Satellite Instrument Calibration and Data Assimilation

Fuzhong Weng, Acting ChiefSatellite Meteorology and Climatology Division

NOAA/NESDIS/Center for Satellite Applications and Research

NOAA Satellite Conference, NCWCP, College Park, MDApril 10, 2013

2

• Calibration: is a process of quantitatively defining the satellite sensor responses to known signal inputs that are traceable to established reference standards, and converting the Earth observation raw signal to Sensor Data Records (SDRs).

• Calibrated SDRs from RDR are the fundamental building blocks for all satellite products, including the radiances for data assimilation in Numerical Weather Prediction (NWP), reanalysis, and fundamental climate data records (FCDRs) for climate change detection.

Why Calibration is Critical

Calibration is the centerpiece of data quality assurance and is part of the core competency of any satellite program

Sensor Data

Records

Environmental

Data Records

Climate Data

Records

Raw Data

RecordsRDR

SDR

EDR

CDR

NOAA Satellite Calibration Tasks

• Conduct prelaunch analyses of thermal vacuum data and provide recommendations for improving instrument design

• Quantify the uncertainty in radiometric calibration (e.g. precision and accuracy) for all categories of instruments

• Quantify the uncertainty in spectral calibration for hyperspectral instruments

• Quantify the errors in instrument geolocation and channel-to-channel co-registration

• Develop a long-term monitoring (LTM) system for trending the instrument performance (e.g. noise, spacecraft and instrument housekeeping )

• Analyze the root-cause for the instrument anomalies and provide the recommendations for mitigating the performance risk associated with all the anomalies

3

NOAA/Metop

ATMS

CrIS

VIIRS

OMPS

CERES

AMSR-2

SSMIS

AMSU

MHS

AVHRR

HIRS

GCOM-W

DMSP

AMSR-2

SSMIS

AMSU

MHS

AVHRR

HIRS

GCOM-W

ATMS

CrIS

VIIRS

OMPS

NPP/JPSS 

DMSP

Input Data Sources:•GRAVITE

(RDR/TDR)•CLASS (TDR/SDR)•DDS (Level 1B)

Input Data Sources:•EMC (GFS/GDAS)•ECMWF (GFS/GDAS)•CLASS (TDR/SDR)•DDS (Level 1B)

NPP/JPSS 

NOAA/Metop

Climate Predictions and Projections

Hurricanes and High Impact Events

Satellite Data and Application Demonstration System (DADS)

Global Forecasts

Regional Forecasts

NWP •  Instrument Status Trending•  Sensor Performance 

Trending•  Spacecraft Operational 

Status•  Sensor/SC Diagnostic 

Datasets

IPMS •  Sensor Data Global Distribution•  Sensor Data Global Bias 

Distribution•  Sensor Data Global Bias 

Trending

SQAS•  Satellite retrieval products•  Inter-sensor calibrated CDR 

products•  High Impact Events Imager

EQAS

Radiative Transfer Model

NPP/JPSS 

ATMS

CERES

OMPS

VIIRS

CrIMSS

GCOM-W AMSR-2

Instrument Performance Monitoring System (IPMS)

SDR Quality Assurance System (SQAS)

EDR Quality Assurance System (EQAS)

STAR ICVS-LTM System

AIRS

AVHRR

MHS

AMSU

IASI

Input Data Sources:

•GRAVITE (TDR/SDR)

•CLASS (TDR/SDR)

•DDS (Level 1B)

Output Products:•IPMS Analysis Data•LTM Trending Plots•Warning

Notification

Output Products:•SQAS Analysis

Data• Sensor Data Global

Distribution• Sensor Data Global

Bias Distribution•LTM Trending Plots•Warning

Notification

Output Products:•T/Q Profiles•Aerosol

Products•Cloud

Products•Ozone

Products•Surface

Products•Energy

Budget

NPP/JPSS Spacecraft 

NOAA/Metop 

Spacecraft Monitoring Example

o Instrument temperatures obtained from S/C telemetry (right)

o S/C PUMA Battery 1 Voltage real time variation during the last 24 hours and LTM trending since launch (below)

ATMS Monitoring Example

o Lunar intrusion effects on ATMS space view readings and channels are different (right)

o ATMS 4-Wire PRTs anomalies are observed in individual readings of all bands (below)

CrIS SDR Data Quality Monitoring

Large imagery part over Australia hot scene caused by invalid bit-trim flag

VIIRS Degradation Monitoring Samples

11

VIIRS Focal Plane Aperture Temperature

12

Suomi NPP TDR/SDR Algorithm Schedule

C

CCCCCCCCCCCCC

Sensor  Beta Provisional (Review Date) Validated (planned) 

CrIS   May 7, 2012 October 23, 2012 2013ATMS            February 22, 2012 October 23  2012 2013

OMPS-EV         March 12,  2012  October 23,  2012  2013VIIRS May 2, 2012  October 23,  2012 2013

Hurricane Sandy Warm Core Anomaly Ascending 1730 UTC, 29 October 2012

Cross section along Longitude 72.9 WCross section along Latitude 38.1 N

At 1800 UTC Oct 29 Max Wind: 90 MPH, Min Pressure: 940 hPa

13

Cross-Track Infrared Sounder (CrIS) SDR Status

• CRIS SDR provisional product review was held on October 23-24, 2012 and the panel recommended its provisional maturity level

• SDR provisional product:• NEdNs are well below

specifications• Spectral uncertainty: < 2 ppm,

well below specification• Radiometric uncertainty: ~0.1K,

well below specification• Geolocation error: < 1.0 km

below specification

14

NOAA AMSR-2 Calibration Status

• AMSR-2 SDR data (aka level 1) are processed at NOAA/STAR

• Biases with respect to TMI and CRTM simulations are evaluated. It is found that AMSR2 brightness temperatures from 6 to 18 GHz have warm biases which are also non-linear.

• Algorithms have been developed to detect and Radio Frequency Interference (RFI) signals in AMSR2 data

2012.10.26.06 2012.10.27.06

2012.10.28.06 2012.10.29.06

AMSR-2 Experimental Rain Water Path

TV Signals Reflected by Ocean – RFI

Satellite downlink beam coverage

Geostationary TV Satellite

18

AMSR2

Emission by ocean

Signals from Geostationary TV Satellite

TV signals reflected by ocean

Geostationary satellite TV signals reflected by ocean surface is a major source of maritime RFI.

RFI signals are mixed with natural emission from pixels interfered by reflected TV signals.

Community Radiative Transfer Model (CRTM) for Satellite Data Assimilation

• Atmospheric gaseous absorption - Band absorption coeff trained by LBL

spectroscopy data with sensor response functions

- Variable gases ( H2O, CO2, O3 etc) . - Zeeman splitting effects near 60 GHz

• Cloud/precipitation scattering and emission- Fast LUT optical models at all phases

including non-spherical ice particles- Gamma size distributions

• Aerosol scattering and emission- GOCART 5 species (dust, sea salt,

organic/black carbon, )- Lognormal distributions with 35 bins

• Surface emissivity/reflectivity - Two-scale microwave ocean emissivity- Large scale wave IR ocean emissivity - Land mw emissivity including vegetation and

snow- Land IR emissivity data base

• Radiative transfer scheme - Tangent linear and adjoints - Inputs and outputs at pressure level coordinate- Advanced double and adding scheme - Other transfer schemes such as SOI, Delta

Eddington

“Technology transfer made possible by CRTM is a shining example for collaboration among the JCSDA Partners and other organizations, and has been instrumental in the JCSDA success in accelerating uses of new satellite data in operations” – Dr. Louis Uccellini, Director of National Centers for Environmental Prediction

19

Satellite Data Critical for Improving Hurricane and Coastal Precipitation Forecasts

• Satellite microwave sounding data – provide hurricane thermal/moisture structure for improving intensity forecast (SSMIS/AMSU-A/MHS/ATMS)

• Satellite infrared sounding data – provide environmental thermal and moisture structure for track and precipitation forecast (HIRS/CrIS/AIRS/IASI)

• Ocean surface wind and temperature from satellite scatterometer and passive microwave imager – provide surface energy flux and surface vortex (ASCAT/AMSR2)

• GPSRO refractivity and bending angle – provide tropical cyclonegenesis information (COSMIC/GRAS)

• Geostationary sounder and imager – provide real-time monitoring and tracking of all severe weather events with a high temporal and spatial resolutions (e.g. GOES etc).

22

HWRF Model and Data Assimilation System

HWRF Model:

• 2012 NCEP-Trunk version 934

• Three telescoping domains:Outer domain: 27km: 75x75o; Inner domain: 9km ~11x10o

Inner-most domain: 3km inner-most nest ~6x6o

Revised Model Level and Top:

• Vertical levels: 61

• Model top: 0.5 hPa

Data Assimilation System:

• HWRF 6 hour forecasts

• GSI (3DVAR)

• The Hurricane Weather Research and Forecasting (HWRF) Model dynamical core is designed based on the WRF model using NCEP Non-Hydrostatic Mesoscale Model (NMM) core with a movable high-resolution nested grid (telescopic)

• Regional-Scale, Moving Nest, Ocean-Atmosphere Coupled Modeling System. Horizontal resolution: 27 km outer grid, 9 km inner grid, 42 vertical levels

• Non-Hydrostaticsystem of equations formulated on a rotated latitude-longitude, Arakawa E-grid and a vertical, pressure hybrid (sigma_p-P) coordinate.

• Advanced HWRF 3D Variational analysis that includes vortex relocation, correction to winds, MSLP, temperature and moisture in the hurricane region and adjustment to actual storm intensity.

• Uses SAS convection scheme, GFS/GFDL surface, boundary layer physics, GFDL/GFS radiation and Ferrier Microphysical Scheme.

• Ocean coupled modeling system (POM/HYCOM).23

Pre

ssur

e (h

Pa)

ATMS Weighting Function

NCEP HWRF Top

STAR HWRF Top

ATMS Weighting Functions

Our approach: Raise the model top to allow for more satellite data assimilated into hurricane forecast model 24

Control Experiment – L61

Conventional Data:

Radiosondes, aircraft reports (AIREP/PIREP, RECCO, MDCRS-ACARS, TAMDAR, AMDAR), Surface ship and buoy observations , Surface observations over land, Pibal winds,Wind profilers, VAD wind, Dropsondes

Satellite Instrument Data:

• AMSU-A (channel 5-14) from NOAA-18, NOAA-19 and METOP-A• HIRS from NOAA-19 and METOP-A • AIRS from EOS Aqua • ASCAT from METOP-A • GPSRO from GRAS/COSMIC

25

Hurricane Sandy Forecasts

Control : L61

Sensitivity Experiments

ATMS: L61+ATMS

IASI: L61+IASI

CrIS: L61+CrIS

Forecast Period: 1800 UTC Oct 22, 2012 -

1800 UTC Oct 29, 2012

Total Cycles: 29

1800 UTC 0000 UTC 0000 UTC day55-day Forecast

HWRF FST Turn on GSI

26

CONV OnlyL61

Impacts of Assimilation of NOAA/METOP Data on Hurricane Sandy’s Track

28

L61:Control Run

Impacts of Assimilation of ATMS Radiances on Hurricane Sandy’s Track

L61+ATMS

29

Comparison of Temperature Increments from ATMS and AMSU-A

Shaded: ATMS Red contour: AMSU-ABlack contour: Conventional

ATMS and AMSU-A (NOAA-19) produce largest temperature innovation in storm regions in similar magnitudes and complementary in spatial coverage

30

L61+IASIL61 L61+CrIS

Impacts of Assimilation of IR Hyperspectral Sounder Radiances on Hurricane Sandy’s Track

31

Multiple Forecasts of Max. Wind Speed for Hurricane Sandy

L61

33

IASI

ATMS

CrIS

Summary and Conclusions

• NOAA satellite instruments are well calibrated for operational applications and environmental data stewardship.

• Suomi NPP ATMS is very unique in resolving hurricane warm core features through its high spatial oversampling and additional channels.

• 2012 HWRF/GSI is re-configured with more vertical layers and higher model top for direct satellite radiance assimilation.

• In general, control and sensitivity experiments show that uses of ATMS/CrIS data in HWRF improve the forecasts in both hurricane intensity and track.

• When hurricane is near landfall, satellite data always have impacts on track, especially with ATMS.

• Satellite data show significant impacts on three day’s track forecasts over open water. It appears that CrIS has also significantly large impact on track forecasts.

• For conventional data only, hurricane track forecast errors increase rapidly when hurricane is near landfall.

35