1 U.S. Department of the Interior U.S. Geological Survey MODIS Direct Broadcast Reception and Processing at the USGS Center for Earth Resources Observations and Science (EROS) Karen Zanter, USGS October 3, 2005
Dec 30, 2015
1
U.S. Department of the Interior
U.S. Geological Survey
MODIS Direct BroadcastReception and Processing
at theUSGS Center for Earth Resources Observations and Science (EROS)
Karen Zanter, USGS
October 3, 2005
2
Agenda
EROS Mission
AmericaView Program
Ground Station Overview
Processing FlowData Capture
Archive
Distribution
Products
Applications
3
EROS Mission Statement
Science: To promote and conduct applications, users, and knowledge of land information to better understand our planet
Data Access: To ensure that scientists, researchers, businesses, decision makers, and the public have ready access to land information
Data Archives: To safeguard and expand the national archive of remotely sensed land data
4
EROS Facility in Sioux Falls, SD
5
AmericaView Mission Statement
“AmericaView is a locally controlled and nationally coordinated program…
… to advance the availability, timely distribution, and widespread use of remote sensing data and technology…
…through education, research, outreach, and sustainable technology transfer to the public and private sectors.”
6
AmericaView Components
AmericaView, Inc. – National Consortium for Remote Sensing Education, Research, and Geospatial Applications
http://www.americaview.org
USGS AmericaView Project at EROS Data Center (included in the Land Remote Sensing Program)
http://americaview.usgs.gov
7
AmericaView Membership Status
8
EROS MODIS DB Ground Station
AmericaView Program funded the creation of an EROS ground station capable of receiving the real-time MODIS data
Began processing and data distribution in July 2002
Two to four images of MODIS DB data are acquired daily
E-mail notifications of each acquisition are automatically sent to registered users
Available for download 2 hours after reception
Coverage of conterminous U.S. and parts of Canada and Mexico
9
EDC MODIS DB Users
Approximately 410 registered users YTD
AmericaView/StateView Consortiums
Non-AmericaView Universities, Individual Students
Government
Industry
Foreign
10
EROS Ground Station Antennas
10m DatronUp to Three Simultaneous X-Band Data Downlink Channels (Image Data Reception)
S-Band TT&C Equipped
5.4m ViaSatUp to Two Simultaneous X-Band Data Downlink Channels (Image Data Reception)
Support for Three X-Band Data Downlink Channels
S-Band TT&C Capability
3m ViaSatUsed to Support AVHRR Data Reception
DCPF Provides Advisory Support and Maintenance
10m and 5.4m Antennas Serve as Back-Up to Each Other as Schedules Permit
10M Antenna
5.4M Antenna
11
EROS Acquisition Circle
12
MODIS DB Processing Flow
CTS
MODIS Processing System
MODISDB.usgs.gov
Raw
L0/L1/L2/L3
AmericaViewcustomer
Linux
Linux
Seamless
Sun/W2K
HTTPpull
L0/L1/L2/L3 cache
1.6 TB RAID
Prod. Svcs. SiloMODIS raw
Working copy
Raw
Offline and offsitearchive copies
L2G/L3/state
compositesProducts
Composites
Seamlessbrowse and
deliveryLinux
DDS
MWDSub-system
Linux
13
Data Capture System (DCS)
Captures X-band Image Data in a Raw Computer Compatible (RCC) Format
Utilized by Multiple EROS MissionsCurrently used for Landsat 5, Landsat 7, MODIS Direct Broadcast
Capable of Capturing a Wide Variety of Image Data
Also ingest Landsat Data via Tape or Network from International Ground Stations
ASN (Alice Springs, Australia)
Gilmore Creek / University of Alaska Fairbanks
PF1 (Poker Flat, Alaska)
DELL Commodity LINUX Servers Hosting Myriad/Mercury Computer Dual-Channel Capture Cards with USGS Device Driver and Application
Sends Data to Processing & Archive Systems
Realtime Moving Window Display Capability
14
Nearline Archive
Implemented Capability for Nearline Archive
Automatic Population of MODIS Level 0 Data after Reception
15
DistributionData can be downloaded at no cost from retrieval site:
http://modisdb.usgs.gov/modisdata/Retrieval site holds the most recent five days of dataRegistration is necessary to access the siteData files are compressed (lossless)
State-based compositesBased on the US composite48 conterminous US states, as well as the District of Columbia, and state composite can be clipped on demand by user’s request.State composites are built using the 500m composite products Users can select any and/or all of bands 1-7 and the NDVI bandThe state composite products can be obtained at no cost at: http://modisdb.usgs.gov/modisdata/state_comp_product.php
Seamless Data Distribution site at:http://seamless.usgs.gov/Daily United States Composite (only)
16
Web Site
http://http://modisdb.usgs.govmodisdb.usgs.gov
MODIS Direct Broadcast provides the most recent 5 days of MODIS Aqua and Terra data and composites of the conterminous United States.
17
MODIS DB Products
L1B (HDF format) 250m product contains bands 1 and 2 at their native 250m resolution
500m product contains bands 3 through 7 at their native 500m resolution, along with bands 1 and 2 aggregated to 500m
1000m product contains bands 8 through 36 at their native 1000m resolution, along with bands 1-7 aggregated to 1000m
L2R (HDF format)Estimates of the true land surface reflectances for the 250m and 500m bands, corrected for atmospheric scattering and absorption. Current and forecast ozone and atmospheric data are used in these corrections
L3V (HDF format)Normalized Difference Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI) products for the 250m and 500m bands, using the surface reflectance products as input
18
MODIS DB Products (cont.)
L2G (HDF and GeoTIFF format)L1B products that have been gridded (projected) to the Lambert Azimuthal equal-area projection and corrected for over-sampling at off-nadir view angles (bow-tie effect)
L2RG (HDF and GeoTIFF format)L2R products that have been gridded (projected) to the Lambert Azimuthal equal-area projection and corrected for over-sampling at off-nadir view angles (bow-tie effect)
L3VG (HDF and GeoTIFF format)L3V products that have been gridded (projected) to the Lambert Azimuthal equal-area projection and corrected for over-sampling at off-nadir view angles (bow-tie effect)
19
MODIS DB Products (cont.)Rolling seven-day composites (GeoTIFF format)
L2R data
7-day period using Aqua acquisitions
The highest NDVI value for each pixel in the composite is then determined from the group of L2R products
The composite products are available in 250m, 500m, and 1000m resolutions
Caveat: Certain pixels are being repeated in the seven-day composites, creating artifacts
Appears only in areas of high terrain (e.g. Rocky Mountains)
Most obvious when using pixels far from nadir
Plans to addressInstall the new L1 processing software, which contains SRTM DEMs
Reassess the terrain correction using the new L1 software
Revise existing composite algorithm to include distance from nadir
20
Seamless Data Distribution
21
Select Area of Interest and Download
22
III. Dissemination to Drought DSS
Weekly Products (Monday a.m.)•VegDRI•VegDRI difference•VegDRI with US drought vectors•VegDRI difference with US Drought vectors
Incorporate VegDRI into USDM
I. MODIS DB NDVI Processing System
2004 Weekly AVHRR NDVI DB
2005
1989
WeeklyMODIS
NDVI DB
2008 …>
Determine
calibration
MODIS DBS
II. VegDRI Modeling System
VegDRIModels
AVHRR & MODIS NDVI
•Smoothing•Metrics
(PASG, SOS Anomaly)
Satellite RS
•Ecoregions•Land cover•Soils•Irrigated Ag
Biophysical Data
•SPI•PDSI
ClimateData
Weekly Real-Time
ModelImplementation
Point DB for Model
Generation /CART
Drought Monitoring
23
Near-RT Drought Monitoring Support
In FY06, NASA and USGS collaborative project to expand operational use of MODIS VI data in Drought Decision Support
MODIS DB to supply U.S. MODIS VI data into VegDRI modeling subsystem
Timely delivery of VI data versus standard MODIS products
Flexibility of composite period (rolling 7-day vs. fixed 16-day)
In the next three years, this system will expand from current 7-state area to 48 states
automate calculation and delivery of VegDRI
Operational use of products in the construction of the U.S. Drought Monitor
24
Evolution of Fire Science Approach
Current NDVIFrom AVHRR
Departure from Average
NDVI~AVHRR
Time Series1989-present
- =AVHRRWeekly
Average NDVI
Transform AVHRRTo MODIS
MODIS DBTime Series2003-present
Include
New Development
Current NDVIFrom MODIS DB
- =Departure from
AverageNDVI
Create Long Term Continuity
Weekly Average NDVIDerived from AVHRR
and MODIS
Current Fire Danger Scenario
25
Improved Fire Danger Monitoring
Greater temporal and spatial resolution (250m) of vegetation condition information
Use of MODIS DB enhances capability to integrate daily vegetation updates with daily weather data
Extends the Decision Support Systems beyond the AVHRR era
New or improved fire danger forecasting algorithms
26
Long-Term Relevance to USGS/EROS
An important component for Terrestrial Monitoring, enabling multi-resolution (1000m, 500m, 250m)
Long term continuity of wide field of view sensor data products with high temporal resolution to complement moderate spatial resolution Landsat data continuity
Continue to develop the institutional knowledge base for developing new products, new uses, and new users
Leverage funding opportunities, cross-project requirements, and strengths in science, engineering, and operations for timely delivery of relevant information for decision support
27
Contact Information
Karen Zanter
AmericaView Project Chief
USGS Center for EROS
Voice: (605) 594-6945
E-mail: [email protected]