ENVISAT: The AATSR

1
ENVISAT The AATSR validation Program
2
Presentation

• Introduction
• The AATSR - Presentation - Instrument
  specification - Data products -
  Application
• Calibration
• validation -Sea Surface Temperature
• -Land Surface
• Conclusion

3
AATSR Introduction

• Designed primarily to measure sea surface
  temperature to lt0.3 K
• Third in a series ATSR-1, ATSR-2, AATSR
• Funded primarily by the UK DETR
• Useful for climate modelling and the detection of
  global change.

4

• ATSR-AATSR toward a long term data record for
  climate studies
• ERS-1/ATSR1 launched in 1991
• Retrieve SKIN SST using a radiative transfer
  model making the SST data set independent
  from in situ measurements
• SST retrieval scheme takes into account the
  stratospheric aerosols effects (Mt Pinatubo)
  and high WV content. It makes the ATSR/AATSR
  accuracy data set unique in the world.
  

Spatially averaged global sea surface temperature
imagery, for September 2000, from ATSR2. Cold
waters are shown in purple/blue, whilst warm
waters are shown in orange/red.
5

• ATSR-AATSR innovative features
• Two viewing angles, Nadir-55 deg forward
• A novel type of stirling cooler to maintain
• the detectors temperature very low, giving
  very high sensitivity
• On-board calibration with two black bodies
  designed for high emissivity, uniformity and
  long term stability.
• Visible calibration performed once per orbit
  using a Russian opal diffuser
• Very good noise performance
• Very good geolocation accuracy
• High digitization allowing fine structures
  detection
• Thanks to ENVISAT improved on board recorder
  capacity, AATSR will be always in full swath
  and full digitization

False colour daytime image of the US eastern
seaboard 16 May 1995 (processed at RAL).
6
Instrument Specifications
7
AATSR Data Products
8
ATSR-AATSR applicationsLand/ice

Hot spot in Indonesia detected from the 3.7 µm
channel in 1997.
A recent 12 µm thermal image, taken by ATSR-2 in
March 2000 showing the break-up of the Ross Ice
Shelf in the Antarctic. A large iceberg, 300
km by 40 km is seen breaking away from the main
ice sheet. (processed at RAL).
Daytime ATSR-2 image of the Middle East, showing
bands 0.55µm, 0.67µm, and 0.87µm
9
ATSR-AATSR applicationsSea / ocean / climate

SST anomaly, 1997 El-Nino observed by
ATSR-2 climate/meteo application
Crete, 1997/07 SST climate, fisheries, pollution
10
ATSR-AATSR applicationsAtmosphere/aerosol
Stereoscopy (3) Fun (and informative!)

Right eye
Left eye
Stereoscopy effect to be viewed by cross-eyed
method
11
ATSR-AATSR Calibration

• INFRARED CALIBRATION.
• AATSR infrared channels are calibrated
  during each scan using a pair of blackbody
• calibration targets spanning the expected
  range of SST one target is cold, typically at
  -10C,
• and the other hot at around 30C. The
  infrared calibration is applied automatically
  during the
• ground processing so users are provided with
  fully calibrated brightness temperatures.

Visible target calibration
Along track earthview
Black Bodies/IR calibration targets
nadir earthview
Absolute accuracy equivalent to 0.1 K (3s) Black
bodies designed and manufactured by AEA
Technology and MSSL
AATSR scan cycle
12
ATSR-AATSR Calibration

• VISIBLE CALIBRATION.Calibration of the AATSR
  visible channels is achieved once per orbit by
  viewing the Sun using a Russian Opal diffuser.
  The radiometric offset is determined by viewing
  the cold black body (dark signal).
• VISIBLE CHANNEL CALIBRATION AND INSTRUMENT
  STABILITY.
• The on-board visible channel calibration system
  on ATSR-2 has worked extremely well. Smith et al.
  (1997) has used a long series of observations
  from the South Eastern Libyan Desert to
  characterize the drift..
• Desert sites, such as this one in Libya, with
  their uniform surfaces, are ideal sites for
  calibration/validation measurements

13
ATSR-AATSR Calibration

Libyan site, ATSR-2 image processed at RAL
(Smith)
14
ATSR-AATSR Calibration

• The on-board calibration system for the
  reflection channels
• shall provide a known source of radiance over
  the AATSR
• reflection channel ranges to an absolute
  accuracy of 5.
• The thermal IR channels shall measure the
  radiance to an accuracy
• equivalent to a temperature error less than 0.1
  K.
• The radiometric noise shall be less than 0.08 K
  at 3.7 and less than
• 0.05 K at 11 and 12 micrometers channels for a
  scene temperature of 270 K.
• Pre-launch end to end calibration has been
  performed. It shows good results.

15
AATSR Calibration / algorithm verification

• AATSR is a selfcalibrating instrument. As such,
  calibration of the instrument after launch is not
  required.
• There will, however, be specific activities to
  check and characterise the instrument
  post-launch.
• - Algorithm verification data processing
  algorithms are verified and fine- tuned. That
  includes
• gt geolocation gt re-gridding process
• gt land-flagging
• gt cloud flagging
• gt spatial averaging
• gt SST / NDVI retrieval
• gt formatting of output product
• - The vicarious calibration of the visible
  channels, can be classified as either
  calibration or validation. Within the AATSR
  validation program, vicarious cal/val is
  treated under the title of validation.

16
AATSR Validation

• The process of assessing by independent means,
  the quality of the data products derived from
  system outputs
• Is AATSR returning acceptable measurements
  and meeting its specifications?
• AATSR validation is performed under the
  coordination of Marianne Edwards from Leicester
  University.
• Validation the core validation program will
  validate the- Level 1b product (GBTR),
  primarily over land.
• - Level 2 product, sea surface temperature.

17
AATSR Core Validation

• Sea Surface Temperature
• 2 type of products
• - Gridded sea surface temperature product
• - Spatially averaged sea surface temperature
  product
• Three levels
• - Global buoy data, and the early indication of
  gross errors in ASST (general measurements)
• - Spot values for gridded data (moderate
  accuracy)
• - Precision measurements (high accuracy)

18
AATSR Core Validation early indication of gross
errors

• ASST data
• Systematic review of buoy data
• Comparison with SST analysis fields
• Inexpensive, global scale, routine basis
• Early indication of gross errors
• UK Met. Office/Hadley Centre/RAL/Meteo-France

19
AATSR Core Validation early indication of gross
errors
Location of buoys used to create a matchup
database for SST (meteo-France)
20
AATSR Core Validation Spot values for gridded
data

• Autonomous measurements on board
  ships-of-opportunity.
• Bolted onto merchant ships of ferries.
• Good coverage, good quality.
• Low cost.
• Dr Ian Barton will co-ordinate validation
  activities in Australia. - One activity uses a
  Tasco radiometer to collect measures of SST
  along the Perth coast.- Other activity uses
  Everest Radiometer. It will operate between
  Townsville and Kelso reef on the Great Barrier
  Reef.
• ISAR, designed by C.Donlon (JRC), will be
  installed on the Brittany Ferry operating between
  Porthmouth and Le Havre.The operation will be
  coordinated by Ian Robinson (SOC).

21
(No Transcript)
22
AATSR Core Validation precision measurements

• Precision measurements using specifically
  designed radiometers.
• Fewer data points and more limited coverage.
• Very accurate.
• Example Instruments - SISTeR (RAL). Already
  Used for ATSR-1/2.- M-AERI (RSMAS, Peter
  Minnett).- DAR011(CSIRO, Ian Barton).
• IR sensor validation and Intercalibration
  Workshop held at Miami (end May). it should
  provide with a high level of confidence for the
  AATSR Envisat validation results.---gt
  Presentation from Ian Barton.

23
The SISTeR (Scanning Infrared Sea Surface
Temperature Radiometer) is a compact and robust
chopped self-calibrating filter radiometer. It
measures approximately 20 20 40cm and weighs
about 20kg. The instrument is divided into three
compartments containing the foreoptics, scan
mirror and reference black bodies, and a
small-format PC with signal processing and
control electronics.
SISTeR mounted on board the foremast of a boat
used in the MUBEX campaign.
24
Land Surface Validation

• Level 1b GBTR product
• Vicarious validation
• Uniform sites, stable over time
• Libyan Desert, Greenland, China
• RAL/CNES/KNMI
• Specific validation campaigns, and the collection
  of ground-based measurements
• - Australian validation campaigns (CSIRO, Fred
  Prata).
• 3 sites, Thangoo, Amburla, Hay,
  representing different environmental conditions
  and land cover type.

25
Land Surface Validation
Amburla site Australia
Thangoo site Australia
26
Conclusion

• ATSR1/2-AATSR SST long time series.
• Very high accuracy responding to the
  climate/ocean community needs.
• Validation plan integrated involves variety of
  different projects.
• Integrated into the ENVISAT validation programme.
• Validation Rehearsal, Oct/Nov 2000. Testing of
  methods and communication channels. 2nd
  rehearsal, June 2001.
• NILU validation database
• Launch in Sep/Oct 2001 (2nd Oct).
• Commissioning phase workshop. L6.
• Validation workshop L9. Each PI will present
  validation results and conclusion drawn.
• Validation loop set up (currently under revision
  by DETR/ESA).
• Long term validation increase number of sites
  and season, monitor long term data product
  quality, validate new product.