Geophysical Ocean Products from AMSR-E

1
Geophysical Ocean Products from AMSR-E WindSAT
Chelle L. Gentemann, Frank Wentz, Thomas
Meissner, Kyle Hilburn, Deborah Smith, and Marty
Brewer www.remss.com
?Retrieval algorithm ? Validation
results ? New directions AMSRE Science Team
Portland, OR, 2012
2
AMSRE WindSAT v7 Ocean Products

• L2, L3, and L4 Data produced in NRT
• Climate quality satellite inter-calibration and
  re-processing
• Geophysical data freely available via FTP
• Browse images, movies, sub-setting at
  www.remss.com

SST
Wind Speed
Rain Rate
Cloud
Water Vapor
3
V07

• There are 3 major V7 algorithm changes
• the water vapor continuum absorption model was
  re-derived
• the clear-sky bias in cloud water was
  removed (cloud data format has changed)
• the beamfilling correction in the rain algorithm
  was modified
• The effects of V7 changes relative to V6  
• increased vapor values in the range of 50-60 mm
  by 1increased vapor values above 60 mm by 2-3
• changed the range of cloud water values to -0.05
  to 2.45 mm (cloud data format has changed)
• increased the global mean rain rates by about
  16 (mostly due to changes in the extratropical
  values)
• New emissivity model (Meissner Wentz)
• Improved diurnal cycle added to Reynolds during
  development

4
Algorithm
AMSR2 AMSRE develop methodology to utilize
adaptive algorithms, when channel is identified
as RFI or side lobe (land) contaminated, will
rely on other channels for retrieval where
possible.
5
AMSR-E v7 SST Buoy Validation
Independent, global validation using GTS in situ
SST database
6
Wind algorithm

• No-rain algorithm (? 10.7 GHz 32 km)
• Physical algorithm
• Trained from Monte Carlo simulated brightness
  temperatures (TB).
• Based on radiative transfer model.
• Wind speed in rain algorithms (? 6.8 GHz 50 km)
• Statistical or hybrid algorithms
• Trained from match-ups between measured TB and
  ground truth wind speeds in rainy conditions.
• Utilizes spectral difference (6.8 GHz versus 10.7
  GHz) in wind/rain response of measured brightness
  temperatures.
• Same method is used by NOAA aircraft step
  frequency microwave radiometers (SFMR) to measure
  wind speeds in hurricanes.
• Global wind speed in rain algo
• trained for all rain conditions
• H-wind algo
• trained for tropical cyclones

7
Winds

• Radiometer winds in rain
• T. Meissner F. Wentz, IEEE TGRS 47(9), 2009,
  3065 - 3083

8
Performance no-rain wind speeds

• Joint PDF WindSat versus validation data

5 years of data. 2003-2008
9
Performance wind speeds in rain
Rain Rate Satellite BUOY Wind Speed m/s Bias Standard Deviation Satellite BUOY Wind Speed m/s Bias Standard Deviation Satellite BUOY Wind Speed m/s Bias Standard Deviation Satellite BUOY Wind Speed m/s Bias Standard Deviation
Rain Rate WindSat all-weather algorithm WindSat all-weather algorithm QuikSCAT Ku 2011 QuikSCAT Ku 2011
no rain 0.04 0.9 0.01 0.9
light rain 0 3 mm/h 0.70 1.6 1.7 2.3
moderate rain3 8 mm/h 0.02 2.0 4.8 3.6
heavy raingt 8 mm/h -0.05 2.5 7.1 4.5
5 years of WindSat - Buoys
10
Comparing Ground-Based GPS to SSM/I water Vapor
Geologists have deployed a HUGE network of
ground-based GPS stations to evaluate earth
movements, etc. GPS works by measuring the
amount of time the signals need to go from the
GPS satellites to the ground. This time depends
on the distance (obviously) but also amount of
water vapor (and air) in the atmosphere above the
station. This is an error source for the
geologists. It is a signal for meteorologists!
11
Binned means of SSM/I v6 GPS PW
Updated and Corrected in AMSRE v7 RTM
All SSMI data (lt50 mm) Mean 0.05 mm STD 1.79
mm collocations 91678
SSM/I v6 vapor appears low above 50 mm
12
Adaptive algorithms

• In region affected by side-lobe contamination
  (near land or sea ice) or RFI, switch algorithms
  to utilize non-affected channels if possible.
• Develop automated way to flag data affected by
  RFI at L2B level

13
AMSR-E Microwave SST Retrievals for Europe
Standard Algorithm
RFI Mitigation

• Color standard deviation of AMSR-E Reynolds
  SST
• RFI mitigation algorithm does not use 10.7 GHz
  channel, and is optimized to use the remaining
  channels SST retrieval possible thanks to
  AMSR-Es 6.9 GHz channel
• RFI around Europe has a strong increasing trend
  over AMSR-Es period of operation
• Data shown descending passes for 2011
• Red spots in North Sea are ground-based RFI (also
  present in ascending passes)

14
Thank you

• Adaptive algorithms to minimize RFI, land
  contamination
• AMSR2 sample data just released.