skip to main content

Title: Isotopic fractionation accompanying CO2 hydroxylation and carbonate precipitation from high pH waters at The Cedars, California, USA
The Cedars ultramafic block hosts alkaline springs (pH > 11) in which calcium carbonate forms upon uptake of atmospheric CO2 and at times via mixing with surface water. These processes lead to distinct carbonate morphologies with ‘‘floes” forming at the atmosphere-water interface, ‘‘snow” of fine particles accumulating at the bottom of pools and terraced constructions of travertine. Floe material is mainly composed of aragonite needles despite CaCO3 precipitation occurring in waters with low Mg/Ca (<0.01). Precipitation of aragonite is likely promoted by the high pH (11.5–12.0) of pool waters, in agreement with published experiments illustrating the effect of pH on calcium carbonate polymorph selection. The calcium carbonates exhibit an extreme range and approximately 1:1 covariation in d13C (9 to 28‰ VPDB) and d18O (0 to 20‰ VPDB) that is characteristic of travertine formed in high pH waters. The large isotopic fractionations have previously been attributed to kinetic isotope effects accompanying CO2 hydroxylation but the controls on the d13C-d18O endmembers and slope have not been fully resolved, limiting the use of travertine as a paleoenvironmental archive. The limited areal extent of the springs (0.5 km2) and the limited range of water sources and temperatures, combined with our sampling strategy, allow us to place tight constraints on the processes involved in generating the more » systematic C and O isotope variations. We develop an isotopic reaction–diffusion model and an isotopic box model for a CO2-fed solution that tracks the isotopic composition of each dissolved inorganic carbon (DIC) species and CaCO3. The box model includes four sources or sinks of DIC (atmospheric CO2, high pH spring water, fresh creek water, and CaCO3 precipitation). Model parameters are informed by new floe D44Ca data (0.75 ± 0.07‰), direct mineral growth rate measurements (4.8 to 8  107 mol/m2/s) and by previously published elemental and isotopic data of local water and DIC sources. Model results suggest two processes control the extremes of the array: (1) the isotopically light end member is controlled by the isotopic composition of atmospheric CO2 and the kinetic isotope fractionation factor (KFF (‰) = (a  1)  1000) accompanying CO2 hydroxylation, estimated here to be 17.1 ± 0.8‰ (vs. CO2(aq)) for carbon and 7.1 ± 1.1‰ (vs. ‘CO2(aq)+H2O’) for oxygen at 17.4 ± 1.0 C. Combining our results with revised CO2 hydroxylation KFF values based on previous work suggests consistent KFF values of 17.0 ± 0.3‰ (vs. CO2(aq)) for carbon and 6.8 ± 0.8‰ for oxygen (vs. ‘CO2(aq)+H2O’) over the 17–28 C temperature range. (2) The isotopically heavy endmember of calcium carbonates at The Cedars reflects the composition of isotopically equilibrated DIC from creek or surface water (mostly HCO- 3, pH = 7.8–8.7) that occasionally mixes with the high-pH spring water. The bulk carbonate d13C and d18O values of modern and ancient travertines therefore reflect the proportion of calcium carbonate formed by processes (1) and (2), with process (2) dominating the carbonate precipitation budget at The Cedars. These results show that recent advances in understanding kinetic isotope effects allow us to model complicated but common natural processes, and suggest ancient travertine may be used to retrieve past meteoric water d18O and atmospheric d13C values. There is evidence that older travertine at The Cedars recorded atmospheric d13C that predates large-scale combustion of fossil fuels. « less
Teagle, Damon A
Award ID(s):
Publication Date:
Journal Name:
Page Range or eLocation-ID:
Sponsoring Org:
National Science Foundation
More Like this
  1. Abstract
    This dataset contains monthly average output files from the iCAM6 simulations used in the manuscript &#34;Enhancing understanding of the hydrological cycle via pairing of process-oriented and isotope ratio tracers,&#34; in review at the Journal of Advances in Modeling Earth Systems. A file corresponding to each of the tagged and isotopic variables used in this manuscript is included. Files are at 0.9° latitude x 1.25° longitude, and are in NetCDF format. Data from two simulations are included: 1) a simulation where the atmospheric model was &#34;nudged&#34; to ERA5 wind and surface pressure fields, by adding an additional tendency (see section 3.1 of associated manuscript), and 2) a simulation where the atmospheric state was allowed to freely evolve, using only boundary conditions imposed at the surface and top of atmosphere. Specific information about each of the variables provided is located in the &#34;usage notes&#34; section below. Associated article abstract: The hydrologic cycle couples the Earth&#39;s energy and carbon budgets through evaporation, moisture transport, and precipitation. Despite a wealth of observations and models, fundamental limitations remain in our capacity to deduce even the most basic properties of the hydrological cycle, including the spatial pattern of the residence time (RT) of water inMore>>
  2. The Medieval Climate Anomaly (MCA; 900e1300 AD) was the most recent period of pre-industrial climatic warming in the northern hemisphere, and thus estimations of MCA signals can illuminate possible impacts of anthropogenic climate change. Current high-resolution MCA climate signals are restricted to mid- and high-latitude regions, which confounds inferences of how the MCA impacted some global/hemispheric climate mechanisms (e.g. North Atlantic Oscillation; NAO). To address this knowledge gap, we estimate seasonally-resolved sea surface temperatures (SSTs) from the oxygen isotope composition (d18O) of serially sampled Phorcus atratus shells from archaeological sites spanning the MCA in the Canary Islands. Twelve archaeological and six modern P. atratus shells were analyzed, and archaeological shells were dated using carbonate-target radiocarbon dating. SSTs were estimated using the published aragonite-water equilibrium fractionation equation. Modern shells showed a mean SST of 20.0 ± 1.5 C, with a seasonal amplitude of 5.3 ± 0.9 C. Archaeological shells exhibited a mean SST of 18.2 ± 0.7 C, with a mean seasonal amplitude of 5.5 ± 1.0 C. Thus, shells that span the MCA in the Canary Islands recorded SSTs that were significantly cooler than the modern (P < .05), contrasting with warming estimates and model predictions elsewhere in the Northernmore »Hemisphere. We propose that the observed cooling resulted from increased upwelling in NW Africa due to a strengthening of the prevailing westerlies and coastal winds along the African shoreline. The intensified upwelling scenario during the MCA is partially supported by in-situ carbon isotope data (d13C) retrieved from the archaeological shells, which was compared to the d13C values of modern shells and dissolved inorganic carbon in the ambient seawater. These results are consistent with other low-latitude temperature/precipitation anomalies associated with a positive NAO mode, suggesting a transition to a positive NAO index during the middle and late MCA that possibly extended later into the 13th century AD.« less
  3. This Chapter considers triple oxygen isotope variations and their 4 Gyr temporal evolution in bulk siliciclastic sedimentary rocks and in granites. The d18O and D'17O values provide new insights into weathering in the modern and ancient hydrosphere and coeval crustal petrogenesis. We make use of the known geological events and processes that affect the rock cycle: supercontinent assembly and breakup that influence continent-scale and global climate, the fraction of the exposed crust undergoing weathering, and isotopic values of precipitation. New data from a 5000 m Texas drillhole into the Oligocene Frio Formation demonstrate minimal isotopic shifts from mudrocks to shales during diagenesis, mostly related to expulsion of water from smectite-rich loosely cemented sediment and its conversion to illite-rich shale. Inversion of triple oxygen isotope fractionations return isotopic values and temperatures along the hole depth that are more consistent with weathering conditions in the Oligocene and modern North America (d18O = -7 to -15‰, and T of +15 to +45°C) rather than d18O from 8 to 10‰ diagenetic water in the drill hole at 175-195°C. More precise T and d18Owater are obtained where the chemical index of alteration (CIA) based detrital contribution is subtracted from these sediments. Triple oxygen isotopes frommore »suspended sediments in major world rivers record conditions (T and d18Ow) of their watersheds, and not the composition of bedrock because weathering is water-dominated. In parallel, the Chapter presents new analyses of 100 granites, orthogneisses, migmatites, tonalite-trondhjemite-granodiorite (TTG), and large-volume ignimbrites from around the world that range in age from 4 Ga to modern. Most studied granites are orogenic and anatectic in origin and represent large volume remelting/assimilation of shales and other metasediments; the most crustal and high-d18O of these are thus reflect and record the average composition of evolving continental crust. Granites also develop a significant progressive increase in d18O values from 6-7‰ (4-2.5 Ga) to 10-13‰ (~1.8-1.2 Ga) after which d18O stays constant or even decreases. More importantly, we observe a moderate -0.03‰ step-wise decrease in D'17O between 2.1 and 2.5 Ga, which is about half of the step-wise decrease observed in shales over this time interval. We suggest that granites, as well as shales, record the significant advent and greater volumetric appearance of low-D'17O, high-d18O weathering products (shales) altered by meteoric waters upon rapid emergence of large land masses at ~2.4 Ga, although consider alternative interpretations. These weathering products were incorporated into abundant 2.0-1.8 Ga orogens around the world, where upon remelting, they passed their isotopic signature to the granites. We further observe the dichotomy of high-D'17O Archean shales, and unusually low-D'17O Archean granites. We attribute this to greater contribution from shallow crustal hydrothermal contribution to shales in greenstone belts, while granites in the earliest 3.0-4.0 Ga crust and TTGs require involvement of hydrothermal products with lower-D'17O signatures at moderately high-d18O, which we attribute to secondary silicification of their protoliths before partial melting. The Chapter further discusses evolution of the shale record through geologic history and discusses the step-wise change in d18O and D'17O values at Archean/Proterozoic transition. Denser coverage for shales in the past 1 billion years permits investigation of the rocks and their weathering in the last supercontinent cycle, with observed lighter d18O values, characteristic for the mid-Phanerozoic at the initiation of Gondwana breakup. The continuing increase in d18O values of the shales since 4 Ga is interpreted to reflect accumulation of weathering products via shale accretion to continents, as low-density and buoyant shales tend to not subduct back into the mantle. The rock cycle passes triple oxygen isotopic signatures from precipitation to sedimentary, metasedimentary, and finally to anatectic igneous rocks. Continental crust became progressively heavier in d18O, lighter in D'17O due to incremental accumulation of high-d18O sediments in accretionary wedges. Second-order trends in d18O and D'17O are due to supercontinent cycles and glacial episodes.« less
  4. Organic and inorganic stable isotopes of lacustrine carbonate sediments are commonly used in reconstructions of ancient terrestrial ecosystems and environments. Microbial activity and local hydrological inputs can alter porewater chemistry (e.g., pH, alkalinity) and isotopic composition (e.g., δ18Owater, δ13CDIC), which in turn has the potential to impact the stable isotopic compositions recorded and preserved in lithified carbonate. The fingerprint these syngenetic processes have on lacustrine carbonate facies is yet unknown, however, and thus, reconstructions based on stable isotopes may misinterpret diagenetic records as broader climate signals. Here, we characterize geochemical and stable isotopic variability of carbonate minerals, organic matter, and water within one modern lake that has known microbial influences (e.g., microbial mats and microbialite carbonate) and combine these data with the context provided by 16S rRNA amplicon sequencing community profiles. Specifically, we measure oxygen, carbon, and clumped isotopic compositions of carbonate sediments (δ18Ocarb, δ13Ccarb, ∆47), as well as carbon isotopic compositions of bulk organic matter (δ13Corg) and dissolved inorganic carbon (DIC; δ13CDIC) of lake and porewater in Great Salt Lake, Utah from five sites and three seasons. We find that facies equivalent to ooid grainstones provide time‐averaged records of lake chemistry that reflect minimal alteration by microbial activity, whereasmore »microbialite, intraclasts, and carbonate mud show greater alteration by local microbial influence and hydrology. Further, we find at least one occurrence of ∆47 isotopic disequilibrium likely driven by local microbial metabolism during authigenic carbonate precipitation. The remainder of the carbonate materials (primarily ooids, grain coatings, mud, and intraclasts) yield clumped isotope temperatures (T(∆47)), δ18Ocarb, and calculated δ18Owater in isotopic equilibrium with ambient water and temperature at the time and site of carbonate precipitation. Our findings suggest that it is possible and necessary to leverage diverse carbonate facies across one sedimentary horizon to reconstruct regional hydroclimate and evaporation–precipitation balance, as well as identify microbially mediated carbonate formation.« less
  5. The δ18O of carbonate minerals that formed at Earth’s surface is widely used to investigate paleoclimates and paleo-elevations. However, a multitude of hydrologic processes can affect δ18O values, including mixing, evaporation, distillation of parent waters, and carbonate growth temperatures. We combined traditional carbon and oxygen isotope analyses with clumped (Δ47) and triple oxygen isotopes (Δ′17O) analyses in oyster shells (Acutostrea idriaensis) of the Goler Formation in southern California (USA) to obtain insights into surface temperatures and δ18O values of meteoric waters during the early Eocene hothouse climate. The Δ47-derived temperatures ranged from 9 °C to 20 °C. We found a correlation between the δ18O of growth water (δ18Ogw) (calculated using Δ47 temperatures and δ18O of carbonate) and the δ13C values of shells. The Δ′17O values of shell growth waters (0.006‰–0.013‰ relative to Vienna standard mean ocean water–standard light Antarctic precipitation [VSMOW-SLAP]) calculated from Δ′17O of carbonate (–0.087‰ to –0.078‰ VSMOW-SLAP) were lower than typical meteoric waters. These isotopic compositions are consistent with oyster habitation in an estuary. We present a new triple oxygen isotope mixing model to estimate the δ18O value of freshwater supplying the estuary (δ18Ofw). The reconstructed δ18Ofw of –11.3‰ to –14.7‰ (VSMOW) is significantly lower than themore »δ18Ogw of –4.4‰ to –9.9‰ that would have been calculated using “only” Δ47 and δ18O values of carbonate. This δ18Ofw estimate supports paleogeographic reconstructions of a Paleogene river fed by high-elevation catchments of the paleo–southern Sierra Nevada. Our study highlights the potential for paired Δ47 and Δ′17O analyses to improve reconstructions of meteoric water δ18O, with implications for understanding ancient climates and elevations.« less