Alan R. Gagnon

Radiocarbon chronology of Black Sea sediments

Accelerator Mass Spectrometer (AMS) radiocarbon analyses have been made on 102 samples from 12 sediment cores and 23 samples from two water column profiles. These materials, collected during the first leg of the 1988 joint U.S.-Turkish Black Sea Expedition, provide the most comprehensive radiocarbon chronology of Black Sea sediments yet attempted. Radiocarbon analyses from carefully collected box cores and a mollusc shell collected live in 1931 suggest the prebomb surface waters had a Δ14C value of −55% (460 years) and that the maximum detrital correction for radiocarbon ages of Unit I sediments is 580 years for the organic carbon and 260 years for the carbonate fractions. Evidence does not support the 1430–2000 years pre-bomb surface water and/or detrital corrections argued for in past studies. The best estimates for the age of the beginning of the final invasion of the coccolithophore Emiliania huxleyi (Unit 12 boundary of Ross and Degens, 1974, The Black Sea—geology, chemistry and biology, pp. 183–199) and the age of the first invasion of E. huxeleyi (Unit I/II boundary of Hayet al., 1991, Deep-Sea Research, 38, S1211–S1235) are 1635 ± 60 and 2720 ± 160 years BP, respectively. Sapropel formation began at approximately 7540 ± 130 years BP at all depths in the basin, a pattern in disagreement with those predicted by existing time-evolution models of sapropel formation for this basin. Our data suggest that the oxic-anoxic interface has remained relatively stable throughout the Holocene, is controlled largely by the physical oceanography of the basin, and has not evolved as assumed by previous workers.

TIC, TOC, DIC, DOC, PIC, POC — unique aspects in the preparation of oceanographic samples for 14C-AMS

Abstract The radiocarbon content of discrete carbon pools (total (T), dissolved (D), and particulate (P) inorganic (I) and organic (O) carbon (C)) is a useful tracer of carbon cycling within the modern and past ocean. The isolation of different carbon pools in the ocean environment and conversion to CO 2 presents unique analytical problems for the radiocarbon chemist. In general, isolation and preparation of inorganic carbon presents few problems; dissolved carbon is easily extracted by acidifying the sample and stripping with an inert gas. Carbon is also readily isolated from particulate carbonate samples; in this case, CO 2 is prepared by hydrolysis of the substrate with phosphoric acid. The isolation and preparation of organic carbon presents a much greater problem. Dissolved organic carbon (DOC) must first be isolated from DIC and then oxidized in the presence of very high salt concentrations. We present results from a closed-tube combustion method in which the DIC-free seawater is evaporated to dryness, transferred to a clean combustion tube, and oxidized overnight at 550°C. Combustion of total organic carbon (TOC) in sediments with a high inorganic carbon content is also difficult. Removal of CaCO 3 with acid leaves severely deliquescent salts which, if not thoroughly dried, cause combustion tubes to explode. Removal of the salts by rinsing can also remove significant amounts of organic matter. Finally, we present results from a local coastal region.

Ten years after – The WOCE AMS radiocarbon program

The National Ocean Sciences Accelerator Mass Spectrometry (NOSAMS) Facility is measuring all of the samples collected as part of the US WOCE Program ‐ over 13,000 samples. We designed our extraction lines so that we also measure precise, oceanographically useful d 13 C-RCO2 values. We have completed the analysis of samples from the Pacific and Southern Oceans and are processing those from the Indian Ocean now. At present, this constitutes the world’s largest AMS data set. Reviews of the Pacific radiocarbon data are available and demonstrate the increased penetration of the ‘‘bomb signal’’ into the water column since the 1970s. Stable isotope data are being combined with those collected as part of NOAA’s Ocean-Atmosphere Carbon Exchange Study to study the ocean’s role in the anthropogenic CO2 cycle. The relationship of d 13 C to other chemical tracers, e.g., PO4 ,O 2 and chlorofluorocarbons, will further our understanding of basic oceanographic processes. We present preliminary results from these studies as well as investigate the relationship of 14 Ct o 13 C in the ocean. ” 2000 Elsevier Science B.V. All rights reserved.

The NOSAMS sample preparation laboratory in the next millenium: Progress after the WOCE program

Since 1991, the primary charge of the National Ocean Sciences AMS (NOSAMS) facility at the Woods Hole Oceanographic Institution has been to supply high throughput, high precision AMS 14C analyses for seawater samples collected as part of the World Ocean Circulation Experiment (WOCE). Approximately 13,000 samples taken as part of WOCE should be fully analyzed by the end of Y2K. Additional sample sources and techniques must be identified and incorporated if NOSAMS is to continue in its present operation mode. A trend in AMS today is the ability to routinely process and analyze radiocarbon samples that contain tiny amounts (<100 μg) of carbon. The capability to mass-produce small samples for 14C analysis has been recognized as a major facility goal. The installation of a new 134-position MC-SNICS ion source, which utilizes a smaller graphite target cartridge than presently used, is one step towards realizing this goal. New preparation systems constructed in the sample preparation laboratory (SPL) include an automated bank of 10 small-volume graphite reactors, an automated system to process organic carbon samples, and a multi-dimensional preparative capillary gas chromatograph (PCGC).

Reproducibility of Seawater, Inorganic and Organic Carbon 14C Results at NOSAMS

The majority of samples processed at the National Ocean Sciences AMS Facility (NOSAMS) thus far were collected as part of the World Ocean Circulation Experiment (WOCE). Due to the long storage time (2-3 yr) required to analyze thousands of samples on the accelerator mass spectrometer (AMS), a test was designed and implemented to determine the effects, if any, of storage time on (super 14) C concentration. We find no systematic offsets in AMS measurements made over a 5-yr period between a total of 16 replicate sets from surface and deep water collected at the same locality. Furthermore, the average delta (super 14) C value from the deepwater AMS replicates (-213.1 per mil, std. dev. 7.3) agrees very closely with the conventional (super 14) C results published for GEOSECS (-212.7 per mil) from station 320 taken 20 yr earlier. A total of 73 WOCE shipboard replicate sets (162 AMS measurements) were analyzed with a mean precision of 4.3 per mil. AMS results from 20 more shipboard replicate sets (44 AMS measurements) submitted as CO (sub 2) from the Stable Isotope Laboratory (SIL) at the University of Washington were analyzed with a mean precision of 3.4 per mil. These results suggests no significant difference between water stripping methods used in each preparation lab. To assess reproducibility, we calculate a pooled estimate of sigma for replicates called s, which we use as an approximation of sigma (sub TOT) for a given sample type. The s for WOCE seawater replicates is 4.9 per mil and 5.8 per mil for SIL gas replicates. These numbers demonstrate an overall reproducibility of seawater AMS results at NOSAMS that is in line with reported errors. We take the difference between total error s and machine error as the overall standard deviation of combined uncertainties associated with preparation of samples and with AMS. For seawater samples processed at NOSAMS, sigma (sub SPL) is calculated to be 2.4 per mil, and for the SIL gas replicates it is 4.8 per mil. Reproducibility of samples prepared with an acid hydrolysis technique is demonstrated using 24 coral samples submitted in triplicate by Dr. R. G. Fairbanks of Lamont Doherty Earth Observatory. Seventy-two replicates were prepared and analyzed at NOSAMS with a mean reported precision of 1.2 per mil. The pooled estimate s for the Fairbanks triplicates is 2.6 per mil. We calculate a laboratory reproducibility uncertainty for coral hydrolysis samples of 2.2 per mil. In 1993, NOSAMS participated in the Third International Radiocarbon Intercomparison (TIRI) Study. We report here 60 AMS analyses of the six TIRI test materials, five of which are organic carbon samples, to validate sample-processing methods for organic carbon sample AMS analyses at NOSAMS.

High-precision AMS radiocarbon measurements of central Arctic Ocean sea waters

Abstract We report on the first sub-5%. precision radiocarbon dataset measured on single targets using accelerator mass spectrometry (AMS). A 13 sample water column profile collected in the Canada Basin (74°N, 150°W, 3850 m water depth) of the central Arctic Ocean in September 1992 has been analyzed in duplicate and the average total precision achieved for the 26 targets is ± 3.2%.. The reproducibility of the 13 paired analyses averages +- 4.8%. as determined by a chi-square fit minimization for a quality factor of unity, and ± 7.8%. using ANOVA. Eliminating two of the 13 paired analyses because of apparent outlier behavior in one of the two analyses comprising the pair results in a total precison of ± 3.2%., a chi-square fit of ± 3.5%., and ANOVA precision of ± 3.5%.. Comparison with a recently published AMS 14 C profile from the same basin suggests these data are accurate as well. Results show that the deep waters of the Canada Basin have a renewal rate of 430 years, in comparison with 250 years estimated for the deep waters of the Eurasian Basin. Although the major requirement of the World Ocean Circulation Experiment (WOCE) for a radiocarbon analysis precision of ± 3 to 4%. for deep water samples can now be met with the AMS technology available at the National Ocean Sciences AMS Facility at the Woods Hole Oceanographic Institution for single-target analyses, a careful program of duplicate analyses should be included to insure the highest quality in the WOCE Δ 14 C dataset.

AMS-graphite target production methods at the Woods Hole Oceanographic Institution during 1986-1991.

In July 1986, an AMS radiocarbon target preparation laboratory was established at the Woods Hole Oceanographic Inst. to produce graphite to be analyzed at the NSF-Accelerator Facility for Radioisotope Analysis at the Univ. of Arizona (Tucson). By June 1991, 923 graphite targets had been prepared and 847 analyzed. The lab procedures during this time included the careful documentation of weights of all starting samples, catalysts and final graphite yields, as well as the volume of CO[sub 2] gas evolved during CaCO[sub 3] hydrolysis or closed-tube organic carbon combustions. From these data, the authors evaluate the methods used in general and in this lab.

Internal and external checks in the NOSAMS sample preparation laboratory for target quality and homogeneity

Abstract In the NOSAMS sample preparation laboratory (SPD we have developed rigorous internal procedures aimed at ensuring that sample preparation introduces as little error into our analyses as possible and identifying problems rapidly. Our three major CO 2 preparation procedures are: stripping inorganic carbon from seawater, hydrolyzing CaCO 3 , and oxidizing organic matter. For seawater, approximately 10% of our analyses are standards or blanks which we use to demonstrate extraction of virtually all the inorganic carbon. Analysis of the stable carbon isotopic composition of the CO 2 extracted from our standards indicates a precision of better than 0.15–0.20‰. We also routinely process 14 C-free CO 2 in our stripping lines to demonstrate the absence of a significant process-dependent blank. For organic combustions and CaCO 3 hydrolyses, we use the carbon yield (% organic carbon (OC) or % CaCO 3 by weight) as a check on our sample procedures. We have analyzed the blank contribution of these procedures as a function of sample size. Our organic carbon blank is constant at approximately 0.4% modern for samples containing greater than 1 mg C and our carbonate blank is less than 0.2% modern for samples containing more than 0.5 mg C. We use a standard Fe/H 2 catalytic reduction to prepare graphite from CO 2 . We check the completeness of our reactions with the pressure data stored during the reaction as well as use a robot to determine a gravimetric yield. All graphite undergoes a visual inspection and is rejected if any heterogeneities are present. We have recombusted graphite made from CO 2 with δ 13 C values ranging from −42 to 1‰ and determined that the δ 13 C of the recombusted carbon agrees with that from the pure gas to within 0.05‰, demonstrating little or no fractionation during the treatment of the sample. The δ 13 C we measure on the CO 2 generated from more than 75% of our samples is compared to the δ 13 C measured on the AMS as a further check of our procedures. As further external checks, we analyzed the International Atomic Energy Association (IAEA) samples during the establishment of our laboratory and are presently participating in the third international radiocarbon intercalibration (TIRI) exercise.

Assessing the blank carbon contribution, isotope mass balance, and kinetic isotope fractionation of the Ramped Pyrolysis/Oxidation instrument at NOSAMS

We estimate the blank carbon mass over the course of a typical Ramped PyrOx (RPO) analysis (150–1000°C; 5°C×min –1 ) to be (3.7±0.6) μg C with an Fm value of 0.555±0.042 and a δ 13 C value of (–29.0±0.1) ‰ VPDB. Additionally, we provide equations for RPO Fm and δ 13 C blank corrections, including associated error propagation. By comparing RPO mass-weighted mean and independently measured bulk δ 13 C values for a compilation of environmental samples and standard reference materials (SRMs), we observe a small yet consistent 13 C depletion within the RPO instrument (mean–bulk: μ=–0.8‰; ±1σ=0.9‰; n =66). In contrast, because they are fractionation-corrected by definition, mass-weighted mean Fm values accurately match bulk measurements (mean–bulk: μ=0.005; ±1σ=0.014; n =36). Lastly, we show there exists no significant intra-sample δ 13 C variability across carbonate SRM peaks, indicating minimal mass-dependent kinetic isotope fractionation during RPO analysis. These data are best explained by a difference in activation energy between 13 C- and 12 C-containing compounds ( 13–12 ∆E ) of 0.3–1.8 J×mol –1 , indicating that blank and mass-balance corrected RPO δ 13 C values accurately retain carbon source isotope signals to within 1–2‰.

Oceanic uptake of CO2 re-estimated through δ13C in WOCE samples

In addition to 14C, a large set of δ13C data was produced at NOSAMS as part of the World ocean circulation experiment (WOCE). In this paper, a subset of 973 δ13C results from 63 stations in the Pacific Ocean was compared to a total number of 219 corresponding results from 12 stations sampled during oceanographic programs in the early 1970s. The data were analyzed in light of recent work to estimate the uptake of CO2 derived from fossil fuel and biomass burning in the oceans by quantifying the δ13C Suess effect in the oceans. In principle, the δ13C value of dissolved inorganic carbon (DIC) allows a quantitative estimate of how much of the anthropogenic CO2 released into the atmosphere is taken up by the oceans, because the δ13C of CO2 derived from organic matter (∼−27‰) is significantly different from that of the atmosphere (∼−8‰). Our new analysis indicates an apparent discrepancy between the old and the new data sets, possibly caused by a constant offset in δ13C values in a subset of the data. A similar offset was reported in an earlier work by Paul Quay et al. for one station that was not included in their final analysis. We present an estimate for this assumed offset based on data from water depths below which little or no change in δ13C over time would be expected. Such a correction leads to a significantly reduced estimate of the CO2 uptake, possibly as low as one half of the amount of 2.1 GtC yr−1 (gigatons carbon per year) estimated previously. The present conclusion is based on a comparison with a relatively small data set from the 70s in the Pacific Ocean. The larger data set collected during the GEOSECS program was not used because of problems reported with the data. This work suggests there may also be problems in comparing non-GEOSECS data from the 1970s to the current data. The calculation of significantly lower uptake estimates based on an offset-related problem appears valid, but the exact figures are tentative because the data set is small and the cause for an offset remains unknown. Therefore, it would be desirable to extend this comparison to the Indian Ocean where it is believed that better GEOSECS δ13C data are available.