Brett D. Walker

Radiocarbon in dissolved organic carbon of the Atlantic Ocean

PUBLICATIONS Geophysical Research Letters RESEARCH LETTER 10.1002/2016GL068746 Key Points: • More than 10% of DOC in the deep North Atlantic contains bomb C • DOC decreases in concentration and increases in C age along the deep ocean conveyor • DOC in the deep ocean is surprisingly dynamic both temporally and spatially Supporting Information: • Supporting Information S1 • Movie S1 Correspondence to: E. R. M. Druffel, edruffel@uci.edu Citation: Druffel, E. R. M., S. Griffin, A. I. Coppola, and B. D. Walker (2016), Radiocarbon in dissolved organic carbon of the Atlantic Ocean, Geophys. Res. Lett., 43, 5279–5286, doi:10.1002/2016GL068746. Received 6 JAN 2016 Accepted 12 MAY 2016 Accepted article online 16 MAY 2016 Published online 31 MAY 2016 Radiocarbon in dissolved organic carbon of the Atlantic Ocean E. R. M. Druffel 1 , S. Griffin 1 , A. I. Coppola 1,2 , and B. D. Walker 1 Department of Earth System Science, University of California, Irvine, Irvine, California, USA, 2 Department of Geography, University of Zurich, Zurich, Switzerland Abstract Marine dissolved organic carbon (DOC) is produced in the surface ocean though its radiocarbon ( 14 C) age in the deep ocean is thousands of years old. Here we show that ≥10% of the DOC in the deep North Atlantic is of postbomb origin and that the 14 C age of the prebomb DOC is ≥4900 14 C year, ~900 14 C year older than previous estimates. We report 14 C ages of DOC in the deep South Atlantic that are intermediate between values in the North Atlantic and the Southern Ocean. Finally, we conclude that prebomb DOC 14 C ages are older and a portion of deep DOC is more dynamic than previously reported. 1. Introduction Marine DOC is the largest pool of reduced carbon (C) in the oceans, about equal to the atmospheric CO 2 reservoir. Though most DOC is produced from photosynthetic uptake of modern DIC in the surface ocean, estimates of the bulk 14 C ages in deep open ocean DOC ranged from 4000 14 C years in the Sargasso Sea to 6000 14 C years in the North and South Pacific [Druffel and Griffin, 2015; Druffel et al., 1992; Williams and Druffel, 1987]. Previous work suggested that the average 14 C age of DOC in the deep Southern Ocean (5600 14 C years) was much closer to that in the deep Pacific, suggesting that the deep North Atlantic DOC contained bomb 14 C, that there was a source of old DOC to the Southern Ocean [Druffel and Bauer, 2000] or that there were diverse isotopic sources of DOC [Follett et al., 2014; Loh et al., 2004; McCarthy et al., 2011; Walker et al., 2011; Ziolkowski and Druffel, 2010]. In the ocean margins, there is evidence of inputs of old DOC in the deep northeast Pacific and the mid-Atlantic Bight off the U.S. coast [Bauer and Druffel, 1998; Ziolkowski and Druffel, 2010] and young DOC to the deep subpolar North Pacific [Tanaka et al., 2010]. Terrestrially derived DOC was found in the deep Arctic Ocean [Griffith et al., 2012]. We find here that the DOC Δ 14 C values in the deep Sargasso Sea in 2012 are lower than those in 1989, indicating that bomb 14 C levels have decreased over a period of two decades, similar to the decrease of surface dissolved inor- ganic C (DIC). Implications for the C cycle in the ocean include the presence of a labile pool of DOC in deep water. 2. Setting and Methods Water samples were collected from the North and South Atlantic Ocean during the Repeat Hydrography Climate Variability and Predictability program. Sampling included surface and subsurface water, northward Antarctic Intermediate Water (AAIW ~700–1200 m, low salinity, high silica) and Upper Circumpolar Deep Water (1000–2000 m), southward North Atlantic Deep Water (NADW 1500–4000 m, high oxygen, low silica), and Antarctic Bottom Water (within a few hundred meters of the bottom, cold and dense) [Jenkins et al., 2015a; Reid, 1989]. A data-constrained, ocean circulation model was used to show that in the South Atlantic, Antarctic water penetrates the NADW in volume-weighted averages that vary from 20 to 40% [DeVries and Primeau, 2011]. Radiocarbon in DOC was measured in seawater samples collected from three stations along 32°S on the A10 cruise in October 2011, four stations along 20°W on the A16N cruise in July/August 2013, and four stations along 65°W on the A22 cruise in March/April 2012 (Figure 1 and Table S1 in the supporting information). Samples shallower than 400 m were filtered using precombusted glass fiber (0.7 μM) filters, and all samples were collected in 1 L Amber Boston Round glass bottles and frozen at 20°C at an angle to avoid breakage until analysis at University of California, Irvine (UCI). Samples were diluted with 18.2 MΩ Milli-Q water (DOC concentration 0.5–0.9 μM), acidified to pH 2 with 85% phosphoric acid, purged with ultra high purity helium gas and UV-oxidized (UVox) [Beaupre et al., 2007; Druffel et al., 2013; Griffin et al., 2010]. Samples for DIC Δ 14 C analyses were prepared according to standard methods [McNichol et al., 1994]. ©2016. American Geophysical Union. All Rights Reserved. DRUFFEL ET AL. The resultant CO 2 from UVox was converted to graphite on iron catalyst for 14 C analysis at the Keck carbon cycle accelerator mass spectrometry (AMS) laboratory at UCI [Southon et al., 2004; Xu et al., 2007]. Total uncertainties for RADIOCARBON IN DOC OF THE ATLANTIC

Linked changes in marine dissolved organic carbon molecular size and radiocarbon age

Marine dissolved organic carbon (DOC) is a major global carbon reservoir, yet its cycling remains poorly understood. Previous work suggests DOC molecular size and chemical composition can significantly affect is bioavailability. Thus, DOC size and composition may control DOC cycling and radiocarbon age (via Δ14C). Here we show DOC molecular size is correlated to DOC Δ14C in the Pacific Ocean. Our results, based on a series of increasing molecular size fractions from three depths in the Pacific, show increasing DOC Δ14C with increasing molecular size. We use a size-age distribution model to predict the DOC and Δ14C of ultrafiltered DOC. The model predicts both large and small surface DOC with high Δ14C, and a narrow range (200-500 Da) of low Δ14C DOC. Deep model offsets suggest different size distributions and/or Δ14C sources at 670-915 m. Our results suggest molecular size and composition are linked to DOC reactivity and storage in the ocean.

High-precision measurement of phenylalanine δ15N values for environmental samples: a new approach coupling high-pressure liquid chromatography purification and elemental analyzer isotope ratio mass spectrometry.

RATIONALE Compound-specific isotope analysis of individual amino acids (CSI-AA) is a powerful new tool for tracing nitrogen (N) source and transformation in biogeochemical cycles. Specifically, the δ(15)N value of phenylalanine (δ(15)N(Phe)) represents an increasingly used proxy for source δ(15)N signatures, with particular promise for paleoceanographic applications. However, current derivatization/gas chromatography methods require expensive and relatively uncommon instrumentation, and have relatively low precision, making many potential applications impractical. METHODS A new offline approach has been developed for high-precision δ(15)N measurements of amino acids (δ(15)N(AA)), optimized for δ(15)N(Phe) values. Amino acids (AAs) are first purified via high-pressure liquid chromatography (HPLC), using a mixed-phase column and automated fraction collection. The δ(15)N values are determined via offline elemental analyzer-isotope ratio mass spectrometry (EA-IRMS). RESULTS The combined HPLC/EA-IRMS method separated most protein AAs with sufficient resolution to obtain accurate δ(15)N values, despite significant intra-peak isotopic fractionation. For δ(15)N(Phe) values, the precision was ±0.16‰ for standards, 4× better than gas chromatography/combustion/isotope ratio mass spectrometry (GC/C/IRMS; ±0.64‰). We also compared a δ(15)N(Phe) paleo-record from a deep-sea bamboo coral from Monterey Bay, CA, USA, using our method versus GC/C/IRMS. The two methods produced equivalent δ(15)N(Phe) values within error; however, the δ(15)N(Phe) values from HPLC/EA-IRMS had approximately twice the precision of GC/C/IRMS (average stdev of 0.27‰ ± 0.14‰ vs 0.60‰ ± 0.20‰, respectively). CONCLUSIONS These results demonstrate that offline HPLC represents a viable alternative to traditional GC/C/IMRS for δ(15)N(AA) measurement. HPLC/EA-IRMS is more precise and widely available, and therefore useful in applications requiring increased precision for data interpretation (e.g. δ(15)N paleoproxies).

EFFECT OF ACIDIFIED VERSUS FROZEN STORAGE ON MARINE DISSOLVED ORGANIC CARBON CONCENTRATION AND ISOTOPIC COMPOSITION

Author(s): Walker, Brett D; Griffin, Sheila; Druffel, Ellen RM | Abstract: AbstractThe standard procedure for storing/preserving seawater dissolved organic carbon (DOC) samples after field collection is by freezing (–20°C) until future analysis can be made. However, shipping and receiving large numbers of these samples without thawing presents a significant logistical problem and large monetary expense. Access to freezers can also be limited in remote field locations. We therefore test an alternative method of preserving and storing samples for the measurement of DOC concentrations ([DOC]), stable carbon (δ13C), and radiocarbon (as ∆14C) isotopic values via UV photooxidation (UVox). We report a total analytical reproducibility of frozen DOC samples to be [DOC]±1.3 µM, ∆14C±9.4‰, and δ13C±0.1‰, comparable to previously reported results (Druffel et al. 2013). Open Ocean DOC frozen versus acidified duplicates were on average offset by ∆DOC±1.1 µM, ∆∆14C± –1.3‰, and ∆δ13C± –0.1‰. Coastal Ocean frozen vs. acidified sample replicates, collected as part of a long-term (380-day) storage experiment, had larger, albeit consistent offsets of ∆DOC±2.2 µM, ∆∆14C±1.5‰, and ∆δ13C± –0.2‰. A simple isotopic mass balance of changes in [DOC], ∆14C, and δ13C values reveals loss of semi-labile DOC (2.2±0.6 µM, ∆14C=–94±105‰, δ13C=–27±10‰; n=4) and semi-recalcitrant DOC (2.4±0.7 µM, ∆14C=–478±116‰, δ13C=–23.4±3.0‰; n=3) in Coastal and Open Ocean acidified samples, respectively.

Radiocarbon Analysis of Individual Amino Acids: Carbon Blank Quantification for a Small-Sample High-Pressure Liquid Chromatography Purification Method

Compound-specific radiocarbon analysis (CSRA) of amino acids (AAs) is of great interest as a proxy for organic nitrogen (N) cycling rates, dating archeological bone collagen, and investigating processes shaping the biogeochemistry of global N reservoirs. However, recoverable quantities of individual compounds from natural samples are often insufficient for radiocarbon ((14)C) analyses (<50 μg C). Constraining procedural carbon (C) blanks and their isotopic contributions is critical for reporting of accurate CSRA measurements. Here, we report the first detailed quantification of C blanks (including sources, magnitudes, and variability) for a high-pressure liquid chromatography (HPLC) method designed to purify individual AAs from natural samples. We used pairs of AA standards with either modern (M) or dead (D) fraction modern (Fm) values to quantify MC and DC blanks within several chromatographic regions. Blanks were determined for both individual and mixed AA standard injections with peak loadings ranging from 10 to 85 μg C. We found 0.8 ± 0.4 μg of MC and 1.0 ± 0.5 μg of DC were introduced by downstream sample preparation (drying, combustion, and graphitization), which accounted for essentially the entire procedural blank for early eluting AAs. For late-eluting AAs, higher eluent organic content and fraction collected volumes contributed to total blanks of 1.5 ± 0.75 μg of MC and 3.0 ± 1.5 μg of DC. Our final measurement uncertainty for 20 μg of C of most AAs was ±0.02 Fm, although sample size requirements are larger for similar uncertainty in late-eluting AAs. These results demonstrate the first CSRA protocol for many protein AAs with uncertainties comparable to the lowest achieved in prior studies.

Radiocarbon in dissolved organic and inorganic carbon of the Arctic Ocean

PUBLICATIONS Geophysical Research Letters RESEARCH LETTER 10.1002/2016GL072138 Key Points: • DOC in the deep Eurasian Basin appears to contain at least 8% bomb C • Modern DOC appears to be selectively lost in the Beaufort Sea and slope water • A relationship between DOC Δ C and total hydrolysable amino acids in the Beaufort suggests that chemical composition controls DOC cycling Supporting Information: • Supporting Information S1 Correspondence to: E. R. M. Druffel, edruffel@uci.edu Citation: Druffel, E. R. M., S. Griffin, C. S. Glynn, R. Benner, and B. D. Walker (2017), Radiocarbon in dissolved organic and inorganic carbon of the Arctic Ocean, Geophys. Res. Lett., 44, doi:10.1002/ 2016GL072138. Received 28 NOV 2016 Accepted 13 FEB 2017 Accepted article online 15 FEB 2017 Radiocarbon in dissolved organic and inorganic carbon of the Arctic Ocean E. R. M. Druffel 1 , S. Griffin 1 , C. S. Glynn 1 , R. Benner 2 , and B. D. Walker 1 Department of Earth System Science, University of California, Irvine, California, USA, 2 Marine Science Program, University of South Carolina, Columbia, South Carolina, USA Abstract Dissolved organic carbon (DOC) in the ocean is thousands of 14 C years old, yet a portion of the DOC cycles on much shorter time scales (days to decades). We present 14 C measurements of DOC in the Arctic Ocean and estimate that ≥8% of the DOC in the deep Eurasian Basin contains bomb 14 C. While this is a limited data set, there appears to be selective loss of modern DOC in the surface and halocline waters of the open Beaufort Sea versus the Beaufort slope. At one of the Beaufort Sea stations, there is a linear relationship between DOC Δ 14 C values and previously measured total hydrolysable amino acid concentrations as reported by Shen et al. (2012), indicating that deep DOC contains small amounts of bioavailable DOC. The 14 C data show that not all of the deep DOC is recalcitrant. 1. Introduction The Arctic Ocean has a large continental shelf region and the largest input of freshwater per unit volume of any ocean. Dissolved organic carbon (DOC) in surface waters of the Arctic is the highest of any ocean, due to the large flux of terrigenous DOC from Arctic rivers [Amon et al., 2012; Benner et al., 2005; Hansell et al., 2004]. This flux of DOC is expected to increase as the Earth continues to warm, caus- ing release of organic C from expansive Arctic permafrost watersheds [Anderson and Amon, 2015; Benner et al., 2004]. There are four main water masses in the Arctic Ocean: (1) polar surface water, (2) halocline water (30–200 m), (3) Atlantic derived water (200–800 m), and (4) deep Atlantic water. Polar surface water is cold and fresh, having been derived from Arctic rivers and sea ice melt. The halocline water is formed on the continental shelves from freezing and brine release and is laterally advected into the Arctic Basin [Aagaard et al., 1981]. Water from the Pacific Ocean flows into the Beaufort Sea between 75 and 100 m depth (Upper Halocline Water), while water from the Atlantic Ocean enters from the Fram Strait and Barents Sea (Lower Halocline Water). The Beaufort Sea (in the southern Canadian Basin) and Eurasian Basin are separated by the Lomonosov Ridge whose sill depth is about 1500 m, keeping the deep waters of the Canada Basin and Beaufort Sea relatively isolated [Ostlund et al., 1987] (Figure 1). Radiocarbon provides quantitative information on the time scale of DOC cycling in the global ocean. Though the existing data are sparse, the available 14 C ages of DOC reveal that the difference between the average, prebomb ( 1000 m depth) and slope (bottom depth from 100 to 1000 m) and the Eurasian Basin to evaluate the distribution and cycling of DOC in the Arctic Ocean. We estimate that ≥8% of the DOC in the Eurasian Basin is of postbomb origin. We find that DOC concentrations and Δ 14 C values in the upper Beaufort Sea and slope are linearly correlated with salinity. Samples from one station in the Beaufort Sea has total hydrolysable amino acid measurements [Shen et al., 2012] that are strongly correlated with DOC Δ 14 C values, which indicates that deep DOC contains little bioavailable DOC. RADIOCARBON IN THE ARCTIC OCEAN

Direct Visualization of Individual Aromatic Compound Structures in Low Molecular Weight Marine Dissolved Organic Carbon

Dissolved organic carbon (DOC) is the largest pool of exchangeable organic carbon in the ocean. However, less than 10% of DOC has been molecularly characterized in the deep ocean to understand DOC’s recalcitrance. Here we analyze the radiocarbon (14C) depleted, and presumably refractory, low molecular weight (LMW) DOC from the North Central Pacific using atomic force microscopy to produce the first atomic-resolution images of individual LMW DOC molecules. We evaluate surface and deep LMW DOC chemical structures in the context of their relative persistence and recalcitrance. Atomic force microscopy resolved planar structures with features similar to polycyclic aromatic compounds and carboxylic-rich alicyclic structures with less than five aromatic carbon rings. These compounds comprise 8% and 20% of the measurable molecules investigated in the surface and deep, respectively. Resolving the structures of individual DOC molecules represents a step forward in molecular characterization of DOC and in understanding its long-term stability.