STATUS OF MULTI-SST ANALYSIS

1
STATUS OF MULTI-SST ANALYSIS

• Eileen Maturi, NOAA/NESDIS/STAR
• Andy Harris, University of Maryland, CICS

The Eighth Science Team Meeting of the Global
Ocean Data Assimilation Experiment High
Resolution Sea Surface Temperature Pilot Project
(GHRSST-PP) Bureau of Meteorology Head Office,
700 Collins St, Docklands, Melbourne, Australia
-Monday 14 May to Friday 18 May, 2007
2
MULTI-SST ANALYSIS
3
MULTI-SST Analysis

• Daily product accumulated from Initialization
  data
• Analysis can be generated every 3 to 6 hours
• Software has been adapted to use different
  spatial resolutions
• (11km and 5km)

4
INTEGRATED PRODUCTS TEAM

• DEVELOPERS
• Jo Murray, Rutherford Appleton Laboratory,
  England
• Paul Fieguth, University of Waterloo, Canada
• PRINCIPLE INVESTIGATORS
• Eileen Maturi
• Andy Harris
• John Sapper, NESDIS/OSDPD
• CONTRACTORS
• Liqun Ma, Perot Systems Corporation
• Heng Gu, SP Systems
• VALIDATORS
• Richard Reynolds, NCDC
• Ken Casey, NODC
• USERS
• CoastWatch Regional Managers of the U.S.

5
OUTLINE

• Objective
• Methodology
• Preliminary Validation
• Summary

6
OBJECTIVE

• Develop estimation scheme for combining
  multi-satellite retrievals of sea surface
  temperature into a single analysis
• Apply complementary SST datasets available from
  polar orbiters, geostationary IR and microwave
  sensors
• Use the computing power available to implement
  this estimation scheme

7
METHODOLOGY

• Initial guess of SST background field
• Initial guess of SST variability
• Observations with well-characterised errors
• Definition of relationship between observational
  datasets (i.e. assume one or more bias terms
  which are spatially correlated)
• Data Quality Control
• Daily SST Update
• Data Error Characterization
• Correlation Map
• Boundaries of Ocean Basins
• Derived Correlation Length Scale

8
OUTPUT

• Daily blended SST analysis at 1/10grid spacing
  using a equal-angle projection
• Original output is MATLAB binary format
• Converted to HDF
• Images of the data will also be generated
• Selected 5-km regions can be generated as
  required

9
Tropical Instability WavesDepicted by SST
Analysis
Click on image to play
10
POES/GOES SSTRegional Analysis-East Coast
Click on image to play
11
VALIDATION

• Evaluated against the RTG_SST ½½ resolution
  (also planning validation against 1/121/12)
  operational NCEP product
• ¼¼ daily OI (with and without MW)
• Screening out bad SST data prior to analysis
• ( use microwave SST (AMSR-E) or other
  independent data)
• Will perform traditional in situ data e.g. buoys
• Ftp web site to view the product for validation

12
Comparisons in Baja, CA
13
More RTG TMI comparisons
14
Validation vs. RTG Analysis TMI
15
SUMMARY

• Pre-operational product at NOAA/NESDIS
• Inputs NOAA-17/18 SST and GOES-11/12 SST
• Validation Process will begin this summer
• NCEP/Ocean Prediction Center
• CoastWatch Regional Managers
• GHRSST
• Operational product January 2008

16
MULTI-SST ANALYSIS
WEB SITE http//www.orbit.nesdis.noaa.gov/sod/ss
t/blended_sst
17
FUTURE PLANS

• SST retrievals from non-NOAA satellites
• Microwave e.g. AMSR-E, WindSat
• Most usefulness is when can make measurements
  along the coasts.
• NESDIS will analyze all satellite SST datasets
  using this methodology to produce a single best
  estimate global analysis

18
REFERENCES

• Fieguth, P.W. et al., Mapping Mediterranean
  altimeter data with a multiresolution optimal
  interpolation algorithm'', J. Atmos. Ocean Tech,
  15 (2) 535-546, 1998.
• Fieguth, P., Multiply-Rooted Multiscale Models
  for Large-Scale Estimation, IEEE Image
  Processing, 10 (11), 1676-1686, 2001
• Khellah, F., P.W. Fieguth, M.J. Murray and M.R.
  Allen, Statistical Processing of Large Image
  Sequences'', IEEE Transactions on geoscience and
  remote sensing, 14 (1), 80-93, 2005