stratification, including
  seasonal variation in the upper ocean
• Forcing from multiple tidal constituents
  simultaneously
• Increased resolution, from our 4 km to 1-2 km.

Model Setup
Surface Current Fields
Mixing
Model runs were made using POM, which is a 3-D,
primitive equation model. The simulations are
carried out in a similar way to Merrifield et al.
2001. We consider only the M2 tidal
constituent, although other runs indicate a
similar response for S2. We have not yet
simulated diurnal tides in this region. Our model
domain is shown below.
Upward heat flux through the pycnocline In
Muench et al. 2002 we estimated mean upward
heat fluxes of 4 W m-2 based on the ridge
generation of Morozov 1995 (500 W m-1 for M2
only), and an arbitrary 1000 km decay scale. The
present model suggests that much less energy is
generated along the ridge than this (100 W m-1),
but most of this energy remains trapped within
200 km of the ridge, hence the predicted mean
heat flux is about the same, 4 W m-2 near the
ridge, with peak fluxes of gt20 W m-2. The
patchiness predicted in Powell Basin by the POM
model is consistent with DOVETAIL
observations. Additional mixing sources? The SSR
topography is extremely rough. Our present model
bathymetry is relatively smooth, interpolated
from our CATS grid (1/4o x 1/12o 10 km) which
is based on ETOPO-5. Higher resolution
bathymetry (e.g., Smith and Sandwell) may provide
much more internal tide generation, perhaps of
high modes see Polzin et al., 1997. Increased
mixing associated with such baroclinic waves will
dilute the dense water outflows through deep
passages through the SSR Naveira Garabato et
al., 2002, one of the paths for the Weddell Sea
contribution to the Global Ocean.
Currents M2 major axes
Divergence
(a)
(a)
References Merrifield et al., 2001 The
generation of internal tides at the Hawaiian
Ridge. Geophys. Res. Lett., 28, 559-562. Morozov,
E.G., 1995 Semidiurnal internal wave global
field. Deep-Sea Res., 42, 784-791. Muench, R.D.,
H. Hellmer, 2002 The international DOVETAIL
program. Deep-Sea Res. II, 48, 4711-4714. Muench
et al., 2002 Upper ocean diapycnal mixing in the
northwestern Weddell Sea. Deep-Sea Res. II, 48,
4843-4861. Naveira Garabato, A.C., et al., 2002
On the export of Antarctic Bottom Water from the
Weddell Sea. Deep-Sea Res. II, 48,
4715-4742. Padman, L., and C. Kottmeier, 2000
High-frequency ice motion and divergence in the
Weddell Sea. J. Geophys. Res., 105 (C2),
3379-3400. Polzin, K.L., et al., 1997 Spatial
variability of turbulent mixing in the abyssal
ocean. Science, 276, 93-96.
(b)
(b)
The model is set up as follows Resolution
1/16 o Longitude x 1/25o Latitude (mean
spacing 3-4 km) 41 sigma levels. Bathymetry
Modified Etopo5. Forcing M2 tidal forcing
only, normal flow at open boundaries. OBC
On all open boundaries, a flow relaxation
scheme in a 10-point sponge layer is
applied to the baroclinic u and v velocities
as well as T and S. Run time 15 days. The edge
of the sponge layer is shown on maps by the
dashed line.
Instantaneous divergence of (a) M2 depth-averaged
current, UBT(M2) and (b) M2 surface current,
US(M2). Area shown is indicated by boxes on plot
to the left. Maximum values are much higher when
baroclinicity is included, because the spatial
scales of the internal waves are much less than
the barotropic variability.
Major axis of (a) M2 depth-averaged current,
UBT(M2) and (b) total M2 surface current, US(M2).
Maximum values are much higher when baroclinicity
is included, and the area affected by tides
expands far out from the ridge.
Acknowledgements The work reported here was
carried out with support from National Science
Foundation grants OPP-9527667 and OPP-9615525 to
Earth Space Research, and is a contribution to
the international DOVETAIL program.