Also known as CMIS

1
WindSat --- the New Competition

• Also known as CMIS

R. A. Brown 2005 LIDAR Sedona
2
Radiometers
Passive Radars
Basic Concepts for The Radiometer
R. A. Brown 2005 LIDAR Sedona
3
Same principal as Scatterometer but signal is
much weaker
Hence speed only from SMMR, SSMI,..
R. A. Brown 2003 U. ConcepciÓn
4
Solar reflectance Brightness Temperature
Two looks at the same spot
R. A. Brown 2004
5
What is Ocean Observer?

• Operational data for Navy and NOAA
• Science data for NASA and NOAA
• RD sensor proof of concept for NASA
• Operational transition for NASA and NOAA
• Team approach to solving mutual problems at
  for OMB
• Oceans mainly

reduced agency cost
6
NPOESS

• WindSat becomes CMIS

7
Primary Contributions to EDRs by Sensor
8
Joint IPO/DoD/NASA Risk Reduction Demo
WindSat/Coriolis
(A stealth mission)

• Description Measures Ocean Surface Wind Speed,
  Wind Direction, Using Polarimetric Radiometer on
  a Modified Satellite Bus, Launched Into a 830 km
  98.7 Orbit by the Titan II Launch Vehicle. 3
  Year Design Lifetime.

?
Launched January 2003
Data release Sept. 2004

• Capability/Improvements
• Measure Ocean Surface Wind Direction (Non-
  Precipitating Conditions). Two looks at same
  spot.
• 25km spatial resolution
• Secondary Measurements
• Sea Surface Temperature, Soil Moisture, Rain
  Rate, Ice, and Snow Characteristics, Water Vapor

R. A. Brown 2004
9
Neil Tysons address/campaignOn the Future of
NASA Jan 20, 2005
Presidents commission --- Vision
(thing)
Winners Space Exploration
Planetary Science
Astrobiology
Astrophysics Astronomy
Losers Einstein prerogatives
Earth Science
LEO (low earth orbits) are old hat and boring.
NASA must do new stuff space
R. A. Brown 2005 LIDAR Sedona
10
WindSAT Cal/Val with SLP Retrievals

• Ralph Foster, Applied Physics Laboratory, U. WA
• Jerome Patoux, R.A. Brown, Atmospheric Sciences,
  U. WA

R. A. Brown 2005 LIDAR Sedona
11
Outline

• Two questions
• How well does WindSAT perform when its working
  at its best?
• Can Sea-Level Pressure (SLP) fields help improve
  model function and ambiguity selection?
• Physics of SLP(U10)
• QuikSCAT example
• Methodology
• WindSAT results
• Comparison with ECMWF SLP Analyses QuikSCAT
  wind distributions
• Ambiguity selection procedure

R. A. Brown 2005 LIDAR Sedona
12
SLP from Surface Winds

• UW PBL similarity model
• Use inverse PBL model to estimate
  from satellite
• Use Least-Square optimization to find best fit
  SLP to swaths
• Extensive verification from ERS-1/2, NSCAT,
  QuikSCAT

(UGN )
(UGN )
R. A. Brown 2005 LIDAR Sedona
13
Surface Pressures
QuikScat analysis
ECMWF analysis
14
Surface Pressure as Surface Truth

• For good quality and consistent U10 input, SLP
  fields are a good match to ECMWF analyses
• SLP/Model-derived U10 is an optimally smoothed
  low-pass filtered comparison data set
• Wind-sensor derived product only
• Model U10 tend to agree with input U10 for good
  swath input
• If SLP fields are wrong, pressure gradients and
  hence U10 are wrong.

R. A. Brown 2005 LIDAR Sedona
15
Dashed ECMWF
16
Dashed ECMWF
17
All four swaths for both WindSAT and QuikSCAT
18
Results

• WindSAT is biased high for U10 lt 8 m/s
• Too few winds U10 lt 5 m/s
• Too many winds 5 lt U10 lt 8 m/s
• Implied grad(SLP) too high when U10 lt 8 m/s
• Implications for assimilation in NWP
• Too few WindSAT winds in 8 lt U10 lt 12 m/s
• Comparable to QuikSCAT 12 lt U10 lt 15 m/s
• SLP agrees better in higher wind regime
• Too small sample to assess higher winds

R. A. Brown 2005 LIDAR Sedona
19
Use SLP to Assess Direction

• Winds derived from SLP are optimal smooth winds
• Arbitrary threshold of 35o from Model U10 used to
  distinguish potentially wrong ambiguity choice
• Look for a WindSAT ambiguity with closer
  direction to Model winds in these cases

R. A. Brown 2005 LIDAR Sedona
20

• Noisy directions
• Front captured
• Changed ambiguities away from clouds low winds
  Why?

21
Conclusions

• There is a lot of wind vector information in the
  WindSAT swaths

• The agreement of the WindSAT-derived SLP fields
  with ECMWF is surprisingly good for a first-cut
  model function.
• Better in higher winds
• An improved model function will produce better
  SLP

• SLP can be used to assess and improve the
  WindSAT wind data

R. A. Brown 2005 LIDAR Sedona
22
Conclusions (cont.)

• SLP fields demonstrate that the current WindSAT
  model function often produces a poor wind speed
  distribution
• Wind speed distribution can be robustly evaluated
  with SLP
• Storm analyses will address high wind
  distribution

• Wind directions are noisy and there there is
  room for ambiguity selection improvement.
• SLP shows promise for this need

R. A. Brown 2005 LIDAR Sedona
23
Next

• SLP adds the robust ECMWF NCEP surface analyses
  and buoy pressure observations to the WindSAT
  Cal/Val data
• We are developing methods to use buoy/analysis
  pressures to identify correct deficiencies in
  model function, e.g. Zeng and Brown (JAM 37 1998)
• Continue development of SLP ambiguity selection
  procedure
• Combining SLP with water vapor, clouds SST will
  greatly improve storms and fronts research and
  analysis

R. A. Brown 2005 LIDAR Sedona
24
(No Transcript)
25
(No Transcript)
26
(No Transcript)
27
(No Transcript)
28
(No Transcript)