Elsevier

Geochimica et Cosmochimica Acta

Volume 199, 15 February 2017, Pages 247-263
Geochimica et Cosmochimica Acta

Investigating the hydrological significance of stalagmite geochemistry (Mg, Sr) using Sr isotope and particulate element records across the Late Glacial-to-Holocene transition

https://doi.org/10.1016/j.gca.2016.10.024Get rights and content

Abstract

The trace element and Sr isotope records in two coeval stalagmites characterized by different growth rates and flow regimes at Savi cave (Grotta Savi, NE Italy) reveal different sources and incorporation mechanisms for Mg and Sr. Mg is sourced primarily from dissolved cave host rock while particulate Mg derived from soil plays a subordinate role. The presence of particulate-borne Mg is inferred from the co-variation of Mg and particle-associated elements (Th, Al and Mn) which are preferentially concentrated in open columnar calcite layers. Variation in Mg concentrations corrected for particle-influenced components, the Mgc parameter, is controlled by water–rock interaction, with higher and lower Mgc during dry and wet phases, respectively. This is thought to reflect incongruent dissolution of Mg-rich phases. Correction of Sr concentrations for contributions from airborne exogenic Sr, based on 87Sr/86Sr ratios, yields the bedrock-only contribution (Src). Src variation in stalagmite calcite is influenced by speleothem growth rate and by variation of the calcite-water Sr partitioning in wet and dry phases, and only to a minor extent by incongruent dissolution of Mg-rich phases. Concentration profiles for Mgc and Srcg (corrected for growth rate effects) show inverse correlations and are inferred to show hydrological significance which is captured in a hydrological index, HI. We suggest HI provides robust information on water–rock interaction related to hydrological changes and can be utilized in both wet and semi-arid environments, provided the corrections for soil Mg and exogenic Sr can be applied with confidence. Application of the HI index allows correction of Grotta Savi oxygen isotope data, to yield a δ18Oc time series that shows when changes in moisture sources and atmospheric reorganization, or changes in moisture amount, were significant. This is especially evident during the Younger Dryas (YD). The Savi record supports the concept of a two-phase YD, marked by an increase of moisture and stronger impact of Adriatic and Mediterranean Sea influences over the northern Adriatic region from 12.3 ka onwards. Then, a large-scale atmospheric reorganization and gradual northward shift of the Polar Front caused a progressive reduction of sea influence over the region from 12.1 ka, supporting the concept of a hemispheric change.

Introduction

Trace element concentrations in speleothems are a proxy for past climate and ecosystem variability over a wide range of environmental conditions (Treble et al., 2003, Borsato et al., 2007, Griffiths et al., 2010, Moreno et al., 2010, Zhou et al., 2011). The incorporation of trace elements into speleothem calcite is a function of both the chemical composition of the parent water and the crystal growth mechanisms. Data for spatial and temporal trace element variability within layers composed of the same crystal fabric allow evaluation of the interaction between climate parameters and dripwater chemistry, the latter being controlled by soil processes and host rock dissolution (Fairchild and Baker, 2012).

Dripwater transports trace elements into the cave in both dissolved and undissolved (particulate-colloidally bound, PCB) form (Fairchild and Treble, 2009). PCB elements include Al, Fe, Mn and Th, which are commonly adsorbed onto organic or inorganic colloidal particles (Hartland et al., 2012). These generally immobile elements are thought to be liberated from soil silicates under low-soil-pH conditions and during high-infiltration events (Kabata-Pendias, 2011). In contrast, the dissolved constituents of dripwater, such as Mg and Sr, are generally mobile elements which are transported as simple inorganic complexes in percolating waters. The concentration of Sr and Mg solutes in cave dripwater is a function of the residence time of the water in the aquifer when the bedrock is their only source (Tremaine and Froelich, 2013, Rutlidge et al., 2014). Consequently, these elements are considered to be reliable hydrological proxies (Hellstrom and McCulloch, 2000, McMillan et al., 2005, Fairchild and McMillan, 2007). Co-variance of Sr/Ca and Mg/Ca ratios indicates increased incongruent dissolution in carbonate phases in the host rock and is related to the residence time of the water in the aquifer (Sinclair, 2001). However, studies invoke incongruent dissolution to explain anti-correlated behavior between stalagmite Sr/Ca and Mg/Ca (e.g. Roberts et al., 1998). Positive covariation between Sr/Ca and Mg/Ca can also be achieved through prior calcite precipitation, which is accompanied by a parallel shift in δ13C due to the changes in the amount of 12C lost through degassing (Fairchild et al., 2000, Tooth and Fairchild, 2003, Johnson et al., 2006, Mattey et al., 2008, Wong et al., 2011, Stoll et al., 2012, Tremaine and Froelich, 2013, Regattieri et al., 2014). Coupled increases in speleothem Mg/Ca and Sr/Ca would be interpreted as marking low infiltration conditions (dry periods) in both the prolonged ‘water-rock’ interaction (WRI) and prior calcite precipitation (PCP) scenarios.

In fossil speleothems, the concentration of trace elements in the parent solution is commonly unknown, and the ‘partition-coefficient’ approach, which implies equilibrium incorporation, cannot be tested because non-equilibrium processes cannot be ruled out. The assumption that modern dripwater composition reflects the past may be valid only when the general climate mode has remained similar (Schimpf et al., 2011). During major climate transitions, this assumption is obviously not applicable. An alternative approach is to assume that fossil dripwater Mg and Sr concentrations can be estimated from the composition of the host rock, when this is the sole source of these two solutes in the dripwater. This assumption is valid when karst soil geochemistry is the result of bedrock weathering, with minimal input from allochthonous materials (e.g., Belli et al., 2013). When the host rock is the major source of solutes, a “speleothem enrichment factor” (SEF) can be calculated by comparing the concentration of a given trace element in the speleothem and in the host rock (Frisia et al., 2012). SEF has been found to be a useful tool for estimating the contribution of dolomite and limestone dissolution to groundwater Mg and Sr, and thus stalagmite, concentrations (Borsato et al., 2016). However, SEF cannot account for a contribution from external sources, such as airborne dust.

Beside the host rock, aeolian dust is a potential source of additional Mg and Sr, particularly in semi-arid regions. Based on data from infiltration experiments in the Wellington caves (NSW Australia), Rutlidge et al. (2014) demonstrated that Mg has both soil and bedrock components, whereas Sr is solely bedrock-derived. Their study highlights that interpretation of speleothem Sr and Mg elemental records should differentiate between seasonally dry and hot (semi-arid) settings such as Wellington caves and more temperate settings. The conclusion that Mg–Sr systematics in speleothems from semi-arid settings are unlikely to follow the single-source model probably also applies to stalagmites formed in periglacial settings, e.g. during a deglaciation when aeolian dust accumulated during glacial stages is mobilized. Wind-blown dust contributions can often be recognized from 87Sr/86Sr studies (e.g. Goede et al., 1998, Ayalon et al., 1999, Musgrove and Banner, 2004).

To contribute to this debate, we report trace element and Sr isotopic records for two stalagmites (SV1 and SV7) from Grotta Savi, in the southern foothills of the Alps, NE Italy. The Grotta Savi site was not glaciated during the Last Glacial Maximum but, similar to most periglacial locations in the Italian Alps, was influenced by deposition of aeolian dust during glacial conditions (Monegato et al., 2011). The proxy records in stalagmites SV1 and SV7 are thought to capture regional temperature and hydrological conditions during the Late Glacial to Holocene transition in the northern Adriatic region. The two stalagmites are characterized by different fabrics which reflect diverse infiltration systems, and are thus good materials to test a novel approach for interpreting anti-correlated Mg/Ca–Sr/Ca elemental records. This approach uses observed relationships between trace element concentrations, mineral fabrics and variations in 87Sr/86Sr to identify if Mg and Sr have multiple sources and allows separation of the various source terms to isolate and model the hydrological-climatic component.

We use this approach to define a hydrological index which is thought to be relevant to stalagmites grown under both wet-temperate and semi-arid to arid conditions, provided the different contributions to observed Mg–Sr records from different sources can be identified from microstratigraphy, PCB element variability and Sr isotopes. Savi speleothems are known to have recorded hydrological changes (Belli et al., 2013) and we use the Mg–Sr based hydrological index to search for significant changes in moisture source or amount in the δ18Oc records. In the case of Savi, this approach helps to clarify the response of climate in SE Alps of the large-scale reorganization of atmospheric and oceanic circulation that characterized the Younger Dryas (YD) (Brauer et al., 2008, Bakke et al., 2009, Rach et al., 2014), which caused important hydrological variability and South to North changes of wind direction (Baldini et al., 2015, Bartolomé et al., 2015).

Section snippets

Study site and speleothem description

Stalagmites SV1 and SV7 were collected 500 m from the main entrance in Grotta Savi, located near the Adriatic coast in NE Italy (45°37′N and 13°53′E). Today the site is characterized by temperate Mediterranean climate (Belli et al., 2013), but in the Late Glacial, proximity to the Alps and a more continental location (a result of sea levels 80–60 m below present) (Lambeck and Bard, 2000) probably supported a different pattern of orographically forced mesoscale circulation (Pichler and Steinacker,

Petrographic observation

Microstratigraphic observations were conducted on 30 micrometer-thick thin sections using a Zeiss Axioplan polarizing microscope at the School of Environmental and Life Sciences of the University of Newcastle. The SV1 microstratigraphic logging is based on the method of Frisia (2015), where fabrics are coded, transformed into numbers and logged against depth/ages. This method allows direct comparison of chemical and morphological data, highlighting when chemical changes are related to crystal

Host rock and dripwater chemical signature

The trace element composition of the limestone host rock is reported in Table 1. The Opicina Limestone is mostly composed of low-Mg calcite (Mg concentration ∼5100 ppm) with a relatively high Sr content (Sr concentration ∼550 ppm). Minor bioclasts (up to 0.5 mm in size) comprise about 8 vol%, are most likely red algae and are characterized by Mg and Sr concentrations almost twice the average of the low-Mg calcite matrix (Fig. S2). This suggests that the original Mg-calcite composition of red algae

Particulate and colloidally-bound (PCB) elements

The concentration profiles of PCB elements (Al, Mn, Th) in stalagmite SV1 show distinctive patterns of centennial-scale abundance peaks (up to 100× baseline levels) superimposed on segments with very low and more or less constant concentrations. In the remainder of this paper, the abundance peaks in the PCB element profiles are referred to as mode-P, defined as intervals where Th concentrations exceed the relevant mean baseline by a factor of 2 or more. On this basis, the profile for SV1

Hydrological index (HI) based on Mg and Sr time series

The variability of Srcg can be used to complement hydrological information from Mg concentrations. For low to moderate flow regimes (mode-F), the two series have a similar but antipathetic pattern. With the maximum bedrock-derived Mg threshold at about 260 ppm (see Section 5.2), we obtain Mgc, the series with minimum interference of particle-associated Mg. Normalizing Srcg and Mgc series and plotting normalized Mgc inversely (Fig. 5A), we can directly compare corrected Sr–Mg variation patterns.

Conclusions

We have demonstrated that dated Mg and Sr concentration profiles in speleothems can be utilized as hydrological proxies provided that contributions from multiple sources to their concentrations in the speleothems can be deconvolved and trace element concentration profiles are tested against fabric microstratigraphy. The SV1 record provides good evidence of multiple sources of Mg and Sr. Exogenic contributions to their concentration variability was removed to obtain the solute uptake by calcite

Acknowledgments

The authors are grateful to F. Corradini (Istituto Agrario S. Michele all’Adige, Italy) for water analyses and F. Cucchi (Università di Trieste, Italy) for cave access and support during sampling. P. Bence (Melbourne University, Australia) is thanked for LA-ICP-MS analysis of host rock and stalagmite thin sections and S. Moggio (Newcastle University, Australia) is acknowledged for technical assistance and contributions to the discussion. This research was supported by a University of Newcastle

References (89)

  • I.J. Fairchild et al.

    Trace elements in speleothems as recorders of environmental change

    Quaternary Science Reviews

    (2009)
  • I.J. Fairchild et al.

    Controls on trace element (Sr–Mg) compositions of carbonate cave waters: implications for speleothem climatic records

    Chemical Geology

    (2000)
  • M. Frechen et al.

    Loess in Europe—mass accumulation rates during the Last Glacial Period

    Quaternary Science Reviews

    (2003)
  • S. Frisia et al.

    Late Holocene annual growth in three Alpine stalagmites records the influence of solar activity and the North Atlentic Oscillation on winter climate

    Earth and Planetary Science Letters

    (2003)
  • P. Gabrielli et al.

    Trace elements in winter snow of the Dolomites (Italy): a statistical study of natural and anthropogenic contributions

    Chemosphere

    (2008)
  • D. Genty et al.

    Timing and dynamics of the last deglaciation from European and North African δ13C stalagmite profiles-comparison with Chinese and South Hemisphere stalagmites

    Quaternary Science Reviews

    (2006)
  • A. Goede et al.

    Aeolian contribution to Sr and Sr isotope variation in a Tasmanian speleothem

    Chemical Geology

    (1998)
  • M.L. Griffiths et al.

    Evidence for Holocene changes in Australian–Indonesian monsoon rainfall from stalagmite trace element and stable isotope ratios

    Earth and Planetary Science Letters

    (2010)
  • A. Hartland et al.

    From soil to cave: transport of trace metals by natural organic matter in karst dripwaters

    Chemical Geology

    (2012)
  • J.C. Hellstrom et al.

    Multi-proxy constraints on the climatic significance of trace element records from a New Zealand speleothem

    Earth and Planetary Science Letters

    (2000)
  • Y. Huang et al.

    Partitioning of Sr2+ and Mg2+ into calcite under karst-analogue experimental conditions

    Geochimica et Cosmochimica Acta

    (2001)
  • K.R. Johnson et al.

    Seasonal trace-element and stable-isotope variations in a Chinese speleothem: the potential for high-resolution paleomonsoon reconstruction

    Earth and Planetary Science Letters

    (2006)
  • G. Kaufmann et al.

    Stalagmite growth and palaeo-climate: an inverse approach

    Earth and Planetary Science Letters

    (2004)
  • K. Lambeck et al.

    Sea-level change along the French Mediterranean coast for the past 30,000 years

    Earth and Planetary Science Letters

    (2000)
  • I. Larocque et al.

    Late-glacial chironomid-based temperature reconstructions for Lago Piccolo di Avigliana in the southwestern Alps (Italy)

    Palaeogeography, Palaeoclimatology, Palaeoecology

    (2008)
  • R.B. Lorens

    Sr, Cd, Mn and Co distribution coefficients in calcite as a function of calcite precipitation rate

    Geochimica et Cosmochimica Acta

    (1981)
  • D. Mattey et al.

    A 53 year seasonally resolved oxygen and carbon isotope record from a modern Gibraltar speleothem: reconstructed drip water and relationship to local precipitation

    Earth and Planetary Science Letters

    (2008)
  • J. McArthur et al.

    A revised Pliocene record for marine-87Sr/86Sr used to date an interglacial event recorded in the Cockburn Formation, Antarctic Peninsula

    Palaeogeography, Palaeoclimatology, Palaeoecology

    (2006)
  • A. Moreno et al.

    A speleothem record of glacial (25–11.6 kyr BP) rapid climatic changes from northern Iberian Peninsula

    Global and Planetary Change

    (2010)
  • J.W. Morse et al.

    The dissolution kinetics of major sedimentary carbonate minerals

    Earth Science Reviews

    (2002)
  • M. Musgrove et al.

    Controls on the spatial and temporal variability of vadose dripwater geochemistry: Edwards Aquifer, central Texas

    Geochimica et Cosmochimica Acta

    (2004)
  • C. Pin et al.

    Concomitant separation of strontium and samarium-neodymium for isotopic analysis in silicate samples, based on specific extraction chromatography

    Analytica Chimica Acta

    (1994)
  • L.N. Plummer et al.

    The solubilities of calcite, aragonite and vaterite in CO2–H2O solutions between 0 and 90 °C, and an evaluation of the aqueous model for the system CaCO3–CO2–H2O

    Geochimica et Cosmochimica Acta

    (1982)
  • R.J. Reeder

    Interaction of divalent cobalt, zinc, cadmium, and barium with the calcite surface during layer growth

    Geochimica et Cosmochimica Acta

    (1996)
  • M.S. Roberts et al.

    Annual trace element variation in a Holocene speleothem

    Earth and Planetary Science Letters

    (1998)
  • H. Rutlidge et al.

    Dripwater organic matter and trace element geochemistry in a semi-arid karst environment: Implications for speleothem paleoclimatology

    Geochimica et Cosmochimica Acta

    (2014)
  • D. Schimpf et al.

    The significance of chemical, isotopic, and detrital components in three coeval stalagmites from the superhumid southernmost Andes (53°S) as high-resolution palaeo-climate proxies

    Quaternary Science Reviews

    (2011)
  • D.J. Sinclair

    Two mathematical models of Mg and Sr partitioning into solution during incongruent calcite dissolution: implications for dripwater and speleothem studies

    Chemical Geology

    (2011)
  • J. Tang et al.

    Sr2+/Ca2+ and 44Ca/40Ca fractionation during inorganic calcite formation: I. Sr incorporation

    Geochimica et Cosmochimica Acta

    (2008)
  • A.J. Tesoriero et al.

    Solid solution partitioning of Sr2+, Ba2+ and Cd2+ into calcite

    Geochimica et Cosmochimica Acta

    (1996)
  • A.F. Tooth et al.

    Soil and karst aquifer hydrological controls on the geochemical evolution of speleothem-forming drip waters, Crag Cave, southwest Ireland

    Journal of Hydrology

    (2003)
  • P. Treble et al.

    Comparison of high resolution sub-annual records of trace elements in a modern (1911–1992) speleothem with instrumental climate data from southwest Australia

    Earth and Planetary Science Letters

    (2003)
  • L.E. Wasylenki et al.

    Nanoscale effects of strontium on calcite growth: an in situ AFM study in the absence of vital effects

    Geochimica et Cosmochimica Acta

    (2005)
  • C.I. Wong et al.

    Seasonal dripwater Mg/Ca and Sr/Ca variations driven by cave ventilation: Implications for and modeling of speleothem paleoclimate records

    Geochimica et Cosmochimica Acta

    (2011)
  • Cited by (41)

    View all citing articles on Scopus
    1

    Present address: Centre for Integrative Biology, University of Trento, 38123 Povo, Italy.

    View full text