CLIVAR/GODAE Synthesis Evaluation

1
CLIVAR/GODAE Synthesis Evaluation
Tony Lee, JPL

• Item5 volume transport, temperature salinity
  fluxes through key straits
• Indonesian throughflow (ITF)
• ACC at Drake Passage
• Florida Current at Florida Strait

2
Some details about legends
3
ITF
4
ITF volume transport (Sv)
Total
Spinup problem
Island Rule doesnt work? Or effect of
geometry/topography?
Anomaly relative to 1993-2001 mean
Remarkable consistency
5
ITF volume transport from 1992 (Sv)
Total
Anomaly relative to 1993-2001 mean
6
ITF temperature flux (PW)
Total
Anomaly relative to 1993-2001 mean
7
ITF temperature flux from 1992
Total
Anomaly relative to 1993-2001 mean
8
ITF salinity flux
Total
Anomaly relative to 1993-2002 mean
9
ITF salinity flux from 1992
Total
Anomaly relative to 1993-2001 mean
10
1993-2001 averaged seasonal anomaly of ITF
11
ITF property flux (volume, heat, and salt)
estimated by various products are overall
consistent (for mean as well as for seasonal,
interannual, and decadal variability e.g., large
magnitude of seasonal cycle relative to mean
value, low ITF during El Nino, lack of decadal
variability.
12
ACC
13
ACC volume transport
Total
Anomaly relative to 1993-2001 mean
14
ACC volume transport from 1992
Total
Anomaly relative to 1993-2001 mean
15
ACC temperature flux
Total
No such large adjustment in volume transport
Relatively consistent decadal variability
Anomaly relative to 1993-2001 mean
16
ACC temperature flux from 1992
Total
Anomaly relative to 1993-2001 mean
17
ACC salinity flux
Total
Anomaly relative to 1993-2001 mean
18
ACC salinity flux from 1992
Total
Anomaly relative to 1993-2001 mean
19
1993-2001 averaged seasonal anomaly of ACC
20
Relatively large difference in mean and seasonal
variability, but relatively consistent on decadal
time scale.
21
Florida Current at Florida Strait
22

• Two different ways of computing Florida Current
  (FC) transport by various groups
• meridional transport near 26.5N between Florida
  (80W) to 75W (more consistent with cable
  measurements),
• zonal transport between Florida and Cuba (better
  defined in models).
• The time mean of (2) is significantly smaller
  than that of (1) because a portion of (1) comes
  through a short circuit north of the Caribbean
  instead of from the loop Current (esp. in models)

23
FC volume transport
Total
Grey curve cable data
Anomaly relative to 1993-2001 mean
24
FC volume transport from 1992
Total
Grey curve cable data
Anomaly relative to 1993-2001 mean
25
Zonal (upper) and meridional (lower) Florida
Current transports
Zonal between Florida Cuba (well-defined)
Meridional 26.5N, 80-75W (near cables)
Grey curve cable data
26
FC temperature flux
Total
Anomaly relative to 1993-2001 mean
27
FC temperature flux from 1992
Total
Anomaly relative to 1993-2001 mean
28
FC salinity flux
Total
Anomaly relative to 1993-2001 mean
29
FC salinity flux from 1992
Total
Anomaly relative to 1993-2001 mean
30
1993-2001 averaged seasonal anomaly of Florida
Current
31
Relatively large spread, partly because of
definition problem. In most models, the flow
going into the Gulf of Mexico is weak some flow
take a short cut north of Cuba to joint the
Florida Current.
32
Summary Statistics
33
1993-2001 mean transports/fluxes For Florida
Current black - meridional transport near 26.5N,
80-75W red zonal transport between Florida
Cuba.
ITF FC ACC ITFt ITFs FCt FCs ACCt ACCs
Sv Sv Sv PW PW PW Kg/s Kg/s Kg/s
K7 -12.4 157
CRFS-A -12.9 16.6 147
CRFS-2 -13.2 24.2 147
ECCO-50yr -11.0 29.3 129 -1.0 -0.4 2.1 1.1 1.4 4.5
ECCO-SIO -10.7 30.6 154 -0.9 -0.4 2.2 1.1 1.6 5.3
ECMWF -10.6 14.8 148
GFDL -15.7 24.3 171 -1.0 -0.6 1.6 0.8 1.7 6.1
INGV -11.9 24.6 127 2.8 0.9
MOVE-G -23.4 179
ECCO-JPLKF -13.2 27.6 132 -1.2 -0.5 2.3 1.0 1.4 4.6
MCTR 3DVAR -18.7 15.0 142
MCTR KF -12.8 21.8 149
ECCO-GODAE -11.1 26.6 148 -0.9 -0.4 1.9 1.0 1.5 5.1
SODA -14.5 27.5 158 -1.2 -0.5 2.1 1.0 1.7 5.6
Obs. Analysis -10 -15 31.9 120160
34
The largest mean ITF (gt 23Sv) consistent with
Island rule (integration of South Pacific curl)?
Or effect of geometry/topography? High mean ACC
transports (170-180 Sv) too strong zonal wind?
lack of form drag due to smooth topography? Small
viscosity?
35
1993-2001 rms variability of volume transport,
rms difference among models, and rms difference
from cable measurements
(all in Sv) ITF FC ACC rms dev. of FC from cable data
K7 3.5 7.0
CRFS-A 5.0 4.7 6.4 4.9
CRFS-2 5.2 4.7 6.2 5.3
ECCO-50yr 2.4 1.6 4.1 2.9
ECCO-SIO.it69 2.5 2.2 4.6 3.1
ECMWF 3.0 2.2 4.0 3.0
GFDL 2.5 2.8 5.0 3.6
INGV 3.1 5.6 3.5 5.5
MOVE-G 4.0 6.3
ECCO-JPL RTS smoother 3.0 3.0 4.1 3.4
MCTR 3DVAR 5.5 5.1 6.6 5.3
MCTR KF 3.2 1.4 4.6 2.8
ECCO-GODAE.v2.199 2.6 2.2 4.4 3.3
SODA 4.3 2.6 4.5 5.0
rms diff. among models 2.3 3.0 3.5
Measurement 2.5