research-articleresearch article12X10.1144/1467-7873/11-RA-066van GeffenTalbot geochemistry
2012
Till and vegetation geochemistry at the Talbot VMS Cu-Zn prospect,
Manitoba, Canada: implications for mineral exploration
Pim W.G. van Geffen1, 2*, T. Kurt Kyser1, Christopher J. Oates3 & Christian Ihlenfeld4
University, Miller Hall, Kingston, Ontario, K7L 3N6, Canada
Solutions Inc, 630-1188 West Georgia Street, Vancouver, BC, Canada, V6E 4A2
3Applied Geochemistry Solutions, 49 School Lane, Gerrards Cross, Buckinghamshire, SL9 9AZ, UK
4Anglo American plc, 20 Carlton House Terrace, London SW1Y 5AN UK
*Corresponding author (e-mail: [email protected])
2ioGlobal
1Queen’s
AbSTrACT: The Proterozoic Talbot VMS occurrence in the Flin Flon-Snow Lake
terrane is buried under more than 100 m of Palaeozoic dolomites and Quaternary
glacial till. Structurally controlled anomalies of Zn, Cu, Ag, Pb, Au, Mn, Hg, Cd, Co,
Bi and Se in the clay fraction of till depth-profiles indicate upward element migration
from the buried volcanogenic massive sulphide mineralisation and near-surface chemostratigraphic deposition. Principal component analysis and molar element ratios
indicate that separation of the <2 µm clay fraction reduces chemical heterogeneity and
increases trace-element yield relative to the <250 µm fraction of the till. The greatest
anomalies occur at or below 30 cm depth and over faults, suggesting that elements
were deposited in the till after upward migration through structures. The ratio Zn/Al
in the <250 µm fraction can be used as a proxy for Zn in the clay fraction, producing
high-contrast anomalies. Carbon isotopic compositions indicate that these anomalies
are related to organic carbon in the clay fraction. Humus, moss and black spruce bark
are of limited use for exploration in this environment, because they accumulate atmospheric Pb and Cd, most likely from the Flin Flon smelter at 160 km NW. Black spruce
tree rings that formed before smelter operations commenced indicate Zn and Mn
anomalies in an uncontaminated sampling material. Much of the initial vertical migration of elements to the surface at the Talbot prospect was driven by upward advection
of groundwater through fractures in the dolomite, resulting from a combination of
subsurface karst collapse and remnant hydrostatic pressure during glacial retreat.
KeyWOrdS: Exploration geochemistry, vertical element migration, clay fraction, carbon isotopes,
VMS Cu-Zn
Discoveries of economic mineral deposits increasingly rely on
the surface expression of deeply buried metal bearing systems
under exotic and transported overburden (Govett 1976; Kelley et
al. 2006; Galley et al. 2007). In many cases these deposits are covered by more than one sequence of allochthonous sediments.
The application of geochemical methods in such environments
requires a secondary expression of the buried ore deposit in surface sample media, which can only exist if elements from the
deposit are mobilised, transported upward, and trapped near the
surface (Cameron et al. 2004), or if other secondary effects cause
distinct patterns of element dispersion in surface sampling media
(Smee 1983; Smee 1998; Hamilton et al. 2004).
A wide variety of surface sampling media and their applications in geochemistry have been studied around the globe. Soil
is commonly used in exploration because of its ubiquity and its
ability to retain indicator and pathfinder elements above buried
mineralisation (Rose et al. 1979). However, different soil horizons and grain-size fractions can have strongly different geochemical compositions at a single sampling site (Reimann 1998;
Reimann et al. 1998; Henderson et al. 1998). Various sorts of
vegetation have been used as sampling media in environmental
Geochemistry: Exploration, Environment, Analysis, Vol. 12, 2012, pp. 67–88
DOI: 10.1144/1467-7873/11-RA-066
studies (Steinnes 1995; Peltola & Åström 2003; Saarela et al.
2005; Savard et al. 2006; Reimann et al. 2007) and in mineral
exploration (Dunn 2007; Kozuskanich et al. 2009). Mosses have
primarily been used as monitors of atmospheric metal deposition (Steinnes 1995). Wood, bark and foliage of trees have many
applications in geochemistry, including indicating buried mineralisation (Dunn 2007). Despite having low metal concentrations,
tree rings provide a historic sampling material that recorded the
local geochemistry and atmospheric conditions of the time in
which the heartwood was formed (Watmough 1997; Padilla &
Anderson 2002; Kozuskanich et al. 2009).
The purpose of this study is to examine the geochemical compositions of surface sampling media that include two size fractions of till, humus (LFH horizon), moss (Sphagnum sp.), and black
spruce (Picea mariana) heartwood and bark and how they can be
used to distinguish geogenic and anthropogenic inputs; specifically, which media and elements can be used for mineral exploration. The area under investigation is the Talbot Lake prospect, in
the Flin Flon-Snow Lake terrane, Manitoba, Canada (Fig. 1). This
Proterozoic VMS Cu-Zn occurrence is buried under more than
100 m of Palaeozoic dolomites and Quaternary till.
1467-7873/11/$15.00 © 2012 AAG/Geological Society of London
van Geffen et al.
68
Fig. 1. Location map of the Talbot prospect, Manitoba, Canada, with outlines
of the major geological terranes. Black
arrows indicate major ice-flow directions
during the last glaciations.
STudy AreA
The Talbot VMS Cu-Zn prospect is situated in the eastern extension of the Paleoproterozoic Flin Flon-Snow Lake terrane (1.92–
1.88 Ga), northwestern Manitoba and northeastern Saskatchewan
(Stern et al. 1995), c. 160 km SE of Flin Flon and 85 km south of
Snow Lake (Fig. 1). Medium to high-grade metamorphic arc
assemblages host the Cu-Zn mineralisation (Syme & Bailes
1993), which occurs as coarse-grained, massive to semi-massive
and disseminated chalcopyrite + sphalerite + pyrite ± pyrrhotite
(Bailes & Galley 1999). Exploration drilling at the Talbot prospect by HudBay Exploration and Development identified massive sulphide lenses of up to 10 m in thickness with 12 % Cu, 3.5
% Zn, 11 g/t Au and 184 g/t Ag that occur along a steeply dipping, north–NE-striking basement structure (Fig. 2).
The Proterozoic basement rocks are unconformably overlain
by 4–6 m of porous, poorly consolidated Ordovician sandstone
of the Winnipeg Formation and c. 100 m of chert-rich, dolomitised carbonates of the Ordovician Red River, Stony Mountain
and Stonewall formations, and the Silurian Interlake Group
(McMartin et al. 1996; Bezys & Bamburak 2001; Matile et al.
2002). Minor intraplate tectonics and glacial loading caused fracturing of the dolomite sequences (Elliott 1996) and NE-striking
normal faults are observed across the Talbot area, with vertical
offsets of up to 4 m. The Palaeozoic dolomites at the Talbot
prospect are overlain by a thin cover (0–2 m) of carbonate rich,
silty till exhibiting poorly developed soil horizons (McMartin et
al. 1996). The till veneer was deposited by the Laurentide Ice
Sheet during the last major glaciation between 13 and 10 ka that
had its centre in Hudson Bay and a dominant southwestward
ice-flow direction in the Talbot area (Klassen 1997). The geology
of the Talbot prospect is represented schematically in Figure 2.
Vertical relief around the prospect is minimal, with less than
3 m on the prospect and up to 10 m in the general area (Fig. 2).
Standing water occupies low-lying areas for most of the year,
creating small bogs with aspen vegetation. A central depression
runs parallel to an airborne EM anomaly, and sub-cropping
dolomite in the western part of the prospect forms a slight
topographic high. The boreal vegetation comprises sphagnum
moss (Sphagnum sp.), black spruce (Picea mariana), jack pine (Pinus
banksiana), trembling aspen (Populus tremuloides), balsam fir (Abies
balsamea), and minor paper birch (Betula papyrifera). The vegetation cover is variable, with the western part of the grid covered
in young black spruce and scrubs growing after a mid-1970s forest fire. Pristine black spruce forest occurs roughly between 400
and 800 m east, whereas the easternmost part of the grid was
clear-cut during recent logging activities. Apart from these logging activities and diamond drilling in the eastern part of the
prospect, the surface above the deposit is relatively undisturbed.
Minor contributions of airborne contamination are of concern,
in particular from the Flin Flon ore smelter at 160 km to the NW
with prevailing north-westerly winds (Phillips et al. 1986), and
potentially from the Thompson smelter at 210 km to the NE.
MeThOdS
Sampling
Till samples were collected at 5–20 cm depth from 151 sites on
the 1000 x 600 m Talbot exploration area before the ground
was significantly disturbed. At each site, an area of 0.5–0.8 m2
was cleared of moss, litter and roots, and one spade-depth of
till was homogenised in the sample pit before a 2 kg split was
collected for analysis. Along Line 2N, moss and humus were
sampled over the same area at each site, directly above the till
sample. The composition and thickness of the humus layer
(LFH horizon) vary strongly depending on soil substrate, vegetation coverage and groundwater tables across the prospect.
Where present, humus samples were collected from between
0–5 cm depth. Sample spacing varied from 25-m intervals near
to the projected mineralisation to 50- and 100-m spacing
further away. To assess the vertical distribution of elements in
the till cover, depth profiles were sampled at 15 sites along
Line 2N of the Talbot exploration grid (Fig. 2). At each site,
five brick-shaped, 20 x 20 cm samples of 10 cm depth intervals
were collected to a depth of 50 cm. Where subcropping dolomite was present, between 100E and 300E, till profiles had
maximum depths of 20 and 30 cm. Black spruce heartwood
was sampled as duplicate north-south oriented tree cores,
using a 5-mm diameter steel increment bore. Tree cores were
transferred into plastic drinking straws and sealed with packing
tape for transportation. Black spruce bark was sampled with a
Talbot geochemistry
69
Fig. 2. Block diagram of the Talbot
prospect geology showing line 2N, with
superimposed topography, geophysics
(SPECTREM airborne electromagnetic
data) and air photo.
paint scraper and a modified plastic dust pan, combining material from 3 trees nearest to the soil sample site. The outer bark
was scraped off between 1.50–2 m above the ground from
around the tree, avoiding branches, cuts and resin.
from either side of the pith to optimise the total sample weight.
The samples were washed with deionised water in an ultrasonic bath for 10 minutes and dried at 70°C overnight. The
dried samples were analysed by Acme Laboratories.
Preparation
Analysis
All till samples were processed at Acme Laboratories in
Vancouver, British Columbia. A split of the till was dried and
sieved to <250 µm before analysis, as a routine preparation
method for soils from exploration surveys. Another split was
used for clay-fraction separation (<2 µm), wherein 500 g of
<250 µm till was agitated in a rolling bottle with 1 L of deionised water for 1 hour. The suspension with clay and silt was
decanted into a 250 mL roughing bottle and centrifuged for 5
minutes at 800 rpm. The resulting clay-rich suspension was
centrifuged at 4000 rpm for 15 minutes to accumulate the clay.
The clay was dried, pulverised and weighed before analysis.
Humus and moss were dried and pulverised, and bark was
milled to powder using a Wiley mill at Acme Laboratories.
Tree cores were prepared for analysis at Queen’s Facility for
Isotope Research at Queen’s University. The tree cores were
lightly sanded to expose the tree ring texture and the age of
each tree was determined by counting the number of rings
under binoculars. The time interval of 1915–30 was cut out
The two till fractions, humus and vegetation were analysed
using a modified aqua-regia digestion followed by ICP-MS and
ICP-AES measurement of 53 elements at Acme Laboratories.
Till was analysed by X-ray diffraction spectrometry (XRD) and
by Portable Infra-red Mineral Analyser (PIMA) for mineralogical characterization at Queen’s University.
Chalcophile elements such as Zn, Pb and Cu are known to
occur in soils and sediments as adsorbed species, metal-organic
complexes, or as integral mineral constituents in oxides, silicates, and carbonates (e.g. O’Day 1999). To distinguish carbonates from organic-carbon phases, the depth-profile till and
its clay extracts were analysed for carbon isotopes, which are
strongly fractionated by biological processes in the till profile
and in soil forming processes (Craig 1953). Concentrations and
stable isotope ratios of carbon were measured at Queen’s
Facility for Isotope Research, Queen’s University, Canada.
Total carbon was extracted using combustion in oxygen at
c. 1700°C with an elemental analyser on-line with a MAT 252
Median Min.
1.18
0.20
0.84
0.14
0.30
0.06
0.21
0.04
0.086 0.014
0.049 0.016
0.016 0.003
0.006 0.003
<0.01 <0.01
131
15
32
6
26
4
25.3
3.8
23.4
3.6
11.7
2.4
11.0
2.0
6.9
2.1
5.7
1.4
4.2
0.6
3.6
0.6
0.9
0.1
0.2
0.1
0.32
0.08
0.19
0.04
0.08
0.02
0.06 <0.01
0.03 <0.01
31
2
11
5
0.4
<0.2
Fe (%)
Al (%)
Mg (%)
Ca (%)
K (%)
Ti (%)
P (%)
Na (%)
S (%)
Mn (ppm)
Ba (ppm)
V (ppm)
Cr (ppm)
Zn (ppm)
Li (ppm)
Ni (ppm)
Sr (ppm)
Pb (ppm)
Cu (ppm)
Co (ppm)
As (ppm)
Se (ppm)
U (ppm)
Mo (ppm)
Bi (ppm)
W (ppm)
Cd (ppm)
Ag (ppb)
Hg (ppb)
Au (ppb)
3.63
2.71
3.01
4.31
0.535
0.116
0.052
0.046
<0.01
826
110
68
76.2
64.8
37.2
40.3
20.4
13.4
27.2
13.9
4.1
0.6
0.77
0.76
0.25
0.14
0.18
169
43
10.5
Max.
0.60
0.43
0.37
0.56
0.073
0.020
0.008
0.004
–
110
19
12
12.5
11.1
5.8
6.4
3.6
2.0
3.7
2.0
0.6
0.1
0.12
0.11
0.04
0.02
0.03
26
8
1.0
4.90
4.19
1.59
0.61
0.636
0.137
0.026
0.017
0.02
442
189
85
115
119
63
63
26.2
14.3
41.8
16.7
3.3
0.4
0.84
0.44
0.29
0.06
0.07
84
31
0.5
1.24
2.32
0.40
0.32
0.172
0.039
0.017
0.005
<0.01
62
98
29
71
42
24
18
10.0
10.5
18.6
3.9
0.9
0.2
0.52
0.19
0.14
0.02
<0.01
14
9
<0.2
Min.
10.42
8.03
1.86
1.90
0.939
0.172
0.193
0.030
0.16
988
401
238
250
255
103
100
35.8
29.2
97.3
25.8
8.5
1.4
1.24
3.77
0.66
0.22
0.65
345
105
2.6
Max.
1.26
0.86
0.25
0.19
0.181
0.034
0.027
0.005
0.02
110
49
32
24
28
9
10
3.7
3.8
12.3
2.4
1.2
0.2
0.13
0.58
0.06
0.04
0.09
54
19
0.5
StDev.
Till <2 µm, n=151
StDev. Median
Till <250 µm, n=151
Element
Material
0.24
0.17
0.095
0.86
0.113
0.004
0.099
0.002
0.13
487
55
4
3.9
53
0.5
4.5
12.0
19.3
9.1
2.2
1.2
0.8
0.11
0.30
0.16
0.03
0.89
66
288
0.4
Median
Table 1. Summary of major and trace element concentrations in Talbot surface sampling media
0.06
0.04
0.040
0.16
0.051
0.001
0.055
0.001
0.08
31
11
<2
1.5
12
0.2
1.4
3.6
4.4
4.8
0.3
0.2
0.4
0.02
0.11
0.06
<0.01
0.40
4
71
<0.2
Min.
1.06
0.72
0.490
2.79
0.270
0.032
0.176
0.008
0.34
5743
132
21
23.5
177
8.7
12.1
26.3
50.7
17.5
18.2
3.5
1.4
0.60
1.12
0.32
3.87
2.31
579
481
7.6
Max.
humus, n=112
0.18
0.13
0.067
0.42
0.031
0.004
0.027
0.001
0.05
965
26
3
3.1
30
1.1
2.3
4.5
8.6
2.3
3.2
0.7
0.2
0.09
0.20
0.06
0.38
0.32
83
80
0.7
Min.
Max.
0.068 0.044 0.959
0.06
0.03
0.74
0.110 0.090 0.290
1.17
0.49
1.64
0.171 0.104 0.199
<0.001 <0.001 <0.001
0.118 0.057 0.138
0.002 <0.001 0.006
0.11
0.05
0.14
994
122
2511
54
20
99
<2
<2
<2
2.6
1.7
18.9
79
36
141
0.2
0.1
8.0
2.4
1.2
11.5
11.8
3.4
15.6
7.0
2.6
21.3
10.6
5.5
14.2
1.1
0.2
6.4
0.3
0.2
2.0
0.6
0.4
0.8
0.04
0.02
0.24
0.16
0.14
0.50
0.08
0.04
0.19
0.05
0.04
0.82
0.65
0.41
1.10
49
23
123
204
51
350
0.5
0.3
0.8
StDev. Median
Moss, n=15
0.231
0.18
0.056
0.34
0.027
0.025
0.001
0.02
718
22
4.2
35
2.0
2.5
3.8
5.7
2.4
1.5
0.5
0.1
0.06
0.09
0.04
0.20
0.18
25
84
0.2
Min.
Max.
0.002
<0.01
0.011
0.13
0.02
<0.001
<0.001
<0.001
0.04
111
11
<2
0.6
16.1
0.03
0.2
2.3
0.48
0.64
0.08
0.6
<0.1
<0.01
0.02
<0.02
<0.1
0.04
172
<1
<0.2
<0.001
<0.01
0.004
0.06
0.01
<0.001
<0.001
<0.001
<0.01
3
4
<2
0.1
0.3
<0.01
0.1
1.0
0.03
0.14
0.01
<0.1
<0.1
<0.01
<0.01
<0.02
<0.1
0.01
28
<1
<0.2
Min.
0.050
<0.01
0.025
0.20
0.04
<0.001
<0.001
<0.001
0.24
255
28
<2
2.1
31.4
0.31
1.6
9.2
7.18
1.52
1.22
2.0
<0.1
<0.01
0.09
<0.02
<0.1
0.34
481
<1
<0.2
0.005
0.003
0.02
0.01
0.06
53
5
0.3
5.3
0.04
0.3
1.0
0.90
0.27
0.14
0.5
0.02
0.04
83
-
Max. StDev.
black spruce heartwood n=124
StDev. Median
0.013
0.005 0.024 0.003
<0.01 <0.01 <0.01
0.023
0.016 0.038 0.004
0.89
0.33
1.59
0.28
0.05
0.02
0.10
0.01
<0.001 <0.001 <0.001
0.017
0.011 0.027 0.003
<0.001 <0.001 <0.001
0.05
<0.01 0.11
0.02
206
59
573
99
89
40
251
38
<2
<2
<2
2.0
1.5
2.7
0.2
47
17
102
14
0.05
<0.01 0.12
0.03
0.4
0.1
1.0
0.2
9.9
3.7
32.5
4.4
1.49
0.37
3.23
0.59
3.27
2.26
5.70
0.51
0.12
0.03
0.34
0.06
0.2
<0.1
0.6
0.1
0.5
0.2
0.9
0.1
<0.01 <0.01 <0.01
0.02
<0.01 0.05
0.01
<0.02 <0.02 <0.02
<0.1
<0.1
<0.1
0.19
0.07
0.8
0.10
13
4
39
6
112
26
262
48
0.2
<0.2
1.3
0.2
StDev. Median
black spruce bark, n=140
70
van Geffen et al.
Talbot geochemistry
71
Fig. 3. Contoured concentrations of Fe, Ca, Al, P, Zn and Cu in near-surface till (<250 µm fraction) from the Talbot exploration grid. Major faults
are indicated as dashed white lines and the surface projection of the mineralisation-bounding basement fault zone as a solid band. Most other trace
element concentrations in the <250 µm soils display similar distribution patterns to Al across the grid, except Mo, Cd, Nb and W.
isotope ratio mass spectrometer (IRMS). Carbon isotopic
compositions of the dolomites were measured on a GasBench
IRMS. Carbon isotope-ratios are expressed as δ13CPDB, with
the Pee Dee belemnite (PDB) as the isotopic standard reference material: δ13C = (13C/12C)sample / (13C/12C)PDB – 1 (in ‰).
reSulTS And dISCuSSIOn
Soil mineralogy
The glacial till on the Talbot prospect is a mixture of silicaterich sediments from the Superior Province (Fig. 1) and carbonates of the Western Interior Basin. As determined by XRD
analysis, the mineral component of the <250 µm fraction of
near-surface till along Talbot Line 2N is composed of, on average, 25 % quartz, 21 % illite, 20 % albite, 14 % chlorite and
minor K-feldspar and dolomite. The <2 µm clay fraction of
the till comprises, on average, 38 % illite, 29 % quartz 16 %
albite and 10 % chlorite. PIMA analysis shows that the clay
fraction is dominated by illite-smectite that overwhelms other,
less reflective minerals such as chlorite. PIMA spectra indicate
minimal variation in clay type with depth and between profiles,
which implies that the composition of the clay fraction is very
uniform across the prospect.
Geochemistry of the <250 µm till fraction
The aqua-regia extractable chemistry of the <250 µm till
fraction is dominated by Fe, Al, Ca and Mg (Table 1).
Distribution of the major elements in the surface grid (1000 x
600 m) is controlled by the mineralogy of till silicates and
dolomite. Aluminum- and Fe-rich till occurs mostly in the
NW of the prospect and parallel to a NE-trending fault zone
(Fig. 3). In the western part of the area, high Ca indicates contributions of the subcropping underlying dolomite to the till.
Low Al, Fe and Ca reflect either organic-rich soil in boggy
areas with high P contents in the northern part of the prospect,
or more sand-rich till in the southeastern section. Zinc, Cu and
most other trace elements in the <250 µm fraction have distribution patterns that are very similar to Al across the prospect.
Most elements have Pearson’s correlation coefficients with Al
that are greater than 0.7, suggesting that aluminous silicates,
particularly the clay minerals, are an important source of many
elements in the till. Exceptions are Mo (PCC = 0.5), W (0.4),
and Cd (0.2), which have concentrations at or near their detection limits (Table 1). None of the elements in the <250 µm
fraction of the near-surface till shows a clear relation to the
projection of buried mineralisation or to its intersection with
fault zones in the cover dolomite.
72
van Geffen et al.
Fig. 4. Contoured concentrations of Ca, Al, Zn, Ag, Mn, Pb, Cu and C, and δ13C values in 50 cm depth-profiles of <250 µm till (left) and its <2
µm clay fraction (right), with a schematic cross section of the subsurface geology. Dashed lines indicate faults in the dolomite cover.
Depth profiles in the <250 µm till fraction along Line 2N
indicate the proximity of underlying dolomite, which is evident
by high Ca, Mg and C contents, and in the maximum profile
depth of 20 cm at 100E and 30 cm at 0E and 200E (Fig. 4). East
of 400E, Al, Fe, K, Zn, Pb, Mn, Ag, Cu and most other trace
elements including Ba, Sn, V, Ti, Zr, rare earth elements and U
are elevated near 30 cm depth, which is interpreted to represent
a developing B-horizon where elements leached from shallower
Talbot geochemistry
Fig. 5. Contoured weight percentage of recovered clay (<2 µm fraction) from Talbot till in 50 cm depth-profiles. Dashed lines indicate
faults in the dolomite cover.
Table 2. Principal component loadings of the major elements in the <250 µm till fraction
from the Talbot exploration grid (n = 151)
Factor
PC1
PC2
PC3
Eigenvalue
Cumulative
variance explained
Al
Fe
Mg
Mn
K
Ti
Na
Ca
P
6.04
67%
1.29
81%
0.82
91%
0.97
0.93
0.90
0.89
0.84
0.81
0.73
0.68
0.54
0.10
−0.15
−0.38
−0.20
0.47
0.52
0.35
−0.68
−0.06
−0.01
−0.07
−0.06
−0.21
0.15
−0.12
−0.18
−0.04
0.83
horizons are accumulated. Above the fault zones at 400E and
900E, Al and trace element concentrations are much lower and
do not show the developing B-horizon. The clay content in the
soil profile is also much lower at 400E and 900E (Fig. 5), which
can be ascribed to enhanced groundwater flow at the faults
causing drainage of the fine fraction from the till and depletion
of elements associated with the clays.
Principal component analysis of soil geochemistry
To unravel mixtures of distinct chemistry or mineralogy, multivariate statistics such as principal component analysis (PCA),
can be used. PCA seeks to reduce the number of variables by
determining vectors (eigenvectors) that describe the maximum
variance in all variables (Anderson 1984). The principal component loadings represent the correlation of each element with a
calculated factor, or principal component. The first factor, PC1,
describes the maximum variance in multivariate (multi-element)
space. The next factor, PC2, describes the maximum variance
uncorrelated (orthogonal) to PC1, and subsequent factors similarly describe ever smaller vectors of maximum variance, orthogonal to the previous ones. The eigenvalue is the proportion of
the total number of variables explained by each factor.
Geochemical differences between the till and its clay fraction may arise from mineralogical variations in the <250 µm
fraction. PCA of the major elements in the till indicates an
Al-rich component for PC1, which correlates strongly with Fe,
Mg, K, Ti, and Mn (Table 2). The strong correlation of elements with PC1, explaining 67 % of all the variance in the
major elements and 94 % of the variance in Al, implies that an
Al-rich phase dominates the till geochemistry on the Talbot
prospect (Fig. 6). The main aqua-regia extractable Al-rich
phases in the till are clay minerals. The strong correlation of
most major elements with clays may be explained by the greater
concentrations of elements in the clay fraction relative to other
phases (Table 1).
PC2 describes a factor that correlates with Ti and K on the
positive axis (Fig. 6), but more strongly with Ca on the negative
73
axis, indicating a Ca-rich phase (dolomite) that increases with
decreasing lithophile phases (till silicates and oxides). The first
two principal components, PC1 and PC2 combined, explain
81 % of the variance in all the major elements (Table 2).
PC3 correlates only with P, representing organic compounds or phosphates in the till. The correlation of P with PC1
suggests a moderate covariance with the clay phase. Although
P correlates somewhat with PC1, it plots quite far from the
unit circle on the PC1 versus PC2 diagram (Fig. 6). PC3 versus
PC1 shows that nearly all variance in P is explained by these
two factors. In effect, the Talbot till geochemistry can be
simplified from 9 major elements to a combination of three
principal components. Sulphur was excluded from the factor
analysis because many measurements were below its analytical
detection limit of 0.01 wt %.
Molar element ratios and soil mineralogy
The presence of dolomite in carbonate-rich till can be determined from the molar concentrations of Mg and Ca, which
have a 1:1 ratio if dolomite is their dominant source. The
Talbot till (<250 µm fraction) contains significant amounts of
dolomite (Fig. 7), but there is another, well defined trend that
indicates a compound containing 4 moles of Mg for every
mole of Ca, which corresponds to the Mg:Ca molar ratio in
the clay fraction. Principal component analysis shows a strong
correlation of most major elements with Al (Fig. 6), which is
interpreted to reflect the control of clay minerals on the distribution of these elements in the <250 µm fraction of the till.
Using molar ratios of the major elements normalised to Al,
the mineralogy of aqua-regia extractable phases in the till can
be estimated if the system does not comprise too many phases.
Molar ratios of the major elements normalized to Al in the
till cluster along discrete lines of constant slope and indicate
K:Al = 1:10, Ca:Al = 1:10, Mg:Al = 2:5, Fe:Al = 3:5, and by
inference, K:Ca = 1:1 and (Fe+Mg):Al = 1:1 (Fig. 7). Titanium
and Na also correlate with Al, with molar ratios of Ti:Al =
1:40 and Na:Al = 1:200. From these molar ratios, a general
composition of the Talbot clay can be deduced as K + Ca + 4
Mg + 6 Fe per 10 Al.
XRD spectra of the clay fraction are dominated by illite
(K,H3O)(Al,Mg,Fe)2(Si,Al)4O10(OH)2•H2O, quartz, feldspars
and chlorite (Mg,Fe)5Al(Si3Al)O10(OH)8 with traces of dolomite and montmorillonite (Na,Ca)0.3(Al,Mg)2Si4O10(OH)2•n(H
2O); PIMA spectra primarily reflect an illite-smectite composition. Chlorite, illite and smectite minerals, which are typical
alteration products of Fe-rich, mafic minerals such as biotite,
amphiboles and clinopyroxenes, can be combined to form a
mixture that resembles the derived clay composition. From the
ratio of (Fe+Mg):Al, a mixture of 2 parts illite plus 1 part chamosite (Fe-chlorite) would match the observed molar ratio
trend that is observed in both fractions of the till (Fig. 7).
The mineralogical variation in the till can be further assessed
by using molar element ratios (MER) with Al as the common
denominator. On a diagram of molar (Fe+Mg)/Al versus Ca/
Al, the average composition of the Talbot clay plots at 0.1 and
1.0 (Fig. 8). Departures from this point indicate mixtures with
other phases present in the sample. Mixing the Talbot clay
composition with kaolinite, Al2Si2O5(OH)4, causes a shift
towards the origin, Fe-oxides and hydroxides produce a positive shift parallel to the y-axis and samples containing a mixture of clays and dolomite/ankerite display a shift along a
mixing line with a slope of 1, since Ca, Mg and Fe are added in
equal proportions. This diagram is a projection from multielement space which omits variations in K and other stoichiometric elements (Benavides et al. 2008).
74
van Geffen et al.
Fig. 6. Correlation of major elements
with principal components of the Talbot
near-surface till geochemistry (<250 µm):
(a) PC2 vs PC1; (b) PC3 vs PC1. The
closer elements plot to the unit circle, the
greater the proportion of their variance is
explained by the corresponding factors. 1,
clays; 2, dolomite; 3, lithophile elements; 4,
organic matter; P, Phosphorus.
Molar ratios of the major elements normalised to Al in the
clay fraction are similar to those of the <250 µm fraction without a significant carbonate trend. The aqua-regia extractable
major-element molar ratios for the clay fraction on the same
MER diagram used for the soils indicate identical clay chemistry in both fractions (Fig. 8), but the <250 µm till is much more
heterogeneous compared to the illite-chamosite clay fraction
with only minor kaolinite.
Geochemistry of the <2 µm till fraction
High concentrations of Ca, Fe, Mn and Pb in the <2 µm clay
fraction in the western part of the Talbot prospect coincide
with slightly elevated topography and well-drained, thin soil
cover on top of subcropping dolomite (Fig. 9). The surface distributions of Pb, Cr, V, As, Bi and Cd show similar patterns to
that of Fe and Mn across the prospect. Calcium concentrations
show a distinct NE-trending lineament across the area and a
minor one parallel to it in the SE corner of the grid, both of
which run parallel to outcropping fault scarps observed in the
area (Fig. 9). The orientation of these lineaments corresponds
to a large-scale structure c. 20 km to the NE that is visible on
regional maps (Fig. 1) and accommodates Gladish Lake.
A well defined Zn anomaly near the centre of the prospect
around 400E along line 200N coincides with the intersection
of the cover fault with the surface projection of buried mineralisation. Minor Zn anomalies are clustered around 900E,
along the eastern cover fault. In the clay fraction of the till
depth-profiles along Line 2N, elevated Zn occurs in the top of
the profiles that overly structures at 400E and 900E, and elevated Ag occurs at 900E (Fig. 4). Anomalies in the concentration of Cu, Ag, Pb, Mn, Hg, Cd, Co, Bi, Au and Se are recorded
in the clay fraction of the lower part of the depth-profile at
400E (Fig. 4). These anomalies are coincident with the fault
that is observed in the Ca grid and intersects the depth-profile
transect (Line 2N) at 400E. The location and orientation of
these anomalies suggest that the elements were introduced to
the soil from below rather than above, as their concentrations
increase with depth in the profile. Unlike Zn, these VMS indicator elements do not show significant anomalies in the nearsurface till sampling grid, in particular Cu, Pb and Mn, because
they are trapped in deeper parts of the till profile and appear
less mobile than Zn in the surface environment.
In contrast to the <250 µm fraction, trace elements do not
correlate with Al in the clay fraction of the till, which has much
higher concentrations of Al as well as trace elements. Moreover,
background concentrations of Zn in the clay fraction are on
average around 120 ppm, with anomalous values ranging from
170–270 ppm, whereas the <250 µm fraction shows a continuum in Zn concentrations. Tukey box plots illustrate the
greater anomaly contrast of Ag, Cu, Mn, Pb, and Zn in the clay
fraction relative to the larger till fraction (Fig. 10). With the
exception of Ti, Ca and K, individual elements do not correlate
between the two size fractions (Table 3), which implies that,
apart from having lower concentrations, elements also have
different distribution patterns in the <250 µm till relative to
the clay fraction.
PCA and MER illustrate that even though the aqua regia
digest does not completely dissolve the till sample, the Al concentration is nonetheless representative of the clay content in
the <250 µm fraction. As such, Al can be used as a proxy for
clay content and trace-element concentrations as ratios to Al
would thus resemble trace-element concentrations in the clay
fraction. The ratio Zn/Al in surface till grid (Fig. 11) does
indeed resemble the distribution observed in the clay fraction
(Fig. 9) and delineates the anomaly around the intersection of
the cover fault with the projected mineralisation even more
clearly, as well as the elevated values around the eastern fault
zone. Because other indicator elements are not present at the
surface as is Zn, their concentrations levelled to Al do not produce anomalies in the till grid.
Carbon isotopes in depth-profiles
Carbon isotopic compositions of the two size fractions of
Talbot till are used to determine the origin of carbon-rich compounds in the depth profile that may act as traps for VMSrelated elements. Combustion of the samples in an elemental
analyser extracts all forms of carbon present, which in the
<250 µm till profiles varies between 0.3–10 wt%, with a
median of 2.4 % and 0.6–8.9 wt% carbon in the clay fraction,
with a median of 1.5 % (Table 4). Till depth-profiles along Line
2N indicate a carbon-rich base and a relatively carbon-poor
upper level, which is more pronounced in the <250 µm than in
the clay fraction (Fig. 4). However, the clay fraction displays
high total-carbon concentrations near the surface at the 400E,
750E and 900E profiles, and anomalous concentrations in the
lower part of the profile at 400E, coincident with observed
trace-metal anomalies in the same fraction.
Although the concentrations of total carbon in the two size
fractions are similar, variations in the δ13C values indicate distinct carbon sources. Total C extracted from the depth profiles
has a wide range of isotopic compositions, with δ13C values of
about 0 to −30 ‰ in both size fractions. Average atmospheric
CO2 has δ13C values of c..−7 ‰ (Peterson & Fry 1987) and the
carbonates derived from this source would have high δ13C values. Carbon isotope fractionation is significant in biological
Talbot geochemistry
75
Fig. 7. Relationships between molar concentrations of major elements in near-surface till from the Talbot exploration grid (black dots, <250
µm; blue diamonds, <2 µm). (a) Mg vs Ca; (b) detail of Mg vs Ca; (c) Ca vs K; (d) K+Ca vs Al; (e) Mg vs Fe; (f) Fe+Mg vs Al, displaying well
defined, discrete ratios.
76
van Geffen et al.
Fig. 8. Molar element ratio diagrams of
Fe+Mg vs Ca normalised to Al for the
(a) <250 µm fraction till; and the (b) <2
µm fraction from the Talbot grid. The
bluepoint indicates the calculated claymineral composition and dashed arrows
indicate mixing trends with other till
constituents.
Fig. 9 Contoured concentrations of Ca, Fe, Zn, Mn, Cu and Pb in the clay fraction (<2 µm) of the Talbot grid till (5–20 cm depth). Major faults
are indicated as dashed white lines and the surface projection of the mineralisation-bounding basement fault zone as a solid band.
processes, which favour the uptake of the lighter isotope 12C in
organisms, resulting in more negative values of δ13C values in
organic C. Soil organic matter (SOM) has a typical average δ13C
value of -26 ‰, whereas microbial activity may fractionate C
to extremely negative values, << −30 ‰ (Peterson & Fry
1987; Nelson et al. 2007).
The Silurian dolomite that forms the upper part of
the Talbot dolomite cover and subcrops on the prospect’s
Talbot geochemistry
77
0 ‰) at the base of the profiles reflect the underlying dolomites (Fig. 4). At 400E, the C-isotopic composition of the clay
fraction is significantly more negative throughout the profile
(δ13C < −20 ‰) compared to elsewhere, indicating a dominantly organic C source, whereas the <250 µm till at this site
indicates dominantly carbonate-sourced C.
Clay-humic complexes
Principal component analysis and molar element ratios reduce
the complexity of the aqua-regia extractable results to two
main components: clays and dolomite in the larger till fraction
and a mixture of illite and chamosite in the clay fraction. MER
diagrams show that the clay fraction reduces chemical heterogeneity and increases trace-element yield relative to the <250
µm till. The relative concentrations of most major elements
do not correlate between the <250 µm till and the corresponding <2 µm clay fraction, but the trend in molar ratios is
the same in both size fractions, except for Ti, Ca, and K
(Table 3). This reflects the control of clay minerals on the
major element molar ratios in both fractions. However, a
major constituent, organic matter, accounts for up to 9 wt%
of the clay fraction in depth profiles, which greatly affects
trace-element geochemistry (Fig. 12).
The greatest concentration anomalies in indicator elements
in the depth profiles occur at or below 30 cm depth and above
fractures (Fig. 4), suggesting that these elements were deposited in the till from fluids or gases percolating upward, particularly along structural conduits. The only indicator element that
has anomalous concentrations towards the top of the profile is
Zn in the samples above the fault zone. The apparent mobility of upward transported elements in the till profile decreases
in the order Zn > Ag ≥ Mn > Pb > Cu, which reflects the
Fig. 10. Log-scale Tukey box-and-whisker plots of Ag, Cu, Mn, Pb,
and Zn in the <250 µm and <2 µm clay fractions of the Talbot till
depth-profiles. Each box is defined by the 25th and 75th percentiles of the data and represents the inter-quartile range (IQR). The
median is represented by a white horizontal line and the mean by
a white dot. Whiskers are extended from the box to the last value
at <1.5 IQR, towards the maximum and the minimum. Samples
beyond the whiskers are considered outliers or anomalies (circles),
and those that plot at >3 IQR from the box are far outliers or highly
anomalous values (triangles).
exploration grid has a δ13C value of +0.8 ‰, and underlying
Ordovician carbonates range from −0.6 to +2.0 ‰ (Table 4).
In the upper part of the till depth-profiles, the δ13C values are
consistent with organic sources (δ13C = −30 to −20 ‰) in
both size fractions, whereas the higher values (δ13C = −8 to
Table 3. Pearson correlation coefficients of elements between Talbot sampling media (bold > 0.5)
Pearson Correlation
Fe
Al
Mg
Ca
K
Ti
P
Na
Mn
Ba
V
Cr
Zn
Li
Ni
Sr
Pb
Cu
Co
As
Se
U
Mo
Bi
W
Cd
Ag
Hg
Au
Till<250/Clay
0.21
−0.06
0.24
0.57
0.64
0.69
0.36
0.25
0.49
0.19
0.33
0.00
−0.01
−0.25
0.11
0.45
−0.36
0.24
0.03
−0.10
−0.11
0.15
0.22
0.06
−0.06
0.33
0.16
−0.16
−0.02
Wood/Bark
Till<250/Wood
Till<250/Bark
Clay/Wood
Clay/Bark
0.04
−0.03
0.31
−0.07
0.21
−0.05
−0.06
−0.09
−0.01
0.03
0.00
0.01
0.09
0.23
−0.10
0.00
−0.03
−0.11
−0.12
0.22
0.63
0.52
0.17
0.14
0.00
0.15
−0.01
−0.17
−0.02
−0.09
0.06
0.06
0.36
0.03
0.19
0.18
0.06
0.03
0.00
0.06
0.49
−0.02
−0.04
0.31
−0.19
0.24
−0.07
0.22
−0.02
0.07
0.15
0.05
0.32
0.12
−0.06
0.07
−0.05
−0.07
−0.10
−0.20
−0.11
0.23
−0.02
−0.07
0.26
0.01
0.03
0.24
−0.06
0.09
0.13
0.27
0.10
0.14
−0.08
0.05
−0.20
−0.03
−0.04
−0.12
0.01
−0.04
0.12
0.02
0.10
van Geffen et al.
78
Fig. 11. Contoured concentration ratios
of Zn/Al in Talbot grid till (<250 µm
fraction). Major faults are indicated as
dashed white lines and the surface projection of the mineralisation-bounding basement fault zone as a solid band.
Table 4. Total carbon concentrations and δ13C values in Talbot till depth profiles, <250 µm and <2 µm fractions, and dolomites
Material
C (%)
δ13C
Till <250 µm, n=68
Till <2 µm, n=68
dolomite*, n=6
Median
Min.
Max.
StDev.
Median
Min.
Max.
StDev.
Min.
Max.
2.38
−16.94
0.32
−28.63
10.39
−0.80
2.18
9.56
1.50
−24.43
0.58
−28.26
8.88
−4.69
1.91
8.28
−0.64
1.99
*Dolomite measured on GasBench IRMS, soil and clay on MAT 252 EA-IRMS
combined effects of pH, Eh, groundwater fluctuations and the
abundance of amorphous Fe-Mn oxides, clay minerals and
organic compounds on the speciation of each element (Tack &
Verloo 1995). The mobility of metal ions in groundwater can
be greatly enhanced through complexing (Leybourne &
Cameron 2010). Zinc is known to form complexes with humic
compounds and is often enriched in organic soil horizons
(Sarret et al. 2004; Christensen & Christensen 2000). In contrast, Cu complexes are considerably less mobile in the surface
environment (Tyler & McBride 1982), which may explain the
observed discrepancy between the distributions of Cu and Zn
in the depth-profile clays.
Humic and fulvic acids are effective scavengers of metal
ions from groundwater in soils, but the process of waterbased gravimetric clay separation would remove these soluble
compounds, unless they bind with clay-sized particles (Slavek
& Pickering 1981; Evangelou et al. 1999). The presence of
clay-humic complexes in the Talbot till is evident from high C
concentrations of the clay fraction in combination with low
δ13C values that indicate organic C (Fig. 12). Values of δ13C in
humic and fulvic acids typically vary between −20 and −28
‰, with δ13C of humic acid c. 1.5 ‰ more negative than fulvic
acids in the same soil (Nissenbaum & Kaplan 1972; Goh et al.
1976; Dzurec et al. 1985). The low δ13C values of the
C extracted from the Talbot clays are consistent with a much
greater proportion of organic C in the clay fraction relative to
the <250 µm fraction. Humic and fulvic acids are known to
form complexes with clay minerals, and clay minerals act as
excellent traps for metal complexes in soil (Liu & Gonzalez
1999). In the depth profiles, anomalous concentrations of
typical VMS indicator elements such as Zn, Cu, Ag and
Mn are only present in the clays with the lowest δ13C values
(Fig. 12), implying that organic compounds in the clay fraction, such as clay-humic complexes, are the dominant traps
for metals in the Talbot till.
Organic media
Dendrochemistry
Trees actively sample large volumes of soil for nutrients and
water through their extensive root systems and redistribute elements into all tissues, recording the local geochemistry in wood,
bark and foliage (Dunn 2007). The outer rings of the tree trunk
serve as active transport vessels and make up the sapwood. The
older, inner tree rings, or heartwood, do not take part in the
vital processes and serve only as support. Heartwood records
the geochemistry of the soil on which a tree grows when the
tree ring forms and, if the tree is old enough, provides a sample
material that predates sources of industrial contamination
(Reimann et al. 2007; Kozuskanich et al. 2009).
Black spruce heartwood from tree cores has the lowest
metal concentrations of all sample media discussed in this
study, with most elements below detection (Table 1). The areal
extent of pre-smelter (1915–30) heartwood is limited to oldgrowth forest and a few older trees in the western part of the
area (Fig. 13). Although areal coverage is limited and element
concentrations are generally low, Mn in tree cores is elevated
around the surface projection of buried mineralisation and its
orientation runs parallel to the faults in the dolomite (Fig. 13).
High Zn concentrations in the centre of the area are distributed
along the surface projection of buried mineralisation, coincident with the central topographic depression (Fig. 2), but do
not correspond to relative Zn concentrations in the surface clay
grid. The Pb anomaly around 600E along the 200N line in
heartwood also might indicate Pb uptake from the subsurface.
Cadmium concentrations in heartwood are very low and erratic,
and do not correspond to Cd in bark (Fig. 13).
Ion selectivity of the roots acts as a biological filter that
causes poor correlation of all measured elements between
black spruce trees and their substrate, in particular for toxic
elements such as Pb and Cd (Table 3). Within black spruce
Talbot geochemistry
79
Fig. 12. Relationship between total carbon contents and δ13C values in the (a) <250 µm till and the (b) clay fraction of the Talbot depth-profiles.
Ranges in carbon concentrations and δ13C values are similar in both size fractions, but carbonate carbon having high δ13C values dominate the
larger fraction and carbon in the clay fraction is mostly from organic sources. Relationships between (c) Zn; (d) Cu; (e) Ag; and (f) Mn concentrations and δ13C values in the clay fraction of the Talbot depth profiles indicate that anomalous metal concentrations occur only in clays that have
low δ13C values. Black dots indicate samples from the profile at 400E.
80
van Geffen et al.
Fig. 13. Contoured concentrations of Mn, Zn, Pb and Cd in black spruce heartwood (left) and bark (right). Major faults are indicated as dashed
white lines and the surface projection of the mineralisation-bounding basement fault zone as a solid band.
trees, Mn, Ba and Sr are the only elements that show significant correlation between heartwood and outer bark, which
may represent less selective distribution or similar physiological functions of these elements in wood and bark tissue
(Watmough 1997). Although root uptake is selective to certain
elements, nutrients such as Ca, Mg and Zn are taken up where
available (Österås 2004). Because the Talbot dolomites are an
important source of the total Zn in the till (Van Geffen 2011),
the availability of carbonate-Zn to trees is correlated to Ca
from dolomites, whereas excess Zn from other sources such as
buried mineralisation is not. To discount for the dolomitederived Zn, the Zn/Ca ratio can be used to define additional
Zn in the till (Fig. 14). This ratio enhances the observed Zn
anomalies in the central part of the grid, in close proximity to
the intersection of the fault with the projected mineralisation,
while reducing dolomite-derived Zn anomalies to the west
Talbot geochemistry
81
Fig. 14. Contoured concentration ratio
of Zn/Ca in black spruce heartwood.
Major faults are indicated as dashed white
lines and the surface projection of the
mineralisation-bounding basement fault
zone as a solid band.
(Fig. 13). To the SE, elevated Zn/Ca is aligned with the underlying fault (Fig. 14).
Metal concentrations in black spruce bark tend to be higher
than in heartwood, although many elements are also below
detection (Table 1). The main sources of metals in bark are
root uptake, with active redistribution into various parts of the
tree, and atmospheric deposition of dust and aerosols onto the
outer bark and foliage (Dunn 2007). Areas of elevated Mn in
bark generally correspond to those in heartwood, suggesting
that most of the Mn input into the bark is from root uptake
(Fig. 13). In contrast, elevated Pb in bark does not overlap Pb
in heartwood. High concentrations of Pb and Cd in bark along
the western edge of the old-growth forest indicate windblown
input of atmospheric sources. Similarly, Zn in bark north of
400N and west of 400E is not supported by elevated Zn concentrations in the heartwood and may indicate a greater contribution from airborne particulates in the bark. With prevailing
winds from the NW, the most likely source is the Flin Flon
Cu-Zn smelter, at 160 km to the NW.
Moss and humus
Median concentrations of Mn, Zn, Pb, Cd and Hg in humus
and moss are high relative to other sampling media (Table 1).
Manganese, Zn, Cd and Ag show similar trends in both moss
and humus along Line 2N across the Talbot prospect, suggesting similar sources of these metals in each medium (Fig. 15)
and Mn, Zn and Ag are elevated in both media in the central
part of the prospect, between 400E and 600E. Relatively high
Fe, Pb and Cd concentrations in humus at 650E correspond to
high concentrations in the clay fraction and to the vertical projection of the VMS mineralisation at depth. The corresponding
patterns of Mn, Cd and Zn in trees with Mn and Zn in moss
and humus reflect the incorporation of forest litter in moss
and humus, and the accumulation of its metal content. Iron
and Pb have similar distributions in the humus as in the clay
fraction of the till (Fig. 15). High concentrations of Pb and Cd
in humus to the west of 400E, much higher than in the till, are
interpreted as atmospheric input. High Fe and Ag in moss and
humus at 900E and 1000E reflect incorporation of dust and
dirt from the disturbed surface in the clear cut area to the east.
Environmental monitors
Much of the atmospheric input of elements from base
metal smelters is retained by organic-rich, upper soil horizons
(Reimann 1998; Reimann et al. 1998). The chemical composition
of the underlying B and C horizons is therefore less affected by
airborne contamination, which allows for the identification of
subtle, secondary indicators of mineralisation in these horizons.
The humic horizon also prevents most metal species from lower
horizons to reach the moss layer above it, making moss a more
suitable sampling medium for the monitoring of airborne contamination. Because of their exposure and texture, moss, humus
and black spruce bark are prone to atmospheric input of dust
and aerosols. Mosses do not have root systems as other plants
do, but take up nutrients from the environment through absorption and are therefore often used as environmental monitors for
atmospheric deposition of metals (Steinnes 1995).
The affinity of VMS-related trace elements such as Zn, Cu,
Pb, Ag and Cd with humic substances would make humus an
ideal sample medium, provided that: (1) these elements are
transported to the top of the soil profile, where they would
precipitate in the LFH horizon; and (2) that the relative element input from other sources such as aerosols is negligible.
Unfortunately, neither condition is met at the Talbot prospect.
The clay fraction of till depth-profiles shows that Zn is the
only indicator element that is transported to the top of the
profile, particularly along major structures in the underlying
dolomite, but this is not reflected in Zn concentrations in
humus or moss along Line 2N.
The parallel distributions of Fe in humus and clay suggest
that Fe in humus is predominantly derived from the underlying
till. The two dominant sources of trace metals in humus and
moss appear to be forest litter (Mn, Zn and Ag) and atmospheric deposition of dust and aerosols (Pb and Cd). Manganese
and Zn in humus and moss have similar distributions as in
black spruce heartwood and bark, reflecting the input of forest
litter, derived primarily from trees as needles, twigs and bark.
The erratic distribution of Pb and Cd in humus and moss
across the prospect, with concentrations greater than the
underlying clays (Fig. 15), as well as the discrepancy between
Pb in black spruce bark and heartwood (Fig. 13), indicate a
significant input of atmospheric Pb and Cd, likely from the
Flin Flon smelter.
Proposed mechanism for upward element migration
The observed indicator and pathfinder element anomalies in
the till cover on the Talbot prospect require an influx of these
elements from the mineralisation, through 100 m of dolomite
cover, to the surface. This influx must have occurred syn- or
post-glaciation, which was responsible for the deposition of
82
van Geffen et al.
Fig. 15. Concentrations of Fe, Mn, Pb, Zn, Cd and Ag in humus (solid black line), sphagnum moss (dashed black line), <250 µm till (dashed grey
line), and <2 µm till (solid grey line) along Line 2N across the Talbot prospect, with a schematic cross section of the subsurface geology and surface vegetation: new growth to the west, old growth in the centre, and clear-cut ground to the east.
the till veneer at between 13–10 ka (Teller & Leverington
2004). Because of the competent nature and lateral extent of
the Palaeozoic dolomites, ion fluxes promoted by a natural
electrochemical cell (Hamilton 1998) or capillary action (Mann
et al. 2005) are unlikely to have played a significant role in this
environment, as these mechanisms require a permeable over-
burden with a great degree of connectivity. Microbial transport
may have been instrumental in ion mobilisation from sulphide
mineralisation and fluxes up structural conduits such as joints
and faults, mostly as a result of microbially produced gas bubbles. Remnants of microbial communities would produce
more negative values of δ13C values around the depth profile
Talbot geochemistry
83
Fig. 16. Schematic overview of processes involved in vertical element migration in the Talbot subsurface. Oxidising sulphides produce sulphuric
acid, creating (a) subsurface karst voids in the cover dolomite; (b) glacial loading deposits till and causes fracturing of the dolomite cover, creating
fluid pathways; (c) remnant excess hydrostatic pressure causes upward fluid flow during glacial retreat; (d) VMS-related elements are deposited in
the soils in the till. Modified after Boulton & Caban (1995).
anomaly (Nelson et al. 2007) and active biological transport of
metal ions over great distances by microbes has not yet been
reported (Aspandiar et al. 2008). Gas-bubble transport in the
saturated groundwater regime, whether microbially induced or
originating from the degassing as ‘geogas’, could potentially
have carried elements upward and contributed to the traceelement anomalies at the surface (Malmqvist & Kristiansson
1984; Kristiansson et al. 1990; Etiope & Lombardi 1996;
Putikov & Wen 2000).
To account for the geologically recent influx of VMS indicator elements, we propose an additional mechanism of upward
fluid transport that involves active advection of groundwater as
a result of remnant glacial tectonic pressure. Oxidising massive
sulphide bodies, particularly those containing significant pyrite,
produce not only sulphate and metal ions, but also considerable
amounts of H+. Carbonates will react with the acidity, producing CO2 and creating karst voids after dissolution (Fig. 16).
Metals that were initially mobilised from the ore body at low pH
may precipitate in these karst voids due to pH-buffered solutions in the carbonate-dominated environment. The process of
sulphide oxidation, acid production and carbonate dissolution
will stagnate unless oxidizing agents are continuously supplied
or sulphide oxidation is catalysed by microbial processes, which
occur even at great depths in the earth’s crust (Moser et al. 2003).
The resultant karst cavities can collapse under glacial tectonic
loading of extensive ice sheets during major glaciations, which
the Talbot area experienced throughout the Quaternary (Klassen
1997). Deep-seated karst voids at the base of the carbonates
have been reported in HBED drill holes from around the Talbot
area and normal faults with vertical offsets of up to 4 m have
been observed.
The enhanced permeability created by subsurface karst
collapse serves as a conduit for groundwater flow and gasbubble migration (Fig. 17). During deglaciation, the rate of
retreat of an ice sheet is much greater than the rate of vertical
isostatic rebound of the lithosphere. The presence of permafrost or competent rocks at the surface maintains excess
hydrostatic pressure, which forces fluids upwards from deep
aquifers in areas of enhanced permeability (Boulton & Caban
1995; Boulton 1996; Fig. 16). Boulton (1996) observed evidence for the physical process of upward groundwater flow
in glacial moraine deposits in Europe, and this process may
also have facilitated upward element migration (Fig. 16). Any
soluble metal ions, metal-organic complexes, or otherwise
mobile compounds released from the oxidizing ore body and
residing at the base of the cover sequence or in faults and collapse breccias are carried to the surface where they are deposited in the glacial sediments that were left behind by the
retreating ice sheet.
This mechanism, likely aided by processes such as microbial mobilisation and gas-bubble transport, may explain how
enhanced mass transport can occur through faults in extensive, competent overburden during geologically recent and
relatively short-lived episodes (Fig. 17). The viability of the
mechanism is supported by several observations, including
documented karst voids occurring at the base of the cover
dolomites discovered in nearby drill holes, fractures in the
cover dolomites, a topographic depression parallel to the orientation of mineralisation, and structurally controlled metal
anomalies in the till profile, showing evidence of upward
transport of VMS-related elements deposited in Quaternary
sediments.
84
van Geffen et al.
Fig. 17. Summary diagram of the Talbot subsurface geochemistry with element reservoirs (square boxes) and major
transport processes (ovals). Elements in
bold font are precipitated in solid phases.
Asterisks indicate deposition of elements
in clay-humic complexes. The dashed
white line marks the δ13C = −20‰ contour in the <2 µm fraction.
Implications for exploration
The Talbot prospect is not unique in its geological and environmental setting. In large parts of Canada, the crystalline
basement is covered by carbonates and glacial deposits, posing
similar challenges for the application of geochemical methods
for exploration. In this study, the greatest anomalies of indicator elements occur in the clay fraction of till depth-profiles, at
or below 30 cm depth and over structures, suggesting that
these elements were deposited in the till from fluids or gases
percolating upward through structural conduits. The only indicator element that appears more mobile towards surface is Zn,
which forms an anomaly above the fault zone in the upper part
of the till profile and in the surface grid clays. An overview of
the geochemical reservoirs and transport processes in the
Talbot system is shown in Figure 17.
As a sampling medium, the <2 µm clay size fraction is preferred over the <250 µm fraction of the till because: (1) the
mineralogical heterogeneity of the sample is reduced; (2) trace
element concentrations in the clay fraction are generally higher;
(3) anomaly contrast is greater; and (4) clay-humic complexes
are the dominant traps for indicator elements in the Talbot till,
as illustrated by carbon isotope-ratios (Fig. 12). In areas of
extensive allochthonous cover and wet, temperate to boreal
climates, the main compounds that host indicators of buried
mineralisation as secondary species in soils are clay minerals,
organic compounds, Fe and Mn hydroxides, or a combination
of these (Cameron et al. 2004), which occur dominantly in the
clay fraction.
Black spruce heartwood from tree cores provides presmelter sampling material that is light-weight and easily accessible, and represents a large volume of soil that is sampled by
the tree’s extensive root system. If the area has sufficient coverage of trees that are older than sources of industrial contamination, tree cores are a potential alternative to soils as a
sample medium, particularly in areas where metal input from
mining and smelting operations is of concern, as well as in
remote areas in which sample weight may otherwise be limiting
to the number of samples in a survey. At the Talbot prospect,
where significant dolomite cover hampers the mobility of
many ore-indicators, the presence of secondary Zn from the
soil in tree cores is best reflected in the Zn/Ca ratio, as this
eliminates the effect of dolomite-derived Zn.
Humus, moss and bark are sensitive to airborne metal input
and are therefore of limited use to vector towards deeply buried ore deposits in areas where sources of atmospheric contamination are of concern. However, in combination with
other media, they can be used to assess relative contributions
of atmospheric metal deposition at great distances from its
source (150–200 km), as recorded by the elevated concentrations of Mn, Pb, Cd, Zn, Ag, and As in humus and moss at
Talbot. Because humus consists of decomposed vegetation, it
integrates and concentrates the influx of airborne contaminants over time and space and can as such be used to estimate
long-term atmospheric deposition.
COnCluSIOnS
The results of this study demonstrate that VMS mineralisation
at the Talbot prospect, buried under more than 100 m of dolomite and glacial till, is expressed in the till as secondary anomalies of Zn, Cu, Ag, Pb, Au, Mn, Hg, Cd, Co, Bi and Se. The
greatest anomalies occur in the clay fraction, above faults and
mostly at or below 30 cm depth, suggesting that elements were
deposited in the till after upward migration through structures.
Chemostratigraphy of the depth-profile clays shows that secondary Zn was deposited near the surface, whereas other indicator elements are trapped in deeper parts of the profile and
hence lack anomalies in the near-surface till. With Al as a proxy
for the clay content, the Zn/Al ratio in the <250 µm till fraction represents Zn concentrations in the clay fraction, which
produce high-contrast anomalies along structures. Carbon isotopic compositions indicate that the metal anomalies in the till
are primarily associated with clay-organic complexes. To
account for the observed influx of indicator elements into the
recently deposited glacial till, a novel mechanism of upward
element migration is proposed that is driven by advection
of groundwater under remnant hydrostatic pressure from a
Talbot geochemistry
retreating ice sheet. Separation of the <2 µm clay fraction produces more consistent element concentrations and increases
trace-element yield relative to the <250 µm fraction, as demonstrated by PCA and MER analysis. Pre-smelter black spruce
heartwood indicates secondary Zn in the Zn/Ca ratio, which
discounts Zn from dolomites. Atmospheric deposition of Pb
and Cd from the Flin Flon smelter at 160 km to the NW is
recognised in humus, moss and black spruce bark, which act as
sensitive environmental monitors for airborne metal input.
We are indebted to Kelly Gilmore and other staff at HudBay Exploration and Development for lending us access to the Talbot property
and providing field assistance. Special thanks to Bill MacFarlane for
sample collection and for editing this paper. Thanks to April Vuletich
and Kerry Klassen for their help with isotope analyses at the QFIR
laboratory. Susan Flasha and Paul Alexandre are acknowledged for
proof-reading of the manuscript. Financial and material support for
this research was provided by Anglo American plc, the Natural Sciences and Engineering Research Council of Canada (NSERC) CRD
program, Queen’s University, the Canadian Foundation for Innovation and the Ontario Research Initiative.
reFerenCeS
Anderson, T.W., 1984. An Introduction to Multivariate Statistical Analysis. 2nd
Edition, Wiley & Sons, New York, USA.
AspAndiAr, M.F., AnAnd, R.R. & GrAy, D.J. 2008. Geochemical dispersion mechanisms through transported cover: Implications for mineral exploration in Australia.
CRC-LEME, CSIRO Open File Report Series.
BAiles, A.H. & GAlley, A.G. 1999. Evolution of the Paleoproterozoic Snow
Lake arc assemblage and geodynamic setting for associated volcanic-hosted
massive sulphide deposits, Flin Flon Belt, Manitoba, Canada. Canadian
Journal of Earth Sciences, 36, 1789–1805.
BenAvides, J., Kyser, T.K., ClArK, A.H., stAnley, C.R. & oAtes, C.J. 2008.
Application of molar element ratio analysis of lag talus composite samples to
the exploration for iron oxide-copper-gold mineralization: Mantoverde area,
northern Chile. Geochemistry: Exploration, Environment, Analysis, 8, 369–380.
Bezys, R. & BAmBurAK, J., 2001. Stratigraphic investigations and corehole
drilling program, 2001. Report of Activities, Manitoba Geological Survey.
Winnipeg, Manitoba, Canada, 159–163.
Boulton, G.S. & CABAn, P. 1995. Groundwater flow beneath ice sheets:
Part II–Its impact on glacier tectonic structures and moraine formation.
Quaternary Science Reviews, 14, 563–587.
Boulton, G.S. 1996. The impact of glaciation on the groundwater regime of
Northwest Europe. Global and Planetary Change, 12, 397–413.
CAmeron, E.M., HAmilton, S.M., leyBourne, M.I., HAll, G.E.M. &
mCClenAGHAn, M.B. 2004. Finding deeply buried deposits using geochemistry. Geochemistry: Exploration, Environment, Analysis, 4, 7–32.
CHristensen, J.B. & CHristensen, T.H. 2000. The effect of pH on the complexation of Cd, Ni and Zn by dissolved organic carbon from leachatepolluted groundwater. Water Research, 34, 3743–3754.
CrAiG, H. 1953. The geochemistry of the stable carbon isotopes. Geochimica et
Cosmochimica Acta, 3, 53–92.
dunn, C.E. 2007. Biogeochemistry in Mineral Exploration. Handbook of Exploration
and Environmental Geochemistry, 9, Elsevier, The Netherlands.
dzureC, R.S., Boutton, T.W., CAldwell, M.M. & smitH, B.N. 1985. Carbon
isotope ratios of soil organic matter and their use in assessing community
composition changes in Curlew Valley, Utah. Oecologia, 66, 17–24.
elliott, C.G. 1996. Phanerozoic deformation in the “stable” craton,
Manitoba, Canada. Geology, 24, 909–912.
etiope, G. & lomBArdi, S. 1996. Laboratory simulation of geogas microbubble flow. Environmental Geology, 27, 226–232.
evAnGelou, V., mArsi, M. & vAndiviere, M. 1999. Stability of Ca 2+-, Cd
2+-, Cu 2+-[illite-humic] complexes and pH influence. Plant and Soil, 213,
63–74.
GAlley, A.G., syme, R. & BAiles, A.H. 2007. Metallogeny of the Paleoproterozoic
Flin Flon Belt, Manitoba and Saskatchewan. Mineral Deposits of Canada: A
Synthesis of Major Deposit-Types, District Metallogeny, the Evolution of Geological
Provinces, and Exploration Methods. Geological Association of Canada, Mineral
Deposits Division, Special Publication, 5, 509–531.
GoH, K.M., rAfter, T.A., stout, J.D. & wAlKer, T.W. 1976. The
Accumulation of Soil Organic Matter and Its Carbon Isotope Content in
a Chronosequence of Soils Developed on Aeolian Sand in New Zealand.
European Journal of Soil Science, 27, 89–100.
85
Govett, G.J.S. 1976. Detection of deeply buried and blind sulphide deposits
by measurement of H+ and conductivity of closely spaced surface soil samples. Journal of Geochemical Exploration, 6, 359–382.
HAmilton, S.M. 1998. Electrochemical mass-transport in overburden: a
new model to account for the formation of selective leach geochemical anomalies in glacial terrain. Journal of Geochemical Exploration, 63,
155–172.
HAmilton, S.M., CAmeron, E.M., mCClenAGHAn, M.B. & HAll, G.E.M. 2004.
Redox, pH and SP variation over mineralization in thick glacial overburden. Part II: field investigation at Cross Lake VMS property. Geochemistry:
Exploration, Environment, Analysis, 4, 45–58.
Henderson, P.J., mCmArtin, I., HAll, G.E.M., perCivAl, J.B & wAlKer, D.A.
1998. The chemical and physical characteristics of heavy metals in humus
and till in the vicinity of the base metal smelter at Flin Flon, Manitoba,
Canada. Environmental Geology, 34, 39–58.
Kelley, D.L., Kelley, K.D., CoKer, W.B., CAuGHlin, B. & doHerty,
M.E. 2006. Beyond the Obvious Limits of Ore Deposits: The Use of
Mineralogical, Geochemical, and Biological Features for the Remote
Detection of Mineralization. Economic Geology, 101, 729–752.
KlAssen, R., 1997. Glacial history and ice flow dynamics applies to drift prospecting and geochemical exploration. In: GuBins, A.G. (ed.) Proceedings of
Exploration. PDAC, Toronto, Canada, 221–231.
KozusKAniCH, J.C., Kyser, T.K., mACfArlAne, W.R. & HAmilton, S.M.,
2009. Dendrochemical variation over the Cross Lake VMS mineralization - a tool for mineral exploration and decoupling anthropogenic input
from background signals. Geochemistry: Exploration, Environment, Analysis,
9, 151–157.
KristiAnsson, K., mAlmqvist, L. & persson, W. 1990. Geogas prospecting: a new tool in the search for concealed mineralizations. Endeavour,
14, 28–33.
leyBourne, M.I. & CAmeron, E.M. 2010. Groundwater in geochemical exploration. Geochemistry: Exploration, Environment, Analysis, 10, 99–118.
liu, a & GonzAlez, R. 1999. Adsorption/Desorption in a System Consisting
of Humic Acid, Heavy Metals, and Clay Minerals. Journal of colloid and interface
science, 218, 225–232.
mAlmqvist, L. & KristiAnsson, K. 1984. Experimental evidence for an
ascending microflow of geogas in the ground. Earth and Planetary Science
Letters, 70, 407–416.
mAnn, A.W., Birrell, R.D., mAnn, A.T., HumpHreys, D.B. & perdrix, J.L.
2005. Vertical ionic migration: mechanisms, soil anomalies, and sampling depth for mineral exploration. Geochemistry: Exploration, Environment,
Analysis, 5, 201–210.
mAtile, G.L.D. et al. 2002. Development of methods for 3-D geological
mapping of southern Manitoba Phanerozoic terrane. Report of Activities,
Manitoba Geological Survey. Winnipeg, Manitoba, Canada, 274–282.
mCmArtin, I., Henderson, P.J., nielsen, E. & CAmpBell, J.E. 1996. Surficial
geology, till and humus composition across the Shield margin, north-central
Manitoba and Saskatchewan: geospatial analysis of a glaciated environment.
Geological Survey of Canada, Ottawa, Open File Report, 3277.
moser, D.P., onstott, T.C., fredriCKson, J.K. et al. 2003. Temporal
Shifts in the Geochemistry and Microbial Community Structure of an
Ultradeep Mine Borehole Following Isolation. Geomicrobiology Journal, 20,
517–548.
nelson, M.A., Kyser, T.K., ClArK, A.H. & oAtes, C.J. 2007. Carbon
Isotope Evidence for Microbial Involvement in Exotic Copper Silicate
Mineralization, Huinquintipa and Mina Sur, Northern Chile. Economic
Geology, 102, 1311–1320.
nissenBAum, A. & KAplAn, I. 1972. Chemical and isotopic evidence for the
in situ origin of marine humic substances. Limnology and Oceanography, 17,
570–582.
o’dAy, P.A. 1999. Molecular environmental geochemistry. Reviews of Geophysics,
37, 249.
Österås, A.H. 2004. Interactions between calcium and heavy metals in Norway spruce.
PhD Thesis, Stockholm University, Sweden.
pAdillA, K.L. & Anderson, K.A., 2002. Trace element concentration in treerings biomonitoring centuries of environmental change. Chemosphere, 49,
575–585.
peltolA, P. & åstrÖm, M., 2003. Urban geochemistry: a multimedia and multielement survey of a small town in northern Europe. Environmental geochemistry and health, 25, 397–419.
peterson, B.J. & fry, B. 1987. Stable Isotopes in Ecosystem Studies. Annual
Review of Ecology and Systematics, 18, 293–320.
pHillips, S.F., wotton, D.L. & mCeACHern, D.B. 1986. Snow chemistry in
the flin flon area of Manitoba, 1981–1984. Water, Air, & Soil Pollution, 30,
253–261.
putiKov, O. & wen, B. 2000. Geoelectrochemistry and stream dispersion.
Handbook of Exploration Geochemistry, 7, 17–79.
86
van Geffen et al.
reimAnn, C. 1998. Comparison of elemental contents in O- and C-horizon
soils from the surroundings of Nikel, Kola Peninsula, using different grain
size fractions and extractions. Geoderma, 84, 65–87.
reimAnn, C. et al. 2007. Element contents in mountain birch leaves, bark
and wood under different anthropogenic and geogenic conditions. Applied
Geochemistry, 22, 1549–1566.
reimAnn, C. et al. 1998. Environmental geochemical atlas of the central Barents region.
Geological Survey of Norway, Norway.
rose, A.W., HAwKes, H.E. & weBB, J.S. 1979. Geochemistry in Mineral
Exploration. Academic Press, London and New York.
sAArelA, K., HArju, L., rAjAnder, J., lill, J., Heselius, S., lindroos, A. &
mAttson, K. 2005. Elemental analyses of pine bark and wood in an environmental study. The Science of the Total Environment, 343, 231–241.
sArret, G., BAlesdent, J., Bouziri, L. et al. 2004. Zn speciation in the organic
horizon of a contaminated soil by micro-X-ray fluorescence, micro-and
powder-EXAFS spectroscopy, and isotopic dilution. Environmental Science
& Technology, 38, 2792–2801.
sAvArd, M.M., BonHAm-CArter, G.F. & BAniC, C.M. 2006. A geoscientific
perspective on airborne smelter emissions of metals in the environment: an
overview. Geochemistry: Exploration, Environment, Analysis, 6, 99–109.
slAveK, J. & piCKerinG, W.F. 1981. The effect of pH on the retention of Cu,
Pb, Cd, and Zn by clay-fulvic acid mixtures. Water, Air, & Soil Pollution, 16,
209–221.
smee, B.W. 1983. Laboratory and field evidence in support of the electrogeochemically enhanced migration of ions through glaciolacustrine sediment.
Journal of Geochemical Exploration, 19, 277–304.
smee, B.W. 1998. A new theory to explain the formation of soil geochemical
responses over deeply covered gold mineralization in arid environments.
Journal of Geochemical Exploration, 61, 149–172.
steinnes, E. 1995. A critical evaluation of the use of naturally growing moss to
monitor the deposition of atmospheric metals. Science of the Total Environment,
160/161, 243–249.
stern, R.A. et al. 1995. Paleoproterozoic (1.90-1.86 Ga) arc volcanism in the
Flin Flon Belt, Trans-Hudson Orogen, Canada. Contributions to Mineralogy
and Petrology, 119, 117–141.
syme, E.C. & BAiles, A.H. 1993. Stratigraphic and tectonic setting of early
Proterozoic volcanogenic massive sulfide deposits, Flin Flon, Manitoba.
Economic Geology, 88, 566–589.
tACK, F.M.G. & verloo, M.G. 1995. Chemical Speciation and Fractionation
in Soil and Sediment Heavy Metal Analysis: A Review. International Journal of
Environmental Analytical Chemistry, 59, 225–238.
teller, J.T. & leverinGton, D.W. 2004. Glacial Lake Agassiz: A 5000 yr
history of change and its relationship to the δ18O record of Greenland.
Geological Society of America Bulletin, 116, 729–742.
tyler, L.D. & mCBride, M.B. 1982. Mobility and extractability of cadmium,
copper, nickel, and zinc in organic and mineral soil columns. Soil Science,
134, 198–205.
vAn Geffen, P.W.G. 2011. Geochemical indicators of buried sulphide deposits under
carbonate cover in glaciated terrain. PhD thesis, Queen’s University, Kingston,
Ontario, Canada.
wAtmouGH, S.A. 1997. An evaluation of the use of dendrochemical analyses in
environmental monitoring. Environmental Reviews, 5, 181–201.
Received 24 January 2011; revised typescript accepted 24 July 2011.