University of Southern California Dissertations and Theses

Marine biogenic halocarbons: potential for heterotrophic bacterial production and seasonality at San Pedro Ocean Time-series

(USC Thesis Other)

Marine biogenic halocarbons: potential for heterotrophic bacterial production and seasonality at San Pedro Ocean Time-series

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Marine biogenic halocarbons: potential for heterotrophic bacterial production and seasonality at San Pedro Ocean Time-series Nick J. Klein PhD Dissertation in Ocean Sciences University of Southern California Faculty of the USC Graduate School Sergio Sañudo-Wilhelmy, adviser 14 December 2016 Klein 1 Table of Contents Acknowledgements 2 Chapter 1: Introduction – The Atmospheric Role of Marine-Derived 3 Halocarbons Chapter 2: The potential for marine heterotrophic bacteria as novel and 21 globally significant halocarbon gas producers Chapter 3: Seasonal and ecological dynamics of halocarbon 42 concentrations and sea-air fluxes at the San Pedro Ocean Time-series Chapter 4: Heterotrophic bacterial production and seasonality 68 of marine biogenic halocarbons: implications and directions for future research References 77 Appendix 1: Regression modeling of a North East Atlantic 85 Spring Bloom transect dataset suggests a previously unrecognized biological role for Mo and V. (Published in Frontiers in Microbiology) Appendix 2: Spatial distribution of dissolved B-vitamins and 109 trace metals and their potential for biological limitation in Lake Michigan Klein 2 Acknowledgements This dissertation represents a journey that began studying trace metals and B-vitamins in lakes (included as appendices). The insights and experiences from that earlier work piqued my interest in marine biogenic halocarbons. Over the last several years I have had the opportunity to work with a variety of laboratory techniques, including microbial culturing and trace chemical analysis. I believe the results presented here offer exciting new insights into the biological production and global significance of these climate-relevant gases. I would like to thank my advisor, Sergio Sañudo-Wilhelmy, for his patient guidance and support, without which this dissertation would not have been possible. I am grateful to my thesis committee members Doug Hammond, Eric Webb, and Josh West for their time and input in composing and revising this manuscript. I have been fortunate to have Lynda Cutter as our lab technician as her expertise was invaluable, particularly in setting up the GC method. And last but by no means least, I would like to thank my family and friends. Klein 3 Chapter 1: Introduction – The Atmospheric Role of Marine-Derived Halocarbons The world’s oceans are both the largest reservoir of halogens and also a net source of them into the atmosphere, where they play important roles in both stratospheric and tropospheric chemistry (Gebhardt 2008, Read et al. 2008, Sturges et al. 2000). Considerable scientific effort has been spent attempting to understand the atmospheric chemistry of anthropogenic chlorofluorocarbons (CFCs), which have long been known to destroy the planet’s protective layer of stratospheric ozone. However, we know relatively little about their short-lived natural analogues, which are mainly from marine sources (Table 1) and largely of biogenic origin (WMO, 2007a). Volatile halogenated hydrocarbons (“halocarbons”) of the general form CH 4-N X N (where X is Cl, Br, or I and N is 1, 2, or 3) participate in a variety of climate-relevant atmospheric chemistry. Being far shorter lived (less than one year; Table 1) than anthropogenic CFCs (whose atmospheric lifetime is on the order of 45-100 years depending on the species), the proportion of these marine biogenic halocarbons (MBHs) that are transported into the stratosphere is relatively small. However, they are correspondingly much more reactive and destructive to ozone on a per-molecule basis than CFCs; for example, each atom of bromine transported to the stratosphere has approximately sixty times more ozone destructive potential than chlorine (WMO, 2007a) and iodine is between 150 and 300 times more powerful (WMO, 2003). The role of short-lived halocarbons is significant and will increase in the near future. Due to international regulatory agreements such as the Montreal Protocol effectively banning usage of most CFCs, the anthropogenic contribution to stratospheric ozone destruction is shrinking relative to natural biological sources. While short-lived halocarbons currently account for around 25% of halogen-mediated ozone depletion, that proportion is predicted to increase to 50% by the year 2050 as levels of manmade CFCs continue to decline (WMO, 2007b). Klein 4 In addition to direct destruction of Earth’s protective ozone layer in the stratosphere, MBHs also play important roles in the tropospheric ozone balance and marine boundary layer chemistry, as summarized in Figure 1. Tropospheric ozone is a pollutant, a human health hazard, and has significant anthropogenic sources, but also mediates the oxidative capacity of Earth’s atmosphere. In fact, tropospheric ozone-mediated oxidation (via production of OH radicals) is the major sink for several other climactically important gases, oxidizing 83% of the greenhouse gas methane (CH 4 ) and 78-96% of carbon monoxide (CO) (Jacob 1999). Similar to ozone in the stratosphere, tropospheric ozone is destroyed by halocarbons. Studies of ozone depletion in the troposphere over the open ocean attribute more than half the rate of ozone destruction to volatile halocarbons (Read et al., 2008). Therefore, by destroying ozone and decreasing the troposphere’s capacity to remove gases such as methane. Thus, halocarbons have a climate feedback role in indirectly affecting the residence time and climate forcing of a major greenhouse gas. Halocarbons are involved in cloud formation, as iodocarbons can polymerize and form condensation nuclei (von Glasgow et al., 2004). Paradoxically, they also interfere with another major cloud-nucleating process involving dimethylsulfoniopropionate (DMSP) (Read et al. 2008). Though cloud nucleation in the marine troposphere is now understood to involve additional complex processes, the role of DMSP is significant and was long believed to constitute a major climate feedback mechanism (Quinn and Bates, 2011). DMSP is a compound produced by several major phytoplankton taxa whose metabolite dimethyl sulfide (DMS) is the most abundant volatile sulfur species in the marine boundary layer. DMS in turn enters the atmosphere as either DMSO (dimethylsulfoxide) or the nuclei-forming SO 2 (sulfur dioxide). The Klein 5 strong oxidizing effect of volatile halocarbons favors the formation of DMSO and causes a net reduction in the stimulated cloud albedo (Read et al. 2008, von Glasgow et al. 2004). Generally, global distributions show halocarbon concentrations highest near areas associated with phytoplankton blooms (Roy et al. 2011), upwelling regions, coastal zones, and oxygen minimum zones (Roy 2010, Roy et al. 2011, Yamamoto et al. 2001) especially during the summer (Butler et al, 2007). Relative concentrations of the various halocarbon species also vary dependent on their reactivity and atmospheric transport patterns, with methyl iodide strongest near the equator and polybrominated compounds generally exhibiting their peak concentrations in mid-tropical and subtropical regions (Butler et al., 2007). Based on studies of the geographic distribution of halocarbons which find a broad correlation between increased sea- surface temperatures and higher halocarbon concentrations, it has been hypothesized that those concentrations will increase with elevated global temperatures due to anthropogenic climate change (Gebhardt 2008). Oceanic oxygen-minimum zones have been expanding in recent decades (Capone and Hutchins 2013, Gilly et al. 2013), and owing to the high observed halocarbon fluxes over these features likely represent a significant and growing source. Despite the importance of marine biogenic halocarbons to the chemistry of several complex and climate-relevant stratospheric and tropospheric systems, relatively little scientific effort has been invested in studying the organisms and environmental conditions relevant to their biosynthesis of these compounds. Research has described halocarbon geographic distribution, atmospheric chemistry, and identified some of the taxa responsible for their biosynthesis (de la Cuesta and Manley, 2009; Hill and Manley, 2009; Johnson et al, 2011; Küpper et al., 1998; Moore et al., 2006; Scarratt and Moore, 1996 and 1997; Sturges and Cota, 1997; Sturges et al., 1992, Urhahn and Ballschmiter 1998). It has partially elucidated the biochemistry of their Klein 6 production (Butler and Carter-Franklin 2004, Itoh et al. 1997). However, most field measurements are only of sea-air fluxes and not depth profiles, no time-series studies exist, and production of halocarbons has not been surveyed in many important marine taxonomic groups, notably heterotrophic bacteria. This thesis seeks to characterize halocarbon synthesis by marine heterotrophic bacteria and better elucidate the seasonal ecological dynamics of biological halocarbon production, allowing for more accurate quantification of global halocarbon sources and sinks and a better understanding of how patterns of halocarbon may shift in tandem with anthropogenic climate change in the coming decades. Biosynthesis of Monohalomethanes Singly halogenated methanes (methyl halides or monohalomethanes) are produced both biologically by the action of methyltransferase enzymes and by abiotic processes (Moore and Zafiriou 1992, Moore 2008). Methyltransferases, a large and diverse family of enzymes, have been shown to catalyze methylation of halide ions (Amachi et al, 2001; Attieh et al., 1995; Coulter et al., 1999) using S-adenosylmethionine (SAM) as the methyl donor. Abiotic photochemical processes are also responsible for some of the CH 3 I and CH 3 Br production via a radical mechanism. Near-surface, UV light produces halogen radicals, which react with dissolved organic carbon (DOC) to produce methyl halides (Happell and Wallace, 1996; Moore and Zafiriou, 1994; Moore, 2008). It is more difficult to address potential environmental controls on biomethylation of halides since the methyltransferase enzymes are neither a uniform group nor well understood in this context. That is, a particular enzyme within the methyltransferase family has not been identified as specifically responsible for methyl halide production like vanadium haloperoxidases (VHPOs) have been for the polyhalocarbons. Indeed, Manley (2002) concludes that methyl Klein 7 halide production is likely a biochemical “accident” caused by some methyltransferase enzymes’ lack of substrate specificity. However, it can be assumed that SAM as the universal biological methyl donor and cofactors required for SAM biosynthesis (vitamin B 12 , cobalt) are likely relevant to biogenic production of methyl halides. In all observed cases of marine biological production of methyl halides, SAM was found to act as the methyl donor (Amachi et al, 2001; Baker et al., 1999; Itoh et al., 1997; Scarratt and Moore, 1996 and 1997). SAM donates a methyl group to a halide ion, producing S-adenosylhomocysteine (SAH) and a methyl halide: SAM + X -  SAH + CH 3 X SAH is then regenerated back into SAM using the cobalt-containing cofactor vitamin B 12 (cobalamin)(Baker et al. 1999). There are robust field data supporting biological production of methyl halides. Gebhardt (2008) found strong correlations between CH 3 I, DMS (a proxy for marine biological productivity), and satellite-derived chlorophyll-a throughout a research cruise from the Amazon plume to the Southern Ocean, suggesting that biosynthesis of methyl halides is linked to photosynthetic activity in the field. I hypothesize that since methylation is a ubiquitous and basic biochemical reaction, if methyl halides are synthesized by nonspecific methyltransferases, production should be taxonomically widespread as well as potentially related to availability of the cofactors necessary for synthesis of the methyl donor SAM (Co and B 12 ). Biosynthesis of Polyhalomethanes: Haloperoxidases The first conclusive studies demonstrating marine biogenic production of polyhalomethanes (those containing multiple halogen atoms, e.g. CH 2 I 2 , CHBr 3 ) can be traced back to the 1970s (Moore 1977). During that period, however, the consensus was that marine halocarbons were chiefly a product of photochemistry or anthropogenic sources (such as water disinfectants and industry) and biogenic sources were relatively insignificant (Helz and Klein 8 Hsu 1978). Berg et al (1983) observed a large, unexpected flux of bromomethanes from the Arctic and suggested macroalgae as a potential biological source. Sturges et al (1992) would later demonstrate for the first time production of bromoform (CHBr 3 ) by sea ice diatoms (chiefly P. glacialis, a pennate diatom) and concluded that while macroalgal synthesis of halocarbons might be regionally important, only production by phytoplankton could account for the observed Arctic bromomethane concentrations. More recent research has identified the haloperoxidase family of enzymes as likely being responsible for biological production of polyhalomethanes. Haloperoxidases catalyze oxidation of halide ions (I - , Br - , or Cl - ) by hydrogen peroxide according to the following general reaction: X - + H 2 O 2  XO - + H 2 O The XO - intermediates are highly reactive and attack methyl groups activated by an adjacent carbonyl or methylsulfone group, forming di and tri-halogenated methanes via the haloform reaction (Gebhardt 2008). XO - and its equilibrium form HOX are both lipophilic and easily diffuse through the cell membranes of organisms (Hill and Manley, 2009) where they react rapidly with dissolved organic matter. Urhahn and Ballschmiter (1998) have suggested that various methyl sulfur compounds present in seawater such as methionine, SAM, DMSP, and DMSO may be the primary carbon sources for polyhalomethane biosynthesis (see summary flowchart of halocarbon production and hypothesized relevant environmental processes and factors in Figure 2). Haloperoxidases isolated from marine organisms are one of only two known enzymes that utilize vanadium in their active site, the other being an unusual V-containing nitrogenase observed in a few soil bacteria (Crans et al 2004). These vanadium haloperoxidases (VHPOs) are capable of catalyzing the peroxidation of both bromide and iodide, but not chloride. Klein 9 To date, VHPO activity has been observed in many macroalgae, several species of diatoms and Phaeocystis (Küpper et al., 1998; de la Cuesta and Manley, 2009; Hill and Manley, 2009; Moore et al., 2006; Scarratt and Moore, 1996 and 1997; Sturges and Costa, 1997; Sturges et al., 1992, Urhahn and Ballschmiter 1998), and recently in the non-diazotrophic cyanobacterial strain Synechococcus CC9311 (Johnson et al, 2011). However, halocarbon production and haloperoxidase activity has not been evaluated for most marine bacteria, despite their ubiquity (Molloy 2012). A cladogram of genes from marine microbes with both high sequence homology to the Synechococcus CC9311 enzyme and a conserved active site structure (an identical binding “pocket” for the metal cofactor) is presented in Figure 3. While several cyanobacteria have genes closely related to the functional Synechococcus CC9311 haloperoxidase enzyme and to that of the purified VHPO from the red alga Corallina officinalis, they are few in number and their taxonomic distribution appears stochastic. It has been suggested that this may be due to a recent horizontal gene transfer, and thus distribution of haloperoxidases in cyanobacteria may be very species and strain-dependent (Johnson et al., 2011). There were no hits found for potential haloperoxidase genes in any of the heterotrophic bacteria in the database, suggesting either a lack of VHPOs or incorrect or missing gene annotations. A high active-site sequence homology to VHPOs and limited haloperoxidase activity have been observed in the widely taxonomically distributed acid phosphatase enzymes (Tanaka et al. 2002), suggesting biosynthesis by acid phosphatases might be a plausible mechanism if heterotrophic bacteria truly lack VHPOs. The biological function of haloperoxidases is uncertain. It has been suggested they play roles in osmoregulation as a means to excrete excess halides (Ni and Hager 1999), chemical defense against grazing and antibacterial activity (Butler and Carter-Franklin, 2004; Gschwend, Klein 10 1985; Manley 2002), or protection against the oxidative stress experienced by all organisms with aerobic metabolisms (Hill and Manley, 2009; Manley 2002). The last of these seems to be the most well-supported hypothesis. VHPOs have been demonstrated to have an antioxidant function in macroalgae (Küpper et al. 1998). Many species of macroalgae have an active iodine metabolism (Kuwabara and North 1980), requiring the element for growth and using haloperoxidases to scavenge cytotoxic H 2 O 2 , releasing significant HOI under oxidative stress (Küpper et al., 1998). Similarly, diatom cultures have been shown to reduce iodate (Wong et al., 2002), take up iodide and iodate (de la Cuesta and Manley, 2009), and release HOI and HOBr (Hill and Manley, 2009). Lending further support to a role for VHPOs in managing oxidative stress are several sets of field data which indicate an unknown but significant biological role for vanadium in phytoplankton. Data from an oceanographic cruise through the North Atlantic Spring Bloom (Klein et al, 2013; Appendix 1) showed a strong correlation between intracellular vanadium concentrations, chlorophyll, and biogenic silica (Figures 4a & 4b). This suggests a relationship between vanadium uptake, photosynthesis, and diatom abundance. As molybdenum and vanadium exist in seawater in chemically analogous forms (as oxyanions molybdate and vanadate), the observed correlations with vanadium but not molybdate imply active, specific V uptake by photosynthetic organisms and diatoms (Crans et al, 2004). Two recent studies advanced a similar hypothesis and found high intracellular vanadium concentrations in the important nitrogen-fixing cyanobacteria Trichodesmium (Nuester et al, 2012; Tovar-Sanchez and Sañudo-Wilhelmy, 2011), with vanadium actually being the most abundant intracellular metal in the former study. Most other enzymes responsible for scavenging hydrogen peroxide (catalase, for example) require Fe-heme cofactors, and phytoplankton possess Klein 11 a small inventory of such enzymes (Manley 2002). It is possible that VHPOs represent a non-Fe alternative pathway for scavenging of reactive oxygen species in marine microbes, and may be relatively more important under Fe-deplete conditions. The San Pedro Ocean Time-series The San Pedro Ocean Time-series (SPOT) is in many ways an ideal location to study any seasonal and ecological patterns in biological halocarbon production. Located in the San Pedro Basin off the coast of Los Angeles, this 890m depth profile station has a pronounced oxygen minimum zone and is hypoxic at depth while also experiencing a strong spring upwelling season (Countway et al. 2010). The station is sampled monthly for a wide variety of chemical and biological data, including over a decade of microbial and eukaryotic abundance and diversity measurements, which aid in interpretation of observed halocarbon concentration and flux seasonality. The dominant species during upwelling and the relatively oligotrophic summer, fall, and winter seasons have been observed to be annually repeatable (Countway et al. 2010, Fuhrman et al. 2006) and all of the taxa with observed halocarbon production are represented at SPOT. During upwelling, enhanced productivity supports a diverse assemblage of diatoms, chiefly Nitzschia and Pseudonitzschia with a variety of other taxa in lesser abundance (Countway et al. 2010). The remainder of the year, Phaeocystis and chlorophytes are relatively more dominant eukaryotic phytoplankton (Table 2). The strain of Synechococcus with observed haloperoxidase activity was isolated from the California current (Johnson et al. 2011) and is ubiquitous throughout the year at SPOT. During the summer season, Synechococcus is typically the dominant phytoplankton along with Prochlorococcus (a demonstrated CH 3 I producer) (Brownell et al. 2010). The combination of a microbial community dominated by many of the few organisms with reported halocarbon production along with a predictable ecological Klein 12 seasonality allow for a robust comparison of halocarbon concentrations and fluxes with seasonal trends in SPOT biological and environmental conditions. Objectives and Hypotheses: A better understanding of the dynamics of marine biogenic halide production is necessary, as they play important roles in both stratospheric and atmospheric chemistry and will increase in relevance in the coming decades. Previous research has largely focused on the contribution of haloperoxidase-containing diatoms and macroalgae and quantifying global sea- air fluxes, while potential biosynthesis by ubiquitous marine heterotrophic bacteria and theseasonality of their production are unexplored. This thesis will aim generally to address the last two topics in order to better understand dynamics of these climate-relevant gases both in the present and in relation to anthropogenic climate change in the near-future. In this thesis I will: 1. Present culture assays of marine heterotrophic bacteria for halocarbon biosynthesis. No studies have yet addressed the potential for production of halocarbons by marine heterotrophic bacteria. If present, these could constitute a significant and previously unaccounted-for fraction of global halocarbon biosynthesis. Culture assays will test the following general hypotheses:  Heterotrophs are not likely to produce polyhalocarbons as the haloperoxidase gene has only been found in photosynthetic organisms and a search of the Integrated Microbial Genome database yielded no likely haloperoxidase genes from heterotrophic genomes, only from several cyanobacterial strains (Figure 3). However, these annotations may not be complete, and production might also be due Klein 13 to acid phosphatases, which have similar active site structure to VHPOs and demonstrated haloperoxidase activity (Tanaka et al. 2002).  Marine heterotrophs may produce methyl halides via methyltransferases. This biosynthesis would be expected to be taxonomically widespread as methyltransferases are diverse and ubiquitous enzymes. 2. Conduct monthly depth profiling to determine halocarbon concentration and fluxes at SPOT for one year alongside ancillary chemical and biological measurements to explore seasonal dynamics of biogenic marine halocarbon production.  As widely observed in previous studies, marine halocarbons such as CH 3 Cl, CH 3 Br, CH 3 I, CH 2 Br 2 , and CHBr 3 are primarily biogenic in origin, and the haloperoxidase products CH 2 Br 2 and CHBr 3 will be associated spatially and temporally with photosynthetic eukaryotes at the chlorophyll maximum, increasing with enhanced productivity in the spring upwelling season. CH 3 Br and CH 3 I have abiotic contributions from UV photolysis and will exhibit near-surface maxima. Any generalized heterotrophic production might also be chiefly associated with the chlorophyll maximum; otherwise it might be maximal in the surface mixed layer, where bacterial abundances and activity are highest.  Distributions might correspond with succession and seasonal dominance of SPOT taxa with reported halocarbon biosynthesis, e.g. diatoms, Phaeocystis, Synechococcus, Prochlorococcus. Diatoms in particular are expected to be the dominant sources of haloperoxidase activity and polyhalocarbon production due to their high observed production rates. If any heterotrophic production is species- Klein 14 specific, numerous ecotypes of marine heterotrophs have predictable seasonal patterns at SPOT. 3. Synthesize these results with existing literature to make tentative predictions for near-future trends in global biogenic halocarbon production and assess needs for future research.  Halocarbon biosynthesis may increase generally due to climate change as it appears positively correlated with sea-surface temperature in previous studies.  Observed SPOT trends in halocarbons and seasonal ecology will allow us to predict how ecosystem shifts due to climate change may affect global halocarbon production.  Any production of halocarbons by heterotrophic bacteria might represent a significant and previously uncharacterized contribution.  Changes in upwelling, expansion of oxygen minimum zones, and broad ecological shifts may favor heterotrophic bacteria relative to eukaryotes with demonstrated halocarbon such as diatoms. Klein 15 Gas Marine Source Strength (Gg / year) Anthropogenic Source Strength (Gg / year) Total Flux (Gg / year) % Marine- derived Lifetime (days) CH 3 Cl 445 - 940 355 - 1420 1743 - 13578 7 - 26% 365 CH 3 Br 30 - 148 44 - 125 77 - 293 39 - 51% 255 CHCl 3 270 - 450 45 - 91 446 - 880 51 - 61% 150 CH 2 Br 2 304 0 304 100% 120 CHBr 3 843 29 862 98% 26 CH 3 I 613 16-29 670 - 694 88 - 91% 7 Table 1: Estimated ranges of source fluxes of major marine biogenic halocarbons, percentage marine contribution to global source, and atmospheric lifetime. Data from that compiled in Gebhardt (2008). Klein 16 Figure 1. Conceptual diagram of atmospheric interactions of marine-derived halocarbons. Halocarbons produced by biology diffuse into the marine boundary layer and interact with the DMS cycle (which is important in cloud formation). Iodocarbons themselves are also capable of nucleating clouds. Halocarbons deplete both stratospheric and tropospheric ozone, in the latter case reducing the ozone-mediated oxidation rate of gases such as methane (CH 4 ) and carbon monoxide (CO). O 3 + CH 4 or CO --> CO 2 Destruction of ozone layer in stratosphere Halocarbons Destruction of tropospheric ozone Interaction with DMS & sulfur cycle DMS Cloud nucleation Biological Processes Klein 17 Figure 2. Flowchart diagram of the hypothesized relationships between environmental factors (chiefly water chemistry), biochemical reactions within organisms, and the halocarbon end products. Atmospheric Chemistry B 12 , Met, Co, SAM V, Fe, oxidative stressors DOC, UV light CH 3 Cl, CH 3 Br, CH 3 I CHBr 3 -> CH 2 Br 2 , CH 2 I 2 Chiefly CH 3 Cl, CH 3 Br Biomethylation Haloperoxidase enzymes Water chemistry Biological Processes Abiotic synthesis Klein 18 Figure 3. Cladogram showing distance (amino acid BLAST-P versus the Integrated Microbial Genome database) from Synechococcus CC9311 VHPO enzyme. Corallina officinalis (labeled with blue dot) is a coralline red algae whose VHPO enzyme has been purified and characterized extensively. Green dots mark other cyanobacterial species. Klein 19 Biogenic silica ( mol L -1 ) 0.0 0.2 0.4 0.6 0.8 1.0 1.2 Intracellular metal (mmol mol P -1 ) 0.0 0.2 0.4 0.6 0.8 1.0 Mo V Chlorophyll a ( g L -1 ) 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 Mo V R 2 = 0.72 p = 2.7 x 10 -6 R 2 = 0.49 p = 0.00024 Figure 4a: P-standardized intracellular vanadium and molybdenum v. biological silica concentrations (left), and 4b: P-standardized intracellular vanadium and molybdenum v. chlorophyll a concentrations (right) from all n=27 surface stations along the 2005 North Atlantic Spring Bloom transect (Klein et al, 2013). Regressions shown with V are significant at p<0.05. Klein 20 Upwelling Summer (July) – Fall (October) – Winter (January) Typically Apr-Jun, though as early as March or late as July Surface (5m) Eukaryotes: Gyrodinium (dinoflagellate), ciliates. Highly diverse assemblage of diatoms, stramenopiles. Chlorophytes. Bacteria and archaea: gamma- Proteobacteria (including SAR11), Roseobacter, Prochlorococcus, Alteromonas, Bacteroidetes, Verrumicrobiales, Synechococcus Eukaryotes: Lingulodinium (dinoflagellate), ciliates. Chlorophytes. Diatoms and stramenopiles. Phaeocystis. Bacteria and archaea: gamma- Proteobacteria (including SAR11), Roseobacter, Prochlorococcus , Verrumicrobiales, Synechococcus Chlorophyll max ca 25-40m Eukaryotes: Gyrodinium (dinoflagellate), ciliates. Highly diverse assemblage of diatoms, stramenopiles. Bacteria and archaea: gamma-Proteobacteria (including SAR11), Roseobacter, Prochlorococcus, Verrumicrobiales, Synechococcus Eukaryotes: Lingulodinium (dinoflagellate), ciliates. Chlorophytes. Diatoms and stramenopiles. Phaeocystis. Bacteria and archaea: gamma- Proteobacteria (including SAR11), Roseobacter, Prochlorococcus, Verrumicrobiales, Synechococcus Mid-depths (150m) Eukaryotes: Gyrodinium (dinoflagellate), ciliates. Highly diverse assemblage of diatoms, stramenopiles. Bacteria and archaea: gamma-Proteobacteria, Rickettsiales, Rhodospirales, Roseobacter, Vibrio, Verrumicrobiales, AOA groups A and B Eukaryotes: Gyrodinium (dinoflagellate), ciliates. Diatoms, stramenopiles. Bacteria and archaea: gamma-Proteobacteria, Rickettsiales, Rhodospirales, Roseobacter, Vibrio, Verrumicrobiales, AOA groups A and B Bottom waters (500m) Eukaryotes: Gyrodinium (dinoflagellate), ciliates. Bacteria and archaea: gamma- Proteobacteria, Rickettsiales, Rhodospirales, Roseobacter, Vibrio, Verrumicrobiales, AOA group B Eukaryotes: Gyrodinium (dinoflagellate), ciliates. Bacteria and archaea: gamma- Proteobacteria, Rickettsiales, Rhodospirales, Roseobacter, Vibrio, Verrumicrobiales, AOA group B Table 2. Dominant eukaryotic, bacterial, and archaeal taxa at SPOT. Listed approximately in order of abundance. Data compiled from Countway et al. 2010, Fuhrman et al. 2006). Klein 21 Chapter 2: The potential for marine heterotrophic bacteria as novel and globally significant halocarbon gas producers Foreword and Acknowledgements This second chapter presents a survey of halocarbon production in laboratory cultures of heterotrophic marine bacteria. After initially becoming interested in marine biogenic halocarbons, I attempted several different culture experiments to confirm I could detect halocarbon production. My previous research involved trace metal and organic cofactors (vitamins), so I was particularly interested in eventually manipulating availability of trace nutrients to determine which are relevant to halocarbon biosynthesis (V, Fe, vitamin B 12 , etc). I grew several axenic, trace-metal clean phytoplankton cultures with isolates from the Caron, Webb, and Capone labs and assayed them for V-haloperoxidase activity following the method employed in Johnson et al. (2011). This produced weak positive results, but the low levels of activity and operationally defined units of the assay were not promising for further studies. Following the purchase and setup of the gas chromatographic method employed in the next two chapters, it was decided to assay marine heterotrophic bacteria as their rapid, dense growth in culture should allow for a detectable halocarbon signal if any are being produced, and no heterotroph assays had been previously reported in the literature despite their global importance. I would like to acknowledge the contributions and assistance of Laura Gomez- Consarnau for providing the heterotroph cultures and training me in their maintenance, the Fuhrman lab for use of their incubator and light microscope for cell counts. Lynda Cutter was invaluable in setting up and developing the GC analytical method. Klein 22 Abstract We explored the synthesis of halocarbon gases by marine heterotrophic bacteria. Six taxonomically diverse strains of ecologically relevant marine phyla were assayed in culture for production of halocarbons. Synthesis of the methyl halides CH 3 Cl, CH 3 Br, and CH 3 I was observed in all strains. Production appeared constitutive during log-growth phase and may potentially be growth-rate dependent. Extrapolations from these culture results based on both exponential growth rate values and biomass concentrations yields world-ocean annual flux estimates of 4.8 – 480 gG CH 3 Br, 140 – 2200 gG CH 3 I, and 14 – 68 gG CH 3 Cl. These fluxes are on the same order of magnitude as those derived from sea-air flux measurements with the exception of CH 3 Cl, which has large anthropogenic sources. The observed heterotrophic production of methyl halides may be a biomethylation process distinct from the haloperoxidase- mediated biosynthesis of poly-halogenated methanes observed previously in eukaryotic phytoplankton and cyanobacteria. These preliminary results suggest that marine heterotrophic bacteria are likely significant producers of these climate-relevant gases in the environment. Introduction Halogenated hydrocarbons (“halocarbons”) are produced from marine systems, with oceanic contributions ranging from 17% (CH 3 Cl) to 100% (CH 2 Br 2 ) of the global budget (WMO 2007a) and with an important biogenic contribution (Hill and Manley 2009). Though less studied than their man-made counterparts (anthropogenic chlorofluorocarbons, CFCs), biogenic halocarbons also deplete atmospheric ozone and play important roles in atmospheric chemistry. These natural halocarbons are only becoming more relevant as levels of CFCs decline following the bans instituted in the Montreal Protocol. Such natural CFC analogues currently represent Klein 23 25% of halogen-mediated stratospheric ozone destruction, but will account for the majority by mid-century (WMO 2007b). The biogenic production of halocarbons seems to occur in different organisms for a variety of reasons. They have previously been observed functioning as antioxidants in macroalgae (Küpper et al. 2008), while it has been hypothesized that their production in eukaryotic phytoplankton (Scarratt and Moore 1996; 1997) and cyanobacteria (Brownell et al. 2010, Johnson et al. 2011) might be a way to excrete excess halide ions (Gschwend et al. 1985, Küpper et al. 2008). Although heterotrophic marine bacteria may be on average the most abundant organisms in seawater, production of halocarbons by these organisms has never been assayed. Here we present evidence of widespread halocarbon biosynthesis by a taxonomically diverse selection of cultured marine heterotrophs representative of the major environmentally relevant alpha- and gamma-Proteobacteria and Flavobacteria phyla. Our results could therefore potentially be applied to the broad range of marine environments in which members of these clades dominate. Materials and Methods Six taxonomically diverse (Figure 1) strains of marine bacteria representing the main heterotrophic oceanic groups Alphaproteobacteria (DFL12 and MED193), Gammaproteobacteria (AND4) and Bacteroidetes (MED134, MED217 and MED217) (Table 1) were assayed in culture. Cultures were grown in 120 mL ZoBell marine broth medium in 250 mL glass vials (ca. 120mL headspace) at 22°C on a shaker at 10rpm and full light conditions using an artificial light source maintained at approximately 125 µmol photons m -2 s -1 until they reached stationary phase, which ranged from ca. 8 to 100 hours. Cultures reached high cell densities, typically on the order of 10 7 - 10 8 cells mL -1 during late log growth phase, and one final “late” time-point was taken 24 hours after the last experimental replicates to assess behavior Klein 24 in stationary phase. Duplicate culture vials were killed by acidification to ca. pH 2.0 with HCl (0.5mL 3mol L -1 solution for 125mL culture volume) HCl and refrigerated prior to analysis, under which conditions they are stable for up to two weeks (EPA 1986). Samples were analyzed for dissolved halocarbon concentrations using a gas chromatography (GC) method adapted from Schall and Heumann (1993) and quantified relative to an internal standard. CH 2 I 2 and CH 2 Br 2 were included in the GC analytical method but detected in none of the cultures. Media blanks to account for potential abiotic synthesis were assayed at the beginning and end of the experiment, and blank values were never more than 10% of those measured from the cultures and are subtracted from data presented. For cell abundance and volume determination, duplicate 1mL samples were fixed with 10% formalin (4% formaldehyde), stained with acridine orange (Hobbie et al. 1977), filtered onto pre-blackened filters and counted with epifluorescence microscopy. All halocarbon concentration and cell count data are presented in Tables 6-8. To assess the genomic potential for halocarbon biosynthesis of the assayed microbes, genome data was obtained from the Integrated Microbial Genome database (IMG, Markowitz et al. 2012, 2013) on enzymes suggested to be responsible for halocarbon biosynthesis. The amino acid sequences of the vanadium-containing bromoperoxidases from Corallina pilulifera (vanadium-dependent bromoperoxidase, Accession number Q8LLW7) (Butler and Carter- Franklin 2004) and Synechococcus sp. WH8202 (vanadium-dependent bromoperoxidase 2, Accession number Q0I6Q3) (Johnson et al. 2011) were compared against genomes of assayed organisms using the Basic Local Alignment Search Tool (BLAST) algorithm. To assess potential for halocarbons as a side-reaction product of biomethylation in a saline chemical environment, methyltransferase abundance was also determined using the IMG database, totaling all PFAM and COG hits for “methyltransferase” and “methylase” gene functions. Klein 25 Purge-and-trap capillary column gas chromatography with electron capture detection (GC-ECD) was employed for dissolved halocarbon analysis (Schall and Heumann 1998). 25mL media samples were purged with ultra-high purity He for 45min at a flow rate of 60mL min -1 through an in-line K 2 CO 3 drying tube and onto a liquid nitrogen trap. The purge vessel is rinsed with methanol and the drying trap replaced with 0.75g fresh K 2 CO 3 between individual analyses. Cryo-concentrated samples were introduced into an Agilent 8390 GC by means of a splitless injection with sweep pressure at 50psi for 1.5min returning to analytical column pressure of 18psi 2.5min after injection. Inlet temperature was set to 60°C to facilitate cryo-focusing on the column. Initial oven temperature was 40°C for 10min increasing to 120°C by 4°C min -1 and held there for another 2min. Temperature was then ramped to a final 240°C at a rate of 5°C min -1 and held for 20min, the last ca. 15min serving as a bakeout and cleaning step in preparation for successive analytical runs. If early-eluting peaks are not sufficiently resolved, reducing initial column pressure below 18psi and slowing temperature ramps can effectively separate the more volatile components. Concentrations were quantified relative to CBrCL 3 as an internal standard, an anthropogenic halocarbon never significantly used industrially (Dyrssen and Fogelqvist, 1981), using PeakSimple data system and software. To account for loss to headspace, Henry’s Law constants were used to determine equilibrium gas-phase concentrations and results are reported as aqueous plus equilibrium gas phase. Results are thus reported as determined aqueous concentration plus equilibrium gas concentration, k H * C aq where k H is the dimensionless gas/air partition coefficient and C aq is the analytically determined aqueous concentration. As halocarbons are moderately soluble in water, calculated gas-phase concentrations were never more than 10% of instrumentally determined aqueous values (Table 2), with the exception of the Klein 26 most volatile analyte CH 3 Cl which partitions 21% into the gas phase at storage (refrigerated) temperature. Detection limits as 3x standard deviation of the lowest standard were 0.24 pmol L -1 CH 3 Cl, 0.18 pmol L -1 CH 3 Br, 0.27 pmol L -1 for CH 3 I, 0.33 pmol L -1 for CH 2 Br 2 , and 0.38 pmol L -1 for CHBr 3 . All analytes were undetectable in Milli-Q blanks. Results and Discussion Production of methyl halides (single halogenations, CH 3 Br, CH 3 I, CH 3 Cl) was observed in all assayed cultures (Table 3, Figures 2b-d), and lower levels of bromoform (CHBr 3 ) were detected only in the Vibrio campbelli AND4 culture (Table 3). Of the methyl halides, CH 3 I was the most highly produced (0.67 – 2.1 amol µm -3 cell volume), with several times to nearly an order of magnitude higher per-cell production rates than CH 3 Br and CH 3 Cl (0.017 – 0.90 and 0.048 – 0.89 amol µm -3 ) (Table 3, Figures 2b-d). These values of bacterial synthesis of halomethanes are on a similar order of magnitude to per-cell concentrations of the metabolic waste product superoxide (0.5 -10 amol µm -3 . Diaz et al. 2013) but several orders of magnitude less than the metabolite dimethylsulfide (42 – 1050 amol µm -3 , Hatton et al 2012, Moran et al. 2012) in other organisms. Halocarbon concentrations over the course of the culture experiments generally fit an exponential curve (Figures 2b-d), similar to the exponential growth of bacterial cells (Figure 2a), potentially indicating that the observed halocarbon production is constitutive or growth-rate dependent. Standardized per-cell halocarbon concentrations (Figures 3a-c) appear relatively consistent over the growth cycle of the assayed organisms, with little clear trend towards lower or higher per-cell concentrations during early (lag) or late (stationary) phase. This suggests an extrapolation may be made to an estimate of environmental production based on log-phase Klein 27 halocarbon concentrations relative to biomass. Vibrio AND4 and Dinoroseobacter DFL12 displayed lower cellular quotas of halocarbons relative to the other strains, and for all strains CH 3 Cl and CH 3 I were more variable over the course of the experiment than CH 3 Br (Table 3, Figures 3a-c). The slopes of log-linearized cell counts or halocarbon concentrations (see example in Figure 3) represent the exponential growth rate constant k in the exponential growth rate expression C (t) = C (i) *e kt , where C is concentration (of cells or halocarbons) at a time (t) as a function of the initial concentration C (i) . Exponential growth rate constant k values are relatively constant during log growth phase and both cell growth and halomethane production display a linear relationship with each other, indicating that the halocarbon production may not be species-dependent but instead growth-rate dependent (Figures 4a-b). Linear regressions (excluding AND4) of halocarbon versus cell growth k values are all statistically significant with R 2 values greater than 0.90 (Figures 5a- b). Those relationships could explain about 89% of the CH 3 Br concentration measured in the AND4 cultures but only 63% and 75% of the CH 3 I and CH 3 Cl respectively. The observed linear relationships have potential for estimating halomethane production rates from bacterial growth rates, or estimating growth from halomethane production rates. Previously reported halocarbon biosynthesis by marine phytoplankton has been attributed to the action of haloperoxidase enzymes (Butler and Carter-Franklin 2004, Johnson et al. 2011), which produce poly-halogenated methanes (e.g. CH 2 I 2 , CHBr 3 ). Haloperoxidases would not be expected to produce the single-halogenation products, and none of the assayed bacteria possess a haloperoxidase gene based on genome annotations or were observed to produce Klein 28 polyhalomethanes in culture, with the exception of low levels of CHBr 3 in AND4 (Table 3). The CHBr 3 produced by AND4 may instead be a product of an uncharacterized enzyme with similar active site structure and related haloperoxidase function, as has been demonstrated in some acid phosphatases (Tanaka et al. 2002). As biomethylation is a fundamental biochemical processes and many methyltransferase enzymes are non-specific, some amount of methylation of halide ions might be expected to occur in a saline environment as a side reaction (Manley 2002). If such methylation of halides is responsible for the observed halomethanes production in heterotrophs, the abundance of methyltransferase enzymes and their specificity may relate to halocarbon biosynthesis. Methyltransferase enzyme abundances both in absolute terms and as a percentage of total genes in the genomes are presented in Table 3. The percentage of methyltransferase/methylase genes relative to genome size is relatively consistent between species, and no clear relationship between methyltransferase quantity and halocarbon production emerge. However the two strains that produced an order of magnitude lower halocarbons (AND4 and DFL12) possess both the greatest number of methyltransferase genes and the largest genome sizes. This might potentially be due to these two organisms having a larger and more specialized suite of biomethylation enzymes, thereby reducing production of halocarbons as side products. Extrapolation of the laboratory results to the global ocean was accomplished using literature cell growth rate and abundance values coupled with per-cell halocarbon abundances and rate constants derived from this culture work, integrated to 30m mixed layer depth. These values are compared to published annual halocarbon sea-air fluxes (Table 5), though these measurements involve large uncertainties. Our cultures were grown under optimum nutrient- enriched conditions, which rarely happens in nature, and there is also a lack of data coverage and Klein 29 large uncertainty in estimated global gas fluxes. However, using reported estimates of marine heterotrophic bacterial growth rates and abundances to extrapolate global annual fluxes (from biomass or exponential growth rates k) yielded values of 4.8 – 480 gG CH 3 Br year -1 and 140 – 2200 gG CH 3 I year -1 , on a similar order of magnitude to estimated annual marine fluxes (Table 5). For CH 3 Cl, however, the observed production rates extrapolate to 14 – 68 gG CH 3 Cl year -1 , 1-2 orders of magnitude less than estimated global CH 3 Cl fluxes, which might be expected given the large anthropogenic sources of this gas (Table 4) (WMO 2007a, Moore et al. 2008, Brownell et al. 2010). The low levels of CHBr 3 observed only in the Vibrio AND4 culture could account for only a small fraction of global CHBr 3 flux, suggesting that the heterotrophic contribution of that halocarbon species is small. Lower extrapolated fluxes based on biomass relative to the growth -rate k calculations may indicate the growth-rate estimate is biased due to the enriched culture media, and the biomass estimate may be more accurate given the nutrient scarcities bacteria may encounter in the environment. Since heterotrophic bacteria are typically the most abundant and fastest-growing organisms in the marine environment, the observed constitutive or growth-rate dependent production of methyl halides by all of the assayed, taxonomically diverse strains would suggest that marine heterotrophic bacteria are important and previously neglected biogenic sources of these climate-relevant gases. Klein 30 Table 1: Strain, species, and phylum of assayed heterotrophic bacterial strains. Strains represent several major marine phyla. Figure 1: Phylogeny of three domains of life, assayed clades highlighted. Source: CSIRO Marine and Atmospheric Research Data Centre, http://www.cmar.csiro.au/datacentre. Species Strain Clade Dinoroseobacter shibae DFL12 alpha-Proteobacteria Roseobacter sp. MED193 alpha-Proteobacteria Dokdonia donghaensis MED134 Flavobacteria Polaribacter sp. MED152 Flavobacteria Leeuwenhoekiella blandensis MED217 Flavobacteria Vibrio campbelli AND4 gamma-Proteobacteria Klein 31 Table 3. Halocarbon production per cell volume. Genomic data on methyltransferase possession obtained from a search of the Microbial Genome database (Markowitz et al. 2012, 2013). Analyte k H Θ k H at 2°C Cg / Ca References CH 3 Cl 0.10 0.19 0.21 Kavanaugh and Trussell 1980 CH 3 Br 0.16 0.65 0.06 Wilhelm et al. 1977 CHBr 3 2.00 6.58 0.01 Wright et al. 1992a CH 3 I 0.19 0.52 0.07 Hunter-Smith et al. 1983 Table 2. Henry’s Law constants (k H ) at standard temperature and at 2°C storage and analysis temperature. C g / C a is the dimensionless partition coefficient in the gaseous relative to aqueous phase. Klein 32 Table 5. Experimentally derived production rates extrapolated to global annual 30m mixed-layer oceanic production, compared to published estimates of global marine halocarbon flux ( a WMO 2007a). Literature flux values are from field measurements ( b Ducklow et al. 1993, c Ducklow et al. 1995, d Carlson et al 1998, e Kirchman et al 1993). Global microbial biomass estimates from Morris et al. 2002 integrated to an assumed mixed-layer depth of 30m Table 4. Cell volumes, culture cell growth and halocarbon exponential growth rate constants k, derived from regression of linearizations of cell counts and halocarbon concentrations using exponential growth expression C t = C i * e kt Observed global annual flux, gG year -1 growth rate (k) Biomass CH 3 Br 30 - 148 39 - 480 4.8 - 260 CH 3 I 270 - 450 740 - 2200 140 - 1800 CH 3 Cl 445 - 940 31 - 68 14 - 26 CHBr 3 843 2.2 1.4 Extrapolation to global ocean based on: Cell volume, um -3 Cell growth rate constant k, day -1 Strain CH 3 Br CH 3 I CH 3 Cl AND4 2.1 17.4 18.3 17 16.8 DFL12 1.3 1.83 2.07 1.87 2.05 MED134 1.7 1.39 1.57 1.6 1.73 MED152 2.4 1.81 2.05 1.79 2.02 MED193 2.2 2.15 2.23 1.99 2.14 MED217 1.9 1.41 1.6 1.5 1.6 Halocarbon production rate constant k, day -1 Klein 33 Cells counts per mL over the course of the culture experiment (log scale) Hours 0 20 40 60 80 100 120 [Log] Cells mL -1 1e+5 1e+6 1e+7 1e+8 1e+9 1e+10 Minutes (for AND4) 0 100 200 300 400 500 DFL12 MED134 MED152 MED193 MED217 AND4 CH 3 I concentration Hours 0 20 40 60 80 100 120 140 CH 3 I, pmol L -1 0 2000 4000 6000 8000 10000 12000 14000 16000 18000 Minutes (for AND4) 0 100 200 300 400 500 DFL12 MED134 MED152 MED193 MED217 AND4 Figures 2a-b. Cell counts (top) and CH 3 I concentration (bottom) over the course of the culture experiments. Note log scale for cell abundances. Regression lines denote exponential (natural log) regression on halocarbon concentration values. Data for each time point consists of two biological replicates. Klein 34 CH 3 Cl concentration Hours 0 20 40 60 80 100 120 140 CH 3 Cl, pmol L -1 0 500 1000 1500 2000 Minutes (for AND4) 0 100 200 300 400 500 DFL12 MED134 MED152 MED193 MED217 AND4 CH 3 Br concentration Hours 0 20 40 60 80 100 120 140 CH 3 Br, pmol L -1 0 500 1000 1500 2000 2500 3000 Minutes (for AND4) 0 100 200 300 400 500 DFL12 MED134 MED152 MED193 MED217 AND4 Figures 2c-d. CH 3 Br (top) and CH 3 Cl concentration (bottom) over the course of the culture experiments. Regression lines denote exponential (natural log) regression on halocarbon concentration values. Data for each time point consists of two biological replicates. Klein 35 Per-cell CH 3 Br over the course of the experiment (log scale) Hours 0 20 40 60 80 100 120 pM CH 3 Br per cell 1e-8 1e-7 1e-6 1e-5 1e-4 Minutes (AND4) 0 100 200 300 400 500 600 DFL12 MED134 MED152 MED193 MED217 AND4 Per-cell CH 3 Cl over the course of the experiment (log scale) Hours 0 20 40 60 80 100 120 pM CH 3 Cl per cell 1e-7 1e-6 1e-5 1e-4 Minutes (AND4) 0 100 200 300 400 500 600 DFL12 MED134 MED152 MED193 MED217 AND4 Figures 3a-b. Per-cell CH 3 Br (top) and CH 3 Cl concentration (bottom) over the course of the culture experiments. Data for each time point consists of two biological replicates. Klein 36 Per-cell CH 3 I over the course of the experiment (log scale) Hours 0 20 40 60 80 100 120 pM CH 3 I per cell 1e-6 1e-5 1e-4 Minutes (AND4) 0 100 200 300 400 500 600 DFL12 MED134 MED152 MED193 MED217 AND4 Figures 3c. Per-cell CH 3 I over the course of the culture experiments. Data for each time point consists of two biological replicates. Klein 37 ln[CH 3 Br] over the course of the culture experiments Hours 0 20 40 60 80 100 ln[CH 3 Br] -4 -2 0 2 4 6 8 10 Minutes (for AND4) 0 100 200 300 400 500 DFL12 MED134 MED152 MED193 MED217 AND4 Figure 4. Natural log transformation of CH 3 Br concentration as an example of exponential relationship between time and halocarbon concentration/production. Klein 38 Cell growth rate constant k (log phase), d -1 1.0 1.5 2.0 2.5 Halocarbon production rate constant k (log phase), day -1 1.0 1.5 2.0 2.5 CH 3 Br CH 3 I CH 3 Cl Cell growth rate constant k (log phase), d -1 0 2 4 6 8 10 12 14 16 18 0 2 4 6 8 10 12 14 16 18 20 CH 3 Br CH 3 I CH 3 Cl Figure 5a-b. Plots of experimentally derived halocarbon production rate k values vs. culture growth rate k. A (left): Linear regression using all six assayed cultures except AND4. All plotted regressions were calculated omitting the higher values of AND4; slope and R 2 values were 0.91 and 0.96 for CH 3 Br, 0.59 and 0.94 for CH 3 I, and 0.68 and 0.90 for CH 3 Cl, respectively. B (right): Linear regression showing predictive power on higher AND4 values. Klein 39 Vibrio campbelli Dinoroseobacter shibae AND4 DFL12 Minutes Cells mL -1 CH 3 Br CH 3 I CH 3 Cl CHBr 3 Hours Cells mL -1 CH 3 Br CH 3 I CH 3 Cl 0 7.80E+05 0.45 10.05 1.37 0.31 0 1.67E+06 0.20 16.34 0.67 0 9.40E+05 0.69 16.19 1.54 0.36 0 2.50E+06 0.28 19.95 1.33 130 1.60E+06 1.08 27.10 2.82 0.78 8 3.19E+06 0.43 29.61 1.31 130 1.94E+06 1.13 30.15 2.89 1.00 8 3.12E+06 0.60 20.77 1.32 240 3.40E+06 2.73 45.22 4.10 1.47 16 7.65E+06 1.90 52.96 2.94 240 2.96E+06 1.82 35.22 4.41 0.96 16 8.67E+06 1.05 26.13 3.57 285 5.26E+06 4.31 58.18 5.68 1.78 24 9.39E+06 2.20 82.45 4.10 285 7.64E+06 5.26 117.66 13.89 3.85 24 1.22E+07 1.49 88.77 8.83 330 1.41E+07 12.93 163.61 17.79 5.07 32 2.86E+07 4.15 184.87 10.67 330 1.57E+07 15.43 237.08 21.82 6.91 32 4.38E+07 5.86 356.74 16.02 370 4.00E+07 42.64 498.40 46.58 21.06 40 1.15E+08 12.58 353.32 34.09 370 3.94E+07 35.18 517.98 45.27 15.76 40 7.01E+07 18.94 505.37 27.36 410 8.90E+07 80.28 1183.70 115.01 33.24 48 2.12E+08 40.76 798.52 113.46 410 7.74E+07 65.97 1028.89 134.02 26.81 48 1.83E+08 21.64 1226.38 86.19 450 1.44E+08 100.03 2571.88 188.96 62.00 56 3.65E+08 54.94 2651.44 128.00 450 1.23E+08 96.61 1477.55 192.52 54.67 56 2.10E+08 47.03 1380.04 173.00 490 2.58E+08 152.35 3034.64 421.39 111.48 64 4.22E+08 63.21 2632.81 315.13 490 1.64E+08 169.02 2152.77 220.67 151.53 64 4.10E+08 86.43 2081.01 204.00 2400 6.84E+08 524.85 6513.86 926.85 233.38 72 3.67E+08 84.40 3856.07 448.29 2400 3.29E+08 397.47 3504.92 558.88 301.53 72 5.06E+08 175.72 4037.39 279.44 80 5.73E+08 228.74 3511.64 507.21 80 3.99E+08 172.50 7492.25 454.27 104 9.86E+08 401.29 5013.22 636.85 104 7.94E+08 336.41 13269.52 1148.06 Concentration (pmol L -1 ) Concentration (pmol L -1 ) Table 6. Raw cell count and concentration data for Vibrio AND4 and D. shibae DFL12 Klein 40 Dokdonia donghaensis Polaribacter sp. MED134 MED152 Hours Cells mL -1 CH 3 Br CH 3 I CH 3 Cl Hours Cells mL -1 CH 3 Br CH 3 I CH 3 Cl 0 2.33E+05 1.87 11.29 0.69 0 8.97E+05 0.96 35.35 2.50 0 7.16E+05 4.97 7.27 1.24 0 1.09E+06 1.29 23.63 2.76 10 7.10E+05 4.97 14.16 2.38 8 1.86E+06 2.05 38.76 4.28 10 9.62E+05 8.29 9.82 3.21 8 1.78E+06 2.24 39.19 4.31 20 1.53E+06 7.65 38.24 4.39 16 4.18E+06 6.23 81.85 6.77 20 1.22E+06 7.86 7.85 2.03 16 3.61E+06 4.15 49.31 7.06 30 3.99E+06 31.74 28.11 4.04 24 5.94E+06 9.83 107.04 8.80 30 2.11E+06 33.24 70.09 9.64 24 6.88E+06 10.31 182.37 24.72 40 5.85E+06 38.55 114.05 16.45 32 1.69E+07 27.03 283.04 27.40 40 7.27E+06 70.86 94.09 17.03 32 1.87E+07 32.56 438.60 39.49 50 1.53E+07 125.47 188.27 42.53 40 4.76E+07 81.87 882.17 68.47 50 1.75E+07 283.46 587.24 36.94 40 4.41E+07 70.71 901.28 78.32 60 3.05E+07 396.39 857.57 148.92 48 1.12E+08 183.84 1905.76 224.26 60 3.00E+07 253.37 789.33 116.72 48 8.51E+07 147.12 1605.07 206.39 70 5.94E+07 433.19 1317.21 203.27 56 1.77E+08 197.07 3677.79 384.54 70 4.84E+07 462.39 580.17 139.12 56 1.34E+08 216.40 2112.89 448.43 80 1.07E+08 544.48 1829.95 605.12 64 2.12E+08 274.23 3884.35 817.50 80 6.14E+07 756.15 827.74 310.87 64 1.82E+08 351.55 3186.09 572.94 90 5.99E+07 1260.08 3177.06 505.08 72 2.31E+08 640.76 4342.00 938.18 90 7.93E+07 1360.16 3873.59 781.22 72 2.10E+08 672.65 5286.00 994.17 100 5.08E+07 1878.88 3207.87 790.74 80 2.34E+08 789.38 5767.73 1117.84 100 1.27E+08 1203.01 2806.00 856.97 80 2.28E+08 628.65 6482.01 1077.44 124 8.63E+07 3673.20 4253.63 1263.60 104 4.07E+08 1057.62 11440.88 2470.20 124 2.62E+08 1809.09 4855.51 1659.43 104 4.99E+08 1555.71 13627.78 3067.47 Concentration (pmol L -1 ) Concentration (pmol L -1 ) Table 7. Raw cell count and concentration data for Dokdonia MED134 and Polaribacter MED152 Klein 41 Rosebacter sp. Leeuwenhoekiella blandensis MED193 MED217 Hours Cells mL -1 CH 3 Br CH 3 I CH 3 Cl Hours Cells mL -1 CH 3 Br CH 3 I CH 3 Cl 0 7.99E+05 0.47 14.48 1.08 0 7.27E+05 0.61 22.98 1.75 0 1.42E+06 2.39 45.69 1.72 0 1.10E+06 1.37 25.52 2.21 8 1.39E+06 2.13 36.37 7.68 10 1.39E+06 1.36 39.53 4.62 8 1.37E+06 1.34 36.77 4.13 10 1.96E+06 2.19 20.77 3.62 16 4.24E+06 2.90 54.66 5.85 20 3.26E+06 3.86 86.76 7.04 16 3.36E+06 4.75 33.63 5.88 20 3.39E+06 2.99 27.12 3.67 24 5.12E+06 16.82 98.32 2.98 30 6.24E+06 9.34 58.87 4.66 24 7.79E+06 18.23 126.38 18.42 30 4.40E+06 10.42 164.13 22.74 32 1.77E+07 23.89 283.33 49.49 40 1.06E+07 18.38 183.98 28.77 32 1.12E+07 15.01 482.81 21.62 40 1.23E+07 17.25 280.70 28.83 40 5.31E+07 49.93 854.64 39.52 50 2.95E+07 46.67 494.01 53.40 40 6.45E+07 127.42 1179.78 111.65 50 4.85E+07 74.95 991.41 84.59 48 1.08E+08 116.07 1574.16 168.64 60 9.53E+07 150.75 1334.03 210.81 48 6.65E+07 172.50 2251.10 256.58 60 9.11E+07 98.57 1364.31 152.73 56 1.37E+08 244.36 3144.51 362.62 70 1.02E+08 157.65 2758.34 307.31 56 1.16E+08 127.37 2334.75 438.68 70 1.03E+08 116.85 1373.38 270.79 64 3.67E+08 257.23 3146.32 617.36 80 2.27E+08 191.96 4195.09 833.85 64 2.22E+08 588.64 3517.45 406.13 80 1.37E+08 365.61 2198.40 410.23 72 3.52E+08 753.02 4306.79 475.00 90 1.25E+08 333.19 7060.31 778.69 72 5.44E+08 538.39 5817.40 724.00 90 2.14E+08 390.14 3827.00 914.64 86 1.11E+09 1784.04 12788.52 1890.62 100 1.64E+08 836.75 6838.83 603.63 86 1.45E+09 1089.35 14248.59 1542.00 100 1.87E+08 320.61 5555.47 829.63 124 4.08E+08 1520.95 11920.09 1698.44 124 5.18E+08 825.92 16311.96 1746.53 Concentration (pmol L -1 ) Table 8. Raw cell count and concentration data for Roseobacter MED193 and D. shibae MED217 Klein 42 Chapter 3: Seasonal and ecological dynamics of halocarbon concentrations and sea-air fluxes at the San Pedro Ocean Time-series Foreword and Acknowledgements This third chapter will present field measurements made at the San Pedro Ocean Time- series (SPOT) site off the coast of Southern California. After the successful measurement of halocarbon production in heterotrophic marine bacterial cultures presented previously, validation of these results in an environmental setting was a logical next step. The SPOT station was attractive due to the presence of a variety of organisms previously demonstrated to produce halocarbons as well as many similar heterotrophs to those I assayed. Additionally, ancillary data from other laboratories cooperating on the SPOT program include detailed characterization of the microbial community. This allows for comparison of trends in seasonal microbial community with measured halocarbon concentrations and derived sea-air fluxes. I would like to acknowledge Troy Gunderson, the SPOT program, and the crew of the R/V Yellowfin for allowing me water samples from their rosette cast and for organizing the research cruises. The Fuhrman lab produced and provided the bacterial abundance and activity data. William Haskell assisted with calculating piston velocities from satellite wind-speed data, and Doug Hammond for guiding me in constructing the numerical box models presented here. Abstract To examine seasonal dynamics of marine biological halocarbon production, monthly concentrations of halocarbon gases were determined at the San Pedro Ocean Time-series station from June 2013 to June 2014. Sea-air fluxes were derived from measured concentrations. Both fluxes and concentrations increased during the Mar-Apr-May upwelling season consistent with Klein 43 enhanced biological productivity during these months. The development of near-surface maxima for CH 3 I and CH 3 Br during upwelling coincided with high bacterial activity, and extrapolations from culture studies of potential heterotrophic bacterial production of these gases suggests bacteria may be their major producers. Increased CHBr 3 and CH 2 Br 2 in the upwelling season is likely associated with the establishment of diverse diatom-dominated assemblages which have previously been observed to exhibit high haloperoxidase activity. Introduction The world’s oceans are both the largest reservoir and source of halogenated hydrocarbons, or halocarbons (WMO 2007a). These gases are produced chiefly biologically and play important roles in both stratospheric and tropospheric chemistry (WMO 2007b). In the stratosphere, they behave much like anthropogenic chlorofluorocarbons (CFCs) and deplete the ozone layer, with implications for the global radiation budget and human health due to its UV filtration properties (Sturges et al 2000). In the troposphere they destroy both ozone and the cloud-nucleating biogenic gas dimethyl sulfide; iodocarbons can also themselves nucleate clouds (von Glasgow et al. 2004). As CFC levels decline following the restrictions of the Montreal Protocol, biogenic halocarbons are becoming a relatively more important contributor to stratospheric ozone destruction and are expected to account for more than 50% by 2050 (Gebhardt 2008). Despite this, the biochemistry and geographic and temporal distribution of their production has been little studied. The San Pedro Ocean Time-series (SPOT) station located in the San Pedro Channel off- shore of Los Angeles, CA provides an excellent opportunity to study seasonality of halocarbon concentrations and sea-air fluxes while obtaining a large number of depth profile measurements Klein 44 relative to the small number that exist in published literature (Baker et al. 2007, Moore and Groszko 1999, Roy et al. 2011, Sturges et al. 1992). SPOT is characterized by a strong spring upwelling season which enhances biological productivity, typically during the months of March through May, while conditions the rest of the year are relatively oligotrophic (Countway et al. 2010). This contrast allows for identification of significant biogenic halocarbon seasonality. Most of the organisms known to produce halocarbons, such as diatoms, the haptophyte Phaeocystis, and the cyanobacteria Prochlorococcus and Synechococcus have representatives that are abundant and predictably seasonal at SPOT (Countway et al. 2010, Fuhrman 2006), allowing for the ecology of halocarbon production to also be explored. This study aims to better understand the seasonal and ecological dynamics of marine biosynthesis of these climate- relevant gases. Materials and Methods Sampling for the analysis of halocarbon concentrations was performed monthly from June 2013 to June 2014 (excluding Dec 2013 due to weather conditions) at the SPOT station. Twelve depths (near-surface, 10m, 20m, chlorophyll maximum, 30m, 50m, 60m, 100m, 250m, 500m, 750m, and near-bottom 890m) were sampled in a single cast using acid-cleaned Niskin bottles Using 250mL silicone-septa, gas-tight serum vials, sub-samples without headspace were taken from the midday (12p.m. – 1p.m.) cast, acidified with 1mL 3mol L -1 HCl to ca. pH 2.0, and stored at 2°C pending analysis. Determination of the aqueous halocarbons CH 3 Br, CH 3 I, CH 2 Br 2 , and CHBr 3 was via gas chromatography with electron capture detection, using the method of Schall and Heumann (1993) with a modified oven program to shorten analysis run- time (see Chapter 2 for GC method details). Concentrations were quantified relative to a CBrCl 3 Klein 45 internal standard and values corrected for chloride substitution after Elliott and Rowland (1993). All raw data and available ancillary parameters are presented in Tables 2-5. Fluxes from the ocean to the atmosphere were calculated for all quantified aqueous halocarbons using piston velocities derived from weighted-average satellite wind speed data (Nightingale et al. 2000, Reuer et al. 2007, Waninkhoff 1992) and integrated to 20m as a typical mixed-layer depth. To estimate the potential biological contribution to the SPOT fluxes, halocarbon production rates from the culture studies presented earlier in this thesis(Chapter 2), typical SPOT eukaryotic phytoplankton and bacterial abundances, and bacterial growth rate were used to calculate a 20m mixed-layer bulk daily biological production rate and compare to the actual measured fluxes. Extrapolations were made for heterotrophic bacteria (median value fromChapter 2 studies), several species of diatoms and the haptophyte Phaeocystis (Hill and Manley, Scarratt and Moore 1996; 1997), as well as for the cyanobacteria Synechococcus CC9311 (Brownell et al. 2010, Johnson et al. 2011) and Prochlorococcus marinus (Brownell et al. 2010). Results and Discussion 1. General trends in halomethane concentrations Depth profiles for CTD in-vivo fluorescence (as a proxy for photosynthetic activity) and bacterial activity by leucine incorporation may be found in Figure 1a-b. Profiles are shown to 150m depth to better present the halocarbon distributions in the upper water column; concentrations of all four analytes were detectable in the deeper 250, 500, and 890m samples, but five to ten-fold lower than near surface and uniform in concentration, ca. 0.15 - 0.25 pmol L -1 CH 3 I, 0.60 – 1.2 pmoL L -1 CH 3 Br, 0.30 – 0.40 pmol L -1 CH 2 Br 2 , and 0.70 – 1.2 pmol L -1 CHBr 3 . Klein 46 Halocarbon depth profiles are displayed in Figures 2a-b for CH 3 Br and CH 3 I, and in Figures 3a-b for CH 2 Br 2 and CHBr 3 . A comparison of the ranges of measured SPOT halocarbons with published literature values is presented in Table 1. All four halocarbons were elevated two to three-fold higher in concentration near the chlorophyll maximum (ca. 20-40m depth, Figure 1b) relative to the rest of the surface 50m, consistent with major biological sources. CH 3 I also developed a surface maximum during the winter and upwelling seasons (Jan-June, Figure 2b). Ranges of halocarbon concentrations measured at SPOT were similar to more open-ocean and oligotrophic values reported in the literature (Table 1), with CH 2 Br 2 and CHBr 3 concentrations in the highly productive Bay of Bengal (Yamamoto et al. 2001) and coastal Arabian Sea (Roy 2010) reaching up to two orders of magnitude greater than those observed at SPOT. Each individual analyte will be treated separately in more detail in the following sections. 2. Methyl Bromide Methyl bromide (CH 3 Br) may be formed at the surface by photochemical generation of methyl radicals, which can proceed to react with halogens to form methyl halides (Itoh 1997, Moore and Zafiriou 1994, Moore 2008). It has been used as a fumigant and pesticide and has anthropogenic sources, however its usage has declined in recent decades (Gebhardt 2008). Biogenic production of CH 3 Br was observed in several eukaryotic phytoplankton (Scarratt and Moore 1996; 1997) and in heterotrophic bacteria (Chapter 2 this thesis). The mechanism of biological production is unknown (Baker et al. 1999, Itoh et al. 1997, Manley 2002), though in the eukaryotes it was correlated with biomass but not photosynthesis (Moore et al. 1996), and was growth-rate dependent in the bacteria (Chapter 2). At SPOT, CH 3 Br ranged in concentration from ca. 1pM at depth to a maximum of 12pM at the chlorophyll max in April (Figure 2a), values similar to those reported in the literature (1.2 – 4.2pM, Table 1). None of the sampled A. Klein 47 months exhibited an elevated surface CH 3 Br concentration suggestive of strong photochemical production or deposition from anthropogenic sources. As the highest concentrations are associated with the chlorophyll maximum and there is no strong evidence of a surface photochemical or anthropogenic source, I hypothesize that CH 3 Br production at SPOT is likely primarily biogenic. 3. Methyl Iodide Similar to CH 3 Br, methyl iodide (CH 3 I) may be produced by either methyl radical photochemistry at the surface (Happell and Douglas 1996) or biologically, though it does not have a significant anthropogenic source (Itoh 1997, Moore and Grozsko 1999, Moore and Zafiriou 1994). Biogenic production has been observed in several diatoms and Phaeocystis (Scarratt and Moore 1996; 1997), the cyanobacterium Prochlorococcus (Brownell et al. 2010), and several marine heterotrophic bacteria (This thesis, Chapter 2). Also in common with CH 3 Br, the biochemical processes for biological synthesis are unknown (Itoh et al. 1997, Manley 2002) but appear growth-rate dependent in bacteria (Chapter 2). CH 3 I concentrations at SPOT ranged from ca. 0.3 pM at depth to a maximum of 6.4 pM at the surface in April (Figure 2b), similar to reported literature values of 0.20 – 19 pM (Table 1). The distribution of CH 3 I appears bimodal, elevated at the chlorophyll max but ca. 150% higher at the surface where a distinct maximum develops Jan-June during the winter and into spring upwelling seasons. The CH 3 I surface max represents some proportion of photochemical production, but might be due largely to biology as the highest CH 3 I concentrations and highest bacterial activity coincide. CH 3 I is produced in a growth-rate dependent manner by marine heterotrophic bacteria (Chapter 2) observed bacterial growth rates were similar or higher near-surface than at the chlorophyll max (Figure1a), and the highest CH 3 I and bacterial growth rate values were both observed near-surface in April (Figure Klein 48 1a, 2b). Therefore, the high near-surface CH 3 I during winter and the spring upwelling seasons might be significantly due to growth-rate dependent biosynthesis by marine heterotrophic bacteria. Additionally, diatoms, Phaeocystis, and Prochlorococcus are all abundant and important autotrophs near-surface at SPOT (Countway et al. 2010, Fuhrman et al 2006) during the upwelling season and might also account for a portion of the increased CH 3 I found there. 4. Dibromomethane Dibromomethane (CH 2 Br 2 , Figure 3a) is produced primarily biologically (Liu et al. 2013), but it is unclear to what extent that is directly by haloperoxidases as observed in several diatoms (Hill and Manley 2009), or by microbial degradation of CHBr 3 (Liu et al. 2013). CHBr 3 hydrolyzes to form CH 2 Br 2 , however the rate of this reaction in natural waters is considered to be too slow to represent more than ca. 4% of observed global CH 2 Br 2 fluxes (Butler et al. 2007). At SPOT, CH 2 Br 2 was highest at or slightly below the chlorophyll maximum and concentrations at ranged from ca. 0.60 below 150m to a maximum of 15 pM at 30m depth in April. These are similar to open ocean surface values (Table 1), though in the highly productive Bay of Bengal and (Yamamoto et al. 2001) and Arabian Sea (Roy 2010) concentrations were up to two orders of magnitude higher. As recent evidence has suggested CH 2 Br 2 is primarily a microbial breakdown product of CHBr 3 (Liu et al. 2013), and CHBr 3 is in turn produced primarily by haloperoxidases in photosynthetic eukaryotes, the increase in the sum of CH 2 Br 2 + CHBr 3 concentrations (Figure 4) sub-surface during the upwelling season likely indicates enhanced production of CHBr 3 by diatoms without a corresponding increase in bacterial degradation rate to CH 2 Br 2 . Klein 49 5. Bromoform Bromoform (CHBr 3 , Figure 3b) is believed to be produced in the marine environment chiefly by the action of vanadium-containing bromoperoxidases (Butler and Carter-Franklin 2007, Moore et al. 1996). These enzymes and CHBr 3 biosynthesis has been observed in a broad assortment of marine taxa: macroalgae (Gschwend et al. 1985), diatoms (particularly polar and sea ice species) (Sturges et al. 1992 Hill and Manley 2009), Phaeocystis (Moore et al. 1996), and the cyanobacterium Synechococcus CC9311 (Johnson et al. 2011). CHBr 3 at SPOT ranged from ca. 1pM at depth to a maximum of 22 pM at the chlorophyll max in April. Similar to CH 2 Br 2 , these concentrations more resemble reported open-ocean values and are one to two orders of magnitude less than concentrations in other highly productive regions such as the Bay of Bengal and coastal Arabian Sea (Table 1) and the coastal Antarctic where sea ice diatoms are a strong source (Carpenter et al. 2012). At SPOT, CHBr 3 was generally associated with the chlorophyll max, and during the upwelling months Mar-May a high concentration develops slightly above the chlorophyll max and in the thermocline (Figure 3b). The vanadium haloperoxidase gene is not widely distributed taxonomically in marine microbes (Johnson et al. 2011) and production has been demonstrated almost exclusively in diatoms with wide variability in rates. The observed Mar-May increase in CHBr 3 in the thermocline is likely due to growth of VBrPO- possessing ecotypes of the diverse assemblage of diatoms that typically bloom during the upwelling season. Phaeocystis and Synechococcus CC9311 are also abundant at SPOT (Countway et al. 2010, Fuhrman et al. 2006) and likely contribute to CHBr 3 production, though likely more so during oligotrophic non-upwelling conditions. CHBr 3 production was observed in cultures of the Vibrio strain AND4 (Chapter 2) but none of the other assayed marine Klein 50 heterotrophic bacteria, and at a rate of production two orders of magnitude lower than for the halomethanes. Heterotrophic bacteria may be minor producers of CHBr 3 . 6. Seasonality and fluxes Monthly halocarbon concentrations for the top 50m (Figure 5) exhibit up to an approx. 150-200% increase for all measured analytes during the Mar-May upwelling season, consistent with the transition from a more oligotrophic regime to enhanced productivity during upwelling driving increased biological production of halocarbons (Figures 1-3, 5). The distinct vertical structures with depth of CH 3 I (Figure 2b) and CHBr 3 (Figure 3b) that develop during the upwelling season likely reflect enhanced contribution from seasonally dominant taxa, potentially heterotrophic bacteria near-surface for CH 3 I and thermocline-dwelling ecotypes of diatoms for CHBr 3 . Sea-air fluxes integrated for the top 20m mixed-layer (Figure 6, Table 6) reflect a similar seasonal trend. Fluxes have their lowest values during the summer months, Jun-Aug, and increased ca. 2-5 fold during the upwelling season to their maxima in April. CH 3 Br fluxes ranged from 5.5 to 13, and CH 3 I from 15 to 82 nmol m -2 day -1 , while CH 2 Br 2 and CHBr 3 varied from 2.9 to 14 and 0.87 to 2.4 nmol m -2 day -1 , respectively. Relative to reported literature halocarbon fluxes (Figure 7), SPOT values are quite similar. Many of the limited selection of organisms assayed in culture for halocarbon production are abundant and important at SPOT (Countway 2010). This allows for an extrapolation of culture production rates using typical SPOT abundances for eukaryotic organisms (Kim et al. 2014) and Synechococcus, and measured bacterial growth rates for heterotrophic bacteria (Chapter 2). Estimated daily production rates, integrated to a typical 20m mixed layer depth are presented in Figures 5 and 6 alongside observed SPOT and reported literature fluxes. Estimates Klein 51 based on heterotrophic bacterial production rates and halocarbons per biomass obtained from Chapter 2 are similar to (CH 3 Br) or overestimate by up to 5-fold (CH 3 I, Figure 6) SPOT fluxes, particularly during Apr-May (Figure 6) when mixed-layer bacterial growth rates are very high (Figure 1a), suggesting that marine heterotrophic bacteria are likely significant or even dominant producers of CH 3 Br and CH 3 I at SPOT, particularly during upwelling. Extrapolated production from the low levels (two orders of magnitude less than other halocarbon analytes) of CHBr 3 produced by Vibrio AND4 was less than 1 nmol -2 day -1 , suggesting heterotrophic contribution may be relatively minor for this particular gas. Extrapolations from diatoms and Prochlorococcus (Figure 7) underestimate SPOT CH 3 I fluxes but are on a similar order of magnitude, suggesting they may also be important contributors for that gas. Diatoms production also appears sufficient, based on these culture extrapolations, to sustain a large proportion of the CHBr 3 flux, while estimates from Phaeocystis and Synechococcus underestimate SPOT CHBr 3 fluxes by an order of magnitude. 7. Diffusion-advection box modeling for CH 2 Br 2 and CHBr 3 Since the hydrolysis rate of both CH 2 BR 2 and CHBr 3 in natural waters is on the order of months and CH 2 Br 2 is primarily a product of microbial degradation of CHBr 3 (Liu et al. 2013, Ichikawa et al. 2015), a box model may be constructed to evaluate the plausibility of strictly biological sources for both CH 2 Br 2 and CHBr 3 (Figure 8a-b). Adapted from the numerical models of Fortin (1981), a 50m surface twelve-layer model was applied to the March 2014 (spring upwelling) depth profile. For the model initial conditions, average measured March surface layer concentrations were used, a uniform 14.7 pmol L -1 CHBr 3 and 6.28 pmol L -1 CH 2 Br 2 . Each layer exchanges with the layers immediately above or below, with the bottom layer fixed at the initial concentration and the top layer losing bulk concentration equal to the Klein 52 calculated daily fluxes described previously. A typical eddy diffusivity of 1 cm 2 s -1 and upwelling rate of 1.25 m day -1 were assumed, several other reasonable values were used and none significantly affected the model output. Under the assumptions that loss rate to hydrolysis is negligible and CH 2 Br 2 is produced by microbial degradation of CHBr 3 in steady state the following should be true: R CHBr3 = F CHBr3 + F CH2Br2 and R CH2Br2 = F CH2Br2 Where R terms are biological production (or microbial degradation) rates and F terms are calculated SPOT sea-air fluxes based on dissolved halocarbon concentrations and satellite windspeed derived piston velocities. I therefore apply a point source of CHBr 3 at the chlorophyll max equal to the sum of CHBr 3 and CH 2 Br 2 sea-air fluxes (10.15 nmol m -2 day -1 ) and a point sink (CH 2 Br 2 source) at the CH 2 Br 2 max equal to calculated CH 2 Br 2 sea-air flux (1.76 nmol m -2 day -1 ). Measured March 2014 SPOT halocarbon concentrations, model initial and 30-day simulation results are presented in Figure 8 and the raw model output data in Table 7. 30-day simulation results correspond reasonably well in overall vertical structure to the measured March concentrations, however at ca. 30m there is a significant excess of CHBr 3 and deficit of CH 2 Br 2 in the simulation relative to actual concentrations. This might suggest that the microbial degradation of CHBr 3 into CH 2 Br 2 is not sufficient to sustain observed concentrations, and other sources and processes such as horizontal transport or direct biosynthesis of CH 2 Br 2 by haloperoxidases are important. Klein 53 Conclusions We present a year of monthly halocarbon depth profiles and sea-air fluxes from SPOT indicate a strong seasonal relationship with the spring upwelling season. During these months, enhanced productivity increases halocarbon biosynthesis generally, and elevated CH 3 Br and CH 3 I concentrations and fluxes are potentially supported in large part by highly active heterotrophic bacteria in the surface mixed layer. A diverse assemblage of diatoms as well as Phaeocystis and Synechococcus are abundant and seasonally dominant at SPOT; with the exclusion of macroalgae they represent all the major marine taxa to have demonstrated CHBr 3 production. Thus most of the CHBr 3 at SPOT can be attributed to members and relatives of these taxa, in particular to ecotypes residing in the thermocline which develop a distinct concentration maximum during upwelling. As the impact of anthropogenic CFCs fade in the coming decades, biogenically produced marine halocarbons will play an increasingly important role in regulating both stratospheric and tropospheric ozone chemistry. Though these data contribute some better understanding of seasonal dynamics of halocarbons and their potential sources, their geographic and temporal distribution as well as underlying biosynthesis are poorly characterized. An increased effort to elucidate marine halocarbon biosynthesis is needed to understand potential their significant, near-term global climate interactions and feedbacks. Klein 54 CTD fluoresence at SPOT (summer-fall-winter) g L -1 0 2 4 6 8 10 Depth (m) 0 50 100 150 400 600 800 June 19, 2013 July 18, 2013 Aug 14, 2013 Sept 18, 2013 Oct 15, 2013 Nov 13, 2013 Jan 15, 2014 Feb 12, 2014 June 18, 2014 CTD fluoresence at SPOT (spring upwelling) g L -1 0 2 4 6 8 10 Depth (m) 0 50 100 150 400 600 800 Mar 12, 2014 Apr 10, 2014 May 21, 2014 SPOT bacterial production by leucine incorporation (spring upwelling) 10 5 cells mL -1 day -1 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 Depth (m) 0 50 100 150 400 600 800 Mar 12, 2014 Apr 10, 2014 May 21, 2014 SPOT bacterial production by leucine incorporation (summer-fall-winter) 10 5 cells mL -1 day -1 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 Depth (m) 0 50 100 150 400 600 800 June 19,2013 July 18, 2013 Aug 14, 2013 Sept 18, 2013 Oct 15, 2013 Nov 13, 2013 Jan 15, 2014 Feb 12, 2014 Figures 1a-b. CTD in vivo fluoresence and (a) bacterial productivity by leucine incorporation (b). Profiles are split between summer-fall-winter (left) and spring upwelling (right) seasons. Note depth scale is condensed below150m. B. A. Klein 55 Depth profile of dissolved CH 3 Br at SPOT (spring upwelling) pM CH 3 Br 0 2 4 6 8 10 12 14 Depth (m) 0 50 100 150 400 600 800 Mar 12, 2014 Apr 10, 2014 May 21, 2014 Depth profile of dissolved CH 3 Br at SPOT (summer-fall-winter) pM CH 3 Br 0 2 4 6 8 10 12 14 Depth (m) 0 50 100 150 400 600 800 June 19, 2013 July 18, 2013 Aug 14, 2013 Sept 18, 2013 Oct 15, 2013 Nov 13, 2013 Jan 15, 2014 Feb 12, 2014 June 18, 2014 Depth profile of dissolved CH 3 I at SPOT (summer-fall-winter) pM CH 3 I 0 2 4 6 8 Depth (m) 0 50 100 150 400 600 800 June 19, 2013 July 18, 2013 Aug 14, 2013 Sept 18, 2013 Oct 15, 2013 Nov 13, 2013 Jan 15, 2014 Feb 12, 2014 June 18, 2014 Depth profile of dissolved CH 3 I at SPOT (spring upwelling) pM CH 3 I 0 2 4 6 8 Depth (m) 0 50 100 150 400 600 800 Mar 12, 2014 Apr 10, 2014 May 21, 2014 Figures 2a-b. Depth profiles of dissolved CH 3 Br and CH 3 I at San Pedro Ocean Time-series, June 2013 to June 2014. Profiles are split between summer-fall-winter (left) and spring upwelling (right) seasons. Note depth scale is condensed below150m. A. Klein 56 Depth profile of dissolved CHBr 3 at SPOT (spring upwelling) pM CHBr 3 0 5 10 15 20 25 Depth (m) 0 50 100 150 400 600 800 Mar 12, 2014 Apr 10, 2014 May 21, 2014 Depth profile of dissolved CHBr 3 at SPOT (summer-fall-winter) pM CHBr 3 0 5 10 15 20 25 Depth (m) 0 50 100 150 400 600 800 June 19, 2013 July 18, 2013 Aug 14, 2013 Sept 18, 2013 Oct 15, 2013 Nov 13, 2013 Jan 15, 2014 Feb 12, 2014 June 18, 2014 Depth profile of dissolved CH 2 Br 2 at SPOT (spring upwelling) pM CH 2 Br 2 0 2 4 6 8 10 12 14 16 Depth (m) 0 50 100 150 400 600 800 Mar 12, 2014 Apr 10, 2014 May 21, 2014 Depth profile of dissolved CH 2 Br 2 at SPOT (summer-fall-winter) pM CH 2 Br 2 0 2 4 6 8 10 12 14 16 Depth (m) 0 50 100 150 400 600 800 June 19, 2013 July 18, 2013 Aug 14, 2013 Sept 18, 2013 Oct 15, 2013 Nov 13, 2013 Jan 15, 2014 Feb 12, 2014 June 18, 2014 Figures 3a-b. Depth profiles of dissolved CHBr 3 and CH 2 Br 2 at San Pedro Ocean Time-series, June 2013 to June 2014. Profiles are split between summer-fall-winter (left) and spring upwelling (right) seasons. Note depth scale is condensed below150m. A. B. Klein 57 CH 3 Br CH 3 I CHBr 3 CH 2 Br 2 SPOT (this study) 0.87 - 12 0.30 - 6.4 1.1 - 22 0.47 - 15 Coastal Antarctic (Carpenter et al. 2012) 2.3 - 4.2 44 - 78 Bay of Bengal (Yamamoto et al. 2001) 1.2 - 2.5 59 - 83 51 - 100 CoastalArabian Sea (Roy 2010) 1.2 - 1100 6.9 - 890 N-S Atlantic transect (Chuck et al. 2005) 0.20 - 19 0.10 - 31 Pacific surface transects (Butler et al. 2007) 0.7 3.1 1.6 NW Atlantic (Moore and Tocarzyk 1993) 0.70 - 15 3.9 - 61 0.35 Range (or median when no range given) of dissolved halocarbons in pmol L -1 Table1. Comparison of SPOT halocarbon concentration ranges with reported values from the literature. Klein 58 Sum of dissolved CHBr 3 + CH 2 Br 2 (spring upwelling) pM CHBr 3 + CH 2 Br 2 0 5 10 15 20 25 30 35 Depth (m) 0 50 100 150 400 600 800 Mar 12, 2014 Apr 10, 2014 May 21, 2014 Sum of dissolved CHBr 3 + CH 2 Br 2 (summer-fall-winter) pM CHBr 3 + CH 2 Br 2 0 10 20 30 Depth (m) 0 50 100 150 400 600 800 June 19, 2013 July 18, 2013 Aug 14, 2013 Sept 18, 2013 Oct 15, 2013 Nov 13, 2013 Jan 15, 2014 Feb 12, 2014 June 18, 2014 Figures 4. Depth profiles of the sum of dissolved CHBr 3 and CH 2 Br 2 at San Pedro Ocean Time- series, June 2013 to June 2014. Profiles are split between summer-fall-winter (left) and spring upwelling (right) seasons. Note depth scale is condensed below150m. Klein 59 Halocarbon concentration ranges (top 50m) over the time series at SPOT CH 3 Br (pmoL L -1 ) 2 4 6 8 10 CH 3 I (pmoL L -1 ) 1 2 3 4 5 6 Jun Jul Aug Sept Oct Nov Jan Feb Mar Apr May June CHBr 3 (pmoL L -1 ) 5 10 15 20 25 CH 2 Br 2 (pmoL L -1 ) 2 4 6 8 10 12 Figure 5. Box plots of monthly halocarbon concentrations for the top 50m, SPOT June 2013 – June 2014. Klein 60 Jun Jul Aug Sept Oct Nov Jan Feb Mar Apr May June CHBr 3 nmol m -2 day -1 2 4 6 8 10 12 14 16 CH 2 Br 2 nmol m -2 day -1 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 Measured halocarbon flux and extrapolated heterotrophic bacterial production integrated to 20m CH 3 Br nmol m -2 day -1 2 4 6 8 10 12 14 16 18 20 Measured flux Extrapolated heterotrophic production CH 3 I nmol m -2 day -1 20 30 40 50 60 70 80 44 280 35 130 Figure 6. Halocarbon sea-air fluxes calculated using satellite wind-speed derived piston velocities (Reuer et al. 2007) and estimation of potential heterotrophic production (Chapter 2) integrated for the 20m surface mixed layer. CHBr 3 production extrapolated from Vibrio AND4 (Chapter 2) cultures was less than 1 nmol m -2 day -1 for all months. Klein 61 Sea-air flux or extrapolated production in mixed layer (20m depth), nmol m -2 day -1 0.01 0.1 1 10 100 1000 CHBr 3 Synechococcus Phaeocystis Diatoms Literature SPOT fluxes Comparison of observed fluxes, reported literature flux values, and extrapolations from culture studies CH 3 Br Heterotrophs Literature SPOT fluxes CH 2 Br 2 Literature SPOT fluxes CH 3 I Prochlorococcus Phaeocystis Diatoms Heterotrophs Literature SPOT fluxes Figure 7. Box-plot comparison of observed SPOT halocarbon fluxes with reported literature values and estimates of 20m mixed layer production based on culture studies (Chapter 2 this thesis, Brownell et al. 2010, Johnson et al. 2012, Moore et al. 1996, Scarratt and Moore 1996). Common log scale. Klein 62 Diffusion-advection modeling, March SPOT CHBr 3 CHBr 3 (pM) 8 10 12 14 16 18 20 22 Depth (m) 0 10 20 30 40 50 Model output, 30 day simulation Measured March SPOT concentrations Model initial conditions Diffusion-advection modeling, March SPOT CH 2 Br 2 CH 2 Br 2 (pM) 2 4 6 8 10 12 Depth (m) 0 10 20 30 40 50 Model output, 30 day simulation Measured March SPOT concentrations Model initial conditions Figure 8. Diffusion-advection model results (30 day simulation runtime) for SPOT March 2014 CHBr 3 and CH 2 Br 2 , compared to measured concentrations and the uniform mixed layer starting condition. Klein 63 Depth Temp Salinity Fluoresence Oxygen CH 3 Br CH 3 I CH 2 Br 2 CHBr 3 m C ppt mV μM pM pM pM pM 6/19/2013 2 19.4 33.61 0.05 241.16 4.40 2.24 2.06 5.80 10 18.43 33.61 0.10 250.10 4.10 1.84 2.48 5.40 20 15.2 33.55 0.70 276.89 3.50 1.62 2.79 12.00 30 14 33.5 3.90 276.89 8.30 2.16 5.80 15.00 40 11.9 33.51 2.30 218.83 3.00 1.63 7.60 16.30 50 11 33.54 0.80 187.57 1.80 1.02 4.40 9.30 60 10.79 33.58 0.50 160.78 1.10 0.48 1.80 5.50 150 8.92 34.05 -0.04 92.89 0.93 0.31 0.39 1.20 250 8.65 34.25 -0.02 41.98 0.87 0.26 0.32 0.89 500 6.64 34.29 -0.01 16.08 0.76 0.28 0.35 1.04 750 5.31 34.38 0.00 4.02 0.84 0.17 0.29 0.78 885 5.19 34.39 0.00 2.68 0.75 0.24 0.40 0.86 7/18/2013 2 20 33.62 0.06 236.70 3.60 2.32 1.82 4.28 10 19.52 33.61 0.09 236.70 3.28 1.91 1.56 7.54 20 16.4 33.6 0.25 263.49 3.42 1.78 1.72 7.82 30 12.5 33.51 5.50 263.49 5.20 1.36 1.89 10.20 40 11.3 33.53 2.50 187.57 2.60 1.34 4.62 12.30 50 10.8 33.58 1.00 160.78 2.32 0.87 2.10 8.24 60 10.4 33.62 0.50 147.38 1.22 0.65 1.42 3.64 150 9.71 33.9 -0.03 107.18 0.88 0.25 0.43 0.97 250 9.08 34.25 0.00 45.55 1.10 0.28 0.28 1.12 500 6.95 34.28 -0.01 19.65 0.72 0.32 0.32 0.84 750 5.36 34.38 -0.01 4.02 0.97 0.15 0.36 0.89 885 5.19 24.39 0.00 2.23 0.81 0.22 0.40 0.87 8/14/2013 2 18.5 33.61 0.08 250.10 4.40 1.44 2.06 4.44 10 18.16 33.61 0.20 250.10 3.28 1.04 2.48 3.82 22.9 13 33.5 2.50 276.90 5.54 1.22 5.80 9.62 30 11.7 33.55 4.70 214.37 6.28 2.00 2.79 11.63 40 11.2 33.51 1.10 200.97 3.24 1.48 7.60 9.88 50 10.9 33.54 0.60 169.71 1.48 1.12 4.40 8.00 60 10.44 33.58 0.30 160.78 0.92 0.52 1.80 4.28 150 9.54 34.05 -0.02 89.32 0.90 0.29 0.39 1.03 250 8.92 34.25 -0.01 46.89 0.85 0.24 0.32 0.84 500 6.57 34.29 -0.02 14.74 0.79 0.29 0.35 0.92 750 5.51 34.37 0.00 4.91 0.88 0.18 0.29 0.76 885.2 5.21 34.39 -0.01 2.23 0.72 0.22 0.40 0.84 Table 2. SPOT profile raw data, June – Aug 2013 Klein 64 Depth Temp Salinity Fluoresence Oxygen CH 3 Br CH 3 I CH 2 Br 2 CHBr 3 m C ppt mV μM pM pM pM pM 9/18/2013 2 19.2 33.61 0.05 250.10 4.53 2.22 2.62 4.93 10 17.8 33.61 0.20 263.50 3.53 1.92 2.28 4.27 20 14.7 33.55 0.25 281.36 2.80 1.94 3.63 9.96 37 12.2 33.50 2.50 223.31 10.38 2.66 5.63 12.30 40 12.08 33.51 1.90 214.37 3.81 1.66 9.50 14.28 50 11.7 33.54 0.90 187.58 1.76 0.96 4.66 9.11 60 11.2 33.58 0.45 165.25 1.09 0.42 2.29 6.38 100 10.29 34.05 -0.02 98.25 1.19 0.23 0.37 1.25 250 8.81 34.25 -0.02 58.06 1.02 0.24 0.23 0.91 500 6.8 34.29 -0.02 16.08 0.81 0.22 0.30 1.15 750 5.32 34.37 0.00 3.13 0.81 0.22 0.26 0.94 885.2 5.21 34.39 0.00 1.79 0.63 0.23 0.45 0.80 10/15/2013 2 19.3 33.60 0.09 245.64 4.93 1.97 1.92 5.22 10 18.6 33.60 0.26 245.64 4.43 2.02 2.53 6.10 20 16.8 33.50 0.26 263.50 4.68 1.44 2.71 12.00 33 13.7 33.36 3.55 259.03 8.72 2.12 6.21 13.80 40 12.9 33.39 1.18 232.24 2.61 1.63 7.75 17.44 50 12.2 33.43 1.50 209.91 1.96 0.93 4.49 10.04 60 11.6 33.47 0.34 192.04 1.09 0.49 1.53 4.73 100 10.45 33.63 -0.02 142.92 0.82 0.35 0.41 1.27 250 9.04 34.17 -0.02 62.53 0.77 0.25 0.35 0.88 500 6.59 34.29 -0.01 13.84 0.73 0.25 0.30 0.99 750 5.34 34.38 0.00 4.02 0.76 0.16 0.26 0.66 886 5.22 34.39 0.00 2.23 0.64 0.21 0.36 0.81 11/13/2013 2 17.67 33.59 0.40 245.64 3.79 1.71 2.62 4.80 10 17.18 33.56 0.65 245.64 4.21 1.98 2.53 8.05 20 17.08 33.55 0.80 245.64 4.45 2.33 1.89 8.52 30 15.9 33.44 0.85 254.57 7.49 2.65 8.38 9.52 36 14.5 33.37 0.85 259.03 3.13 1.48 9.69 13.95 50 12.7 33.40 0.60 209.91 1.22 0.89 3.68 10.24 60 12.1 33.43 0.45 192.04 0.88 0.57 1.04 4.07 100 10.74 33.54 -0.01 160.78 0.71 0.24 0.53 0.89 250 8.73 34.16 -0.03 59.85 0.82 0.21 0.43 0.82 500 6.65 34.29 -0.02 14.74 0.96 0.26 0.41 0.76 750 5.33 34.38 -0.01 3.13 0.85 0.21 0.25 0.75 886 5.27 34.39 0.00 2.68 0.81 0.23 0.45 0.74 Table 3. SPOT profile raw data, Sept – Nov 2013 Klein 65 Depth Temp Salinity Fluoresence Oxygen CH 3 Br CH 3 I CH 2 Br 2 CHBr 3 m C ppt mV μM pM pM pM pM 1/15/2014 2 17.67 33.59 0.29 245.64 4.06 2.88 3.18 6.24 10 17.18 33.56 0.66 245.64 4.23 2.12 2.25 9.45 20 17.08 33.55 0.82 245.64 4.53 1.82 3.07 10.39 30 15.9 33.44 0.73 254.57 5.25 2.20 6.38 7.43 36 14.5 33.37 0.73 259.03 3.95 2.27 8.79 15.35 50 12.7 33.4 0.77 209.91 1.91 1.00 5.12 11.06 60 12.1 33.43 0.36 192.04 0.77 0.59 1.40 6.10 100 10.74 33.54 -0.01 160.78 1.03 0.19 0.72 0.67 250 8.73 34.16 -0.02 59.85 1.07 0.23 0.56 1.02 500 6.65 34.29 -0.03 14.74 1.01 0.34 0.61 0.72 750 5.33 34.38 -0.01 3.13 1.09 0.35 0.30 0.80 886 5.27 34.39 0.00 2.68 0.93 0.28 0.62 0.97 2/12/2014 2 17.67 33.59 -0.02 245.64 4.34 3.20 3.11 7.49 10 17.18 33.56 0.04 245.64 5.65 1.97 3.81 7.59 20 17.08 33.55 0.51 245.64 4.67 1.61 3.74 9.36 25 14.5 33.37 0.23 254.57 6.61 2.13 6.57 10.25 30 15.9 33.44 1.13 259.03 5.05 2.22 8.79 14.43 50 12.7 33.4 0.38 209.91 1.64 0.67 5.52 8.52 60 12.1 33.43 0.68 192.04 0.78 0.56 1.67 4.94 100 10.74 33.54 -0.03 160.78 0.89 0.13 0.72 0.91 250 8.73 34.16 -0.03 59.85 0.95 0.26 0.36 0.73 500 6.65 34.29 -0.02 14.74 0.75 0.32 0.66 0.62 750 5.33 34.38 -0.01 3.13 1.09 0.33 0.48 0.67 886 5.27 34.39 -0.02 2.68 0.84 0.27 0.36 0.67 3/12/2014 2 17.67 33.59 -0.02 245.64 5.04 2.73 5.48 8.54 10 17.18 33.56 0.04 245.64 7.00 2.43 3.52 21.06 20 17.08 33.55 0.60 245.64 4.76 2.96 3.74 15.81 30 15.9 33.44 0.35 254.57 6.48 3.49 7.82 16.91 35 14.5 33.37 1.40 259.03 8.33 3.09 11.25 12.84 50 12.7 33.4 0.95 209.91 1.86 0.80 5.30 9.54 60 12.1 33.43 0.55 192.04 1.14 0.62 2.12 7.86 100 10.74 33.54 -0.04 160.78 1.30 0.18 1.16 1.34 250 8.73 34.16 -0.02 59.85 0.97 0.29 0.43 1.21 500 6.65 34.29 -0.02 14.74 1.18 0.25 0.37 1.20 750 5.33 34.38 -0.01 3.13 1.19 0.29 0.42 1.12 886 5.27 34.39 -0.02 2.68 0.76 0.26 0.39 1.09 Table 4. SPOT profile raw data, Jan – Mar 2014 Klein 66 Depth Temp Salinity Fluoresence Oxygen CH 3 Br CH 3 I CH 2 Br 2 CHBr 3 m C ppt mV μM pM pM pM pM 4/10/2014 2 17.67 33.59 0.80 245.64 5.24 7.10 6.10 10.08 10 17.18 33.56 0.25 245.64 5.61 3.54 6.03 22.11 20 17.08 33.55 0.80 245.64 6.10 3.86 4.64 22.45 23 14.5 33.37 3.00 259.03 8.42 4.93 10.94 17.93 30 15.9 33.44 0.85 254.57 12.17 3.09 14.85 16.31 50 12.7 33.4 0.50 209.91 2.30 1.04 6.47 10.88 60 12.1 33.43 0.35 192.04 1.54 0.87 3.14 8.02 100 10.74 33.54 -0.02 160.78 1.61 0.21 1.52 1.01 250 8.73 34.16 -0.01 59.85 1.30 0.30 0.38 1.00 500 6.65 34.29 -0.02 14.74 1.25 0.28 0.43 0.90 750 5.33 34.38 -0.01 3.13 0.99 0.32 0.42 1.06 886 5.27 34.39 -0.01 2.68 1.08 0.34 0.51 1.30 5/21/2013 2 18.535 33.6 0.03 243.40 4.82 4.67 4.08 7.94 10 17.805 33.585 0.15 247.87 4.86 2.69 4.26 13.76 20 16.14 33.55 1.20 261.26 4.80 2.74 3.72 17.23 27 14.25 33.435 9.50 267.96 8.36 3.54 8.37 16.46 30 13.9 33.475 1.10 236.70 7.58 2.36 11.23 16.30 50 11.85 33.47 0.55 198.74 2.05 1.03 5.44 10.09 60 11.445 33.505 0.45 176.41 1.32 0.68 2.47 6.76 100 9.83 33.795 -0.02 126.84 1.27 0.26 0.95 1.20 250 8.69 34.205 -0.02 50.91 1.08 0.28 0.35 1.05 500 6.645 34.29 -0.02 15.41 1.26 0.28 0.69 0.97 750 5.32 34.38 -0.02 3.57 1.01 0.25 0.51 0.92 886 5.23 34.39 -0.02 2.68 0.91 0.29 0.54 1.08 6/18/2014 2 19.4 33.2739 0.05 255.63 6.86 2.02 2.84 3.77 10 19.1672 33.2739 0.11 227.59 7.01 2.43 2.41 2.92 20 14.592 32.5435 0.67 287.97 5.78 1.90 3.79 14.04 27 14.7 32.495 3.51 296.27 8.05 2.66 7.54 16.20 40 11.305 32.8398 2.37 212.27 3.42 2.79 12.01 15.97 50 10.89 34.5462 0.87 170.69 2.81 1.05 6.95 7.44 60 11.0058 33.9158 0.49 168.81 0.96 0.86 1.71 5.01 100 9.366 34.05 -0.04 100.32 1.17 0.41 0.35 1.18 250 8.996 33.565 -0.02 45.34 1.00 0.35 0.29 0.58 500 6.5736 33.9471 -0.01 15.76 1.31 0.22 0.42 0.67 750 5.31 34.0362 0.00 3.90 0.97 0.23 0.32 0.94 886 5.3976 34.0461 0.00 2.47 0.94 0.21 0.45 0.89 Table 5. SPOT profile raw data, Apr – June 2014 Klein 67 Calculated sea-air flux, nmol m -2 day -1 Month CH 3 Br CH 3 I CHBr 3 CH 2 Br 2 Jun 16.2 47 9.42 -0.8 Jul 9.4 12 5.44 2.56 Aug 5.68 13 -0.68 4.48 Sept 6.72 28 3.28 -3 Oct 9.4 19 2.1 -1.6 Nov 4.5 12.6 1.28 2.4 Jan 7.22 14.6 3.2 2.92 Feb 2.5 18.2 2.93 2.6 Mar 13.4 57 29 3.12 Apr 10.2 102 6.52 -1.2 May 14.8 34 16.8 1.48 Layer depth (m) CHBr 3 CH 2 Br 2 CHBr 3 CH 2 Br 2 2 14.52 3.33 14.93 3.24 4.5 16.44 3.62 16.76 3.65 7 18.21 3.81 18.38 4.05 9.5 19.56 3.99 19.58 4.45 12 18.74 4.21 18.60 4.86 14.5 17.99 4.53 17.70 5.30 17 17.30 4.97 16.88 5.75 19.5 16.65 5.54 16.10 6.22 22 16.01 6.21 15.34 6.68 24.5 15.35 6.92 14.58 7.10 27 14.65 7.62 13.83 7.47 29.5 13.88 8.21 13.07 7.75 32 13.06 8.65 12.31 7.93 34.5 12.23 8.84 11.60 7.97 37 11.75 8.47 11.25 7.57 39.5 11.24 7.94 10.88 7.12 42 10.75 7.30 10.51 6.64 44.5 10.29 6.61 10.15 6.16 47 9.87 5.90 9.82 5.69 49.5 9.54 5.30 9.54 5.30 Model output 15 days 30 days Table 6. Calculated sea-air fluxes at SPOT for the surface 50m, derived from measured halocarbon concentrations and satellite windspeed derived piston velocities. Table 7. Diffusion-advection model output at 15 and 30 days runtime, pmol L -1 . Klein 68 Chapter 4: Heterotrophic bacterial production and seasonality of marine biogenic halocarbons: implications and directions for future research Introduction Halocarbon gases, produced primarily by biology in the marine environment (WMO 2007a), play important and climate-relevant roles in the chemistry of both the troposphere (von Glasgow et al. 2004) and stratosphere (Sturges et al. 2000). Their significance continues to increase relative to analogous anthropogenic chlorofluorocarbons (CFCs) as atmospheric levels of CFCs decline following restrictions on their usage. As a result, marine biogenic halocarbons are expected to account for the majority of halogen-mediated stratospheric ozone depletion by 2050 (WMO 2007b). Despite their growing relevance and the observation of unprecedented ozone depletion over the Arctic in recent years (Manley et al. 2011), their sources, sinks, and underlying biochemical mechanisms are poorly understood. This thesis has generated several important new observations regarding production and seasonality of halocarbons, namely: 1) in culture, marine heterotrophic bacteria appear to universally produce methyl halides in a growth-dependent fashion, and 2) at the coastal upwelling San Pedro Ocean Time-series (SPOT) station, halocarbon concentrations and sea-air fluxes were broadly similar to more open-ocean literature values (notably lower than those observed above oxygen minimum zones) and correspond with enhanced productivity during the upwelling season as well as diatom and likely heterotrophic bacterial abundance. Though numerous uncertainties remain, this chapter will explore the implications of these results and avenues for future research. Klein 69 Heterotrophic bacterial production Though biosynthesis of halogenated methanes has been reported in macroalgae (Gschwend et al. 1985, Kupper et al. 1998), a number of important eukaryotic phytoplankton (especially diatoms) (Scarratt and Moore 1996, 1997; Hill and Manley 2008), and the cyanobacteria Synechococcus (Johnson et al. 2011) and Prochlorococcus (Brownell et al. 2010), no previous studies have addressed the potential for halocarbon production in marine heterotrophic bacteria. As molecular techniques have improved in recent decades and allowed for better quantification and identification of marine bacterioplankton, it has emerged that heterotrophic bacteria typically represent the dominant fraction of biomass in the marine environment and play central ecological roles (Morris et al. 2002). I have reported, for the first time, growth-rate dependent production of the methyl halides CH 3 Cl, CH 3 Br, and CH 3 I in all of a taxonomically broad and environmentally relevant selection of marine heterotrophic bacteria (This thesis, Chapter 2). When extrapolated to annual global fluxes, heterotrophic production can plausibly support the observed sea-air methyl halide fluxes at SPOT (SPOT chapter Figure 6) and likely represents a significant and previously unrecognized proportion of global halocarbon production (SPOT chapter Table 1). The growth-rate dependence of observed halocarbon production in these cultures implies that biosynthesis may be constitutive and part of the cell’s basic metabolic and reproductive machinery, such that halocarbon production is continuous throughout the cell cycle. That the observed biosynthetic products were almost exclusively methyl halides (low levels of CHBr 3 were detected in Vibrio AND4 only) suggests a methylation reaction as the underlying biochemical mechanism (Toda and Itoh, 2011) rather than haloperoxidases which would yield only poly-halogenated methanes (Butler and Carter-Franklin 2004). Several hypotheses have Klein 70 been proposed to explain halide biomethylation. They may be excreted as a grazing deterrent (Butler and Carter-Franklin, 2004; Gschwend, 1985; Manley 2002) or a means of osmoregulation by removing excess halide ions (Ni and Hager, 1999). However, there is no direct evidence to support either of these arguments, and the low picomolar concentrations of methyl halides in seawater would seem to contradict their role in salt balance or predation avoidance. Manley (2002) proposed that methyl halide biosynthesis was instead a metabolic “accident,” simply enzymes with low substrate specificity performing halogenation as a side product. Although logically consistent, this mechanism cannot account for all marine methyl halide biosynthesis, as activity of halide/thiol methyltransferases (HTMTs) has been observed in a number of eukaryotic phytoplankton (Ohsawa et al, 2001, Yokouchi et al. 2014) and higher plants (Itoh et al. 2009). Although thiol is the favored substrate of this enzyme in plants (Attieh et al. 1995), measured HTMT kinetics indicate HTMT instead favors halides in phytoplankton (Ohsawa et al. 2001), that the preferred methyl donor in vivo is S-adenosyl methionine (SAM), that halocarbon production rate is dependent primarily on iodide substrate concentration, and that the enzyme does not contain a bound metal cofactor (Yokouchi 2014). The potential for halomethane production by methylating agents other than SAM, such as the osmolyte dimethylsulfoniopropionate (DMSP) and methionine (Urhahn and Ballschmiter 1998) has been evaluated in several studies. In all cases, enzyme activity with SAM was at least an order of magnitude greater than with the other methylating agents (Itoh et al. 2009, Toda and Itoh 2011, Yokouchi 2014). Interestingly, some organisms can perform the reverse reaction and utilize methyl halides as methylating agents to regenerate s-adenosylhomocysteine to SAM. Some debate exists as to whether methionine and therefore SAM could have been synthesized Klein 71 abiotically under early Earth conditions (Waddell et al. 2000), and it is possible that methyl halides may act physiologically as methyl carriers. While its function in vivo is unknown and no annotated HTMT genes were found in the genomes of the surveyed heterotrophic bacteria, all contained several genes with high sequence homology that are likely related methyltransferases, suggesting SAM-dependent HTMT or similar enzyme systems may be responsible for halocarbon biosynthesis in heterotrophs. If this is the case, it is expected that environmental availability of halides (especially the most favorable substrate iodide) as well as methionine, vitamin B 12 , and other SAM precursors may control rates of methyl halide biosynthesis. The observed low-level production of CHBr 3 by only the cultures of Vibrio AND4 (Chapter 2) may be due to haloperoxidase enzymes, however none are annotated in the genomes of the assayed heterotrophic bacteria. If the annotations are correct and heterotrophs do not possess haloperoxidases, the observed CHBr 3 production may instead be due to the ubiquitous acid phosphatase enzymes which exhibit high active site sequence homology to haloperoxidases and limited haloperoxidase activity in vivo (Tanaka et al. 2002). Climate perspective: Past and present Halocarbons, as climate-relevant gases with primarily biological sources which are becoming more important as CFC concentrations decline, their response to changes in the global ocean system as a result of anthropogenic climate change is of particular interest. Based on previous research and new insights provided by this thesis, the factors most relevant to global halocarbon production are likely biosynthesis by haloperoxidase in diatoms (particularly sea-ice and polar species) (Sturges et al. 2000), biosynthesis by heterotrophic bacteria (Chapter 2) Klein 72 general ocean productivity, and the extent of oxygen-minimum zones (OMZs) owing to the order of magnitude higher halocarbon fluxes and high iodide concentrations typically found in OMZs (SPOT table 1, Roy et al. 2011). Table 1 presents these factors and corresponding hypothetical signs of changes in global annual halocarbon production for the last glacial maximum (LGM) and future ocean under climate change scenarios, relative to the present day. During the LGM, sea-ice extent was much greater and sea-surface temperatures lower at high latitudes, though mid-latitudes and the tropics especially were similar in temperature and salinity to today (Benway et al. 2006). Shutdown of formation of North Atlantic Deep Water in the LGM radically altered heat and nutrient transport throughout the global ocean, resulting in generally reduced productivity and upwelling, although the extent of oxygen minimum zones (OMZs) was likely similar to today (McManus et al. 2004). If not more abundant, diatoms were relatively more dominant than in modern oceans, and extensive sea-ice cover provided greater habitat for sea-ice dwelling diatoms (Benway et al. 2006). Global halocarbon production in the LGM may have been lower overall due to decreased ocean productivity, but that may be balanced to an extent by enhanced contribution by diatoms at high latitudes. Conversely, studies on the impact of anthropogenic climate change on global oceans in the near future (Table 1) suggest that in addition to higher sea-surface temperatures, global ocean productivity will likely decrease, increased winds may enhance upwelling in many regions (Sydeman et al. 2014), and large-scale ecological shifts will favor heterotrophic bacteria (Poloczankta et al. 2013). Diatoms, and especially polar and sea-ice species which are likely to be affected disproportionately by climate change, are expected to decrease in relative abundance and play a less dominant role (Edwards and Richardson, 2004). Oxygen minimum zones, where the highest halocarbon flux measurements (Roy et al. 2011) and high iodide concentrations (to Klein 73 support potential methyl halide production) (Chance et al. 2014) may be found, have been expanding in recent decades (Capone and Hutchins 2013, Gilly et al. 2013). Relative to the LGM and present-day scenarios, halocarbon production in the future ocean may be enhanced at low latitudes and in expanding OMZs and more dominated by heterotrophs. Our present understanding of halocarbon sources in the context of the future global ocean suggest a potentially more heterotroph-dominated and low-latitude (perhaps associated with OMZs) source signal. As halocarbons produced at low latitudes are more easily transported into the stratosphere via the intertropical convergence zone, this might suggest a stronger biogenic halocarbon impact on stratospheric ozone depletion in the future. In contrast, conditions during the LGM likely favored a greater relative contribution from diatoms, especially at high-latitudes. Outlook and directions for future research The new observations contributed by this thesis research only reinforce the need for continued and intensive study of the distribution, sources, and biochemistry of marine biogenic halocarbons. Despite playing important roles in atmospheric chemistry and likely increasing in abundance and/or shifting in global production patterns due to climate change, our knowledge of them remains lacking to nonexistent in many areas. The biochemical mechanisms for halocarbon production, while well-studied for haloperoxidases, are similarly lacking for methyltransferases and their methyl halide products. These are the primary species produced by heterotrophic bacteria in culture (Chapter 2), and heterotrophic production may increase relative to the contribution from other marine sources (diatoms, Table 1). There is therefore a particular need for further research into the biochemistry and functional role of methyltransferases in marine heterotrophic bacteria. Klein 74 The culture experiments presented in this thesis were conducted in enriched media and consist of only a relatively small selection of organisms—assays of heterotrophic halocarbon production in a minimal media would allow for manipulation experiments involving physical and chemical variables such as light, temperature, and nutrient concentrations. Coupled with molecular techniques to determine halide methyltransferase activity, this would give insight into the contribution and importance of HTMT enzymes in heterotrophic halocarbon production as well as more quantitative measures of the effects of environmental parameters on halocarbon biosynthesis. Additionally, more quantification of potential production of halocarbons by other heterotrophs as well as the many eukaryotic phytoplankton that have not been assayed for halocarbon production will allow for better extrapolation of culture results to the environment and a better understanding of the interplay between ecological succession and halocarbon biosynthesis. Armed with a better understanding of the underlying biochemical mechanisms, further culture manipulation and field incubation studies could evaluate the relevance of availability of trace metals, organic cofactors, and iodide on halocarbon production. As algal halide methyltransferase activity has observed to be SAM-dependent, if this is also the chief mechanism for production of methyl halides in heterotrophs it might be expected that the distribution of biomethylation cofactors such as B 12 (often undetectable in the oceans, Sañudo-Wilhelmy et al. 2012), methionine, and cobalt to some degree will control heterotrophic halocarbon synthesis. The SAR-11 clade of ubiquitous marine heterotrophs, for example, require exogenous sources of reduced sulfur (including methionine) (Tripp et al. 2008) and therefore any production by this group is likely dependent on the availability of methionine or other biomethylating agents. In addition to laboratory work, field data and particularly depth profiles and seasonal studies are Klein 75 lacking in the literature. Water overlying oxygen minimum zones might be regions of specific sampling interest, particularly during bloom events as that would allow for studying the succession dynamics and cofactor chemistry of the bloom in a likely highly halocarbon- productive area. Klein 76 Source or forcing Last glacial maximum Future ocean Diatoms +? - Diatoms (sea-ice) + - Heterotrophic bacteria -? + Ocean productivity - -? Oxygen minimum zone extent =? +? Upwelling - +? Hypothesized sign of change in halocarbon production relative to present-day Table 1. Proposed sign of changes in global annual halocarbon production as a result of changes in corresponding sources or factors, relative to present-day. Last glacial maximum = 18kya. Data from Benway et al. 2006, Capone and Hutchins 2013, McManus et al. 2004, Sydeman et al. 2014. Klein 77 References Amachi, S., Kamagata, Y., Kanagawa, T., & Muramatsu, Y. (2001). Bacteria mediate methylation of iodine in marine and terrestrial environments. Applied and environmental microbiology, 67(6), 2718-2722. Attieh, J.M., Hanson, A.D., and Saini, H.S. (1995). Purification and Characterization of a Novel Methyltransferase Responsible for Biosynthesis of Halomethanes and Methanethiol in Brassica oleracea. The Journal of Biological Chemistry 270(16): 9250-9257. Baker, J.M., et al. (1999). Biological production of methyl bromide in the coastal waters of the North Sea and open ocean of the northeast Atlantic. Marine Chemistry 64: 267-285. Benway, H. M., Mix, a. C., Haley, B. A. & Klinkhammer, G. P. Eastern Pacific Warm Pool paleosalinity and climate variability: 0-30 kyr. Paleoceanography 21 (2006). Brownell, D. K., Moore, R.M., and Cullen, J. J. (2010). Production of methyl halides by Prochlorococcus and Synechococcus. Global Biogeochemical Cycles 24 (2) Butler, A. and Carter-Franklin, J.N. (2004). The role of vanadium bromoperoxidase in the biosynthesis of halogenated marine natural products. Nat. Prod. Rep. 21: 180-188. Butler, J., et al. (2007). Oceanic distributions and emissions of short-lived halocarbons. Global Biogeochemical Cycles 21:GB1023. Capone, D.G., and Hutchins, D.A. (2013) Microbial biogeochemistry of coastal upwelling regimes in a changing ocean. Nature Geoscience 6: 711-717 Carlson, C. A., Ducklow, H. W., Hansell, D. A., and Smith, W. O. J. (1998) Organic carbon partitioning during spring phytoplankton blooms in the Ross Sea polynya and the Sargasso Sea. Limnol Oceanogr 43: 375-386 Atkinson, H. M., Huang, R. J., Chance, R., Roscoe, H. K., Hughes, C., Davison, B., ... & Liss, P. S. (2012). Iodine emissions from the sea ice of the Weddell Sea. Atmospheric Chemistry and Physics, 12(22), 11229-11244. Chance, R., Baker, A. R., Carpenter, L., & Jickells, T. D. (2014). The distribution of iodide at the sea surface. Environmental Science: Processes & Impacts, 16(8), 1841-1859. Chuck, A. L., Turner, S. M., & Liss, P. S. (2005). Oceanic distributions and air ‐sea fluxes of biogenic halocarbons in the open ocean. Journal of Geophysical Research: Oceans (1978–2012), 110(C10). Coulter, C., et al. (1999). Halomethane:bisulfide/halide ion methyltransferase, an unusual corrinoid enzyme of environmental significance isolated from an aerobic methylotroph using chloromethane as the sole carbon source. Applied and Environmental Microbiology 65(10): 4301-4312. Klein 78 Countway, P.D., Vigil, P., Schnetzer, A., Moorthi, S., Caron, D.A. (2010). Seasonal analysis of protistan community structure and diversity at the USC Microbial Observatory (San Pedro Channel, Pacific Ocean). Limnol. and Oceanogr. 55: 2. Crans, D. C., Smee, J. J., Gaidamauskas, E., and Yang, L. (2004). The chemistry and biochemistry of vanadium and the biological activities exerted by vanadium compounds. Chem.Rev. 104, 849–902. de la Cuesta, J.L. and Manley, S.L. (2009). Iodine assimilation by marine diatoms and other phytoplankton in nitrate-replete conditions. Limnol. Oceanography. 54(5): 2653-1664. Ducklow, H. W., & Harris, R. P. (1993). Introduction to the JGOFS North Atlantic bloom experiment. Deep Sea Research Part II: Topical Studies in Oceanography, 40(1), 1-8. Ducklow, H. W., Quinby, H. L., & Carlson, C. A. (1995). Bacterioplankton dynamics in the equatorial Pacific during the 1992 El Nino. Deep Sea Research Part II: Topical Studies in Oceanography, 42(2), 621-638. Edwards, M., & Richardson, A. J. (2004). Impact of climate change on marine pelagic phenology and trophic mismatch. Nature, 430(7002), 881-884. Fuhrman, J. A., Hewson, I., Schwalbach, M. S., Steele, J. A., Brown, M. V., & Naeem, S. (2006). Annually reoccurring bacterial communities are predictable from ocean conditions. Proceedings of the National Academy of Sciences, 103(35), 13104-13109. Gebhardt, S. (2008). Biogenic Emission of Halocarbons. PhD dissertation, Johannes Gutenberg-Universtät, Mainz, Germany. Gilbert, J.A., Steele, J., Caporaso, J.G., Steinbruck, L., Reeder, J., Temperton, B., Huse, S., Joint, I., McHardy, A.C., Knight, R., Somerfield, P., Fuhrman, J.A., Field, D. (2012). Defining seasonal marine microbial community dynamics. The ISME Journal 6: 298. Gilly, W.F., Beman, J.M., Litvin, S.Y, and Robinson, B.H. (2013) Oceanographic and Biological Effects of Shoaling of the Oxygen Minimum Zone. Ann Rev Mar Sci 5: 393-420 de Gruijl, F. and Leun, J. (2000). Environment and health: 3. Ozone depletion and ultraviolet radiation Canadian Medical Association Journal 163: 851-855 Ducklow, H. W., Kirchman, D. L., Quinby, H. L., Carlson, C. A., and Dam, H. G. (1993) Stocks and dynamics of bacterioplankton carbon during the spring phytoplankton bloom in the eastern North Atlantic Ocean. Deep-Sea Res 40: 245 – 263 Ducklow, H. W., Quinby, H. L., and Carlson, C. A. (1995) Bacterioplankton dyanmics in the equatorial Pacific during the 1992 El Nino. Deep-Sea Res. II 42: 621-638 Dyrssen, D., & Fogelqvist, E. (1981). Bromoform concentrations of the Arctic Ocean in the Svalbard area. Oceanologica Acta, 4(3), 313-317. Klein 79 Elliott, S. and Rowland, F. S. (1993). Nucleophilic substitution rates and solubilites for methyl halides in seawater. Geophysical Research Letters 20 (11): 1043-1046 Fortin, Michelle A. (1981). A numerical model for the prediction of the dispersion of certain passive discharges in the marine environment. MS thesis, University of Southern California Fuhrman, J.A., Hewson, I., Schwalback, M.S., Steele, J.A., Brown, M.V., Naeem, S. (2006). Annually reoccurring bacterial communities are predictable from ocean conditions. PNAS 103(35): 13104-13109 Gebhardt, S. (2008). Biogenic Emission of Halocarbons. PhD dissertation, Johannes Gutenberg-Universtät, Mainz, Germany. Gilly, W. F., Beman, J. M., Litvin, S. Y., & Robison, B. H. (2013). Oceanographic and biological effects of shoaling of the oxygen minimum zone. Annual Review of Marine Science, 5, 393-420. Gschwend, P.M., Macfarlane, J.K., and Newman, K.A. (1985). Volatile Halogenated Organic Compounds Released to Seawater from Temperate Marine Macroalgae. Science 227(4690): 1033-1035. Happell, J.D. and Douglas, W.R. (1996). Methyl iodide in the Greenland/Norwegian Seas and the tropical Atlantic Ocean: Evidence for photochemical production. Geophysical Research Letters 23(16): 2105-2108. Harper, D. B. (2000) The global chloromethane cycle: biosynthesis, biodegradation, and metabolic role. Natural Product Reports 17: 337-348 Hatton, A. D., Shenoy, D. M., Hart, M. C., Mogg, A., & Green, D. H. (2012). Metabolism of DMSP, DMS and DMSO by the cultivable bacterial community associated with the DMSP-producing dinoflagellate Scrippsiella trochoidea. Biogeochemistry, 110(1-3), 131-146. Helz, G.R. and Hsu, R.Y. (1978). Volatile chloro- and bromocarbons in coastal waters. Limnol. Oceanogr. 23(5): 858-869. Hobbie, J. E., Daley, R. J., and Jasper, S. (1977). Use of nucleopore filters for counting bacteria by fluorescence microscopy. Applied and environmental microbiology 33(5): 1225-1228 Hill, V.L. and Manley, S.L. (2009). Release of reactive bromine and iodine from diatoms and its possible role in halogen transfer in polar and tropical oceans. Limnol. Oceanogr. 54(3): 812-822. Hunter ‐Smith, R. J., Balls, P. W., & Liss, P. S. (1983). Henry's Law constants and the air ‐sea exchange of various low molecular weight halocarbon gases. Tellus B, 35(3), 170-176. Itoh, N., et al. (1997). Formation and emission of monohalomethanes from marine algae. Phytochemistry 45(1): 67-73. Klein 80 Jacob, D.J. (1999). Introduction to Atmospheric Chemistry. Princeton University Press. Jakopitsch, C., Regelsberger, G., Furtmüller, P. G., Rüker, F., Peschek, G. a, & Obinger, C. (2001). Catalase-peroxidase from synechocystis is capable of chlorination and bromination reactions. Biochemical and biophysical research communications, 287(3), 682-7. doi: 10.1006/bbrc.2001.5616. Johnson, T. L., Palenik, B., and Brahamsha, B. (2011). Characterization of a functional vanadium-dependent bromoperoxidase in the marine cyanobacterium Synechococcus Sp. CC9311. J. Phycol. 47, 792–801. Kavanaugh, M. C., & Trussell, R. R. (1980). Design of aeration towers to strip volatile contaminants from drinking water. Journal (American Water Works Association), 684- 692. Kirchman, D. L, and Ducklow, H. W. (1993) Estimating conversion factors for the thymidine and leucine methods for measuring bacterial production. In P. Kemp, B. Sherr, E. Sherr, and J. J. Cole, eds Handbook of Methods in Microbial Ecology. Lewis Publishers, Boca Raton, FL, pp 513-518. Ichikawa, K., Kurihara, M., Tamegai, H., & Hashimoto, S. (2015). Decomposition of brominated organic halogens by cultures of marine proteobacteria: Phaeobacter, Roseobacter, and Rhodobacter. Marine Chemistry, 176, 133-141. Klein, N.J., Beck, A.J., Hutchins, D.A., Sañudo-Wilhelmy, S.A. (2013) “Regression modeling of the North East Atlantic Spring Bloom suggests previously unrecognized biological roles for V and Mo.” Frontiers in Microbiology. Kuwabara, J.S., and North, W.J. (1980). Culturing microscopic stages of Macrocystis pyrifera (Phaeophyta) in Aquil, a chemically defined medium. J. Phycol. 16: 546-549. Küpper, F.C., Schweigert, N, Ar Gall, E., Legendre, J.M., Vilter, H., and Kloareg, B. (1998). Iodine uptae in Laminariales involves extracellular, haloperoxidase-mediated oxidation of iodide. Planta 207: 163-171. Li, H.J., Yokouchi, Y., and Akimoto, H. (1999). Measurement of methyl halides in the marine atmosphere. Atmospheric Environment 33: 1881-1887. Liu, Y., Yvon-Lewis, S. A., Thornton, D. C. O., Butler, J. H., Bianchi, T. S., Campbell, L. H., and Smith, R. W. (2013). Spatial and temporal distributions of bromoform and dibromomethane in the Atlantic Ocean and their relationship with photosynthetic biomass. Journal of Geophysical Research: Oceans 118 (8): 3950-3965 Manley, S.L. (2002). Phytogenesis of halomethanes: A product of selection or a metabolic accident? Biogeochemistry 60: 163-180. Klein 81 Manoj, K. M., & Hager, L. P. (2006). A colorimetric method for detection and quantification of chlorinating activity of hemeperoxidases. Analytical biochemistry, 348(1), 84-6. doi: 10.1016/j.ab.2005.10.014. Markowitz, V. M., Chen, I. M. A., Palaniappan, K., Chu, K., Szeto, E., Grechkin, Y., ... & Huntemann, M. (2012). IMG: the integrated microbial genomes database and comparative analysis system. Nucleic acids research, 40(D1), D115-D122 Markowitz, V. M., Chen, I. M. A., Palaniappan, K., Chu, K., Szeto, E., Pillay, M., ... & Anderson, I. (2013). IMG 4 version of the integrated microbial genomes comparative analysis system. Nucleic acids research, gkt963. McManus, J., Francois, R. & Gherardi, J.-M. Collapse and rapid resumption of Atlantic meridional circulation linked to deglacial climate changes. Nature 428, 834–7 (2004) Moore, R.E. (1977). Volatile compounds from marine algae. Acc. Chem. Res. (10): 40-47. Moore, R.M. (2008). A Photochemical Source of Methyl Chloride in Saline Waters. Environ. Sci. Technol. 42: 1933-1937. Moore, R.M. and Groszko, W. (1999). Methyl iodide distribution in the ocean and fluxes to the atmosphere. Journal of Geophysical Research 104(C5): 11,163-11,171. Moore, R.M., Webb, M., Tokarczyk, R., and Wever, R. (1996). Bromoperoxidase and iodoperoxidase enzymes and production of halogenated methanes in marine diatom cultures. Journal of Geophysical Research 101(C9): 20,899-20,908. Moore, R.M. and Zafiriou, O.C. (1994). Photochemical production of methyl iodide in seawater. Journal of Geophysical Research 99(D8): 16,415-16,420. Levine, N. M., Varaljay, V. A., Toole, D. A., Dacey, J. W., Doney, S. C., & Moran, M. A. (2012). Environmental, biochemical and genetic drivers of DMSP degradation and DMS production in the Sargasso Sea. Environmental microbiology, 14(5), 1210-1223. Morris, R. M., Rappé, M. S., Connon, S. A., Vergin, K. L., Siebold, W. A., Carlson, C. A., & Giovannoni, S. J. (2002). SAR11 clade dominates ocean surface bacterioplankton communities. Nature, 420(6917), 806-810. Ni, X. and Hager, L. (1999). Expression of Batis maritima methyl chloride transferase in Escherichia coli. PNAS 96(7): 3611-3615. Nightingale, P. D., C. S. Law, A. J. Watson, P. S. Liss, M. Liddicoat, J. Boutin, and R. C. Upstill-Goddard (2000). In situ evaluation of air-sea gas exchange parameterizations using novel conservative and volatile tracers. Global Biogeochem. Cycle, 14(1), 373–387 Nuester, J., Vogt, S., Newville, M., Kustka, A.B., and Twining, B.S. (2012). The Unique Biogeochemical Signature of the Marine Diazotroph Trichodesmium. Front Microbiol. 3: 150 Klein 82 Ohsawa, N., Tsujita, M., Morikawa, S., & Itoh, N. (2001). Purification and characterization of a monohalomethane-producing enzyme S-adenosyl-L-methionine: halide ion methyltransferase from a marine microalga, Pavlova pinguis. Bioscience, biotechnology, and biochemistry, 65(11), 2397-2404. Panzeca, C., Beck, A. J., Leblanc, K., Taylor, G. T., Hutchins, D. a, & Sañudo-Wilhelmy, S. a. (2008). Potential cobalt limitation of vitamin B 12 synthesis in the North Atlantic Ocean. Global Biogeochemical Cycles, 22(2), 1-7. doi: 10.1029/2007GB003124. Passardi F., Theiler G., Zamocky M., Cosio C., Rouhier N., Teixera F., Margis-Pinheiro M., Ioannidis V., Penel C., Falquet L., Dunand C. (2007). PeroxiBase: The peroxidase database. Phytochemistry (12):1605-11. Poloczanska, E. S., Brown, C. J., Sydeman, W. J., Kiessling, W., Schoeman, D. S., Moore, P. J., ... & Duarte, C. M. (2013). Global imprint of climate change on marine life. Nature Climate Change, 3(10), 919-925. Quinn, P.K., and Bates, T.S. (2011). The case against climate regulation via oceanic phytoplankton sulphur emissions. Nature 480: 51-56. Read, K.A., et al. (2008). Extensive halogen-mediated ozone destruction over the tropical Atlantic Ocean. Nature 453: 1232-1235. Reuer, M. K., B. A. Barnett, M. L. Bender, P. G. Falkowski, and M. B. Hendricks (2007), New estimates of Southern Ocean biological production rates from O 2 /Ar ratios and the triple isotope composition of O 2 , Deep-Sea Res., 54, 951–974. Redeker, K.R., Manley, S.L., Walser, M., and Cicerone, R.J. (2004). Physiological and biochemical controls over methyl halide emissions from rice plants. Global Biogeochemical Cycles 18: GB1007 Roy, R., Pratihary, A., Narvenkar, G., Mochemadkar, S., Gauns, M., and Naqvi, S.W.A. (2011). “A study of volatile halocarbons in relation to phytoplankton pigments during a Trichodesmium bloom in the coastal eastern Arabian Sea.” Estuarine, Coastal, and Shelf Science, vol. 95; 110-118. Sañudo-Wilhelmy, S. A., Cutter, L. S., Durazo, R., Smail, E. A., Gómez-Consarnau, L., Webb, E. A., ... & Karl, D. M. (2012). Multiple B-vitamin depletion in large areas of the coastal ocean. Proceedings of the National Academy of Sciences, 109(35), 14041-14045. Sañudo-Wilhelmy, S.A., Gomez-Consarnau, L., Suffridge, C.P., and Webb, E.A. (2014). The Role of B Vitamins in Marine Biogeochemistry. Annual Review of Marine Science 6: 339- 37. Scarratt, M.G. and Moore, R.M. (1996). Production of methyl chloride and methyl bromide in laboratory cultures of marine phytoplankton. Marine Chemistry 54: 263-272. Scarratt, M.G and Moore, R.M. (1997). Production of methyl bromide and methyl chloride in laboratory cultures of marine phytoplankton II. Marine Chemistry 59: 311-320. Klein 83 Schall, C. and Heumann, K. (1993). GC determination of volatile organoiodine and organobromine compounds in Arctic seawater and air samples. Fresenius J Anal Chem 346: 717-722. Sturges, W.T., Cota, G.F., and Buckley, P.T. (1992). Bromoform emission from Arctic ice algae. Nature 358: 660-662. Sturges, W.T., Oram, D.E., Carpenter, L.J., and Penkett, S.A. (2000). Bromoform as a source of stratospheric bromine. Geophysical Research Letters 27(14): 2081-2084 Sydeman, W. J., García-Reyes, M., Schoeman, D. S., Rykaczewski, R. R., Thompson, S. A., Black, B. A., & Bograd, S. J. (2014). Climate change and wind intensification in coastal upwelling ecosystems. Science, 345(6192), 77-80. Tanaka, N., Dumay, V., Liao, Q., Lange, A.J., Wever, R. (2002) Bromoperoxidase activity of vanadate-substituted acid phosphatases from Shigella flexneri and Salmonella enterica ser. Typhimurium. European Journal of Biochemistry 269 (8): 2162-2167 Toda, H., & Itoh, N. (2011). Isolation and characterization of a gene encoding a S-adenosyl-L- methionine-dependent halide/thiol methyltransferase (HTMT) from the marine diatom Phaeodactylum tricornutum: Biogenic mechanism of CH 3 I emissions in oceans. Phytochemistry, 72(4), 337-343. Tripp, H. J., Kitner, J. B., Schwalbach, M. S., Dacey, J. W., Wilhelm, L. J., & Giovannoni, S. J. (2008). SAR11 marine bacteria require exogenous reduced sulphur for growth. Nature, 452(7188), 741-744. Tovar-Sanchez A., Sañudo-Wilhelmy S. A. (2011). Influence of the Amazon River on dissolved and intra-cellular metal concentrations in Trichodesmium colonies along the western boundary of the sub-tropical North Atlantic Ocean. Biogeosciences 8, 217–225 Urhahn, T. and Ballschmiter, K. (1998). Chemistry of the biosynthesis of halogenated methanes: C1-organohalogens as pre-industrial chemical stressors in the environment? Chemosphere 37(6): 1017-1032. von Glasow, R., von Kuhlmann, R., Lawrence, M. G., Platt, U., and Crutzen, P. J. (2004). Impact of reactive bromine chemistry in the troposphere. Atmospheric Chemistry and Physics 4: 2481-2497. Wang, D., and Sañudo-Wilhelmy, S.A. (2009). Vanadium speciation and cycling in coastal waters. Marine Chemistry 117(1-4): 52-58 Waddell, T. G., Eilders, L. L., Patel, B. P., & Sims, M. (2000). Prebiotic methylation and the evolution of methyl transfer reactions in living cells.Origins of Life and Evolution of the Biosphere, 30(6), 539-548. Wanninkhof, R. (1992), Relationship between gas exchange and wind speed over the ocean. J. Geophys. Res. 97(C5), 7373 –7381 Klein 84 White, R. H. (1982) Analysis of Dimethyl Sulfonium Compounds in Marine Algae, Journal of Marine Research (40): 529-536, 1982. Wilhelm, E., Battino, R., & Wilcock, R. J. (1977). Low-pressure solubility of gases in liquid water. Chemical reviews, 77(2), 219-262. WMO (2003). Scientific Assessment of Ozone Depletion 2002, Chapter 2: Very Short-Lived Halogen and Sulfur Substances, Global Ozone Research and Monitoring Project -Report No. 47 World Meteorological Organization, Geneva, Switzerland: 2.1-2.57. WMO (2007a). Scientific Assessment of Ozone Depletion 2006, Chapter 2: Halogenated Very Short-Lived Substances, Global Ozone Research and Monitoring Project – Report No. 50, World Meteorological Organization, Geneva, Switzerland: 2.1-2.57. WMO (2007b). Scientific Assessment of Ozone Depletion 2006, Chapter 8: Halocarbon Scenarios, ODPs ans GWPs, Global Ozone Research and Monitoring Project -Report No. 50, World Meteorological Organization, Geneva, Switzerland: 8.1-8.39. Wong, A., Piumsomboon, U., and Dunstan, W.M. (2002). The transformation of iodate to iodide in marine phytoplankton. Mar. Ecol. Prog. Ser. 237: 27-39. Wright, D. A., Sandler, S. I., & DeVoll, D. (1992). Infinite dilution activity coefficients and solubilities of halogenated hydrocarbons in water at ambient temperatures. Environmental science & technology, 26(9), 1828-1831. Yamamoto, H., Yokouchi, Y., Otsuki, A., & Itoh, H. (2001). Depth profiles of volatile halogenated hydrocarbons in seawater in the Bay of Bengal.Chemosphere, 45(3), 371- 377. Yokouchi, Y., Ooki, A., Hashimoto, S., & Itoh, N. (2014). A Study on the Production and Emission of Marine-Derived Volatile Halocarbons. M. Uematsu, Y. Yokouchi, YW Watanabe, S. Takeda, and Y. Yamanakaeds., Western Pacific Air-Sea Interaction Study, TERRAPUB, 1-25. Klein 85 Appendix 1: Regression modeling of a North East Atlantic Spring Bloom transect dataset suggests a previously unrecognized biological role for Mo and V Published in Frontiers in Microbiology Nick Klein, Department of Earth Sciences, University of Southern California, Los Angeles, CA, USA Aaron Beck, Department of Physical Sciences, Virginia Institute of Marine Sciences, Gloucester Point, VA, USA David A. Hutchins, Department of Biological Sciences, University of Southern California, Los Angeles, CA, USA Sergio Sañudo-Wilhelmy, Department of Biological Sciences and Department of Earth Sciences, University of Southern California, Los Angeles, CA, USA 1. Abstract In order to identify the physico-chemical parameters controlling pCO 2 , total chlorophyll a, and dimethylsulfide (DMS) concentrations during the North East Atlantic Spring Bloom (NASB), we used previously unpublished particulate and dissolved elemental concentrations to construct several linear regression models; first by hypothesis-testing, and then with exhaustive stepwise linear regression followed by leave-one-out cross-validation. The field data was obtained along a latitudinal transect from the Azores Islands to the North Atlantic, and best-fit models (determined by lowest predictive error) of up to three variables are presented. Total chlorophyll a is predicted best by biomass (POC, PON) parameters and by pigments characteristic of picophytoplankton for the southern section of the transect (from the Azores to the Rockhall-Hatton Plateau) and coccolithophores in the northern portion (from the Rockhall- Hatton Plateau to the Denmark Strait). Both the pCO 2 and DMS models included variables traditionally associated with the development of the NASB such as mixed-layer depth and with the Fe and Si-deplete conditions (dissolved Fe, dissolved and biogenic silica, dissolved PO 4 3- ). Klein 86 However, the regressions for pCO 2 and DMS also include intracellular V and Mo concentrations, respectively. Mo is involved in DMS production as a cofactor in dimethylsulfoxide reductase. No significant biological role for V has yet been determined, although particulate V is significantly correlated (p-value < 0.05) with biogenic silica (R 2 = 0.72) and total chlorophyll a (R 2 = 0.49) while the same is not true for its biogeochemical analogue Mo, suggesting active uptake of V by phytoplankton. Our statistical analysis suggests these two lesser-studied metals may play more important roles in bloom dynamics than previously thought, and highlights a need for studies focused on determining their potential biological requirements and cell quotas. Keywords: trace nutrients, North Atlantic Spring Bloom, B-vitamins, vanadium, molybdenum 2. Introduction The North Atlantic Spring Bloom (NASB) is a large annual phytoplankton bloom event triggered by a decrease in mixed-layer depth in March or April. It is typically characterized by early domination of diatoms, depletion of dissolved Si, and later succession by coccolithophores and other non-silicifying organisms (Sieracki et al, 2003). The dynamics of the NASB strongly influence the partial pressure of carbon dioxide (pCO 2 ) in the region (Ducklow and Harris, 1993). The bloom is of particular interest in light of global climate change, owing to its status as a significant sink for anthropogenic CO 2 (Gruber, 2006). The NASB 2005 program set as its goals to describe the phytoplankton community structure during the late stages of the NASB and determine relative contributions of the major phytoplankton taxa (e.g., diatoms and coccolithophores) in export of carbon and biominerals (LeBlanc et al, 2009). The NASB 2005 cruise yielded a large amount of data, including a broad spectrum of phytoplankton pigments, atmospheric CO 2 , DMS, and trace metal and B-vitamins Klein 87 (B 12 and B 1 ) concentration data. This paper includes previously unpublished field trace metal dissolved and intracellular metals data, which is not abundant in the literature for that geographical region. In aggregate, the NASB 2005 data set represents a broad-based and extensive sampling of trace cofactors and phytoplankton pigments yet produced for that region. This publication aims to utilize the trace metal and B-vitamin data in combination with pigment and other environmental data to more fully describe nutrient limitation conditions observed during the 2005 NASB cruise, as well as to employ correlative statistical methods to produce predictive models describing any relationships between pCO 2 , chlorophyll a, and DMS with the wealth of other variables in the dataset. The three variables were selected to explore the relationship between primary production (represented by chlorophyll a) and production of the climactically important gases CO 2 and DMS. 3. Area of study Sampling was conducted from 6 June to 3 July 2005 aboard the R/V Seaward Johnson II along a south-to-north transect of the northeast Atlantic Ocean (Figure 1), generally following the 20°W meridian. Real-time satellite data was monitored during the cruise, and the route adjusted slightly to sample areas where satellite data indicated coccolithophore blooms. 4. Methods Near-surface seawater (5-10 m depth) was pumped onboard using an acid-washed all- Teflon trace-metal clean pumping system (Osmonics Bruiser) extended away from the ship on a boom. Water was pumped directly into a trace metal clean van and filtered through a 0.22 µm acid-washed polypropylene capsule filter directly into 1 L acid-washed LDPE bottles. Dissolved Klein 88 trace metal samples were acidified to pH < 2 with 6 N quartz-distilled HCl (Optima-grade) and preconcentrated following Bruland et al. (1985). Particulate samples for metals determination was filtered onto duplicate acid-washed polycarbonate filter membranes (0.2 µm pore size) from between 0.13 and 4 L of seawater, depending on plankton abundance. For total metals content, particulates collected on one of the filters was rinsed with Chelex-cleaned trace metal-free seawater. For intracellular metals quotas, biomass on the second filter was washed to remove surface-adsorbed metals using 10 mL of oxalate reagent (Tovar-Sanchez et al., 2003), although the reagent was not cleaned prior to use. Instead, biomass was rinsed following the oxalate wash with 5 aliquots of 10 mL Chelex-cleaned trace metal-free seawater (Tang and Morel, 2006). To monitor the rinse efficiency and confirm that there was no contamination from the oxalate reagent, a blank filter was subjected to the oxalate wash and rinse procedure with every sample (n=36). Phytoplankton biomass was digested with 2 mL aqua regia and 50 µL HF (all acids Optima-grade). The digests were evaporated to dryness, and the residue taken up in 2 mL of 1 N Optima-grade HNO 3 . Dissolved trace metal extracts and filter digests were analyzed by high resolution inductively-coupled plasma mass spectrometry (ICPMS; ThermoFisher Element 2) using indium as an internal standard. The ancillary dataset was compiled from surface transect data (depth = 10m) presented in LeBlanc et al (2009) for a total of 51 variables across 27 surface transect stations. Dissolved trace metal and nutrient data were compared to published literature stoichiometry to assess potential limitation. All statistical work was performed in the R statistical analysis program (R Development Core Team, 2010). Shorthand abbreviations (e.g. Klein 89 DIC for dissolved inorganic carbon) for each variable are used in the figures presented here, and a key for their interpretation may be found in Appendix A. Missing values were estimated using nearest-neighbor imputation (Hastie et al, 2010) and the data scaled and centered. Hypothesis-driven regression models were constructed from variables of interest identified using existing literature (e.g. mixed-layer depth as a trigger for the NASB) and from the nutrient stoichiometry analyses. Following hypothesis-driven analysis, an exhaustive stepwise linear regression algorithm (Lumley 2009) was employed and statistically significant regressions of up to three variables were retained for further consideration. As stepwise linear regression amounts to data mining and introduces the risk of Type III statistical errors (formulating hypotheses suggested from the data), leave-one-out cross-validation was performed to aid in selection of linear regression models better reflective of real trends and not data noise (Canty and Ripley, 2010). 5. Results and Discussion Prior to statistical analysis, the dataset was subdivided into two sections on the basis of a distinct surface salinity and temperature front separating the southern transect from the northern transect section (see LeBlanc et al, 2009, Fig. 2). This hypothesis was tested via application of k-means clustering (R Development Core Team, 2010) on the dataset, which produced two main clusters divided by the observed front, confirming the hypothesis. The following results and discussion consider the northern and southern transect sections separately, with the southern section stations (n = 13) corresponding to the waters over the Porcupine Abyssal Plain and Rockhall-Hatton Plateau, while the northern section (n = 14) represents those stations north of station number 23 and from waters overlying the Icelandic Basin and Shelf. Klein 90 Dissolved and intracellular metal concentrations for bioactive trace metals considered in the nutrient limitation and stoichiometry (section 5.1) calculations (Fe, Cu, Co, Cd, and Mo) are presented in Figures 2 and 3, respectively. A distinct concentration gradient was observed for dissolved Fe and Co, generally increasing northward (from 0.5-1 nmol L -1 and 20-35 pmol L -1 respectively), with a sharp peak observed in the Denmark Straight influenced by ice melt-waters (2 nmol L -1 Fe and 80 pmol L -1 Co) (Figure 2). Dissolved Mo and Cd ranged from 116-137 nmol L -1 and from 0.58-0.74 nmol L -1, respectively, with neither element displaying a clear latitudinal trend (Figure 2). Intracellular trace metal concentrations (Figure 3) were plotted with typical literature phytoplankton cellular quota values derived from laboratory culture experiments (Ho et al, 2003) as dashed lines for reference. Intracellular Fe concentrations ranged from less than 0.01 to 0.14 mmol • mol -1 P and were generally below or near the typical literature culture value of 7.5 mmol • mol -1 P, with stations at the extreme south and north of the transect being enriched by an order of magnitude. Intracellular Cu concentrations ranged from 0.02 to 1.61 mmol • mol -1 P, and were generally below the literature value of 0.38 mmol • mol -1 P in the northern transect, excepting portions of the southern transect where they were enriched twofold. In contrast, observed intracellular Co concentrations (0.01 to 0.16 mmol • mol -1 P) were below the typical laboratory culture value of 0.19 mmol • mol -1 P at all stations. Intracellular Cd levels (0.02 to 0.35 mmol • mol -1 P )are generally below or slightly above the literature value of 0.21 mmol • mol -1 P, while Mo (0.01 to 0.52 mmol • mol -1 P) is enriched above the culture value of 0.03 mmol • mol -1 P at most stations. 5.1 Nutrient Limitation Stoichiometry To assess nutrient limitation and the relative importance of the various elements during the 2005 NASB cruise, median ratios between measured dissolved nutrients and typical Klein 91 intracellular concentrations derived from laboratory culture experiments are presented in Table 1. In this treatment of nutrient limitation, a ratio value greater than one suggests the element is enriched in the dissolved phase relative to stoichiometric requirements and should therefore not be limiting, while a ratio value of less than one is indicative of a nutrient that is depleted relative to (and more limiting than) P. Dissolved inorganic N:P is near Redfield stoichiometry for both transect sections with values (Table 1) of 0.89 and 0.93 for the southern and northern sections, respectively. Literature data on B-vitamin requirements for phytoplankton is very limited, and so the median of cellular stoichiometry values compiled in Tang et al (2010) was used. On a stoichiometric basis, B 12 would appear to be present in excess, with ratios of 11.0 and 5.9 (Southern and Northern transect sections, respectively) representing for auxotrophic phytoplankton along the entire transect, while B 1 is slightly replete or near literature stoichiometric values, with ratios of 2.0 and 1.0, respectively. Both vitamins are enriched relative to literature requirements by about a factor of two in the southern transect relative to the northern transect. When observed B-vitamin concentrations are compared to literature K s half-saturation constants for growth (Figure 4), B 12 again appears replete while observed B 1 concentrations are an order of magnitude lower than literature K s values. As B 1 is required for 49% of assayed dinoflagellate species, 15% of diatoms, and 83% of coccolithophores (Tang et al, 2010), it may be limiting to growth rates of those taxa and may therefore selectively favor prototrophic species. Our analysis suggested that phytoplankton in the NASB was not limited by Mo and Cd as both elements appear replete (ratio >1). Conversely, low median ratio values for Co of 0.49 and 0.56 (Southern and Northern portions of the transect, respectively) and for Fe of 0.34 (same median value for both transect portions) is indicative of potential depletion and limitation. There Klein 92 is other evidence for Fe as a likely a limiting or co-limiting element, as previous studies have demonstrated Fe limitation both during the development of the NASB (Moore et al 2006) and during post-bloom conditions (Nielsdóttir et al, 2009), and Fe addition experiments during the 2005 NASB cruise stimulated chlorophyll a concentrations above control (LeBlanc et al, 2009). Overall, the comparison of nutrient stoichiometric ratios support the conclusions of LeBlanc et al (2009) that the NASB at the time of sampling was in its late stages and had progressed beyond initial diatom dominance, which is reflected in the strong depletion of and limitation by dissolved Si. The stoichiometry supports previous findings of mid- and post-bloom Fe limitation in the North Atlantic, and indicates that Co and vitamin B 1 (thiamine) may also have been limiting or co-limiting at the time of sampling. 5.2 Linear regression modeling of pCO2, chlorophyll a, and DMS Linear regression models for pCO 2 , chlorophyll a, and DMS were constructed first with hypothesis-testing based on potential nutrient limitation as discussed in section 5.1 (dissolved Si, inorganic N, Fe, B 1 , and Co) and with mixed-layer depth, which is classically thought to trigger the NASB (Ducklow and Harris, 1993). Asterisked variable combinations in Tables 2, 3, and 4 (variable abbreviations defined in Appendix) denote statistically significant models constructed from this initial hypothesis-testing. Following this, a stepwise linear regression algorithm (Lumley, 2009) was employed to exhaustively calculate polynomial regressions versus the three response variables for all possible combinations of up to three variables. Statistically significant models were retained and leave- one-out cross-validation performed (Canty and Ripley, 2010) to minimize overfitting. These models and relevant statistical metrics are presented in Tables 2, 3, and 4 with the best-fit model Klein 93 bolded. Best-fit regressions for pCO 2 , DMS, and chlorophyll a for both Northern and Southern transect sections yield good fits and are plotted versus observed field data in Figure 5. These models and relevant statistical metrics are presented in Tables 2, 3, and 4 with the best-fit model in boldface font. 5.3 pCO 2 Modeling The best-fit models for pCO 2 (Table 2, Figure 5) involves DOC (dissolved organic carbon), PERI (peridinin, a pigment characteristic of dinoflagellates), and QV (intracellular vanadium concentrations) for the southern transect and DFe (dissolved Fe), Zm (mixed-layer depth), and BSi (biogenic silica) for the northern transect section. For the south, DOC and intracellular V in particular are present in many of the pCO 2 regression models. DOC alone yields a statistically significant (p-value < 0.05) regression with pCO 2 with an R 2 of 0.78. During the 1989 Joint Global Ocean Flux Study experiment in the North Atlantic, depth- integrated DOC was found to be 10x greater than POC (particulate organic carbon), and bacterial production was 30% of total primary production (Lochte et al, 1992). The authors hypothesized that this bacterial production likely metabolized a significant amount of DOC, and this microbial utilization of the DOC pool could explain the inclusion of DOC in the pCO 2 models. This would support a causal link between bacterial utilization of the DOC pool, and the inclusion of the dinoflagellate pigment peridinin in the best-fit regression agrees with the observed domination by dinoflagellates of the Southern transect reported in LeBlanc et al (2009). Biological roles for vanadium are not well-understood, but the inclusion of intracellular V in many of the best regression models presented here as well as statistically significant correlations between intracellular V alone and both biogenic silica and chlorophyll a across the Klein 94 entire transect (Figure 6) suggest and important relationship. V and Mo are chemically analogous, so correlation with V but not intracellular Mo (as shown) implies selective uptake of V by biology. V-containing bromoperoxidase activity has been identified in a number of polar and temperate diatoms (Hill and Manley, 2009), but the enzyme’s function and thus potential relation to pCO 2 is not well understood. The best-fit model for pCO 2 in the north contains variables more typically associated with bloom development (mixed-layer depth, biogenic silica) as well as dissolved Fe, which is likely limiting based on stoichiometric ratios presented here. Regression with dissolved Fe alone yields a statistically significant model with an R 2 of 0.38. The two regressions with VIO (violaxanthin, a pigment characteristic of coccolithophores) have predictive errors much greater than the other models and as such are not considered further here. 5.4 Chlorophyll a Modeling Models for chlorophyll a (Table 3, Figure 5) contain mostly biomass variables in the southern transect section (PON, POC, POP, BSi) and chiefly other pigments in the northern transect section (size-fractionated chlorophyll a, eukaryotic accessory pigments chlorophyll c2 and chlorophyll c3). The best-fit model for the southern transect includes particulate organic nitrogen, the nanophytoplankton fraction of chlorophyll a, and alloxanthin, which is a pigment characteristic of cryptophytes (Roy et al, 2006). Regressions containing intracellular V concentrations also occur here, also. The best-fit model for chlorophyll a in the northern transect section includes chlorophyll c3 and 9’hexanoyloxyfucoxanthin (both characteristic of the then- dominant coccolithophores) and violaxanthin, a pigment characteristic of dinoflagellates. Klein 95 5.5 DMS Modeling DMS linear regression models (Table 4, Figure 5) for the southern transect subset include dissolved inorganic nutrients (phosphate, DIN, Si) as well as chlorophyll a. For the northern section, they involve mostly biomass indicators (PON, POC) and intracellular Mo concentrations. The best-fit model for the southern section comprises dissolved silica, chlorophyll a, and mixed-layer depth—all variables associated with the classical NASB progression. For the north, the best-fit model involves Mo and V intracellular levels and POC. As referenced earlier, Mo is a cofactor in DMSO reductase (Schindelin et al, 1996). Mo is relatively more depleted stoichiometrically (Table 1) in the northern transect section than in the south. This along with the inclusion of intracellular Mo in many of the DMS regressions for the north suggests that Mo may be important for the production of DMS. No relationship between V and DMS production has been previously suggested in the literature, and a better understanding of the biological role of V is needed to understand the relationship between V and DMS implied by the inclusion of intracellular V in the best-fit regression for the northern transect. 6. Conclusions The 2005 NASB data analyzed here indicate, on the basis of nutrient stoichiometry, that the bloom was both Si and Fe-limited at the time of sampling, and Co and B 1 (thiamine) concentrations were also potentially limiting . Linear regression modeling confirmed the importance of mixed-layer depth and dissolved Si and Fe concentrations in relation to pCO2 and DMS concentrations. The inclusion of Mo and V intracellular concentrations alongside parameters traditionally of importance in the NASB (mixed layer depth, dissolved Fe, Si) in the models for DMS and pCO 2 , respectively, suggests an important relationship for these lesser- Klein 96 studied trace metals, perhaps particularly in the case of V where biological functions are not well elucidated. Our results suggest that further investigations are needed into the possible linkages between V and phytoplankton biology, and between Mo quotas and DMS production in the oceans. Klein 97 6. References Beck, A.J., Panzeca, C., Hutchins, D.A., LeBlanc, K., Gueguen, C., Sañudo-Wilhelmy, S.A. (2007) Trends in dissolved and intracellular trace metals in the North Atlantic Ocean. Poster presentation, American Society of Limnology and Oceanography meeting. Bruland, K. W.; Coale, K. H.; Mart, L. Analysis of seawater for dissolved Cd, Cu, and Pb-An intercomparison of voltammetric and atomic-absorption methods. Mar. Chem. 1985, 17, 285- 300. Bruland, K. W., R. P. Franks, G. A. Knauer, and J. H. Martin. 1979. Sampling and analytical methods for the determination of copper, cadmium, zinc and nickel at the nanogram per liter level in sea water. Anal. Chim. Acta 105: 233–245. Brzezinski, M.A. (1985). The Si:C:N ratio of marine diatoms: interspecific variability and the effect of some environmental variables. Journal of Phycology, Vo. 21, pp. 347-357 Canty, Angelo and Ripley, Brian (2010). boot: Bootstrap R (S-Plus) Functions. R package version 1.2-43. Ducklow, H. W. and Harris, R. P. (1993): Introduction to the JGOFS North Atlantic bloom experiment, Deep Sea Res. II, 40, 1–8 Gruber, N. (1996): Anthropogenic CO2 in the Atlantic Ocean. Global Biogeochem. Cy., 12, 165–191 Hastie, Trevor; Tibshirani, R; Balasubramanian, N and Chu, G (2010). impute: Imputation for microarray data. R package version 1.24.0. http://CRAN.R-project.org/package=impute Hill, V.L. and Manley, S.L. (2009) Release of reactive bromine and iodine from diatoms and its possible role in halogen transfer in polar and tropical oceans. Limnology and Oceanography 54(3): 812-822. Ho, T.-yuan, Quigg, A., V, Z., Milligan, A. J., Falkowski, P. G., & Morel, M. M. (2003). THE ELEMENTAL COMPOSITION OF SOME MARINE PHYTOPLANKTON 1 Franc, 1159, 1145-1159. Hutchins, D. A., G. R. DiTullio, Y. Zhang, K. W. Bruland, 1998. An iron limitation mosaic in the California upwelling regime. Limnol. Oceanogr., 43(6), 1037-1054 LeBlanc, K, Hare, CE, Feng, Y., Berg, GM, DiTullio, GR, Neeley, A, Benner, I, Sprengelm C, Beck, A, Sañudo Wilhelmy, SA, Passow, U, Klinck, K, Rowe, JM, Wilhelm, SW, Brown, CW, Hutchins, DA. 2009. Distribution of calcifying and silicifying phytoplankton in relation to environmental and biogeochemical parameters during the late stages of the 2005 North East Atlantic Spring Bloom. Biogeosciences. 6: 2155-2179. Lochte, K., Ducklow, H.W., Fasham, M.J.R., and Stienen, C. (1992). Plankton succession and carbon cycling at 47N 20W during the JGOFS North Atlantic Bloom Experiment. Deep Sea Res. II, 40(1): 91-114 Klein 98 Lumley, T. using Fortran code by Miller, A. <tlumley@u.washington.edu> (2009). leaps: regression subset selection. R package version 2.9. http://CRAN.R- project.org/package=leaps Moore, C. Iron limits primary productivity during spring bloom development in the central North Atlantic. Global change biology. 2006;12:626-634. Nielsdóttir, M. C., Moore, C. M., Sanders, R., Hinz, D. J., & Achterberg, E. P. (2009). Iron limitation of the postbloom phytoplankton communities in the Iceland Basin. Global Biogeochemical Cycles, 23(3), 1-13. doi: 10.1029/2008GB003410. R Development Core Team (2010). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0, URL http://www.R-project.org Roy, R., Pratihary, A. Mangesh, G., Naqvi, S.W.A. (2006). Spatial variation of phytoplankton pigments along the southwest coast of India. Estuarine Coastal and Shelf Science 69: 189-195 Sañudo-Wilhelmy, S.A., Tovar-Sanchez, A., Fu, F.X., Capone, D.G., Carpenter, E.J., and Hutchins, D.A. (2004). The impact of surface-adsorbed phosphorus on phytoplankton Redfield stoichiometry. Nature 432: 897-901. Saito, M., and Moffett, J. (2001). Complexation of cobalt by natural organic ligands in the Sargasso Sea as determined by a new high-sensitivity electrochemical cobalt speciation method suitable for open ocean work. Marine Chemistry 75: 49-68. Schindelin, H., Kisker, C., Hilton, J., Rajagopalan, K., and Rees, D. (1996). Crystal Structure of DMSO Reductase: Redox-Linked Changes in Molybdopterin Coordination. Science 272: 1615-1621/ Sieracki, M. E., Verity, P. G., and Stoecker, D. K. (1993). Plankton community response to sequential silicate and nitrate depletion during the 1989 North Atlantic spring bloom. Deep-Sea Res. II, 40, 213–225 Tang, Y. Z., Koch, F., & Gobler, C.J. (2010). Most harmful algal bloom species are vitamin B1 and B12 auxotrophs. Proceedings of the National Academy of Sciences, 107(48), 20756- 20761. National Acad Sciences. doi: 10.1073/pnas.1009566107/- /DCSupplemental.www.pnas.org/cgi/doi/10.1073/pnas.1009566107. Tang, D., Morel, F.M.M., 2006. Distinguishing between cellular and Fe-oxide-associated trace elements in phytoplankton. Mar. Chem. 98(1), 18-30. Taylor, G.T., & Sullivan, C. W. (2008). Vitamin B12 and cobalt cycling among diatoms and bacteria in Antarctic sea ice microbial communities. Limnol. Oceanogr, 53(5), 1862– 1877. Tovar-Sanchez, A., Sanudo-Wilhelmy, S. A., Garcia-Vargas, M., Weaver, R. S., Popels, L. C., and Hutchins, D. A. 2003. A trace metal clean reagent to remove surface-bound Fe from marine phytoplankton. Mar. Chem., 82: 91–99. Klein 99 Appendix A: Abbreviations of model variables Abbreviation Variable ALLO Alloxanthin B1 Dissolved vitamin B1 (thiamine) B12 Dissolved vitamin B12 (cobalamin) Bact Bacterial abundance BSi Biogenic silica BUT 19'-butanoyloxyfucoxanthin Chla chlorophyll a Chlb chlorophyll b Chlc2 chlorophyll c2 Chlc3 chlorophyll c3 Chlides total chlorophyllides DCd dissolved Cd DCo dissolved Co DCu Dissolved Cu DFe dissolved Fe DIADINO diadinoxanthin DIN dissolve dinorganic nitrogen (NO3- + NO2- + NH4+) DMo dissolved Mo DMS dissolved dimethyl sulfide DNi dissolved Ni DOC dissolved organic carbon DON dissolved organic nitrogen DV dissolved vanadium DZn dissolved zinc FUCO fucoxanthin HEX 19’Hexanoyloxyfucoxanthin pChla pico fraction of chlorophyll a pCO2 partial pressure of CO2 PERI peridinin PFe particulate Fe PIC particulate inorganic carbon PMn particulate Mn PO4 dissolved ortho-phosphate POC particulate organic carbon PON particulate organic nitrogen POP particulate organic phosphorus QCd intracellular quotas of Cd QCo intracellular quotas of Co QCu intracellular quotas of Cu QFe intracellular quotas of Fe QMn intracellular quotas of Mn QMo intracellular quotas of Mo Klein 100 QNi intracellular quotas of Ni QV intracellular quotas of V Si dissolved silicic acid TEP transparent exopolymer particles uChla micro fraction of chlorophyll a VIO violoaxanthin ZEA zeaxanthin Zm depth of the mixed layer Zn depth of the nitracline Klein 101 Figure 1. Transect route, shown with major surface currents. Surface sampling stations are numbered 1-37. Figure from LeBlanc et al (2005). Klein 102 Dissolved Fe and Cu (nmol L -1 ) 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 Dissolved Co (pmol L -1 ) 10 20 30 40 50 60 70 80 90 Fe Cu Co Station number 0 5 10 15 20 25 30 35 40 Dissolved Cd (nmol L -1 ) 0.56 0.58 0.60 0.62 0.64 0.66 0.68 0.70 0.72 0.74 0.76 Dissolved Mo (nmol L -1 ) 115 120 125 130 135 140 Cd Mo Figure 2. Dissolved trace metal concentrations along the NASB transect (depth = 10m). Vertical dashed line separates the Southern from Northern transect sections. Klein 103 Fe (mmol L -1 mol -1 P) 0 20 40 60 80 100 120 Cu (mmol L -1 mol -1 P) 0.0 0.2 0.4 0.6 0.8 1.0 1.2 Co quotas (mmol L- -1 mol -1 P) 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 Fe Cu Co Station number 0 5 10 15 20 25 30 35 40 Cd (mmol L -1 mol -1 P) 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 Mo (mmol L -1 mol -1 P) 0.00 0.05 0.10 0.15 0.20 0.25 Cd Mo Figure 3. Intracellular trace metal quotas along the NASB transect (depth = 10m). Vertical dashed line separates the Southern from Northern transect sections. Horizontal dashed lines are color-coded by element and correspond to typical phytoplankton cellular quotas for that element from Ho et al (2003). All quotas for Co were below the literature value of 0.19. Klein 104 Dissolved / Literature Stoichiometric Ratio (P-standardized) N Si B 12 B 1 Mo Cd DCo Cu Fe Southern Transect 0.93 0.13 11 2.0 24000 17 0.49 13 0.34 Northern Transect 0.89 0.12 5.9 1.0 15000 11 0.56 10 0.34 Table 1. Median ratios of dissolved nutrients vs. stoichiometric values from the literature, standardize to P. Trace metal and N stoichiometry is after Ho et al. (2003), B-vitamin stoichiometric values are from those complied in Tang et al. (2011), and Si values are from Brzezinski (1985). Klein 105 B 12 (pmol L -1 ) 0 1 2 3 4 B 1 (pmol L -1 ) 0 50 100 150 200 Literature K s Southern Transect Northern Transect Figure 4. Comparison of observed B-vitamin concentrations with literature Ks half-saturation constants for growth (Taylor and Sullivan, 2008; Tang et al 2010). Klein 106 Chlorophyll a ( g L -1 ) 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 Station number 0 5 10 15 20 25 30 35 40 DMS (nmol L -1 ) 0 2 4 6 8 10 12 Observed Modeled pCO 2 (ppm) 240 260 280 300 320 340 360 380 400 420 440 North = -1.1 + 20 QMo + 0.25 POC - 3.1 QV R 2 = 0.92, p < 0.01 South = 4.3 -1.4 Si + 2.2 Chla - 0.10 Zm R 2 = 0.89, p < 0.01 North = 0.51 + 0.011 Chlc3 - 0.0016 HEX + 0.0041 VIO R 2 = 0.99, p < 0.01 South = 0.071 + 0.24 POC + 2.5 nChla - 0.025 ALLO R 2 = 0.97, p < 0.01 South = 17 + 5.0 DOC + 510 QV - 0.49 PERI R 2 = 0.94, p < 0.01 North = North = 490 - 49 DFe - 2.7 Zm - 56 BSi R 2 = 0.75, p < 0.01 Figure 5. Observed versus modeled pCO 2 , chlorophyll a, and DMS along the NASB transect. Models graphed are those with lowest predictive error as determined by leave-one-out cross-validation. Formulas, R 2 , and p-values are given for each regression. Klein 107 Table 4. DMS model Model Variables R 2 p PE Southern Transect *Si 0.31 0.05 0.58 DON 0.35 0.03 0.52 *DIN 0.41 0.02 0.51 PO4 Chlides 0.70 0.00 0.63 PO4 BUT 0.71 0.00 0.59 TEP Si 0.72 0.00 0.82 DIN QV 0.74 0.00 0.41 PO4 Chla 0.81 0.00 0.48 PO4 Chla QNi 0.88 0.00 0.39 Si Chla Zm 0.89 0.00 0.22 PO4 Chla QCd 0.90 0.00 0.29 PO4 Chla QCo 0.90 0.00 0.35 PO4 BUT PFe 0.90 0.00 0.50 Northern Transect nChla 0.63 0.00 0.34 HEX 0.67 0.00 0.30 PON 0.68 0.00 0.29 POC 0.71 0.00 0.25 BUT 0.74 0.00 0.23 ALLO POC 0.81 0.00 0.21 QMo PON 0.81 0.00 0.23 QMo POC 0.84 0.00 0.21 QMo BUT 0.84 0.00 0.26 QMo PON DCd 0.91 0.00 0.14 QMo POC DCd 0.91 0.00 0.14 PMn POC DFe 0.92 0.00 0.57 B12 FUCO Chlc2 0.92 0.00 0.12 QMo POC QV 0.92 0.00 0.09 Table 2: pCO2 model Model Variables R 2 p PE Southern Transect POC 0.50 0.00 0.86 DNi 0.51 0.00 0.75 DZn 0.55 0.00 0.78 PON 0.56 0.00 0.83 *DOC 0.78 0.00 0.98 DNi POP 0.81 0.00 0.86 QFe DZn 0.84 0.00 1.73 *DOC QNi 0.85 0.00 1.61 DOC QV 0.85 0.00 1.78 DNi QV QFe 0.93 0.00 3.04 DOC QV QCd 0.94 0.00 1.29 DOC QV PERI 0.94 0.00 0.40 DOC QNi TEP 0.94 0.00 1.46 DZn QV DV 0.95 0.00 1.98 Northern Transect *DFe 0.38 0.02 0.68 VIO Chlb 0.51 0.02 1.10 POP DON DNi 0.72 0.00 0.43 DFe Zm Chlides 0.74 0.00 0.47 DFe Zm BSi 0.75 0.00 0.42 VIO DON DMo 0.80 0.00 1.00 Table 3: Chlorophyll a model Model Variables R 2 p PE Southern Transect Chlides 0.66 0.00 0.42 BSi 0.67 0.00 0.39 QV 0.71 0.00 0.34 PON 0.77 0.00 0.34 POC 0.84 0.00 0.24 POC QV 0.89 0.00 0.19 PON DIADINO 0.89 0.00 0.24 POC uChla 0.89 0.00 0.24 POC DIADINO 0.90 0.00 0.25 POC PFe 0.91 0.00 0.17 POP Bact PFe 0.96 0.00 0.25 POC DCu PFe 0.96 0.00 0.12 POC BUT PFe 0.96 0.00 0.14 PON nChla ALLO 0.97 0.00 0.06 POC nChla ALLO 0.97 0.00 0.09 Northern Transect *DCo 0.48 0.01 0.46 *POP 0.48 0.00 0.50 uChla 0.57 0.00 0.43 ALLO 0.60 0.00 0.38 Chlc2 0.63 0.00 0.33 Chlc2 PFe 0.90 0.00 0.10 Chlc2 QFe 0.90 0.00 0.09 Chlc2 DIADINO 0.93 0.00 0.08 Chlc3 DCo 0.94 0.00 0.08 Chlc3 QFe 0.94 0.00 0.08 Chlc3 VIO nChla 0.97 0.00 0.03 Chlc3 DCo uChla 0.98 0.00 0.03 Chlb pChla nChla 0.98 0.00 0.03 Chlc3 Chlb pChla 0.98 0.00 0.03 Chlc3 HEX VIO 0.99 0.00 0.01 Tables of linear models and diagnostic statistics for 2) pCO 2 , 3) chlorophyll a, and 4) DMS Only statistically significant regressions for up to three variables are presented. Models with the lowest predictive error (PE), determined by leave-one-out cross validation, are bolded. Models produced from hypothesis-testing are marked with an asterisk. Klein 108 Biogenic silica ( mol L -1 ) 0.0 0.2 0.4 0.6 0.8 1.0 1.2 Intracellular metal (mmol mol P -1 ) 0.0 0.2 0.4 0.6 0.8 1.0 Mo V Chlorophyll a ( g L -1 ) 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 Mo V Figure 6. Intracellular V and Mo quotas versus biogenic silica and total chlorophyll a across the entire NASB transect. Klein 109 Appendix 2: Spatial distribution of dissolved B-vitamins and trace metals and their potential for biological limitation in Lake Michigan Klein, N.J., Aguilar, C., Cuhel, R., Cutter, L., and Sañudo-Wilhelmy, S.A. Abstract In order to evaluate the spatial distribution and potential biological effect of dissolved B- vitamins in freshwater systems, samples were taken on 20 July 2009 from an eight-station surface transect on Lake Michigan from Milwaukee, WI and two depth profile stations over the mid-lake Sheboygan Reef Complex. Samples were analyzed for thiamin (vitamin B 1 ), biotin (vitamin B 7 ) and cobalamin (vitamin B 12 ) as well as for the macronutrients N, P, and Si, chlorophyll a, and twelve trace metals. This represents the first data set of dissolved B-vitamins from a freshwater system obtained using modern direct-analysis LC-MS techniques. These methods are a large improvement over older bioassay methods, yielding much lower detection limits and avoiding other complications associated with the bioassays. . Macronutrients and dissolved Fe, Al, and Ti decreased off-shore along the surface transect, suggesting a coastal source for these three metals and nutrients. The three B-vitamin concentrations as well as the other trace metals (Ag, Cd, Co, Cu, Mn, Mo, Ni, Pb, and V) varied throughout the spatial transect. In the depth profiles, none of the trace metal data suggest atmospheric deposition as a source, and Fe, Al, Ti, Co, Mn, Cd, Cu, and Pb appear to have a sedimentary source during our sampling. B-vitamins are generally uncorrelated with other parameters and with each other, though all three were present in the upper water column (top ca. 50m depth) and decreased to undetectable or near-undetectable with depth. B 1 and B 12 concentrations displayed maxima near the deep fluorescence maximum (DFM), a layer where high abundances of putative vitamin producers have previously been reported. B 7 maxima were Klein 110 above the DFM. These vitamin depth distributions are broadly similar to those observed previously in both marine and other freshwater systems. A comparison of median ambient dissolved stoichiometric ratios from Lake Michigan with literature values from phytoplankton cultures suggests the lake was strongly P-limited and relatively replete in N and major bioactive trace metals during our sampling. When compared to published half-saturation constants for growth (K s ), observed concentrations of B 12 are similar or slightly lower, B 1 is significantly lower, and B 7 much greater than literature values. This comparison suggests B 1 and possibly B 12 are limiting to growth rates of auxotrophs in Lake Michigan during our cruise. Alongside the growing body of contemporary literature indicating the importance of B-vitamins in co-limitation of primary production and as regulators of community structure, these results suggest the need for a better understanding of the distribution and organismal requirements of B-vitamins in both marine and freshwater systems. Introduction The importance of trace metals and trace organic cofactors (such as the B-vitamins) as limiting or co-limiting elements in marine systems has been recognized for several decades (Carlucci and Silbernagel 1969, Provasoli 1973). B-vitamins are a group of water-soluble organic cofactors that are vital to many basic biochemical reactions but present in concentrations in the pmol • L -1 range in natural waters (Sañudo-Wilhelmy et al. 2012; 2014). Cobalt- containing B 12 (cobalamin, B 12 hereafter) plays a central role in methionine synthesis and assimilation of inorganic carbon as a cofactor for methylmalonyl CoA mutase (Lindemans and Abels 1985). B 1 (thiamin, B 1 hereafter) is involved in decarboxylation and several other reactions and enzyme systems vital to amino acid synthesis and carbohydrate metabolism Klein 111 (Vandamme 1989). B 7 (biotin, B 7 hereafter) functions as a cofactor for carboxylase enzymes, and thus is involved in metabolism of lipids, carbohydrates, and some amino acids (Dakshinamurti et al. 1985). A majority of assayed eukaryotic phytoplankton cultures are auxotrophic for one or more B-vitamins (Croft et al. 2006; Tang et al. 2010; Sañudo-Wilhelmy et al. 2014), and must therefore obtain them from their environment. Some prokaryotes, especially diazotrophs (Bonnet 2010), but also non-diazotrophic cyanobacteria and heterotrophic bacteria, are important producers of B-vitamins in the marine environment (Provasoli 1974; Provasoli and Carlucci 1974; Sañudo-Wilhelmy et al. 2014). Auxotrophs and higher trophic levels are then ultimately reliant on some prokaryote and eukaryote B-vitamin synthesizers (Sanudo-Wilhelmy et al. 2014), and relationships between prokaryotic producers and auxotrophic eukaryotic consumers have been proposed, at least for vitamin B 12 (Croft et al. 2006). The consensus of the original bioassay and culture studies on B-vitamins in aquatic environments (as summarized in Droop 2007) seems to have been that natural concentrations of dissolved B-vitamins should be sufficient to support observed marine and freshwater biomass. However, amendments of picomolar quantities of B 12 have been demonstrated to preferentially increase growth of larger phytoplankton (Sañudo-Wilhelmy et al. 2006), and B-vitamins may co- limit (with Fe) primary production in the high nutrient low chlorophyll (HNLC)) regions of the Southern Ocean (Panzeca et al. 2006; Bertrand et al. 2007). A recent study of the physiology of several diatoms (a group responsible for a large portion of marine primary productivity) identified several mechanisms used to obtain B 12 under deplete conditions, and found widespread transcriptomic evidence for these B 12 acquisition mechanisms in environmental samples (Bertrand et al. 2012). In addition to limitation and co-limitation of primary productivity, B- vitamins logically could exert a strong influence on phytoplankton community structure given Klein 112 widespread auxotrophy among eukaryotes. For example, low-B 1 environmental concentrations might select against coccolithophores, as 83% of assayed haptophytes require B 1 (Provasoli, 1961; Tang et al. 2010; Sañudo-Wilhelmy et al., 2014). Recent advances in instrumentation and methodology allowing for direct and simultaneous analysis of B-vitamins and other trace organics in natural waters down to the fmol • L -1 range (Okbamichael and Sañudo-Wilhelmy 2004; Suarez-Suarez et al. 2011; Sañudo- Wilhelmy et al. 2012) have sparked renewed interest in the role of such cofactors in aquatic ecosystems. Despite a number of recent studies on the cycling of B-vitamins in different marine and estuarine environments (Panzeca et al. 2006; Sañudo-Wilhelmy et al. 2006; Bertrand et al., 2007; Bertrand et al. 2012; Gobler et al. 2007; Panzeca et al. 2008; Panzeca et al. 2009; Bonnet et al. 2010; Tang et al. 2010, Sañudo-Wilhelmy et al. 2012; 2014), B-vitamin studies from freshwater systems utilizing modern direct-analysis techniques are rare at best. Existing freshwater literature relied on bioassays, and are now several decades old (Daisley 1969; Carlucci and Bowes 1972; Ohwada and Taga 1972; Cavari and Grossowicz 1977; Parker 1977; Kurata and Kadota 1981; Kurata et al. 1982). Previous data are also not generalizable to freshwater systems as a whole, with major ecosystems completely uncharacterized at present. Methodologies vary widely and the diversity of bioassay organisms employed respond differently to various forms (“vitamers”) of the B-vitamins. Likewise, trace metal data from freshwater systems in general is also scarce, though some limited data from Lake Michigan (Shafer and Armstrong, 1990) exists, and neighboring Lake Superior (Sherrell 2004) has been well characterized. This study seeks broadly to describe the distribution of dissolved B-vitamins and trace elements in Lake Michigan and its potential implications for biological processes by applying a Klein 113 Redfield-type stoichiometric analysis. This is the first application of modern direct-analysis techniques for B-vitamins to freshwater systems, and one of only a few broad-spectrum trace metal analyses for lakes. Therefore, these data should help to establish a baseline for comparison to other studies and in similar major lakes and contribute toward a better understanding of the role and importance of bioactive B-vitamins and trace elements in freshwater systems. Methods Near-surface samples were collected from a teflon-lined boat-bottom intake pump in July 2009 along a transect from Milwaukee, Wisconsin to the Sheboygan Reef North 145m (N145; 145m) profile station at 43° 37.5019’ N, 87° 09.7747’ W (Figure 1) . Depth profiles were obtained at N145 as well as at another station, Sheboygan Reef South 80m (S80; 80m) located at 43° 14.8143’ N, 87° 10.1188’ W. An underwater camera was deployed above the sediment to observe invasive Quagga mussel (Dreissena rostriformis bugensis) populations at both depth profile sites. All sampling materials were acid-washed using standard trace metal and B-vitamin protocols (Flegal et al. 1991; Sañudo-Wilhelmy et al. 2006). Water for trace metal and B- vitamin analysis was obtained using an acid-clean, externally sprung 2.5L Niskin bottle (General Oceanics) rigged on a Kevlar line (Cortland Cable) and operated on a separate, plastic-coated winch arrangement separate from the forward winch used for the CTD and non-trace-metal clean bottle casts. Water for vitamin analysis was stored in methanol-rinsed 1L HDPE amber bottles and for trace metals in 250mL acid-clean LDPE bottles. All samples were filtered using 0.2µm acid-washed cartridges. Vitamin samples were immediately frozen at -20°C and trace Klein 114 metal samples double-bagged prior to shipment to USC for analysis. (What about the amazing sample concentration step from the blood bags hanging in the cold room?) Trace metal analysis was performed according to Field and Sherrell (2003). In a class- 100 clean laboratory, trace metal samples were acidified to pH 2.0 with QHNO3 (Fisher Trace grade) and stored for six weeks prior to analysis. Samples were diluted 1:1 with 5% QHNO 3 and analyzed for 15 elements via high resolution inductively coupled plasma mass spectrometry (HR-ICPMS, Thermo-Finnigan Element II). Concentrations of dissolved trace metals were quantified using both an indium internal standard and a mixed-metal external calibration curve. Vitamin samples were thawed and pre-concentrated via solid phase extraction in a class- 100 clean laboratory. One-liter samples were acidified with HCl to pH 6.5 and passed through a column containing 7.5mL C 18 resin (HF Bondesil 120µm) at a rate of no greater than 1mL min- 1. Samples were then re-acidified to pH 2.0 and the extraction repeated (Sañudo-Wilhelmy et al. 2012). Columns were eluted with 3x 5mL aliquots LC-MS grade methanol and evaporated to dryness under vacuum, then reconstituted in 1mL QH 2 O. Triplicate analysis of each sample was performed on a Thermo Scientific TSQ Quantum Access LC-MS/MS using chromatographic conditions from Sañudo-Wilhelmy et al. (2012). The B-vitamins in the analytical blanks were essentially undetectable and therefore we calculated limits of detection as 3x standard deviation of lowest calibration standard. Those were 3, 20, and 10 fmol L -1 for B 12 , B 1 , and B 7 , respectively. Temperature and in vivo chlorophyll fluorescence data were obtained in situ by a conductivity-temperature-depth instrument (CTD) instrument package (SBE25; SeaBird Electronics. Extracted chlorophyll a was determined fluorometrically following the method of Klein 115 Shoaf and Lium (1976), NO 3 - + NO 2 - determined by Cd-reduction flow injection analysis using sulfanilamide (APHA 2002). Dissolved Si was determined colorimetrically using standard methods (APHA 2002). Particulate ("biogenic") silicate was concentrated on 0.8µm pore size polycarbonate filters, extracted in 0.05M carbonate at 85°C for 2h, neutralized, and assayed as dissolved Si. Results Surface transect. Transect sampling compared 2 m water just inside the main gap of Milwaukee Harbor (positive control for most analytes) with 9 stations approximately equally spaced between the harbor and the first mid-lake station (N145) almost 90 km ENE of Milwaukee (Figure 1). Biomass and critical element constituents dropped rapidly from the harbor to the open lake and were typically enriched at least several-fold within the harbor (Table 1). Particulate silica, a proxy for diatoms and scaled Chrysophytes (cf. Sandgren REF), as well as chlorophyll a both reached minimum values at 25 km distance from the harbor. Biomass nutrients displayed very low spatial variability as well, with adequate N and Si. Total P remained very low (<0.15 µmol L -1 ) in all transect stations. Phosphate was heavily depleted to less than 10 nmol L -1 , a characteristic of Lake Michigan (Brooks and Edgington 1994). Vitamin concentrations differed little between harbor and transect waters, though the variation among stations was nearly equal to the average for most (Table 1). Large maximum- minimum ranges were also characteristic, from below detection to several times the mean of all stations. Vitamins did not vary systematically with distance from the harbor, but all did have a maximum value at the 63 km sample (Stn. 6; 6th black point outbound on the map). Vitamin B 12 Klein 116 as well as the related amino acid methionine were low or even undetectable in the harbor sample that contained the largest algal and diatom populations (ca. 10x open lake stations). Trace metals were not assessed in the harbor sample (first outbound station) to reduce risks of subsequent contamination. Once distanced from the anthropogenic impacts of the harbor, samples were collected at regular intervals with other parameters. The variability among stations offshore for metals was substantially less than for vitamins (Table 1), rarely reaching half of the mean and the range of values rarely exceeded 4-fold. Even so, these ranges were far wider than those for either biomass or nutrient elements. Systematic spatial variability was limited to Fe, Ti, and Al, whose concentrations decreased by approximately half along the outbound transect (from 11, 10, and 0.9 nmol L -1 to open-lake values of 5, 7, and 0.4 nmol L -1 respectively). Several other metals showed distinct high values at 37 km (Stn. 5). Depth profile and basin hydrography Physical and biogeochemical water column structure was well defined for the three Sheboygan Reef stations (Figure 2) and varied systematically among them. The most remote basin station north of the reef plateau, N145, exhibited a simple density structure with a single sharp thermocline at 12m; by 20 m the coldest deep water was encountered. A broad, distinct particle-rich zone extended from 12-57 m and contained several sub-layers within it. The lower hundred meters of the N145 water column were physically unremarkable. Atop the reef pinnacle (P40) there were 3 distinct temperature-density regimes, uplifted relative to the upstream basin station and reflecting the upwelling and turbulent dissipation predicted in the MLRC model of Cuhel and Aguilar (2013). Light scattering by particles was maximal at 9.5 m, and water clarity increased below 27 m in the deep isothermal bottom water overlying the quagga mussel Klein 117 populations. Finally, the downstream station S80 retained the very shallow principal pycnocline at 7m, but the underlying water lacked the multiple density shelves of the pinnacle station. Particle influence began to increase just below the 7m temperature shelf, increasing to a very strong turbidity maximum (76% transmission) just under the bottom of the stronger temperature gradient. Again water in the deepest isothermal bottom water (55-78 m) was very clear at 96% transmission. Physical features were associated with biological phenomena that are relevant to interpretation of macro- and micronutrient distribution. Using N145 as a control (most open- water like), the pinnacle station had several distinct characteristics. Foremost was the strong evidence of mussel excretion of ammonium at the benthos-water interface and its advective diffusion throughout the apparently well-mixed 15m thick bottom water (Figure 2). Sheboygan Reef North 145m (N145) Station Sheboygan Reef North 145m was located north of the Sheboygan Reef Complex (Figure 1) in the Ludington Basin and this station was chosen to be representative of mid-lake, open water conditions. A sharp thermocline was observed at 15m (Figure 2), separating surface waters of ~18°C from deep waters of ~4°C. The extracted chlorophyll maximum spanned a depth of ~20-40m, with a maximum value of 2µg L -1 . The in-vivo Deep Fluorescence Maximum (DFM) lay below the chlorophyll maximum at a depth of 30-50m, and dissolved TP, NO 3 - + NO 2 - , and Si generally increased with depth (Figure 3). Maximal values were 130 nmol L -1 TP (between the depths of 20 and 40m), 32 µmol L -1 Si, and 29 µmol L -1 NO 3 - + NO 2 - near-bottom. Dissolved B-vitamins were detectable in the upper water column and decreased to near- or undetectable concentrations at depths greater than 100m (Figure 3). Each had a maximum Klein 118 within the zone of maximum fluorescence. Dissolved B 12 exhibited two maxima, near-surface at ~2pmol L -1 and again at the DFM (1.8 pmol L -1 ). B1 had a maximum value of ~ 5 pmol L -1 at the DFM, and B 7 presented a broad peak spanning the chlorophyll maximum at 20-40m with a maximum concentration of 40 pmol L -1 . As observed in the surface transect (Table 2), the depth distribution of the “terrigenous” metals, Fe, Al, and Ti covaried with each other (Figure 3). Their concentrations were relatively constant from the surface to 60m depth (8, 12, and 0.5 nmol L -1 , respectively) and then increased linearly towards near-sediment maxima (33, 45, and 1.3 nmol L -1 respectively). In contrast, the “bioactive” dissolved metals (Co, Mn, and Cd) also showed comparable depth distributions (Figure 3), strong maxima at 60m (485 pmol L -1 , 4.6 nmol L -1 , 1100 nmol L -1 , and 303 pmol L -1 , respectively); Co, Mn, and Cd were also enriched several times near the bottom of the lake relative to their near-surface concentrations. Dissolved Cu and Pb concentrations covaried with each other (Figure 3) and increased from near-surface values of ~10 nmol L -1 and ~300 pmol L -1 , respectively, to 44 nmol L -1 and 1300 pmol L -1 . Dissolved Ni, V, and Mo shared similar trends, with Ni concentrations ranging from ~10 to 25 nmol L -1 and V and Mo near 5.2 and 10 nmol L -1 , respectively (Figure 3). In contrast to the depth gradients observed for most of the trace elements, dissolved Ag had a strong maximum of 25 nmol L -1 at a depth of 25m (Figure 3). Sheboygan Reef Pinnacle 40m (P40) Station The Sheboygan Reef Pinnacle 40m station was located at the shallow pinnacle of the mid-lake reef complex. Sampling resolution for B-vitamins and trace elements at this station was lower than at the other two, and Quagga mussel beds with densities >10 4 individuals m -2 were observed via underwater camera. SRP40m had two thermoclines; one with a larger Klein 119 temperature gradient of ~10°C at 7-10m depth and a second smaller thermocline at 27m (Figure 4). The chlorophyll maximum and DFM both occurred between these two thermoclines at depths of 15m and 21m, respectively (Figure 4). No clear structure in the B-vitamin profiles in relation to other water column features is evident, though all three are undetectable above the sediments and B 12 is undetectable at all depths (Figure 4). Concentration ranges for all B- vitamins were similar to those observed at N145. Trace metal concentrations at P40 generally shared maxima at 5m depth and were otherwise relatively uniform throughout the water column (Figures 4), though Cd and Mo appear to increase slightly with depth below the DFM (Figure 4). Dissolved Fe ranged from 5.9 to 11, Cu from 7.2 to 15, and Al from 11 to 13 nmol L -1 (Figure 4). Dissolved Ni ranged from 6.0 to 10, Cd from 0.09 to 0.15, and Mo from 11 to 28 nmol L -1 (Figure 4). Sheboygan Reef South 80m (SRS80) Station Sheboygan Reef South 80m station was located south of the reef pinnacle over the mid- lake plateau facing the southern Chippewa Basin of the lake (Figure 1). It is on the downstream side of the reef peak that intercepts a persistent southerly current (c.f. Cuhel and Aguilar 2013). The thermocline was observed from 8 to 20m (Figure 5) and was both more gradual in slope and shallower in the water column than at N145 (Figure 5), though it spans a similar range of temperatures from ~18°C near-surface to ~4°C at depth. The extracted chlorophyll maximum (2.7µg L -1 ) and DFM were both observed at a depth of 30m. NO 3 - + NO 2 - and Si were similar in concentration to (Figure 5) and increased with depth to near-sediment maxima of 29 and 32 µmol L -1 , respectively. TP concentrations were relatively constant throughout the water column, but presented a maximum of 0.18 µmol L -1 at the chlorophyll maximum/DFM. Klein 120 At this station, we detected the highest concentrations of vitamins B 12 and B 1 in surface waters (3.1 pmol L -1 and 66 pmol L -1 , respectively; Figure 4). In fact, these were the highest vitamin concentrations measured throughout the lake by an order of magnitude for B 1 and by a factor of ~1.5 for B 12 . A secondary B 12 maximum of 1.0 pmol L -1 corresponded with the chlorophyll maximum/DFM. B7 was depleted at the surface (8pmol L -1 ) relatively to its highest concentration of 35 pmol L -1 at a depth of 5m). All three vitamins decreased with depth below 30m, though B 1 was elevated slightly (8pmol L -1 ) near-sediment relative to the overlying deep water, where it was undetectable. It is more difficult to discern specific trends in trace metal distributions at this location due to the lower sampling resolution (Figure 5). Trace metal concentrations were in general lower and varied within a much more narrow range than at N145 (Figure 3). The terrigenous elements Fe, Al, and Ti (Figure 4) increased from their minima at the 30m chlorophyll maximum/DFM to maximum near-sediment values of 6.8, 9.5, and 0.44 nmol L -1 , respectively, an order of magnitude lower than at N145. Variance in Co, Mn, Zn, and Cd (Figure 5) concentrations was mostly within the error bars and much lower than the range observed at N145 (e.g. Co was 98 – 470 pmol L -1 at the northern station vs. 96 – 110 pmol L -1 at S80). Dissolved Cu and Pb again shared similar trends (Figure 5) as they did at N145, however, they did not increase with depth and only Fe, Al, and Ti (Figure 5) appeared to have a sediment source. Dissolved V and Mo again covaried within a narrow concentration range (Figure 4f, ca. 5.0 and 9.5 nmol L -1 respectively), however at the S80 station Ni resembled more closely the depth distributions of Cu, Pb, and Zn at this station. Klein 121 Discussion Depth Profiles and Hydrography The Mid-Lake Reef Complex (MLRC) of southern Lake Michigan provides a perfect venue for testing biogeochemical cycling concepts (Cuhel and Aguilar 2013). In the 2000s it played host to abundant and diverse biological sources and sinks for both macro- and micronutrients. The upstream northern MLRC basin behaved typically for a deep open-water aquatic system (Figure 2), with smooth depth gradients of temperature, single extended deep fluorescence and turbidity maxima, upper water column depletion of macronutrients including nitrogenous compounds and silicate, and relatively uniform water column characteristics in the thick layer of water overlying a surficially-oxic silty mud sediment surface. On the pinnacle of the Sheboygan Reef (Station P40), quagga mussels carpeted the surfaces at densities of 25,000 / m 2 or more, excreting digestion products. This could represent a potential source of dissolved vitamins different from the bacterially-mediated excretion in water and sediment. The horizontal surfaces upon which they grow became organic-matter rich and contained zones susceptible to ephemeral and/or extended local anoxia, favoring mobilization of redox-sensitive trace metals, for example. Although the peak at 38-43m is far below the principal thermocline at 10m, the influence of the shoal upon hydrographic conditions was clearly evident in the multiple stepped and uplifted thermocline system and a distinctly well mixed near-bottom layer (Figure 4). Mussel excretion, possibly augmented by localized diagenetic ammonium efflux, lead to greatly enhanced near-bottom ammonium in this layer, with especially pronounced enrichment in the bottom 2m (Figure 4). The peak was at a depth equivalent to open Klein 122 water deep chlorophyll maxima (Figure 3), but mussel grazing and hydrodynamic anomalies brought phytoplankton down to deepwater concentrations in the near-bottom layer as well. Downstream (south) of the pinnacle, excreted nutrients may have been entrained in a sweeping current into midwater enriched phytoplankton populations. The largest chlorophyll a and particle maxima clearly occurred in the south basin station (Figure 5). The thermocline was slightly less steep than north of the reef, and the characteristic near-bottom mixed layer was evident from 50m to the bottom (Figure 2). Ammonium was likewise intermediate between deepwater lows and pinnacle highs, with definitive but mild enrichment in the bottom layer, which contained smaller but significant populations of quagga mussels on a silty-sand surface. Station N145 exhibited a broad, abundant deep phytoplankton chlorophyll and turbidity maxima overlying a thick deepwater zone showing some evidence of sediment nutrient efflux (Figure 3). No vitamin was enriched in near-sediment samples, and in fact most were at or below detection limits, arguing against a sedimentary source. In contrast, near-sediment water was enriched in all metals except Mo, V, and Zn (similar) and Ni and Ba (decreased near-bottom). Distribution of dissolved B-vitamins Broadly, B-vitamin concentrations and distributions observed in Lake Michigan are consistent with our understanding of their biological sources within the photic zone and agree with trends observed in several older and contemporary studies. The concentration range measured during our study was similar to that of other relatively oligotrophic lake systems measured using the bioassay technique, suggesting that the two methods obtain similar results (Table 2), with the exception of our measurements of B 1 in Lake Michigan, which was an order of magnitude lower than concentrations previously measured by bioassays in other lakes. High Klein 123 near-surface concentrations of B 12 and B 1 (Figure 3, Figure 5) above the thermocline may reflect relatively low biomass production (low total chlorophyll and thus biological vitamins demand). Cyanobacteria and diazotrophs in particular are important producers of B-vitamins (Bonnet et al. 2010; Sañudo-Wilhelmy et al., 2014), and Synechococcus spp. were the dominant phytoplankton group observed. The B 12 and B 1 maxima associated with the DFM are likely reflective of production by the high abundance of prokaryotes at these depths, which both synthesize and may be the largest consumers of B-vitamins in the marine environment (Koch et al. 2012). The surface transect and depth profile data (Figures 2-5, Table 2) are not suggestive of either a coastal or sediment source for B-vitamins in Lake Michigan. Cavari and Grossowicz (1977) found an increase in B 12 concentrations measured with depth in Lake Kinneret, Israel, and several studies have implicated remobilization from sediments as a potential source as their concentrations increased toward the sediments andwere several orders of magnitude greater than in the water column (Gillespie and Morita, 1972; Ohwada and Taga, 1972; Nishijima and Hata, 1978). Dissolved B-vitamin concentrations in Lake Michigan are low and often below limits of detection near-sediment at all two profile stations. Therefore, sediments do not appear to be a source of those vitamins to the water column in Lake Michigan during our sampling. However, those previous studies involved highly eutrophic lake systems, and the observed near-sediment B-vitamin concentrations were highest in November and December (Cavari and Grossowicz, 1977). While the sediments of Lake Michigan were not a source of B-vitamins to the water column at the time of sampling, this is likely not generalizable to other seasons or freshwater systems in general. In exception to this trend, B 1 is slightly elevated (~8 pmol L-1) near- sediment at the S80 station, however this could be due to excretion by the dense population (on the order of 10 4 individuals • m -2 ) invasive Quagga mussels (Dreissena rostriformis bugensis) Klein 124 found on the benthos at all profile stations. These overall trends of decrease with depth match those observed by Carlucci and Bowes (1972) in Lake Tahoe and in some coastal oceans (Sañudo-Wilhelmy et al., 2012). Distribution of Dissolved Trace Metals Trace metal concentrations measured in Lake Michigan during our study were, for most elements, on the same order of magnitude of those reported in other freshwater systems (Table 3, e.g. Fe was 3.5 – 33 nmol L -1 in Lake Michigan, 0.91 – 23nmol L -1 in Lake Tahoe, and 1.1 – 100 nmol L -1 in Lake Superior) compared to but elevated relative to other oligotrophic lake systems (Table 3, e.g. Co was 0.08 – 0.49 nmol L -1 Co in Lake Michigan, 0.01- 0.04 nmol L -1 in Lake Tahoe, and 0.01 – 0.28 in Lake Superior). Furthermore, our metals data agrees well with what limited trace metal data exists from Lake Michigan (Shafer and Armstrong 1990). Decrease of terrigenous Fe, Al, and Ti offshore along the surface transect (Figure 2) suggests a coastal source for these elements, and no other trace metals or vitamins exhibited a similar pattern. The enrichment of Fe, Al, Ti, Cu, Pb, Co, Mn, and Cd with depth at N145 (and for Cu and Pb at S80) is suggestive of benthic remobilization from bottom sediments (Figures 3, 5). Dissolved V, Mo, and Ni, share similar depth distributions at both N145 and S80 (Figure 4, Figure 5), but neither increase with depth (indicative of a benthic source) nor possess distributions suggesting atmospheric deposition. None of the bioactive trace metals such as Fe or Cu exhibit a “nutrient-like” profile indicative of significant draw-down by biology, with the possible exception of the relatively depleted values for Co at 0m and 60m in the N145 profile (Figure 3). These depths correspond with the maxima for B 12 , which requires Co for its synthesis (Figure 3). Co, Mn, and Cd all Klein 125 display similar trends with depth at N145 (Figure 3) and have a strong maximum at 60m immediately below the DFM, likely as a result of remineralization of sinking particles. One source of trace metal exchange for redox-active metals (e.g., Fe, Mn, others) is efflux from anaerobic respiration in anoxic sediments. This avenue has been closed as a source in Lake Michigan because of permanently oxic surface sediments – there are no anoxic basins known for the open lake habitat. This effect is enhanced by the lower sedimentation rates as a result of the Quagga Mussel invasion and associated high filtration rates. On the other hand, formerly bare sands are now experiencing substantial organic matter loading from mussel feces and may be exchanging reduced metals into the water during huffing, a potentially new source. Because of a lengthy deep mixing period during December through March or April, concentrations of chemical constituents tend to be uniform in the water column each spring. This means that processes promoting vertical structure and chemical gradients must be sufficiently strong to produce a signal in the short stratified season. Physical, macronutrient, and biomass parameters explored in this study display clearly identifiable, discrete zones of biogeochemistry. However, B-vitamins and trace metals rarely exhibited obvious associations with biogeochemical processes. Assessment of Nutrient Limitation Nutrient limitation was assessed by comparing ratios of those dissolved constituents to literature laboratory culture results (Ho et al 2003, Taylor and Sullivan 2008, Tang et al 2010) standardized to P (Table 4). These “extended Redfield ratios” allow us to consider the relative abundances of these nutrients to assess potential limitations. With the exception of the B- vitamins, all considered nutrients are one or more orders of magnitude more abundant (relative to Klein 126 P) than the stoichiometries observed in laboratory culture data. For example, the median dissolved N:P observed in Lake Michigan in this study was 570 mol • mol P -1 , compared to the 16 mol • mol P -1 classical Redfield ratio. That all but the B-vitamins are enriched orders of magnitude relative to the culture values suggests that none of those nutrients are more limiting than P, and therefore that the lake is P-limited. Indeed, phosphorus limitation is widely understood to be the dominant regime in the Great Lakes (Sterner 2008, Sterner et al 2007). However, co-limitation of other elements likely occurs on shorter time and spatial scales (Sterner 2008), and Fe was observed to quickly become limiting after P addition in incubations from neighboring Lake Superior (Sterner et al 2004), which possesses similar Fe concentrations to those reported here for Michigan. B 1 and B 12 ratios (medians of 4.7 and 35 nmol • mol P -1 , respectively) are more similar to existing culture data (medians of 150 and 4.1 nmol • mol P -1 ) and indeed suggest that B 1 may be a limiting nutrient for auxotrophs. Strong depletion of thiamin (detectable in only 7% of samples) relative to B 12 has previously been observed in the Southern Indian Ocean (Carlucci and Cuhel 1975), suggesting such deficiencies may occur in diverse regimes. A thiamin deficiency impacting reproduction has been reported in salmonid fish in the Great Lakes, though this has been attributed to high thiaminase activity in prey species as B 1 content in all assayed prey fish were high enough to satisfy fish nutritional requirements (Tillitt et al 2005). Additionally, thiaminase activity 5-100 times greater than that observed in Great Lakes fishes has recently been reported in Quagga mussels (Tillitt et al 2009), which were present in great abundance at both depth profile sites. Though there may be no direct connection between the relatively low dissolved concentrations of B 1 reported here and thiamin deficiency in salmonid Klein 127 fishes, cycling of B 1 in the Lake Michigan ecosystem is likely complex and merits further study due to its impact on salmonid reproduction. Several caveats accompany this treatment of nutrient limitation. The literature stoichiometry for metals is from cultured marine organisms, albeit a broad sampling of major taxa also present in freshwater systems. The literature values for B-vitamin stoichiometryare from a small number of marine organisms, and the B-vitamin limitation ratios in Table 3 are relevant only for auxotrophic organisms. The proportion of auxotrophs for each vitamin in Lake Michigan is unknown, and thus establishing a “true” requirement or a bulk stoichiometric ratio is not feasible. Additionally, the distribution of B-vitamins is highly heterogeneous, with undetectable or near-undetectable concentrations in many samples. Using a median value for stoichiometric calculations may not adequately describe the spatially heterogeneous distribution of B-vitamins. Adaptation and application of direct analytical techniques for intracellular or particulate B-vitamins should allow for better determination of typical B-vitamin stoichiometry in both the environment and in laboratory culture and exploration of the relationship between dissolved and intracellular pools. A second approach to assessing B-vitamin limitation is a comparison of observed values with published phytoplankton half-saturation constants for growth (K s ) as in Figures 6a-c. Several caveats apply; as with the B-vitamin stoichiometric calculations described above and reported in Table 4, literature K s values are derived from a small set of primarily marine organisms in culture and thus may not be totally applicable to Lake Michigan.However, we believe that this is a valid comparison because the dissolved vitamin concentrations measured in the lake are similar to those found in marine systems. Secondly, seasonal changes in B-vitamin uptake rates have been observed to be much greater than changes in B-vitamin concentrations, Klein 128 suggesting that organisms may vary their uptake rate to match vitamin availability (Parker 1977). Based on what laboratory values are available, B 12 measured concentrations in Lake Michigan (0.0 -3.1 pmol L -1 ) would appear to be similar to or slightly lower than the half-saturation constant for growth (Figure 6a), B 1 is significantly lower (Figure 6b), and B 7 much higher (Figure 6c). This, in tandem with the preceding stoichiometric calculations, suggests that B 1 and possibly B 12 are limiting for auxotrophs both stoichiometrically and on growth rates. Of more than three hundred cultured phytoplankton species, a majority required B 12 , about a quarter require B 1 , and a smaller proportion (8%) B 7 (Sañudo-Wilhelmy et al 2014). It is likely, then, that B-vitamin availability might significantly affect phytoplankton growth rates and dynamics in Lake Michigan. Conclusions The data presented here represent the first direct analysis of B-vitamins in a freshwater environment, and one of only a few broad-spectrum trace metal surveys in lakes. Metals are comparable in concentration to other oligotrophic lakes. Fe, Al, and Ti appear to have both a coastal and a sediment source to the water column. B-vitamin concentrations and trends agree broadly with previous literature, with concentrations ranges similar to those observed in older bioassay studies in lakes and vertical distributions corresponding to bacterial production in the upper water column. A stoichiometric nutrient limitation calculation finds all metals and macronutrients replete relative to P, and the lake therefore appears to be chiefly P-limited in agreement with previous findings (Sterner 2008). Although little data exists for B-vitamin stoichiometry and half-saturation constants for growth, what is available suggests that B 1 and possibly B 12 are Klein 129 potentially limiting on biomass and growth rates of auxotrophs. As lake systems are quite heterogeneous, each system requires individual sampling and consideration to determine trace nutrient limitation or co-limitation. A larger set of laboratory culture data on B-vitamin requirements and half-saturation constants for growth as well as field observations of dissolved B-vitamin concentrations is required to accurately assess the importance of B-vitamins to freshwater systems. Klein 130 References American Public Health Association (APHA) (2002) Standard methods for the examination of water and wastewater, 20th edn. APHA, New York Arar E. J., Collins G. B. (1997). Method 445.0: in vitro Determination of Chlorophyll a and Pheophytina in Marine and Freshwater Algae by Fluorescence. Washington, DC: US Environmental Protection Agency. Bertrand, E. M., Saito, M. A., Rose, J. M., Riesselman, C. R., Lohan, M. C., Noble, A. E., et al. (2007). Vitamin B12 and iron colimitation of phytoplankton growth in the Ross Sea. Limnol. Oceanogr52(3): 1079-1093 Bonnet, S., Webb, E. A., Panzeca, Caterina, Karl, D. M., Capone, D., Sañudo-Wilhelmy, S. A. (2010).Vitamin B12 excretion by cultures of the marine cyanobacteria Crocosphaera and Synechococcus. Limnol. Oceanogr55(5): Brooks, A. S. and Edgington D. N. (1994). Biogeochemical control of phosphorus cycling and primary production in Lake Michigan. Limnol. Oceangr 39 (4): 961-968 Brzezinski M. A. (1985). The Si:C:N ratio of marine diatoms: interspecific variability and the effect of some environmental variables. J. Phycol. 21: 347–357 Carlucci, A. F. (1972). Production and utilization of dissolved vitamins by marine phytoplankton. In: Effects of the ocean environment on microbial activities. University Park Press, Baltimore MD: 449-456 Carlucci, A. and Bowes, P. M. (1972). Determination of Vitamin B 12, Thiamine, and Biotin in Lake Tahoe Waters Using Modified Marine Bioassay Techniques. Limnol. Oceanogr17(5): 774–777 Carlucci, A. F., and Cuhel, R. L. (1975). Vitamins In The South Polar Seas, I. Distribution and Significance of Dissolved and Particulate Vitamin B12, Thiamine, and Biotin in the Southern Indian Ocean. From the Proceedings of the Third SCAR Symposium on Antarctic Biology Cavari, B., and Grossowicz, N. (1977). Seasonal distribution of vitamin B12 in Lake Kinneret.Applied and environmental microbiology34(2): 120-4 Croft, M. T., Warren, M. J., Smith, A. G. (2006). Algae need their vitamins. Eukaryotic cell5(8): 1175-83 Cuehl, R. L. and Aguilar, C. (2013). Ecosystem Transformation of the Laurentian Great Lake Michigan by Nonindigenous Biological Invaders. Annu. Rev. Mar. Sci. 5: 289-320 Klein 131 Daisley, K. (1969). Monthly survey of vitamin B 12 concentrations in some waters of the English Lake District.Limnol. Oceanogr14(2): 224–228 Dakshinamurti, K., Chalifour L., Bhullar R.P. (1985). Requirement for Biotin and the Function of Biotin in Cells in Culture.Annals of the New York Academy of Sciences 447: 38-55 Davies, A., and Leftley, J. (1985). Vitamin B12 binding by microalgalexocrines: dissociation constant of the vitamin-binder complex determined using an ultrafiltration technique. Marine Ecology Progress Series21: 267-273 Droop, M. R. (2007). Vitamins, phytoplankton and bacteria: symbiosis or scavenging? Journal of Plankton Research29(2): 107-113 Field, M. P. and Sherrell, R. M. (2003). Direct determination of ultra-trace levels of metals in fresh water using desolvatingmicronebulization and HR-ICP-MS: application to Lake Superior waters. Journal of Analytical Atomic Spectrometry18(3): 254-259 Flegal, A.R., Smith, G.J., Gill, G.A., Saudo-Wilhelmy, S., Anderson, L.C.D. (1991). Dissolved trace element cycles in the San Francisco Bay estuary. Marine Chemistry 36: 1-4: 329- 363 Ho, T.-yuan, Quigg, A., V, Z., Milligan, A. J., Falkowski, P. G., Morel, M. M. (2003). The elemental composition of some marine phytoplankton.J. Phycol. 39: 1145-1159. Jones M. N. (1984). Nitrate reduction by shaking with cadmium: alternative to cadmium columns. Water Res.18:643–646 Kurata, A., and Kadota, H. (1981). Annual changes of vitamin B1, biotin and vitamin B12 in water in Lake Biwa.Journal of nutritional science and vitaminology27(4): 301 Kurata, A., Kawase, T., Kadota, H. (1982).Diurnal Fluctuation in the Concentrations of В Group Vitamins in Water in the Southern Lake Biwa.Bull. Jap. Soc. Sci. Fish.48(3): 43-44 Messina, D. S. and Baker, A. L. (1982). Interspecific growth regulation in species succession through Vitamin B 12 competitive inhibition.Journal of Plankton Research4(1)L: 41-46 Montagnes, D. J. S., Berges, J. a, Harrison, P. J., & Taylor, F. J. R. (1994). Estimating carbon, nitrogen, protein, and chlorophyll a from volume in marine phytoplankton. Limnology and Oceanography39(5): 1044-1060 Moss, B. (1973). The influence of environmental factors on the distribution of freshwater algae: An experimental study. 3. Effects of temperature, vitamin requirements and inorganic nitrogen. J. ecol, 61(1), 179-192. Klein 132 Nishijima, T.,Hata, Y. (1977).Distribution of thiamine, biotin, and vitamin B12 in Lake Kojima. I. Distribution in lake water. Bull. Jap. Soc. Sci. Fish.43 Nishijima, T., Hata, Y. (1978).Distribution of Thiamine, Biotin, and Vitamin B12 in Lake Kojima—II.Bull. Jap. Soc. Sci. Fish.44(8): 815–818 Ohwada, K., Otsuhata, M., Taga, N. (1972). Seasonal cycles of vitamin B 12, thiamine and biotin in the surface water of Lake Tsukui. Bull. Jap. Soc. Sci. Fish.38: 817–823 Ohwada, K. and Taga, N. (1972). Vitamin B 12, Thiamine, and Biotin in Lake Sagami. Limnol. Oceanogr17(2): 315–320 Okbamichael, M. andSañudo-Wilhelmy, S. (2004).A new method for the determination of Vitamin B12 in seawater.AnalyticaChimicaActa517(1-2): 33-38 Panzeca, C., Beck, A. J., Leblanc, K., Taylor, G. T., Hutchins, D. A., Sañudo-Wilhelmy, S. A. (2008).Potential cobalt limitation of vitamin B 12 synthesis in the North Atlantic Ocean.Global Biogeochemical Cycles22(2): 1-7 Panzeca, Caterina, Beck, A. J., Tovar-Sanchez, A., Segovia-Zavala, J., Taylor, Gordon T., Gobler, Christopher J., et al. (2009).Distributions of dissolved vitamin B12 and Co in coastal and open-ocean environments.Estuarine, Coastal and Shelf Science85(2): 223-230 Paode, R. D., Sofuoglu, S. C., Sivadechathep, J., Noll, K. E., Holsen, T. M., Keeler, G. J. (1998). Dry Deposition Fluxes and Mass Size Distributions of Pb, Cu, and Zn Measured in Southern Lake Michigan during AEOLOS. Environmental Science & Technology32(11): 1629-1635 Parker, M. (1977). Vitamin B12 in Lake Washington, USA: Concentration and rate ofuptake. Limnol. Oceanogr22(3): 527-538 Provasoli, L. A. A. F. C. (1974). Vitamins and growth regulators. (W. D. P. Stewart, Ed.)Algal Physiology and Biochemestry, 741787.Blackwell Scientific Publications. Sandgren, C. D., Smol, J. P., Kristiansen J. (eds) (1995). Chrysophyte algae: ecology, phylogeny and development. Cambridge University Press. Sañudo-Wilhelmy, S.A., Cutter, L.S., Durazo, R., Smail, E.A., Gomez-Consarnau, L., Webb, E.A., Prokopenko, M.G., Berelson, W.M., Karl, D.M. (2012) Multiple B-vitamin depletion in large areas of the coastal ocean.Proc. Natl. Acad. Sci. USA 109 (35): 13888- 13899 Sañudo-Wilhelmy, S.A., Gobler C.J., Okbamichael M., and Taylor G.T. (2006).Regulation of phytoplankton dynamics by vitamin B12.Geophysical Research Letters 33 Klein 133 Sañudo-Wilhelmy, S.A., Gomez-Consarnau, L., Suffridge, C., Webb E.A. (2014).The Role of B Vitamins in Marine Biogeochemistry.Annual Reviews 6: 339-367 Sañudo-Wilhelmy, S.A, Kustka, A B., Gobler, C J, Hutchins, D A, Yang, M., Lwiza, K., et al. (2001).Phosphorus limitation of nitrogen fixation by Trichodesmium in the central Atlantic Ocean.Nature, 411(6833): 66-9 Shoaf, W. T. and Lium, B. W. (1976). Improved extraction of chlorophyll a and b from algae using dimethyl sulfoxide.” Limnol. Oceanogr 21: 926-928 Sterner, R. W. (2008).On the Phosphorus Limitation Paradigm for Lakes.Internat. Rev. Hydrobiol. 93(4-5): 433-445. Sterner, R.W., Anagnostou, E., Brovold, S., Bullerjahn, G. S., Finlay, J. C., Kumar, S., et al. (2007).Increasing stoichiometric imbalance in North America’s largest lake: Nitrification in Lake Superior.Geophysical Research Letters34(10): 1-5 Sterner, R.W., Smutka, T. M., McKay, R.M.L., Xiaoming, Q., Brown, E. T., Sherrell, R.M. (2004). Phosphorus and trace metal limitation of algae and bacteria in Lake Superior. Limnol. Oceanogr49(2): 495–507 Suarez-Suarez A., Tovar-Sanchez A., Rossello-Mora R. (2011).Determination of cobalamins (hydroxo-, cyano-, adenosyl- and methyl-cobalamins) in seawater using reversed-phase liquid chromatography with diode-array detection.AnalyticaChimicaActa 701(1): 81-85 Swift, D. G. (1980).Vitamins and phytoplankton growth. In I. Morris (Ed.), The Physiological Ecology of Phytoplankton (pp. 329-368). Blackwell Scientific Publications. Tang, Y. Z., Koch, F., Gobler, C.J. (2010). Most harmful algal bloom species are vitamin B1 and B12 auxotrophs. Proceedings of the National Academy of Sciences107(48): 20756-20761 Taylor, G.T. and Sullivan, C. W. (2008). Vitamin B12 and cobalt cycling among diatoms and bacteria in Antarctic sea ice microbial communities. Limnol. Oceanogr53(5): 1862–1877. Tillitt, D. E., Riley, S.C., Evans A. N., Nichols, S. J., Zajicek, J. L., Rinchard J., Richter, C.A., Krueger, C. C. (2009). Dreissenid Mussels from the Great Lakes Contain Elevated Thiaminase Activity. Journal of Great Lakes Research 35(2): 309-312 Tillitt, D. E., Zajicek, J. L., Brown, S. B., Brown, L. R., Fitzsimons, J. D., Honeyfield, D. C., Holey M. E., Wright, G. M. (2005). Thiamine and Thiaminase Status in Forage Fish of Salmonines from Lake Michigan.Journal of Aquatic Animal Health 17: 13-25 Klein 134 Full Station 2m Underway sample P40 S80 N145 63 km Figure 1. Bathymetric map of Lake Michigan study area Black symbols demark approximate locations of surface transect stations 1-8. Yellow symbols indicate locations of the three depth profile sites in the basins to the north and south as well as on the pinnacle of the mid-lake reef complex. Klein 135 0 25 50 75 100 125 150 75 80 85 90 95 100 0.25 m PATH BEAM TRANSMISSION ( % ) 0.25 m BIN DEPTH ( m ) S80 N145 P40 0 25 50 75 100 125 150 0 5 10 15 20 CTD TEMPERATURE ( ºC ) 0.25 m BIN DEPTH ( m ) N145 P40 S80 TEMP %TX 0 25 50 75 100 125 150 0.0 0.5 1.0 1.5 2.0 2.5 AMMONIUM ( µM ) DEPTH ( m ) N145 S80 P40 0 25 50 75 100 125 150 0.0 0.5 1.0 1.5 2.0 2.5 3.0 EXTRACTED CHLOROPHYLL a ( µg / L ) DEPTH ( m ) P40 S80 N145 N145 CHL NH4 Figure 2. Temperature, transmissivity, extracted chlorophyll a, and ammonium depth profiles at all three depth stations. Klein 136 B 12 (pmol L -1 ) 0.0 0.5 1.0 1.5 2.0 2.5 0 20 40 60 80 100 120 140 B 1 (pmol L -1 ) 0 1 2 3 4 5 6 B 7 (pmol L -1 ) 0 10 20 30 40 50 In vivo fluorescence (V) 0.1 0.2 0.3 0.4 0.5 0.6 Depth (m) 0 20 40 60 80 100 120 140 T (C) 0 2 4 6 8 10 12 14 16 18 20 Chl a ( g L -1 ) 0.0 0.5 1.0 1.5 2.0 2.5 NO 3 - + NO 2 - ( mol L -1 ) 25 26 27 28 29 30 TP ( mol L -1 ) 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 Si ( mol L -1 ) 26 27 28 29 30 31 32 33 Depth (m) Fe (nmol L -1 ) 5 10 15 20 25 30 35 40 45 Ti (nmol L -1 ) 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 Al (nmol L -1 ) 10 20 30 40 50 Figures 3a-d. Depth profiles from Sheboygan Reef North 145m Station. Dashed horizontal lines indicate approximate location of thermocline (blue), chlorophyll maximum (green), and fluoresence maximum (black). c a d a a a a b a a Klein 137 Pb (pmol L -1 ) 0 200 400 600 800 1000 1200 1400 Depth (m) 0 20 40 60 80 100 120 140 Cu (nmol L -1 ) 0 10 20 30 40 50 Ag (pmol L -1 ) 0 5 10 15 20 25 30 Co (pmol L -1 ) 0 100 200 300 400 500 600 Mn (nmol L -1 ) 0 1 2 3 4 5 Zn (nmol L -1 ) 0 200 400 600 800 1000 1200 Cd (pmol L -1 ) 50 100 150 200 250 300 350 V (nmol L -1 ) 4.8 5.0 5.2 5.4 5.6 5.8 Depth (m) 0 20 40 60 80 100 120 140 Ni (nmol L -1 ) 0 5 10 15 20 25 30 Mo (nmol L -1 ) 8.5 9.0 9.5 10.0 10.5 11.0 11.5 Ba (nmol L -1 ) 0 1000 2000 3000 4000 5000 Figures 3e-g. Depth profiles from Sheboygan Reef North 145m Station. Dashed horizontal lines indicate approximate location of thermocline (blue), chlorophyll maximum (green), and fluoresence maximum (black). e a f a g a Klein 138 B 12 (pmol L -1 ) 0.0 0.5 1.0 Depth (m) 0 10 20 30 40 B 1 (pmol L -1 ) 0 1 2 3 4 5 B 7 (pmol L -1 ) 0 10 20 30 40 50 Fe (nmol L -1 ) 0 2 4 6 8 10 12 Cu (pmol L -1 ) 40 60 80 100 120 140 160 180 Al (nmol L -1 ) 0 5 10 15 20 25 NO 3 - + NO 2 - ( mol L -1 ) 24 25 26 27 28 29 30 TP ( mol L -1 ) 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 Si ( mol L -1 ) 27 28 29 30 31 32 In vivo fluorescence (V) 0.15 0.20 0.25 0.30 0.35 0.40 0.45 Depth (m) 0 10 20 30 40 T (C) 0 5 10 15 20 Chl a ( g L -1 ) 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 Figures 4a-d. Depth profiles from Sheboygan Reef Pinnacle 40m Station. Dashed horizontal lines indicate approximate location of thermocline (blue), chlorophyll maximum (green), and fluoresence maximum (black). a a a b a a c a a d a a Klein 139 Cd (nmol L -1 ) 0.00 0.05 0.10 0.15 0.20 0.25 Depth (m) 0 10 20 30 40 Ni (nmol L -1 ) 0 2 4 6 8 10 12 Mo (nmol L -1 ) 0 5 10 15 20 25 30 Figure 4e. Depth profile from Sheboygan Reef Pinnacle 40m Station. Dashed horizontal lines indicate approximate location of thermocline (blue), chlorophyll maximum (green), and fluoresence maximum (black). e a a Klein 140 B 12 (pmol L -1 ) 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 Depth (m) 0 20 40 60 80 B 1 (pmol L -1 ) 0 10 20 30 40 50 60 70 B 7 (pmol L -1 ) 0 10 20 30 40 In vivo fluorescence (V) 0.0 0.2 0.4 0.6 0.8 1.0 1.2 Depth (m) 0 20 40 60 80 T (C) 0 2 4 6 8 10 12 14 16 18 20 Chl a ( mol L -1 ) 0.0 0.5 1.0 1.5 2.0 2.5 3.0 NO 3 - + NO 2 - ( mol L -1 ) 24 25 26 27 28 29 30 TP ( mol L -1 ) 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18 0.20 Si ( mol L -1 ) 24 26 28 30 32 34 Fe (nmol L -1 ) 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 Ti (nmol L -1 ) 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 Al (nmol L -1 ) 7.0 7.5 8.0 8.5 9.0 9.5 10.0 Figures 5a-d. Depth profiles from Sheboygan Reef South 80m Station. Dashed horizontal lines indicate approximate location of thermocline (blue), chlorophyll maximum (green), and fluoresence maximum (black). a a a b a a c a a d a c a c a a Klein 141 Pb (pmol L -1 ) 70 80 90 100 110 120 130 140 150 Depth (m) 0 20 40 60 80 Cu (nmol L -1 ) 6 7 8 9 10 11 Ag (pmol L -1 ) 0 2 4 6 8 V (nmol L -1 ) 4.7 4.8 4.9 5.0 5.1 5.2 Ni (nmol L -1 ) 6 7 8 9 10 11 Mo (nmol L -1 ) 8.0 8.5 9.0 9.5 10.0 10.5 Ba (nmol L -1 ) 900 1000 1100 1200 1300 1400 1500 1600 1700 Co (pmol L -1 ) 70 80 90 100 110 120 Depth (m) 0 20 40 60 80 Mn (nmol L -1 ) 1.6 1.7 1.8 1.9 2.0 2.1 Zn (nmol L -1 ) 40 60 80 100 120 140 160 180 200 Cd (pmol L -1 ) 60 70 80 90 100 110 120 Figures 5e-g. Depth profiles from Sheboygan Reef South 80m Station. Dashed horizontal lines indicate approximate location of thermocline (blue), chlorophyll maximum (green), and fluoresence maximum (black). e a c a c a a f a e c a c a a g a c a c a a Klein 142 PARAM UNITS HARBOR OUTBOUND TRANSECT MGAP MEAN STDEV MINIMUM MAXIMUM BIOMASS COMPONENTS CHLa ug/L 10.21 0.65 0.24 0.45 1.10 pSIL uM 10.62 1.48 0.23 1.09 1.76 TP uM 1.04 0.08 0.03 0.06 0.15 BIOMASS NUTRIENTS NH4+ uM 2.34 0.05 0.04 <0.02 0.10 NO3+NO2 uM 62.73 22.79 2.26 20.00 27.08 dSiO2 uM 31.92 25.51 3.05 20.22 28.20 SRP uM 0.107 0.009 0.002 0.005 0.012 VITAMINS B1 pM 10.33 8.05 7.81 <0.020 24.79 B2 pM 23.73 9.66 6.49 1.90 19.72 B7 pM 21.70 19.01 18.86 <0.010 51.04 Met pM BDL 23.15 64.46 BDL 182.65 METALS Fe pM ND 15.60 9.97 7.43 33.97 Al nM ND 21.17 10.48 11.40 39.53 Co nM ND 0.13 0.04 0.10 0.20 Mo nM ND 12.69 5.47 8.95 24.79 Mn pM ND 2.65 0.69 1.79 3.44 Cu nM ND 9.45 2.73 5.02 12.27 Zn nM ND 172.48 97.52 78.41 350.32 Cd nM ND 0.12 0.05 0.07 0.23 P nM ND 76.89 24.63 42.38 123.43 Ti nM ND 0.45 0.20 0.31 0.88 V nM ND 5.46 0.38 4.89 6.05 Ni nM ND 15.02 16.96 5.91 53.08 Ag nM ND 0.01 0.01 BDL 0.02 Pb nM ND 0.54 1.17 0.02 3.19 Table 1. Ship's science pump (2m) concentrations of biomass, nutrient, vitamin, and trace metal constituents of Lake Michigan during the transect from Milwaukee Harbor to mid- lake station N145. Nine stations approximately 9 km apart were sampled outside of the harbor gap ( 5.5 - 88.5 km ENE from Milwaukee ). ND, not determined; BDL below detection limit if undefined. Klein 143 Study Study Site LOD Min Max Bioassay LOD Min Max Bioassay LOD Min Max Bioassay Daisley (1969) Several UK lakes 0.738 0.74 2.22 E. gracilis Carlucci & Bowes (1972) Lake Tahoe, CA 0.074 0.81 0.81 C. nana 1.64 2.46 6.96 A. carterae 11.86 14.82 112.67 M. lutheri Clear Lake 0.074 3.32 1.64 26.2 11.86 1185.99 Ohwada & Taga (1972) Lake Sagami, Japan 0.369 0.74 14.76 L. leichmanii 0.41 28.65 204.66 Achromobacter 29.65 148.25 1185.99 C. albedus Nishijima & Hata (1977) Lake Kojima, Japan * 1.55 26.57 L. leichmanii * 9.41 6.55 L. plantarum * 59.3 6819.46 L. fermenti Cavari & Grossowicz (1977) Lake Kinneret, Israel 0.738 1.11 73.8 E. coli Parker (1977) Lake Washington, WA * 2.73 6.94 O. malhamensis Kurata & Kadota (1981) Lake Biwa, Japan * 0.15 4.47 L. leichmanii * 0.57 17.07 L. arabinosus * 12.16 316.07 L. fermenti Kurata et al (1982) Lake Biwa, Japan * 0.63 5.99 L. leichmanii * 12.69 116.65 L. arabinosus * 237.2 1719.69 L. fermenti This study Lake Michigan, WI 0.003 0.13 3.1 0.01 1.2 51 0.02 0.59 66 B 12 (pmol L -1 ) B 7 (pmol L -1 ) B 1 (pmol L -1 ) Al P Ti V Mn Fe Co Ni Cu Zn Mo Cd Pb Ag This study (Lake Michigan) 6.4 - 45 22 - 130 0.21 - 1.33 4.9 - 6.4 1.7 - 4.6 3.5 - 33 0.08 - 0.49 4.9 - 88 6.7 - 44 63 - 1100 9.1 - 12 0.07 - 0.30 0.02 - 0.26 0.002 - 0.024 Lake Michigan (Shafer and Armstrong 1990) 10 9 0.17 0.25 Lake Tahoe, California (Klein and Sañudo-Wilhelmy unpublished data) 2.7 - 65 7.4 - 76 0.03 - 0.34 11 - 13 0.25 - 1.2 0.91 - 23 0.01 - 0.04 0.09 - 1.5 0.64 - 5.9 4.4 - 70 32 - 36 0.03 - 0.08 0.005 - 0.088 0.001 - 0.034 Lake Superior (Sherrell 2004) 0.3 - 8.0 1.1 - 100 0.005 - 0.28 10 - 13 10 - 13 1 - 7 0.05 - 0.13 N Fe Zn Cu Co Cd Mo B 12 B 1 B 7 Lake Michigan ambient mol : mol P 236 6.7 x 10 -2 1.5 0.11 1.2 x 10 -3 9.5 x 10 -4 9.0 x 10 -2 6.9 x 10 -12 7.1 x 10 -10 1.7 x 10 -9 Literature mol : mol P 16 7.5 x 10 -3 8.0 x 10 -4 3.8 x 10 -4 1.9 x 10 -4 2.1 x 10 -4 3.0 x 10 -5 2.3 x 10 -12 7.8 x 10 -10 7.3 x 10 -13 Ambient : Literature 15 9 1900 280 6.4 4.5 3000 3.1 0.91 2400 Table 2. Survey of select published B-vitamin data from lakes including method limits of detection, minimum detectable and maximum observed values, and bioassay organisms utilized. *No LOD given. Table 4. Nutrient limitation stoichiometry, standardized to P. Literature values for N and trace metals are from Ho et al (2003), B- vitamins use median values from data compiled in Taylor and Sullivan (2008) and Tang et al (2010). Data reported in cell volume yields (µm 3 biomass) were converted to moles C after Montagnes et al (2004). Table 3. Comparison of concentrations of dissolved trace metals (nmol L -1 ) in this study with older Lake Michigan data and oligotrophic lakes Tahoe and Superior. Klein 144 Lake Michigan Ks B 7 (pmol L -1 ) 0 10 20 30 40 50 60 B 1 (pmol L -1 ) 0 50 100 150 200 B 12 (pmol L -1 ) 0 2 4 6 8 10 12 14 Figures 6a-c. Box-and-whisker plots of observed Lake Michigan vitamin concentrations vs. literature Ks values for phytoplankton half-saturation growth constants (Taylor and Sullivan 2008, Tang et al 2010). a b c

Abstract (if available)

Abstract Small molecular weight, halogenated hydrocarbons (halocarbons) are known to be produced by marine phytoplankton and macroalgae. Halocarbon gases play important roles in climate and atmospheric chemistry and are analogues to ozone-depleting anthropogenic chlorofluorocarbons banned by the Montreal Protocol in 1989. This dissertation presents evidence from laboratory culture experiments and field sampling at the San Pedro Ocean Time-series that marine heterotrophic bacteria are also likely globally significant producers of halocarbons.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Multi-dataset analysis of bacterial heterotrophic variability at the San Pedro Ocean Time-series (SPOT): an investigation into the necessity and feasibility of incorporating a dynamic bacterial c...

PDF

The impact of the concentration and distribution of dissolved and particulate B-vitamins and their congeners on marine microbial ecology

PDF

Concentration and size partitioning of trace metals in surface waters of the global ocean and storm runoff

PDF

Biological nitrogen fixation associated with living and decomposing macroalgae

PDF

The impact of mesoscale and submesoscale physical processes on phytoplankton biomass, community composition, and carbon dynamics in the oligotrophic ocean

PDF

Identifying functional metabolic guilds: a computational approach to classifying heterotrophic diversity in the marine system

PDF

Microbe to microbe: monthly microbial community dynamics and interactions at the San Pedro Ocean Time-series

PDF

Thermal acclimation and adaptation of key phytoplankton groups and interactions with other global change variables

PDF

Comparative behavior and distribution of biologically relevant trace metals - iron, manganese, and copper in four representative oxygen deficient regimes of the world's oceans

PDF

B-vitamins and trace metals in the Pacific Ocean: ambient distribution and biological impacts

PDF

Viral, archaeal and bacterial colonization of native marine rock substrata

PDF

Microbial communities in marine sediments affecting and effecting biogeochemical cycling: influence of microbial ecology on geochemical transformations in two contrasting marine settings

PDF

Ecosystem export efficiency in an upwelling region: a two-year time series study of vertical transport, particle export and in-situ net and gross oxygen production

PDF

The dynamic regulation of DMSP production in marine phytoplankton

PDF

The distribution of B-vitamins in two contrasting aquatic systems, and implications for their ecological and biogeochemical roles

PDF

The distribution and speciation of copper across different biogeochemical regimes

PDF

Future impacts of warming and other global change variables on phytoplankton communities of coastal Antarctica and California

PDF

Dynamics of marine bacterial communities from surface to bottom and the factors controling them

PDF

Iron-dependent response mechanisms of the nitrogen-fixing cyanobacterium Crocosphaera to climate change

PDF

Ecological implications of daily-to-weekly dynamics of marine bacteria, archaea, viruses, and phytoplankton

Asset Metadata

Creator Klein, Nicholas J. (author)

Core Title Marine biogenic halocarbons: potential for heterotrophic bacterial production and seasonality at San Pedro Ocean Time-series

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Ocean Sciences

Publication Date 12/14/2016

Defense Date 03/11/2016

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Biogeochemistry,Climate,halocarbons,marine heterotrophic bacteria,OAI-PMH Harvest

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Sanudo-Wilhelmy, Sergio (committee chair), Hammond, Doug (committee member), Webb, Eric (committee member), West, Joshua (committee member)

Creator Email nicholjk@usc.edu,njklein87@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-325429

Unique identifier UC11214565

Identifier etd-KleinNicho-4966.pdf (filename),usctheses-c40-325429 (legacy record id)

Legacy Identifier etd-KleinNicho-4966.pdf

Dmrecord 325429

Document Type Dissertation

Format application/pdf (imt)

Rights Klein, Nicholas J.

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA