Groundwater in geochemical exploration

M.I. Leybourne1,* & E.M. Cameron2

1
GNS Science, P.O. Box 30-368, Lower Hutt 5040, New Zealand
2
Eion Cameron Geochemical Inc., 865 Spruce Ridge Road, Carp, Ontario, K0A 1L0, Canada
*Corresponding author (e-mail: m.leybourne@gns.cri.nz)

ABSTRACT: Groundwater is an important medium for geochemical exploration of

many different styles of mineralization, including porphyry copper, volcanogenic
massive sulphide (VMS), sandstone uranium, and gold. Groundwater recharges to
depth, resulting in greater likelihood of interacting with buried mineralization
compared to surface geochemical methods, and thus providing a three-dimensional
perspective. Advances in the understanding of ore formation processes, water-rock
interaction and element transport/attenuation in the secondary environment are
enhancing the efficacy of groundwater geochemical exploration. This paper
describes key techniques and methodologies for sampling, analysis and interpretation
of groundwater geochemical data, and provides two different approaches for use by
industry: routine exploration and research approaches. New advances in analytical
methods are providing new isotopic systems and improving the cost and speed of
traditional isotopic techniques, which can greatly aid in interpretation of water
sources, water-rock reactions and fingerprinting of ore sources. Case studies are
presented for the use of groundwater geochemistry around a porphyry copper
deposit in the hyperarid Atacama Desert of Chile, and VMS mineralization in a
mature mining camp in Canada. This paper also summarizes key elemental
associations for successful utilization of aqueous geochemistry in mineral explora-
tion. The most successful aqueous-phase indicators of mineralization are those that
are associated with the ore and are mobile in solution.

KEYWORDS: groundwater, mineral exploration, review, deposit types, isotopes

Because many surface and near surface mineral deposits have concentrations are low, enhancing anomaly contrast (Giblin
been discovered, the challenge for mineral exploration is to find 1994). Although aqueous geochemistry as an exploration tool
new, more deeply buried deposits, particularly in areas where has been advocated and studied for some time (Cameron 1977,
thick cover exists (Cameron et al. 2004; Reith et al. 2005) and in 1978; Miller et al. 1982; Earle & Drever 1983; Giblin & Snelling
established mining camps (Goodfellow et al. 2003). Ground- 1983; Taisaev & Plyusnin 1984; Boyle 1988; Giblin 1994; Cidu
water has the potential to be a powerful mineral exploration et al. 1995; Mingqi et al. 1995; Cameron et al. 1997; Leybourne
tool in these environments for several reasons: (1) recent and et al. 1998, 2002; Leybourne et al. 2003; Cameron et al. 2004;
continuing advances in analytical methods, in particular induc- Phipps et al. 2004; Cameron & Leybourne 2005; Leybourne &
tively coupled plasma-mass spectrometry (ICP-MS) and, more Cameron 2006a; Eppinger et al. 2007; Leybourne & Cameron
recently multi-collector ICP-MS (MC-ICP-MS) and reaction/ 2008), there have been relatively few detailed studies, in
collision cell ICP-MS, have resulted in lower detection limits, particular on the utility of groundwaters. Another significant
more rapid sample throughput, and new methods for overcom- change in mineral exploration since Exploration ’97 and the
ing mass interferences for key species (e.g. Se, As) and for review of aqueous geochemistry in exploration (Taufen 1997) is
measurement of non-traditional isotopes (e.g. Fe, Zn, Cu, Se, the increase in the use of partial (generally weak) extractions of
Zn, Hg); (2) groundwaters recharge to depth so that there is soils and stream sediments, which in some cases have been
potential to penetrate deeply into the Earth’s crust, with the shown to produce greater anomaly to background contrast than
result that groundwater geochemistry offers the potential to traditional strong-acid (e.g. aqua regia) digestions (Cameron et al.
explore into the third dimension and detect more deeply buried 2004). One of the difficulties of these weak leaches has been
mineralization than other geochemical methods; (3) ground- the interpretation of the results owing to a lack of understand-
waters are chemically reactive with mineralization and host ing of the mechanisms by which ore-related species migrate
rocks, in particular where waters are O2-bearing; (4) ground- from depth to the surface environment (Goldberg 1998; Smee
water flows away from the site of reaction with mineralization, 1998; Cameron et al. 2004). Processes postulated to account for
providing a potentially broader exploration target (footprint) these surface anomalies over deeply buried mineralization
than lithogeochemistry; (5) sample preparation is relatively include capillary migration of ions (Mann et al. 2005), vapour-
simple, i.e. most modern analytical techniques operate with a transport (Smee 1998), ground and soil water advection or
fluid sample; and (6) for many species of interest, background diffusion (Govett 1976), electrochemical transport (Hamilton
Geochemistry: Exploration, Environment, Analysis, Vol. 10 2010, pp. 99–118 1467-7873/10/$15.00  2010 AAG/Geological Society of London
DOI 10.1144/1467-7873/09-222
100 M. I. Leybourne & E. M. Cameron

groundwaters is practical: (1) in areas where domestic or

agricultural wells already exist; (2) springs, seeps and gaining
streams (and some lakes) represent areas where groundwater
returns to surface; and (3) where exploration holes are drilled,
the additional cost of adding groundwater chemistry is minor
relative to the cost of drilling. Because groundwaters flow away
from mineralization, there may be geochemical vectors to
mineralization even in an apparently barren hole. Drilling is
expensive and typically only occurs after a suitable exploration
target has been identified. In Western Australia, thick regolith
cover is an impediment to geochemical exploration of any kind.
In this environment, mineral exploration commonly involves
drilling through cover to bedrock. The advantage of ground-
water geochemistry is that groundwaters can provide a broader
anomaly target than regolith geochemistry, with the result that
the density of drill holes can potentially be reduced (Carey et al.
2003). Pauwels et al. (2002), for example, used some domestic
Fig. 1. Conceptual model for the development of soil geochemical
anomalies via movement of groundwaters to surface during seismic wells in their surveys around two deposits in the Iberian Pyrite
events. In this model, saline fluids interact with porphyry copper Belt, Spain, and were able to use groundwater geochemistry
mineralization and mix with a more regional shallow groundwater (anomalies in Cu, Zn, Pb and Cd) to vector to known
system. The resultant soil geochemical anomalies reflect the com- mineralization.
position of the saline groundwaters (in the case of the Spence
deposit, Na + Cl) and porphyry copper mineralization (Cu, Se, Re,
Mo, As). Modified from Cameron & Leybourne (2005). GROUNDWATER SAMPLING TECHNIQUES
There are a number of different ways that groundwaters in
1998), and seismic pumping (Cameron et al. 2002; Cameron exploration holes or other wells can be sampled, with varying
et al. 2004). Given the wide variety of terrains and climates that degrees of complexity and practicality and quality of the
occur, it is likely that no one mechanism will explain soil resultant information. Some of the more commonly used
geochemical anomaly development, and that in many environ- methods include flow-through bailers, down-hole pumps (e.g.
ments, groundwaters will be directly or indirectly involved. For double-valve or Grundfos), and straddle-packer systems.
example, recent work in the Atacama Desert of northern Chile Bailers are commonly used because they are inexpensive,
has indicated that seismic pumping of groundwaters interacting simple to operate, and permit relatively rapid sampling of
with mineralization is a viable mechanism for producing groundwater. Bailers consist of a PVC tube, generally holding
significant anomalies in soils and gravels over porphyry copper a litre, with neutrally buoyant stop valves that remain open
mineralization (Cameron et al. 2002; Cameron et al. 2004). (Fig. during bailer descent and close on retrieval. Bailers work best in
1). Clearly, some of the postulated mechanisms for these unconfined aquifers of porous media where there is little
anomalies in soils will have limited applicability to aqueous variation in hydraulic conductivity with depth. However, bailers
geochemical exploration, but in some cases, as noted for are difficult to decontaminate between samples within a well
porphyry copper deposits in Chile, hydromorphic transport is and between wells, the volumes are generally limited (although
fundamentally important in the development of geochemical any given depth can be resampled if needed), and where there
anomalies and understanding the aqueous geochemistry of are significant hydraulic gradients, the groundwater from the
water-deposit interaction can greatly improve our ability to most dominant fractures may be the only one sampled,
predict styles and composition of surficial anomalies. regardless of depth (Leybourne et al. 2002). Bailers also tend to
This paper represents a synthesis of sampling methods, disturb the well, which can result in an anomalous increase in
analytical methods and types of anomalies relevant to mineral suspended sediment concentrations. However, analysis of the
exploration; two case studies are also presented to demonstrate suspended sediment can prove useful; for example, adsorption
these aspects of hydrogeochemical exploration. of Pb to particle surfaces is typically much greater than is the
Geochemical mineral exploration is predicated on the devel- case for Zn. Leybourne (2001) showed that Pb/Zn ratios of
opment of elemental or isotopic anomalies relative to the groundwater suspended sediments increased towards VMS
regional or local geochemical background. In order to develop (volcanogenic massive sulphide) mineralization.
good conceptual models and therefore effective exploration Down-hole pumps are similar to bailers in that they are
strategies, it is important to understand the mechanisms by typically used in an open borehole. In the case of double-valve
which geochemical anomalies develop in different settings and pumps, a one-way valve is opened by the overlying water
deposit types and how these are then reflected in different column and groundwater flows into the sample chamber. A
sample media. Surface geochemical methods rely on develop- source of clean gas (compressed air or N2 ) is forced down one
ment of anomalies by mechanical means (i.e. via glacial trans- PVC line closing the valve and returning sample to surface via
port, solifluction, or mass wasting), hydromorphic (water) a second PVC or Teflon line. The double-valve pump can then
transport and precipitation, or metal migration via one of the be evacuated multiple times, effectively purging the well of its
mechanisms discussed above. In contrast, anomalies observed standing water until electrical conductivity (as a proxy for
via groundwater geochemistry are fundamentally developed by salinity), Eh (redox state) and pH reach steady-state conditions
hydromorphic dispersion. reflecting those of water in the aquifer. Once steady-state
In some environments, such as northern Chile or parts of conditions are reached, the sample is taken on the assumption
Australia, grid drilling is common and ideal for the purposes of that the sample represents groundwaters flowing out of the
groundwater geochemical exploration. However, even where aquifer at the sampled depth (Leybourne & Cameron 2006a).
exploration wells are not typically densely drilled, there are a This repeated sampling to develop the well also serves to
number of ways in which analysis of the geochemistry of decontaminate the sampling apparatus between boreholes or
 Groundwater in geochemical exploration 101

difference between the two methods in terms of major ions and

ore-related metals and metalloids, significant differences were
observed for Eh (Fig. 2).
Rocks dominated by fracture-flow complicate recovery of
samples that represent groundwater flow at a specific depth.
These complications include: (1) contamination by drilling
waters; (2) mixing within open boreholes by groundwaters
entering the hole from different depths; and (3) commonly low
hydraulic conductivities (Bottomley et al. 1994). Sampling under
these conditions is best achieved using a straddle-packer
system, which permits sampling of groundwaters from discrete
intervals in a borehole (typically 1–3 m in length) (Leybourne
et al. 2002). The straddle-packer system is optimal for collecting
samples that represent groundwater flow at the sampled depth,
but suffers from high cost and time to collect samples. For
exploration holes that are older than a few years, groundwater
flow should have removed drilling fluids.
Hydrogeochemical methods in general are not used as
commonly as they should be by the exploration industry
because of a perceived complexity of sampling and treatment.
Depending on the nature of the exploration program, ground-
water geochemical exploration can be carried out in a simple,
routine, and easy to learn manner. Techniques that are more
complex are needed for orientation studies that provide more
information but also require more training and equipment.
These two approaches are discussed in more detail below. As
an example, aqueous geochemists typically filter waters in situ,
either to <0.45 or to <0.20 µm. These filter pore sizes are
operationally defined (Hall et al. 1996), as they do not separate
metals that are truly dissolved from those that are bound to fine
colloids; many colloids are as small as <5 nm in size (e.g.
Zanker et al. 2002). However, filtering is not always necessary;
Cameron et al. (1997) showed that for Canadian northern lakes,
pH and suspended sediment contents were sufficiently low that
filtering and acidification were not required, with the caveat that
analyses were to be carried out relatively rapidly after collection.
In that study, some 5000 samples of lake waters were collected
in six weeks and analysed in the field for Zn, allowing for
immediate follow-up. Similarly, Leybourne & Cameron (2008)
showed that for groundwaters collected around the Spence
porphyry Cu deposit in northern Chile, many of the species of
Fig. 2. Eh versus pH plots for groundwaters from the Spence interest (e.g. S, Se, Re, Mo) showed little or no difference
deposit, Chile, showing speciation of Se, Fe, SO4 and As. Note that between filtered (<0.45 µm) and unfiltered aliquots, although
the samples collected by double-valve pump have generally lower Eh metal cations (e.g. Cu, Zn) showed large differences owing to
values than flow-through bailer samples. (A) Se and As speciation, adsorption to Fe, Mn and Al oxyhydroxides (Fig. 3).
with [Se] = 10
6, [As] = 10
6, [SO4] = 10
2, [HCO3] = 10
3, [Fe]
= 10
4. (B) SO4 speciation, with [SO4] = 10
2, [HCO3] = 10
3,
[Cl] = 10
1. Fe speciation with [Fe] = 10
5. For the As diagram, HYDROLOGY
formation of realgar, orpiment, scorodite and claudetite were sup-
pressed, as they are kinetically unlikely to form in these waters. For In order for a geochemical anomaly to develop in groundwater,
Fe speciation, formation of goethite, magnetite and hematite were clearly that water must flow through mineralization; therefore
suppressed. Plots created using The Geochemists Workbench interpretation of groundwater geochemistry in mineral explora-
(Bethke 1994). Figure modified from Leybourne & Cameron (2008). tion is greatly enhanced where there is information regarding
the local and regional hydrology because of the ability to trace
between depths within a well. Some down-hole systems are the anomaly back to its source (Carey et al. 2003; Leybourne &
pressurized prior to deployment in the well, so that the pressure Goodfellow 2003; Gilliss et al. 2004). Interpretation of hydrol-
valves do not open until the pump reaches the depth at which ogy is simplified in regions dominated by porous flow in
the head of the overlying water column exceeds the pressure in unconfined aquifers (e.g. basal gravels in northern Chile that are
the lines (Pauwels et al. 2002). Electric submersible pumps (e.g. in direct contact with porphyry mineralization exposed along
Grundfos) operate continuously via a motor, also allowing the underlying basement unconformity and Athabasca sand-
purging of the standing water in a well, with the added stone U deposits), and significantly more complicated where
advantage that flow rates can be varied. Flow-through cells can flow is fracture-dominated (Leybourne et al. 1998), or where
be connected to compressed gas and electric pump systems, significant density contrasts between different groundwater
which permit real-time monitoring of changes in salinity, pH, flow systems exist (Carey et al. 2003). In some cases, faults/
and Eh. Both bailer and double-valve pump were used in a fractures that originally focused fluid flow during ore formation
study of groundwaters interacting with porphyry copper min- may also serve as conduits to subsequent groundwater flow into
eralization in northern Chile (Fig. 2). Although there was little and out of the deposit (e.g. Spence deposit, Fig. 1) (Cameron &
102 M. I. Leybourne & E. M. Cameron

mobile than Pb), till overlying the deposit has relatively low Zn
and Cu contents (Parkhill & Doiron 2003). Thus, groundwaters
interacting with this type of gossan will likely not have Zn and
Cu anomalies. At Halfmile Lake, Pb contents are elevated in till
down-ice, and Pb isotopic ratios demonstrate that the Pb is
associated with gossan mineralization over primary VMS min-
eralization (Hussein et al. 2003).

MAJOR IONS
A full understanding of groundwater geochemistry requires
determination and interpretation of the major ion composition
(i.e. for most waters this means Ca, Mg, Na, K, Cl, SO4 and
dissolved inorganic carbon, typically HCO3 ). These chemical
parameters permit classification and interpretation of, for
example, water type, mixing, evaporation, water-rock reaction
and recharge. However, some studies have also suggested that
the major ion composition of groundwater can provide direct
vectors to mineralization, in particular owing to the generally
Fig. 3. Bar chart showing average elemental values for filtered/ larger target provided by alteration halos around mineralization
unfiltered samples of groundwater from the Spence deposit, Chile. relative to the size of primary ore mineralization (e.g. Giblin
The error bars represent the standard deviation. From Leybourne &
Cameron (2008). 1994; Taufen 1997; Phipps et al. 2004; de Caritat & Kirste
2005b). The major ion chemistry in a hydrogeochemical survey
should be interpreted and considered in conjunction with the
Leybourne 2005). Furthermore, the Eh (and concentration of trace metal data. However, owing to the complexities of
dissolved oxygen) and pH conditions greatly influence the water-rock reactions, cation exchange reactions and formation
extent and style of ore mineral reaction and the mobility and of secondary minerals, the major element chemistry of water
dispersion of ore-related metals/metalloids. Thus, if the Eh and should be used with great caution as an exploration guide in the
pH conditions are not conducive to mineral dissolution/ absence of other (i.e. deposit-diagnostic) trace element data
replacement and element transport, aqueous geochemical (Table 1). As an example, soil Br and Cl anomalies develop
anomalies will be subdued or may not form. For example, over some deposit types as a result of electrochemical, gaseous
groundwaters recovered close to the buried Tillex Cu–Zn or advective transport (Clark 1999; Hamilton 1999). However,
deposit in Ontario, Canada have low Eh and near neutral pH the extent to which these elements are diagnostic of mineralized
(Gilliss et al. 2004). These pH and redox conditions prevent zones is unclear. For example, work on groundwaters around
oxidation of sulphide to dissolved sulphate, which restricts the Spence deposit in northern Chile suggests that saline
deposit sulphide mineral weathering and therefore prevents the groundwaters migrate to surface along fractures during seismic
development of a significant aqueous geochemical plume. events (Cameron et al. 2002; Cameron & Leybourne 2005).
Although some types of deposit are stratabound or con- Because saline basinal fluids in northern Chile have elevated I,
trolled by faults and fractures, other deposits are channelized. Br, Cl and S, soils over fractures will be anomalous in I and Br
This restriction of mineralization to channels has been noted in regardless of whether or not the waters have interacted with
particular for some types of Cu–Ni–PGE and unconformity U porphyry copper mineralization. In order to explore effectively
deposits (Brand 1999; Pirlo & Giblin 2004). Channelization of using geochemical techniques, the explorationist needs to be
mineralization is an important consideration for aqueous geo- able to identify mineralized rocks and be able to distinguish
chemical exploration because the orientation of the channels them from non-mineralized rocks of the same composition, so
with respect to local hydrology will determine the extent to that attention should be placed on what is geochemically
which mineralization is in contact with groundwaters. For distinct about the mineralization and this is then applied to the
example, at the Kambalda deposit, Ni–Cu–PGE mineralization compositon of groundwater (Tables 1 and 2).
is associated with komatiite flows, which were channelized by Clearly, the major element composition of rocks differs as a
pre-existing topography. Mineralization at Kambalda forms function of deposit type, host rock composition, and degree
linear features over 3 km long, but less than 300 m wide, and and style of alteration, and these have been used to vector to
less than 5 m thick (Brand 1999). mineralization based on alteration mineralogy and geochemistry
Groundwater flow can also be strongly controlled by faults (Large et al. 2001a, b). However, water-rock interaction pro-
that may have originally focused ore-forming fluids. For cesses are such that major elements in groundwaters are
example, at the Spence porphyry-Cu deposit in Chile, deep difficult to interpret with regard to how they indicate proximity
groundwaters are interpreted to have come to surface along to ore mineralization. For example, groundwaters recovered in
long-lived structures beneath mineralization (Cameron et al. and around several kimberlites in northern Ontario displayed
2002). elevated pH (up to 12.4) and low Eh conditions, with concomi-
Another issue is the pre-cover history of the rocks hosting tant loss of Mg from solution and increase in dissolved K
mineralization. Regolith/saprolite development and gossan for- contents (Fig. 4) (Sader et al. 2007). However, these major ion
mation can affect the type of anomaly developed in ground and changes occur regardless of whether the kimberlite is diamond-
surface waters. For example, at the Halfmile Lake deposit, bearing or not; any water–ultramafic rock interaction will likely
Bathurst Mining Camp (BMC), Canada, mineralization crops produce similar changes in major ion chemistry of groundwater.
out as a gossan (Boyle 2003). As a result of differential mobility In contrast, Giblin (2001) suggested that waters interacting
in gossan-forming groundwaters (e.g. Zn and Cu are more with ultramafic rocks will have Mg > Ca, Na, K contents.
 Groundwater in geochemical exploration 103

Table 1. Major components of different deposit types.

Type of deposit Major components Minor components Labile componentsA Relatively immobile componentsB
VMS Fe, S, Cu, Zn, Pb Cd, Hg, Au, As, Sb, Ba, Bi, In Fe, S, Zn, Cu, As, Cd, Hg, Sb Pb, Bi, In, Au, Ag, Ba
Porphyry Cu  Mo Cu, Mo, S Fe, Ag, Au, Se, Re, As Cu, Mo, S, Fe, Se, As, Re Ag, Au
SEDEX Fe, S, Cu, Zn, Pb Ag, Au, Ba, Cd Fe, S, Zn, Cu, Cd Pb, Ba, Au, Ag
Gold (vein) Au, Ag As, Sb, Se, Te, S, Hg S, Se, As, Hg, Te, Sb Au, Ag
Ni–Cu–PGE Ni, Cu, PGE Cr, Co, S Cu, S, PGE Co, Ni, Cr
Kimberlite (diamond) Sr, Nb, Ba, Cr, Ni LILE, HFSE, REE Sr, LILE Ba, HFSE, Nb, Ba, Cr, Ni, REE
Unconformity uranium U Se, Mo, V, Cu, Pb U, Se, Cu, Mo U, Pb, V
A
Under oxidizing and near neutral conditions
B
Under normal conditions; e.g. Ba is immobile in the presence of S as SO4 owing to insolubility of barite.
LILE, large ion lithophile elements; HFSE, high-field strength elements; REE, rare earth elements
Table modified after McMartin McClenaghan (2001) and Leybourne (2007)

Table 2. Summary of key aqueous geochemical pathfinders for different deposit types.

Deposit type Main pathfinders Secondary pathfinders Key methods Key publications Notes
VMS Zn Low pH, Pb, SO4 ICP-MS – metals Cameron (1978); Leybourne Sulfide–Pb sources typically
IC, ICP-AES – S et al. (2003); Leybourne & isotopically distinct; Pb
Goodfellow (2003) isotopes can fingerprint ore
versus non-ore Pb
Porphyry Cu  Distal – Se, Re, Mo, Pb, Zn ICP-MS – metals Cameron & Leybourne S isotopes also useful as a
Mo As IC, ICP-AES – S (2005); Cameron et al. complementary vector
Proximal – Cu (2002); Leybourne &
Cameron (2006), (2008)
SEDEX Zn Ag, Au, Ba, Cd ICP-MS – metals Goodfellow (1983); Jonasson Sulfide–Pb sources typically
IC, ICP-AES – S et al. (1989); Kelley & Taylor isotopically distinct; Pb
TIMS, MC-ICP-MS – (1997); de Caritat et al. isotopes can fingerprint ore
Pb isotopes (2005) versus non-ore Pb
Gold (vein) Au Se, As, Sb Activated carbon Carey et al. (2003); Hall et al.
preconcentration or (1986)
BrCl (see text)
ICP-MS, NAA
Cu–Ni–PGE Ni, Cu, Pd As, Cr, Co, S, PGE ICP-MS – metals Hattori & Cameron (2004) Pd mobility is enhanced under
IC, ICP-AES – S alkaline conditions relative to
other pathfinders
Kimberlite Low Mg, elevated Ni, Co, Cr, high ICP-MS – metals Sader et al. (2003), (2007) Also, formation of Mg
(diamond) K/Mg, pH n 10 Co/Mg and Ni/Mg IC, ICP-AES – S hydroxides (brucite), silicates
(serpentine) and carbonates
(magnesite)
Unconformity Oxidizing – U, radon Se, Mo, As, V, Cu, ICP-MS – metals Deutscher et al. (1980); Radiogenic 207Pb/204Pb and
206
uranium Reducing – Se, Mo Pb IC, ICP-AES – S Dickson & Giblin (2006); Pb/204Pb but
TIMS, MC-ICP-MS – Earle & Drever (1983); non-radiogenic 208Pb/204Pb
Pb isotopes Giblin & Snelling (1983); should prove useful
Langmuir & Chatham (1980)

METAL SATURATION, SPECIATION, AND affects outcomes. Most geochemical modelling codes are
MOBILITY inherently based on equilibrium thermodynamics, so that
kinetic (reaction rate) factors are not considered. Finally,
Several computer codes and thermodynamic databases exist much of the information of relevance to exploration is heavily
that permit calculations of mineral saturation indices and dependent on the input values for pH and the redox state of
metal speciation to be undertaken relatively easily. These the system, so that these data need to be as well constrained as
programs include, for example, PHREEQC (Parkhurst 1995; possible.
Parkhurst & Appelo 1999; Toran & Grandstaff 2002), MINT- Despite these caveats, calculating mineral saturation and
EQA2 (Allison et al. 1991), and The Geochemists Workbench element speciation can be instructive (Giblin 1994; Leybourne
(Bethke 1994). Some geochemical speciation/saturation codes 1998; Gilliss et al. 2004; Pirlo & Giblin 2004), for example,
are only available commercially, whereas some are available as to better understand metal mobility or attenuation, and the
free downloads via the Web. These programs can be very occurrence of secondary mineralization, such as transported
useful in interpreting reaction paths for groundwaters and the gossan (Boyle 2003) or secondary copper mineralization
extent of water-rock reaction. However, the programs are only (Cameron & Leybourne 2005). Sader et al. (2007) showed that
as good as the underlying thermodynamic database and the reaction path modelling of water-kimberlite and water-gneiss
knowledge of the user; some understanding of expected interaction enhances understanding of likely meteoric water
output greatly enhances the utility of these programs. In alteration assemblages, which can also be used as an exploration
addition, quality and completeness of the input data greatly guide.
104 M. I. Leybourne & E. M. Cameron

Fig. 5. (A) Solubility of anglesite (PbSO4 ) in groundwater, Bathurst

Mining Camp, Canada. Waters collected in the mineralized zone are
closest to saturation. (B) Precipitates of Pb and Fe sulphates on filter
paper from waters closest to anglesite saturation (from Leybourne
2001). (C) Anglesite crystals in gossan from the Restigouche deposit.
Fig. 4. (A) pH vs K/Mg; (B) pH vs Co/Mg; and (C) Co/Mg vs
Ni/Mg for groundwaters around kimberlites in northern Ontario The degree to which a water sample will be anomalous
(Sader et al. 2007), VMS deposits in the Bathurst Mining Camp with respect to proximal or distal mineralization depends
(Leybourne & Goodfellow 2003), and porphyry Cu deposits in fundamentally on the solubility of the species that reflect that
northern Chile (Leybourne & Cameron 2006a). style of mineralization under the prevailing pH, Eh, and salinity
conditions (Fig. 5). Ideally, the geochemical explorationist
 Groundwater in geochemical exploration 105

would like any sample media to have anomalies in all species A 15

that define the style of mineralization. In reality, this is rarely, if Groundwaters
ever, the case. However, although waters and sediments will not
10 É H East of deposit
necessarily have anomalies in all the relevant elements, it is Within deposit
important to understand why this is the case and avoid the West of deposit
temptation to include elements that are not diagnostic of the ÑÉ
ÉÑ F North of deposit

! 18OVSMOW (‰)
ÉÑ É
Ñ
style of mineralization (Tables 1 and 2). Furthermore, care 5
ÉÉÑ ÉÑ
Ñ É
ÑF

M
must be taken to interpret aqueous geochemical data with F

or
H
ÑF
ÑÑ

es
consideration of geochemical theory and analytical constraints. É
Ñ

ali
0 Ñ

M
ne
For example, although Ti is elevated in mafic and ultramafic

or
Ñ

ed
igneous rocks compared to high-Si igneous and sedimentary Ñ Ñ Ñ

ilu
lithologies, Ti is unlikely to be of much use in groundwater

te
-5 Ñ
geochemical exploration because: (1) Ti is not particularly
anomalous in mineralized mafic and ultramafic rocks compared H
to non-mineralized equivalents; (2) Ti is scarcely soluble in -10 H H
water, except in very rare cases; and (3) analytically, Ti suffers H HHH
HHHH
from several potential spectral overlaps in an ICP plasma.
Similarly, several studies have suggested that the rare earth -15
elements (REE) may be useful in mineral exploration for
0 1 2 3 4 5 6 7 8
non-REE deposits. In some cases, this may be reasonable. For
example, metal-rich hydrothermal fluids recovered from active ! 34SCDT (‰)
vents at mid-ocean ridge crests show strong positive Eu
anomalies, and these anomalies are commonly observed in B 8 Ñ
VMS deposits. In contrast, strong positive Eu anomalies are H
not common in other geological environments, so that the 7
presence of these Eu anomalies could be used as an explora-
tion guide (Leybourne et al. 2000b; Volesky et al. 2000; Volesky 6 H
2002; Leybourne & Cousens 2005; Leybourne et al. 2006b). HH
H
! 34SCDT (‰)

Low-pH groundwaters proximal to VMS mineralization have 5 H

H
H
also preserved the positive Eu anomalies (Leybourne et al. HH
2000b). Similarly, ground and surface water REE profiles Ñ
4 Ñ
have, in some cases, reflected the host lithology, although this FÑ
H Ñ ÑÑ
ÉÉÑ Ñ
can be complicated by the behaviour of the REE in aqueous Ñ FFÑHÑÉÑÉ
3 Ñ Ñ Ñ É Ñ
systems (Johannesson & Zhou 1999; Johannesson & Hendry É
2000). Diamond-bearing kimberlites commonly have distinc- ÑÉ É
2 Ñ Ñ
tive large ion lithophile element (LILE) and high field É
strength element (HFSE, including the REE) associations
(McMartin & McClenaghan 2001). In this case, aqueous 1 É
anomalies in mobile species (K, Rb, Sr and the REE in some
waters) may assist in the distinction between rock types, but 0
may not be as useful in distinguishing mineralized from 0 100 200 300 400 500 600 700 800
unmineralized rocks. Se (µg/L)
Fig. 6. Stable isotopes of S and O and Se concentrations in
groundwaters from the Spence porphyry Cu deposit, Chile. These
ISOTOPIC APPLICATIONS data are consistent with mixing between deep saline waters and
regional dilute meteoric waters. Increase in Se indicates reaction with
Traditionally, stable and radiogenic isotopes have not been used Cu mineralization. After Leybourne & Cameron (2006a).
in routine mineral exploration programs. Use of isotopes has
been hampered by cost, additional requirements for sampling
and perceived difficulties in data interpretation. Although this Analytical considerations include the ease of sampling,
underutilization of isotopes in exploration is unlikely to change storing, sample processing and cost of determination. For
for large exploration programs, there are applications that example, traditionally, Sr and Pb isotope analyses have been
warrant their incorporation. The light stable isotopes (e.g. O, carried out by thermal ionization mass spectrometry (TIMS).
H, S) can greatly assist in the interpretation of water sources, Although Pb isotopes have received some attention (Gulson
mixing and water sulphide interaction (Kirste et al. 2003; 1986; Gulson et al. 1988; Gulson et al. 1992; Bell & Franklin
Cameron & Leybourne 2005; de Caritat et al. 2005; Leybourne 1993; de Caritat et al. 2005; Leybourne et al. 2009a), measure-
& Cameron 2006a; Leybourne et al. 2006a) (Fig. 6). In addition, ment by TIMS is time-consuming owing to the need for
advances in analytical methods in terms of detection limits, chemical separation of the species of interest. Newer quadru-
sample throughput and lower costs are such that some isotope pole (especially those with dynamic reaction cells, DRC),
systems may become as routine as ICP-MS elemental analyses high resolution (HR) and multi-collector ICP-MS instruments
are today. Isotopes of potential use in mineral exploration (e.g. de Caritat et al. 2005) allow Pb isotopes to be measured
include the traditional stable isotopes (O, H, C, S), radiogenic with accuracy and precision in some cases rivalling that of
isotopes (Pb, Sr), and non-traditional isotope systems, such as TIMS, with significantly less sample preparation and greater
Fe, Se and Cu (Ehrlich et al. 2004; Johnson & Bullen 2004; throughput.
Rouxel et al. 2004; Severmann et al. 2004; Layton-Matthews et al. Stable isotopes are a critical component in many aqueous
2006). geochemical studies because they provide important information
106 M. I. Leybourne & E. M. Cameron

regarding water sources, groundwater flow history, water-rock

reactions, and groundwater evolution. For a good recent
summary of the light stable isotopes in aqueous studies, see
Clark & Fritz (1997). Stable and radiogenic isotopes have not
found much application in exploration, owing to cost and
perceived difficulty in interpretation. Although stable isotope
determinations are not overly expensive per isotope, especially
compared to commercial rates for Pb and Sr by TIMS, adding
a full suite (e.g. O, H, C, S) to an exploration program would
significantly add to the cost of a survey, with additional
resources required for sample collection, preparation, and data
interpretation. However, although the light stable isotopes may
not become routine, they can be a very useful addition to an
orientation or more detailed study. For example, future suc-
cesses in surficial geochemical exploration will rest with the
ability of groundwaters to penetrate to depth, with subsequent
return of pathfinder element-bearing fluids to surface. In that
context, sampling of overland or shallow soil water flow
following storm events will not be as useful as sampling
base-flow conditions for stream water surveys, or deeper
groundwater flow in the case of springs. In a study of stream,
seep, and ground waters from two undisturbed VMS deposits
in the Bathurst Mining Camp (Halfmile Lake and Restigouche
deposits), Leybourne et al. (2006a) showed that surface waters
were similar to shallow groundwaters based on the
18O and

2H compositions, indicating that surface water flow repre-
sented shallow groundwater discharge. In addition, saline
groundwaters at the Restigouche deposit were found to be
isotopically distinct from shallow groundwaters, consistent with
the interpretation that the saline waters were paleowaters
(Leybourne et al. 2006a). These saline waters were also Zn-, Co-
and Ni-rich, but enhanced metal contents were a function of
increased Cl-complexing and, possibly, to deeper mineralization
unrelated to VMS mineralization at the Restigouche deposit
(Leybourne & Goodfellow 2003). Studies of groundwaters
recovered around the undisturbed Spence porphyry copper
deposit, in the hyperarid Atacama Desert of northern Chile
have shown using O and H isotopes that saline waters
interacting with copper mineralization (as shown by elevated
Cu, As, Se, Re, Mo and low
34S; Fig. 6 shows Se and
34S) are
palaeowaters, possibly derived from sedimentary basin brines.
These palaeowaters have mixed with regional groundwater
flow, derived by recharge in the high Andes (Cameron & Fig. 7. Plots of Pb isotopes for BMC waters and host rocks, galena
Leybourne 2005; Leybourne & Cameron 2006a). Understand- (Thorpe et al., 1997, pers. comm.) and local Ordovician granitoids
ing the recharge history of groundwater and whether sampled (Whalen et al. 1998). Modified from (Leybourne et al. 2009a).
waters are modern meteoric or palaeowaters is important in Groundwaters can provide a useful vector to mineralization, even
where Pb concentrations are low.
terms of understanding local hydrology, in particular where
groundwater composition is being used to vector to minerali-
zation.
In contrast to the isotopes of O and H, which are unlikely to suggested that the S and O isotopic composition of dissolved
yield relationships directly linked to mineralization, sulphur sulphate can be used to vector to mineralization and provide a
isotopes can provide vectors to mineralization in addition to broader halo than groundwater base metal (Zn, Pb, Cu)
aiding interpretations of water sources and rock reactions. For contents (de Caritat & Kirste 2005a).
example, Whitford et al. (1997) suggested that groundwater The generally low solubility of Pb in most waters means that
sulphur isotopic compositions provide a vector to gold miner- ground and surface water Pb contents may not be sufficiently
alization in Australia. Similarly, determination of the sulphur elevated proximal to mineralization to warrant consideration, in
isotopic composition of groundwaters at the Spence porphyry particular where pH values are buffered and/or mineralization
copper deposit in Chile shows that groundwater
34S values is more deeply buried. However, because sulphide minerals do
decrease from values typical of regional waters and salars not incorporate U and Th into their structure, Pb derived from
(Pueyo et al. 2001; Rech et al. 2003; Rissmann 2003) to values sulphide minerals will retain the isotopic composition of the
closer to those typical of porphyry copper mineralization as ore-forming fluids, whereas Pb from surrounding U- and
waters flow into and through mineralization (Leybourne & Th-rich host rocks will typically have more radiogenic Pb
Cameron 2006a). This increasing influence on groundwater isotopic compositions. Therefore, Pb isotopes can ‘fingerprint’
sulphur isotopic composition is also evident in Se (Fig. 6), in Pb sources, and indicate subsurface mineralization, even where
addition to Re, Mo, and As. Recent work in the area that hosts aqueous Pb contents are low (Leybourne 1998; de Caritat et al.
the giant Broken Hill SEDEX deposit in Australia has also 2005; Leybourne et al. 2009a) (Fig. 7).
 Groundwater in geochemical exploration 107

Fig. 8. Flowchart illustrating the major

aspects of a groundwater geochemical
exploration program for ‘routine
exploration’.

RECOMMENDATIONS FOR SAMPLING AND knowledge base does not exist, or where techniques require
ANALYTICAL PROTOCOLS OF WATERS IN refining, the research approach should be adopted in an
MINERAL EXPLORATION orientation survey.
Ideally, powder-free plastic gloves (cheap, disposable)
From the above discussion and the case studies discussed should be worn during collection of water samples to ensure
below, there are clearly a large number of parameters that can uniformity in sampling and prevent contamination. Different
be used to aid in the detection of undiscovered mineralization research groups and laboratories have different recommenda-
and there are a number of potential issues and pitfalls that need tions regarding bottles and, for ‘exploration research’ filter
to be avoided. Despite these issues, groundwater geochemistry type and size (Hall 1998; Reimann et al. 1999a, b). The best
can be a powerful tool for the exploration geochemist for the bottles for aqueous geochemical exploration are high-density
reasons outlined in the introduction. In this section, recom- polyethylene bottles (HDPE). These HDPE bottles may be
mendations are made regarding sampling and analysis that pre-cleaned by rinsing with 18.2 M
water or with weak (5%
should aid exploration geochemists in realizing this relative v/v) HNO3, but should not be precharged with concentrated
simplicity. The collection and analysis of groundwaters in HNO3 prior to sampling (Hall 1998). Experiments by Hall
mineral exploration can be broken down into two different (1998) found that different filters produce different levels of
approaches, here referred to as the ‘routine exploration’ (Fig. 8) contamination and retention of colloids. Hall (1998) found that
and ‘exploration research’ (Fig. 9) approaches. Giblin (2001) optimal filters were Acrodisk syringe filters (Supor membrane;
also provides useful guidelines for groundwater sampling in Gelman) and Sterivex syringe filter capsules (Durapore mem-
geochemical exploration. The routine approach is applicable brane; Millipore). In North America, filters of 0.45 µm pore
in areas where some knowledge of the geochemical landscape size are typically used, whereas in Europe, 0.2 µm filters are
and, ideally, the hydrology already exists. In areas where this commonly used. The differences between these two pore sizes
108 M. I. Leybourne & E. M. Cameron

Fig. 9. Flowchart illustrating the major

aspects of a groundwater geochemical
exploration program for ‘exploration
research’.

are nominal (Reimann et al. 1999b), and neither traps all GROUNDWATER GEOCHEMICAL EXPLORATION:
colloids. Some measurements are best made in the field because CASE STUDIES
the parameters are unstable (pH, Eh) and because they can
provide useful guides during exploration. For example, large This section presents some case studies of groundwater explo-
changes in pH may indicate the presence of unbuffered ration related to several deposit types. The case studies pre-
sulphide oxidation (decrease) or proximity to ultramafic igne- sented here present concepts and methodologies that are
ous rocks (increase). Similarly, changes in electrical conductivity applicable to other styles of mineralization. Table 1 indicates
(i.e. salinity) may reflect additions of different, potentially the major geochemical characteristics of the deposit types
deeper, groundwater. For routine exploration, conductivity can discussed below, and Table 2 summarizes the key results of
serve as a proxy for total dissolved solids. Where possible, aqueous studies around those deposit types summarized in
SO42
should be measured by ICP-AES if anions are not Table 2.
determined by ion chromatography. Meters for the determi-
nation of pH and conductivity are relatively robust, inexpen- Porphyry copper deposits
sive, and simple to calibrate. Cameron et al. (1997) showed that Porphyry copper and porphyry molybdenum deposits are
using portable anode stripping voltammetry (ASV) to measure typically associated with intermediate to felsic intrusive igneous
dissolved Zn was useful in guiding the exploration program; rocks. Some of the largest porphyry copper deposits occur in
waters were analysed immediately following each day’s collec- northern Chile and include the giant Chuquicamata, El Salvador
tion. Comparison of these field Zn analyses with subsequent and Escondida deposits. Recent reviews of porphyry deposits
ICP-MS analyses showed effectively no difference for Zn include those by Sillitoe (2000), Richards (2003) and Sinclair
>1 µg/l. (2007). Key geochemical characteristics include anomalous
 Groundwater in geochemical exploration 109

Fig. 10. Location of the Spence

deposit, Atacama Desert, Chile. West–
east cross-section of the southern part
of the Spence deposit. Intermediate
porphyry intrudes andesites. After
Cameron et al. (2007).

accumulation of As, Mo, Se and Re in addition to Cu, and in (Cameron et al. 2002; Cameron & Leybourne 2005; Leybourne
many deposits Au, Pd and Pt (Table 1). There have been a & Cameron 2006a, 2008).
small number of studies of the aqueous geochemistry of waters Groundwaters at the Spence deposit are variable between
associated with porphyry deposits, with the most detailed two end-members; a relatively fresh regional groundwater
around the Spence deposit in the Atacama Desert northern end-member (with TDS typically < 1000 mg/l) and a saline
Chile (Cameron et al. 2002; Cameron & Leybourne 2005; end-member (with TDS values up to and in excess of seawater;
Leybourne & Cameron 2006a). These studies around the Fig. 11) that has clearly extensively interacted with Spence
Spence deposit also highlight the utility of stable isotopes in deposit porphyry copper mineralization (Cameron et al. 2002;
process interpretation. Cameron & Leybourne 2005; Leybourne & Cameron 2006a)
RioChilex discovered the Spence deposit in 1996–1997 by (Fig. 6). Up-gradient of the Spence deposit, groundwaters
grid drilling through piedmont gravels that cover the deposit, typically have low concentrations of porphyry copper-related
which displays a typical supergene-enriched sequence (Fig. 10). species, such as Cu, Mo, Se, Re and As, and sulphate has
Resources recoverable by open-pit mining comprise 50 Mt of isotopic compositions typical of salars and non-mineralized
oxide ore with 1.4% Cu, 200 Mt of enriched sulphide ore with groundwaters in the Atacama Desert (Fig. 6). Oxygen and
1.3% Cu, and 150 Mt of primary sulphide ore with 0.6% Cu. hydrogen isotopes of these up-flow waters are consistent with
Ore is associated with three quartz-feldspar porphyry bodies relatively recent recharge at elevations around 4000 m in the
intruded along a NE axis into andesitic volcanic rocks. Intru- high Andes (Fig. 12). Within the deposit, As, Se, Re and Mo
sions are cut by tourmaline-quartz-sulphide hydrothermal contents increase and remain elevated for at least 2 km down-
breccias. The irregular deposit surface is covered by 30–180 m gradient from the deposit, owing to the fact that under the pH
of gravels. Groundwaters were sampled in and around the and Eh conditions present (near-neutral, moderately oxidizing),
Spence deposit in exploration and definition wells in 1999 these elements form oxyanions, which are less readily adsorbed
and 2000, using a combination of flow-through bailer and to oxyhydroxide or clay mineral surfaces than the metal cations
down-hole pump, as described above. This groundwater study (Cameron et al. 2002; Cameron & Leybourne 2005; Leybourne
was undertaken to enhance understanding of geochemical & Cameron 2006a). The water table shallows in a salar (salt
anomalies in the soils overlying the Spence deposit, which are pan) 5 km down-gradient from the deposit, and salar sediments
similar to surficial anomalies elsewhere in northern Chile are strongly enriched in these porphyry indicator elements
(Cameron et al. 2002). As such, this groundwater study can be (Leybourne & Cameron 2006b). In contrast, Cu is clearly
considered an example of a ‘research exploration’ survey (Fig. anomalous in groundwaters in and around mineralization,
9). Thus, samples were collected and analyzed for all major and but Cu contents decrease rapidly down-gradient, owing to
trace metals, anions and trace oxyanions, stable isotopes (O, H, adsorption of this cation to mineral surfaces. Base metal
C, S), and both filtered (< 0.45 µm) and unfiltered aliquots contents in Spence deposit groundwaters are limited, as in other
110 M. I. Leybourne & E. M. Cameron

Fig. 11. Outline of the Spence deposit,

including location of major porphyries,
groundwater sample locations, soil
sample traverse and generalized
groundwater flow. After Leybourne &
Cameron (2006a). Salinities indicated
next to well location.

environments, by the generally near-neutral pH conditions. 1981; Taufen 1997). Waters around Casa Grande appear to
Thus, at the Spence deposit, mineralization can be detected for have very similar characteristics in terms of metal association as
long distances down-flow of mineralization (at least several km) waters from the Spence deposit. In summary, porphyry copper
by anomalous contents of As, Se, Re and Mo, with local deposits appear to be excellent candidates for groundwater
definition of the target based on coincident anomalies in Cu geochemical exploration; these deposits commonly contain
abundance and S isotopes. Cameron and Leybourne also anomalous concentrations of species that tend to be relatively
studied soils and waters around a large Cu (soil MMI) geo- mobile in neutral to slightly alkaline, suboxic solutions (e.g. As,
chemical anomaly located in the Pampa del Tamarugal region Se, Re and Mo). As shown at the Spence deposit in Chile, these
of northern Chile (Leybourne & Cameron 2006b). Although species form oxyanions and therefore migrate down-flow for
soils here show elevated Cu, analyses of the waters and soils significant distances, providing a broad exploration target.
suggest that, different to Spence, the Cu in the Tamarugal Anomalies in these oxyanions can be refined using S/Se ratios
anomalous soils is the result of long periods of accumulation and S isotopes, to aid distinction from far-travelled saline fluids.
owing to evaporation of Cu-bearing waters from a distal source. Copper migration is typically restricted owing to adsorption to
S/Se ratios of these waters are similar to the regional waters hydroxide surfaces, which also likely explain the lack of a
up-flow of the Spence deposit, and significantly elevated copper anomaly at Casa Grande.
compared to waters that have clearly interacted with Spence
mineralization.
There has been relatively little research on groundwater Volcanogenic massive sulphide deposits
geochemistry around porphyry copper deposits elsewhere; Volcanogenic massive sulphide (VMS) deposits are variable in
groundwaters were recovered from around the Casa Grande terms of host rock and sulphide compositions, but are charac-
porphyry copper-molybdenum system, Arizona (Ficklin et al. teristically dominated by metal sulphides, in particular by pyrite,
1981). Groundwaters at Casa Grande appear to be anomalous pyrrhotite, sphalerite, galena and chalcopyrite. Details of the
in Mo, As, K and SO4 proximal to mineralization, although As geology and geochemistry of VMS mineralization are presented
anomalies are displaced from Mo and SO4 anomalies, compli- elsewhere (Barrie & Hannington 1999; Galley et al. 2007).
cating interpretation in the absence of groundwater flow data. These deposits are dominated by Fe, Zn, Pb, and Cu, with
Groundwater Cu contents are not anomalous (Ficklin et al. generally minor Cd, Hg, Au, As, Sb, Ba, Bi and In (Table 1).
 Groundwater in geochemical exploration 111

in suspended sediments from the Restigouche deposit)

(Leybourne 2001) (Fig. 5).
Based on these studies in the BMC and on other studies
around VMS mineralization, it appears that the best and most
widely dispersed indicator of this type of mineralization in both
ground and surface waters is Zn. In many cases, there are
coincident anomalous concentrations of other ore-related
species, in particular Pb, Cu, As, Mo, Sb and sulphate. How-
ever, some of these species are relatively immobile under
near-neutral conditions, so anomalies in these species are not
always present. Sulphate, although commonly anomalous
down-flow of VMS deposits, is not in itself a reliable indicator
of mineralization, given that it is a major anion in many waters.
Similar results were found for groundwaters associated with the
Myra Falls deposit, Vancouver Island, Canada (Phipps et al.
2004). The mountainous terrain and well-developed fracture
network have resulted in rapid recharge so that groundwaters
interacting with ore have generally low salinity and are oxygen-
ated, which should promote sulphide mineral oxidation. Zinc is
the best indicator of massive sulphide mineralization, with a
threshold value of 20 µg/l (Phipps et al. 2004). Zinc anomalies
at Myra Falls are commonly associated with more subtle
Fig. 12. Stable isotope of oxygen and hydrogen for Spence deposit anomalies in a suite of pathfinder elements associated with
groundwaters. GMWL = global meteoric water line of Craig (1961), primary VMS mineralization and hydrothermal alteration,
and local MWL after Aravena et al. (1999). Note that the most saline including Cu, Pb, Cd, Mn, As, Sb, Ba and U.
waters at the Spence deposit are similar to basinal brines; fields
based on data from Hoefs (2004) and Taylor (1974). Figure after
Leybourne & Cameron (2006a).
Other deposit types
The main impediment to Au exploration using hydrogeochem-
VMS deposits have been studied in variable detail in terms istry is the typically low solubility of Au in low temperature
of surface and groundwater geochemical exploration. Ground waters (Vlassopoulos & Wood 1990). Despite this, several
and surface waters were investigated for their efficacy in studies in various environments have found that the best
exploring for more deeply buried mineralization in the mature predictor of Au mineralization is aqueous Au (Cidu et al. 1995;
Bathurst Mining Camp, New Brunswick, Canada (Leybourne Grimes et al. 1995; Gray 2001; Carey et al. 2003). These studies
et al. 1998, 1999; Leybourne 2001; Leybourne et al. 2002; indicate that aqueous Au contents provide the most consistent
Leybourne et al. 2003; Leybourne & Goodfellow 2003; vectors to Au mineralization, although confidence in Au
Leybourne & Cousens 2005; Leybourne et al. 2006a; Leybourne anomalies may be increased with coincident As and/or Sb
& Johannesson 2008; Leybourne et al. 2009a, b). These studies anomalies. Water samples can be difficult to preserve for Au
involved investigation of surface water geochemistry over a and Ag analyses, as these species tend to adsorb strongly to the
moderately large scale (c. 400 km2 ) with comparison to coinci- walls of HDPE bottles acidified with nitric acid (Leybourne
dent stream sediments, with samples taken about every 500 m et al. 2000a). A separate aliquot can be acidified to 1% with
along all streams in the area. At the Restigouche and Halfmile BrCl; this keeps the Au, Ag and Hg in solution, which can then
Lake deposits, groundwaters were collected from diamond be analysed by ICP-MS to <2 ng/l. Alternatively, Au (and Ag)
drill-holes within and distal from the deposits (Fig. 13). can be preconcentrated in the field using activated carbon, with
Groundwaters proximal to the Restigouche deposit have addition of HCl to adjust the pH to c. 2 in order to ensure
elevated As, SO4 and metal (Zn, Pb, Cu, Mo, Sb, Cd) contents complete adsorption of Au onto the charcoal (Hall et al. 1986).
that are significantly greater than local surface waters and After digestion of the activated carbon to release the adsorbed
groundwaters distal from mineralization (Leybourne & Au, analysis is performed by ICP-MS.
Goodfellow 2003) (e.g. Zn, Pb; Figs 14–17). Metals will enter There has been little research into aqueous geochemical
solutions because of oxidation of massive sulphide minerals in exploration for magmatic Ni–Cu–PGE mineralization. There is
the deposit. In most cases, oxidation of sulphide minerals good potential for water chemistry in this regard, in particular
occurs where O2-bearing waters interact with mineralization, for those deposits that are sulphide-poor (especially in pyrrho-
although oxidation of sulphides can occur in the absence of tite) and therefore do not present a good geophysical target
dissolved oxygen (Hamilton 1998). This control by dissolved (Hattori & Cameron 2004). The PGE are variably mobile in
O2 is shown in Figure 17, which is a pseudo cross-section of low-temperature aqueous systems with Pd the most mobile
groundwater geochemical data (Pb, Zn) at the Restigouche (Wood 2002; Hattori & Cameron 2004). Palladium is mobile
deposit. Where wells intersect mineralization at greater depth, in surface waters around Lac des Iles PGE mineralization
metal contents are low in the waters recovered from those as hydroxide complexes (e.g. Pd(OH)20, Pd(OH)3
and
depths indicating that O2-bearing waters do not penetrate to Pd(OH)42
); the low SO4 and Cl contents of these waters
these depths. As with the surface waters, groundwaters around precludes sulphate and chloride forming significant Pd-
undisturbed VMS deposits in the BMC typically have near- complexes (Hattori & Cameron 2004). As discussed above for
neutral pH values, which generally restricts the mobility of porphyry copper systems, where metals and metalloids are
base metal cations owing to adsorption, primarily to Fe- complexed as neutral or negatively charged species, adsorption
oxyhydroxide minerals, and in rare cases due to precipitation to clays and Fe- and Mn-oxides and oxyhydroxides is inhibited
of metal-sulphate minerals (e.g. anglesite, PbSO4, is present at neutral to slightly alkaline pH conditions (owing to surface
112 M. I. Leybourne & E. M. Cameron

Fig. 13. Geology of the (a) Halfmile

Lake deposit and (b) Restigouche
deposit, Bathurst Mining Camp,
Canada, showing the location of the
boreholes sampled. Inset map shows
the location of the study area in
northern New Brunswick. After
Leybourne et al. (2006a).

charge considerations), consistent with the potential for signifi- result in relatively elevated pH values as a result of hydration of
cant aqueous dispersion of Pd (Hattori & Cameron 2004). In mafic Mg- and Fe-bearing minerals (Fritz et al. 1992; Stefánsson
contrast to Lac des Iles, many Ni–Cu–PGE deposits have et al. 2001; Sader et al. 2007). In the case of PGE mineralization
abundant Ni- and Fe-sulphides. In this case, Pd (and other in the absence of significant Fe-sulphides, pH values are likely
PGE) mobility in solution should be enhanced, in particular to remain elevated (i.e. up to pH 10 or so), and Pd may prove
where pyrrhotite is abundant, as dissolution of Fe-sulphides to be the optimal pathfinder in solution, along with S and As,
promotes more acidic conditions, which in turn will result in owing to their likely occurrence as oxyanions (Cameron &
greater leaching of the PGE (Hattori & Cameron 2004). Given Hattori 2003). Where significant Fe-sulphides are present, as at
the mafic-ultramafic rock association of magmatic Ni–Cu–PGE most Cu–Ni–PGE deposits, pH values are likely to be anoma-
deposits, typically water-rock reactions can be expected to lously low compared to waters more distal from mineralization,
 Groundwater in geochemical exploration 113

Fig. 14. Plot of pH vs total metal load

(Ficklin diagram) for groundwaters in
the Bathurst Mining Camp. The
majority of the groundwaters from the
undisturbed Halfmile Lake and
Restigouche deposits are near-neutral
with low to moderate metal loads. Also
shown are waters from the Red Dog
and Lik deposits (Kelley & Taylor
1997) and acid-mine drainage (Plumlee
et al. 1999).

Fig. 15. Sulphate versus metals for

groundwaters from the Bathurst Mining
Camp, Canada. Lead concentrations are
highest in Restigouche deposit
groundwaters closest to where the
orebody crops out (see Figs 16 and 17).

and the lower pH values will likely promote solution of Ni and There has been almost no research investigating the use of
Cu in addition to the PGE, so that multi-element anomalies groundwater chemistry in diamond/kimberlite exploration,
should be more significant (Table 2). although there have been a number of studies that have
114 M. I. Leybourne & E. M. Cameron

Fig. 16. Proportional circle maps of Zn

and Pb in groundwaters within and
around the Restigouche deposit,
Bathurst Mining Camp, Canada. After
Leybourne (2007).

investigated the geochemical consequences of water-ultramafic waters are uncommonly Mg-rich and in equilibrium with
rock interaction (Fritz et al. 1992; Palanallen & Seyfried 2004; phengite (Yanase et al. 1995a; Dickson & Giblin 2006). One
Palandri & Reed 2004; Pokrovsky & Schott 2004). Sader et al. avenue that requires more research is using Pb and/or U-series
(2007) have shown that where groundwater interacts with isotopes (Earle & Drever 1983; Toulhoat & Beaucaire 1991;
kimberlite (in one instance, diamond-bearing), hydration of Yanase et al. 1995b).
olivine and clinopyroxene produces elevated groundwater pH,
up to in excess of pH 12, and results in loss of Mg and increase FUTURE DIRECTIONS
in K (Fig. 4). Further work is needed to better define trace
element and/or isotopic anomalies in groundwaters that might The role of bacteria in controlling redox reactions (e.g. sulphide
provide vectors towards diamond-bearing ultramafic bodies, oxidation), silicate hydrolysis, metal mobility and metal attenu-
although preliminary data (Sader & Leybourne, 2004, pers. ation is not well understood. In recent years there has been
comm.) suggests that some kimberlite waters are enriched in increasing appreciation that many reactions that were pre-
Co and Ni and have elevated Co/Mg and Ni/Mg ratios (Fig. 4). viously considered inorganic, are now recognized as at least
Exploration for uranium using groundwater geochemistry being facilitated, if not controlled, by biological processes.
was investigated in the 1970s and early 1980s (e.g. Langmuir Recent studies have suggested that alteration of mid-ocean
1978; Dyck 1979; Langmuir & Chatham 1980; Runnells & ridge basalt glass is bacterially mediated (Torsvik et al. 1998).
Lindberg 1981; Earle & Drever 1983; Giblin & Snelling 1983) For example, Reith et al. (2005) suggest that there is an
and has received renewed interest in the last few years (Singh association between Au mineralization and Bacillus cereus in soils
et al. 2002; Pirlo & Giblin 2004). Despite the fact that U is in Australia. Supergene sphalerite mineralization was recently
immobile under reducing conditions, in terms of elemental identified at the Mike deposit in Nevada (Bawden et al. 2003).
abundances, U appears to be the best indicator of unconformity- Here, sulphur isotopes of framboidal sphalerite are very nega-
type U mineralization. Clearly, exploration success will be tive (as low as
70 ‰ 34S), illustrating the control of
enhanced if there are coincident anomalies in other species that sulphur-reducing bacteria on this unusual mineralization
are typically enriched in U deposits, such as Se, V, Mo, and Cu (>400 000 tonnes of Zn).
(Table 2). Molybdenum and Se may especially be useful in Greater emphasis needs to be placed on distinguishing real
sulphide-poor reducing waters. There is some indication that from false geochemical anomalies, in particular development of
 Groundwater in geochemical exploration 115

2. Determine the type of sampling apparatus to be used:

routine exploration may only require a flow-through bailer,
whereas, more detailed studies will likely need some form of
down-hole pump or straddle-packer system, particularly in
fractured-rock media to provide optimal understanding of
the redox state of the groundwater system.
3. If resources permit, isotopic analyses can prove invaluable.
In particular, Pb isotopes can be used to fingerprint metal
sources and S isotopes (in both SO42
and HS
, as
appropriate) can provide vectors to mineralization. Analyti-
cal methods are rapidly advancing in terms of technology
and more rapid and cost-effective techniques. It is likely that
isotopic analyses, such as those for Pb and S, will become
cheap and routine in a small number of years.

We thank W. Goodfellow, J. Peter, D. Boyle, D. Layton-Matthews,

B. McClenaghan, S. Hamilton, I. Jonasson, G. Hall, S. McCutcheon,
D. Lentz and T. Al for various discussions over the last few years on
aqueous geochemistry, mineral exploration and ore deposits.
Research in Chile, the Bathurst Mining Camp, and Kirkland Lake
was carried out with funding from CAMIRO (Deep Penetrating
Fig. 17. Proportional circle plots of Zn and Pb concentrations in Geochemistry Phase II), BHP Ltd (now BHP-Billiton Ltd) and the
groundwaters at the Restigouche deposit, Bathurst Mining Camp, GSC (TGI, Targeted Geoscience Initiative and EXTECH-II,
Canada. Modified from Leybourne & Goodfellow (2003). EXploration TECHnology). Tony Christie (GNS Science) and
Patrice de Caritat (Geoscience Australia) are thanked for reviewing
the manuscript. We thank G. Hall, R. Jackson, G. Closs and S.
better models that relate redox and pH conditions of ground Romberger for reviewing the paper.
and surface waters with the element associations that can be
expected given different styles of mineralization. With the REFERENCES
increasing use of MC-ICP-MS instruments, in particular in
the area of non-traditional isotopes, there is real potential in the A, J.D., B, D.S. & N-G, K.J. 1991. MINTEQA2/
PRODEFA2, a geochemical assessment model for environmental systems. EPA/
future to use isotopes to fingerprint ore from non-ore signa- 600/3-91/021, U.S. Environmental Protection Agency.
tures in groundwaters. For example, Zn, Fe, Cu and Se isotopes A, R., S, O., P, H., P, A., F, H. &
are relatively new systems being investigated in a variety of G, A. 1999. Isotopic composition and origin of the precipitation in
systems (Layton-Matthews et al. 2006; de Jong et al. 2007; John northern Chile. Applied Geochemistry, 14, 411–422.
et al. 2008; Petit et al. 2008). In the future these isotopic systems B, C.T. & H, M.D. 1999. Volcanic Associated Massive
will become understood sufficiently that they may become Sulfide Deposits: Processes and Examples in Modern and Ancient Settings.
Reviews in Economic Geology, 8, 1–408.
useful in mineral exploration. In addition, more research is B, T.M., E, M.T. et al. 2003. Extreme 34S depletions in ZnS at
needed on analytical methods that lower the cost, sample the Mike gold deposit, Carlin Trend, Nevada: evidence for bacteriogenic
preparation steps, and speed of isotope analyses. supergene sphalerite. Geology, 31, 913–916.
There is a continued need for better integration of real B, K. & F, J.M. 1993. Application of lead isotopes to mineral
aqueous geochemical data and thermodynamic modelling pro- exploration in glaciated areas. Geology, 21, 1143–1146.
grams, in particular the incorporation of kinetic thermodynam- B, C.M. 1994. The Geochemist’s Workbench. University of Illinois, USA.
B, D.J., G, D.C. & R, K.G. 1994. Saline groundwaters
ics and adsorption to Mn and clay minerals. In addition, the and brines in the Canadian Shield: geochemical and isotopic evidence for
thermodynamic databases for saline waters (greater than sea- a residual evaporite brine component. Geochimica et Cosmochimica Acta, 58,
water) are limited with respect to most of the trace metals of 1483–1498.
interest to exploration. Accurate hydrological models are B, D.R. 1988. Application of groundwater geochemistry in mineral
needed in a study area in order to connect surface aqueous exploration--L’utilisation de la geochimie des eaux souterraines dans le
geochemistry to groundwater flow and metal sources. domaine de l’exploration minerale. In: B, M.J. (ed.) Current activities
forum 1988, program with abstracts--Forum des travaux en cours 1988, programme et
One of the great advantages of collecting groundwater resumes. Geological Survey of Canada, Ottawa.
samples during exploration is that the data are also critical in B, D.R. 2003. Preglacial weathering of massive sulfide deposits in the
establishing local and regional geochemical baselines should the Bathurst Mining Camp: Economic geology, geochemistry, and exploration
program be successful in finding mineralization and the deci- applications. In: G, W.D., MC, S.R. & P, J.M.
sion made to initiate mining. (eds) Massive Sulphide Deposits of the Bathurst Mining Camp, New Brunswick, and
Northern Maine. Economic Geology Monograph, 11, 689–721. Society of
Economic Geologists, Littleton, CO, USA.
CONCLUSIONS B, N.W. 1999. Element ratios in nickel sulphide exploration: vectoring
towards ore environments. Journal of Geochemical Exploration, 67, 145–165.
The following recommendations can be made with respect to C, E.M. 1977. Geochemical dispersion in lake waters and sediments
the use of groundwater geochemistry in mineral exploration: from massive sulphide mineralization, Agricola Lake, Northwest Terri-
tories. Journal of Geochemical Exploration, 7, 327–348.
1. In new areas or new styles of mineralization, orientation C, E.M. 1978. Hydrogeochemical methods for base metal exploration
studies are important, in particular with a view to determi- in the northern Canadian Shield. Journal of Geochemical Exploration, 10,
ning the elemental associations that are diagnostic of that 219–243.
C, E.M. & H, K.H. 2003. Mobility of palladium in the surface
style of mineralization. In these kinds of studies, emphasis environment: data from a regional lake sediment survey in northwestern
should be placed on characterization of groundwaters (via Ontario. Geochemistry: Exploration, Environment, Analysis, 3, 299–311.
wells, springs), surface waters and partial leaches of stream C, E.M. & L, M.I. 2005. Relationship between ground-
(and groundwater suspended) sediment. water chemistry and soil geochemical anomalies at the Spence copper
116 M. I. Leybourne & E. M. Cameron

porphyry deposit, Chile. Geochemistry: Exploration, Environment, Analysis, 5, G, A., H, M. & J, I. 2007. Volcanogenic massive
135–145. sulfide deposits. In: G, W.D. (ed.) Mineral Resources of Canada: A
C, E.M., H, S.M., L, M.I., H, G.E.M. & Synthesis of Major Deposit-types, District Metallogeny, the Evolution of Geological
MC, B. 2004. Finding deeply-buried deposits using geochem- Provinces, and Exploration Methods. Mineral Deposits Division, Geological
istry. Geochemistry: Exploration, Environment, Analysis, 4, 7–32. Association of Canada, Special Publications 5, 141–161.
C, E.M., L, M.I. & K, D.L. 2002. Exploring for G, A. 1994. Natural waters as sample media in drainage geochemistry.
deeply covered mineral deposits: Formation of geochemical anomalies in In: H, M. & P, J.A. (eds) Handbook of Exploration Geochemistry. Vol.
northern Chile by earthquake-induced surface flooding of mineralized 6, Elsevier Science, Amsterdam, 269–303.
groundwaters. Geology, 30, 1007–1010. G, A. 2001. Groundwaters: Geochemical Pathfinders to Concealed Ore Deposits.
C, E.M., L, M.I. & P, C. 2007. Atacamite in the CSIRO Exploration and Mining, North Ryde, NSW, Australia.
oxide zone of copper deposits in northern Chile: involvement of deep G, A.M. & S, A.A. 1983. Application of hydrogeochemistry to
formation waters. Mineralium Deposita, 42, 205–218. uranium exploration in the Pine Creek geosyncline, Northern Territory,
C, E.M., M, C. & H, G.E.M. 1997. Water: A medium for Australia. Journal of Geochemical Exploration, 19, 33–55.
exploration in northern terrains. Explore, 96, 1–6. G, M.L., A, T.A., B, D.W., H, G.E.M. & ML, B. 2004.
C, M.L., MP, D.C. & T, P.M. 2003. Groundwater flow in Geochemical dispersion in groundwater from a weathered Cu–Zn deposit
playa lake environments: impact on gold and pathfinder element distribu- in glaciated terrain. Geochemistry: Exploration, Environment, Analysis, 4, 291–
tions in groundwaters surrounding mesothermal gold deposits, St. Ives 305.
area, Eastern Goldfields, Western Australia. Geochemistry: Exploration, G, I.S. 1998. Vertical migration of elements from mineral deposits.
Environment, Analysis, 3, 57–71. Journal of Geochemical Exploration, 61, 191–202.
C, R., F, L., S, P., E, W.M., D, L.V.T. & G, G, W.D., MC, S.R. & P, J.M. 2003. Massive
R. 1995. Hydrogeochemical exploration for gold in the Osilo area, Sardinia, Sulphide Deposits of the Bathurst Mining Camp, New Brunswick, and
Italy. Applied Geochemistry, 10, 517–529. Northern Maine. Economic Geology Monograph, 11. Society of Econ-
C, I.D. & F, P. 1997. Environmental Isotopes in Hydrogeology. Lewis omic Geologists, Littleton, CO, USA.
Publishers, Boca Raton, NY, USA. G, G.J.S. 1976. Detection of deeply buried and blind sulphide deposits
C, J.R. 1999. Concepts and models for interpretation of enzyme leach by measurement of H+ and conductivity of closely spaced surface soil
data for mineral and petroleum exploration In: Geoanalysis: With Emphasis on samples. Journal of Geochemical Exploration, 6, 359–382.
Selective Extractions. Association of Exploration Geochemistry Short Course, 19th G, D.J. 2001. Hydrogeochemistry in the Yilgarn Craton. Geochemistry:
International Geochemical Exploration symposium, April 10, 1999. Association of Exploration, Environment, Analysis, 1, 253–264.
Exploration Geochemists, Vancouver, BC, Canada. G, D.J., F, W.H., M, A.L. & MH, J.B. 1995. Anomalous
C, H. 1961. Isotopic variations in meteoric waters. Science, 133, 1702– gold, antimony, arsenic, and tungsten in ground water and alluvium around
1703. disseminated gold deposits along the Getchell Trend, Humboldt County,
Nevada. Journal of Geochemical Exploration, 52, 351–371.
 C, P. & K, D. 2005a. Groundwater chemical and isotopic composition:
G, B.L. 1986. Lead Isotopes in Mineral Exploration. Elsevier,
a mineral exploration tool for areas of regolith cover – lessons from the Curnamona
Amsterdam.
Province. CRC LEME-GSWA Mineral Exploration Seminar, 81–82.
G, B.L., C, S.E., M, K.J. & M, A.L. 1992. Lead isotopes
 C, P. & K, D. 2005b. Hydrogeochemistry applied to mineral
in iron and manganese oxide coatings and their use as an exploration guide
exploration under cover in the Curnamona Province. MESA Journal, 37,
for concealed mineralization. Applied Geochemistry, 7, 495–511.
13–17.
G, B.L., L, R.R. & P, P.M. 1988. Gold exploration using
 C, P., K, D., C, G. & MC, M. 2005. Groundwater
lead isotopes at Tennant Creek, Australia. Applied Geochemistry, 3, 243–254.
in the Broken Hill region, Australia: recognising interaction with bedrock
H, G.E.M. 1998. Cost-effective protocols for the collection, filtration and preservation
and mineralisation using S, Sr and Pb isotopes. Applied Geochemistry, 20,
of surface waters for detection of metals and metalloids at ppb (µg l-1) and ppt (ng l-1)
767–787.
levels. Aquatic Effects Technology Evaluation Program (Task Force on
 J, J., S, V. et al. 2007. Precise measurement of Fe isotopes in Water Quality Issues), AETE Project 3.1.3, Canmet, Ottawa, http://
marine samples by multi-collector inductively coupled plasma mass spec- www.nrcan-rncan.gc.ca/mms/canmet-mtb/mmsl-lmsm/enviro/reports/
trometry (MC-ICP-MS). Analytica Chimica Acta, 589, 105–119. 313.pdf.
D, R.L., M, A.W. & G, A.M. 1980. Ground water geo- H, G.E.M., B-C, G.F. et al. 1996. The effect of using
chemistry in the vicinity of the Jabiluka deposits. In: F, J. & different 0.45 µm filter membranes on ‘dissolved’ element concentrations
G, A.B. (eds) Uranium in the Pine Creek Geosyncline; proceedings of the in natural waters. Applied Geochemistry, 11, 243–249.
International uranium symposium on the Pine Creek Geosyncline. International H, G.E.M., V, J.E. & B, S.B. 1986. Field and laboratory
Atomic Energy Agency, Vienna, Austria, Sti/pub/555, 477–486. procedures for determing gold in natural waters; relative merits of
D, B. & G, A. 2006. New horizons for groundwater geochemistry in preconcentration with activated charcoal. Journal of Geochemical Exploration,
uranium exploration. AESC2006. 26, 191–202.
D, W. 1979. Application of hydrogeochemistry to the search for uranium. H, S.M. 1998. Electrochemical mass-transport in overburden: a new
In: H, P.J. (ed.) Geophysics and Geochemistry in the Search for Metallic Ores. model to account for the formation of selective leach geochemical
Proceedings of Exploration 77, an International Symposium. Ottawa, Canada, anomalies in glacial terrain. Journal of Geochemical Exploration, 63, 155–172.
October 1977. Geological Survey of Canada Economic Geology Report H, S.M. 1999. Transport mechanisms responsible for selective leach
31, 489–510. anomalies. In: Association of Exploration Geochemists (ed.) Geoanalysis
E, S.A.M. & D, G.L. 1983. Hydrogeochemical exploration for with Emphasis on selective Extractions. Association of Exploration Geochem-
uranium within the Athabasca Basin, northern Saskatchewan. Journal of ists, Short Course. Vancouver, BC, Canada.
Geochemical Exploration, 19, 57–73. H, K.H. & C, E.M. 2004. Using the high mobility of palladium
E, S., B, I., H, L., R, D., O, A. & in surface media in exploration for platinum group element deposits:
M, A. 2004. Experimental study of the copper isotope fraction- evidence from the Lac des Iles region, northwestern Ontario. Economic
ation between aqueous Cu(II) and covellite, CuS. Chemical Geology, 209, Geology, 99, 157–171.
259–269. H, J. 2004. Stable Isotope Geochemistry. Springer-Verlag, Berlin.
E, R.G., B, P.H. et al. 2007. Environmental geochemistry at Red H, A.A., L, C. & B, K. 2003. Application of Pb isotopes to
Mountain, an unmined volcanogenic massive sulphide deposit in the mineral exploration in the Halfmile Lake area, Bathurst, New Brunswick.
Bonnifield district, Alaska Range, east-central Alaska. Geochemistry: Explora- In: G, W.D., MC, S.R. & P, J.M. (eds) Massive
tion, Environment, Analysis, 7, 207–223. Sulphide Deposits of the Bathurst Mining Camp, New Brunswick, and Northern
F, W.H., P, D.J., W, E.P. & N, G.A. 1981. Analyti- Maine. Economic Geology Monograph, 11, 679–688. Society of Economic
cal results for seventy eight ground-water samples from the Casa Grande vicinity, Geologists, Littleton, CO, USA.
Arizona, U.S. Geological Survey Open File Report 81-960, USGS. J, K.H. & H, M.J. 2000. Rare earth element geochemistry
F, P., C, I.D., F, J.-C., W, M.J. & F, E. 1992. of groundwaters from a thick till and clay-rich aquitard sequence,
Deuterium and 13C evidence for low temperature production of hydrogen Saskatchewan, Canada. Geochimica et Cosmochimica Acta, 64, 1493–1509.
and methane in a highly alkaline groundwater environment on Oman. In: J, K.H. & Z, X. 1999. Origin of middle rare earth element
K, Y. & M, A.S. (eds) Proceedings of the 7th International Symposium enrichments in acid waters of a Canadian High Arctic lake. Geochimica et
on Water–Rock Interaction. Balkema, Rotterdam, 793–796. Cosmochimica Acta, 63, 153–165.
 Groundwater in geochemical exploration 117

J, S.G., R, O.J., C, P.R., E, A.M. & B, E.A. L, M.I. & J, K.H. 2008. Rare earth elements (REE)
2008. Zinc stable isotopes in seafloor hydrothermal vent fluids and and yttrium in stream waters, stream sediments, and Fe-Mn oxyhydroxides:
chimneys. Earth and Planetary Science Letters, 269, 17–28. fractionation, speciation, and controls over REE + Y patterns in the
J, T.M. & B, T.D. 2004. Mass-dependent fractionation of surface environment. Geochimica et Cosmochimica Acta, 72, 5962–5983.
selenium and chromium isotopes in low-temperature environments. In: L, M.I., B, D.R. & G, W.D. 2003. Interpretation of
J, C.M., B, B.L. & A̀, F. (eds) Geochemistry of Non- stream water and sediment geochemistry in the Bathurst Mining Camp,
Traditional Stable Isotopes. Mineralogical Society of America, Geochemical New Brunswick: Applications to mineral exploration. In: G,
Society, Reviews in Mineralogy and Geochemistry, 55, 289–317. W.D., MC, S.R. & P, J.M. (eds) Massive Sulphide Deposits of the
J, I.R., G, W.D., W, D.A. & J, L.E. 1989. Bathurst Mining Camp, New Brunswick, and Northern Maine. Society of
Secondary zinc mineralization in post-glacial sediments and talus at Economic Geologists, Inc., Economic Geology Monograph, 11, 741–762.
Howard’s Pass, Yukon, Canada – mineralogy, geochemistry and genesis. In: Society of Economic Geologists, Littleton, CO, USA.
Weathering; its Products and Deposits, Volume II, Products–Deposits– L, M.I., G, W.D. & B, D.R. 1998. Hydrogeochemi-
Geotechnics, 523–577. Theophrastus Publications S. A., Athens, Greece. cal, isotopic, and rare earth element evidence for contrasting water-
K, K.D. & T, C.D. 1997. Environmental geochemistry of shale- rock interactions at two undisturbed Zn–Pb massive sulphide deposits,
hosted Ag-Pb-Zn massive sulfide deposits in northwest Alaska: natural Bathurst Mining Camp, N.B., Canada. Journal of Geochemical Exploration, 64,
background concentrations of metals in water from mineralized areas. 237–261.
Applied Geochemistry, 12, 397–409. L, M.I., G, W.D. & B, D.R. 1999. Stream waters and
K, D., D C, P. & D, R. 2003. The application of the stable stream sediment geochemistry of the Restigouche – Murray Brook area, Bathurst Mining
isotopes of sulfur and oxygen in groundwater sulfate to mineral explora- Camp, New Brunswick (parts of NTS 21O7, 21O8, 21O9 and 21O10). GSC
tion in the Broken Hill region of Australia. Journal of Geochemical Exploration, Open-File D3763.
78–79, 81–84. L, M.I., G, W.D. & B, D.R. 2002. Sulfide oxidation
L, D. 1978. Uranium solution-mineral equilibria at low temperatures and groundwater transport of base metals at the Halfmile Lake and
with applications to sedimentary ore deposits. Geochimica et Cosmochimica Restigouche Zn–Pb massive sulfide deposits, Bathurst Mining Camp, New
Acta, 42, 547–569. Brunswick. Geochemistry: Exploration, Environment, Analysis, 2, 37–44.
L, D. & C, J.R. 1980. Groundwater prospecting for L, M.I., G, W.D., B, D.R. & H, G.E.M. 2000a.
sandstone-type uranium deposits: a preliminary comparison of the merits Form and distribution of anthropogenic gold mobilized in surface waters
of mineral–solution equilibria, and single-element tracer methods. Journal of and sediments from a gossan tailings pile, Murray Brook massive sulfide
Geochemical Exploration, 13, 201–219. deposit, New Brunswick, Canada. Applied Geochemistry, 15, 629–646.
L, R.R., A, R.L., B, M.D. & H, W. 2001a. Hydro- L, M.I., G, W.D., B, D.R. & H, G.M. 2000b.
thermal alteration and volatile element halos for the Rosebery K Lens Rapid development of negative Ce anomalies in surface waters and
volcanic-hosted massive sulfide deposit, Western Tasmania. Economic contrasting REE patterns in groundwaters associated with Zn-Pb massive
Geology, 96, 1055–1072. sulphide deposits. Applied Geochemistry, 15, 695–723.
L, R.R., G, J.B., P, H. & H, D.L. 2001b. The L, M.I., C, I.D. & G, W.D. 2006a. Stable isotope
alteration box plot: a simple approach to understanding the relationship geochemistry of ground and surface waters associated with undisturbed
between alteration mineralogy and lithogeochemistry associated with massive sulfide deposits; constraints on origin and water-rock reactions.
volcanic-hosted massive sulfide deposits. Economic Geology, 96, 957–971. Chemical Geology, 231, 300–325.
L-M, D., L, M.I., P, J.M. & S, S.D. 2006. L, M.I., P, J.M., L-M, D., V, J. & B,
Determination of Selenium Isotopic Ratios by Continuous-Hydride- D.R. 2006b. Mobility and fractionation of rare earth elements during
Generation Inductively-Coupled-Plasma Mass Spectrometry. Journal of supergene weathering and gossan formation and chemical modification of
Analytical Atomic Spectroscopy, 21, 41–49. massive sulfide gossan. Geochimica et Cosmochimica Acta, 70, 1097–1112.
L, M.I. 1998. Hydrochemistry of ground and surface waters associated with L, M.I., C, B.L. & G, W.D. 2009a. Lead isotopes
massive sulphide deposits, Bathurst Mining Camp, New Brunswick: Halfmile Lake in ground and surface waters: fingerprinting heavy metal sources in mineral
and Restigouche deposits. PhD Thesis, University of Ottawa. exploration. Geochemistry: Exploration, Environment, Analysis, 9, 115–123.
L, M.I. 2001. Mineralogy and geochemistry of suspended sediments L, M.I., H, S.M., M, J.C., G, W.D.
from groundwaters associated with undisturbed Zn–Pb massive sulfide & B, D.R. 2009b. Hydrogeochemical and geophysical evidence for in
deposits. Canadian Mineralogist, 39, 1597–1616. situ oxidation of an undisturbed massive sulphide deposit and development
L, M.I. 2007. Aqueous geochemistry in mineral exploration. In: of self-potential anomalies, Stratmat Main Zone, New Brunswick. Geochem-
G, W.D. (ed.) Mineral Resources of Canada: A Synthesis of Major istry: Exploration, Environment, Analysis, 9, 39–50.
Deposit-types, District Metallogeny, the Evolution of Geological Provinces, and M, A.W., B, R.D., F, M.A.F. &  S, H.A.F. 2005.
Exploration Methods. Mineral Deposits Division, Geological Association of Vertical ionic migration: mechanisms, soil anomalies, and sampling depth
Canada, Special Publications, 5, 1007–1034. for mineral exploration. Geochemistry: Exploration, Environment, Analysis, 5,
L, M.I. & C, E.M. 2006a. Composition of groundwaters 201–210.
associated with porphyry-Cu deposits, Atacama Desert, Chile: Elemental MM, I. & MC, M.B. 2001. Till geochemistry and sampling
and isotopic constraints on water sources and water-rock reactions. techniques in glaciated shield terrain: a review. In: MC, M.B.,
Geochimica et Cosmochimica Acta, 70, 1616–1635. B, P.T., H, G.E.M. & C, S.J. (eds) Drift Exploration in
L, M.I. & C, E.M. 2006b. Composition of soils and Glaciated Terrain. Geological Society, London, Special Publications, 185,
groundwaters at the Pampa del Tamarugal, Chile: anatomy of a fossil 19–43.
geochemical anomaly derived from a distant porphyry copper deposit. M, W.R., F, W.H. & L, R.E. 1982. Hydrogeochemical
Economic Geology, 101, 1569–1581. prospecting for porphyry copper deposits in the tropical-marine climate of
L, M.I. & C, E.M. 2008. Source, transport, and fate of Puerto Rico. Journal of Geochemical Exploration, 16, 217–233.
rhenium, selenium, molybdenum, arsenic, and copper in groundwater M, W., T, R., Y, L. & S, L. 1995. Stream
associated with porphyry-Cu deposits, Atacama Desert, Chile. Chemical water, an ideal medium for rapidly locating polymetallic mineralization in
Geology, 247, 208–228. forest-swamp terrain: a case study in the Da Hinggan Mts., northern
L, M.I. & C, B.L. 2005. Rare earth elements (REE) and Nd Heilongjiang, China. Journal of Geochemical Exploration, 55, 257–264.
and Sr isotopes in groundwater and suspended sediments from the P, D.E. & S, W.E.J. 2004. Serpentinization and heat
Bathurst Mining Camp, New Brunswick: water-rock reactions and elemen- generation: constraints from Lost City and Rainbow hydrothermal systems.
tal fractionation. In: J, K.H. (ed.) Rare Earth Elements in Geochimica et Cosmochimica Acta, 68, 1347–1354.
Groundwater Flow Systems. Springer Verlag, Berlin, 254–293. P, J.L. & R, M.H. 2004. Geochemical models of metasomatism in
L, M.I. & G, W.D. 2003. Processes of metal solution ultramafic systems: serpentinization, rodingitization, and sea floor carbon-
and transport in ground waters interacting with undisturbed massive ate chimney precipitation. Geochimica et Cosmochimica Acta, 68, 1115–1133.
sulfide deposits, Bathurst Mining Camp, New Brunswick. In: P, M.A. & D, A. 2003. Quaternary geology of the Bathurst
G, W.D., MC, S.R. & P, J.M. (eds) Massive Mining Camp and implications for base metal exploration using drift
Sulphide Deposits of the Bathurst Mining Camp, New Brunswick, and Northern prospecting. In: G, W.D., MC, S.R. & P, J.M.
Maine. Society of Economic Geologists, Inc., Economic Geology Mono- (eds) Massive Sulphide Deposits of the Bathurst Mining Camp, New Brunswick, and
graph, 11, 723–740. Society of Economic Geologists, Littleton, CO, Northern Maine. Economic Geology Monograph, 11, 631–660. Society of
USA. Economic Geologists, Littleton, CO, USA.
118 M. I. Leybourne & E. M. Cameron

P, D.L. 1995. User’s guide to PHREEQC—a computer program for S, S., J, C.M. et al. 2004. The effect of plume processes on
speciation, reaction-path, advective-transport, and inverse geochemical calculations. U.S. the Fe isotope composition of hydrothermally derived Fe in the deep
Geological Survey, Denver, CO, USA. ocean as inferred from the Rainbow vent site, Mid-Atlantic Ridge,
P, D.L. & A, C.A.J. 1999. User’s guide to PHREEQC (version 3614 N. Earth and Planetary Science Letters, 225, 63–76.
2); a computer program for speciation, batch-reaction, one-dimensional transport, and S, R.H. 2000. Gold-rich porphyry deposits: descriptive and genetic
inverse geochemical calculations. Water-Resources Investigations – U. S. Geo- models and their role in exploration and discovery. Reviews in Economic
logical Survey, Report WRI 99-4259. Geology, 13, 315–344.
P, H., T-W, M.-L. et al. 2002. Chemical characteristics of S, W.D. 2007. Porphyry deposits. In: G, W.D. (ed.)
groundwater around two massive sulphide deposits in an area of previous Mineral Resources of Canada: A Synthesis of Major Deposit-types, District Metallog-
mining contamination, Iberian Pyrite Belt, Spain. Journal of Geochemical eny, the Evolution of Geological Provinces, and Exploration Methods. Mineral
Exploration, 75, 17–41. Deposits Division, Geological Association of Canada, Special Publications,
P, J.C.J.,  J, J., C, L. & M, N. 2008. Development of 5, 223–243.
Cu and Zn isotope MC-ICP-MS measurements: Application to suspended S, R.V., S, R.M., B, B.S. & B, D.C. 2002. Hydrogeo-
particulate matter and sediments from the scheldt estuary. Geostandards and chemical exploration for unconformity-related uranium mineralization:
Geoanalytical Research, 32, 149–166. example from Palnadu sub-basin, Cuddapah Basin, Andhra Pradesh, India.
P, G.C., B, D.R. & C, I.D. 2004. Groundwater geochemistry Journal of Geochemical Exploration, 76, 71–92.
and exploration methods: Myra Falls volcanogenic massive-sulphide S, B.W. 1998. A new theory to explain the formation of soil geochemical
deposits, Vancouver Island, British Columbia, Canada. Geochemistry: Explo- responses over deeply covered gold mineralization in arid environments.
ration, Environment, Analysis, 4, 329–340. Journal of Geochemical Exploration, 61, 149–172.
P, M.C. & G, A.M. 2004. Application of groundwater-mineral S́, A., G́, S.R. & A́, S. 2001. Dissolution of
equilibrium calculations to geochemical exploration for sediment-hosted primary minerals in natural waters II. Mineral saturation state. Chemical
uranium: observations from the Frome Embayment, South Australia. Geology, 172, 251–276.
Geochemistry: Exploration, Environment, Analysis, 4, 113–127. T, T.T. & P, A.M. 1984. Hydrogeochemical prospecting of
P, G.S., S, K.S., M, M.R., F, W.H. & M, E.L. gold in an alpine bald mountain zone. Journal of Geochemical Exploration, 21,
1999. Geologic controls on the composition of natural waters and mine 355–360.
waters. In: F, L.H. & P, G.S. (eds) The Environmental Geochem- T, P.M. 1997. Ground waters and surface waters in exploration
istry of Mineral Deposits Part B: Case Studies and Research Topics. Reviews in geochemical surveys. In: G, A.G. (ed.) Proceedings of Exploration 97:
Economic Geology, 6B, 373–432. Society of Economic Geologists, Fourth Decennial International Conference on Mineral Exploration. Prospectors
and Developers Association, Toronto, ON, Canada, 271–284.
Littleton, CO, USA.
T, H.P. Jr. 1974. The application of oxygen and hydrogen isotope
P, O.S. & S, J. 2004. Experimental study of brucite dissolu-
studies to problems of hydrothermal alteration and ore deposition.
tion and precipitation in aqueous solutions: surface speciation and chemical
Economic Geology, 69, 843–883.
affinity control. Geochimica et Cosmochimica Acta, 68, 31–45.
T, L. & G, D. 2002. Phreeqc and Phreeqci: Geochemical
P, J.J., C, G. & J, A. 2001. Neogene evaporites in desert
modeling with an interactive interface. Ground Water, 40, 462–464.
volcanic environments: Atacama Desert, northern Chile. Sedimentology, 48,
T, T., F, H., M, K., T, I.H. & T, O.
1411–1431.
1998. Evidence for microbial activity at the glass-alteration interface in
R, J.A., Q, J. & H, W.S. 2003. Isotopic evidence for the source of
oceanic basalts. Earth and Planetary Science Letters, 162, 165–176.
Ca and S in soil gypsum, anhydrite and calcite in the Atacama Desert,
T, P. & B, C. 1991. Comparison between lead isotopes,
Chile. Geochimica et Cosmochimica Acta, 67, 575–586. 234
U/238U activity ratio and saturation index in hydrogeochemical explo-
R, C., S, U., S, H. & B, D. 1999a. Does bottle ration for concealed uranium deposits. Journal of Geochemical Exploration, 41,
type and acid-washing influence trace element analyses by ICP-MS on 181–196.
water samples?: A test covering 62 elements and four bottle types: high V, D. & W, S.A. 1990. Gold speciation in natural waters: I.
density polyethene (HDPE), polypropene (PP), fluorinated ethene pro-
Solubility and hydrolysis reactions of gold in aqueous solution. Geochimica et
pene copolymer (FEP) and perfluoroalkoxy polymer (PFA). The Science of Cosmochimica Acta, 54, 3–12.
the Total Environment, 239, 111–130. V, J.C. 2002. Massive sulfide deposits of the Wadi Bidah mineral district, Saudi
R, C., S, U., S, H. & B, D. 1999b. Influence of Arabia; geologic control of mineralization, remote sensing and mineral exploration,
filtration on concentrations of 62 elements analysed on crystalline bedrock geochemical exploration and petrogenesis. PhD Thesis, University of Texas, Dallas.
groundwater samples by ICP-MS. The Science of the Total Environment, 234, V, J.C., L, M.I. & B, D.R. 2000. Mobility of rare
155–173. earth elements during weathering of massive sulfide deposits, Bathurst Mining
R, F., MP, D.C. & C, A.G. 2005. Bacillus cereus, gold and Camp, New Brunswick, Canada. Geological Society of America Annual
associated elements in soil and other regolith samples from Tomakin Park Meeting, Geological Society of America, Reno, Nevada.
Gold Mine in southeastern New South Wales, Australia. Journal of Geochemi- W, J.B., R, N.,  S, C.R., L, F.J., J, G.A.
cal Exploration, 85, 81–98. & W, J.A. 1998. Geochemical and isotopic (Nd, O) data from
R, J.P. 2003. Tectono-magmatic precursors for porphyry Cu-(Mo- Ordovician felsic plutonic and volcanic rocks of the Miramichi Highlands:
Au) deposit formation. Economic Geology and the Bulletin of the Society of petrogenetic and metallogenic implications for the Bathurst Mining Camp.
Economic Geologists, 98, 1515–1533. Canadian Journal of Earth Sciences, 35, 237–252.
R, C.F.W. 2003. Dilute hydrothermal fluids within the groundwaters of the W, D., A, A., C, G. & G, A. 1997. Multi-isotopic
Monturaqui Basin, northern Chile: the role of Volcano Socompa. MSc Thesis, studies of groundwaters: exploration for concealed mineralization. Explo-
University of Texas, Dallas. ration and Mining Research News, 7, 5–7.
R, O., F, Y. & L, J.N. 2004. Subsurface processes at W, S.A. 2002. The aqueous geochemistry of the platinum group elements
the Lucky Strike hydrothermal field, Mid-Atlantic Ridge: evidence from with applications to ore deposits. Canadian Institute of Mining and Metallurgy
sulfur, selenium, and iron isotopes. Geochimica et Cosmochimica Acta, 68, Special Volume, 54, 211–249.
2295–2311. Y, N., P, T.E. & S, K. 1995a. Groundwater geochemistry in
R, D.D. & L, R.D. 1981. Hydrogeochemical exploration the Koongarra ore deposit, Australia (I): implications for uranium migra-
models for sedimentary uranium deposits: use of the computer model tion. Geochemical Journal, 29, 1–29.
WATEQFC. Journal of Geochemical Exploration, 15, 51–62. Y, N., P, T.E. & S, K. 1995b. Groundwater geochemistry in
S, J.A., L, M.I., MC, B.M. & H, S.M. the Koongarra ore deposit, Australia (II): activity ratios and migration
2007. Low temperature serpentinization processes and kimberlite ground- mechanisms of uranium series radionuclides. Geochemical Journal, 29, 31–54.
water signatures in the Kirkland Lake and Lake Timiskiming kimberlite Z, H., M, H. et al. 2002. The colloid chemistry of acid rock drainage
fields Ontario, Canada: Implications for diamond exploration. Geochemistry: solution from an abandoned Zn–Pb–Ag mine. Applied Geochemistry, 17,
Exploration, Environment, Analysis, 7, 3–21. 633–648.

Received 23 July 2008; revised typescript accepted 2 February 2009.