Geochemistry and the challenge posed by exotic cover · Geochemistry and the challenge posed by...
Transcript of Geochemistry and the challenge posed by exotic cover · Geochemistry and the challenge posed by...
Geochemistry and the challenge posed by exotic cover
David CohenUniversity of New South Wales
and
Association of Applied Geochemists
Recent Practical Advancesin Mineral Exploration
Technologies
September 11, 2009
Ahmad MokhtariIsfahan University of Technology, Isfehan
Neil RutherfordRutherford Mineral Resource Consultants, Sydney
Simon GatehouseHellman and Schofield, Sydney
David Kelley MMG – Minerals and Metals Group, Denver
Ravi AnandCRCLEME / CSIRO Exploration and Mining, Bentley
William CokerBHP Billiton World Exploration, Vancouver
Iain DalrympleActlabs, Perth
RMRC
SMEDG -AIG ‘09
SMEDG -AIG ‘09
Intro
Models
Mandy
SEx
Conclusions
Data
The problematic terrains for geochemical exploration
Glacial depositsThick gravel and scree depositsThick alluvium or colluvium + deep weatheringAeolian deposits
?IceVolcanic ash
IntroModels
MandySEx
ConcData
SMEDG -AIG ‘09
Data: Australian Minerals Atlas
IntroModels
MandySEx
ConcData
L. Mesozoic to Tertiary basins
Thompson Orogen
Lachlan Fold Belt
SMEDG -AIG ‘09
Significant mineral occurrences
⇒ Systematic approach to identifying regolith materials and landforms
⇒ Link evolution of regolith and its components to geochemical processes and sampling strategies
Weathered profiles typically have RESIDUAL and TRANSPORTED components of varying ages
Several distinct phases of creation and destruction of secondary Fe, Ca, Si and Al minerals – each with total or partial resetting of trace element geochemistry
Anand and Smith, 2005
IntroModels
MandySEx
ConcData
Red-brn sandy clay
Pisolitic ferricrete
Silicd sandy, silty + gravel colluv & alluv
Mega mottle palaeo-channel clays
Transported gravel
Clays
Mottled saprolite
Saprolite
Saprock
Base of sedimentsLateritic residuum
Bedrock
Reg
olith
dev
elop
ed o
n an
d in
prim
ary
litho
logy
Tran
spor
ted
Ove
rbur
den
Tert
iary
Qa
min
eral
isat
ion
?Regolith Classification
unconformity
SMEDG -AIG ‘09
(data: Lawrance, 1999)
MesozoicCover
ProterozoicBedrockCu - Au
Mineralisation
< 15 (ppm)15 - 3030 - 5050 - 200200 - 10001000 - 2000>2000
Cu
Unconformity
< 1 %1 - 22 - 33 - 88 - 1515 - 20> 20
Fe
Unconformity
100 m
40m
Dispersion mechanisms?
Dispersion timing?
Osborne Cu-Au deposit, Qld
IntroModels
MandySEx
ConcData
Does dispersion occur through deep, transported cover?
SMEDG -AIG ‘09
IntroModels
MandySEx
ConcData
CSA Mine Lease, Cobar
C.S.A.
Recent Drainage
Line ALine B
Line D
Line C
1km
U. Amphitheatre Gp.CSA SiltstoneChesney Fmn. (grwk)
Girilambone Gp.Chesney Fmn. (slts)
Palaeo-drainage
Shen and Cohen, 2005
Line DCoarse silt + lagTransported silt + clay
Residual regolith
0 m 150
Chesney Fmn. siltstone Girilambone Gp
3000
600
50m
Pb
Fe
Zn
SMEDG -AIG ‘09
Does dispersion always occur?
McKinnons deposit, Cobar
Intro
MandySEx
ConcData
Models
SMEDG -AIG ‘09
Mixed pediment / alluvium
Au mineralised structure Cohen et al, 1996
Alluvial fillPediment
Saprolite
line 79600N
300 m
205 m
195 m
Dispersion models – vegetation effects
0
0.3ppm
0
15ppm
As
300ppm
0
Cypress pineneedles
Soil total
Lag total
magnetic
non-mag
Dispersion models – gases and aerosols
Deployment arrangement
Collector assembly
polystyrene-coatedglass slide
surface
hole
plastic cap(dust cover)
collector
funnel
backfill
50 cm
Rutherford et al., 2005
Clay in droplet
50 μm
20 μm
Amorphous silica
10 μm
NaCl + Na2CO3
Intro
MandySEx
ConcData
Models
SMEDG -AIG ‘09
Cu
Fe
Cl
K
Ore Zoneng cm-2
20
10
0
100
50
0
0.8
0.4
0
20
10
0
0 500 1000 1500 m
Osborne Cu-Au deposit, Qld
Intro
MandySEx
ConcData
ModelsPlus a range of other transition and group I and group II elements
SMEDG -AIG ‘09
Osborne (Kulthor) Cu-Au deposit, Qld
Intro
MandySEx
ConcData
Models+6 +7 +8 +9 +10 +11 +12 +13
chalcopyrite
pyrrhotite
pyrite
gypsum in soil
regional soil background
δ34S
SMEDG -AIG ‘09
Figure 5 - Model of Cameron et al
Cameron et al., 2004
Dispersion models – dilatancy pumping
Intro
MandySEx
ConcData
Models
SMEDG -AIG ‘09
Spence porphyry Cu deposit, Northern Chile
0
15ppm Cu
0
2ppm As
0.6%
0
Eastern Fracture
Zone
Spence Deposit
Na
500 m
Surface regolith samples;weak selective extraction
Vertical fracturewith saline soil
Intro
MandySEx
ConcData
ModelsBurial under 250m of Tertiary gravels
SMEDG -AIG ‘09
Dispersion models – electrochemical
Govett, 1977Smee, 1983
Intro
MandySEx
ConcData
Models
SMEDG -AIG ‘09
Till
Varvedclays
Humus
Sulp
hideCountry
rock
SPSurface SP
responseCationconc.
Cation conc. in humus
Cation migration
Cation migration Current
lines
Equipotential lines
e-
Anode: 4FeS2 + 40H2O →4FeO.OH + 76H+ + 8SO4
2- + 60e-
Cathode: 15O2 + 30H2O + 60e- → 60OH-
Zone of increased cation conc. created by organic chelation of diffusive flux
Ions chelated by humus
Cation diffusion
Zone of increased cation conc. created by electrochemical migration
A horizon
B horizon
Clay
Volcanic Rocks
Water Table
Sulphide
Zone of CaCO3Accumulation
Upper limit of Carbonate
Oxidation ZoneLow pH
Carbonate Loss
Zn Cd
Zone of Metal Accumulation
O2
Zn2+ + Cu2+ + Fe2+
Reduced Column
Hamilton, 1998
Intro
MandySEx
ConcData
Models
Dispersion models – electrochemicalSMEDG -
AIG ‘09
A horizon
B horizon
Clay
Volcanic Rocks
Water Table
Sulphide
Oxidation ZoneLow pH
Carbonate Loss
Zn Cd
Zn2+ + Cu2+ + Fe2+
Reduced Column
Hamilton, 1998
Intro
MandySEx
ConcData
Models
Dispersion models – electrochemical
pHCa, REE
SMEDG -AIG ‘09
Transported Cover
Inherited metal content in regolith
[met
als]
with overprint due to dispersion away from mineralisation OR other features
Minn
A need to isolate this component or signal
which may constitute <0.1% of total signal
A basic premise
BedrockMandy
Models
Conc
Intro
Data
SEx
SMEDG -AIG ‘09
Integration of dispersion over timeProgressive accumulation of metalsStabilisation and fixation of pattern
Progressive degradationReworking of metals
Hydration - dehydration reaction cycling
Acid-base and redox reaction recycling
Increasing age of mineral phase in regolithLess transitory metal contents
Resistate minerals
SilicatesCryst. Fe-oxides
Mn-oxides &
am. Fe-ox.Carbonates
Adsorbed & Exch.species
Solublephases
Organics
Selective Extractions
Mandy
Models
Conc
Intro
Data
SEx
SMEDG -AIG ‘09
Mandy
Models
Conc
Intro
Data
SEx
(Cohen et al, 2001; based on Gray, 1999)
Carbonates Resistate mineralsSilicates
Mn-oxides &
am. Fe-ox.
Solublephases
Cryst. Fe-oxides
Adsorbed & Exch.species
Organics
SMEDG -AIG ‘09
Acetate + HOAc
HF / fusion
Aqua regia
Mixed acids (Canadian wine)
Weakly acidified NH2OH
EDTA / H+
Water (Qld Beer)
Strongly acidified NH2OH
Regoleach
HCl
Ammonium acetate
MMI
Guinness
Enzyme Leach / H2O2
Na-pyro / H2O2t
Selective Extractions
Mandamah, NSW
Lower Saprolite
50 m
30 mMHACD 228
MACD 312
206m @ 0.51 g/t Au0.37 % Cu
+ Mo
Ser-py altn
Andesite
Ser-c
hl-m
t-alb
altn
SEx
Conc
ModelsIntro
DataMandy
Mandamah
NSW Transported Regolith
SMEDG -AIG ‘09
Incr conc of some trace elementse.g. Cu, Ni and Znin SEx
Increasing Cs U Be Th in all SEx
Ba accum.
Higher leachableFe and Al
Na, S accum.
NaCl + SO42-
Increase in gypsum
Organic matter accumulation
Fe-enriched mottling zoneAccumulation of Cu, U & REE
Carbonate formationdol, gyp and Na2CO3
0 cm
25
50
75
100
125
150
200
SEx
Conc
ModelsIntro
DataMandy
5 6 7 8 9 10pH
SMEDG -AIG ‘09
mineralisation
pH 5 K-acetate
0.01
0.1
1
10
100
1,000
10,000ppm
Northern block Central block Southern block
CaMg
Ce
Cu
Mo
30 cm depth,
mineralisation
SEx
Conc
ModelsIntro
DataMandy
SMEDG -AIG ‘09
Mineralization
0
50
100
150m0.2123 % Cu
0 100 200 300 400 500 600 700mE
0
1.8m
CEC18
28 cmolc/kg
23
S0
240 ppm
120
Na0
1400 ppm
700
Ca0
1.0 %
0.5
pH
5.5
9.5
7.5
CEC
S
Na
Ca
pH
SEx
Conc
ModelsIntro
DataMandy
SMEDG -AIG ‘09
Mineralization
0
50
100
150m0.2123 % Cu
0 100 200 300 400 500 600 700mE
0
1.8m
Cu by aqua regia
8
20 ppm
14
Cu by NH2OH
2
6
4
Cu by NH4.OAc
0.0
0.24
0.12
Dep
th b
elow
su
rfac
e (m
)
ppm
ppm
SEx
Conc
ModelsIntro
DataMandy
SMEDG -AIG ‘09
0
40
80
120
160
200
ECa(mS/m)
EM38 vertical coil modeEM38 horizontal coil mode
240
CEC18
28 cmolc/kg
23
Ca0
1.0 %
0.5
Mineralization
0
50
100
150m0.2123 % Cu
0 100 200 300 400 500 600 700mE
0
1.8m
CEC
Ca
SEx
Conc
ModelsIntro
DataMandy
SMEDG -AIG ‘09
Initial weathering phase (period of elevated rainfall)
Dispersion model for Mandamah? archaeo-electrochemical
Ore body
Host rock
Weathering of mineralization and H+ production due to oxidn of sulfide minerals
In-situ30 m
Leaching of elements
Oxidn of mineralization and production of H+
and ore-related ionsSEx
Conc
ModelsIntro
DataMandy
Movement of Fe2+ and hydrolysis generating large amounts of H+
SMEDG -AIG ‘09
Ore body Host rock
In-situ
Transported
Migration of H+ into transported regolith and clay alteration
Pedogenic carbonate
General loss of pedogenic carbonate or limited formation
“Prograde” Phase(still high rainfall)
50 m
30 m
SEx
Conc
ModelsIntro
DataMandy
SMEDG -AIG ‘09
Ore body Host rock
In-situ
Transported
Addition of H+ to surface ceases or
diminishes
Pedogenic carbonate Zone of (palaeo) clay modification
with low CEC+
Non-carbonate alkalinity
Recent migration of mobile alkalis (e.g. Na).
“Retrograde” Phase(Post onset of aridity)
SEx
Conc
ModelsIntro
DataMandy
SMEDG -AIG ‘09
Data Analysis
Mul
tivar
iate
dis
tanc
e
Probability function
Objective – detecting samples whose geochemistry appears “anomalous”
101101001101011001010100010010010101100100101100101011001001011000101010010101010010101011010100101011010110111111010101011001010101111001101011100111111011010111001101110111100111110101011110100110001011101011011010010100101011101001011010110100001011010100011010111101011010111001100011101010110101101010110101000101010101001100011110011011011011001110110011100011010110110101010110101101100111011010110100010101001010
Conc
SExModels
Intro
Mandy
Data
Traditional approach – satellite spotting
SMEDG -AIG ‘09
Data transformed and standardised
-3
-2
-1
0
1
2
3
-3 -2 -1 0 1 2 3
Sm
Y
-3
-2
-1
0
1
2
3
-3 -2 -1 0 1 2 3
Fe
Al
Weak hydroxylamineAqua regia
Reality for SEx data – very complex multivariate signals
Mandamah101101001101011001010100010010010101100100101100101011001001011000101010010101010010101011010100101011010110111111010101011001010101111001101011100111111011010111001101110111100111110101011110100110001011101011011010010100101011101001011010110100001011010100011010111101011010111001100011101010110101101010110101000101010101001100011110011011011011001110110011100011010110110101010110101101100111011010110100010101001010
Conc
SExModels
Intro
Mandy
Data
SMEDG -AIG ‘09
Sample from above mineralised sites
Sample from above non-mineralised sites
1.2ppm
Cu
As1.0ppm
0.0
0.0
0.0
Ave
0.8
Enzyme Leach Aqua Regia 800
0
Chimborazo porphyry Cu, Chile
As average7-point moving As variance
gravel
0 500 1000 1500 2000 2500 3000 3500m
2500m
2300
Supergeneenrichment
Monzonite porphyry
Porphyryticandesite
Libra zone Virgo zone
101101001101011001010100010010010101100100101100101011001001011000101010010101010010101011010100101011010110111111010101011001010101111001101011100111111011010111001101110111100111110101011110100110001011101011011010010100101011101001011010110100001011010100011010111101011010111001100011101010110101101010110101000101010101001100011110011011011011001110110011100011010110110101010110101101100111011010110100010101001010
Conc
SExModels
Intro
Mandy
Data
SMEDG -AIG ‘09
SExModels
Data
Intro
Mandy
Conc
ConclusionsSMEDG -
AIG ‘09
RED scheme (Anand et al, 1998)
Summary of various forms of interactions of deeply buried mineral deposits with hydrosphere and biosphere
GW Table
Higher pHHigher pH
Dec
reas
ing
Eh
Transported cover
Bedrock
20 m
In-situregolith
Low pH
Sulphides
Reduced column
Cu2+
Vapor orgroundwater
dispersion
Microbial Redox
Cyclical dilatancypumping alongstructural zone
Biologicalaccumulation
SExModels
Data
Intro
Mandy
Conc
ConclusionsSMEDG -
AIG ‘09
SExModels
Data
Intro
Mandy
Conc
Seeing Geochemical Dispersion Haloes
OBJECTIVE:
Separate processes that have affected the elemental and mineralogical composition of regolith (both gain and loss)
METHODS:
Isolate signals that can be related to the effects or presence ofmineralisation through:
PHYSICAL (the extraction) and/or
NUMERICAL (data processing) methods
ConclusionsSMEDG -
AIG ‘09
SExModels
Data
Intro
Mandy
Conc
research into fundamental processes of geochemical dispersion development of new approaches to definition and detection of geochemical patterns related to the effects of underlying mineralisation
SEx not (yet) the silver bullet for thick transported or deeply weathered cover, but has prompted:
The bests tool for detecting deeply buried, sulphide mineralisation, may yet prove to be a simple pH meter and a bottle of dilute HCl
Developments in media and analysis, based on process modellingApplication of isotopes to refine understanding of dispersion mechanisms and regolith evolution modelsAbandoning “baselines”, “backgrounds” and “anomalies” in favour of new data processing methods that recognise geochemical signals related to processes of interest
Future Research Directions
Conclusions SMEDG -
AIG ‘09
Advance!
Dispersion models
Analytical improvements
General Ponsonby leads the Scots Greys against the
Polish lancers at Waterloo
Retreat!Old thinking on
defining geochem“anomalies”
Emperor Napoleon leads his remaining troops back
from Moscow
War of Attrition
Selective extractions
Exploring in deep transported
cover
Data modelling
Field Marshal Haig leads his troops
nowhere in particular
SExModels
Data
Intro
Mandy
Conc
SMEDG -AIG ‘09
Phys Geog GeologyFirst year (excludes engineering servicing)
Mineralogy Ore deposits Geology honsUpper years
Students
Students
0
50
100
150
200
1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 20090
10
20
30
2 43
1 1
40
SExModels
Data
Intro
Mandy
Conc
SMEDG -AIG ‘09
People