3D, nonhydrostatic numerical simulation of nonlinear internal wave generation in the South China Sea

1
3D, nonhydrostatic numerical simulation of
nonlinear internal wave generation in the South
China SeaOliver FringerCollaborator Steve
Ramp (MBARI) Ph.D. Student Zhonghua Zhang
Environmental Fluid Mechanics Laboratory
Stanford University, Stanford, CA10 June 2009
Support ONR grant N00014-05-1-0294
TexPoint fonts used in EMF. Read the TexPoint
manual before you delete this box. ??A?????
2
SCS NLIWI with SUNTANS

• Resolution
• 1 5 km horizontal
• 10 150 m vertical
• 12 million grid cells
• Initial conditions
• quiescent velocity
• horiz. uniform stratificationfrom observations
• Boundary conditions
• Tidal currents from OTIS TPXO6.2
• Bathymetry
• 1800 m resolution Ocean Data Bank of the National
  Center for Ocean Research, Taiwan.
• Simulation time
• 14 days real 7 days wall on 64 procs11,000 cpu
  hours on ARL JVN/MJM

China
Taiwan
B1
500 x 300 km
3
Temperature field at a depth of 60 m (17-30 June
2005)
4
Free-surface signature
A
B
5
(No Transcript)
6
(No Transcript)
7
(No Transcript)
8
(No Transcript)
9
(No Transcript)
10
(No Transcript)
11
(No Transcript)
12
A
B
13
(No Transcript)
14
(No Transcript)
15
(No Transcript)
16
(No Transcript)
17
(No Transcript)
18
(No Transcript)
19
(No Transcript)
20
A
B
21
(No Transcript)
22
(No Transcript)
23
(No Transcript)
24
(No Transcript)
25
(No Transcript)
26
(No Transcript)
27
A
28
(No Transcript)
29
(No Transcript)
30
B1
31
A waves at B1
Observation (S. Ramp, MBARI)
Prediction

• Good phase prediction, bad amplitude prediction
• Bathymetry
• Grid resolution

32
28 June 2005, 1220
27 June 2005, 1130
28 June 2005
B
A
30 June 2005, 1401
29 June 2005, 1311
Generation is strongly influenced by
three- dimensional topography.
Apparent sources inferred from wave fronts are
not aligned with sills in Luzon. A-wave
source is at 120.5oE.
33
Generation through superposition
34
Is the nonhydrostatic pressure necessary?
Nonhydrostatic code
Isotherms 16, 20, 24, 28 degrees C
Simulation time 2X
Hydrostatic code
Simulation time X
35
Nonhydrostatic effects
Hydrostatic
Nonhydrostatic
both at 1230 am on 25 Jun 2005

• Nonhydrostatic simulation produces more realistic
  results because of the influence of
  nonhydrostatic dispersion which has a
  regularizing effect.

36
Comparison to SAR imagery
37
Summary

• Apparent sources of waves may be a result of
  superposition of multiple wave fronts.
• Phase is predicted well but this is due to
  accurate prediction of mostly linear phenomenon
  of internal tidal propagation in deep basin.
• Onshelf propagation is more nonlinear and
  nonhydrostatic, and hence nonhydrostatic code
  produces more realistic dispersive behavior of
  waves on shelf.
• Hydrostatic code will produce steeper waves that
  decay faster and hence have smaller amplitude
  may alter phase prediction by O(hours).
• To predict peak amplitude of waves at B1 and
  onshelf, resolution of O(100 m) and better
  bathymetry are required.