AIP INPUT TO WGOMD GFDL, June 2004

1
AIP INPUT TO WGOMDGFDL, June 2004

• Dan Wright
• Wrightdan_at_mar.dfo-mpo.gc.ca

2
Stating the obvious
1. Model-model intercomparisons are useful to
determine both robustness and areas of
uncertainty.
2. We need checks against observations to
determine connections to reality. gt
need to check sensitivities to uncertain forcing.
3. Validated models can help fill holes in gappy
data. (Observationalists wont believe model
results just because they agree with each
other.)
3
Some Comments

• Observations have not (yet?) provided conclusive
  evidence of
• the role of the MOC in decadal climate
  variability. Real need
• for model results!

It would be very useful to get guidance from
ocean models on how to design an appropriate
sustained observation program for this purpose.

• Identification of indices of MOC (and heat flux)
  variability would
• be very useful, particularly if such an index
  could be obtained from
• historical sea level and hydrographic
  observations.

4

• Robust correlations under realistic forcing
  conditions between
• things that are of climatic relevance (e.g. MHF
  and tropical SST)
• and things that are more easily monitored
  (e.g., boundary current
• transports) are anxiously anticipated.

5
Some Questions

• Decadal Oscillations
• What controls the damping timescale of decadal
  oscillations?
• Are the processes reliably represented?
• How important are nonlinear effects in
  determining
• decadal variations? (Dewar, J. Clim.,
  2001)

• Large-scale freshening
• Sensitivity to source of freshening
• Canadian Archipelago throughflow
• East Greenland Current
• Denmark Strait Overflow
• NE Alantic Deep Water
• Surface fluxes

6

• Robustness
• Model intercomparisons suggest robust OT
  variability.
• Great!

• Watermass properties
• Can model comparisons identify the root cause of
  biases?
• Do different models agree on the causes?
• Can biases be reliably reduced?

• Deep conditions
• How well do models represent overflows and deep
  flows?
• How steady or intermittent are these flows?
• Where are the best monitoring locations?

Why is Gulf Stream separation so different in
different models? -- are improvements reliable?
7
(No Transcript)
8
Restating the obvious
We need checks against observations to
determine connections to reality.
9
Long Time Series?
10
The depth of the ?t28.0 isopycnal in the
upstream reservoir of the Norwegian Sea
controls the pressure head that drives overflow
through the Faroe-Bank Channel (Hansen et al
2001). Observations show that this head is
changing.

11
Volumetric T-S census of the northwest corner of
the North Atlantic
12
(No Transcript)
13
Time Variations T-S in Labrador Sea
Observations
Model
T
S
14
Western Boundary Transports at 42N Clarke,
Hendry Yashayaev (1998)
15
CFCs in the Atlantic Overturning
Circulation Rhein, Fischer, Smethie,
Smythe-Wright, Weiss, Mertens, Min, Fleischmann
Putzka, JPO, 2002 Andrie, Rhein, Freudfenthal
Plahn, DSR I, 49(2), 2002 Andrie, The South
Atlantic, 1996
CFC-11 inventory in the subpolar gyre
Increase over the 90s of
1983 1993
CFC-11 in DWBC at the Equator
16
Summary of Proposed Reality checks

• Spatial and temporal variation of tracers

• Variations in sea ice and relation to SST
• Response to the Great Salinity Anomaly(ies)

• Major gyre transports (Curry and McCartney)
• Upper ocean shelf-slope transports (Dickson and
  Yashayaev)

• Transport through A2/AR19 section (Grand Bank to
  England)
• Florida Current transport
• Transport through full A5/AR1 section (24N)
• Sections across 10N (MOVE/GAGE), 10S, 30S

17
Reality Checks (contd)

• Drainage of Greenland Sea deep water
• Production and drainage of Labrador Sea (and
  deeper) Waters
• Watermass pathways into the South Atlantic

• The magnitude and structure of the ocean warming
  over past
• 50 years revealed by Levitus et al. (Science,
  2000)

• The spatial structure of the changes in Atlantic
  fresh water
• content revealed by Curry, Dickson and
  Yashayaev (Nature, 2003)