The Mineral Map of Australia:

The realisation of a dream

Thomas Cudahy, Mike Caccetta, Simon Collings, Carsten Laukamp, Matilda Thomas, Cindy Ong,
Ian Lau, Andrew Rodger, Rob Hewson, Joanne Chia, Mike Abrams
11 October 2013
• EARTH SCIENCE AND RESOURCE ENGINEERING
ASTER system
1st (only) geoscience-tuned satellite sensor
• Japanese designed/built (ERSDAC now JSS)
• USA satellite platform (TERRA)
• Launched December 1999
• Basic specifications
• 60 by 60 km image
• 14 spectral bands:
– 3 VNIR (Landsat config) @ 15 m
– 6 SWIR @ 30 m
– 5 TIR @ 90 m
• Push-broom for VNIR and SWIR
• Whiskbroom for TIR
• Significant Instrument/data Issues
– atmospheric correction, SWIR X-talk, TES
• SWIR module failed 2009
• Data now only available through ERSDAC GDS
http://www.gds.aster.ersdac.or.jp/gds_www2002/index_e.html

NT ASTER Geoscience Map Workshop

 Importance of ASTER for Geosciences
• ASTER is the only operational satellite-borne sensor specifically
designed for geoscience/mineral mapping applications;
• The advantage of ASTER over other satellite data are
• 6 bands in SWIR region to distinguish AlOH, FeOH, MgOH mineral groups;
• 5 bands in TIR region to distinguish silica, quartz, mafic, carbonate mineral groups;
• Bands in the VNIR to infer ferric oxides (overlaps with other satellites);
Spatial Spectral
1.65

(30,60m TIR)
1.45

HyMap
Laboratory
126 bands
1.25

ARGUS
/ AVIRIS
1.05

HYMAP

HyMap 0.85
HYMAP

ASTER
(3-30m,swath
2.5km@5m) 0.65
Landsat TM

0.45
1.9 2.1 2.3 2.5

Wavelength (micrometer) =>(um)

 ASTER Mineral Map of Australia Workflow
Visualisation
35,000 L1B Scene selection, QA scripts.
(Analysis) candidate
ASTER scenes ENVI/IDL code 1 Automated process
ArcMAP 10 scenes
~10Tb /work/?
Multiple parallel sessions l Interactive process
/source/
IO intensive

Geo-registration, sensor Exoatmospheric correction CPU intensive

3600 gain, resampling.
1 ENVI/IDL code 2,3 Iterative
scenes
ENVI/IDL code
Core workflow

Invariant target mask

Ratios Threshold Mosaicing
2 ENVI/IDL code ENVI ENVI 3
(Multiple parallel sessions)

Cross scene calibration Mosaicing

3 4
R script, c code ENVI

Visualisation
Conversion to reflectance Information Extraction
4 (Analysis) 5
ENVI/ IDL code ENVI/IDL code
ENVI

Geoscience product Validation

5 generation ENVI
ENVI/IDL code Multiple parallel sessions

MC
Methodology - Overview
• Scene selection
• L1B radiance@sensor data
• where possible summer, cloud free scenes are used.
• Crosstalk correction
• Standard ERSDAC software see Iwasaki* for algorithm details
• VNIR and SWIR re-sampled into single 30m dataset
• Conversion to exo-atmospheric reflectance
• Solar irradiance, sun zenith

*Iwasaki & Tonooka (2005) Validation of Crosstalk Correction Algorithm fo ASTER/SWIR.

IEEE TRANS ON GEOSCIENCE AND REMOTE SENSING, VOL. 43, NO. 12, DECEMBER
Workflow considerations
• ASTER Instrument Effect – Crosstalk
• Leakage of photons from band 4
• Caused by multiple reflections between aluminum coatings in detector
• Effects all bands but most apparent in bands 5 and 9
• Additive, spatially varying effect
• ERSDAC software available which partially compensates for effect.
Residual effects can remain
4 5 6

7 8 9
Methodology – Overview (continued.)
• Cross scene calibration
• CSIRO in-house developed calibration software*
• Developed by group that established calibration techniques for Australian
continental Landsat mosaic
• Empirical approach utilising statistics of invariant sites in the areas of
scene overlap to calculate per scene per band calibration coefficients.
• Semi automated approach
Before After

*Collings S. et al. (2010) Techniques for BRDF Correction of Hyperspectral Mosaics. IEEE Trans. on Geos. and
Remote Sensing, Vol 48 (10) pg3733
Methodology – Overview (continued.)
• Calibration to apparent surface reflectance
• ASTER mosaic calibrated to Hyperion data;
• Australian archive of 18 Hyperion scenes processed to apparent
surface reflectance and used as reference;
• Least squares regression using 422 invariant targets;
• Calibration curves with R2 0.77-0.91 for each band established
Methodology – Overview (continued.)
• Information Extraction – Mineral Mapping
• Ratios and continuum band ratios used to identify geoscience information e.g.
(A+C)/B
– (B5+B7)/B6 = AlOH group abundance
– B5/B7 = AlOH group composition etc.
– B4/B3 = Ferric oxide abundance
• Robust measures which normalise out topographic differences
• Ratios map absorption features specific to mineral groups
• Geoscience products further refined by masking for green veg, cloud, water bodies
and recent fire scars

Caccetta, M, et. al, A calibration methodology for continental scale mapping using ASTER imagery, Remote Sensing
of Environment, Volume 139, December 2013, Pages 306-317, ISSN 0034-4257,
http://dx.doi.org/10.1016/j.rse.2013.08.011.
Version 1 ASTER Geoscience products
A set of ~14 GIS-compatible mosaics generated from 3,600 ASTER scene .

VNIR-SWIR products include:

• False colour
• Green vegetation content
• Regolith ratios
• Ferric oxide content
• Ferric oxide composition
• Ferrous iron content
• Opaque index
• AlOH group content
• AlOH group composition
• Kaolin group index Silicate index
low high
• FeOH group content
TIR products
• MgOH group content • Silica index
• MgOH group composition • Quartz index
• Ferrous iron content in MgOH • Gypsum index

NT ASTER Geoscience Map Workshop

 ASTER geoscience product descriptions
https://datanet.csiro.au/dap/landingpage?execution=e1s2&_eventId=viewDescription.

NT ASTER Geoscience Map Workshop

Minerals

Eastern
Arunta Inlier

Silicate index
low high

View the ASTER maps using World Wind

http://www.ga.gov.au/aster-viewer
View and download ASTER maps:
• http://portal.auscope.org/portal/gmap.html;
• https://sarig.pir.sa.gov.au/Map;
• http://geoscience.nt.gov.au/giws or
•https://datanet.csiro.au/dap/landingpage?execution=e1s2
& eventId=viewDescription
 The journey Chemotherapy
September 2012 launch
of Min Map of Oz at IGC
October 2010 TC diagnosis of
Waldenstrom’s Macroglobulinemia 6 months
Pulmonary
embulism
Chemotherapy
3 months

2009 ASTER archive of OZ,

35K scenes, 75% cloud free
1999 launch of ASTER

Early 1990s, Gabell and Cudahy

on ASTER science team

Life's a journey, not a destination. Steve Tyler, American Musician 1948

Building upon our pass: Viewing across the surface
and into the earth through time

When you arrive at your destination, pay absolutely no attention to the

thing people call jetlag.
Lara St John, 1971 (Canadian Musician)

“
National Virtual Core Library
National Geochemical Survey of Australia (NGSA).
– Geoscience Australia + State and Territory
Geological Surveys
http://www.ga.gov.au/energy/projects/national-geochemical-
survey.html
– Gap = no National geochemical coverage
– Transported regolith samples at the outlet of
large catchments
– Two depths: 0-10 cm; 60 and 80 cm
– 1390 catchments covering 91 per cent of
Australia
– 60 elements using mainly XRF and collision cell
ICP-MS + particle size, pH, LOI….

CET Guest Lecture| NGSA meets ASTER 26th July 2013

 FieldSpectra-XRD
Validation Results
Results

UWA 3rd year 2011

Broken Hill Survey
• 150 Gigabytes of airborne
HS data collected in 2002 for
GSNSW
• L1b and L2 products made
publicly available but only 1-2
copies downloaded
• C3DMM processed to L3
mineral maps in 2009
• together with other C3DMM
data, over 60,000 products
downloaded
Continental scale mineral mapping using ASTER data

http://www.ga.gov.au/aster-viewer
http://portal.auscope.org/portal/gmap.html
Use of ASTER mineral map to identify iron
ore deposit

Channel
iron ore

iron oxide
5 40
Airborne hyperspectral mineral maps

Channel
iron ore

airborne HS kaolin disorder

iron oxide content

iron oxide
crysallinity
low5 40
high
Clay composition – spectra & XRD

In collaboration with Murchison Metals

Iron oxide content – spectra & XRF

In collaboration with Murchison Metals

Access of ASTER geoscience maps
http://portal.auscope.org/portal/gmap.html ; http://www.ga.gov.au/aster-viewer ;
IN-02-C2: “Development of Regional/Global Information and Cross-cutting
Datasets) http://www.earthobservations.org/ts.php?id=136 ;
http://geossregistries.info/geosspub/

NT ASTER Geoscience Map Workshop

Thank you
CSIRO Earth Science and
Resource Engineering
Carsten Laukamp
Geoscientist
t +61 8 6436 8754
e [email protected]
w http://c3dmm.csiro.au/
Western Australian Centre of Excellence for 3D Mineral Mapping/ MDU Flagship
References
• Cudahy, T. (1997): Geometry of the spectral-mineralogical alteration halo at Birthday
South, Fimiston Gold Deposit, Western Australia.- CSIRO Report 421R.
• Cudahy, T., Caccetta, M., Cornelius, A., Hewson, R., Wells, M., Skwarnecki, M., Halley, S.,
Hausknecht, P., Mason, P., Quigley, M. (2005): Regolith, Geology and Alteration Mineral
Maps from New Generation Airborne and Satellite Remote Sensing Technologies.-
MERIWA Report 252.
• Haest, M., Cudahy, T., Laukamp, C., Gregrory, S. (2012): Quantitative mineralogy from
visible to shortwave infrared spectroscopic data - I. Validation of mineral abundance and
composition products of the Rocklea Dome channel iron deposit in Western Australia.-
Economic Geology, 107, 209 - 228.
• Haest, M., Cudahy, T., Laukamp, C., Gregrory, S. (2012): Quantitative mineralogy from
visible to shortwave infrared spectroscopic data - II. 3D mineralogical characterisation of
the Rocklea Dome channel iron deposit, Western Australia - Economic Geology, 107, 229
- 249.

IGARSS 2012 - Continent-Scale Mineral information from ASTER multi-spectral SATELLITE data