Oregon State
Hyperspectral Imager for the Coastal
Ocean (HICO): Overview and Coastal
Ocean Applications
C. O. Davis1, N. B. Tufillaro1, J. Nahorniak1
M. Corson2, B.-C. Gao2, R. Lucke2 and Z.-P. Lee3
1CEOAS,
Oregon State University, Corvallis, OR, USA
2Naval Research Laboratory, Washington, DC, USA
3U. Mass. Boston, MA, USA
Oregon State
Oregon State
Introduction and Outline
• Why Hyperspectral Imaging
• The Hyperspectral Imager for the Coastal Ocean
(HICO)
– How it came to be
– The challenge of operating on the International
Space Station (ISS)
• What we can see with HICO?
– Collecting and processing HICO data
– Comparison to other ocean color data
– Scientists and scenes from around the World
• Access to HICO data via CEOAS HICO website
– http://hico.coas.oregonstate.edu
Oregon State
Optical Components of a Coastal Scene
• Multiple light paths
• Scattering due to:
– atmosphere
– aerosols
– water surface
– suspended particles
– bottom
• Absorption due to:
– atmosphere
– aerosols
– suspended particles
– dissolved matter
• Scattering and
absorption are convolved
Physical and biological modeling of the scene is often required to analyze the
hyperspectral image.
Oregon State
Introduction to Hyperspectral Imaging
• A hyperspectral imager records a spectrum of the light from each pixel in
the scene
• Hyperspectral image analysis exploits this extra spectral information
For an open land scene:
Total spectrum for a pixel is a
Hyperspectral imager
weighted sum of the spectra
of what is in that pixel
Grass
Spectral
Decomposition
for an open land
scene
+
Dry Soil
+
k
S  i Ei  N
Leaves
i 1
Spectrum for each pixel
+
Camouflage
The imager and method of exploitation must be
tailored to the scene and the desired products.
Oregon State
HICO based on experience with PHILLS
AN-2 Aircraft
Airborne Experiments with the Portable
Hyperspectral Imager for Low-Light
Spectroscopy (PHILLS) demonstrated:
 Sensor design.
 Processing algorithms.
 Shallow water bathymetry, hazards to
navigation, and beach trafficability from
hyperspectral remote sensing data.
PHILLS image of shallow
water features near Lee
Stocking Island,
Bahamas used to
develop and validate
hyperspectral algorithms
for bathymetry, bottom
type and water clarity.
PHILLS Sensor
Oregon State
NRL Airborne Coastal Environmental
Hyperspectral Program
• 20 years end-to-end development of airborne coastal environmental hyperspectral imaging
Sensor Performance Modeling
Signal to Noise Ratio
10 nm Spectral Bins
350
Signal to Noise
GSD = 100m
Albedo = 5%
GMC = 1
f/2
300
250
f/2.8
200
150
100
f/4
50
0
400
500
600
700
800
900
1000
Wavelength (nm)
Sensor
Development
Nonlinear
Manifold
Analysis
Spiral
Development
Product
Evaluation
Flight
Campaigns
Ground / Water
Truth
Data
Processing
300
250
PHILLS-1
4
ORASIS
Spectral
Identification
Product
Extraction
Reflectance X 10
PURSUIT
Pattern
Recognition /
Classification
Sensor
Calibration
Requirements
Evaluation
200
Ground Truth ASD
150
100
50
0
0.4
0.5
0.6
0.7
Wavelength (microns)
0.8
0.9
TAFKAA
Atmospheric
Removal
Algorithm
Georectification
Oregon State
What is the Hyperspectral Imager for the Coastal
Ocean (HICO)?
HICO is an experiment to see what we gain by imaging the coastal ocean
at higher resolution from space.
The HICO sensor:
first spaceborne imaging spectrometer for coastal oceans
samples coastal regions at <100 m (400 to 900 nm: at 5.7 nm)
high signal-to-noise ratio to resolve the complexity of the coastal
ocean
Sponsored as an Innovative Naval Prototype (INP) by the Office of Naval
Research: Goal to reduce cost and a greatly shortened schedule.
Start of Project to Sensor Delivery in 16 months
Launched to the ISS September 10, 2009
HICO image of
Hong Kong, October
2, 2009.
HICO is integrated
and flown under the
direction of DoD’s
Space Test Program
Oregon State
HICO Flight Sensor - Stowed position
spectrometer
camera
lens
View port
Oregon State
HICO meets Performance Requirements
Parameter
Performance
Rationale
Spectral Range
380 to 960 nm
All water-penetrating wavelengths plus
Near Infrared for atmospheric correction
Spectral Channel Width
5.7 nm
Sufficient to resolve spectral features
Number of Spectral
Channels
102
Derived from Spectral Range and
Spectral Channel Width
Signal-to-Noise Ratio
for water-penetrating
wavelengths
> 200 to 1
Provides adequate Signal to Noise Ratio
for 5% albedo scene
after atmospheric removal
(10 nm spectral binning)
Polarization Sensitivity
< 5% (430-1000 nm)
Sensor response to be insensitive to
polarization of light from scene
Ground Sample
Distance at Nadir
92 meters
Adequate for scale of selected coastal
ocean features
Scene Size
42 x 192 km
Large enough to capture the scale of
coastal dynamics
Cross-track pointing
+45 to -30 deg
To increase scene access frequency
Scenes per orbit
1 maximum
Data volume and transmission
constraints
Oregon State
HICO Installed on the ISS on September 24, 2009
HICO
Japanese Module
Exposed Facility
Oregon State
HICO docked at ISS – Now What?
HICO Viewing Slit
Oregon State
HICO Image Locations
Locations chosen based on:
1. Location – within latitude limits of ISS orbit
2. Type – ocean, coast, land
3. Use – CalVal, Science, Navy, etc
- Currently ~300 locations identified
- New sites can be added which may
mean fewer observations of each site
due to “competition” between sites
Oregon State
HICO Processing
Level 01a –
Navigation
Level 0
Level 1bCalibration
Vicarious
Calibration
Level 1b :
Calibrated data
Multispectral
Hyperspectral
Level 1c – Modeled
Sensor bands
MODIS
MERIS
OCM
SeaWIFS
Level 2a:
Sunglint
Level 2c:
Standard APS
Multispectral
Algorithms
Products
QAA,
Products
At, adg,
Bb, b. CHL
(12)
NASA:
standards
OC3, OC4,
etc
(9)
Level 2b –
TAFKAA
Atmospheric
Correction
K532
Etc
(6)
Level 2c- :
Hyperspectral
L2genAtm Correction
Atmospheric
Correction
Methods
Level 2d:
Hyperspectral
Algorithm Derived Product
Navy Products
Diver Visibility
Laser performance
Level 2f:
Cloud and
Shadow
Atm Correction
HOPE
Optimization
(bathy, optics, chl,
CDOM ,At, bb ..etc
Hyperspectral
QAA
At, adg,
Bb, b. CHL
(12)
Level 3: Remapping Data and Creating Browse Images
CWST - LUT
Bathy,
Water Optics
Chl, CDOM
Coastal
Ocean Products
Methods
Oregon State
Radiometric Comparison of HICO to MODIS (Aqua)
Nearly coincident HICO and MODIS images of turbid ocean off
Shanghai, China demonstrates that HICO is well-calibrated
HICO
Date: 18 January 2010
Time: 04:40:35 UTC
Solar zenith angle: 53
Pixel size: 95 m
East China Sea off Shanghai
Image location
50 km
R.-R. Li, NRL
MODIS (Aqua)
Date: 18 January 2010
Time: 05:00:00 UTC
Solar zenith angle: 52
Pixel size: 1000 m
Top-Of-Atmosphere Spectral Radiance
Oregon State
Chlorophyll Comparison of HICO to MODIS (Aqua)
Nearly coincident MODIS and HICOTM images of the Yangtze River, China taken on
January 18, 2010. Left, MODIS image (0500 GMT) of Chlorophyll-a Concentration
(mg/m3) standard product from GSFC. The box indicates the location of the HICO image
relative to the MODIS image. Right, HICOTM image (0440 GMT) of Chlorophyll-a
Concentration (mg/m3) from HICOTM data using ATREM atmospheric correction and a
standard chlorophyll algorithm. (Preliminary Results by R-R Li and B-C Gao.)
Oregon State
Andros Island, Bahamas, Oct 22, 2009
RGB image
bathymetry
absorption
Oregon State
Internal waves at the Straits of Gibraltar
Generation of the
Internal Wave
Camarinal Sill
And the Tidal Boar.
Internal Wave packets
Straits of Gibraltar
HICO Image Dec 5, 2009
(R. Arnone analysis)
Oregon State
Microcystis bloom in Lake Erie
HICO Image of a
massive Microcystis
bloom in western Lake
Erie, September 3,
2011 as confirmed by
spectral analysis.
Oregon State
Birth of a New Island, Canary Islands
HICO Image of the new
underwater volcano off
the small Canary Island
of El Hierro, December
22, 2011.
Oregon State
HICO Data Distribution at OSU
• Developed HICO Public Website at OSU using published and approved for
distribution data, publications and presentations.
• Includes some example HICO data that are approved for distribution.
• OSU HICO Web site will be portal for data requests and distribution
– Data requests require a short proposal and data agreement signed by
the requestor and their institution and approved by NRL.
• Example data and data requested by that user will be available to them.
– http://hico.coas.oregonstate.edu
• Full description of the data and directions for use on the website
Oregon State
Besides the work
ongoing at NRL and
OSU and their
collaborators a
number of US and
international
Scientists have
submitted proposals
and been approved
to work with HICO
data. Projects are
listed under the
Current Project Tab
on the HICO
website.
Current HICO users
Oregon State
The HICO Team
NRL – DC
• Michael Corson, PI
• Robert Lucke, Lead
Engineer
• Bo-Cai Gao
• Charles Bachmann
• Ellen Bennert
• Karen Patterson
• Dan Korwan
• Marcos Montes
• Robert Fusina
• Rong-Rong Li
• William Snyder
NRL – SSC
•
•
•
•
•
•
•
•
•
•
Bob Arnone
Rick Gould
Paul Martinolich
Will Hou
David Lewis
Ronnie Vaughn
Theresa Scardino
Adam Lawson
Alan Weidemann
Ruhul Amin
Academic
• Curt Davis, OSU, Project
Scientist
• Jasmine Nahorniak, OSU
• Nick Tufillaro, OSU
• Curt Vandetta, OSU
• Ricardo Letelier, OSU
• Zhong-Ping Lee, U. Mass
Boston
Industry
• John Fisher, Brandywine
Photonics
Special thanks to our sponsors the Office of Naval Research, the Space Test
Program, and to NASA and JAXA who made this program possible.
Oregon State
HICO Summary (HICO Docked on the Space Station)
Japanese Exposed Facility
HICO
• Built and launched in 28 months
• Over 5000 scenes so far
• One more year of operations
• Data from OSU HICO website
Oregon State
Are we there Yet? Need beyond HICO
A new Free flying Hyperspectral Imager (CWR) would offer several key
advantages over HICO:
• Broader bandwidth. CWR will collect data from 380 to 1650 nm. The longer
infrared wavelengths will make it possible to characterize beaches and plants
along the shoreline and snow and ice in alpine regions and in the Arctic.
• Finer spatial resolution. CWR will collect data with 30-m GSD, which will be
able to re-solve ocean bottom, coastal, glacier and arctic features.
• Optimized orbit. CWR will fly on a smallsat in a near-noon equatorial crossing
time polar orbit. It will be able to image locations in high latitudes every day and
locations in mid to low latitudes every 2nd or 3rd day.
• Wide field of regard. Agile spacecraft to rotate CWR through 60 degrees, from
-30˚ to +30˚ from nadir. Allows rapid revisit (1-3 days) to study sites.
• Optimized radiometric calibration. CWR will calibrate signal brightness onorbit by scanning the full moon once a month.
• Optimized spectral calibration. CWR will calibrate signal wavelengths on-orbit
to within 0.2 nm by comparing observed and known spectral features including
Fraunhofer absorption lines and atmospheric absorption features.
• Complete data processing system. HICO data processed to level 1B, need
routine full processing to geolocated products, including reprocessing as
needed.