Argo Floats

1
Argo Floats and
Modelling the North Pacific Ocean
Michael W. Stacey Department of Physics Royal
Military College of Canada Kingston, Ontario,
K7K 7B4 Stacey-m_at_rmc.ca 15 June 2006
2
Major Collaborators Jennifer Shore (RMC) Dan
Wright (BIO) Keith Thompson (Dalhousie) Howard
Freeland (IOS) Bill Crawford (IOS)
3
Motivation Meteorologists have provided regular
weather forecasts for years, using atmospheric
observations and numerical models. Observations
of the sub-surface ocean are much harder to make,
but oceanographers are now at the stage where
they can begin to make comprehensive observations
in almost real-time. These observations are
obviously useful in providing better information
about the state of the ocean. Also, they provide
useful information to scientists studying climate
change, since the oceans influence the climate.
4
Basic Requirements 1. Observations - Argo
Floats - Satellite observations - etc 2.
Computational Resources - HPCVL
5

• Outline
• Argo float program
• Numerical Model
• Spectral Nudging
• Model simulations of the North Pacific Ocean
• Zoom in on the Northeast Pacific Ocean
• Eddy formation and propagation in the NE Pacific
• Summary

6
Argo Floats (images from
http//www.argo.ucsd.edu/)
7
Argo floats record salinity, temperature and
pressure Approx 15K/Float, plus another 15K
for handling and running
8
- Each floats lasts for about 140 cycles, or 4
years. - Observations can be downloaded from the
Argo website free of charge.
9
- Each floats lasts for about 140 cycles, or 4
years. - Observations can be downloaded from the
Argo website free of charge.
10
(No Transcript)
11
Approximately 100,000 profiles/yr once 3000
deployed
12
Isopycnals are deepening in the NE Pacific
13
The Model
The Model Parallel Ocean Program (POP) -Finite
difference, numerical model, parallelized at Los
Alamos, modified at the Bedford Institute of
Oceanography (Halifax) for the Atlantic Ocean,
modified at RMC for the North Pacific
Ocean. -Uses Spectral Nudging to prevent drift
of the mean model fields. -Run at HPCVL. A
single, twenty year simulation takes about 10
days of CPU time, using 20 processors.
14

• - Horizontal Resolution approx. 30 km
•  
• Vertical Resolution (23 layers)
• Layer Depths (m)
• Top 200 m Bottom 4000 m
• 1. 10 9. 260 16. 2200
• 2. 20 10. 360 17. 2700
• 3. 35 11. 510 18. 3200
• 4. 55 12. 710 19. 3700
• 5. 75 13. 985 20. 4200
• 6. 100 14. 1335 21. 4700
• 7. 135 15. 1750 22. 5200
• 8. 185 23. 5700
• -

15

• Spectral Nudging
• - Available observations of temperature and
  salinity are temporally averaged to obtain the
  observed climatological mean.
•  
• Want the model to have the same mean as the
  observations. Energy for the eddies comes from
  the energy in the mean flow, so need reasonable
  mean in order to get the eddy field right.
•  
• Models can be constrained to have the same
  mean as the observations by nudging them
  towards the climatology as the simulation
  proceeds through time, but because the observed
  mean is smooth, standard nudging suppresses
  the formation of eddies in models.
•  
• Spectral nudging constrains only the mean
  component of the model to
• remain close to the climatology, so eddies can
  form in the model.

16
Standard Nudging vs. Spectral Nudging
Elevation (cm)
Elevation (cm)
Standard nudging prevents model T and S from
drifting away from climatology but suppresses
eddy variability.
Spectral nudging constrains the mean T and S
fields while keeping higher frequency variations.
17
Sea Surface Height
18
Temperature at 100 m
19
Northeast Pacific (Gulf of Alaska)
20
Sea Surface Height
21
Temperature at 100 m
22
Thomson and Gower (1998)
Snapshot of the Sea-Surface Temperature off the
Coast of B.C.
Observation (Thomson Gower, 1998)
Model
- Eddy trains form during the winter. - The
eddies are about 200 km across and at least 500 m
deep. - They propagate towards the west at about
3 km/day.
23
Eddy Sizes and Speeds
24
Standard Deviation of the sea surface height
variability. Stacey et al (2006)
25
Crawford et al., (2000)
26
Summary - The Argo program has significantly
improved our ability to make observations of
the sub- surface ocean. It is a
multi-national program for which Canada is
a significant contributor. - HPCVL has
provided the computational resources for
simulations of the North Pacific Ocean to
take place at RMC.
27
Thank You
References Crawford, W. R., J. Y. Cherniawsky,
and M. G. G. Foreman (2000), Multi-year meanders
and eddies in the Alaskan Stream as observed by
TOPEX/Poseidon altimeter, Geophys. Res. Letters,
27, 1025-1028. Stacey, M. W., J. Shore, D. G.
Wright, and K. R. Thompson (2006), Modeling
events of sea-surface variability using spectral
nudging in an eddy permitting model of the
northeast Pacific Ocean, J Geophys. Res., 111,
C06037, doi10.1029/2005JC003278. Thomson, R.
E., and J. F. R. Gower (1998), A basin-scale
oceanic instability event in the Gulf of Alaska,.
J. Geophys. Res., 103, 3033-3040. Yelland, D.,
and W. R. Crawford (2005), Currents in Haida
Eddies, Deep-Sea Res. II, 52, 875-892.