AESC2016 Mauger Ehrig 29 June Final

Prévia do material em texto

Hyperspectral applications in 
economic geology – Case Study: 
Alteration at the Olympic Dam 
IOCG-U Deposit 
Mauger, A.J., Ehrig, K., 
Kontonikas-Charos, A., 
Ciobanu, C.L. & Cook, N.J. 
29 June 2016, AESC2016, Adelaide 
www.minerals.statedevelopment.sa.gov.au 
AJES Special Monograph 
Hyperspectral Applications 
• What is the attraction of the term 
“hyperspectral”? 
• Origins in Remote Sensing – bands 
• The more, and narrower, the “bands” the closer 
the digital record approximates an analogue 
signal 
• Usually considered to be > 100 bands 
 
 
What are we referring to with 
“hyperspectral”? 
• Specifically the mineralogical response to reflected radiation 
• VIS-NIR 300-1300 nm 
• SWIR 1300-2500 nm 
• TIR 6000-14500 nm 
• Instrumentation (a selection) 
– Hyperion 
– HyMap 
– HyLogger 
– Corescan 
– Terracore 
• This presentation will focus on HyLogger 
 
What has been the impact of 
hyperspectral? 
• Greater confidence in material identification – not just 
discrimination 
• Improved modelling of data corrections/calibration 
• All scales of observation 
– Satellite 
– Airborne 
– Laboratory 
– Portable 
 
Transition from Remote Sensing to 
Spectroscopy 
 
• Objective to have accurate, repeatable, 
unambiguous mineral identification and 
quantification from mixed spectra. 
• What are the applications of spectroscopy in 
the geological exploration-discovery-
exploitation-recovery cycle? 
What have been the game changers? 
• PIMA study by CSIRO (AMIRA P435, 1999) identified 
white mica and chlorite chemistry as key vectors to 
mineralization 
• CSIRO/AMIRA Project P685: HyLogger Prototype 2002 
• NCRIS/Auscope: Every Australian State Geological Survey 
equipped with HyLogger 2 (VSWIR) 2009 
• Thermal Infrared attached: HyLogger 3 (VSWIR-TIR) 2011 
Not just hyperspectral…. 
…but almost continuous recording at the cm scale 
• Why is this important? 
• Because mineralogy just became an Exploration tool. 
• HyLogger provides detailed datasets 
• HyLogger provides more than the best geologist can see 
• HyLogger enhances the geologist’s ability to describe and interpret 
• HyLogger provides consistent and quantitative information at the 
scale of mineralogy 
• Mineral Systems focus on minerals 
 
HyLogger 3-3 
Quick recap: 
• Visible – Thermal Infrared semi-automated core scanner 
• Interprets minerals responsive to the measured 
wavelengths of reflected radiation 
• Can scan 500m diamond core per day – spectra every cm 
• Provides a high resolution image of the core 
• Alteration mineralogy is a key focus 
… Applications in Economic Geology 
• For the purposes of this presentation the focus will be 
on South Australian examples 
• Cu, Au, U, Pb, Zn, Ag have been the main commodities 
of interest in SA 
• The Copper Strategy 
South Australia’s Copper Strategy outlines the plan to 
triple the State’s copper production to 1 million tonnes 
per year within the next two decades and ensure the 
South Australian community benefits from this increase. 
Copper 
Gold 
Current Cu-Au Deposits & HyLogger 
OD 
HyLogger & Mineral Systems 
• Early identification of spectral vectors 
• Focus on the Olympic Domain (Cu, Au, U) 
• Establish the application of chlorite and 
sericite chemical gradients in IOCG-U systems 
• Development of TIR spectral vectors 
• Quintessential example – Olympic Dam itself 
Olympic Dam 
• Refer to Monday!! 
• World’s largest endowment of contained metal 
• 20,000 m of core in a 14 km cross section 
• Au, Cu, Ag, U – highly economic 
• What can HyLogger tell us that BHP Billiton don’t 
already know? 
Location of 
Olympic Dam 
In relation to 
Gawler Silicic 
Large Igneous 
Province 
 
After McPhie et al 
Location 
• 14 km 
• 30 Drill holes 
• 20,000 m 
Roxby Downs 
Olympic Dam 
RD 
6750
00E 
66300
00N 
Roxby 
Downs 
Granite 
Olympic Dam 
Breccia Complex 
biotite ‘out’ 
altered, weakly brecciated 
RDG 
Fe ~5% 
N 
Resource outline 
6800
00E 
66250
00N 
66350
00N 
* dykes projected to -350mRL 
Gairdner dykes 
 
undiff mafic dykes 
 
sediment: hm-qtz 
 
sediment: mafic 
 
sediment: hm 
 
conglomerate 
 
volcanic breccia 
 
Fe (>20%) 
hem-rich bx 
 
Fe (5-20%) 
gr- to hem-rich bx 
 
Fe (<5%) 
 
RDG 
drill hole collars 
(this study) 
Simplified Geological Map (at -
350mRL) 
0 1 2 km 
RD2488 
RD2499 
RD451 
RD302 
RD2773 
RD2786A 
RD2923 
RD2751 
RD2347 
RD2336 
RD2326 
RD480 
RD2684 
RD2568 
RD2137 
RD16/16W1 
RD2151 RD2153 
RD1603 RD1605 
RD1604 RD1606 
RD2162 
RD16/16W1 
RD76 
RD252 
RD33 
RD49 RD2715 
RD1614 inset 
-1400 
-1000 
-600 
-200 
ksp-ser 
ksp>ser 
-1400
-1000
-600
-200
ser (minor ksp) 
no ser 
hem-alt 
What minerals can be measured? 
• Not all minerals have a distinctive spectral 
response at the wavelengths HyLogger measures 
• SWIR 
– OH Bond predominantly, harmonics 
• TIR 
– Primary molecular vibrations 
What the HyLogger saw… 
SWIR 
00 = Opal 
01 = Dickite ‘ 
02 = Kaolinite-PX 
03 = Kaolinite-WX 
04 = Nacrite 
05 = Muscovite 
06 = Paragonite 
07 = Phengite 
08 = Montmorillonite 
09 = Nontronite 
10 = Saponite 
11 = Diaspore 
12 = Gibbsite 
13 = Prehnite 
14 = Pyrophyllite 
15 = Topaz 
16 = Chlorite-Fe 
17 = Chlorite-FeMg 
18 = Chlorite-Mg 
19 = Biotite 
20 = Phlogopite 
21 = Actinolite 
22 = Hornblende 
23 = Tremolite 
24 = Riebeckite 
25 = Serpentine 
26 = Brucite 
27 = Talc 
28 = Epidote 
29 = Zoisite 
30 = Tourmaline 
31 = Tourmaline-Fe 
32 = Rubellite 
33 = Ankerite 
34 = Siderite 
35 = Calcite 
36 = Dolomite 
37 = Magnesite 
38 = Alunite-K 
39 = Alunite-Na 
40 = Alunite-NH 
41 = Gypsum 
42 = Jarosite 
43 = Palygorskite 
44 = Vegetation-Dry 
45 = IsaWhite 
46 = IsaYellow 
47 = PlasticChipTray 
48 = Teflon 
49 = WhiteMarker 
50 = Wood 
51 = YellowMarker 
52 = MuscoviticIllite 
53 = ParagoniticIllite 
54 = PhengiticIllite 
 
 
TIR 
000 = Opal 
001 = Quartz 
002 = Anorthoclase 
003 = Microcline 
004 = Orthoclase 
005 = Albite 
006 = Anorthite 
007 = Bytownite 
008 = Labradorite 
009 = Andesine 
010 = Oligoclase 
011 = Andradite 
012 = Grossular 
013 = Uvarovite 
014 = Almandine 
015 = Spessartine 
016 = Augite 
017 = Diopside 
018 = Hedenbergite 
019 = Enstatite 
020 = Fayalite 
021 = Forsterite 
022 = Olivine 
023 = Zircon 
024 = Andalusite 
025 = Cordierite 
026 = Marialite 
027 = Meionite 
028 = Vesuvianite 
029 = Analcime 
030 = Chabazite 
031 = Heulandite 
032 = Laumontite 
033 = Mesolite 
034 = Natrolite 
035 = Phillipsite 
036 = Thomsonite 
037 = Kaolinite 
038 = Kaolinite-PX 
039 = Illite 
040 = Muscovite 
041 = Paragonite 
042 = Mont-
morillonite 
043 = Mont-
morillonite-Na 
044 = Nontronite 
045 = Smectite-Fe 
046 = Saponite 
047 = Axinite 
048 = Prehnite 
049 = Pyrophyllite 
050 = Topaz 
051 = Chlorite 
052 = Biotite 
053 = Phlogopite 
054 = Stilpnomelane 
055 = Actinolite 
056 = Amphibole-
ML48 
057 = Edenite 
058 = Ferro-
hornblende 
059 = Hornblende 
060 = Kaersutite 
061 = Tschermakite 
062 = Anthophyllite 
063 = Gedrite 
064 = Grunerite 
065 = Holmquistite 
066 = Mangano 
cummingtonite 
067 = Arfvedsonite 
068 = Glaucophane 
069 = Riebeckite 
070 = Antigorite 
071 = Chrysotile072 = Lizardite 
073 = Talc 
074 = Clinozoisite 
075 = Epidote 
076 = Zoisite 
077 = Tourmaline 
078 = Cerussite 
079 = Smithsonite 
080 = Strontianite 
081 = Witherite 
082 = Azurite 
083 = Malachite 
084 = Ankerite 
085 = Rhodochrosite 
086 = Siderite 
087 = Aragonite 
088 = Calcite 
089 = Dolomite 
090 = Dolomite-Fe 
091 = Magnesite 
092 = Alunite-K 
093 = Alunite-Na 
094 = Barite 
095 = Gypsum 
096 = Jarosite 
097 = Apatite 
098 = Vonsenite 
099 = Cassiterite 
100 = Goethite 
101 = Hematite 
102 = Ilmenite 
103 = Magnetite 
104 = Chromite 
105 = Gahnite 
106 = Psilomelane 
107 = Pyrolusite 
108 = Rutile 
109 = Palygorskite 
 
 
163 
substances 
 
82 - OD 
Which wavelength region is best? 
What the HyLogger saw… 
320
430 
224 
424 
378 
157 
445
873 
245 
2084 
380
85 
374
602 
319 
531 
328 
457 
318
693 
327
376 
306
391 
306
583 
295
623 TIR Summary Plots 
Showing approximate vertical 
relationships 
NE SW 
Feldspar Story 
• Absence of Albite (?) 
• HyLogger identified Plagioclase 
• Alkis revealed presence of Plagioclase 
• HyLogger maps Orthoclase and Microcline as 
distinct K-Feldspar species 
• Plotted the HyLogger results in a QAP diagram 
K-Feldspar in the TIR 
Orthoclase 
Microcline 
Mixtures compare slopes 
For the sake of modelling 
• Calculated single value per hole 
• Used “metres of mineral” as an approximation 
of abundance in a hole. 
• Wavelengths – used averages 
• Assays – used maximums 
Copper values 
Sericite and Chlorite 
Se
ri
ci
te
: 
D
ec
re
as
in
g 
A
l 
C
h
lo
ri
te
: 
in
cr
ea
si
n
g 
Fe
 
K-Feldspar and Plagioclase 
Microcline 
Orthoclase 
Oligoclase 
Albite 
Feldspar Ratios 
O:M 
A:O 
QAP (modified) 
Biotite Out 
Distal 
Barren Core 
Ore Q – Quartz 
A – K-Feldspar 
P – Plagioclase 
From 1D to 2D 
• Next few slides depict cross sections derived 
from HyLogger data and presented in GoCAD 
• Blue is low number (wavelength or 
abundance). Red is a high number. 
Cu ppm 
Orthoclase 
Microcline 
Albite 
> An10 
Scale: 3,000 m 
Illite 
Muscovite 
Phengite 
W2200 
Scale: 3,000 m 
W2250 
Fe-Chlorite 
Mg-Chlorite 
K/Al ratio 
Scale: 3,000 m 
Opportunities 
• Improve the sophistication of the 
quantification of mineralogy from TIR 
• Improve understanding of QAP (and similar 
tools) and limitations 
• Refine the TIR vectors 
• Undertake further transects to build 3D 
South Australian Experience 
• 2002 AMIRA Prototype – Emmie Bluff 
• Signature Holes 
– 2003 Hired Prototype – 1 month (1,000m/day) 
– 2005 Hired Prototype – 1 month 
• 2009 NCRIS/AuScope VSWIR HyLogger 2-3 
• 2011 NCRIS/AuScope +TIR HyLogger 3-3 
• 2016: 810 Openfile drillholes on SARIG/NVCL 
 
Projects 
• Significant scientific uptake 
• Cariewerloo Basin 
• Stuart Shelf IOCG-U 
– Olympic Dam 
– Emmie Bluff 
– Punt Hill 
– Hillside 
• Paris 
• DETCRC 
• Frome Embayment - U 
What has HyLogging revealed? 
• Every hole scanned delivers something new 
• 800 metre thick “monotonous red bed 
sandstone” is not monotonous at all. 
• Significant sections of Dickite, Kaolinite, 
Muscovite, Illite, Phengite and Paragonite 
whose geometry and spatial relationships 
required new thinking. 
Key Mineral Chemistry Gradients 
• Fe- Mg- content of Chlorite 
– A significant component of IOCG-U systems in the 
Olympic Domain 
• Al- content of White Mica 
– Muscovite, phengite, “paragonite” 
• Abundance gradients e.g. White Mica 
• K-Feldspar – Plagioclase - Quartz 
Future of Spectral Geology 
• One tray is a waste of time 
• One hole is interesting but really doesn’t do it 
• One section is a start 
• 3D model is where you want to be 
• Clusters of holes: where you start to 
appreciate the power of dense mineralogy 
Future of National Virtual Core Library 
• The HyLoggers were gifted to the States in order to 
effectively digitise their extensive drill core collections. 
• This work is only partly complete – there are at least 10 
years more work in SA alone. 
• This task is ideally suited to HyLoggers. Acquisition 
speeds and data volumes are manageable. Current 
geological programs can be used to prioritise the work. 
• Internet is geared up to distribute the data efficiently 
• New knowledge is emerging from these repositories. 
Hyperspectral Technology… 
… has made its mark in the geological community. 
Surveys are providing free pre-competitive data. 
There is a growing desire to discover how it can 
assist exploration and deposit modelling. 
Current need for greater education – u-grad, post 
grad, professional development. 
 
Disclaimer 
www.minerals.statedevelopment.sa.gov.au 
The information contained in this presentation has been compiled by the Department of State 
Development (DSD) and originates from a variety of sources. Although all reasonable care has been 
taken in the preparation and compilation of the information, it has been provided in good faith for 
general information only and does not purport to be professional advice. No warranty, express or 
implied, is given as to the completeness, correctness, accuracy, reliability or currency of the 
materials. 
 
DSD and the Crown in the right of the State of South Australia does not accept responsibility for and 
will not be held liable to any recipient of the information for any loss or damage however caused 
(including negligence) which may be directly or indirectly suffered as a consequence of use of these 
materials. DSD reserves the right to update, amend or supplement the information from time to 
time at its discretion. 
 
HyLoggerTM is a trade mark of CSIRO <www.csiro.au/org/HyLoggingSystemsGroup.html> 
AuScope Ltd <www.auscope.org.au> is funded under the National Collaborative Research 
Infrastructure Strategy (NCRIS) an Australian Commonwealth Government Programme 
 
Contact 
www.minerals.statedevelopment.sa.gov.au 
Department of State Development 
Geological Survey of South Australia 
Level 4, 101 Grenfell Street 
Adelaide, South Australia 5000 
GPO Box 320 
Adelaide, South Australia 5001 
T: +61 8 8463 3062 
E: dsdreception@sa.gov.au