1. Evaluation of Recent VIIRS Sensor Performance in the Coastal Ocean
Sherwin Ladner1, Robert Arnone2, Adam Lawson1, Giulietta Fargion3,
Jennifer Bowers4 , Paul Martinolich4 , Michael Ondrusek5 , Michelle Wood6,7
1Naval Research Laboratory, Stennis Space Center, MS, USA
2University of Southern Mississippi, Stennis Space Center, MS, USA
3San Diego State University, San Diego, CA, USA
4QinetiQ Incorporated, Stennis Space Center, MS, USA
5NOAA NESDIS, College Park, MD, USA
6NOAA/AOML Ocean Chemistry Division, Miami, FL, USA
7Current Address: Dept. Biology and IE2, Univ. Oregon, Eugene, OR, USA
We would like to acknowledge our sponsors at the
JPSS Program Office for support of this work.
FALL AGU San Francisco, California – December 03-07, 2012

2. Objectives
Evaluate current NRL and IDPS VIIRS Environmental Data Products
nLw (l) – Water Leaving Radiance
Chlorophyll, backscattering (bb551)
Coastal AERONET OC Platform Time Series - nLw(l)
AAOT - Venice, Italy
Calibration & Validation Ship Cruises
Hawaii (NOAA/Ondrusek)  nLw(l)
Gulf of Mexico (NMFS Cruise – Mitch Roffer, John Lamkin)  CHL
US East Coast (GOMECC2 Cruise – M. Wood)  nLw(l)
Navy Panama City Slocum Glider (Optics Only)  bb(551)
Inter-sensor comparison with MODIS (nLw(l),bb(551),CHL)

3. VIIRS – NPP Operational Software for Ocean Color Processing
n2gen software (NRL,NASA) – R&D
Calibration applied to SDR (AFWA/NAVO)
Atmospheric correction – GW NIR w/ 80 aerosol models
Glint / Cloud Removal
In water Algorithms – QAA - Coastal iteration
NOAA Operational IDPS From CLASS (Mx6.0-Mx6.5)
Research Software #1 used to evaluate VIIRS sensor
and identify issues and improve IDPS VOCC software

4. CURRENT IDPS OPERATIONAL SYSTEM Mx 6.2 CURRENT NRL APS OPERATIONAL13
CURRENT NRL APS
OPERATIONAL
SYSTEM 4 9 39
August 03, 2012
FUTURE Mx6.6 IDPS
OPERATIONAL
SYSTEM PLANNED
FOR
JANUARY 2013
Inland
lakes
Cloud
Masks
Similar
Higher
Coastal
CHL
Aqua
4.7
Masked
Pixels
(in Black)

5. Coastal Chlorophyll Retrievals from VIIRS
Martha Vineyard’s Coastal Observatory (MVCO)
Average Chl range MVCO = 4-6 mg/m3 (August 2011)
NRL vs. IDPS v6.4
August 03, 2012 US East Coast
NRL Chlorophyll
IDPS v6.4 Chlorophyll
1:1
Line
Some
Coastal
Chlorophyll
Values
Exceeding
200 mg/m3
1. IDPS processing Chlorophyll is to high
2. N2gen retrieving accurate Chl .
VIIRS Sensor is showing good agreement with observations

6. n2gen Inter-Sensor Matchups (VIIRS vs MODIS n2gen)
US East Coast – August 03, 2012
VIIRS = 1803 GMT AQUA = 1810 GMT
nLw nLw nLw 486
nLw Backscattering Chlorophyll
551nm QAA OC3M
VIIRS and MODIS Aqua agree well and are within time and space uncertainty of the products

7. NASA AERONET-OC (GLOBAL NETWORK) 15 Sites that have approved L2 data – nLw(l)
These sites have met the deployment requirements:
Platform for unobstructed sky and sea viewing;
2) Bottom effects
3) Superstructure perturbations
AAOT
P-height
S-distance
Water depth at which the bottom perturbation adds a contribution of 1% to the SeaPRISM LWN as a function of seawater diffuse attenuation coefficient Kd and irradiance reflectance R (the latter values are defined by the curves in black) assuming a Lambertian seabed irradiance reflectance RB=0.10).
Just going to show time series matchup at AAOT. Are doing many others to monitor radiances and gain stability. Right part of slide is requirement of platform
3 New Coastal US Sites:
LISCO Site: Western Long Island Sound
The observed surface area should be at a distance from the main superstructure larger than the height of the superstructure itself.
WaveCIS_Site_CSI_6' site: Gulf of Mexico
SeaPrism
Eureka - UCS site: South California Bight (Newport Beach)

8. n2gen - nLw Matchup @ AAOT – Venice Italy
April – October 2012
V V6
V V6
Operational July 19th
V V6
V V6
What is V5 and V6 – The basis of this reprocessing is the MODIS Calibration Support Teams (MCST) look-up table (LUT) V (Based on SeaWiFS Cross Calibration). This is the same LUT that is being applied for the Collection 6 reprocessing of MODIS Aqua Land and Atmosphere products (completed in early 2012 for L1B products, on-going as of this writing (April 2012) for L2 products; note that OBPG actually used the update V for reprocessing). It includes time-varying corrections for changes in response versus scan angle (RVS) for bands 8 and 9, which were derived using observations of desert sites. These desert-based corrections largely replicate what the OBPG had previously derived from cross-calibration to SeaWiFS, and previously applied in recent Aqua reprocessings. The results using this LUT alone, however, still showed significant artifacts (residual cross-scan variations and detector/mirror-side striping in derived products). To mitigate these effects, the OBPG developed a cross-calibration approach similar to Meister et al. 2011, but using global MODIS-Aqua Level-3 water-leaving radiance retrievals as the vicarious calibration source. This Aqua-to-Aqua calibration (see below, 'Additional Information') is used to derive temporal adjustments to RVS, relative to the MCST calibration, that significantly reduce residual striping and cross-scan artifacts.
SeaPrism:MODIS SeaPrism:VIIRS MODIS:VIIRS
wavelength slope r^2 slope r^2 slope r^
MCST Look-up Tables (Temporal RVS)
V5 = Calibration Issue with 412 & 443nm (SeaWiFS Cross-Calibration)
V6 = Calibration Correction (Desert Sites)

9. n2gen VIIRS/MODIS Rrs vs Hyperpro Matchup
Hawaii: September 08-18, 2012
Rrs = nLw/f0
Honolulu
2 Stations
MODIS
9/11/12
MODIS
9/12/12
Chl
MODIS(2)=Cloud
Chl
IDPS(2) = Masked
VIIRS
9/11/12
VIIRS
9/12/12
Mike/NOAA/Hyperpro. Similar matchups with NOAA MSL12 and NASA l2gen.
1 Station
IDPS(1) = Masked
Black Pixels
Are Masked
Mx6.2
VIIRS IDPS
9/11/12
Black Pixels
Are Masked
Mx6.2
VIIRS IDPS
9/12/12
Data Provided
By M. Ondrusek
NOAA Hyperpro

10. n2gen VIIRS/MODIS Rrs vs Hyperpro Matchup
Hawaii: September 08-18, 2012
Rrs = nLw/f0
4 Stations
Honolulu
VIIRS
9/15/12
VIIRS IDPS
9/15/12
Black Pixels
Are Masked
Mx6.2
Chl
MODIS(all)=Glint
IDPS(all) = Masked
Data Provided
By M. Ondrusek
NOAA Hyperpro
Honolulu
VIIRS
9/17/12
Mike/NOAA/Hyperpro. Similar matchups with NOAA MSL12 and NASA l2gen.
VIIRS IDPS
9/17/12
Black Pixels
Are Masked
Mx6.2
Chl
2 Stations
MODIS(all)=Glint
IDPS(all) = Masked

11. n2gen VIIRS/MODIS Rrs vs Insitu Matchup
GOMECC2: July 22 – August 13, 2012
Aqua VIIRS Aqua VIIRS
FL FL FL FL
Glint
Glint
Glint
Rrs = nLw/f0
1 Station
2 Stations
Michelle/NOAA/ASD.
IDPS(1) = NA CLASS
VIIRS(2) = Glint
IDPS(2) = Masked
Data Provided
By M. Wood
NRL ASD

12. n2gen VIIRS/MODIS Rrs vs Insitu Matchup
GOMECC2: July 22 – August 13, 2012
VIIRS VIIRS IDPS Mx6.0
Glint
Delaware
Bay
NewYork
Rrs = nLw/f0
2 Stations Station
Michelle/NOAA/ASD.
MODIS(1)= Glint
MODIS(all) = Cloud
IDPS(all) = NA
Data Provided
By M. Wood
NRL ASD

13. Gulf of Mexico Cruise: April 02 – May 28, 2012
190 Stations
April 2012 Monthly Mean
Chlorophyll
Data from Mitch Roffer and John Lamkin
and NMFS cruise
CTD – Fluorometer (surface ) NOT HPLC!

14. Gulf of Mexico Cruise – April 02 – May 28, 2012
NRL n2gen
~55% More Retrievals
Than IDPS (Masking)
IDPS
Errors in VIIRS CHL compared to insitu CHL
Bias [avg(logC1-logC2)] = 0.088
RMSE[avg(logC1-logC2)2] = 0.029
Average Relative Error = 13.4%
N = 71
One problem from literature (Campbell & O’Reilly, Metrics for Quantifying the Uncertainty in a Chlorophyll Algorithm,2006) with using dataset like NOMAD is that the distribution of the data is not representative of the global chlorophyll (seawifs). Errors increase with increasing chl values. NOMAD Matdhup y-axis is bias. Check literature to make sure.
log(VIIRS CHL) – log(insitu CHL)

15. Panama City, FL Slocum Glider Deployment
November 08 – November 16, 2012
bb532
11/08/12
bb532
11/09/12
bb532
11/10/12
IDPS
IDPS
IDPS
Panama Panama Panama
City, FL City, FL City, FL
VIIRS
VIIRS
VIIRS
Average Relative Error = 5.75%, 13.24%
Insitu (glider) vs. Satellite (VIIRS,IDPS)
Backscattering 532nm
Glider Near Surface (<2m)
BB Slope Algorithm (Gould, et.al - ). Atmospheric Correction / different aerosol models. Linear relationship between wavelength. Know bb at one bb estimate for another.
11/08
11/09
11/10
Glider
n2gen
IDPS
Glider
1514 – 2131
VIIRS
1850
Glider
1850 – 2212
VIIRS
1905
Glider
1500 – 2112
VIIRS
1846
Differences in Processing
Aerosol models, IOP Algorithm, etc.

16. Summary : VIIRS Ocean Color
Evaluated VIIRS ocean color EDR products using processing from NRL’s n2gen (R&D) and VOCCO (IDPS)
VIIRS Coastal Ocean Color EDR’s (nLw/Rrs, Chlorophyll and bb 551) compare well with insitu Ship/AERONET-OC and MODIS
IDPS EDR’s products are over masking (Mx ) – limited matchups
Update of IDPS EDR’s (Jan Mx6.6)
- suggest setting / using quality flags
Monitoring sensor and radiometric calibration stability using global validation network  requires IDPS updates (Mx6.6)
Follow-on cruises and validation planned in near future
Ongoing monitoring and evaluation of VIIRS products
Navy’s initial assessment of VIIRS ocean color products indicates ocean products are of high quality and plans for operational use
APS v4.10 currently going thru OpEval at NAVO

17. Sponsor Acknowledgements:
Questions?
Sponsor Acknowledgements:
JPSS Program Office
NOAA Ocean Acidification Program (GOMECC2 Cruise)
Disclaimer (NOAA co-authors):
The opinions expressed here are those of the individual scientist/authors and not the organizations.

18. n2gen VIIRS/MODIS Rrs vs Insitu Matchup
GOMECC2: July 22 – August 13, 2012
VIIRS
VIIRS
MODIS Aqua
MODIS Aqua
VIIRS
Y=1.097x
R2=0.82
VIIRS
Y=1.0357x
R2=0.88
MODIS Aqua
y = x
R2 = 0.86
MODIS Aqua
y = x
R2 = 0.99
MODIS Aqua
MODIS Aqua

19. Comparison of Processing Software Packages
IDPS and NRL APS (n2gen)

20. Panama City, FL Slocum Glider Deployment
November 08 – November 16, 2012
bb 532 Time Series (Entire Deployment)
Slocum Glider
Bottom = *
VIIRS bb 551nm -> 532nm
Insitu (glider) vs. Satellite (VIIRS) – Near Surface (<2m)
11/08
11/09
11/10
11/08
11/09
11/10
Glider
1514 – 2131
VIIRS
1850
1850 – 2212
1905
1500 – 2112
1846
Average Relative Error = 5.75%
Slope=0.95
R2=0.84
Panama
City, FL
11/10/12
7daylp
BB Slope Algorithm (Gould, et.al - ).

21. Inter-Sensor Matchups (VIIRS vs MODIS n2gen)
Navy Ocean Optical Backscattering Product – US East Coast
Bb 551nm
QAA
Bb 547nm
QAA
08/03/2012
MODIS
1810 GMT
NPP VIIRS
1840 GMT
U.S.
East
Coast
U.S.
East
Coast
1/m

22. n2gen VIIRS/MODIS Rrs vs Insitu Matchup
GOMECC2: July 22 – August 13, 2012
VIIRS
VIIRS
MODIS Aqua
MODIS Aqua
GOMECC2 Satellite (VIIRS-n2gen) vs. Insitu (ASD) Rrs
Average Relative %Error:
410nm nm nm nm
VIIRS(5)
MODIS(3)
VIIRS
VIIRS
MODIS Aqua
MODIS Aqua 22