Fall AGU Meeting (Dec. 2002) COAST abstracts:

• Allen
  
• Bane et al.
  
• Barth et al.
  
• Carney et al.
  
• Couch et al.
  
• Cowles et al.
  
• Dale et al.
  
• Durski et al.
  
• Egbert et al.
  
• Gan et al.
  
• Hales et al.
  
• Kosro et al.
  
• Kuebel et al.
  
• Kurapov et al.
  
• Perlin et al.
  
• Pierce et al.
  
• Ruttenberg et al.
  
• Samelson
  
• Spitz et al.
  
• Wheeler
  

Abstracts should be cited as:

EOS Trans. AGU, 83 (47),
Fall Meet. Suppl.,
Abstract XXXXX-XX, 2002

OS62A-0241

Comparison of Hydrostatic and Nonhydrostatic Models in the Coastal
Ocean

S Durski and J S Allen

Models employing the hydrostatic approximation have long been
the standard for numerical simulations of the coastal ocean. As
computational performance has improved and our understanding of
ocean physics developed, these models have been pushed to higher
resolution to uncover the role smaller scale phenomena play on
the mesoscale circulation. However as the vertical and horizontal
length scales of the processes of interest converge nonhydrostatic
effects which are neglected by these models become of first-order
importance. The dynamics at these scales can involve energetic mixing
which may significantly feedback into the larger scale circulation.
Examples include the breaking of internal waves and enhanced
mixing over small- scale topography. In this study we compare the
performance of a hydrostatic model to that of a non-hydrostatic
model in two-dimensional coastal ocean settings. The hydrostatic
model to be employed is ROMS. The non-hydrostatic model is a version
of Clark's cloud-scale meteorological model which has been adapted
for oceanic application. Both models are formulated in generalized
sigma-coordinates which make them favorable for simulating flow over
continental shelf topography. The two models are configured with
identical spatial resolutions to accommodate direct comparisons.
The generation and propagation of internal waves are compared
in simulations of wind and tidally forced stratified flow over
topography including a case representative of the circulation at
Stonewall Bank off the Oregon coast.  Shoreward propagation of large
amplitude internal waves established by an imposed displacement
of the pycnocline at model initialization is also explored. The
comparisons reveal both areas of parameter space where the solutions
are comparable and ones where they differ significantly. The most
apparent difference in these experiments rests in the inability of
the hydrostatic model to represent the balance between non-linear
and dispersive effects that arise in internal solitary waves.