Evidence from Monitoring Soil CO2, Drip Water, and Modern Speleothem Calcite in Central Texas
Total Page:16
File Type:pdf, Size:1020Kb
Available online at www.sciencedirect.com ScienceDirect Geochimica et Cosmochimica Acta 142 (2014) 281–298 www.elsevier.com/locate/gca Interpretation of speleothem calcite d13C variations: Evidence from monitoring soil CO2, drip water, and modern speleothem calcite in central Texas Kyle W. Meyer a,⇑, Weimin Feng a,b, Daniel O. Breecker a, Jay L. Banner a, Amber Guilfoyle a,1 a Department of Geological Sciences, The University of Texas at Austin, Austin, TX, United States b Department of Physics, Astronomy, and Geosciences, Valdosta State University, Valdosta, GA, United States Received 10 August 2013; accepted in revised form 28 July 2014; available online 8 August 2014 Abstract We studied the sources and transport of carbon in two active karst systems in central Texas, Inner Space Cavern (IS) and Natural Bridge North and South Caverns (NB), to provide new insights into the interpretation of speleothem (cave calcite 13 13 deposit) carbon isotope compositions. We have determined the d C values of soil CO2 (d Cs) in grassland and savanna 13 13 À above these caves with d C values of cave drip water (d CHCO3 ) and modern speleothem calcite grown on artificial substrates 13 13 À (d Ccc). We compare d CHCO3 values from direct drip sites, where water was sampled immediately upon discharging from the cave ceiling, to values from indirect sites, where water was sampled after flowing along a prolonged path within the cave 13 À that allowed for longer CO2 degassing and have found that direct drip sites yield consistently lower d CHCO3 values. The 13 À & & d CHCO3 values of direct drip water below savanna (À10.6 ± 0.5 and À12.6 ± 0.2 , in NB and IS, respectively) are indis- 13 À tinguishable from (IS) or similar to (NB) calculated d CHCO3 values in equilibrium with measured soil CO2 beneath trees (À13.5& to À11.3& for juniper trees above NB, and À13.6& to À12.6& for mixed oak and elm trees above IS, respectively). 13 À & & At IS, the d CHCO3 values of direct drip water are higher below grassland (À9.7 ± 0.3 ) than below savanna (12.6 ± 0.2 ). 13 À These results suggest that the d CHCO3 values of drip waters that initially enter the caves are controlled by deep-rooted plants, 13 À where present, and are minimally influenced by host-rock dissolution and/or prior calcite precipitation (PCP). The d CHCO3 values of indirect drip water vary seasonally with relatively low values during the summer (À10.8 ± 0.8& and À9.2 ± 0.4& 13 À under juniper savanna at NB and under grassland at IS, respectively) that are similar to the direct drip d CHCO3 values & & 13 (À10.6 ± 0.5 and À9.7 ± 0.3 under savanna at NB and under grassland at IS, respectively). The relatively high d CHCO3- 13 À À & & values of indirect drip sites during the winter (d CHCO3 = À8.6 ± 0.8 at NB and 8.0 ± 0.1 at IS) result from CO2 degas- sing of water along in-cave flow paths. We also present decade-long records of modern calcite d13C values from direct and 13 indirect drip sites at IS. The d Ccc values vary seasonally with lower values during the summer and higher values during the winter, and with smaller amplitude variations at the direct drip site. Such seasonal variations can be used as a geochro- 13 nological tool in some speleothems that do not contain visible lamina. The summer d Ccc values of direct drip calcite are 13 similar to d Ccc values predicted from soil CO2 collected beneath trees above that drip site. The occurrence of highest 13 d Ccc values during the winter, when cave CO2 concentrations are low, highlights the significance of ventilation-driven 13 changes in cave-air pCO2. Modern calcite d C values are also negatively correlated with drip rate, which suggests that 13 d Ccc variations are controlled by kinetic effects during degassing and calcite precipitation associated with the drip water ⇑ Corresponding author. Current address: Department of Earth and Environmental Sciences, University of Michigan, Ann Arbor, MI, United States. E-mail address: [email protected] (K.W. Meyer). 1 Current address: Environmental Services Business Group, CH2M Hill, Asheville, NC, United States. http://dx.doi.org/10.1016/j.gca.2014.07.027 0016-7037/Ó 2014 Elsevier Ltd. All rights reserved. 282 K.W. Meyer et al. / Geochimica et Cosmochimica Acta 142 (2014) 281–298 13 exposure time to a low-pCO2 environment. In all, at the caves we investigated, variability in speleothem d Ccc values primar- ily reflect presence/absence of deep-rooted vegetation and kinetic isotope effects. We therefore infer that increased aridity may 13 13 result in higher d C values of vegetation, lower drip rates and more drip water degassing, and thus higher d Ccc values of speleothem calcite. Ó 2014 Elsevier Ltd. All rights reserved. 13 1. INTRODUCTION of soil CO2 (d Cs), DIC in drip waters (particularly À focused on the carbonate species of bicarbonate, HCO3 ; 13 À Speleothems (i.e., cave calcite deposits) serve as impor- d CHCO3 ), and modern calcite grown on artificial sub- 13 tant terrestrial proxies for paleoenvironmental and paleocli- strates (d Ccc) in the caves. A two-phase approach was matic conditions (McDermott, 2004; Fairchild et al., 2006). adopted to investigate: (1) processes in soil and epikarst 13 À Hendy and Wilson (1968) first evaluated the potential of that influence d CHCO3 value before water enters caves; speleothem oxygen isotope compositions (d18O) in preserv- and (2) in-cave processes that influence the d13C value of ing climatic information. The development of uranium- drip water DIC and calcite. series geochronology (e.g., Edwards et al., 1986; Musgrove et al., 2001) allowed for precise age constraints 2. BACKGROUND of changes in speleothem d18O values. Since then, variations in speleothem calcite d18O values have been used to docu- 2.1. The formation of karst landscapes ment environmental and climatic changes associated with, among others, Heinrich events (Bar-Matthews et al., Speleothems in most regions are cave-calcite deposited 1999), East-Asian monsoon intensity (Wang et al., 2001) from groundwater that travels through the vadose zone dis- and shifts in the Intertropical Convergence Zone solution region and into the calcite precipitation region (Dykoski et al., 2005). In general, speleothem d18O values (Fairchild et al., 2006). The dissolution region constitutes are controlled by the d18O value and temperature of the the soil horizon and the epikarst (the weathered/fractured water entering a cave and to a variable extent by kinetic iso- bedrock where groundwater conduits and flow paths tope effects (termed “kinetic effects” hereafter) occurring develop). In this region, CO2 respired by plants and micro- during CO2 degassing and calcite precipitation (Hendy bial communities dissolves into vadose zone waters and and Wilson, 1968; Hendy, 1971; Mickler et al., 2004, dissociates into a series of carbon-bearing ions, collectively 2006; Feng et al., 2012). known as dissolved inorganic carbon (DIC), according to 13 Speleothem calcite carbon isotope compositions (d Ccc) the following reactions: have also been employed for paleoclimate reconstructions 13 CO2ðgÞ $ CO2ðaqÞ ð1aÞ (e.g., Dorale et al., 1992). The interpretations of d Ccc val- ues are myriad, complex, and often region- or case-specific. H2O þ CO2ðaqÞ $ H2CO3ðaqÞ ð1bÞ 13 þ À Proposed factors affecting speleothem calcite d C values H2CO3ðaqÞ $ H þ HCO3ðaqÞ ð1cÞ 13 include: the d C value of atmospheric CO2 (Baskaran À þ 2À HCO3ðaqÞ $ H þ CO3ðaqÞ ð1dÞ and Krishnamurthy, 1993), the relative proportions of C3 and C4 vegetation above caves (Dorale et al., 1992; The resulting waters are acidic, and dissolve host limestone. Dorale, 1998), the soil/vegetation productivity affected by The pH of water interacting with limestone bedrock is com- temperature and precipitation (Genty et al., 2006), carbon monly between 7.5 and 8.5, such that the dominant DIC À 2+ isotope fractionation during decomposition of soil organic species is bicarbonate (HCO3 ). Water containing Ca(aq) matter (Baker et al., 2011), the proportion of host rock car- and DIC is transported by matrix, fracture, or conduit flow bon (e.g., Hendy, 1971; Genty et al., 2001; Fairchild et al., to open cavities in the subsurface (Atkinson, 1977; White 2006), and in-cave kinetic effects (Mickler et al., 2004, 2006; and White, 2005). The calcite precipitation region is defined Spo¨tl et al., 2005; Mattey et al., 2010; Baker et al., 2011; by the entrance of these waters into any void space within Frisia et al., 2011; Lambert and Aharon, 2011). The kinetic the subsurface with a lower pCO2 than the epikarst, and effects are primarily caused by two mechanisms: (1) degas- where precipitation of calcite occurs by the overall reaction: 12 sing, whereby CO2 preferentially escapes from oversatu- 2þ À 13 Ca þ 2HCO $ CaCO þ CO þ H O ð2Þ rated waters resulting in higher d C values of residual ðaqÞ 3ðaqÞ 3ðsÞ 2ðgÞ 2 ðlÞ dissolved inorganic carbon (DIC); and (2) precipitation, whereby carbonate ions are incorporated into rapidly pre- 2.2. Surface and shallow-subsurface processes: Soil horizon cipitating CaCO3 without equilibrium isotope exchange and the epikarst between solid CaCO3 and solution. This study aims to determine the factors controlling the Several factors control the d13C values of soil pore space d13C values of speleothem calcite in three central Texas 13 CO2 (i.e., soil CO2; d Cs). Soil CO2 is a mixture of caves by investigating the processes that influence C isotope atmospheric CO2 and soil-respired CO2, which is derived compositions during transmission of water from the surface from root/rhizosphere respiration and soil organic 13 to the site of speleothem growth.