Laurence A. Coogan, Randall R. Parrish, and Nick M.W. Roberts. S3. Table of all data used in age determination as an Excel file.

Transcription

1 GSA Data Repository Early hydrothermal carbon uptake by the upper oceanic crust: Insight from in situ U-Pb dating Laurence A. Coogan, Randall R. Parrish, and Nick M.W. Roberts This supplementary information consists of three parts: S1. Background information about the study sites S2. Analytical techniques and data reduction S3. Table of all data used in age determination as an Excel file. S1. Study locations DSDP Sites 163/164: DSDP Sites 163 and 164, drilled on Legs 16 and 17, are in the equatorial central Pacific (van Andel et al., 1973; Winterer et al., 1973). Based on the seafloor age maps of Muller et al. (2008), Sites 163 and 164 have crustal age of 80.8 and Myr respectively (throughout the crustal ages are taken from GeoMapApp: The age for Site 163 is consistent with the Campanian biostratigraphic age derived from the deepest portion of the 276 m of sediment overlying basement in this hole (van Andel et al., 1973). Site 164 is overlain by 264 m of sediment but there is poor control on the biostratigraphy in the basal sediments; these are at least Albian in age (Winterer et al., 1973), consistent with the assigned age.

2 DSDP Site 307: DSDP Site 307, drilled on Leg 32, is in the NW Pacific; basement is overlain by 307 m of sediment (Larson et al., 1975). Based on the seafloor age maps of Muller et al. (2008) the crustal age is Myr. Based on the radiolarian assemblage the basal sediments are Valanginian to Berriasian (132.5 to 145 Myr; Larson et al., 1975) consistent with the assigned crustal age. DSDP Sites 417/418: DSDP Sites 417D and 418A are in the western Atlantic, with Site 417D about 8 km north of Site 418A (Donnelly et al., 1979). Based on the seafloor age maps of Muller et al. (2008) Sites 417D and 418A have crustal age of and Myr respectively and the basement is covered by 352 and 323 m of sediment respectively (Donnelly et al., 1979). At both sites biostratigraphy suggests that the oldest sediments are Aptian (Donnelly et al., 1979) consistent with the assigned crustal age. DSDP Site 543: DSDP Site 543A is in the western Atlantic just outboard of the deformation front associated with the Less Antillies subduction zone (Biju-Duval et al., 1984). Based on the seafloor age maps of Muller et al. (2008) the basement age is 80.8 Myr, consistent with the Campanian age derived from biostratigraphy (Bergen, 1984). The 411 m of sediment overlying this site has been thickened due to deformation and does not reflect the abyssal sedimentation history. DSDP Site 595B: DSDP Site 595B lies just to the north of the Osbourn trough in the western Pacific outboard of the Tonga-Kermadec trench (Billen and Stock, 2000). Based on the seafloor age maps of Muller et al. (2008) crust here has an age of 90.3 Myr however the local tectonics

3 are complex with two orthogonal spreading directions in this area potentially increasing the uncertainty on the model age. The Muller et al. (2008) crustal age is somewhat younger than the total fusion 40 Ar- 39 Ar ages of 96.8±0.6 and 101.5±0.6 for two ferrobasalts from this site (R. Duncan pers. comm. reported in Montgomery and Johnson, 1987). Because of this uncertainty in the crustal age in the figures we assign a crustal age of 95 ± 10 Myr to be conservative. This area of the Pacific has a very slow sedimentation rate with only 68 m of sediment covering this site. Troodos ophiolite: The Troodos ophiolite formed ~91.6 Myr ago based on U-Pb dating of zircon from plagiogranites (Mukasa and Ludden, 1987) in a supra-subduction zone setting. It is perhaps the only ophiolite that preserves a typical seafloor low-temperature (i.e. off-axis) alteration history with extensive preservation of volcanic glass and no overprint of highertemperature (obduction related) metamorphism (e.g. Gillis and Robinson, 1990). S2. Analytical techniques and data reduction Samples consisting of small chips of carbonate, all reasonably clear, were mounted in epoxy, polished and cleaned prior to insertion into a laser ablation cell (Fig. DR1). In general, methods of U-Pb analysis closely resemble those used for routine U-Pb zircon dating. Calcite was analysed for U-Pb geochronology mainly using a Nu Instruments Attom HR sector-field single-collector inductively coupled plasma mass spectrometer (SC-ICP-MS) following the method of Mottram et al. (2014) and Li et al. (2014). Laser ablation was performed with a New Wave Research UP193ss laser ablation system, using a New Wave Research large-format cell, or a New Wave Research 193UC with a TV2 cell; both feature a moveable cup with an ablation volume of ca. 3-4 cm 3 which, combined with <1 m of tubing to the plasma torch, leads to a

4 washout time of < 1 second. The ablation parameters were a 100 μm static spot, a repetition rate of 10 Hz, a fluence of ~5-6 Jcm -2, a 15 second washout period between analyses, and a 30 second ablation time. Data processing used the time-resolved analysis on the Nu Instruments software, an in-house Excel spreadsheet for data reduction, including subtraction for gas blank counts on Pb isotopes, and uncertainty propagation, and Isoplot (Ludwig, 2003) for age determination. All data are plotted at the 2σ confidence interval. The Nu Attom ICP-MS is used in peak-jumping mode with measurement on a MassCom secondary electron multiplier (SEM). The following masses are measured in each sweep: 202 Hg, 204 Pb+Hg, 206 Pb, 207 Pb and 238 U. Each data integration records 80 sweeps of the measured masses, which roughly equates to 0.4 seconds. A standard-sample-bracketing technique was used to normalise the data using the WC-1 calcite that has been measured by isotope dilution to yield an age of ~254±7 Ma (2σ); this material has ~5 ppm U and is approximately 90% radiogenic lead (although this is slightly variable). This procedure involved calculation of U-Pb normalisation factors based on the 254 Ma age of the WC-1 calcite using an assumed initial 207 Pb/ 206 Pb of 0.83±0.02. The normalisation is not sensitive to the initial ratio due to the highly radiogenic lead within the WC-1 calcite. It was analysed at regular intervals during each session, before and after groups of unknown analyses. The method of normalisation is to bracket based on a stable portion of an analytical session, which may vary from an entire day, to several blocks within one day. Because the reference material used for normalisation is heterogeneous (~2.75%), small (<~2%) variations in the normalisation procedure may add some bias to the final age. The reproducibility of the reference material is propagated to the final age uncertainty, which

5 should account for this effect, but it is worth noting this caveat to the method. 207 Pb/ 206 Pb normalisation used NIST 614 glass and values from Woodhead and Hergt (2001). Overall reproducibility achieved on the calcite in-house reference material was 4% (2 standard deviations) for U-Pb ages and this value is propagated onto the datapoint uncertainties as an excess variance. Plots of the WC-1 standard, after normalisation, are shown in Fig. DR2 for the three analytical sessions of this study over three years. Regressions of data from each sample were performed after eliminating most data with signals of Pb so weak that uncertainties in ratios were ~20% (1σ). Data with no U but higher Pb, were generally retained with an assumed U/Pb of zero, as these allow a good measurement of the initial lead isotope ratio. Intersections with the Tera-Wasserburg concordia were calculated using Isoplot (mostly Model 1, occasionally Model 2). In the final ages the additional 2.75% (2σ) uncertainty of the WC-1 standard was added in quadrature, and all final ages are quoted at 2σ in Table 1. Only one of the samples was anchored to a value (0.83±0.03) of common lead isotopic ratio (Table S3) because of its lack of any unradiogenic datapoints. S3. Interpretation of ages There are some samples that clearly have excess scatter; this may arise by more than one reservoir of initial common lead, or sampling zones that may have been altered or undergone more than one growth stage. It could also arise from the inherent difficulty of measuring very small Pb signals approaching the background uncertainty in Pb counts in the blank gas because there is temporal variation the magnitude of the gas blank counts. This latter aspect cannot always be fully accounted for in uncertainty propagation of low-pb samples and therefore may lead to some bias and additional unquantifiable uncertainty in some analyses. Most samples,

6 however, form a good array with low MSWD, suggesting but not proving that the arrays are mainly a simple mixture of radiogenic ingrowth with initial common lead with a single age of growth. In some cases the ages appear slightly older than the age of the host oceanic crust. The measurements, normalized to the WC-1 calcite appear satisfactory and we have no reason to suspect the measurements or uncertainties on arrays of data, but in cases such as 595B-2R1, 595B-3R2, or 2012-CL26 (i.e. the bottom row in Fig. 1), which are relatively non-radiogenic and involve large extrapolation to calculate an age, the uncertainties have to be treated as a minimum. It is possible, but not likely, that a compositional or textural bias of samples relative to the standard could introduce additional bias, but we have no way of evaluating this as of yet, given that we only have one calcite reference/normalization material. On the basis of our wider measurement experience of carbonate using this WC-1 material, including some by isotope dilution, if there is any compositional or textural bias in measuring an unknown against the WC- 1 material, it is likely to be a few percent bias at most, but again this cannot be fully evaluated at the present. This remains a caveat using this method.

7 Figure DR1. Photographs of carbonate grains showing analysis pits. Larger pits show the preablation to clean the surface. Smaller pits are older analyses (2014) that have been partially polished away during preparation for the next analytical session. Brighter images are from the 2011 analytical session. Scale bars are 1 mm in all parts.

8 2011 WC-1 measurements 2014 WC-1 measurements 2015 WC-1 measurements Figure DR2. Tera Wasserburg plots for the carbonate standard from each analytical session

9 References Van Andel, T.H., Heath, G.R., et al., 1973, Initial Reports of the Deep Sea Drilling Project, Volume 16, in US Government Printing Office; doi: /dsdp.proc Bergen, J.A., 1984, Calcareous nannoplankton from Deep Sea Drilling Project Leg 78A: evidence for imbricate underthrusting at the Lesser Antillian active margin, in B. Biju- Duval, J.C.M. et al. ed., Initial Reports of the Deep Sea Drilling Project, (U.S. Government Printing Office, Washington, D.C., 1984), p Biju-Duval, B., Moore J.C., et al., 1984, Initial Reports DSDP Leg 78A: Washington (U.S.Govt. Printing Office). Billen, M.I., and Stock, J., 2000, Morphology and origin of the Osbourn Trough: Journal of Geophysical Research: Solid Earth, v. 105, p , doi: /2000JB Donnelly, T., Francheteau, J., Bryan, W., Robinson, P., Flower, M., and Salisbury, M., 1979, Initial Reports of the Deep Sea Drilling Project, v. 51, 52, 53: U.S. Government Printing Office, Washington. Gillis, K.M., and Robinson, P.T., 1990, Patterns and processes of alteration in the lavas and dykes of the Troodos Ophiolite, Cyprus: J. Geophys. Res., v. 95, p. 21, ,548. Larson, R.L., and Moberly, R., 1975, Initial Reports of the Deep Sea Drilling Project, Volume 32: U.S. Government Printing Office. Li, Q., Parrish R.R., Horstwood M.S.A., and McArthur J.M., (2014), U-Pb dating of cements in Mesozoic ammonites: Chemical Geology, 376, Ludwig, K.R., User s Manual for Isoplot Berkeley Geochronology Center, Berkeley, CA.

10 Montgomery, A.F., and Johnson, H.P., 1987, Paleomagnetic studies of Leg 91 basalts and sediments, in Menard, H.W., Natland, J.H., Jordan, T.H., and Orcutt, J.A. eds., Init. Repts. DSDP, 91: Washington (U.S. Govt. Printing Office), p Mukasa, S.B., and Ludden, J.N., 1987, Uranium-lead isotopic ages of plagiogranites from the Troodos ophiolite, Cyprus, and their tectonic significance: Geology, v. 15, p Muller, R.D., Sdrolias, M., Gaina, C., and Roest, W.R., 2008, Age, spreading rates, and spreading asymmetry of the world s ocean crust: Geochemistry Geophysics Geosystems, v. 9, doi: /2007gc Winterer, E.L., and Ewing, J.I., 1973, Initial Reports of the Deep Sea Drilling Project, Volume 17: U.S. Government Printing Office. Woodhead, J.D., and Hergt, J.M, 2001, Strontium, neodymium and lead isotope analyses of NIST glass certified reference materials: SRM 610, 612, 614. Geostandards Newsletter, v. 25, p