Geoscience Australia: Advances on Analysis Ready Data (ARD)
Geoscience Australia: Advances on Analysis Ready Data (ARD) Surface Reflectance from Landsat, Sentinel-2, SPOT Surface Brightness Temperature from Landsat 5,7,8 Land Surface Imaging Virtual Constellation meeting #3 (LSI-VC3), Frascati, March 20-21, 2017 15 minutes The demand for ARD is growing radar time series The demand for ARD is growing landcover
The demand for ARD is growing multi-sensor approaches are replicating pre-processing chains! The demand for ARD is growing per pixel metadata And there are benefits to users Time saved using [GA] for all of the calculations; The team that participated in the trial has estimated a 30% time saving, compared to current work practices. This is based on work done on the two trial areas.
However once the team has access to Version 2, and become more familiar with the platform, productivity time will improve significantly Efficiencies will come from obtaining the standard products of the data cube and being able to use indices which are not being used in current projects. Analysis Ready Data for the Murray Darling Basin Geometric median of Landsat 8 SR No visible artefacts (satellite passes are not visible in the data!)
Standardisation of the illumination and view angles is effective. NBAR corrections are applied 2015 calendar year Bands 6,5,3 Sentinel-2 SR Longreach QLD Corrections applied to Landsat
can be applied also to Sentinel-2 Geoscience Australia in development Queensland Uni well advanced (separate presentation) SPOT-5 ARD Stuart Phinn (Queensland Univ.) has a process for atmospheric correction, terrain
correction and NBAR for SPOT-5. The SEO team was planning to send him a few SPOT-5 files over Kenya to process into Surface Reflectance (SR). We would select an area where we have Kenya data cube content. Stuart would process the data into SR products. The SEO team would test them in QGIS (visual inspection against Landsat data). If the data looks promising, we will need to create a document for the processing steps, create a Python script for users to perform the same processing, and create a Data Cube ingestor. Example on right is SPOT-5 over Africa in Jan 1-8, 2014 (RED) using the COVE tool.
*Supplied by Brian Killough, CEOS Systems Engineering Office (SEO) CEOS ARD Status for Data Cubes *Supplied by Brian Killough, CEOS Systems Engineering Office Surface Brightness Temperature (SBT) and Analysis Ready Data Surface Brightness Temperature (TB) is another family of analysis ready data
proposed as Threshold CARD4L by Geoscience Australia and USGS. TB can be retrieved from thermal channels on Landsat (L5, 7, 8) using radiative transfer models (MODTRAN 5). The key input is globally available ancillary atmospheric data. Geoscience Australia compared approaches in 2015 with 60 m Landsat. The study used 3 dates x 3 sites, x 4 global datasets Sites included Darwin (coastal, 12.5 S), Moree (inland, 29.5 S), Wollongong (coastal, 34.4S) Target CARD4L would require use of Land surface emissivity corrections Surface Brightness Temperature in K (02/05/2003)
Balloon NCEP1 coincident, ground released (this is the reference) NCEP2
Reference: Li et al., 2015 Land surface brightness temperature retrieved from Landsat data. http://www.mssanz.org.au/modsim2015/L11/li.pdf MERRA ECMWF (best overall) NCEP1 (NOAA NCEP/NCAR Reanalysis I
data) NCEP2 (NOAA NCEP-DOE Reanalysis II) MERRA (NASA Modern Era RetrospectiveAnalysis for Research and Applications)
ECMWF (European Centre for MediumRange Weather Forecasts) MODSIM 2015, Gold Coast, Australia, November 29 to December 4, 2015 Landsat Surface Brightness Temperature (oK) image of Landsat 7 image on October 6, 2000; (a) using ground based radiosonde (atmospheric profile) measurements;
(b) using MERRA atmospheric profile; (c) using NCEP1 atmospheric profile and; (d) using NCEP2 atmospheric profile Brightness temperature comparison Best results, Moree, 2000 Worst results,
Darwin, 2003 * Presented in Land Surface Brightness Temperature Retrieval from Landsat Data at 21st International Congress on Modelling and Simulation MODSIM 2015, Gold Coast, Australia, November 29 to December 4, 2015 Data Cube Ingestion and Interoperability Analysis Ready Data, Ingestion and Indexing User
User Raw data Analysis Ready Data This saves users expertise and resources! ($)
User Ingester Data Cube User User User Data Provision
Data Exploitation This allows the time series to be leveraged Ingestion to Open Data Cube is a simple process Longitude latitude range s2a_l1c = dc.load(product='s2a_level1c_granule',x=(147.36, 147.41), y=(-35.1, -35.15), measurements=['04','03','02'], output_crs='EPSG:4326', resolution=(-0.00025,0.00025))
Spectral bands / measurements of interest Analysis grid representation http://agdc-v2.readthedocs.io/en/stable/index.html Open Data Cube Ingest 1) Decode existing data files Read multiple file formats 2) Define translation of data files to
Storage Units Target storage unit type, format and shape Reproject or resample data if desired Reformat or reshape if desired 3) Pass job to a Storage Unit Writer Handle file locking etc http://www.datacube.org.au/wiki/Ingest How Sentinel-2 bands (20m) would look in the AGDC
database Band names from S2 User Guide Alias band name JSON Spectral response for Spectral Ignorance tools
Spectral Ignorance concept, sensor band equivalence Supporting multi-sensor AGDC Sentinel-2A Band 2 (Blue) and Landsat 5 TM Band 1 (Blue) Phone: +61 2 6249 9111 Web: www.ga.gov.au Email: [email protected] Address: Cnr Jerrabomberra Avenue and Hindmarsh Drive, Symonston ACT 2609 Postal Address: GPO Box 378, Canberra ACT 2601
What is Optimized Academic Scheduling? Charged by the Provost. Changing the way in which the University assigns general use classroom space. Utilizing EMS, used for many years for campus reservations, to optimize the assignment of these spaces
What's in it for Me? You want to advance your interests (so do they) You want the other side to agree and be happy about it. Or as Daniele Vare said "Negotiation is the art of letting them have your...
(Kant, I.: Crítica del juicio) "Y resulta que el sentimiento de placer surge cuando en la representación de un objeto el sujeto siente modificada su subjetividad del siguiente modo: surgimiento de una perfecta coincidencia entre la facultad de las intuiciones...
J. Trzeciak. November 2012 "provides best practice research and practical advice to academic, business, and student affairs leaders" "teams of consultants and analysts dedicated to uncovering the best ideas from across higher education" ...
Psalm 1. Which Way. Psalm 1:1-6. Choices…we are our choices. Jean Paul Sarte. the way of the godly. the . PRACTICE. of the BLESSED. Is the counsel helping me to deal with the truth or just helping me to believe...
Higher power reveals that bone formation is by a "metaplastic" process from the fibrous stroma rather than by rows of osteoblasts forming lamellar bone This is a cellular fibrous stroma containing immature bone less cellular and active fibrous connective tissue...
Herman & Chomsky (1988) highlight volume of coverage as a key aspect of their media propaganda model, and Carvlho points out that "[t]he volume of media coverage is the first indicator of the relative salience awarded to an issue over...
Ready to download the document? Go ahead and hit continue!