SST from VIIRS on NPP: prelaunch preparations and post-launch validation Peter J Minnett & Robert H...

SST from VIIRS on NPP: prelaunch preparations and post-launch

validation

Peter J Minnett & Robert H Evans

Meteorology & Physical Oceanography

Rosenstiel School of Marine and Atmospheric Science

University of Miami

Miami FL USA

NASA SST Science Team MeetingSeattle, November 2010

Outline

• Description of VIIRS – Visible/Infrared Imager/Radiometer Suite

• SST retrievals• Cal/Val approach

All information about VIIRS is from publicly accessible sources.


NPP payload

From http://modis.gsfc.nasa.gov/sci_team/meetings/201001/presentations/plenary/gleason.pdf


VIIRS

• The Visible/Infrared Imager/Radiometer Suite collects visible/infrared imagery and radiometric data.

• Applications include atmospheric clouds, earth radiation budget, clear-air land/water surfaces, sea surface temperature, ocean color, and low light visible imagery.

• Primary instrument for satisfying 22 Environmental Data Records (EDRs) and 2 Key Performance Parameters (KPPs): Imagery & sea surface temperature.

• Multiple VIS and IR channels between 0.3 and 14 μm• Imagery (I) Spatial Resolution: ~370m @ nadir / 750m @ edge of

swath• Moderate (M) Spatial Resolution: ~740m @ nadir / 1500m @ edge of

swath• Swath width ~3000km


VIIRS Components• Spectral Bands:

– Visible/Near IR: 9 plus Day/Night Band– Mid-Wave IR: 8– Long-Wave IR: 4

• Imaging Optics: 18.4 cm Aperture, 114 cm Focal Length

• Band-to-Band Registration (All Bands, Entire Scan)

> 80% per axis• Orbital Average Power:

240 W • Mass: 275 Kg


VIIRS innovations

• Rotating telescope primary optics• Two-sided “Half-Angle Mirror” (HAM)• Multiple detectors (16) per spectral band• On-board pixel aggregation


VIIRS


Risk reduction by using components derived from heritage instruments:

• Rotating Telescope from SeaWiFS

• Black-body from MODIS

• Multiple Focal Plane Arrays and Multiple Detector Assemblies from MODIS

Risk reduction by using components derived from heritage instruments:

• Rotating Telescope from SeaWiFS

• Black-body from MODIS

• Multiple Focal Plane Arrays and Multiple Detector Assemblies from MODIS



Pixel Aggregation

• Each “pixel” has three rectangular detectors in the scan direction

• Detectors have a 3x1 aspect ratio

• These are aggregated in threes, then twos, then no aggregation, across the scan.

• This is an attempt to provide near uniform spatial resolution across the swath.


VIIRS vs MODIS spatial resolution

From http://www.ipo.noaa.gov/ams/2010/posters/AGU_AMS-RAY_NGAS-VIIRSHeritageSystems-SNODGRASS_GUENTHER_ANDREAS-WE_PRINT-PR.pdf


VIIRS SST Bands

GSD = Ground sampling distance

GSD = Ground sampling distance

Spectral bands are a subset of MODIS bands


These are very promising


VIIRS SST Uncertainty Estimates

• The sources of error the VIIRS SSTs fall into two categories:– associated with imperfections in the instrument– arise from imperfections in the atmospheric correction algorithm.

• The instrumental effects include:– The inherent noise in the detectors, the Noise Equivalent Temperature

Difference (NEΔT)– Band-to-band registration (BBR)– Modulation Transfer Function (MTF) – Imperfections in the knowledge of angular dependence of the reflectivity

of the “Half Angle Mirror”– Calibration errors, such as imperfections in the knowledge of the

emissivity and surface temperature of the on-board black body target, and of stray radiation falling on the detectors.

• Uncertainties will be established soon after launch using multiple techniques.


VIIRS SST algorithms

Daytime NLSST algorithm:

where a0, a1, a2, a3 are coefficients derived by regression analysis, T11 is the measured brightness temperature at 11 µm (VIIRS band M15), T12 is the measured brightness temperature at 12 µm (VIIRS band M16), RSST is a modeled, first guess SST, and z is the sensor zenith angle.

Night-time NLSST algorithm:

where a0, a1, a2, a3 are coefficients derived by regression analysis (but are different from those in Equation 12), T3.7 is the measured brightness temperature at 3.7 µm (VIIRS band M12).


Post launch validation

The approach will be based on experience gained from AVHRR, (A)ATSR and MODIS, and will involve comparisons with:• Other validated satellite data sets (e.g. AVHRR,

AATSR, MODIS…)• Drifting and moored buoys• Ship-based radiometers – M-AERI, M-AERI

Mk2, ISAR…..


SST validation using ship-board

radiometersRadiometers installed on ships for the validation of MODIS skin SSTs. Top: the ISAR mounted above the bridge of the M/V Jingu Maru. Middle: M-AERI mounted on the NOAA S Ronald H. Brown. Bottom: M-AERI mounted on an upper deck of the Explorer of the Seas.


M-AERI validation data

M-AERI cruises since the launch of Terra used for the validation of MODIS skin SSTs


M-AERI Mk 2

19NASA SST Science Team MeetingSeattle, November 2010

ISAR VOS cruises for SST validation

Real-time transmission of data via Iridium, on-the-

fly validation is feasible.

20

SST radiometers - 2009 3rd Miami IR Radiometry Workshop

Traceability to SI references is a prerequisite for CDRs


Validation with buoys

Buoys provide many more opportunities of “matchups ” than radiometers.


GHRSST Diagnostic Data Set

Location of the 250 HR-DDS global data comparison locations for SST in situ and satellite retrievals.


DDS time series

Example of time series of DDS data including multiple satellite data, in situ measurements, NWP analysis fields and OI fields. This allows rapid comparison between VIIRS SSTs and other SSTs.

In situ data → LUT generation to product validation

Gather in situBuoy

MAERI, ISARReal time or retrospective

Generate extraction files

Quality control

Acquire, load SDR and reference

field inputs

1

1

ProcessSDR,

Navigate → EDR,

Matchuprecords

Analyze Matchups → Quality Test Hypercube

LUT

Update L2gen with revised

LUT and tables

2

2

Process VIIRS SDR

→ EDR, Diagnostics

Analyze Diff wrt

Reference, Time Series Hovmueller

plots

Correct algorithm as necessary,

update and re-process

0

0

A B C

D

G

E

I

F

H


Current status at L-351

• Instrument level T/V testing completed, and some optical cross-talk issues identified – but not expected to be dominant source of SST error

• Instruments integrated on NPP spacecraft at Ball Aerospace & undergoing testing• Post-launch SST validation plans being set up: coordination between May

(NAVOCEANO), Ignatov (NOAA –STAR), Emery (U. Colorado) & Evans – Minnett (U Miami)

• New validation sensors (M-AERI Mk2) being developed• Real-time data transmission being tested• Software being installed and tested, including match-ups “on the fly”• Data streams being established and tested• Anticipated validation data:

– Satellite fields (MODIS, AVHRR, AATSR)– Buoys– Radiometers (2 M-AERIs; 2 M-AERI Mk2s, 2 ISARS)

• Logical framework for feedback to improve retrievals being established


VIIRS & NPP


Summary

• VIIRS has the potential to provide high quality SSTs.• Post launch validation will focus on comparison

with:– Satellite SST fields– Buoys– Radiometers

• Contribution to SST CDR requires validation with NIST-traceable radiometers – facilitated through Miami Infrared Radiometry Workshops.


• Additional slides in reserve


Major VIIRS Objectives

• High resolution imagery with near constant resolution across scan

• Increased resolution of SST retrievals• Disaster monitoring (Volcanic ash, Suspended

Matter, Floods, Fires, …)• Increased accuracy/resolution of aerosols and

cloud properties• Climate relevant accuracies……


In situ and proxy data tasks

In Situ Measurements

MAERI


ISAR

A

A1

A2

Matchup database

RTEsimulation


MAERI

E1

E

I1

I

Telescope / HAM Synchronization Angles

Note – successive rotations of the Rotating Telescope Assembly use alternate sides of the HAM

Note – successive rotations of the Rotating Telescope Assembly use alternate sides of the HAM

FU w/ MIB Fix, F/: 6.20 FU w/ MIB Fix, F/: 6.20

M1 0.412 Ocean Color High 135.0 44.9 352 434 23.3 352 451 28.3Aerosols Low 615.0 155.0 316 717 126.6 316 747 136.2




I1 0.640 Imagery Single 718.0 22.0 119 164 37.6 119 166 39.3M5 0.672 Ocean Color High 59.0 10.0 242 294 21.4 242 297 22.7

Aerosols Low 651.0 68.0 360 549 52.5 360 556 54.3M6 0.746 Atmospheric Corr'n Single 41.0 9.6 199 320 60.7 199 323 62.1I2 0.865 NDVI Single 349.0 25.0 150 251 67.1 150 252 68.3


M8 1.240 Cloud Partical Size Single 164.9 5.4 74 122 65.2 74 122 65.2M9 1.378 Cirrus/Cloud Cover Single 77.1 6.0 83 171 107.2 83 171 107.2I3 1.610 Binary Snow Map Single 72.5 7.3 6 123 1956.7 6 123 1956.7

M10 1.610 Snow Fraction Single 71.2 7.3 342 463 35.3 342 463 35.3M11 2.250 Clouds Single 31.80 0.12 10 20 95.4 10 20 95.4M12 3.700 SST Single 353 270 0.396 0.182 117.4 0.396 0.182 117.4

I4 3.740 Imagery Clouds Single 353 270 2.500 0.549 355.5 2.500 0.549 355.5M13 4.050 SST High 343 300 0.107 0.058 85.3 0.107 0.058 85.3

Fires Low 634 380 0.423 0.316 33.8 0.423 0.316 33.8

M14 8.550 Cloud Top Properties Single 336 270 0.091 0.067 35.4 0.091 0.067 35.4M15 10.763 SST Single 343 300 0.070 0.030 133.9 0.070 0.030 132.7

I5 11.450 Cloud Imagery Single 340 210 1.500 0.414 262.5 1.500 0.414 262.5M16 12.013 SST Single 340 300 0.072 0.029 146.7 0.072 0.029 146.7

EOL Predicts, Nom Tolerance BOL Predicts, Nom Tolerance

SNR @ Ltyp or NEDT(K) @ Ttyp

Required PredictedSNR Margin

(%)

Gain Range

Ltyp or Ttyp

SNR @ Ltyp or NEDT(K) @ Ttyp

Required PredictedSNR Margin

(%)

Lmax or Tmax

Driving EDRs

PV

HC

T

LWIR

BandWave-length (µm)

Sili

con

PIN

Dio

des

PV

HgC

dTe

(H

CT

)

S/M

WIR

VIS

/NIR

FP

A

Noise Component Breakdown

(single sample, no TDI or Aggregation)

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1

Band, Gain

Rel

ativ

e M

agn

itu

de

#REF!

#REF!

#REF!

#REF!

VIIRS BandsSpectral bands are a subset of MODIS bands




ISAR validation data

Real-time transmission of data via Iridium, on-the-fly validation is feasible


Temperatures are traced to NIST

1. On-board black-body cavities have thermometers calibrated to NIST-traceable thermometers (SSEC)

2. Periodic calibration using a 3rd black body in M-AERI zenith view.

3. Periodic calibration of M-AERI system with a NIST-designed Water-Bath Black-Body target at RSMAS, using NIST-traceable reference thermometers.

4. RSMAS Water-Bath Black-Body target characterized with NIST EOS TXR

NIST EOS TXR

TXR characterizing the RSMAS WBBB

36NASA SST Science Team MeetingSeattle, November 2010

NIST water-bath black-body calibration target

See: Fowler, J. B., 1995. A third generation water bath based blackbody source, J. Res. Natl. Inst. Stand. Technol., 100, 591-599


M-AERI

Input aperture

Interferometer

Cold finger, Dewar and detectors

Cold finger, Dewar and detectors

Stirling cycle coolerStirling cycle cooler

Aft opticsAft optics


The innards


Wavelength calibration

Wavelength calibration provided by a HeNe laser

Date post:	19-Dec-2015
Category:	Documents
View:	217 times
Download:	3 times

SST from VIIRS on NPP: prelaunch preparations and post-launch validation Peter J Minnett & Robert H...

Documents