Upload
others
View
13
Download
0
Embed Size (px)
Citation preview
The DESIS Hyperspectral Instrument A New Space-Based Tool for Coastal Zone Monitoring
February 6, 2017 Ray Perkins1, Rupert Müller2, Emiliano Carmona2
1 Teledyne Brown Engineering, 2 Deutsches Zentrum für Luft- und Raumfahrt (DLR)
Teledyne and DLR Partnership
2
► Teledyne & DLR, the German Space Centre, have partnered for the DLR Earth Sensing Imaging Spectrometer (DESIS) to be hosted on Teledyne’s Multi-User System for Earth Sensing (MUSES) mounted on the International Space Station (ISS)
► MUSES is an Earth-imaging platform designed, built, owned and operated by Teledyne
• A commercialization project for the ISS • Hosts up to 4 robotically installed & removed instruments • Provides precision pointing for hosted earth observing instruments • Provides all EO mission planning, control, and data downlink • MUSES commercial remote sensing license from NOAA
► DESIS is a Visible to Near-InfraRed (VNIR) Imaging Spectrometer
• Designed and built by DLR • Operated by Teledyne • DLR uses DESIS for scientific research and humanitarian purposes • Teledyne uses DESIS for commercial purposes
Multi-User System for Earth Sensing (MUSES)
3
► MUSES Platform ► Located on ELC 4-2 ► Inertially stabilized ► Precise pointing and Earth surface
target tracking ► Up to 4 robotically installed
instruments ► Total data downlink ~225 GB/day ► Teledyne owns the platform,
determines pointing schedules, and retains data rights in cooperation with partners
► Instruments launched in “soft stowage”
MUSES Location on ELC-4
4 4
Teledyne’s Geospatial Solutions
5
► Turn-Key Instrument Missions • Development, integration, and operations. • Negotiable Template: 30-month engagement;18-month
integration,12-month flight on MUSES (6 month for primary) • TRL maturation, prove-out of instrument design, early data
products prior to free-flyer missions. ► Space Qualification of Hardware
• Reusable container for the MUSES platform • Standardized power, data, and mechanical interfaces
► Hyperspectral Data Sales • Archive, tasking, subscription • Non-exclusive, limited exclusivity, exclusivity options
► Data Analytics and Decision Information Products • Commercial, scientific, humanitarian value • Collaborative Research and Development • Multi-sensor fusions (MUSES & non-MUSES sensors)
► Payload Operations as a Service • Plan and operate ISS and free-flying satellite payloads from
the Teledyne Tele-Science Center ► Content as a Service
• Host partner geospatial data (imagery, point clouds, etc.) using Teledyne’s AWS cloud-based archive & point of sale infrastructure
► Hosted Workflows • Host partner workflows on Teledyne’s AWS cloud-based
flexible processing architecture
Platform Capabilities
6
Pointing Knowledge ≤ 30 arc seconds (~ 60 m on ground from 400 km altitude) Field of Regard 5° outboard cross-track
45° inboard cross-track +/- 25° along- track
Star Tracker Sodern SED26 Inertial Measurement Unit Honeywell Miniature Inertial Measurement Unit (MIMU) Precision Time Sourced from the ISS GPS, ± 250 usec to MUSES
instruments Location knowledge Sourced from the ISS GPS, ± 50 meters, RMS Orbit 51.6° Inclination, 400 km altitude ± 5% (nominal) Data Processing Linux Server on-board ISS with redundant 8 TB storage Daily Downlink Capacity 225 GB
DLR Earth Sensing Imaging Spectrometer
7
Characteristic DESIS-30 Features
Ground Sampling Distance 30 m @ 400 km altitude
Ground Swath 30 km @ 400 km altitude
Spectral Range 400 nm – 1000 nm
Spectral Bins Measured: 235 @ 2.55 nm Programmable binning on-orbit
Quantization 12 bits + 1 gain bit
Signal to Noise Ratio @ 550 nm 205:1 sampled at 2.55 nm 406:1 binned to 10.2 nm
On-board calibration Dark Field for DSNU LED Array for PRNU
Independent Pointing Pointing Unit ±15° Along Track
Independent Time and Location On-board GPS
Management of Agricultural and Forest
Ecosystems
Hazard Assessment
Inland Water Quality Monitoring
Drought Impact Assessment
Hyperspectral Application Areas
8
► Multi-sensor Data Fusion • Enhance spatial resolution
► Land Use/Land Cover • Land cover classification • Vegetation health assessment • Biomass assessment • Drought impact assessment • Hazard assessment
► Forestry • Forest species identification and
assessment • Forest health monitoring • Biomass assessment
► Aquaculture • Coastal zone monitoring • Water quality monitoring • Oil spill monitoring and assessment
Fusion of Multispectral and Hyperspectral Data
9
WV-2 (~2 m, MS 8 bands) Fusion Simulated DESIS (30 m, HSI)
Land Use and Land Cover
► Land cover classification using hyperspectral data • Improved classification accuracy1, 2 • Improved differentiation of crop canopy and crop growth stages3
• Is effective in urban land cover classification4
► Vegetation health assessment • Solar-induced chlorophyll fluorescence5
• Biophysical indicators (Wet biomass, LAI, N, Chlorophyll A & B)6, 7
► Biomass assessment • High potential for estimating biomass of grassland habitats8
10
1. Thenkabail P.S.; et. al. Hyperspectral Remote Sensing of Vegetation, CRC Press, Boca Raton, FL, 2012, pp 24-31 2. Wilson, J. H.,; et. al. Separating Crop Species in Northeastern Ontario Using Hyperspectral Data, Remote Sens. 2014, 6, 925-945; doi:10.3390/rs6020925 3. Vijayan, D., Shankar,;et. al. Hyperspectral Data for Land use /Land Cover Classification, ISPPR Archives, Vol XL-8, 2014. 4. Tan, Q.; Wang, J. Hyperspectral Versus Multispectral Satellite Data for Urban Land Cover and Land Use Mapping – Beijing, an Evolving City, ASPRS 2007 Annual
Conference, May 2007. 5. Frankenberg, C., et. al. Remote sensing of near-infrared chlorophyll fluorescence from space in scattering atmospheres: implications for its retrieval and interferences
with atmospheric CO2 retrievals, Atmos. Meas. Tech., 5, 2081-2094, doi:10.5194/amt-5-2081-2012, 2012. 6. Zarco-Tejada P. J.,; et. al. Temporal and spatial relationships between within-field yield variability in cotton and high-spatial hyperspectral remote sensing
imagery," Agronomy Journal 97(3), 641-653 (2005). 7. Cammarano, D.; et. al. Assessing the Robustness of Vegetation Indices to Estimate Wheat N in Mediterranean Environments, Remote Sens. 2014, 6, 2827-2844;
doi:10.3390/rs6042827. 8. Psomas, A.; et. al. Hyperspectral remote sensing for estimating aboveground biomass and for exploring species richness patterns of grassland habitats. International
Journal of Remote Sensing, Vol. 32, No. 24, Dec 2011, 9007-9031
Forestry – Monitoring and Assessments
► Forest Stand Mapping • Use of spectral unmixing and endmember classification to characterize
stands in mixed forests1, 2
► Pest Infestation/Damage • Use of spectral signatures to recognize pine beetle infestation3
► Forest monitoring • Generation of above-ground carbon, reforestation, afforestation and
deforestation maps4
11
1. Darvishsefat, A.; Kellenberger, T.; Itten, K. Application of Hyperspectral Data for Forest Stand Mapping, Symposium on Geospatial Theory, Processing and Applications, Ottawa, 2002
2. Tong, Q.; Jiao, Q., Zhang, X.: Forest Species Classification Based on Hyperspectral and Multitemporal CHRIS Images, 4th Chris Proba Workshop, Session 3.
3. White, J.; et. al., 2007; Detecting mountain pine beetle red attack damage with EO-1 Hyperion moisture indices, International Journal of Remote Sensing. Vol. 28, No. 10, pp. 2111-2121.
4. Goodenough, D.; et. al., Monitoring Forests From Space: Hyperspectral And Kyoto Products, World Forestry Congress, 2003, http://www.fao.org/docrep/ARTICLE/WFC/XII/0888-B1.HTM.
Aquaculture – Coastal Zone Monitoring (HySpex Hyperspectral Data)
Before Denoising After Denoising
12
Starnberger See, Germany Noisy & ‘clean’ bands
Zoomed Areas
0
1
[mg/l]
Absorption Estimation (WASI Tool) of Coloured Dissolved Organic Matter
(Error in model fit drops 50% after denoising)
Aquaculture – Estuary and Inland Water Monitoring
13
Hyperspectral Imager for the Coastal Ocean (HICO): Application of Space-based Hyperspectral Imagery for the Protection of the Nation’s Coastal Resources, Dr. Darryl Keith ([email protected]), July 2013
Using atmospherically corrected HICO imagery and a comprehensive field validation program, regionally-tuned algorithms were developed to estimate the spatial distribution of chlorophyll a, colored dissolved organic matter, and turbidity for four estuaries along the northwest coast of Florida from April 2010 – May 2012.
“HICO helped show us that it is possible for a hyperspectral space-based sensor to produce products that meet the needs of EPA.”
Chlorophyll a Turbidity Colored Dissolved Organic Matter
Oil Spill Identification and Tracking
► Adding VNIR hyperspectral imagery to radar can:
• Extract oil spill chemical composition signatures1
• Aid in spill classification1 and reduce the false alarms2 of suspected oil spills3
• Reduce errors in low and high winds2, 4
• Detect oil spills below the surface5
14
1. Salem, F.; Kafatos, P. Hyperspectral image analysis for oil spill mitigation. Proc. ACRS 2001—22nd Asian Conference on Remote Sensing, 5-9 November 2001, Singapore, vol. 1 (pp. 748– 753)
2. Mityagina, M.; Lavrova, O. Satellite Survey of Inner Seas: Oil Pollution in the Black and Caspian Seas. Remote Sens. 2016, 8, 875; doi:10.3390/rs8100875. 3. Brekke, C.; Solberg, A. Oil Spill Detection by Satellite Remote Sensing. Remote Sensing of Environment 95 (2005), 1-13; DOI: 10.1016/j.rse.2004.11.015. 4. Ferraro, G.; et. al. Long Term Monitoring of oil spills in European Seas. International Journal of Remote Sensing, Vol. 30, No. 3, 10 February 2009, 627–645 5. Otremba, Z. Oil Droplet Clouds Suspended in the Sea: Can They Be Remotely Detected? Remote Sens. 2016, 8, 857; doi:10.3390/rs8100857.
Current MUSES and DESIS Status
15
► MUSES • Completed closeout of pre-launch verifications • Completed pre-launch performance characterizations • Ship to launch site 6 February, 2017 • Manifested for launch on SpaceX-11, April, 2017 • MUSES commissioning during Q3/Q4, 2017
► DESIS • Critical Design Review completed June 2016 • Planned launch on SpaceX-13, Q4, 2017 • DESIS commissioning during Q1/Q2,2018
www.teledyne.com 16
17
Backup Slides
Earth Observation From the ISS – Why It Works
18
► Coverage of ~90% of populated Earth
► Coverage of ~100% of ocean shipping lanes and major navigational ports
► Coverage of 100% of tropics and equatorial region
► Sophisticated spacecraft bus with required resources
► Upgrade and exchange of instruments as technology and/or markets evolve
► Traditional barriers to entry minimized
MUSES Imaging Opportunity Analysis
19
• Selected 12 point targets • From ~50° North to ~50° South latitude • Generally at 10° latitude intervals
MUSES Imaging Opportunities
20
DLR Earth Sensing Imaging Spectrometer
21
► Teledyne and DLR have partnered to build and operate the DLR Earth Sensing Imaging Spectrometer (DESIS) from the Teledyne-owned MUSES Platform on the ISS
► Teledyne retains rights for commercial use
► DLR retains rights for scientific use
► Launch planned for Q4, 2017
► The DESIS Instrument will be used to • Enable scientific RESEARCH • Expand HUMANITARIAN response • Provide COMMERCIAL value
DESIS Signal to Noise (SNR)
22 SNR for 2.55 nm sampling distance and spectral binning by factor 4
DESIS Pointing Unit
23
► Changes sight ±15° in the along-track direction
► Earth Sensing Mode • 11 measurement positions ±15° (every 3°) • Repeatability / accuracy 20 arc minutes • Target replacement time ≤ 0.5 seconds
► Stereo Mode • Collection of up 3 image tiles at different
angles • Used for BRDF & altitude extraction
► Forward Motion Compensation Mode (experimental)
• Used to increase SNR for specific targets • Speed 0.6 deg/sec and 1.5 deg/sec • Accuracy 0.06 degrees (1/10 GSD) • Range of rotation ±15°
TBE GS
Partnership: DLR and TBE Data Processing
24
Teledyne Tele-Science Center
DLR GS
DLR DESIS Data Management System
Data Storage DESIS Data + Auxiliary Data
Data Request
L1B – ToA Correction
L1C - Orthorectification
L2A – Atm. Correction
DLR Requestor
L1B – ToA Correction
L1C - Orthorectification
L2A – Atm. Correction
L1A – Transcription
Order Management
Catalog DESIS Archive +
Auxiliary Data
Tasking
Image Data & Telemetry
Tasking
Data Request Products
DESIS data + Auxiliary data
Image Data & Telemetry
L1A – Transcription
Products
Share Calibration Share Validation
Teledyne Earth Sensor Portal
Processing
Commercial Customers Scientific Partners
Processing Chain
25
• Extract & evaluate dark current data • Tiling & reformatting of raw HSI • Processing of Calibration Data • Screen raw data • Derive dead/suspicious pixel mask • Generate quicklook image • Metadata generation
L1A
• Input L1A HSI tiles + dark current data • Apply systematic and radiometric
corrections (housekeeping and AOCS data appended).
• Extract quality indicators • Top of Atmosphere radiance.
L1B
• L1B data is orthorectified and resampled • Direct georeferencing using DEM • Map projection.
L1C
• Atmospheric corrections using ATCOR • At Surface reflectance L2A
Long Term Archive L1A Data
Level 1A Processor Transcription
Earth Imagery Experimental Imagery
Calibration Measurements
Calibration & Reference Products
Position & Attitude Products
Auxiliary Data
Screening
In-flight Calibration
Process (offline)
Calibration & Reference Products
Update Cal Tables
Level 1B Processor Systematic &
Radiometric Correction L1B Product
Level 1C Processor Orthorectification L1C Product DEM Database
REF Database
Level 2A Processor Atmospheric Correction
L2A Product Atmospheric LUT