Controls on chemistry during fracture-hosted flow of cold CO2-bearing mineral waters, Daylesford, Victoria, Australia: Implications for resource protection
Introduction
Mineral springs provide an opportunity to investigate the geochemistry of groundwater flow systems without needing to rely on observation bores. Compared to non-mineralised groundwater, mineral waters are characterised by high solute contents and commonly contain elevated concentrations of minor and trace elements such as Si, Sr, Li, heavy metals, Ba, and B (Parry and Bowman, 1990, Harris et al., 1997, Céron et al., 1998, Cruz et al., 1999, Grasby et al., 2000). Mineral waters also commonly contain substantial amounts of dissolved gases such as CO2 that can lead to effervescence during discharge (e.g., Harris et al., 1997, Céron et al., 1998, Siegel et al., 2004). Although many mineral springs are hydrothermal (e.g., Goldie, 1985, Parry and Bowman, 1990, Cruz et al., 1999, Grasby et al., 2000), a few discharge at close to ambient temperatures (e.g., Harris et al., 1997, Céron et al., 1998, Siegel et al., 2004).
Mineral water from springs in the Daylesford region of Victoria, SE Australia (Fig. 1) discharges at ambient temperature. It is characterised by high HCO3, Na, Ca, and Mg concentrations and, unlike groundwater from other parts of southeastern Australia (Herczeg et al., 2001, Cartwright et al., 2004), by low Cl concentrations relative to the concentrations of other anions. The springs contribute significantly to the local economy providing mineral water for commercial bottling plants and spas, and act as a focal point for regional tourism. As such, sustainable use of this resource and its protection from contamination are critically important. Numerous potential sources of contamination exist in the region, including discharge from agricultural land, concentrated livestock farming, recreation areas, and septic systems. The major and minor ion chemistry of these mineral waters, in conjunction with stable isotopes and other parameters, are used here to constrain water–rock interaction, groundwater flow paths and mixing, and to evaluate the vulnerability of the mineral water resource.
Over 100 recorded discharges of mineral water occur in the Daylesford region of the Central Highlands of Victoria (Fig. 2) over an area of approximately 60 km by 30 km (e.g., Wishart and Wishart, 1990, Shugg and Knight, 1994, Shugg, 1996). Most of the springs discharge in river valleys and creek beds through Ordovician turbidites of the Lachlan Fold Belt (e.g., Hepburn Springs and Sailors Falls). However, at a few locations, discharge occurs through overlying Newer Volcanic basalts (e.g., Turpins Falls) or through Pliocene to Recent alluvial sediments in river valleys. Although the Lachlan Fold Belt covers approximately 400,000 km2 from the east coast of New South Wales to the Adelaide region of South Australia, the Daylesford area is the only region characterised by extensive CO2-rich mineral water springs.
The Ordovician bedrock comprises metamorphosed and indurated shales and sandstones of the 4500-m-thick turbidite sequence of the Lachlan Fold Belt (Cas and VandenBerg, 1988). Throughout the Lachlan Fold Belt, the turbidites are extensively faulted and fractured (Gray and Foster, 1998). Quartz occurs both as the most common mineral in the turbidites and as veins. Granitic batholiths were emplaced into the region during the Devonian (e.g., White et al., 1988, Gray, 1997) and several crop out in the Daylesford region (Fig. 2). Pliocene to Recent Newer Volcanics basalt flows cover part of the region (Coulson, 1954) and are up to 30 m thick in the Daylesford area. These flows erupted as recently as 470 ka, from approximately 40 local eruption points. The Daylesford area has the highest concentration of eruption points in the Newer Volcanics Province (Fig. 2) (Nicholls et al., 1993). The Quaternary Newer Volcanics extend into South Australia and are younger in the southwest (Gill and Gibbons, 1969) at Tower Hill and Mount Gambier (Fig. 1). However, seismic tomography (Graeber et al., 2002) shows that a high-velocity p-wave anomaly occurs at about 45 km depth beneath the eastern part of the Newer Volcanics, coinciding with the highest density of eruption centres in the Daylesford region. This may represent a zone of higher heat flow related to the volcanic province (Graeber et al., 2002). The high density of eruption points suggests that the feeder system to the volcanic field below the Daylesford area is substantial, with large volumes of mafic intrusions likely to exist at depth in the Lachlan Fold Belt metasediments in this region.
The mineral water resides primarily in fractures and fissures in the Ordovician sediments that form the Ordovician Aquifer (Lawrence, 1969, Laing, 1977, Shugg and Knight, 1994, Shugg, 1996). As is common in SE Australia, groundwater in the shallowest Ordovician Aquifer is locally brackish to saline (Shugg, 1996) due to evapotranspiration both during recharge and from regions where the water table is within 1–2 m of the surface (e.g., Herczeg et al., 2001, Cartwright et al., 2004). Mining records show that, early in the 20th century, mineral water discharge was encountered in Au mines at depths of up to 450 m indicating that fracture-hosted groundwater flow occurs at considerable depth (Shugg and Knight, 1994). Records also indicate that several spring flows ceased during periods of extensive dewatering associated with mining (Laing, 1977).
The upper 50 m of the Ordovician sediments are highly weathered and contain kaolinite throughout the sediments and Fe oxyhydroxides along fracture zones. In this weathered zone, groundwater flow occurs both through fractures and the weathered bedrock. Where they are sufficiently thick, the basalt flows that cover parts of the weathered Ordovician sediments form the unconfined Quaternary basalt aquifer (Fig. 2). Groundwater from this aquifer has total dissolved solids (TDS) contents of 200–500 mg/L (Shugg, 1996) and is used for stock watering and irrigation.
Recharge to the mineral water system was assumed by Shugg and Knight, 1994, Shugg, 1996 to occur mainly through fractured basalts on the crest of the Great Dividing Range (a range of mountains and hills running along most of Australia’s eastern seaboard that consists of hills approximately 50–60 m above general elevation levels in this region). The mineral springs occur up to 45 km north and 60 km south of this watershed. Locally, recharge may be focussed around the many eruption points in the basalt aquifer (Fig. 2). Stable isotope data presented by Cartwright et al. (2002) indicate that the mineral water currently discharging at the springs was recharged under cooler climates, probably prior to 4 ka.
In their natural state, spring discharges in the region are identified by brown-stained seeps, and gas and mineral water discharge in stream beds. However, most of the mineral springs in the Daylesford region have been developed into discrete outlets, and some have been developed several times. Development involves installing pumps over the spring eyes, driving pipes into stream banks, or digging culverts into weathered rock (Wishart and Wishart, 1990). Previous studies have reported mineral water composition (Schaefer and Kecskemeti, 1981, Wishart and Wishart, 1990, Bannister, 1992, Shugg and Knight, 1994) and many have discussed the source of entrained gases (Skeats, 1914, McLaughlin and Macumber, 1968, Lawrence, 1969, Laing, 1977, Chivas et al., 1983, Shugg and Knight, 1994, Cartwright et al., 2002). In spite of these reports, there has been little discussion of the development of the major ion chemistry of the mineral water. In this paper, the authors identify the sources of the solutes, discuss the controls on the groundwater chemistry, and use groundwater chemistry as the basis for understanding mineral water flow paths and vulnerability.
Section snippets
Methods and results
The sampled springs cover the extent of the mineral spring area (Fig. 2), extending ∼60 km south of the Great Dividing Range to Ballan, and ∼35 km north to Turpins Falls. Detailed descriptions of the localities are provided by Wishart and Wishart, 1990, Cartwright et al., 2002. Mineral water samples were collected in 1998 and 1999 using polyethylene tubing from either pipe or pump outlets. At pump outlets, the pump was started and samples were collected when pH had stabilised unless water volumes
CO2 sources
To determine how the composition of these mineral waters developed, the H2O, dissolved gases, and solutes, must all be considered. Stable isotope data indicate that meteoric water, most probably recharged during cooler climatic conditions (Cartwright et al., 2002), is the main source of water. Most samples lie either on, or to the left of, the local meteoric water line, with δ2H values of −44‰ to −37‰ V-SMOW (Fig. 4(a)). However, samples from Hepburn-Soda and Vaughan have δ2H values of −32‰ and
Conclusions
The major aqueous species in the mineral water from the Daylesford springs are derived from distinct sources during water–rock interaction and groundwater mixing along the flow path and at discharge. The variability in cation concentrations, including Ca, Na, Mg, Sr, Ba, and Li, between even closely adjacent springs implies that these cations are derived primarily from water–rock interaction along fracture-hosted flow paths and that groundwater from individual springs remains isolated from
Acknowledgements
This study was supported by grants from the Australian Research Council and the Victorian Mineral Waters Committee (VMWC), Department of Natural Resources and Environment. We thank A. Shugg and members of the VMWC for providing background information. P. Chow and the late E. Gerasimova (CEAF) provided cation analyses. We thank A. Bath and H.N. Waber for comments on an earlier version of this manuscript and K. Durocher, P. deCaritat, and I. Hutcheon for their comments on this version.
References (42)
- et al.
The chemistry of geothermal waters in Iceland. III. Chemical geothermometry in geothermal investigations
Geochim. Cosmochim. Acta
(1983) - et al.
Stable isotope geochemistry of cold CO2-bearing mineral spring waters, Daylesford, Victoria, Australia: sources of gas and water and links with waning volcanism
Chem. Geol.
(2002) - et al.
Hydrogeochemical and isotopic constraints on the origins of dryland salinity, Murray Basin, Victoria, Australia
Appl. Geochem.
(2004) - et al.
Isotopic identification of CO2 from a deep origin in thermomineral waters of southeastern Spain
Chem. Geol.
(1998) - et al.
Chemistry of water from Furnas volcano, São Miguel, Azores: fluxes of volcanic carbon dioxide and leached material
J. Volcanol. Geotherm. Res.
(1999) - et al.
Strontium and carbon isotope constraints on carbonate–solution interactions and inter-aquifer mixing in groundwaters of the semi-arid Murray Basin, Australia
J. Hydrol.
(2002) Geothermal solute equilibria. Derivation of Na–K–Mg–Ca geoindicators
Geochim. Cosmochim. Acta
(1988)- et al.
The influence of water–rock interaction on the chemistry of thermal springs in western Canada
Appl. Geochem.
(2000) - et al.
Character and kinematics of faults within the turbidite-dominated Lachlan Orogen; implications for tectonic evolution of eastern Australia
J. Struct., Geol.
(1998) - et al.
Chemical and stable isotopic models for Boundary Creek warm springs, southwestern Yellowstone National Park, Wyoming
J. Volc. Geotherm. Res.
(1990)
Hydrogeochemistry of the Goulburn Valley region of the Murray Basin, Australia: implications for flow paths and resource vulnerability
Hydrogeol. J.
Physical volcanology; Explosive eruption centres
Ordovician
The volcanic rocks of the Daylesford District
Proc. Roy. Soc. Victoria
An Introduction to the Rock Forming Minerals
Constraints on the circulation of meteoric water in hydrothermal systems imposed by the solubility of quartz
Abstracts With Programs – Geol. Soc. Am.
The behaviour of silica in hydrothermal solutions
Cited by (18)
Tracing the migration of mantle CO<inf>2</inf> in gas fields and mineral water springs in south-east Australia using noble gas and stable isotopes
2019, Geochimica et Cosmochimica ActaStructural controls on the location and distribution of CO<inf>2</inf> emission at a natural CO<inf>2</inf> spring in Daylesford, Australia
2019, International Journal of Greenhouse Gas ControlCitation Excerpt :The mineral waters have a residence time of ˜4.5 ka (Cartwright et al., 2002) and are assumed to recharge within the nearby Great Dividing Range, as well as through local volcanics that outcrop at higher elevations in the regional topography (Shugg, 1996). The Daylesford mineral springs are high in calcium, magnesium, and bicarbonates, and so are quite different from typical Australian Na-Cl rich groundwater (Cartwright et al., 2002; Weaver et al., 2006). Individual spring water chemistry has changed little during 20 years of detailed measurements although some springs exhibit mixing with fresh water during discharge (Weaver et al., 2006).
The influence of oxygen isotope exchange between CO<inf>2</inf> and H<inf>2</inf>O in natural CO<inf>2</inf>-rich spring waters: Implications for geothermometry
2017, Applied GeochemistryCitation Excerpt :CO2 is reportedly mantle-sourced, based on their close proximity to the eruptive centres (Lawrence, 1969), 3He/4He gas data (Chivas et al., 1983), and gas carbon isotope values (δ13C; Cartwright et al., 2002). Spring waters are Na-HCO3 type with a significant solute composition variation between individual springs (Weaver et al., 2006). The Pah Tempe springs discharge in the damage zone of the currently active Hurricane fault at Timpoweap Canyon, Utah, USA (Fig. 3b).
Geochemistry of CO<inf>2</inf>-rich waters in Iceland
2016, Chemical GeologyCitation Excerpt :Water samples from Snæfellsnes and Torfajökull plot above the local meteoric water line but have δ18O values less than the range of values observed in local groundwater by 0.15 to 0.90‰, which is outside the range of error for the δ18O measurements (Section 3.2) (Fig. 10). Such a negative shift in δ18O relative to local meteoric water, observed previously in other CO2-rich waters, has been attributed to oxygen isotope exchange between CO2 (g) and H2O (aq) (Delalande et al., 2011; Kristmannsdóttir and Sveinbjoernsdóttir, 2012; Marques et al., 2006; Matthews et al., 1987; Newell et al., 2008; Pauwels et al., 1997; Siebe et al., 2007; Vuataz and Goff, 1986; Weaver et al., 2006). During exchange, 18O is preferentially incorporated into CO2 (g) over H2O (aq) as indicated by the fractionation factor (αH2O(aq)CO2(g)>1) (Bottinga, 1968; Friedman and O'Neil, 1977), which leaves the sampled aqueous phase depleted in 18O relative to the gas phase.
Trace metal contamination of mineral spring water in an historical mining area in regional Victoria, Australia
2013, Journal of Asian Earth Sciences