MERSEA GODAE Intercomparison project for the Mediterranean sea

1
MERSEA - GODAEIntercomparison project for the
Mediterranean sea

• Laurence Crosnier (Mersea, France)
• Christian Le Provost (Mercator, France)

2
GOAL Evaluate the strength/weakness of 3
European 2 US Ocean monitoring and forecasting
systems.
3
Metrics have been defined

• Fields provided Daily mean best estimate
  day(T06) forecasts.
• Time period from JUNE 2003 on.
• CLASS1 T,S,U,V,SSH,MLD,BSF,TX,TY,Qtot,E-P-R.
• Interpolation (non conservative) on 1/8
  horiz. grid with 12 vert. levels in ATL and 8
  vert. levels in MED.
• CLASS2 High resolution(T,S,U,V)
  sections/moorings.
• CLASS3 Integrated quantities daily transports
  
  
  
  
  through sections/MHT/Overturning.
• CLASS4 Test performance of analysis and
  forecasts.
• COMPARISON model/model, model/climatology,
  model/insitu obs.

4
CLASS 2 (T,S,U,V) SECTIONS
5

• Data stored on OPENDAP servers
• http//opendap.mercator-ocean.fr/dodsC/
• http//www.sincem.unibo.it8080/dodsC/
• http//www.nerc-essc.ac.uk9090/
• http//mersea.nersc.no/dodsC/
• http//hycom.rsmas.miami.edu/dodC/
• http//NLOM....
• GRAPHIC TOOLBOX
• LAS (Live Access Server)
• DODS/MATLAB or DODS/IDL command line client
• http//www.unidata.ucar.edu/
• gtgt loaddods('http//www.nerc-essc.ac.uk9090/FOAM_
  NATL_120_8th_ARC?temperature00')
• Reading http//www.nerc-essc.ac.uk9090/FOAM_NATL
  _120_8th_ARC
• Constraint temperature00
• Server version catalogaggserver/0.8
• Creating matrix temperature (1 x 1 x 441 x 869)
  with 383229 elements.
• Creating vector time with 1 elements.
• Creating vector depth with 1 elements.

6
October mean Zonal Wind Stress (Tx) (ci0.01Pa)
MERCATOR
TOPAZ
FOAM
7
September mean Zonal Wind Stress (Tx) (ci0.01Pa)
MERCATOR
TOPAZ
FOAM
8
October mean Meridional Wind Stress (Ty)
(ci0.01Pa)
MERCATOR
TOPAZ
FOAM
9
October mean Qnet (ci20W/m2)
MERCATOR
TOPAZ
FOAM
10
October mean Fresh Water Flux (ci0.2 e-5kg/m2/s)
11
OCTOBER mean SSH (ci0.05m) (BASIN MEAN SSH
SUBTRACTED)
MERCATOR
TOPAZ
FOAM
12
October mean Barotropic Stream Function (ci2Sv
or .1Sv)
MERCATOR
TOPAZ
FOAM
13
October mean Mixed layer Depth (m) ci5 m
14
October mean potential temperature at 5 m depth
(ci.5degC)
LEVITUS01
15
October mean salinity at 5m depth (ci0.2psu)
MERCATOR
TOPAZ
LEVITUS01
FOAM
16
October mean potential temperature at 200 m depth
(ci0.2degC)
MERCATOR
TOPAZ
LEVITUS01
FOAM
17
October mean salinity at 200m depth (ci0.1psu)
MERCATOR
TOPAZ
LEVITUS01
FOAM
18
October mean potential temperature at 1000 m
depth (ci0.05degC)
MERCATOR
TOPAZ
LEVITUS01
FOAM
19
October mean salinity at 1000m depth (ci0.05psu)
MERCATOR
TOPAZ
LEVITUS01
FOAM
20
October mean potential temperature at 2000 m
depth (ci0.05degC)
MERCATOR
TOPAZ
LEVITUS01
FOAM
21
October mean salinity at 2000m depth (ci0.025psu)
MERCATOR
TOPAZ
LEVITUS01
FOAM
22
SEPT2003 TEMPERATURE SECTION35-40N
MERCATOR
MFS
FOAM
23
SEPT2003 SALINITY SECTION35-40N
MERCATOR
MFS
FOAM
24
EKE SECTION35 - 40N
FOAM
MERCATOR
MFS
25
SEPT2003 TEMPERATURE SECTION39-35N
MERCATOR
MFS
FOAM
26
SEPT2003 SALINITY SECTION39-35N
MERCATOR
MFS
FOAM
27
EKE SECTION39-35N
MERCATOR
MFS
FOAM
28
SEPT2003 TEMPERATURE SECTION34-5.5E
MERCATOR
MFS
FOAM
29
SEPT2003 SALINITY SECTION34-5.5E
MERCATOR
MFS
FOAM
30
EKE SECTION34-5.5E
MFS
FOAM
MERCATOR
31
SEPT2003 TEMPERATURE SECTION41-19E
MERCATOR
MFS
FOAM
32
SEPT2003 SALINITY SECTION41-19E
MERCATOR
MFS
FOAM
33
EKE SECTION41-19E
MFS
MERCATOR
FOAM
34
SEPT2003 TEMPERATURE SECTION43 -28E Rhodes gyre
MERCATOR
MFS
FOAM
35
SEPT2003 SALINITY SECTION43 -28E Rhodes gyre
MERCATOR
MFS
FOAM
36
EKE SECTION43 -28E Rhodes gyre
MFS
MERCATOR
FOAM
37
Conclusions

• CLASS1/CLASS2 data analysis enable to describe
  the agreement and differences between system
  outputs.
• With 3 systems available in the MED basin, Its
  easy to pinpoint problems in one system, as
  usually 2 of the systems agree in front of a
  third one.
• CLASS2 higher resolution sections allow more
  quantitative and finer diagnostics than CLASS1
  fields.
• Systematic comparison to In situ Independent
  Data set ( T/S data sets, TOPEX SLA, moorings,
  tide gauge) will allow better understanding of
  the differences and similarities of each systems.
• Other diagnostics soon to be done - (T/S)
  diagram, velocity arrow plot, surface EKE.
• CLASS3 (volume and heat) transport data are
  still needed.
• CLASS 4 (forecast performance evaluation) m
  ethodology agreed basic reference data sets
  need to be fixed
• Methodology for a synthesis of these
  intercomparison products ? Next workshop
  Bologna, 23 24 feb. 2004