An observing System Simulation Experience OSSE for SMOS

1
An observing System Simulation Experience (OSSE)
for SMOS

• Results of simulating data SMOS assimilation

2
Study plan

• Objectives of the study
• Model and assimilation system used
• Description of the OSSE
• Impact validation
• Physical results
• Prospects
• Conclusions

3
I.1 Objectives of the study

• Today Little Sea Surface Salinity (SSS)
  measurement
• Depends of commercial boats
• Salinity has NEVER been measured over 24 of the
  ocean (see next slide).
• Importance of Sea Surface Salinity (SSS)
• Salinity is key to understanding interactions
  among the ocean, climate, and water cycle
• Influence on the ocean salinity, temperature and
  density, and also ocean stratification
• Mean Mixed Layer key role on the exchange
  ocean/atmosphere
• Barrier Layer Importance on the mixing near
  fresh water flux.
• A good understand of the ocean physical needs a
  good knowledge of SSS

4
I.2 Objectives of the study
End of 2007 launch of a new satellite SMOS
(Soil Moisture Ocean Salinity)
global cover of the SSS accuracy of 0.1 psu for
a 10-30 day average for an open ocean area of 200
x 200 km We have assimilate Simulating SMOS data
in order to study the impact of this future
knowledge
Simulated seasonal (winter) sea-surface salinity
map
5
II.1 Model and assimilation system used the
model

• Ocean Global Circulation Model  OPA8.1 (Madec
  Imbard, 1996)
• Configuration
• MNATL North Atlantic and Tropics (-20S 70N
  98.5W, 20W)
• Resolution 1/3
• Time step 30mn
• Forcing every day
• Ice is diagnostic

6
II.2 Model and assimilation Data System used

• Data assimilated
• Satelital
• Altimetry (SSH)
• SST NOAA15, NOAA16, NOAA17 SST Reynolds
• In situ
• Measurement
• Drifter
• Voluntary and oceanographical vessels
• System used SAM2 (System dAssimilation
  Mercator 2) with a filter SEEK Kalman filter
  derivated

7
III.1 Description of the OSSE SMOS Data's
simulations

• Interpolated Levitus Climatology 2001 everyday
• Integration all the 7 days
• Two differents errors
• Constant
• Variable more realistic

8
III.2 Description of the OSSE Three simulations
Validation of the SSS assimilation impact REF
vs. SMOS_CTE Study the importance of the error
and of the assimilation sensibility REF vs.
SMOS_VAR
SMOS_VAR is a better representation than SMOS_CTE
of the future data SMOS !
9
IV.1 Impacts validationSalinity field
10
IV.2 Impacts validationTemperature field
11
IV.3 Impacts validation Effects on the currents
12
V.1 Physical results Sea Surface Salinity
13
V.2 Physical results Impact on the EKE
14
V.3 Physical results The Mean Mixed Layer
15
V.4 Physical results The barrier Layer
16
VII. Conclusions

• Mercator can assimilate the futur data SMOS
• This study showed that SSS has an impact on
• Thermohalins field and currents
• Mean mixed layer and barrier layer
• But data assimilated hasnt the annual
  variability signal !
• Assimilate SSS is important to
• Predict the ocean state
• Simulate exchange between ocean and atmosphere

17
VI. Prospects

• A new study with data simulated by PSY2v1 more
  realistic
• Paper about the results ?
• Newsletter Mercator ?