NOAA18 Instrument Calibration and Validation Briefing

1
NOAA-18 Instrument Calibration and Validation
Briefing

• NOAA/NESDIS/Office of Research and Applications
• July 6, 2005
• For archived activities and latest news, please
  visit
• http//www.orbit.nesdis.noaa.gov/smcd/spb/n18calva
  l

2
Our Team

• Mitch Goldberg ORA/SMCD Division Chief, -
  Management and Technical Oversight
• Fuzhong Weng ORA/SMCD/Sensor Physics Branch
  Chief and NOAA-18 cal/val team leader, instrument
  asymmetry and microwave products and algorithms,
  radiance bias assessments for NWP model
  applications
• Changyong Cao HIRS instrument calibration
• Fred Wu AVHRR VIS/IR instrument calibration
• Tsan Mo AMSU/MHS instrument calibration
• Jerry Sullivan AVHRR thermal channel
  calibration/ NDVI validation
• Tony Reale HIRS/AMSU/MHS sounding
  channel/products validation
• Mike Chalfant HIRS/AMSU/MHS sounding
  channel/products validation /geolocation
• Ralph Ferraro AMSU/MHS window channels/MSPPS
  products validation
• Larry Flynn SBUV product validation
• Tom Kleespies AMSU on-orbit verification
• Hank Drahos Sounding product validation
• Dan Tarpley AVHRR product NDVI monitoring
• John LeMarshall (JCSDA) Impacts assessments of
  NOAA-18 data for NWP applications
• Stephen English (UK) tests of direct readout
  data and NWP impacts demonstration

3
Outline for this Brief

• ORA NOAA-18 Cal/Val Activities
• Current Assessments
• Summary
• Next Step

4
Activities of the ORA NOAA-18 Cal/Val Team

• Monitor and quantify instrument noises though
  analyzing calibration target counts and channel
  space view measurements
• Assess instrument geolocation biases and
  co-registration and provide recommended solutions
  for satellite roll and pitch adjustments
• Characterize other systematic biases in radiance
  through rigorous forward modeling and
  inter-satellite calibrations
• Provide initial demonstration and assessments of
  NOAA-18 data for improving numerical weather
  prediction
• Validate product algorithms (e.g. ATOVS and
  MSPPS, TOAST, UV index, NDVI, SST, AOD) for
  transition into operation
• Communicate with NOAA-18 OV team, instrument
  vendors and users with timeliness diagnostics of
  instrument performances and provide root cause
  analyses

5
Long-Term Goals of ORAs Cal/Val Program

• Develop an Integrated Cal/Val System Framework to
  enhance ORAs capability and efficiency to
  provide outstanding calibration and validation to
  METOP, NPP NPOESS, and GOES-R
• Outcome gtgtgt Provide timely and accurate
  assessments of NOAA instruments on-orbit
  performances and the impacts of noise and loss of
  channels on operational products and data
  assimilation

6
NOAA-18 Instrument Assessments

• AVHRR/3 Advanced Very High Resolution Radiometer
• HIRS/4 High Resolution Infrared Sounder
• AMSU-A Advanced Microwave Sounding Unit-A
• MHS Microwave Humidity Sounder
• SBUV/2 Solar Backscattered Ultraviolet Radiometer

7
ORA NOAA-18 Instrument Cal/Val Status Report
Support to NOAAs Critical Satellite Program
Noise, Bias and Anomaly
Navigation and Co-registration
Navigation errors are more than one scan lines
and/or fovs and their causes have not been
identified
All channels do not meet the noise specification,
or large unknown biases and anomaly
Some channels do not meet the noise
specification, no major anomaly and biases
Small navigation errors which can be fixed
through recommended roll and pitch adjustments
All channels meet the noise specification
No navigation errors
NWP Readiness and Product Demo
Analysis and Recommended Solutions
Measurements are assessed and not ready for
product and NWP applications
No analyses and recommended solutions for out-of
spec instrument parameters
R
Analysis for out-of spec parameters but solutions
have not been recommended
Some channels are assessed and may be useful for
product and NWP applications
Y

Analysis and recommended solutions are proposed
for out-of spec instrument parameters
G
Measurements are fully tested and ready for
product and NWP applications
8
ORA NOAA-18 HIRS/4 Cal/Val Findings Support to
NOAAs Critical Satellite Program
Noise, Bias and Anomaly
Navigation and Co-registration

• HIRS/4 noise is fluctuating at 2-3 times the
  specification for most long-wave channels.
• The noise among all channels is highly correlated
• Space view count at channel 1 remains saturated
  and the noise has not been quantified

• HIRS/4 geolocation has a mean shift of 3-4 AVHRR
  pixels, which is more related to AVHRR
  geolocation error

NWP Readiness and Product Demo
Analysis and Recommended Solutions

• ORA has developed a website for real-time
  monitoring of noise trend
• ORA had several telecons with OV team and
  instrument vendor to present the noise assessment
  results
• ORA proposed several possible coherent noise
  generators, 1) clamp system itself, 2)
  interference with the clamp system or 3) through
  the clamp system
• These root cause analyses are likely confirmed by
  MIT/Lincoln Lab.

• Coefficients in radiative transfer model and data
  assimilation system are updated
• Brightness temperatures at most of HIRS channels
  are not ready for sounding and NWP applications

9
ORA NOAA-18 HIRS/4 Cal/Val Findings Support to
NOAAs Critical Satellite Program
Noise, Bias and Anomaly
Navigation and Co-registration
HIRS Pixel Number
NWP Readiness and Product Demo
Analysis and Recommended Solutions
Reference voltage for the clamp target which may
be fluctuating
10
ORA NOAA-18 AMSU-A Cal/Val Findings Support to
NOAAs Critical Satellite Program
Noise, Bias and Anomaly
Navigation and Co-registration

• AMSU-A noise is evaluated and shows all channels
  meet the specification.
• Overall AMSU calibration algorithms are healthy
  with reasonable gains, and variability in cold
  and warm calibration targets
• AMSU-A1 module displays a slightly larger
  cross-track asymmetry

• AMSU-A1 is correctly navigated
• AMSU-A2 is off by 0.7 MHS scanline along track
  and 0.36 MHS FOVs across track.

Analysis and Recommended Solutions
NWP Readiness and Product Demo

• Coefficients in radiative transfer model and data
  assimilation system are updated
• Impacts of AMSU on weather prediction are being
  assessed
• AMSU microwave surface and precipitation products
  are demonstrated

• Recommend to OV team for possible roll and pitch
  adjustments to AMSU-A2 module navigation errors
• Recommend to OV team to investigate AMSU-A1
  cross-track asymmetry with various antenna
  pattern correction algorithms

11
ORA NOAA-18 AMSU-A Cal/Val Findings Support to
NOAAs Critical Satellite Program
Noise, Bias and Anomaly
Navigation and Co-registration
NWP Readiness and Product Demo
Analysis and Recommended Solutions
A1 cross-track asymmetry
12
ORA NOAA-18 MHS Cal/Val Findings Support to
NOAAs Critical Satellite Program
Noise, Bias and Anomaly
Navigation and Co-registration

• Display 0.9 MHS scanline error along track

• MHS noises are quantified and show all channels
  meet the specification and are better than AMSU-B

NWP Readiness and Product Demo
Analysis and Recommended Solutions

• Coefficients in radiative transfer model and data
  assimilation system are updated
• Impacts of MHS on weather prediction are being
  assessed
• MHS microwave surface and precipitation products
  are demonstrated

• Recommend to OV team for a possible roll
  adjustment to correct MHS navigation errors

13
ORA NOAA-18 MHS Cal/Val Findings Support to
NOAAs Critical Satellite Program
Noise, Bias and Anomaly
Navigation and Co-registration
MHS channel 1 reveals a coast line shift along
track
NWP Readiness and Product Demo
Analysis and Recommended Solutions
A high quality of rainfall distribution from MHS
14
ORA NOAA-18 AVHRR/3 Cal/Val Findings Support to
NOAAs Critical Satellite Program
Noise, Bias and Anomaly
Navigation and Co-registration

• The blackbody temperature changes are monitored
  and shows in a small range of variability (only
  2K)
• Thermal channel (3-5) calibration is healthy
• Wus Vis/IR calibration algorithms are working
  well

• AVHRR/3 has possible 2-3 pixels navigation errors
  along track

NWP Readiness and Product Demo
Analysis and Recommended Solutions

• AVHRR/3 produces a good quality of NDVI
• AVHRR/3 produces SST and aerosol optical depth
  with smaller biases

• Recommend to OV team to have a possible roll
  adjustment

15
ORA NOAA-18 AVHRR/3 Cal/Val Findings Support to
NOAAs Critical Satellite Program
Noise, Bias and Anomaly
Navigation and Co-registration
AVHRR NDVI product shows AVHRR has a 2-3 pixels
shift in coastline along track
NWP Readiness and Product Demo
Analysis and Recommended Solutions
NOAA-18 AVHRR derived NDVI is based on NOAA-16
vicarious calibration algorithm. This NDVI
product is selected for our initial assessments
of visible channel calibration because it uses
both visible/near infrared channel data.
16
Responses from Program Managers

• From Stephen English, The Metoffice UK Thanks
  for sending this Fuzhong. For your information we
  intend to put NOAA-18 "passively" into a parallel
  suite next week which will allow us to
  characterise bias characteristics more
  thoroughly. It will go active in the parallel
  suite in two weeks time (all being well) and
  fully operational in 4 weeks time. It seems to us
  that NOAA-18 is in pretty good shape except for
  HIRS (which we won't use initially) and the
  issues which are being discussed are mostly what
  could be described as fine-tuning. When it goes
  operational we will drop Aqua AMSU-A from our
  assimilation system. Thanks for this. We will
  try and put together a report today on our
  monitoring of NOAA-18 against our model and send
  it to you to use as you see fit in NOAA. We won't
  distribute to anyone else. As far as we are
  concerned AMSU-A and MHS data quality is good
  enough to assimilate with our normal bias
  correction and quality control procedures in
  place. We will be spinning up bias corrections
  over the next 2 weeks and putting it into a
  parallel suite in two weeks time which will run
  for several weeks. We are disappointed that we
  are still being told not to pass the data to
  ECMWF and Meteo-France. It would be useful to see
  their monitoring of it
• From Jim Silva, NDE Program Manager The
  attached analysis of the NOAA 18 HIRS and AMSUA
  data anomalies demonstrates the level of
  expertise available within our Office of Research
  and Application to investigate sensor problems
  with instrument manufacturers, test hypotheses,
  and find acceptable work-arounds
• From Bill Mazur, NESDIS Polar Satellite
  Acquisition Manager Excellent work by your
  folks, Fuzhong.  Thanks for including me in the
  distribution
• From Tom Schott, NESDIS Polar Satellite Product
  Manager Graeme,Here's some more initial finding
  from ORA on MHS
• From Al Goldberg, IPO Jim,Thanks for forwarding
  this to me.  It looks like very good work, and
  highlights the types of analytical tools,
  approaches, and experience reserves which will be
  irreplaceable if we expect to get good
  performance from the NPP/NPOESS sensors.  I'll
  forward the presentation to Karen and Joe
  Mulligan.

17
Overall Summary

• Most instruments are assessed under ORA cal/val
  project and their noises have met specification
  with the current exception of HIRS
• Navigation and co-registration errors for all
  instruments are examined and quantified. The
  required roll and pitch adjustments are
  recommended to OV team
• AMSU instrument cross-track biases are quantified
  and A1 asymmetry is being studied
• NOAA-18 AVHRR NDVI product appears to be
  consistent with NOAA-16
• The NOAA-18 MHS NE?T for each channel was
  calculated and is better than AMSU-B by a factor
  of 2
• AMSU-A/MHS hydrological parameters (rainfall,
  total precip. water) appear to be better quality
• ORA has delivered all required software and codes
  to NCEP for assimilating all NOAA-18 measurements

18
Next Step

• Develop an integrated calibration and validation
  system for timely and accurate calibration
  analyses for POES instrument on-orbit
  performances
• Validate all existing operational products
  clouds, aerosol, NDVI, SST, ATOVS, MSSPS, TOAST,
  SBUV
• Study NOAA-18 unique products improvements (e.g.
  HIRS 10 km sounding, MHS cloud ice and rainfall
  products)
• Test ORA Multi-sensor Integrated Retrieval system
  with NOAA-18 measurements

19
Backup Slides show ORA Long-Term Strategy for An
Integrated Cal/Val System Framework
20
ORA Pre-launch Calibration Support
Level 0 (RDR)
Level 1B (SDR)
Calibration
Radiances from Earth Target
Counts from Earth Targets
Sensor Calibration
Counts from Calibration Targets (ICT, SD, SPACE)
Calibration Coefficients
Meta Data to CLASS Long Term Monitoring Re-process
ing
Pre-launch Sensor Characterization
SOH Telemetry
21
ORA Post-Launch Integrated Cal/Val System
Framework
22
Cal/Val System Performance Example Detecting
DMSP SSMIS Anomaly
Shown is the difference between simulated and
observed SSMIS 53.6 GHz. The SSMIS is the first
conical microwave sounding instrument, precursor
of NPOESS CMIS. The calibration of this
instrument remains unresolved after 2 years of
the lunch of DMSP F16. The outstanding anomalies
have been identified from three processes 1)
antenna emission after satellite out of the earth
eclipse which contaminates the measurements in
ascending node and small part in descending node,
2) solar heating to the warm calibration target
and 3) solar reflection from canister tip, both
of which affect most of parts of descending node.

23
Cal/Val System Performance Example NOAA-16
AVHRR Calibration Slope Update
Shown is NOAA-16 AVHRR channel 1slope calibration
coefficient during the past three years.
This calibration is done through a vicarious
technique because there is no on-board visible
channel calibration target for AVHRR
24
Cal/Val System Performance Example WindSat
Stokes channel Biases
Shown are the time series of the mean 4th Stokes
parameters over Amazon rainforests. the line
with diamond, triangle and square corresponds to
10.7, 18.7 and 37 GHz, respectively Monthly mean
of 4th Stokes components over Amazon rainforests
should be zero because of surface roughness and
heterogeneity relative to azimuthal direction.
The residual of this mean is largely due to the
instrument calibration biases. The bias (-0.5K)
at 18.7 GHz will result in substantial bias in
wind direction retrievals because of the actual
wind induced signal is on the order of a couple
of degrees in Kelvin
25
Transition to NPP/NPOESS Cal/Val
NPOESS verification and validation. After the
delivery of the NPP/NPOESS ground systems and
with the launch of the NPP and NPOESS satellites,
algorithm verification and product validation
will berequired to ensure that data and products
that meet spec are being delivered from the IDPS.
The government will have to evaluate and make a
determination of the quality of NPP and NPOESS
EDRs. This job must be done in NESDIS by ORA and
resources must be available to do it. There has
been no commitment from IPO to fund this task.
Without this funding, NESDIS will take delivery
on the IDPS system and EDRs without independent
verification of product accuracy or if the data
stream and products even meet spec.