Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Determination of dissolved trace metals in the western North Pacific Yang, Lu 1993

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1993_spring_yang_lu.pdf [ 4.24MB ]
Metadata
JSON: 831-1.0061143.json
JSON-LD: 831-1.0061143-ld.json
RDF/XML (Pretty): 831-1.0061143-rdf.xml
RDF/JSON: 831-1.0061143-rdf.json
Turtle: 831-1.0061143-turtle.txt
N-Triples: 831-1.0061143-rdf-ntriples.txt
Original Record: 831-1.0061143-source.json
Full Text
831-1.0061143-fulltext.txt
Citation
831-1.0061143.ris

Full Text

DETERMINATION OF DISSOLVED TRACE METALS IN THEWESTERN NORTH PACIFICbyLU YANGB. Sc. (Hons.), Nanjing University, 1987A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF SCIENCEinTHE FACULTY OF GRADUATE STUDIES(Department of Chemistry)We accept this thesis as conformingo the required standardTHE UNIVERSITY OF BRITISH COLUMBIAMarch, 1993©Lu Yang, 1993In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature)Department of The University of British ColumbiaVancouver, CanadaDate^4v;\it 6^(99,2 DE-6 (2/88)ABSTRACTA method for analyzing dissolved cadmium, zinc, nickel, copper, lead, aluminum,manganese, cobalt, cerium, and gallium in seawater was developed using a Chelex-100resin for preconcentration/separation of these metals from seawater matrix, followed byICP-MS or GFAAS detection. This method was used to investigate trace metals in thewestern North Pacific, a region for which little trace metal data are available.Concentrations of these elements are found to be in the range of 0.7 pmol/kg to 12nmol/kg, analytical precision is better than 7%, and the seawater detection limits are in thelow picomolar range (2-40 pmol/kg) after a 125-fold preconcentration. The accuracy ofthis technique was verified by the analysis of standard reference materials from the NationalResearch Council of Canada for Cd, Ni, Zn, Cu, Pb, Mn, and Co. Results are in goodagreement with the certified values, within 99% confidence limit.Determining the spatial distribution of trace elements in new oceanic regions helps toelucidate the processes which control the chemical composition of seawater. Data from 6vertical profiles in the western and central North Pacific Ocean and 14 surface samplesalong a transect from Japan to Hawaii, collected in the spring of 1991, are reported here.Vertical profiles for dissolved Cd, Ni, Zn, Cu, Pb, Mn, and Co from the central gyrestation (28°N, 175°E), close to Hawaii, agree well with previous results (Bruland, 1980;Landing and Bruland, 1987; Flegal and Patterson, 1983; Bruland, et al., 1993),demonstrating reliable sampling and analytical procedures. Remarkably similar trace metalconcentrations and vertical distributions are found in the Kuroshio Current station (35°N,155°E), about 3500 km west of the central gyre station. In the coastal region near Japan(38°N, 146°E), enriched trace metal concentrations are observed in the upper watercolumn, with a shallower maximum for those elements with a mid-depth maximum.Dissolved Al, Ce, and Ga in the central North Pacific show similar major trends to otherstudies (Orians and Bruland, 1985, 1988; de Barr, et al., 1985), with the exception ofmaximum Al concentrations found at 100-250 m rather than at the surface, and higherintermediate values. The sub-surface dissolved maximum observed in this study could bedue to advection, via the North Equatorial Current, of low Al waters from the coast ofNorth America to this area. The higher mid-depth Al concentrations could be due to inputsfrom the Karin Ridge.The surface transect (25m depth) for dissolved Cd, Ni, Zn, Cu, Pb, and Mn in thewestern North Pacific generally shows high coastal values and low concentrations in theopen ocean. However, at a station midway along the transect from Japan to Hawaii (34°N,157°E), high values are observed for all elements. This is interpreted as the advection ofcoastal waters being carried out into the central gyre via a cold-core ring, broken off fromthe Kuroshio Current. This feature is observed in both salinity and temperature as well asin the trace metal data. Station HS-2, west of the Kuroshio Current, shows upwelling.Nutrient-type elements which have a surface minimum (Cd, Zn, Cu, Ni) are elevated,while scavenging-type elements, such as Mn, which have a surface maximum are depletedat this station relative to the coastal station, HS-1. The data from surface waters show ahigher degree of variability than observed in vertical profiles. Since good agreement wasfound between vertical profiles of these metals in this study and other previous results(Bruland, 1980; Landing and Bruland, 1986; Flegal and Patterson, 1983), the surface dataare believed to reflect the real variation of these metals in surface seawater, not analytical orsampling errors. One of the most surprising features of the surface transect is the lack of astrong gradient from increased eolian inputs closer to the coastal region. Dust and leadfluxes are both thought to vary by nearly two orders of magnitude across this transect, yetthe range observed for the trace elements is only 2 to 4 fold. Much of the expected inputappears to be balanced by increased scavenging in the coastal region. It is also likely thatsampling closer to the coast of Japan (samples are not available from this area) would haveshown the effects of these increased inputs.iiiTABLE OF CONTENTSPageABSTRACT^ iiTABLE OF CONTENTS^ ivLIST OF TABLES viiiLIST OF FIGURES^ ixLIST OF ABBREVIATIONS^ xiACKNOWLEDGMENTS xiiCHAPTER 1:INTRODUCTION^ 11.1 BACKGROUND 11.2 DETERMINATION OF TRACE METALS IN SEAWATER^ 21.3 MARINE GEOCHEMISTRY OF Cd, Zn, Ni, Cu, Al, Pb, Ga, Ce,Mn, and Co^ 41.3.1 CADMIUM, ZINC AND NICKEL^ 51.3.2 COPPER^ 61.3.3 ALUMINUM AND LEAD^ 61.3.4 GALLIUM^ 71.3.5 CERIUM 81.3.6 MANGANESE AND COBALT^ 81.4 AIM OF PRESENT STUDY^ 9CHAPTER 2:EXPERIMENTAL^ 112.1 INSTRUMENTATION^ 112.1.1 ICP-MS^ 112.1.2 GFAAS 12iv2.1.3 OTHERS^ 132.2 MATERIAL AND REAGENTS ^132.2.1 CHELEX- 100 132.2.2 AMMONIUM ^132.2.3 ACIDS 142.2.4 AMMONIUM ACETATE pH ADJUSTING SOLUTION ^ 142.2.5 DISTILLED DEIONI7P.D WATER 142.2.6 METAL STANDARDS ^142.2.7 ENRICHED 62Ni ISOTOPE 152.2.8 REFERENCE MATERIALS ^152.3 OPERATING CONDITION AND OPTIMIZATION 162.3.1 ICP-MS ^162.3.2 QUANTITATIVE ANALYSIS ON ICP-MS 182.3.2.1 STANDARD CALIBRATION WITH INTERNAL STANDARD-182.3.2.2 ISOTOPE DILUTION TECHNIQUE ^ 182.3.3 GFAAS 192.4 STUDY AREA ^ 202.5 SEAWATER SAMPLES 22CHAPTER 3:METHOD DEVELOPMENT ^233.1 OPTIMIZATION OF CHELEX-100 PRECONCENTRATION 233.1.1 pH ^233.1.2 FLOW RATE 253.1.3 VOLUME OF CHELE.X-100- ^253.1.4 ELUENT CONCENTRATION AND VOLUME 273.1.5 EQUILIBRATION TEST FOR NICKEL ANALYSIS ^273.2 GENERAL PROCEDURE^ 283.3 ANALYTICAL FIGURES OF MERIT^ 283.3.1 RECOVERY^ 283.3.2 PRECISION 303.3.3 ACCURACY^ 313.3.4 LIMITS OF DETECTION^ 323.4 CONCLUSION^ 33CHAPTER 4OCEANOGRAPHIC RESULTS^ 354.1 VERTICAL PROFILES 354.1.1 NUTRIENT-TYPE ELEMENTS^ 354.1.1.1 DISSOLVED Cd 394.1.1.2 DISSOLVED Zn^ 444.1.1.3 DISSOLVED Ni 484.1.2 NUTRIENT-TYPE WITH SCAVENGING ELEMENT (Cu)-^ 534.1.3 SCAVENGED TYPE ELEMENTS^ 584.1.3.1 SCAVENGED TYPE ELEMENTS WITHONLY EXTERNAL SOURCES (Al, Pb and Ga)-^ 584.1.3.2 OXIDATIVE SCAVENGING TYPE ELEMENT (Mn, Co and Ce)-- 664.2 SURFACE TRANSECT^ 764.3 CONCLUSIONS 81CHAPTER 5:DISSOLVED MANGANESE AND COBALT IN A COASTAL^835.1 INTRODUCTION^ 835.2 METHOD 845.2.1 STUDY AREA^ 84vi5.2.2 SAMPLING METHOD-^ 845.2.3 ANALYSIS METHOD- •855.3 RESULTS AND DISCUSSION^ 865.4 CONCLUSIONS^ 93CHAPTER 6:SUMMARY^ 94REFERENCES 98viiLIST OF TABLESTable^ Page2.3.1 ICP-MS operation conditions^ 172.3.2.2 62Ni/60Ni ratio measurements in standard solutions^ 192.3.3 GFAAS operation conditions^ 203.1.3 Effect of resin volume on metal recoveries^ 273.3.1.1 Recoveries from seawater matrix at pH 8.0±0.1 293.3.1.2 Recoveries from seawater matrix at pH 6.5±0.2^ 293.3.2 Precision of triplicate analyses^ 303.3.3.1 The determination of dissolved Mn, Ni, co, Pb, Cd,Zn, and Cu in seawater reference materials^ 313.3.3.2 Statistical analysis on NASS-3^ 323.3.4 Detection limits obtained for ICP-MS analysis of seawater samples^ 334.1.1 Dissolved Ni, Cd, Zn, Cu, Pb, Mn and Co inthe western North Pacific sample^ 364.1.2 Dissolved Ni, Cd, Zn, Cu, Pb, Mn and Co inthe central North Pacific samples 374.1.3 Dissolved Al, Ga, and Ce in the central North Pacific samples^ 385.3.1 Dissolved Mn, Co, and oxygen in seawater in Howe Sound^ 875.3.2 The statistical analysis for dissolved Mn and Co (nmol/kg) in seawatercollected in Howe Sound using CID rosette Niskin bottles with astainless wire and "Go-flo" sampler with a galvanized wire^ 92viiiLIST OF FIGURESFigure Page1.3 Three major types of distributions^ 52.1 PlasmaQuad overall schematic 122.4 Study area^ 213.1.1.1 Recoveries at various pH^ 243.1.1.2 Recoveries study at various flow rates^ 264.1.1.1 Dissolved Cd distribution at stations HS-1, HS-5 and HS-1(}^ 404.1.1.2 Dissolved Cd distribution at stations HS-14, HS-15 and HS-16- ^ 414.1.1.3 Dissolved Cd vs. nutrients^ 434.1.1.4 Dissolved Zn distribution at stations HS-1, HS-5 and HS-10^ 454.1.1.5 Dissolved Zn distribution at stations HS-14, HS-15 and HS-16-^ 464.1.1.6 Dissolved Zn vs. dissolved Si^ 474.1.1.7 Dissolved Ni distribution at stations HS-1, HS-5 and HS-10^ 494.1.1.8 Dissolved Ni distribution at stations HS-14, HS-15 and HS-16-^ 504.1.1.9 Dissolved Ni vs. nutrients^ 514.1.1.10 Dissolved Ni vs. nutrients, upper 800m^ 524.1.2.1 Dissolved Cu distribution at stations HS-1, HS-5 and HS-la^ 544.1.2.2 Dissolved Cu distribution at stations HS-14, HS-15 and HS-16- ^ 554.1.2.3 Dissolved Cu vs. salinity and temperature vs. salinity plotsat stations HS-1 and HS-10-^ 574.1.3.1 Dissolved Pb distribution at stations HS-1, HS-5 and HS-1a^ 594.1.3.2 Dissolved Pb distribution at stations HS-14, HS-15 and HS-16-^ 614.1.3.3 Dissolved Al distribution at stations HS-14, HS-15 and HS-16-^ 634.1.3.4 Dissolved Ga distribution at stations HS-14, HS-15 and HS-16-^ 654.1.3.5 Dissolved Mn and oxygen at stations HS-1, HS-5 and HS-10^ 674.1.3.6 Dissolved Mn and oxygen at stations HS-14, HS-15 and HS-16 -^ 694.1.3.7 Dissolved Co and oxygen at stations HS-land HS-5^ 71ix4.1.3.8 Dissolved Co and oxygen at stations HS-14, HS-15 and HS-16-^ 734.1.3.9 Dissolved Ce distribution at stations HS-14, HS-15 and HS-16^ 754.2.1 Dissolved Cu, Zn, Ni, and Cd in surface waters^ 774.2.2 Dissolved Pb, Mn and Co in surface waters 784.2.3 Temperature and salinity in surface waters^ 795.3.2 Dissolved Mn, Co and oxygen in Howe Sound 885.3.3 Dissolved Mn vs. dissolved Co in seawater, Howe Sound^ •895.3.4 The logCo and logMn vs. logO2 seawater profiles from 50mto 250m at Howe Sound^ 91LIST OF ABBREVIATIONSAAS^Atomic absorption spectroscopyAV Research vessel "Aleksandr Vinogradov"DI^Distilled deionized waterFI-ICP-MS^Flow injection inductively coupled plasma mass spectrometryGFAAS Graphite furnace atomic absorption spectroscopyHAc^Acetic acidHEPA High efficient particle airICP-AES^Inductively coupled plasma atomic emission spectroscopyICP-MS Inductively coupled plasma mass spectrometryID-ICP-MS^Isotope dilution inductively coupled plasma mass spectrometryIOS^Institute of Ocean Sciencekg Kilogramml^Millilitermiz Mass to charge ratioN^NormalityNAA Neutron activation analysisnM^Nanomolarnmol/kg^Nanomole per kilogrampM Picomolarpmol/kg^Picomole per kilogramppb Parts per billionppm^Parts per millionRF Radio frequencyRSD^Relative standard deviationXRF X-ray fluorescence spectroscopyxiACKNOWLEDGMENTSI want to take this opportunity to thank the many people who have helped to make thiswork possible. I thank Kristin J. Orians for her unending support and supervisionthrough-out this study. I thank my lab mates and friends, Brad McKelvey, Bert Mueller,Robert Mugo, Sanny Chan, Helen Nicolidakis, Lucila Lares and John Crusius for theirvaluable help and friendship. Kristin J. Orians, Brad McKelvey, Robert Mugo and HelenNicolidakis are also thanked for collecting the seawater. Institute of Ocean Science isthanked for providing the salinity, temperature, oxygen, and nutrients data. Finally, Ithank my husband, Yuxiang Gong for encouraging me and believing in me from thebeginning. He is largely responsible for any successes I may have in my life.xiiCHAPTER 1INTRODUCTION1.1 BACKGROUNDTrace metals are defined as metals whose concentrations are less than a few parts perbillion (ppb). Trace metal analysis is very important for understanding environmentalsystems. It can help us to elucidate the roles trace metals play in biogeochemical cycles,and how human activities impact on the natural environment. Trace metals can also be usedas water mass tracers in the study of ocean circulation. Modern analytical measurementsmust therefore be capable of precise and accurate measurements at low concentrations.Several analytical techniques are currently available for trace metals analyses in seawater(Henshaw, et al., 1989). Inductively coupled plasma atomic emission spectroscopy(ICP-AES) and X-ray fluorescence spectroscopy (XRF) are multielemental techniques, butthey suffer from overlapping emission line interferences, particularly for heavy ions, orfrom matrix problems. In addition, detection limits are too high for open ocean trace metalanalysis. Graphite furnace atomic absorption spectroscopy (GFAAS) has low detectionlimits, but it is a single element technique and, therefore, is time consuming formultielemental analysis. Neutron activation analysis (NAA) also provides good precisionat trace levels, but is time consuming and expensive.Inductively coupled plasma mass spectrometry (ICP-MS) is a new multielementalanalytical technique which provides rapid sample throughput, low detection limits, a largeworking range, and the ability to make isotope measurements. ICP-MS uses the ability ofan argon ICP to efficiently generate singly charged ions from the elemental species within aliquid or solid sample. These ions are then directed into a quadrupole mass analyzer whichseparates the ions according to their mass to charge ratio. The selected ions are quantified1in the channel electron mulitipler. ICP-MS has provided the analyst with an important andpowerful tool for trace elements analyses. Since ICP-MS became commercially available in1983, trace metals analyses in variable sample matrices have been performed. It has beenused for elemental analysis in earth science, water resources, food science, the petroleumindustry and metallurgical laboratories (Date and Gray, 1989). In this study dissolved Cd,Zn, Ni, Cu, Al, Pb, Mn, Ce, and Ga were analyzed in seawater using ICP-MS (and Co byGFAAS), after simultaneous preconcentration on Chelex-100 resin.1.2 DETERMINATION OF TRACE METALS IN SEAWATERSeawater is a highly complex matrix with a large concentration of dissolved salts(3.5% by weight). The major ions: Cl - , Na+, K+, SO4 2- , Ca2+, and Mg2+ account for99% of these dissolved salts. Trace elements, such as Cd, Zn, Ni, Cu, Al, Pb, Mn, Co,Ce, and Ga, on the other hand, exist at concentrations of less than lgg/kg (ppb). Toaccurately determine trace metals in the complex seawater matrix, contamination controlsduring sampling, storage and preconcentration/separation procedures are required, inaddition to low instrumental detection limits.In 1979, a technique for seawater sampling which is contamination-free for traceelements was described by Bruland and co-workers (Bruland, et al., 1979). This involvesthe use of Teflon0-lined "Go-flo" (General Oceanic) bottles, suspended on a Kevlar® line,with Teflon® messengers to trigger bottle closure. In addition, a trace metal cleanenvironment is made onboard ship to avoid exposing the samples to the ship'senvironment.Commonly used techniques for preconcentration and separation involve solventextraction (Bruland, et al., 1979; Knauer, et al., 1982; Orians and Bruland, 1986; Landingand Bruland, 1987), co-precipitation (Boyle, et al., 1977; German and Elderfield, 1990),ion exchange and chelating resins (Berman, et al., 1980; Boyle, et al., 1987; Orians andBruland, 1988). Chelex-100, which has an iminodiacetate functional group, is one of most2commonly used chelating resins for preconcentrating trace metals from seawater (Kingston,et al., 1978; Sturgeon, et al., 1980; Bruland, et al., 1979; Berman, et al., 1980; Paulson,1986; Pai, et al., 1988: Van Berkel, et al., 1988) due to its low blanks, high capacity,strong preference for transition metals over alkali metals and commercial availability.Combining clean sampling and analytical techniques with sensitive analyticalinstruments, chemical oceanographers have been able to accurately measure trace metalconcentrations in seawater in the last 15 years. For example, Bruland (1980) reported thevertical profiles of dissolved Cd, Zn, Ni and Cu in the North Pacific ocean by usingGFAAS detection with dithiocarbamate extraction preconcentration. Berman, et al. (1980)measured Fe, Mn, Cu, Ni and Zn in coastal seawater by using ICP-AES withpreconcentration on Chelex-100. Landing and Bruland (1987) measured Mn and Fe inseawater by using GFAAS with preconcentration by an 8-hydroxyquinoline liquid-liquidorganic extraction technique. German and Elderfield (1990) measured the rare earthelements (REEs) in the NW Indian Ocean using mass spectroscopy with ferric hydroxideco-precipitation preconcentration. Wilhelmy and Regal (1991) measured Pb, Cd, Mn, Feand Zn in coastal seawater using GFAAS after preconcentration on Chelex-100. Resingand Mottl (1992) measured Mn in seawater using flow injection analysis with on-linepreconcentration and spectrophotometric detection.The recent introduction of ICP-MS has provided analysts with a more powerful tool fortrace metals analyses in aqueous samples. McLaren, et al. (1985) determined Mn, Co, Ni,Cu, Zn, and Pb in a coastal seawater reference material, CASS-1, by both standardadditions and isotope dilution techniques, using ICP-MS with preconcentration on silica-immobilized 8-hydroxyquinoline. Shabani, et al. (1990) measured the rare earth elementsand yttrium in seawater by ICP-MS after preconcentration with solvent extraction (withethylhexyl hydrogen phosphates in heptane) followed by back extraction (octyl alcohol).Other recent studies have involved determination of gold in seawater, using flow injectionICP-MS (FI-ICP-MS) with preconcentration on an anion exchange resin (Falkner and3Edmond, 1990), barium in pore waters using isotope dilution ICP-MS (ID-ICP-MS)without preconcentration (Klinkhammer and Chan, 1990), thorium in sediments usingID-ICP-MS after preconcentration via coprecipitation and separation via ion exchange(Shaw and Francois, 1991), and titanium, gallium and indium in seawarer using ICP-MSafter preconcentration on 8-hydroxyquinoline resin (Orians and Boyle, 1993). Thesestudies would have been much more difficult without this new technique.1.3 MARINE GEOCHEMISTRY OF Cd, Zn, Ni, Cu, Al, Pb, Ga, Mn, CoAND CeThe oceanic distributions and biogeochemical behavior of dissolved Cd, Zn, Ni, Cu,Al, Pb, Ga, Mn, Co and Ce are controlled by complex interactions among inputs, internalcycling and removal processes, coupled with physical transport and mixing in the oceans(Bruland, 1983; Whitfield and Turner, 1987). Our understanding of the marinebiogeochemistry of these elements is preliminary, but has been growing rapidly in the lastfifteen years.Trace element behavior can be, in a general sense, divided into three categories.Conservative distributions (fig. 1.3 A) are observed for elements whose concentrationsvary only with salinity, due to very slow removal from the oceans. These elements usuallyhave high concentrations relative to their crustal abundance and have long residence timesin the ocean. The trace metals Rb and Cs are two examples of conservative elements.Some trace metals exhibit nutrient-type behavior (fig. 1.3 B). These elements are notnecessarily nutrients, but they do show a strong correlation with nutrient distributions.Vertical distributions of these elements show a surface depletion and increasingconcentrations with depth, reaching a maximum coherent with the nutrient maximum. Theincrease of these nutrient-type trace metals in the deep waters is explained by the sinking ofdead biota and fecal pellets which then decompose at depth, releasing nutrients andassociated metals to the water column. Cadmium, Zn, and Ni are three elements with4nutrient-type distributions. Scavenged distributions (fig. 1.3 C) are observed for elementswhose distributions primarily reflect their external sources, with decreasing concentrationsaway from the sources due to rapid removal from the water column, typically by adsorptiononto sinking particles (scavenging removal). As a result, these metals have short oceanicresidence times and variable concentrations in different regions of the oceans. Scavengedmetals are valuable as tracers for elucidating the transport and mixing mechanisms in theoceans. Lead and Al are two scavenged elements.1.3.1. Cadmium, Zinc and NickelCadmium, a multi-isotopic, divalent, second row transition metal below Zn in theperiodic table, exists as a CdC12 complex in seawater. It is a non-essential element, yetshows a nutrient-type distribution in the oceans. Since the size of Cd2+ is similar to Zn2+,it may be indiscriminately taken up by phytoplankton in their quest for Zn (Knauer andMartin, 1980; Price and Morel, 1990). Nickel and Zn, also multi-isotopic, divalent, firstrow transition metals, exist as a mixture of CO3 2- , OH- complexes and some simplehydrated ions in seawater. They are micro-nutrients in seawater and also co-vary withmajor nutrients (such as phosphate and silicic acid) in the oceans (Bruland, 1980; Yeats and5Campbell, 1983; Westerlund and Ohman, 1991). These relations show that Cd, Zn, andNi are linked to the marine bioproduction cycle. Vertical transport with biogenic particles isthe major process responsible for the distribution of Cd, Zn, Ni and other nutrient-typetrace metals. In general, the vertical profiles of these three metals show a surface depletion,increasing concentrations with depth, reaching a maximum coherent to the nutrientsmaxima. Below this maximum the concentrations of these dissolved metals decreaseslightly in the deep sea, due to deep ocean circulation patterns and the intensity ofrespiration.1.3.2 CopperCopper, a multi-isotopic, divalent, first row transition metal, exists as a mixture ofCO32- and OH- complexes, as well as some hydrated Cu 2+ in seawater. It is also arequired micro-nutrient in seawater. Studies of Cu in different oceans (Bruland, 1980,1983; Yeats and Cambell, 1983; Westerlund and Ohman, 1991) have shown that thedissolved Cu distribution is controlled by internal cycles, scavenging processes in theintermediate and deep waters, and a bottom source. The vertical profiles of dissolved Cushow a surface depletion, and a gradual increase in concentration towards the bottom. Thistype of distribution can be used to estimate the rate of scavenging removal in the deep oceanusing a vertical advection-diffusion model. The residence times of Cu have been reportedto be around 1000 years in the deep water (Boyle, et al., 1977; Bruland, 1980; Bruland andFranks, 1983).1.3.3 Aluminum and LeadAluminum is a mono-isotopic, trivalent metal, which tends to hydrolyze in seawater toform Al(OH)3 and Al(OH)4 - . Lead is a multi-isotopic, divalent heavy metal, which existsas a mixture of Cl - and CO3 2- complexes in seawater. Both elements are non-essentialelements with scavenged distributions. Unlike dissolved Cd, Zn and Ni, which are mainly6controlled by internal cycles within the oceans, the vertical distributions of dissolved Al andPb are driven by scavenging removal processes and external inputs (Hydes, 1979; Oriansand Bruland, 1985; Boyle, et al., 1986; Flegal and Patterson, 1983). Aluminum has twomajor sources to the ocean. Eolian input has been proposed as the dominant source ofdissolved Al to the surface waters of the open ocean. Diffusion out of the sediments and/orsediment surface remineralization have/has been proposed as deep water sources.Dissolved Al vertical distributions therefore have a surface maximum, low concentrationsin intermediate and deep waters, and higher concentrations again near the bottom. Lead isbrought into the oceans mainly by atmospheric inputs, from alkyl leaded gasoline. Onceintroduced into the ocean, Pb is rapidly removed from the water via scavenging, leading todecreasing concentrations with depth in the deep sea. Lead concentrations in surfacewaters have been steadily decreasing with the declining use of leaded gasoline over the lastcouple of decades, resulting in a subsurface maximum in most regions (Boyle, et al., 1986;Regal and Patterson, 1983).1.3.4 GalliumGallium, a multi-isotopic, trivalent metal, which is below Al in the periodic table ofelements, tends to hydrolyze in seawater to form Ga(OH)3 and Ga(OH)4- . The primarysources of dissolved Ga to the open ocean may be atmospheric inputs of crustal dust to thesurface water and diffusion out of the sediments and/or a sediment surface remineralizationto the deep water (Orians and Bruland, 1988). The vertical distribution of dissolved Ga inprevious studies (Orians and Bruland, 1988; Shiller, 1988) shows a sub-surfacemaximum, low concentrations in the intermediate waters and increasing levels with depthinto the deep water. The subsurface maximum may be caused by either horizontaladvection and/or a vertical process involving exchange with sinking particles. Thesevertical profiles of dissolved Ga suggest complex controls on dissolved Ga with multiple7sources, reversible exchange and scavenging processes all contributing to the distributionof dissolved Ga in the ocean.1.3.5 CeriumCerium, a rare earth element, exists as CeCO3+, hydrated Ce3+ and CeC12+ in seawater(Whitfield and Turner, 1987; de Barr, et al., 1985; Bruland, 1983). Cerium is primarilybrought into the ocean by river and eolian inputs. As a sensitive redox element, Ce(HI)tends to be oxidized to insoluble Ce(IV) in oxygenated water, which adsorbs onto Fe andMn oxyhydroxides (Whitfield and Turner, 1987). In water with low 02, Ce(IV) can bereduced to soluble Ce(III). The vertical profiles of dissolved Ce in the Pacific and Atlantic(de Barr, et al., 1985) show a surface maximum, a sub-surface minimum, enrichedconcentrations in the 02 minimum zone, and decreasing concentration in the deep waters.German and Elderfied (1990) reported similar vertical distribution of dissolved Ce in theIndian ocean, except a slightly increasing concentration of Ce towards the bottom wasfound in their study.1.3.6 Manganese and CobaltManganese and Co, mono-isotopic, second transition metals, exist as Mn 2+, MnC1+and Coll- , CoCO3, CoC1+, respectively, in seawater. They are essential elements (Landingand Bruland, 1987; Yeats and Bewers, 1985; Knauer, et al., 1982). Manganese and Cohave multiple sources and complex marine geochemical controls. Manganese is broughtinto the oceans mainly by river, atmospheric and submarine hydrothermal inputs (Landingand Bruland, 1987; Wilhelmy and Flegal, 1991; Yeats and Bewers, 1985). Onceintroduced, Mn participates in a wide range of biogeochemical processes. Manganese maybe released from eolian particles in the surface of the ocean (Landing and Bruland, 1987),and dissolved Mn may also adsorb onto particles (Martin and Knauer, 1984). As arequired micro-nutrient, Mn may be incorporated into organic tissue and be involved in8internal cycling (Landing and Bruland, 1987). Manganese can exist in a variety ofoxidation states under environmental conditions; in low 02 water it exists as solubleMn(II), in high 02 water it is slowly oxidized to insoluble Mn (Ill, IV) oxyhydroxides.When Mn (III, IV) oxyhydroxides form, many trace metals (Co and Ce, for example) arescavenged onto these mineral phases (Knauer, et al., 1982; Kremling and Hydes, 1988).The dissolved Mn vertical profiles exhibit a surface maximum, a subsurface minimum, amaximum coincident with the oxygen minimum zone, and lowest values in the deep waters(Landing and Bruland, 1987; Yeats and Bewers, 1985). If in proximity to hydrothermalsources, another maximum may be observed at 2500-3000m depth. Although thedissolved Co concentration is much lower than that of dissolved Mn in seawater, it has asimilar behavior to Mn (Knauer, et al., 1982) due to scavenging onto Mn(III, IV)oxyhydroxides. There are still a paucity of dissolved oceanic Co data, due to the extremelylow concentrations of Co in the ocean and the resulting sampling and analytical difficulties(Knauer, et al., 1982; Sakamoto-Arnold and Johnson, 1987; Jickells and Burton, 1988).The preliminary data indicate a similarity between dissolved Co and Mn profiles,suggesting that dissolved Co may be involved in the same biogeochemical processes as Mnin the ocean.1.4 AIM OF PRESENT STUDYThe desire to increase our understanding of the factors that control the distribution oftrace metals in the oceans, and their importance in biological processes, has led to thesearch for sensitive, rapid and accurate techniques for their analyses in the seawater matrix.This study was therefore aimed at the development of a multielemental method for thedetermination of dissolved Cd, Zn, Ni, Cu, Pb, Al, Ga, Mn, Co and Ce in seawater bypreconcentration on a Chelex-100 resin with ICP-MS detection, and to use this method toinvestigate the distribution of these ten trace metals in the western and central North Pacificbetween Japan and Hawaii. The western North Pacific is a region where few data on trace9metal distributions are available. This study will help to elucidate the relative importance ofvariations in riverine and eolian sources, variations in primary productivity and associatedscavenging intensity, and advective transport on the distribution of trace metals in thewestern north Pacific Ocean.CHAPTER 2EXPERIMENTAL2.1 INSTRUMENTATION2.1.1 ICP-MSA VG "PQ2 Turbo Plus" ICP-MS was used in this study. The main components andarrangement of the VG ICP-MS system are shown in the figure 2.1. The system isequipped with a VG ICP torch and extraction system, an SX 300 quadrupole massanalyzer, and a channel electron multiplier (Galileo Electro-Optics Corp). The datahandling is performed on a Compaq 386 computer equipped with PQ vision 4.05 software.Pure (99.998%) argon gas, purchased from Medigas (Vancouver, Canada) was used as theplasma and carrier gas.The sample solution is usually introduced into the spray chamber as fine aerosoldroplets by a pneumatic nebulizer. The analyte is dried, vaporized, atomized and furtherionized into mostly singly charged ions in the plasma. These ions are extracted by thesampling and skimmer cones and then focused by the ion lenses and transported into themass analyzer, which separates ions by their mass to charge ratio. The selected ions travelto the channel electron multiplier for quantitation.11Figure 2.1 PlasmaQuad Overall Schematic2.1.2 GFAASGraphite furnace atomic absorption spectroscopy (GFAAS), using a Varian SpectraAA300 with Zeeman background correction, was used to determine dissolved Co inseawater samples. GFAAS measures absorbance of a vaporized and atomized sampleinjected into a graphite tube, and the concentration of the analyte is determined by12comparison with a standard calibration curve. GFAAS is a single element technique withlow detection limits for many elements.2.1.3 OtherspH measurements were performed using an Orion SA 520 pH meter equipped with a91-02 general purpose combination electrode.A masterflex cartridge pump (Cole-Parmer Instrument Co., Chicago, USA) was usedto pump seawater through the Chelex-100 columns for preconcentration.An analytical balance with precision of ± 0.001g (Mettler PM1200, Fisher Scientific,Vancouver, Canada) was used to quantify the weight of seawater, eluents, and reagents.2.2 MATERIAL AND REAGENTS2.2.1 Chelex-100Chelex-100 resin, a polystyrene divinyl benzene structured resin with iminodiacetatefunctional groups (Bio-Rad Laboratories., Richmond, USA), was used to preconcentrateand separate trace metals from the seawater matrix after acid cleaning (with 60 ml 2.0 NHNO3 by gravity through) and adjustment of pH.2.2.2 AmmoniumAmmonium (double subboiling distilled in quartz), purchased from Seastar ChemicalsInc.(Sidney, Canada), was used to adjust the pH of the acidified seawater samples and tomake ammonium acetate solution (see below).132.2.3 AcidsAcetic and hydrochloric acid (double subboiling distilled in quartz, Seastar ChemicalsInc., Sidney, Canada) were used to make an ammonium acetate solution (see below) and toacidify the seawater samples, respectively. Analytical reagent grade hydrochloric acid(BDH Inc. Toronto, Ontario, Canada) was used to clean polyethylene bottles and pipettips.Nitric acid (68-71%), environmental grade (Anachetnia Science Division of AnachemicCanada Inc.) was used to make 2.0 N HNO3 for cleaning and eluting the Chelex-100columns.2.2.4 Ammonium Acetate pH Adjusting SolutionAnimonium acetate (NH4Ac), 0.15 M, made from mixing ammonium and acetic acid indeionized water at pH 8.0 or 6.5, was used to pH adjust and rinse the Chelex-100columns.2.2.5. Distilled Deionized Water (DI H20)Water was first distilled with a Barnstead system (Barnstead/Thermolyne Corporation,Dubuque, Iowa) then passed through a "Milli-Q" deionization system (Millipore WatersAssociates, Milford, USA) installed in the laboratory at UBC, in the Department ofOceanography.2.2.6. Metal StandardsCertified atomic absorption standards for Mn(II), Ni(II), Co(H), Cu(II), Zn(II), Pb(II),Cd(II), Al(lE), Ga(III), and Ce(III) were used to prepare standard solutions for ICP-MSmeasurements. Stock standards of 1000 µg/ml Mn in 5% HNO3 (Johnson Matthey Inc.,Seabrook, NH, USA); 1000 ppm of Co, Ni, Pb, Cd, Cu and Zn in 0.3 M nitric acid (J. T.Baker Chemical Co., Phillipsburg, NJ); 1000 ppm of Al in 5% HC1, Ga and Ce in 5%14HNO3 (Johnson Matthey Inc., Seabrook, NH, USA) were diluted with 1% HNO3 to makeprimary standards. Indium, 1000 ppm in 5% HNO3, purchased from Johnson MattheyInc. Seabrook, NH, USA, was used as the internal standard in this study.Mixed aqueous standards were prepared by serial dilution of the primary standardsolutions with 2.0 N HNO3. The calibration standards prepared were 10 ppb, 20 ppb, 40ppb, 80 ppb, and 160 ppb in each metal. The In internal standard was at 10 ppb level in allsamples, blanks, and standards.2.2.7. Enriched IsotopeA 62Ni enriched stable isotope in oxide form (95.96% 62Ni) (Oak Ridge NationalLaboratory, Oak Ridge, USA) was dissolved in conc. HNO3 and diluted with DI water tomake a 7 ppm primary standard. This was used for isotope dilution analysis of Ni inseawater.2.2.8. Reference Materials — CASS-2 and NASS-3The Nearshore Seawater Reference Material (CASS-2) and the Open Ocean SeawaterReference Material (NASS-3), filtered and stored acidified, were purchased from NationalResearch Council of Canada (Ottawa, Canada), and used to assess the accuracy of themethod developed for trace metal analyses in seawater.152.3 OPERATING CONDITIONS AND OPTIMIZATION2.3.1 ICP-MSThe operating conditions of the ICP-MS were optimized daily for accurate results. Thesignal can be varied with RF power, flow rate, the distance between sampling cone andload coil, as well as the lens settings. The ICP-MS conditions were selected to ensure thehighest possible sensitivity as well as good resolution for the elements of interest. Table3.2.1 shows typical optimized parameters used for all quantitative work. A tuning solutioncontaining 10 ppb of Co, In, Pb, Bi and U was used to optimize the resolution and peakshape by using the ICP-MS in scanning mode. The resolution (R) and peak width at 5%peak height (AM) were selected so that a good separation of the Pb isotopes was achievedwhile maintaining the highest possible sensitivity. The short term stability of the ICP-MSwas determined using the tuning solution in peak jump mode, on a weekly basis. Thesampler and skimmer cones were replaced or cleaned daily.The instrument can be operated in either scanning or peak jump mode. In this study,the instrument was used in multichannel peak jump mode with a 20gs dwell time, one pointper peak and a 60 second acquisition time. Calibration was accomplished by measuring theblank and five standard solutions, normalized to the internal 115In standard (used to correctfor intensity drift during the measurements). The matrix of the blank and standards was2.0 N HNO3. The blank and standards (from low to high conc.) were analyzed before thesamples. Between each standard or sample 2.0 N HNO3 was used to wash the uptakesystem for two minutes. The wash time was extended to five minutes after measurement ofhighest standard (160 ppb). The standards and samples were blank subtracted (2.0 NHNO3) after normalizing to the 10 ppb In internal standard.16Table 2.3.1. ICP-MS Operation ConditionsParameter• Typical ValueRF power (kW) 1.35Argon gas flow rate (L/min.):Cooling gas 13.87Auxiliary gas 0.505Nebulizer gas 0.800Sampling position (mm above load coil) 13Sampler cone (nickel) orifice (mm) 1.0Skimmer cone (nickel) diameter (mm) 0.7Ion lens setting:Extraction lens(volts) -160 - -210Collector lens(volts) -2.0 - 0L1(volts) -2.5 - 0L2(volts) -60 - -30L3(volts) 0 - +5L4(volts) -50 - -42Pole Bias(volts) -3 - -4Operating pressure (mbar):Interface running pressure <0.1x10-4Expansion pressure 2.7 - 2.9Analyzer pressure 2.2x10-6 - 2.8x10-6172.3.2. QUANTITATIVE ANALYSIS ON ICP-MS2.3.2.1. Standard Calibration with Internal StandardUse of In as an internal standard to correct for instrument instability, signal drift andmatrix effect, requires low background levels of In in the samples. 115In was found to beinsignificant in seawater samples (< 0.4% of the added In signal). The calibration curves,obtained by the five blank subtracted standards with a 10 ppb 115In internal standard, for55Mn, 59Co, 1 ilcd, 65Cu , 66Zn, 208pb, 140Ce, 27A1 and 71 Ga were excellent. Theregression coefficients of calibration curves were within the range of 0.9998-1.000. Theprecision of ICP-MS measurements is 3%.2.3.2.2. Isotope Dilution TechniqueIsotope dilution was used for the Ni determination. Tests of the isotope ratio accuracyof the ICP-MS measurement were made by looking at 62Ni/60Ni ratios of a few standards.Standards were made from a mixture of the natural abundance standard solution (1.0 ppm)and the enriched isotope standard solution (7.53 ppm). Results are shown in table 2.3.2.2.62Ni/60Ni ratios measured in these mixed standards, over a range of concentrations similarto those measured in the seawater concentrates, agree well to the theoretical values,showing little mass fractionation.The Ni concentration in seawater can be calculated by the following equation (Date andGray, 1989):CNi (nmol/kg) = [Wsp / (Ws * Wisp)] * [(62p) sp - R * (60A) sp[ / [R * (60A) 5 - (62A)s]R is the ratio of 62Ni/60Ni in the final mixture, measured on ICP-MS. ( 62A)sp and (60A)spare isotope abundances (95.96% and 1.48%) of 62Ni and 60Ni in the enriched isotopestandard solution. (60A)s and (62A) s are isotope abundances (26.1% and 3.59%) of 60Ni18and 62Ni in the natural seawater. Wsp is the weight (ng) of the enriched 62Ni added to thesample. Ws is weight (kg) of seawater used. Msp is the atomic weight (61.813 g/mol) ofthe enriched isotope of 62Ni.Table 2.3.2.2 62Ni/60Ni Ratio Measurements in Standard SolutionsStandardsMeasured 62Ni/60Ni ratioRM RT RSD(%)Si:(8.6 ppb 60Ni / 3.1 ppb 62Ni)1.32 1.35 1.8S2:(80 ppb 60Ni / 18 ppb 62Ni)0.93 0.95 1.4S3:(331 ppb 60Ni / 68 ppb 62Ni)0.86 0.88 2.3RM:^Measured 62Ni/60Ni ratioRT:^Calculated 62Ni/6°Ni ratioRSD: Relative standard deviation of three replicate measurements2.3.3 GFAAS determinationThe GFAAS was optimized, using 1, 2 and 4 ppb Co standard solutions, for highsensitivity and a linear calibration curve by varying the sample ashing and atomizing timesand temperatures. The instrument was operated in hot (105 °C), multiple injection mode(three 32 ill injections) for dissolved Co measurements in order to optimize the detection oflow Co concentrations. The atomization temperature was 2350°C. Table 2.3.3 shows theconditions used for quantitative measurements of Co. Standard additions provide amonitor and a correction for signal suppression/enhancement caused by the sample matrixwhich would not be detected by external calibration. Standard addition was used for a fewsamples and no significant matrix effects were found, therefore dissolved Co wasdetermined only using standard calibration.19Table 2.3.3 GFAAS Operation ConditionsDescription ConditionsInstrument Mode AbsorbanceMeasurement Mode Peak AreaLamp Current 7 mASlit Width 0.2 nmWavelength 242.5 nmSample Introduction Sampler Automixin .5Furnace Parameters:Step No.Temp.(°C)Total time(sec.)Gas Flow(L/min.)1 105 5.0 3.02 150 47.0 3.03 800 11.0 3.04 2350 2.9 0.05 2350 1.0 3.06 40 12.8 3.02.4 STUDY AREASeawater samples were collected from the western North Pacific (Figure 2.4), with sixvertical profiles and a surface transect (25m) from Japan to Hawaii. Station HS-1 andHS-2 are closest to Japan, on the west side of the Kuroshio Current. This region ischaracterized by low temperature and low salinity waters, which are carried from the northby the Oyashio Current, entrained waters from the Sea of Japan, and some river inputs.Stations HS-3, HS-4 and HS-5 are in the Kuroshio Current, which has warmer, highersalinity surface waters, advected northwards from the equatorial region. Stations HS-7 toHS-11 are in the western half of the central north Pacific gyre. Station HS-6 has colder,lower salinity surface waters compared to the two adjacent stations. As discussed inmoredetail later, this is believed to be a cold-core ring, bringing water from the coastalregion across the Kuroshio and into the central gyre. Station HS-14, HS-15, and HS-1620Figure 2.4 Study Area, Water Samples Were Collected in A SurfaceTransect From Stations HS-1 to HS-16 with Vertical Profiles at StationsHS-1, HS-5, Hs-10, HS-14, HS-15 and HS-1621are in the North Equatorial Current, which brings water from the coast of North Americaacross the Pacific. Karin Ridge is beneath station HS-15, and thick ferro-manganese crustsare found at its top (depth 1500m). These enriched manganese crusts may provide anexternal source or sink for some redox or scavenged elements, which are incorporated ontothese mineral phases, and may affect the distributions of some elements at this area.2.5 SEAWATER SAMPLESSamples were collected by Kristin J. Orians, Brad McKelvey, Robert Mugo and HelenNicolidakis from departments of Chemistry and Oceanography of UBC, using methodsdeveloped by Bruland and co-workers (Bruland, et al., 1979), with Tefion0-lined, 30L,"Go-flo" bottles (General Oceanic), suspended on Kevlar® line. The "Go-fio" bottles,sealed prior to entering the ocean, were opened at 5-10 m depth, below the surface layerwhere potential contaminants are higher. Bottles were then lowered to the desired depths,and termination of sampling was triggered by a Teflon® messenger. Samples were filteredat sea, in a clean area built onboard ship, through 0.45gm polycarbonate membrane filtersusing 10 psi N2 overpressure, then acidified to pH 2 with 2 ml 6 N HC1/liter and stored inacid leached polyethylene bottles (HDPE). Acidified samples were transported to a shore-based clean laboratory for further processing and analysis.CHAPTER 3METHOD DEVELOPMENT3.1 THE OPTIMIZATION OF CHELEX-100 PRECONCENTRATIONThe optimum conditions for the preconcentration of cadmium, zinc, nickel, copper,lead, aluminum, gallium, manganese, cobalt and cerium from seawater were determined onthe basis of the recoveries of metal spikes to a seawater matrix. The spike, 200 ng of eachmetal, was added to clean seawater (from station "P", 145°W, 50°N, in the NortheastPacific), which had been stored acidified at pH 2. A series of seawater "blanks" with thesame amount of reagents but without added metals were run alongside the spiked samples.The following factors were investigated: the sample pH, pump flow rate, the volume ofChelex-100 resin, and eluent (HNO3) strength and volume. The recoveries of these metalswere quantified by the use of calibration curves obtained from mixed standards describedpreviously.3.1.1 pHThe pH was adjusted by the addition of NH4OH solution to the previously acidifiedsamples (pH 2) to obtain pH values between 3 and 8.5. These samples were thenpreconcentrated as in the general procedure.Quantitative recoveries (-100%) for all elements except Pb are generally achieved at pH6.5 to pH 8.0 (See figure 3.1.1.1). The recoveries of these elements are observed todecrease at lower or higher pH. Two pH conditions, 8.0±0.1 and 6.5±0.2, were used inthis study. Samples collected from stations 1, 5, 10 and surface stations 1-11 wereadjusted to pH 8.0±0.1 for dissolved Cd, Zn, Ni, Cu, Pb, Mn and Co measurements.23PHFigure 3.1.1.1 Recoveries at Various pHSamples from stations 14, 15, and 16 were adjusted to pH 6.5±0.2 for determination ofdissolved Cd, Zn, Ni, Cu, Pb, Al, Ga, Mn, Co and Ce. The lower pH used for thesesamples was modified to achieve 100% recoveries of Al and Ce without significant loss ofMn recovery.3.1.2 Flow RateThe effect of flow rate on metal recoveries from seawater was investigated for Mn, Ni,and Co on Chelex-100 resin. Results presented in Figure 3.1.1.2 show that slow pumpingis required for optimal recoveries. The recoveries dropped from 100% to 75% for Ni and78% for Co when the flow rate is increased to 2.5 ml/minter. In general, at flow ratesbetween 0.6 and 0.8 ml/min., 100% recovery is found for most metals. The pH studydescribed above was performed at this flow rate (0.6-0.8 ml/min.), verifying 100%recovery for all elements except Pb. The Pb recovery was lower (84 ± 2.1%), yetconsistent and easily corrected for.3.1.3 Volume of Chelex-100 ResinThe effect of resin volume on recovery was investigated using Mn, Ni, and Co. Theresults, given in table 3.1.3, show that the recoveries of these three metals are independentof resin volume in the range of 2 ml to 4 ml fresh resin (in sodium form). Smaller volumeswere not tested, due in part to the difficulty in handing small amounts of resin. In thisstudy, 2 ml resin was chosen for convenience, low cost, and to minimize column blankswhile avoiding the possibility of capacity breakthrough. The resin volume would need tobe reoptimized for samples at much higher concentrations. Breakthrough was not observedfor mixed standard spikes of up to 50 nM total metals, in addition to the naturally occurringions in the seawater matrix. The capacity of Chelex-100 is high (700 pmols Cu 2+/m1)relative to the trace levels of transition metals found in seawater.Figure 3.1.1.2 Recoveries Study at Various Flow Rates2.5 3.00.0^0.5^1.0^1.5^2.0Flow Rate (ml/min.)120100806040200120100gs'^80N 60a . )8^40rz42001201008060402000 a _000— Co0 0 0 0^01 t i t iMn..... ....0^.0000— Nii 1 t t iTable 3.1.3. Effect of Resin Volume on Metal RecoveriesVolume of Chelex-100 resin(sodium form)RecoveriesMn (%) Ni (%) Co (%)2 ml 101 100 1013 ml 102 103 1024 ml 99 104 1003.1.4 Eluent Concentration and VolumeThe effect on recovery from varying the HNO3 strength, and volume used for elutingthe columns, was tested for Mn, Ni and Co at concentrations of 1.6 N to 2.4 N, andvolumes of 2 ml to 12 ml. HNO3 concentration in the range of 2.0 to 2.4 N shows noeffect on recovery. Low recoveries of 80% for Ni and Co were obtained when using1.6 N HNO3. The concentration of 2.0 N HNO3 was chosen for ensuring quantitativerecovery and decreasing cost. 100% recovery is obtained using an eluent volume between5 to 12 ml of 2.0 N HNO3. The recoveries of these elements are observed to decreasewhen volumes less than 5 ml were used. In this study, 8 ml 2.0 N HNO3 was used toelute the columns for quantitative analysis. The recovery study (described later in section3.3.1) was performed using 8 ml 2.0 N HNO3 for elution, yielding 100% recovery for allelements with the exception of Pb (84%).3.1.5 Equilibration Test for Nickel AnalysisAccurate results for Ni measurements using isotope dilution depend on the equilibrationachieved between the two isotopes. The equilibration time was investigated from 0.5 to 24hours, and no significant changes were found after 2 hours. The spiked seawater sampleswere left overnight to ensure full equilibration between the two isotopes.273.2. GENERAL PROCEDUREAll sample processing was performed in a trace metal clean lab (high efficient particleair filtered - HEPA, positive pressure air supply) at UBC, in the Department ofOceanography. Prior to use, Teflon® columns (1 cm x 10cm) with 2 ml Chelex-100 resinwere cleaned with 60 ml 2.0 N nitric acid by gravity elution. After be rinsed with 4 ml DIwater, the columns were pH adjusted with NH4Ac to the optimal extraction pH. Filtered,acidified seawater samples (1000-2000g) were measured by weight. Next, 100 - 250g1 ofa 0.7173 ppm enriched 62Ni solution was added to the seawater samples and left overnightto equilibrate the natural 60Ni and the enriched 62Ni spike. The seawater samples werethen pumped through the Chelex-100 columns at a flow rate of 0.6-0.8 ml/min. afteradjusting the seawater to either pH 8.0±0.1 or 6.5±0.2. Next, the columns were rinsedwith 6X1 ml NH4Ac (0.15 M) and 3x1 ml DI water, at either pH 8.0±0.1 or pH 6.5±0.2,and eluted with 8x1 ml of 2.0 N HNO3 into 15 ml acid washed polyethylene bottles.Column eluents were then analyzed on the ICP-MS (or GFAAS) at this stage, or sealed andstored at room temperature for later analysis. Along with each batch of five samples, aspiked sample (200 ng Mn, Ni, Co, Cd, Cu, Zn, Pb, Ga, Al, and Ce) was processed tocheck recoveries. The CASS-2 and NASS-3 standard reference materials were treated inthe same manner at a pH of 8.0±0.1 and 6.5±0.2, respectively. The analytical blank wastreated in the same manner with the exception that no 62Ni was added and 50 ml of DIwater was used instead of seawater. The temperature, salinity and nutrient data wereprovided by the Institute of Ocean Sciences, Sidney, B.C., Canada.3.3. ANALYTICAL FIGURES OF MERIT3.3.1 RecoveryThe recovery studies of Cd, Zn, Ni, Cu, Pb, Al, Ga, Mn, Co, Ce were performed byspiking known amounts of these metals into seawater samples and then subjecting the28samples to the full preconcentration procedure. The recoveries were calculated from thedifference between spiked and unspiked seawater samples. Two replicates for each samplewere analyzed, as shown in table 3.3.1.1 and table 3.3.1.2. Quantitative recoveries ofMn, Ni, Co, Cd, Cu, Zn, Al, Ga, Ce (-100%) were obtained while there was a lowerrecovery for Pb (84%). Since the precision of Pb recovery is excellent (RSD=2.1%), therecovery was corrected for in the Pb data reported.Table 3.3.1.1 Recoveries from Seawater Matrix at pH 8.0±0.1SeawaterSample No.Recoveries (%)Mn Ni Co Pb Cd Zn Cu1 104 102 107 84.9 98.3 107 1022 101 98 108 85.8 101 103 1033 102 102 104 82.9 98.1 108 1004 105 106 102 81.9 97.9 104 945 108 107 106 82.2 101 99 99Ave. 104 104 105 84 99 104 100SD 3 3 3 2 2 4 4Table 3.3.1.2 Recoveries from Seawater Matrix at pH 6.5±0.2SeawaterSample No.Recoveries (%)Mn Ni Co Pb Cd Zn Cu Al Ga Ce1 99.3 104 97.7 85.9 104 107 96 106 98.5 1012 98.9 106 103 82.1 99.9 103 101 103 97.0 99.13 105 105 101 84.1 102 106 106 101 104 98Ave. 101 105 100 84 102 105 101 103 100 99SD 3 1 3 2 2 3 5 3 3 2293.3.2 PrecisionThe precision of this technique was evaluated by performing triplicate ICP-MSmeasurements of Mn, Ni, Pb, Cd, Zn and Cu (Co measured on GFAAS) on triplicateseawater samples, subdivided after collection and storage. The seawater sample used forthis purpose was collected from the central North Pacific (station HS-15 - see figure 2.4) at300m depth, and had been stored acidified at pH 2 for over 1 year. The results are shownin table 3.3.2. Better than 7% precision of this method was obtained for all elements.Table 3.3.2. Precision of Triplicate AnalysesName Measured Concentration (nmol/kg)Mn Ni Co Pb Cd Zn CuX1 0.285 ± 4.40 ± 0.033 ± 0.0347 ± 0.434 ± 1.05 ± 0.865 ±SDI 0.010 0.13 0.002 0.0008 0.009 0.07 0.035X2 0.292 ± 4.32 ± 0.029 ± 0.0330 ± 0.406 ± 1.07 ± 0.848 ±SD2 0.009 0.13 0.002 0.0009 0.009 0.07 0.035X3 0.287 ± 4.37 ± 0.029 ± 0.0323 ± 0.415 ± 1.01 ± 0.867 ±SD3 0.009 0.13 0.002 0.0013 0.011 0.07 0.035Ave. 0.288 4.36 0.030 0.0333 0.418 1.04 0.860SD 0.010 0.13 0.002 0.0013 0.014 0.07 0.036RSD 4.6% 3.0% 6.7% 3.9% 3.3% 6.9% 4.2%Xn : The mean of triplicate measurements on ICP-MSSDn: The standard deviation of triplicate measurements on ICP-MSAvg.: The average of triplicate samplesSD: Standard deviationRSD: Relative standard deviation3.3.3 AccuracyThe accuracy of this technique was assessed by the analyses of dissolved Mn, Ni, Co,Pb, Cd, Zn and Cu in trace metal seawater reference standards from the National ResearchCouncil of Canada. Two reference seawater samples, the Nearshore Seawater ReferenceMaterial (CASS-2) and the Open Seawater Reference Material (NASS-3), were analyzedfor these metals using the technique developed in this study. Triplicate measurements ofCASS-2 were determined only with GFAAS detection, due to accidental loss of most ofthis sample eluent. Triplicate measurements of NASS-3 were performed with ICP-MSdetection. Both are shown in table 3.3.3.1. Statistical analysis, comparing the dataobtained using this method with the certified values, using a Student-t test with a=0.01(Holscher, 1971), was conducted on NASS-3 (table 3.3.3.2). Results show no significantdifference between measured values and certified values within the 99% confidence limit.Table 3.3.3.1 The Determination of Dissolved Mn, Ni, Co, Pb, Cd, Znand Cu on Seawater Reference MaterialsNameConcentration (ng/L)Mn Ni Pb Cd Zn CuCASS-2found 2072 ± 313 ± 21.1 ± 16.5 ± 23.4 ± -- --110 14 1.1 1.9 0.5CASS -2certified 1990 ± 298 ± 25 ± 19 ± 19 ± 1978 + 675 +150 36 6 6 4 120 39NASS -3found 26.9 ± 248 ± 3.55 ± 36.5 ± 29.9 ± 181 ± 103 ±1.5 9 0.4 1.5 1.0 13 4.3NASS-3certified 22± 257± 4± 39± 29± 178± 109±7 27 1 6 4 25 11*: Co was measured on GFAAS for both reference materials--: The sample for dissolved Zn were believed to be contaminated. Dissolved Cu isnot available due to accidental lose of sample eluent.31Table 3.3.3.2 Statistical Analysis on NASS-3Metals Scertified(ng/L)Sfound(ng/L)SD df t value tcritical( a =0.01)Cd 4 1 0.88 37 1.02 < 2.71Zn 25 13 8.8 31  0.34 < 2.74Ni 27 9 7.2 31 1.25 < 2.71Cu 11 4.3 3.1 37 1.94 < 2.71Pb 6 1.5 1.5 25 1.67 < 2.79Mn 7 1.5 1.9 19 2.58 < 2.86Co*  1 0.4^_ 0.33 19 1.36 < 2.86Standard deviation from certified valuesStandard deviation from measured dataStandard deviation of difference between certified and measured dataDegrees of freedomMeasured on GFAASScerdfied:Sfound:Sa:df:*.3.3.4 Limits of DetectionTwo types of blanks were used to determine the limits of detection. Replicatemeasurements of the acid blank (2.0 N HNO3) were used to estimate the instrumentaldetection limit (IDL), as shown in table 3.3.4. Replicate measurements of the analyticalblank, in which 50 ml DI water was treated the same way as seawater samples, were usedto estimate the seawater detection limit (SDL) for the entire procedure. The data are shownin table 3.3.4. The SDL is given as three times the standard deviation of the six calculatedanalytical blank concentrations. The contribution to the analytical blank from the resin andcolumns was found to be low. Column blanks (CB), estimated from replicatemeasurements of the column eluents, shown in table 3.3.4., demonstrate that Chelex-100resin is well suited for trace metal preconcentration at low seawater levels.32Table 3.3.4. Detection Limits Obtained for the ICP-MS Analysis ofSeawater SamplesELEMENT IDLa(ppb)SDLb(nmol/kg)CBS(ng)Mn 0.05 0.009 0.16Ni 0.08 0.011 0.24Co* 0.05 0.008 0.54Cd 0.04 0.003 0.11Zn 0.05 0.006 0.16Cu 0.05 0.006 0.16Pb 0.03 0.002 0.08Al 0.10 0.037 0.36. Ga 0.04 0.005 0.11Ce 0.03 0.002 0.08a: Instrumental detection limit (average over 1 month, N=100)b: Seawater detection limit (N=6)c: Column blanks: the mean of six measurements on column blank*: Co measured on GFAAS3.4 CONCLUSIONA method for dissolved Cd, Zn, Ni, Cu, Pb, Al, Ga, Mn, Co and Ce measurements inseawater has been developed. 100% recovery is obtained for all elements except Pb, whichhas 84% recovery (RSD=2.1%). This technique was tested by analyzing standardreference materials from the National Research Council of Canada for Cd, Ni, Zn, Cu, Pb,Mn, and Co. Results are in good agreement with the certified values, within 99%confidence limit. The precision of this method is found to be better than 7% for theseelements. The seawater detection limits are in the low picomolar range (2-40 pmol/kg) after33a 125-fold preconcentration. This accurate multielemental technique will be useful forproviding information on the distribution of a suite of trace metals in various oceanographicregimes. These distributions help to elucidate the biogeochemical controls on seawatercomposition and are useful as physical tracers of water mass movement in the oceans. Thefollowing chapter presents data from the western North Pacific, obtained using thismethod, and demonstrates the use of multiple trace metals as physical tracers of water massmovement in this area. In addition, the processes which control the distributions of theseelements will be further elucidated.CHAPTER 4OCEANOGRAPHIC RESULTSDissolved Cd, Zn, Ni, Cu, Pb, Al, Ga, Mn, Co and Ce in seawater samples from thewestern North Pacific, with six depth profiles and fourteen surface samples, have beenmeasured. The results are listed in tables 4.1.1, 4.1.2, and 4.1.3. The data are presentedin the following sections.4.1 VERTICAL PROFILES4.1.1 Nutrient-Type ElementsA nutrient-type profile exhibits a surface depletion and increasing concentrations withdepth, reaching a maximum coherent with the maximum observed for nutrients, such asnitrate, phosphate and silicic acid. These observed correlations can be explained by theuptake of nutrient-type metals in the surface water from the dissolved phase to biogenicparticulate matter fixed by phytoplankton. This biogenic carrier phase sinks and isregenerated via oxidation processes and chemical dissolution, releasing the nutrients andassociated trace metals. Some nutrient-type elements, which can be almost totallydepleted in the surface water, are biolimiting constituents (like nitrate, phosphate, silicicacid, cadmium and zinc). Alternatively, some nutrient-type elements, which are partiallydepleted in the surface water, are biointermediate constituents (such as nickel andcopper). There are three variants of nutrient-type profiles: (1) A distribution caused byshallow regeneration leading to a mid-depth maximum, similar to that found for the labilesoft tissue nutrients, phosphate and nitrate. Cadmium is an example of a nutrient-type35Table 4.1.1 Dissolved Ni, Cd, Zn, Cu, Pb, Mn and Co in theWestern North PacificStation Depth (m)Trace Metal ConcentrationsNi(nmol/kg) Cd(nmol/kg) Zn(nmol/kg) Cu(nmol/kg) Pb(pmol/kg) Mn(nmol/kg) Co(pmol/kg)HS 1 25 5.60 0.186 1.48 1.08 76 1.95 3450 5.12 0.219 1.97 1.17 95 1.93 2975 6.37 0.280 1.58 1.15 65 1.39 34100 6.13 0.299 1.52 1.25 66 1.10 27250 6.86 0.440 2.99 1.29 53 1.63 47499 9.35 0.844 7.02 1.48 40 1.32 30770 10.68 0.970 8.86 1.56 40 1.25 261036 11.53 0.966 9.24 1.48 39 0.72 161489 11.76 0.984 9.83 2.37 47 0.63 112326 11.09 0.861 9.14 2.52 23 0.37 122759 10.88 0.837 8.77 2.87 23 0.34 94689 10.87 0.823 8.14 3.86 20 0.34 6HS 2 25 7.07 0.333 1.84 1.20 72 1.55 NAHS 3 25 5.36 0.058 0.56 0.77 79 1.17 NAHS 4 25 5.18 0.063 1.06 0.84 87 1.27 NAHS 5 25 4.53 0.024 0.37 0.56 78 0.96 2250 4.62 0.060 0.37 0.74 61 1.09 2375 4.40 0.027 0.45 0.53 83 0.97 16100 4.67 0.028 0.34 0.65 76 0.96 16262 4.71 0.047 0.66 0.59 77 0.87 19410 6.68 0.208 1.10 0.85 59 0.57 33692 8.28 0.637 3.91 1.08 47 0.55 381013 11.02 0.865 6.44 1.48 39 1.01 401502 11.62 0.960 7.80 1.43 39 0.60 242526 11.94 0.896 8.08 1.67 28 0.35 243051 11.78 0.918 8.36 2.09 20 0.33 18HS 6 25 7.31 0.130 2.28 1.61 115 1.95 NAHS 7 25 5.43 0.094 0.70 0.87 77 1.37 NAHS 8 25 5.07 0.064 0.89 0.84 68 1.35 NAHS 9 25 5.23 0.072 0.74 0.92 62 1.38 NAHS 10 25 3.73 0.01 0.48 0.55 62 1.03 NA50 3.80 0.018 0.44 0.55 63 0.97 NA75 3.61 0.014 0.50 0.52 53 0.96 NA100 3.68 0.021 0.56 0.57 67 0.95 NA238 3.97 0.076 0.82 0.59 71 0.38 NA388 4.78 0.216 1.16 0.75 59 0.25 NA511 5.91 0.397 5.43 1.01 59 0.27 NA1009 10.30 0.969 7.99 1.60 43 0.60 NA1492 11.28 1.002 8.13 1.72 24 0.35 NA2615 11.28 0.928 9.15 2.55 31 0.24 NA3116 11.51 0.879 8.64 2.53 12 0.32 NA4599 10.31 0.788 8.12 3.12 20 0.19 NAHS 11 25 4.81 0.065 0.56 0.89 42 1.41 NANA = not analyzed36Table 4.1.2 Dissolved Ni, Cd, Zn, Cu, Pb, Co and Pb in theCentral North PacificStation Depth (m)Trace Metal ConcentrationsNi(nmol/kg) Cd(nmol/kg) Zn(nmol/kg) Cu(nmol/kg) Pb(pmol/kg) Mn(nmol/kg) Co(pmol/kg)HS 14 25 3.30 0.086 0.377 0.424 31 0.971 2550 2.88 0.103 1.044 0.471 39 0.798 16100 3.00 0.113 0.940 0.408 50 1.099 12300 4.89 0.478 1.131 0.632 44 0.368 30500 7.27 0.896 2.977 0.630 28 0.414 28716 8.59 1.008 4.043 0.736 23 0.409 19916 9.05 1.049 5.276 0.910 15 0.297 271115 9.59 1.066 5.7% 0.972 16 0.337 271432 10.26 1.081 6.975 1.345 13 0.308 242352 10.74 1.017 7.548 1.559 9 0.290 222646 10.38 1.017 8.164 2.134 7 0.097 224058 9.87 0.945 8.455 2.869 8 0.146 13HS 15 25 3.824 0.139 0.558 0.522 48 0.843 3950 3.812 0.133 1.868 0.600 35 1.009 23100 3.445 0.145 1.093 0.599 42 0.412 18250 5.552 0.598 1.287 0.784 52 0.189 46400 5.011 0.385 2.651 0.871 (94) 0.455 40600 8.343 0.852 3.485 1.047 45 0.532 43800 9.839 1.081 5.610 1.216 36 0.368 341000 10.311 1.102 6.395 1.488 22 0.418 281200 11.286 1.114 7.124 1.588 18 0.412 341500 11.496 1.080 8.268 1.745 21 0.297 21HS 16 25 3.29 0.102 0.575 0.517 46 1.036 2650 3.22 0.101 0.560 0.580 40 1.036 12100 3.32 0.098 0.538 0.534 47 1.022 6300 4.88 0.451 1.065 0.697 55 0.239 33500 7.96 0.968 3.519 1.308 31 0.391 28706 9.27 1.019 5.095 1.339 22 0.372 27906 9.63 1.024 6.185 1.270 20 0.313 181204 10.76 1.063 7.451 1.744 18 0.409 261504 11.01 1.058 8.805 1.730 13 0.260 222032 11.36 1.030 9.205 2.492 10 0.263 202532 10.91 0.990 9.734 2.412 9.7 0.213 143034 10.98 0.982 9.671 2.684 12 0.188 143534 11.17 0.967 9.522 2.815 12 0.199 144041 10.79 0.956 9.853 3.509 20 0.165 9( ): indicates sample believed to be contaminated37Table 4.13 Dissolved Al, Ga and Ce in the Central North PacificStationTrace metal concentrationsDepth (m)/k ) Ga (pmol/kg) Ce (pmol/kg)HS 14 25 3.43 10.9 2.550 5.33 11.3 2.4100 9.38 15.2 2.4300 1.41 10.5 1.2500 (14.70) 10.5 1.6716 0.97 8.4 1.8916 1.52 8.3 2.71115 1.07 11.3 1.71432 0.95 12.4 1.82352 1.51 19.4 2.62646 (20.09) 28.2 2.14058 2.39 33.3 2.5HS 15 25 6.27 15 2.650 5.14 17 2.5100 4.99 13 2.4250 6.99 14 0.7400 4.04 15 1.2600 2.39 13 1.7800 1.73 12 2.11000 1.65 11 1.71200 1.69 12 1.51500 1.88 14 2.0HS 16 25 5.40 12 2.450 4.55 11 2.6100 6.09 14 2.1300 2.06 10 1.4500 1.53 10 1.4706 1.95 7 1.6906 1.73 7 1.91204 1.57 11 1.61504 1.45 14 1.52032 1.46 18 1.72532 1.49 16 2.33034 1.69 25 2.23534 2.56 31 2.04041 2.92 31 2.7( ): indicates sample believed to be contaminatedtrace element with a soft-tissue distribution. (2) A distribution caused by deepregeneration, leading to deep maximum similar to that found for the "hard-part" nutrient,Si. "Hard-part " nutrients are those which have been used by organisms to construct theirskeletal structures (such as calcite, CaCO3) or opal (SiO2) shells. Zinc is a nutrient-typetrace element with deep regeneration. (3) A distribution resulting from a combination ofshallow and deep regeneration. Nickel is an example of this kind of nutrient-type traceelement.4.1.1.1 Dissolved Cd - Nutrient-Type Distribution with Shallow RegenerationThe depth profiles of dissolved Cd in three western and three central North Pacificstations are plotted in figure 4.1.1.1 and figure 4.1.1.2. In the central gyre station (HS-10,27°46.5' N, 174°59.4' E), Cd shows the expected soft-tissue, phosphate or nitrate-typedistribution. A surface depletion of 0.010 nmol/kg and increasing concentrations withdepth, reaching a maximum of 1.00 nmol/kg at 1000-1500m, are observed. This agreeswell with previous results in the central North Pacific (Bruland, 1980). Approximately3500km to the northwest, at HS-5 in the Kuroshio Current, the dissolved Cd verticaldistribution and concentrations are remarkably similar to that in the central gyre. Incontrast, the coastal station (HS-1) has a higher surface Cd value of 0.19 nmol/kg with arapid increase to a broader maximum (0.97-0.98 nmol/kg) at a shallower depth (600-1500m). At depths below 1000m the profiles are indistinguishable. Three depth profilesfrom the central North Pacific also show the expected nutrient-type distribution withsurface values of 0.086-0.10 nmol/kg and maximum concentrations at 1000-1500m(figure 4.1.1.2).The correlation between Cd and the labile nutrients, phosphate and nitrate, wasinitially observed by Boyle, et al., (1976) and confirmed by Bruland (1980) in the centralFigure 4.1.1.1 Dissolved Cd Distribution at Stations HS-1, HS-5, and HS-1040Figure 4.1.1.2 Dissolved Cd Distribution at Stations HS-14, HS-15, and HS-16North Pacific. This correlation between Cd and labile nutrients is also found to hold inthe western North Pacific. Plots of dissolved Cd versus phosphate and nitrate arepresented in figure 4.1.1.3. The data from 400m and 600m at station HS-14 were notused in the plots, due to errors in the depth reading during sampling. The phosphate andnitrate data were provided by IOS (Institute of Ocean Science in Sidney, B. C.). A linearrelationship is observed for dissolved Cd vs. phosphate, with a slope of 0.325 Cd/P (nmolkg-1/µM) , a Y-intercept of 0.013 nmol/kg Cd, and a regression coefficient, R 2, of 0.986.Comparing this slope with that found in east Pacific Ocean, 0.33-0.37 (Bruland, 1980;Knauer and Martin, 1980), and in Antarctic Ocean, 0.31 (Westerlund and Ohman, 1991),it is clear that the Cd/P ratio in western North Pacific is not significantly different fromprevious observations. In the Atlantic ocean, however, the reported slopes show a widerrange, 0.14-0.30 (Burton, et al., 1982; Bruland and Franks, 1983; Yeats and Campbell,1983; Danielsson, et al., 1985; Yeats, 1988), with lower ratios found for the eastern andArctic parts of the Atlantic ocean. The difference in Cd/P ratio in Atlantic ocean may becaused by a different deep water source to the eastern Atlantic. Both the Pacific (Reidand Lynn, 1971) and the western Atlantic (Whitehead and Worthington, 1982) aresupplied with deep water from the Antarctic, while the major source for the easternAtlantic deep water is from the Arctic (Rudels, 1987). The Cd/NO3 relationship for alldata in the western and central North Pacific also show a linear correlation, with a slopeof 0.0226 Cd/NO3 (nmol kg -1/µM), a intercept of 0.0413 nmol/kg Cd and R 2 of 0.982.This agrees with Bruland's data (1980) from the central North PacificaCd]=0.0228[NO3]1-0.0437). The strong correlation of Cd to both PO4 and NO3 verifiesthat the dissolved Cd distribution is largely driven by vertical transport via shallowregeneration, and that Cd can be used as a proxy for PO4 throughout the western NorthPacific.4210 20 30 40 500.5^1.0^1.5^2.0^2.5Phosphate (pM)1.2 ^1.00.80.60.40.20.00.01.21.00.80.60.40.20.003.53.0Nitrate (pM)Figure 4.1.1.3 Dissolved Cd vs. Nutrients434.1.1.2 Dissolved Zn - Nutrient-Type with Deep RegenerationThe vertical profiles of dissolved Zn from three western and three central NorthPacific stations are presented in figure 4.1.1.4 and figure 4.1.1.5. The vertical Zndistributions show the expected "hard-part", Si-type distribution in the central gyre station(HS-10), with a depleted surface value of 0.48 nmol/kg and increasing concentrationswith depth to a deep maximum of 9.33 nmol/kg at depths between 1500-3500m. Thisagrees well with previous results from the central North Pacific (Bruland, 1980).Dissolved Zn in the Kuroshio Current station, HS-5, exhibits a similar distribution andconcentrations to that observed in central gyre. The vertical profile at the coastal station(HS-1) shows the highest dissolved Zn surface concentration of 1.48 nmol/kg, with arapid increase to a broader and shallower maximum of 9.83 nmol/kg between1000-2000m. At depths below 1500m, the profiles are indistinguishable. The depthprofiles at stations HS-14, HS-15 and HS-16 show similar distributions with a surfacedepletion and deep maximum (fig. 4.1.1.5).The correlation between Zn and the "hard-part" nutrient element, Si, is estimated byplotting the concentration of dissolved Zn vs. Si for all samples (fig. 4.1.1.6). A linearrelationship between Zn and Si is observed, with a slope of 0.0544 Zn/Si (nmol kg -1/µM),a intercept of 0.629 nmol/kg Zn, and an R 2 of 0.981. The Zn/Si ratio found in this studyshows no significant difference to that observed in the eastern central Pacific, 0.0535(Bruland, 1980), but is significantly different to that found in the Atlantic, 0.17 (Yeatsand Campbell, 1983; Danielsson, et al., 1985). This high Zn/Si ratio in the Atlantic hasbeen explained by the mixing of Antarctic bottom water with North Atlantic deep water,which has a much lower dissolved Si concentration (Westerlund and Ohman, 1991). Thelinear relationship between Zn and Si in this study supports the premise that Zn isremoved from the surface waters by processes related to the construction of the hard-parts(e. g. biogenic opal) of phytoplankton, and transported to the deep sea where it isregenerated with Si.441^I^I 1 14^6^8^10 12 0^2^4^6^8^10 1250010001500200025003000350040004500500055000 ^01i ' I0 a '^I^'^I^10_ 0,,-^00\- HS-5. 1yoI0PRE- HS-101^12^4^6^8^10 12 0^20\o\o' •r....43^I^'^I^'^I^'^I- ion 0'. \..^/0?10- HS-1Dissolved Zn Concentration (nmol/kg)Figure 4.1.1.4 Dissolved Zn Distribution at Stations HS-1, HS-5, and HS-10. 1 i . 1A.^I^'^1^i^I^I"-------o...., oo NoOM\001010DIY I ' 1^1a- "------42 0_HS-14.. 1 1 elr9 ' I ' I ' I0s"--`0,NO—--------O■—0\0— 0HS-15 HS-16.0^2^4^6^8^10 12 0^2^4^6^8^10 12 0^2^4^6^8^10 1250010001500-4. 2000,a'-a.' 2500-5 3000a.0Q 350040004500500055000Dissolved Zn Concentration (nmol/kg)Figure 4.1.1.5 Dissolved Zn Distribution at Stations HS-14, HS-15, and HS-16Figure 4.1.1.6 Dissolved Zn vs. Dissolved Si474.1.1.3 Dissolved Ni — Nutrient-Type With a Combination of Shallow and DeepRegeneration CyclesThe vertical profiles of dissolved Ni in the western and central North Pacific stationsare plotted in figures 4.1.1.7 and 4.1.1.8. The Ni concentrations found in this study aresimilar to other findings in different regions (Bruland, 1980; Bruland, et al., 1993; Yeatsand Campbell, 1983; Danielsson, et al., 1985). The Ni profile in the central gyre showsthe expected biointermediate nutrient-type distribution, with a partial surface depletion(3.73 nmol/kg) and increasing concentrations in the upper 1500 meters to a broadmaximum of 11.5 nmol/kg between 1500 and 3500m. Like Cd and Zn, Ni has similarconcentrations and vertical distributions in the Kuroshio Current station (HS-5) and thecentral gyre station. In the coastal station (HS-1) Ni shows a higher surface value (5.6nmol/kg) and a shallowing of the mid-depth maximum (11.8 nmol/kg) to 1000-2500m.The depth profiles of dissolved Ni at station 14, 15 and 16 show distributions similar tothose found in three western North Pacific stations.Dissolved Ni versus phosphate, nitrate and silicic acid are shown in figure 4.1.1.9 andfigure 4.1.1.10. The relationships between Ni and the soft-part nutrients, phosphate andnitrate, show regression coefficients, R 2, of 0.941 and 0.949 in the western and centralNorth Pacific. A better correlation between Ni and Si is observed, with an R 2 of 0.967.This may indicate that Ni distribution is largely controlled by the "hard-part" formationand regeneration. Plots of Ni vs. PO4 and Ni vs. Si, however, both show curvaturethroughout the water column. The correlation of Ni versus phosphate and nitrate for onlythe upper 800m show no improvements. This is different to Bruland's finding (Bruland,1980). He found better correlations with PO4 in the upper 800 meters and bettercorrelations with Si in the deep waters, in the eastern North Pacific. In general, all sixprofiles of dissolved Ni show a biointermediate nutrient-type distribution with acombination of deep regeneration and shallow regeneration cycles throughout the watercolumn.48A■C)Figure 4.1.1.7 Dissolved Ni Distribution at Stations HS-1, HS-5 and HS-10Figure 4.1.1.8 Dissolved Ni Distribution at Stations HS-14, HS-15, and HS-16Ulii•■•.,Figure 4.1.1.9 Dissolved Ni vs. Nutrients, Full Depth ProfilesFigure 4.1.1.10 Dissolved Ni vs. Nutrients , Upper 800m524.1.2 Nutrient-Type with Scavenging Element — CuThe vertical distributions of dissolved Cu at three western and three central NorthPacific stations are plotted in figure 4.1.2.1 and figure 4.1.2.2. These depth profiles showclear evidence of surface depletion, intermediate and deep water scavenging, and bottomsources from the sediments. In the central gyre station (HS-10), the dissolved Cu verticalprofile shows a surface depletion to 0.55 nmol/kg and gradually increasing concentrationswith depth to a maximum of 3.12 nmol/kg in the bottom water. This is in agreement withprevious data from this region (Bruland, 1980). Like the other nutrient-type elements, Cushows little difference between the Kuroshio Current profile (HS-5) and the central gyre,but is higher in the coastal station (HS-1). The vertical profiles of dissolved Cu atstations HS-14, HS-15 and HS-16 are similar to Cu profile at station HS-10. The lack ofan intermediate maximum of dissolved Cu, in comparison with other nutrient-typeelements, such as Cd, Zn, and Ni, is due to dissolved Cu being scavenged onto particlesin the intermediate and deep waters (Bruland, 1980; Bruland and Franks, 1983).A simple vertical advection-diffusion model can be used to estimate the deep-waterscavenging times for non-conservative elements with bottom sources. The model can beexpressed as (Boyle, et al., 1977):0 = K (a2 [Cu]/a2z) - W (a[Cu]/az) + JIt can be shown that, by rearranging the solution to the above differential equation, theadvection-diffusion model may be expressed by:[Cu] = a + (J/W) z + pewhere K is the vertical eddy diffusivity, and W is the vertical advection velocity (positivedownwards), [Cu] is the concentration of Cu, z is the depth, J is a term that accounts forscavenging removal, a and 13 are constants, and 0 is the potential temperature.531 2 3 40 1 2^3^4 0 1 2 3. 1OM5u10..,00NoI.0100\01 .5001000150020002500300035004000450050005500  0 ^0NOM_HS-10N1 i l'ab '^I0.00i_HS-1PM- HS-5Dissolved Cu Concentration (nmol/kg)Figure 4.1.2.1 Dissolved Cu Distribution at Stations HS-1, HS-5, and HS-10r 1IIII't ' I ' Io D WI ! ! I^II^I Io ...........004:s c,101■■- HS-16r^1-HS-14i r rI^I^I11.111i^Ioo_10._HS-150.005001000150020002500300035004000450050005500Dissolved Cu Concentration (nmol/kg)0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0Figure 4.1.2.2 Dissolved Cu Distribution at Stations HS-14, HS-15, and HS-16Alternately, potential temperature can be replaced by a conservative tracer such assalinity in a linear temperature-salinity (T-S) region. By using multiple linear regression,a, J/W and f can be determined, and thus, J, the scavenging term may also bedetermined, assuming an upwelling rate of 4 m/year (Boyle, et al., 1977).At stations HS-1, HS-5, and HS-10, the T-S plots at depths below 1000 m show alinear relationship, allowing application of this vertical, one-dimensional, advection-diffusion model. Dissolved Cu data from below 1000 m, plotted with respect to salinityshow a concave contour, indicative of scavenging removal (figure 4.1.2.3). Thescavenging rate for dissolved Cu are estimated to be 0.000385, and 0.000426nmol/kg/year, equivalent to a "half-life" of 1036, and 1066 years at stations HS-10 andHS-1, respectively. This advection-diffusion model was not applied to data from stationHS-5 due to lack of Cu values below 3500m, even though a linear T-S plot is found inthis station. Stations HS-14, HS-15 and HS-16 have T-S plots that are not linear,therefore preclude the use of this model. The scavenging times ("half-life") found inwestern North Pacific are similar to that seen in other regions, 830-1100 years (Boyle, etal., 1977; Bruland, 1980; Bruland and Franks, 1983).56Figure 4.1.2.3 Dissolved Cu vs. Salinity and Temperature vs. Salinity Poltsat Stations HS-1 and HS-10574.1.3 Scavenged ElementsA scavenged profile generally shows a surface and/or bottom maximum, due toexternal sources, and rapidly decreasing concentrations with distance from these sourcesdue to scavenging removal. Scavenging via adsorption onto particulate matter, whichthen sinks, is the dominant removal process. Pb is an adsorptive scavenged-type element,with a surface source, primarily from the atmospheric, due to the use of leaded gasoline.Scavenged profiles can also be produced by oxidative scavenging. A striking example ofremoval via oxidative scavenging is provided by Mn, which is soluble in the reducedform, Mn(II), but is oxidized in the water column to produce insoluble Mn(III, IV)oxyhydroxides. When Mn(III, IV) oxyhydroxides form, several trace metals (Co forexample) are scavenged onto these mineral phases (Knauer, et al., 1982; Kremling andHydes, 1988).4.1.3.1 Scavenged Elements with Only External SourcesScavenged Element With Surface Source — Dissolved Pb Vertical DistributionsDissolved Pb in three western and three central North Pacific stations are plotted infigure 4.1.3.1 and figure 4.1.3.2. The vertical profiles of dissolved Pb support thecontention that external sources and scavenging removal processes are the majorprocesses responsible for Pb distribution in the seawater. The depth profile in the centralgyre shows trends similar to those in previous studies in the Atlantic (Boyle, et al., 1986)and the Pacific (Flegal and Patterson, 1983). At HS-10, the central gyre station,dissolved Pb increases with depth from the surface (62 pmol/kg) to a sub-surfacemaximum (71 pmol/kg) around 250m. Below this depth, dissolved Pb decreases withdepth down to a deep water value of 12 pmol/kg, due to scavenging throughout the watercolumn. In contrast, the coastal station (HS-1) has higher values in the upper waters with58ulV:)Figure 4.1.3.1 Dissolved Pb Distribution at Stations HS-1, HS-5, and HS-10a surface or near surface maximum (75-95 pmol/kg) above 75m depth, which indicatesintensified inputs near Japan. The Kuroshio Current station, HS-5, with a surfacemaximum of —80 pmol/kg in the upper 100m, is intermediate between the two extremes(HS-1 and HS-10). At depth below 500m the profiles are indistinguishable. The sub-surface maximum of dissolved Pb in the central gyre could be due to a number of factors.Lead could be released from falling particles, accumulating in the water column until theconcentration builds up to where the deep scavenging flux equals the decomposition flux.Physical transport processes may supply Pb to this depth by isopycnal ventilation fromthe west. Alternatively, the decline in leaded gasoline consumption since 1974 couldsimply induce this maximum. The lack of a pronounced sub-surface maximum closer tothe Asian continent supports the second theory.The vertical profiles of dissolved Pb in the central North Pacific stations (HS-14,HS-15 and HS-16) show distributions similar to that in the central gyre station HS-10.All show a sub-surface maximum between 100 and 400m and low values in the deepwaters. These profiles primarily reflect the location and intensity of external sources,with scavenging removal throughout the water column.Figure 4.1.3.2 Dissolved Pb Distribution at Stations HS-14, HS-15, and HS-16Scavenged Element with Both Surface and Bottom Sources - Dissolved Al VerticalDistributionsThree vertical profiles of dissolved Al in the central North Pacific are shown in figure4.1.3.3. The three depth profiles show major trends similar to those of previous results(Orians and Bruland, 1986) with the exception that maximum Al concentrations arefound at 100-250m rather than at the surface, and higher intermediate values areobserved. Two samples believed to be contaminated at station HS-14, at 500 and 2646mwere ignored. At station HS-14, dissolved Al shows a sub-surface maximum at 100m(9.4 nmol/kg, which could be due to contamination), decreasing concentrations withdepth reaching the lowest value (0.95 nmol/kg) in the intermediate waters (-1500m), andincreasing concentrations towards the bottom, reaching 2.4 nmol/kg. Similardistributions were observed at stations HS-15 and HS-16.The high surface concentrations of dissolved Al, relative to intermediate or deepwaters, are caused by atmospheric Al inputs, since Al is highly abundant in the earth crust(8.23% by weight, Taylor, 1964), similar to findings in the eastern North Pacific and theAtlantic (Orians and Bruland, 1985; Hydes, 1979; Measures, et al., 1984). The sub-surface maximum observed in this study could be caused by advection via the NorthEquatorial current, bringing low Al content waters from the coast of North America tothis area in the surface. The Al data in surface waters from Orians and Bruland (1986)show decreasing trends from 5.1 nmol/kg in the eastern part of central gyre (20 000' N,160000' W) to 0.6 nmol/kg in the eastern coastal station (36°26' N, 122°30' W). Eolianinput from Asian dust sources would predict that dissolved Al concentrations increasetowards the west. The sub-surface maximum would be expected if the North Equatorialcurrent brings the low Al coastal water of North America to this area on top of centralNorth Pacific water. This sub-surface maximum should be further investigated in futurestudies. The decreasing concentrations in the intermediate waters indicate scavenging inthe water column. The increasing concentrations in the bottom water indicates a62ONLaFigure 4.13.3 Dissolved Al Distribution at Stations HS-14, HS-15, and HS-16bottom source of Al, from pore water diffusion out of the sediments or remineralizationon the sediment surface. The mid-depth values are three to four fold higher thanobserved in previous studies in the North Pacific (Orians and Bruland, 1986). Thesemid-depth values are believed to reflect real elevated Al levels in this area, not analyticalor sampling errors. First, there is a good agreement between bottom Al values (2.4 and2.9 nmol/kg at stations HS-14, HS-15) in this study and previous results (2.4 nmol/kg) ineastern North Pacific gyre (Orians and Bruland, 1986). Second, the precision of theanalytical technique for Al measurements is estimated to be less than 20%, even in theworst case recovery study. Even 20% error in Al measurements still can not make a threeto four fold difference in intermediate Al values. A possible source for this highermid-depth Al could due to the inputs from the Karin Ridge.Vertical Distributions of Dissolved Ga in the Central North PacificThree depth profiles of dissolved Ga in the central North Pacific are shown in figure4.1.3.4. These preliminary data generally agree with previous results (Orians andBruland, 1988; Shiller, 1988). At station HS-14, the vertical profile of dissolved Gashows a low surface value (10.9 pmol/kg), a sub-surface maximum at 100m depth (15.2pmol/kg), a mid-depth minimum at 500-1000m (8.4-8.3 pmol/kg), and increasingconcentrations with depth and reaching a maximum of 33.3 pmol/kg in the deep water.The vertical profiles of dissolved Ga at HS-15 and HS-16 show similar trends.The source of dissolved gallium to the surface has been proposed to be fromdissolution of atmospheric dust (Orians and Bruland, 1988; Shiller, 1988). Eolian inputfrom the Asian dust sources (Duce, et al., 1991) would predict that dissolved galliumconcentrations in the central North Pacific should increase towards the west. The highersurface values (11-12 pM) found in this study compared to those found further east (<10pM, Orians and Bruland, 1988), support this theory. The sub-surface maximum and64C71VIFigure 4.1.3.4 Dissolved Ga Distribution at Stations HS-14, HS-15, and HS-16the mid-depth minimum may be caused by either horizontal advection and/or a verticalprocess involving exchange with sinking particles. Gallium taken up at the surface maybe released in this region to give the subsurface maximum, which might be expected ifbiological uptake into soft tissues is occurring in surface waters. The sub-surfacemaximum observed in this area, which is closer to Asian dust sources, is at a depthshallower than that noted in previous results from further east (Orians and Bruland,1988). This supports that horizontal advection is major process causing this sub-surfacemaximum. After release, gallium may then be scavenged by sinking particles resulting inthe minimum observed at 1000m, as was suggested previously (Orians and Bruland,1988). The increasing concentration towards bottom indicate a bottom source which maycome from diffusion out of the sediments and/or sediment surface remineralization.4.1.3.2 Oxidative Scavenging Type ElementsVertical Profiles of Dissolved MnThree depth profiles of dissolved Mn and 02 (provided by IOS) in the western NorthPacific are plotted in figure 4.1.3.5. Dissolved Mn data from the central gyre (station HS-10) agree well with previous results in this area (Landing and Bruland, 1987). Amaximum value of 1.0 nmol/kg is found in the surface, indicating the fluvial (riverine),eolian (atmospheric), and photoreductive and /or biological reductive sources (Sunda andHuntsman, 1988). The subsurface minimum of dissolved Mn (0.25 nmol/kg) around 300-400m, where high oxygen concentrations were found (> 200 pm) indicates the oxidativescavenging of dissolved Mn. The secondary maximum of dissolved Mn (0.60 nmol/kg)in the oxygen minimum zone (1000-1500 m, [02]<60 p.m), indicates the reduction ofMn(III, IV) oxyhydroxides to soluble Mn(II) and a horizontal advection from the6640003000A0100020003000400050000100020005000010002000300040005000HS-100 ^HS-11■■HS-5I^1^1^10-P00—^0\00,\0‘.1^1HS-100 50 100 150 200 250 300 0.0 0.5 1.0 1.5 2.0 2.5Dissolved Oxygen (pM)^Dissolved Mn (nmol(kg)Figure 4.1.3.5 Dissolved Mn and Oxygen at Stations HS-1, HS-5, and HS-1067continental shelf. Below the 02 minimum zone, dissolved Mn decreases with depth andreaches a value of 0.20 nmol/kg around 4500 m, due to oxidative scavenging removal. Instation HS-5, the Kuroshio Current station, a similar distribution of dissolved Mn wasfound to that in the central gyre station (HS-10), except that a higher secondary maximum(1.0 nmol/kg) was observed in the oxygen minimum zone. The coastal station (HS-1) hasa higher surface Mn value of 2.0 nmol/kg, and the secondary maximum of dissolved Mnis intensified and shifted upwards to 250-750m depth, coincident with a shallowing of the02 minimum zone.The dissolved Mn maxima in the 02 minimum zone show a decreasing trend awayfrom the coastal regions. Dissolved Mn values in this zone decrease from 1.6 nmol/kg atHS-1 to 1.0 nmol/kg at HS-5 to 0.60 nmol/kg at HS-10. This supports the contention thatthe Mn maximum in the 02 minimum zone is primarily due to horizontal advection fromreducing shelf sediments, rather than from in situ reduction of particulate manganeseoxides in the open ocean, as observed in other regions (Landing and Bruland, 1987;Saager, et al., 1989; Gordeyev and Atnashev, 1990). The oxygen concentration at theminimum is actually higher in the coastal region than in the central gyre (5011M at HS-1vs. 40 1.1M at HS-10). If Mn maximum is due to in situ reduction of particulatemanganese oxides and thus cannot explain the coastal Mn enrichment. Without data onthe amount of particulate Mn, which is likely to be higher near the continent, it is difficultto make an accurate estimate of the rate of in situ dissolved Mn production at the twolocations. The most likely explanation for the coastal Mn enrichment, however, appearsto be horizontal advection from the reducing shelf sediments.Three depth profiles of dissolved Mn and dissolved 02 at station HS-14, HS-15, andHS-16 are shown in figure 4.1.3.6. All show distributions similar to those foundelsewhere except for the broader maxima of 0.34 to 0.42 nmol/kg at 500-1500m depth.This broad maximum may be due to Mn inputs from horizontal advection from thecontinental shelf, which become broader from the mixing of different water masses,68Figure 4.1.3.6 Dissolved Mn and Oxygen at Stations HS-14, HS-15, and HS-1669and/or the changing of water mass movements caused by the Karin Ridge in this area.The reason for this broad maximum is not clear at this stage. The ridge is thought to be asmall sink for Mn, due to the slow formation of ferro-manganese crusts. The growth rateof these crusts is very slow (1-3 mm/million years, Halbach, et al., 1983) and dissolvedMn concentrations in the water column are therefore not expected to be affected by thisprocess over the short time-scales of ocean mixing (<1000 years). The vertical profiles ofdissolved Mn show the combined effects of external sources, horizontal advection,oxidative scavenging and reductive processes.Vertical Profiles of Dissolved CoDissolved Co was measured by GFAAS, due to interference found by ICP-MS(potential interferences: 42C a 170 , 42ca 160H , 43C a 160 , 40Ar1801H and 24Mg35C1).Since ICP-MS was used to measure all metals initially, the measurements of dissolved Cowere only made on the available sample eluents remaining from five depth profiles.Two depth profiles of dissolved Co and 02 from the western North Pacific are plottedin figure 4.1.3.7. The Kuroshio (HS-5) and coastal (HS-1) stations show a somewhatmanganese-type profile. The coastal station (HS-1) has a higher surface Co value of 34pmol/kg, which may indicate riverine or eolian continental inputs, as observed in otherareas (Knauer, et al., 1982; Sakamoto-Arnold and Johnson, 1987; Jickell and Burton,1988). The mid-depth maximum of dissolved Co at 250-750m depth, coincident with the02 minimum zone, is similar to the Mn maximum. Below the 02 minimum zone,dissolved Co decreases with depth and reaches a low deep sea value of 18 pmol/kgaround 4500m, due to continuous scavenging. In addition to the similarities, there aresome distinct differences between Co and Mn. In the Kuroshio, the Co mid-depthmaximum is broader and shallower than that observed for Mn, whereas in the coastalstation Mn and Co are more tightly coupled. It is not clear why this is so. The possibleexplanations for this may due to complex inputs and removal processes in the open70Figure 4.1.3.7 Dissolved Co and Oxygen at Stations HS-1, HS-5, and HS-1071ocean: 1) input from the shallow regeneration of sinking particle could cause this shallowmaximum if dissolved Co is taken up by phytoplankton and regenerated in the nutriclineto a greater extent than Mn; 2) dissolved Co may also be released from Mn(III, IV)oxyhydroxides which are reduced and transported from the coastal reducing shelf regionin the 02 minimum zone.Three depths profiles of dissolved Co at station HS-14, HS-15, and HS-16 are shownin figure 4.1.3.8. The depth profiles at these three stations show trends similar to those ofthe two western Pacific stations, with the exception of a broader maxima at 500-1500m inthe central north Pacific stations. These broader maxima at 500-1500m depth may becaused by a combination of several processes described previously for station HS-5. Inaddition, water mass mixing caused by the Karin Ridge may contribute to these broadermaximum. The reason for this broad maxima is still not clear. The dissolved Codistributions need to be further investigated with larger sample size, since the detectionlimit of this technique for one liter seawater samples is close to the values measured. Thevertical profiles of dissolved Co show major trends similar to those of the dissolved Mnprofiles with the exception that the dissolved Co maximum in 02 minimum zone isshallower and broader than dissolved Mn maximum. In general the combination ofexternal sources, horizontal advection, oxidative scavenging and reductive processes, areall responsible for the distribution of dissolved Co in the ocean.00100020003000400050000100020003000400050000 ^10002000300040005000HS-15#.0O --0 °'cle —64HS-140000HS-14^0^HS-1500 00014S,—^ 00—HS-16^0\01^1^1^10o00HS-160 50 100 150 200 250 0 10 20 30 40 50 60Dissolved Oxygen (pM)^Dissolved Co (pmol/kg)Figure 4.1.3.8 Dissolved Co and Oxygen at Stations HS-14, HS-15, and HS-1673Dissolved Ce In Central North Pacific.Three vertical profiles of dissolved Ce in the central North Pacific are shown in figure4.1.3.9. These depth profiles agree reasonably well with previous studies in the IndianOcean (German and Elderfield, 1990) and in the Pacific and Atlantic oceans (de Barr, etal., 1985). In general, the vertical profiles of dissolved Ce exhibit a surface maximum,decreasing concentrations in the intermediate waters, enriched concentrations in theoxygen minimum zone, and gradually increasing concentrations towards the bottom.At station HS-14, dissolved Ce shows a surface maximum (3.1 pmol/kg), a sub-surface minimum (1.2 pmol/kg) at 300m depth, a slightly enriched concentration ofdissolved Ce (up to 2.7 pmol/kg) in the oxygen minimum zone (400-1200m), and slightlyincreasing concentrations towards the bottom (2.5 pmol/kg). The surface maximum mayindicate atmospheric inputs or a photoreductive source. The sub-surface minimum mayindicate the oxidative scavenging of dissolved Ce. The slightly enriched concentration inthe oxygen minimum zone may be caused by the reduction of Ce(IV) and Mn(III, IV)oxyhydroxides with subsequent release of the associated Ce(III), and/or horizontaladvection from coastal reducing shelf sediments. The slightly increasing concentrationtowards the bottom may indicate the diffusion of dissolved Ce out of the sediments or asediment surface remineralization. At stations HS-15 and HS-16, dissolved Ce showssimilar distributions, except for a higher value (2.0 pmol/kg) just above Karin Ridge(1500m) at HS-15. This high value of dissolved Ce in the bottom water may indicatebottom inputs by the diffusion of dissolved Ce from the Ridge surface.74751^2^3^4^5^6Dissolved Ce (pmolikg)0^50 100 150 200 250 0Dissolved Oxygen (pM)010002000300040005000010002000a.40005000010002000300040005000Figure 4.1.3.9 Dissolved Ce and Oxygen at Stations HS-14, HS-15, and HS-160.^00 09tpHS-14HS-15000*000IHS-16,o ooHS-14op HS-15!till4.2 Surface TransectThe dissolved Cd, Zn, Ni, Cu, Mn, and Pb distributions along the surface transectfrom Japan to Hawaii are plotted in figure 4.2.1 and figure 4.2.2. The surface transectfrom this cruise provides a chance to examine oceanic surface water Cd, Zn, Ni, Cu, Pb,and Mn levels, and to elucidate the relative importance of variations in riverine and eoliansources, variations in primary productivity and associated scavenging intensity, andadvective transport on the distribution of these metals in the western north Pacific Ocean.The surface data on dissolved Co at a few stations are also presented here (see figure4.2.2). The temperature and salinity, provided by IOS, are shown in figure 4.2.3. Thesurface data presented here show larger variations than those found in vertical profiles.Since good agreements were obtained between the vertical profiles in this study and otherprevious results (Bruland, 1980; Landing and Bruland, 1986; Flegal and Patterson, 1983),the surface data are believed to reflect the real variation of these metals in surfaceseawater, not analytical or sampling errors.The nutrient-type elements, Cd, Zn, Ni and Cu, generally show high concentrations inthe coastal station where higher nitrate, phosphate and silicic acid values are found, anddecrease by 2 to 4 times into the central gyre. The scavenged-type elements, Mn and Pb,also show high concentrations in the coastal station and decrease by a factor of 2 into theopen ocean. These indicate the river and continental inputs near the coast, diminishingatmospheric inputs away from the continent, as well as horizontal transport of thesemetals from the western boundary to the open ocean. Most metals show little changes inKuroshio Current (HS-3, HS-4 and HS-5) compared to the open ocean except Pb, whichdecreases gradually from west to east. In general, dissolved Co in the surface watershows no significant changes, though more detailed data are required to verify this.Midway along this transect, at station HS-6, high dissolved Cd, Zn, Ni, Cu, Mn and Pbare found. This feature is also found as a lower temperature and salinity at station HS-6in comparison to its neighboring stations. This may be caused by coastal waters being76155E 160E 165E 180 175W 170W145E 150 E 170E 175ENi0Io^o/\^\O o - 0^0 ■■,... 0.O8640.40.30.22.01.61.20.80.40.00.10.03.02.52.01.51.00.50.0Cd_1^/0,0^00 - '0, /^0 *^0■•elI^I^C°^I^I0AA kit 1 t i.^A_ A-A-^A- a AtZn. V \- .v^/V\-v v^'--- ffI^I^I^I^I^I^I^I^I^ICuLongitudeFigure 4.2.1 Dissolved Ni, Zn , Cu, and Cd in Surface Waters77*^Pb-o <><> 0 I\. O .,._1^.^_l_^.^1^.^I^1^I.^l.^I^_.^10.002.01.51.00.50.0145E0.050.040.030.020.01145E 150E 155E 160E 165E 170E 175E 180 175W 170W0.00Co0_O^ o01^.^1^.^I^i^1 1145E 150E 155E 160E 165E 170E 175E 180 175W 170WLongitudeFigure 4.2.2 Dissolved Pb, Mn and Co in Surface Waters0.120.080.04150E 155E 160E 165E 170E 175E 180 175W 170W781rv/-^V-v-vV _V'_ / - vvVI^I^I^I^I^I^I^I^I^I25300145E 150E 155E 160E 165E 170E 175E 180 175W170W5O_ 0. 0 // 0 - .0 ,000IIIIIIIIII36.035.62^35.2.1^34.834.434.0145E 150E 155E 160E 165E 170E 175E 180 175W170WLongitudeFigure 4.2.3 Temperature and Salinity in Surface Waters79carried out via a cold-core ring, broken off from the Kuroshio Current. The cold-corering can be formed by a meander pinching off from a strong current (in this case isKuroshio Current), carrying coastal water out into the central gyre or bring a deep waterfrom a couple of hundred meters to the surface. Both processes result in a decrease intemperature and salinity. Upwelling from a couple of hundred meters as the source oflow temperature and salinity can be eliminated since higher concentrations of bothnutrient-type elements (Cd, Zn, Ni, and Cu) with minimum concentrations in the surface,and the scavenging-ype element, Mn, with maximum concentration in the surface, wereobserved. A comparison of the surface concentrations in the two stations west of theKuroshio (HS-1 and HS-2) shows upwelling at station HS-2. Elements which have asurface minimum (Cd, Zn, Cu, Ni), are elevated at this station, while the element whichhas a surface maximum (Mn) is depleted. These features (upwelling and a cold-core ringwhich bring water from the west coast) can not be identified by temperature and salinitydata alone, since both processes result in lower temperature and salinity in the surfacewaters. The use of multiple trace metals is specially useful for elucidating theseprocesses.One of the most surprising features of the Pb surface transect is the lack of a stronggradient from increased eolian inputs closer to the coastal region (Asian dust andanthropogenic lead sources). Dust and lead fluxes are both thought to vary by nearly twoorders of magnitude across this transect (Duce, et al., 1991), yet the range observed forPb is only two-fold. Much of the expected input may be balanced by increasedscavenging in the coastal region. It is also likely that sampling closer to the coast ofJapan (not possible from the Russian ship) would have shown the effects of theseincreased inputs.804.3 CONCLUSIONSThe concentrations of dissolved Cd, Zn, Ni, Cu, Pb, Al, Ga, Mn, and Ce weredetermined in seawater samples from the western North Pacific by preconcentration on aChelex-100 resin followed by analysis with ICP-MS. Dissolved Co was measured onGFAAS due to Ca and Mg interferences on ICP-MS. Concentrations of these elementsare found to be in the range of 0.7 pmol/kg to 12 nmol/kg, consistent with previousobservations.Vertical profiles for dissolved Cd, Ni, Zn, Cu, Pb, Mn, and Co from the central gyrestation, closest to Hawaii (28°N, 175°E), agree well with previous results (Bruland, 1980;Landing and Bruland, 1987; Flegal and Patterson, 1983; Bruland, et al., 1993).Remarkably similar trace metal concentrations and vertical distributions are observed inthe Ku-roshio Current station (35°N, 155°E), about 3500 km west of the central gyrestation. At the coastal station (38°N, 146°E), enriched trace metal concentrations arefound in the upper water column, with a shallower maximum for those elements with amid-depth maximum. Dissolved Al, Ce, and Ga in the central North Pacific show majortrends similar to those of other studies (Orians and Bruland, 1985, 1988; de Barr, et al.,1985), with the exception of maximum Al concentrations found at 100-250 m rather thanat the surface, and higher intermediate Al values. The sub-surface maximum observed inthis study could be caused by the advection via the North Equatorial current, bringing lowAl content waters from the coast of North America to this area in the surface. The highermid-depth Al could be due to the inputs from the bottom of Karin Ridge.The surface transect (25 m depth) for dissolved Cd, Ni, Zn, Cu, Pb, and Mn in thewestern North Pacific shows generally high coastal values and low concentrations in theopen ocean. Upwelling can be seen at station HS-2 by enrichments of elements with asurface minimum (Cd, Zn, Cu, Ni) and depletion of the element with a surface maximum(Mn). Eddy transport, in the form of a cold-core ring, results in higher concentrations ofall elements at station HS-6. The data from surface waters, presented here, show a higher81degree of variability than observed in vertical profiles. These data indicate a real surfacevariability, not a sampling or analytical artifact. One of the most surprising features ofthe surface transect is the lack of a strong gradient from increased eolian inputs closer tothe coastal region. Dust and lead fluxes are both thought to vary by nearly two orders ofmagnitude across this transect, yet the range observed for the trace elements is only 2 to 4fold. Much of the expected input appears to be balanced by increased scavenging in thecoastal region. It is also likely that sampling closer to the coast of Japan (samples notavailable from this area) would have shown the effects of these increased inputs.CHAPTER 5DISSOLVED MANGANESE AND COBALT IN A COASTAL INLET5.1 INTRODUCTIONThe oceanic distributions and biogeochemical behavior of dissolved Mn and Co arecontrolled by complex interactions among input, internal cycling, and removal processes,coupled with physical transport and mixing in the oceans (Bender, et al., 1977; Bruland, etal., 1979; Klinkhammer and Bender, 1980; Landing and Bruland, 1980; Martin andKnauer, 1980; Knauer, et al., 1982; Gordeyev and Atnashev, 1990). Over the last fifteenyears, the use of relatively contamination-free sampling and analytical methods has resultedin a significant increase in our understanding of the marine biogeochemistry of Mn(Bender, et al., 1977; Landing and Bruland, 1986). However, there is still a paucity ofdissolved oceanic Co data due to extremely low concentration of Co in the ocean and theresulting sampling and analytical difficulties (Knauer, et al., 1982). The goal of thischapter was to establish a contamination-free method for collecting seawater samples and touse this method to investigate the behavior of dissolved Mn and Co with changing 02levels in Howe Sound, B. C., Canada.Mn is brought into the ocean mainly by river and atmospheric inputs. Mn also hassubmarine hydrothermal inputs (Yeats and Bewers, 1985; Statham and Burton, 1986;Landing and Bruland, 1986; Wilhelmy and Flegal, 1991). Once introduced, Mnparticipates in a wide range of biogeochemical processes. Mn may desorb fromatmospherically introduced particles in the surface of the ocean (Hodge, et al., 1978), anddissolved Mn may also adsorb onto particles (Knauer, et al., 1982; Martin and Knauer,1984). As a required biochemical micro-nutrient, Mn may be incorporated into organictissue and be involved in internal cycling (Landing and Bruland, 1986). Mn can exist in avariety of oxidation states under different environmental conditions; in low 02 water it83exists as soluble Mn(II), in high 02 water it is oxidized to insoluble Mn(III, IV)oxyhydroxides. When Mn(III, IV) oxyhydroxides form, dissolved Co is scavenged ontothese mineral phases (Knauer, et al., 1982; Krernling and Hydes, 1988). Dissolved Cotherefore has a behavior similar to that of Mn (Knauer, et al., 1982).In the present study, two well developed techniques were used to collect samples forMn and Co analysis from Howe Sound: the C Ill rosette Niskin bottles and the "Go-flo"sampler. From a statistical analysis (the Student-t-test, a=0.05) of triplicate seawatersamples (collected from both the CTD rosette Niskin bottles and the "Go-flo" sampler) at50m depth, the possibility of contamination in sampling was evaluated. Finally, bycomparing these measurements with the 02 levels in the study area, the behavior ofdissolved Mn and Co as a function of oxygenation in Howe Sound was investigated.5.2 METHODS5.2.1 The study areaSeawater samples were collected in Howe Sound, 490 37.7' N, 123013.8' W. Below50m in this region, the dissolved oxygen concentration decreased with depth and reached aminimum value (6 tmol/L) in the bottom water.5.2.2 Sampling methodOn March 4, 1992, seawater samples were collected from Howe Sound using aTeflon8-coated "Go-flo" sampler (30L) attached to a galvanized wire (the distance betweenthe "Go-flo" and the wire was 2 cm) at five depths: 50, 100, 150, 200 and 250m.Triplicate samples were collected at 50m depth using both the "Go-flo" sampler and theNiskin bottles (1.7L) attached to the CID rosette on a stainless wire (the distance betweenthe Niskin bottles and the wire was 50 cm) for the study of contamination caused by84equipment used during sampling. Surface samples were collected directly in storage bottlesupstream from the ship using a pole sampler. All samples were stored in a freezer in 1Lacid washed polyethylene bottles on the ship. Samples used for oxygen measurementswere collected from the CTD rosette Niskin bottles at eleven depths: 2, 5, 10, 20, 30, 50,75, 100, 150, 200 and 250m.5.2.3 Analysis methodSamples were analyzed in a trace metal clean lab (HEPA filtered - high efficient particleair - positive pressure air supply) at the Department of Oceanography, UBC. Thepreconcentration and separation of dissolved Co from the seawater matrix was conductedusing an 8-hydroxyquinoline (8-HOQ) resin on a silica gel support. Prior to use, Teflon®columns with 4 ml resin were cleaned by gravity elution with 45 ml 2.37 N HNO3 (madefrom double distilled, HNO3, purchased from Seastar Chemicals Inc.) and rinsed withNH4Ac solution (prepared from high quality concentrated HAc and NH4OH purchasedfrom Seastar Chemicals Inc., Sidney, Canada). The seawater samples were filteredthrough acid washed, 0.4 gm, nuclepore polycarbonate membrane filters, and adjusted tothe optimum pH (8-8.5) for extraction. Subsamples (100-400 g) of the seawater werepumped through the resin columns at 0.5-0.7 ml/min. The columns were then eluted with2.37 N HNO3 after rinsing with 8 ml NH4Ac and 20 ml DI water (pH is 8.0-8.5).Finally, the column eluents were analyzed in duplicate by graphic furnace atomicabsorption spectroscopy (GFAAS, Varian spectra AA 300 with Zeeman backgroundcorrector, equipped with a graphite tube analyzer) for dissolved Co.For dissolved Mn, the seawater samples were analyzed in duplicate by GFAAS withdirect injection after a 4-10 fold dilution with 2.37 N HNO3. Dissolved Mn in columneluents was above the linear range for GFAAS in all samples except those from 50m depth.These samples used for the contamination test were processed through the 8-HOQ resin asdescribed for dissolved Co measurements.85Oxygen concentrations were measured using the Winkler titration technique (Carritt andCarpenter, 1966) on the ship by the Oceanography Departmental technician from UBC.The measurements of two triplicate samples for the contamination sources study werecompared using a t-test (a=0.05) with the null hypotheses being: [Co] in "Go-flo" samplesequal [Co] in Niskin bottle samples, and [Mn] in Go-Flo samples equal [Mn] in Niskinbottle samples.5.3 RESULTS AND DISCUSSIONThe dissolved Mn, Co, and oxygen data are presented in table 5.3.1. Vertical profilesof dissolved Mn, Co and oxygen are shown in figure 5.3.2. Oxygen concentrationsdecreased from the surface to 30m, and then increased and reached a subsurface maxima at50m (Fig. 5.3.2). Below 50m, oxygen concentrations decreased with depth and reachedthe lowest value in the bottom water (6 gm/L at 250m). In general, the dissolved Mnconcentration decreased from the surface (0.262±0.031 gmol/kg) down to 50m depth(0.024±0.004 p.mol/kg) (Fig. 5.3.2). Below this depth, dissolved Mn concentrationsincreased with depth and decreasing 02, and reached a maximum value in the bottom water(3.22±0.35 gmol/kg at 250m, Fig. 5.3.2).The slightly elevated dissolved Mn concentration in the surface water may result fromriver and/or atmospheric inputs to Howe Sound, as observed in other oceanic regions(Bender, et al., 1977; Bruland, et al., 1979; Klinkhammer and Bender, 1980; Landing andBruland, 1980; Martin and Knauer, 1980). The subsurface minimum in Mn concentration,where high oxygen concentration was found, may indicate the oxidizing and scavenging ofdissolved Mn. The dramatic increase in Mn concentration at low levels of 02 results fromthe reduction of Mn(III, IV) oxyhydroxides to soluble Mn(II).The dissolved Co vertical profile (Fig. 5.3.2) shows trends similar to that of Mn. Anelevated dissolved Co concentration was found in surface seawater (0.66±0.17 nmol/kg),86Table 5.3.1 Dissolved Mn, Co, and Oxygen in Seawater in Howe SoundDepth (m) Dissolved Mn(nmol/kg)Dissolved Co(nmol/kg)Dissolved 02(gmo1/1-)2 0.26 0.66 2805 26210 24120 21030 20050 0.024 0.31 21675 185100 0.076 0.60 113150 0.095 0.65 61200 1.40 3.37 18250 3.22 7.13 6probably indicating river and atmospheric inputs as in the open ocean (Knauer, et al.,1982). The lower dissolved Co concentration at 50m (0.31±0.06 nmol/kg) may be due toscavenging onto Mn oxyhydroxides. The increased dissolved Co concentration from 50mto the bottom (7.13±0.22 nmol/kg) may result from the release of Co from Mn(III, IV)oxyhydroxides, as they are reduced to soluble Mn(II) in low oxygen bottom waters. Thesimilarity of the vertical profiles of dissolved Co and Mn can be viewed by plotting thedissolved Mn concentration vs. the dissolved Co concentrations (Fig. 5.3.3). This870000Figure 5.3.2 Dissolved Mn, Co and oxygen in Howe Sound0 0VDFigure 5.3.4 The IogCo, logMn vs. log oxygen Seawater Profiles from 50m to 250m at Howe Soundshows a linear relationship, with R2 =0.999, and demonstrates the quantitative similaritybetween dissolved Co and dissolved Mn at this study area. The relationship betweendissolved Mn, Co and 02 levels in Howe Sound can be demonstrated by plotting log Mnvs. log 02 (Fig. 5.3.4) and log Co vs. log 02 (Fig. 5.3.4). These data show good linearrelationships, with R2 of 0.984 and 0.987, respectively. These are linear because log 02controls the reducing potential of the environment, which controls the Mn solubility andtherefore the Co concentration. Thus, the reducing potential has a linear relationship withlog Mn, log Co, and log 02 (Manahan, 1991). The surface data were not used in plottingFig. 5.3.4, due to alternate sources of dissolved Mn and Co to the surface such as river andatmospheric inputs, not related to the relationship with 02. These results quantitativelydemonstrate the sensitive redox behavior of dissolved Mn and Co in a coastal marineregion. The agreement between duplicate measurements averaged around 10% of themean.The possibility of contamination due to the two different sampling methods used in thisproject was examined using a t-test (a=0.05) (using the two triplicate samples collected byboth Niskin and "Go-flo" bottles at 50m depth in Howe Sound). There was no significantdifference between the two groups of samples in dissolved Mn concentrations (Table5.3.2), suggesting both sampling methods are suitable for collecting dissolved Mn samplesat these high coastal levels. The high variability in Mn data may be due to the low resincapacity (6 gmol/g for Cu2+) which therefore affects Mn recovery. For dissolved Co,however, there were higher concentrations in the "Go-flo" samples than those in the CTDrosette Niskin bottle samples (Table 5.3.2). This finding suggests that there wascontamination during "Go-flo" sampling, which may result from the galvanized wire used(the distance between the wire and the "Go-Flo" is only 2 cm). The Niskin bottles attachedto the CID rosette were about one-half meter away from the stainless hydraulic wire. Evenif there was the same amount of Mn and Co in the hydraulic wire used in the sampling, thecontamination may not be seen in dissolved Mn samples due to the 60 fold higher dissolved90Figure 5.3.3 Dissolved Mn vs. Dissolved Co in Seawater, Howe SoundTable 5.3.2 The Statistical Analysis (Student-t test) for Dissolved Mn andCo (nmol/kg) in Seawater Collected From 50m in Howe Sound Using CTDRosette Niskin Bottles with a Stainless Wire and "Go-flo" Sampler with aGalvanized WireDissolved Co (nmol/kg) Dissolved Mn (nmol/kg)Sampling methods Go-flo Niskin Go-flo NiskinResults 0.338 0.109 15.39 26.610.237 0.115 21.01 15.430.339 0.109 24.97 30.47Average 0.305 0.109 20.46 24.17Degrees of freedom 2 2 2 2S 0.059 0.0059 13.27 7.81SD 0.042 10.88t value 4.68>1.388 (critical t) 0.34<1.388 (critical t)Conclusion significant differencebetween two groupsno significant differencebetween two groupsSD : Standard deviation between two groupsS:^Standard deviationMn concentration at 50m. This could explain the lower dissolved Co concentrations foundin the C ID rosette Niskin bottle samples, whereas Mn concentrations were not significantlydifferent (a=0.05). Manganese and Co concentrations in the open ocean samples,collected with "Go-flo" sampler on Kevlarg line, however, show a good agreement withother previous results (see chapter 4). This shows that there was no contaminationproblem from the "Go-flo" sampler itself. The higher dissolved Co values, found in HoweSound samples collected with "Go-flo" sampler attached on gavanized wire, show theimportance of using suitable wire for sampling. Fortunately, the contamination in the "Go-92flo" samples for dissolved Co (a 0.19 nmol/kg difference) is smaller than theconcentrations of dissolved Co in the seawater samples collected in Howe Sound (0.66-7.13 nmol/kg), except at 50m depth. The linear relationship found between dissolved Mnand dissolved Co should not be changed from contamination of dissolved Co during thesampling, as this should affect all samples by the same amount.5.4 CONCLUSIONSThe vertical distributions of dissolved Mn and Co were found to be similar to previousreports (Knauer, et al., 1982; Yeats and Bewers, 1985; Statham and Burton, 1986;Landing and Bruland, 1986; Kremling and Hydes, 1988; Wilhelmy and Flegal, 1991),although dissolved Mn and Co had very high concentrations in this inlet compared to thecoastal and open ocean. The dissolved Mn and Co concentrations decreased from thesurface down to 50m, reaching minimum values (0.024±0.003 p.mol/kg for dissolved Mn,0.31±0.06 nmol/kg for dissolved Co) at an oxygen subsurface maxima (50m). Below50m depth, the concentrations of dissolved Mn and Co dramatically increased with depthand reached maximum values in the bottom water (7.13±0.22 nmol/kg for Co, 3.22±0.35gmol/kg for Mn) at the lowest 02 level. A linear relationship is found in log 02 vs. log Coand log Mn (Fig. 5.3.2). Similar behavior of dissolved Co and Mn is found, with a linearrelationship between dissolved Mn and Co. This finding suggests that dissolved Co andMn are controlled in similar biogeochemical processes.The results of the contamination study show that both sampling methods ( "Go-flo"sampler and the C11) rosette Niskin bottle sampler) are suitable for collecting dissolved Mnsamples at these gmol/kg levels. The sampling method using the "Go-flo" samplerattached to a galvanized wire is not suitable for collecting dissolved Co samples at nmol/kglevels.93CHAPTER 6SUMMARYAn accurate, multi-elemental technique, using ICP-MS for the "simultaneous"determination of dissolved Cd, Zn, Ni, Cu, Pb, Mn, and GFAAS for the determination ofdissolved Co in seawater, after preconcentration on Chelex-100 resin at either pH 8.0±0.1or 6.5±0.2, was developed. This technique was tested by the analysis of standardreference materials from the National Research Council of Canada for Cd, Ni, Zn, Cu, Pb,Mn, and Co. Results are in good agreement with the certified values, within 99%confidence limit. Analytical precision is better than 7% for these metals. This method isalso suited for dissolved Al, Ga, and Ce measurements in seawater. The seawaterdetection limits are in the low picomolar range (2-40 pmol/kg) after a 125-foldpreconcentration. This multi-elemental technique, which has the ability to measure thesetrace metals in seawater, provides important information which helps to elucidate thebiogeochemistry of these metals in marine environments.Concentrations of these ten elements are found to be in the range of 0.7 pmol/kg to 12nmol/kg. Vertical profiles of Cd, Zn, Ni, Cu, Pb, Mn and Co from the North Pacificcentral gyre (28°N, 175°E) agree well with previous results from nearby stations (Bruland,1980; Landing and Bruland, 1986; Flegal and Patterson, 1983; Bruland, et al., 1993),demonstrating contamination-free sampling and reliable analytical procedures. Cadmium,Zn, and Ni show nutrient-type distributions, with surface depletion and maximumconcentrations coherent nutrient maximum throughout the western North Pacific. TheCd/PO4 and Zn/Si correlation observed in other oceanic regions still hold in this area. Thisindicates that the dissolved Cd distribution is largely driven by vertical transport viashallow regeneration, and that Cd can be used as a proxy for PO4 throughout the western94North Pacific. Zinc is removed from the surface waters by processes related to theconstruction of the hard-parts of phytoplankton, and transported to the deep sea where it isregenerated with Si. The linear relationships of Ni/PO4 and Ni/Si support the contentionthat the dissolved Ni distribution is largely controlled by internal cycles with both deepregeneration and shallow regeneration. Unlike Cd, Zn and Ni, whose vertical distributionsare only controlled solely by internal cycles, Cu shows clear evidence of surface depletion,intermediate and deep water scavenging, and bottom sources from the sediments. Theresidence time for dissolved Cu in deep water is found to be 1036 to 1066 years in thewestern North Pacific.The vertical profiles of dissolved Pb support the contention that external sources andscavenging removal processes are the major processes responsible for the Pb distribution inseawater. Horizontal transport Pb from the west appears to be important in resulting thesub-surface maximum in the open ocean. The vertical distributions of dissolved Mn in thewestern North Pacific exhibit a surface maximum, a sub-surface minimum, a maximum inthe oxygen minimum zone and decreasing concentrations with depth towards the bottom.Horizontal advection appears to be important for supporting the dissolved Mn maxima inthe 02 minimum zone in the open ocean. The vertical profiles of dissolved Co show majortrends similar to those of dissolved Mn, with the exception that the dissolved Co maximumis shallower and broader than the dissolved Mn maximum in the 02 minimum zone.Dissolved Al, Ce, and Ga in the central North Pacific show major trends similar tothose of other studies (Orians and Bruland, 1985, 1988; de Barr, et al., 1985). Themaximum Al concentrations at 100-250m, rather than at the surface, may be caused byadvection via the North Equatorial current, bringing low Al content waters from the coastof North America to the surface waters in this area. The vertical profiles of dissolved Ceexhibit a surface maximum, decreasing concentrations in the intermediate depth, enrichedconcentrations in the oxygen minimum zone, and gradually increasing concentrationstowards the bottom. Gallium distributions found at this area show intermediate surface95concentrations, a slight subsurface maximum at 100 m, a minimum at —1000 m and abottom maximum. This type of distribution suggests a subsurface and a bottom watersource with scavenging removal throughout. All trace metal data from the central NorthPacific show that the Karin Ridge (beneath station HS-15), where thick ferro-manganesecrusts are found, does not provide a significant sink for Mn and other scavenged-typeelements (e.g. Co and Ce) in terms of their distributions, controlled by oceanic short timescale processes, but may still be important geologically. The Karin Ridge may be a sourcefor higher intermediate Al values observed, and may also play some role in changing watermass movement in this area, and therefore causing a broader maximum in Mn and Covertical profiles.The dissolved Cd, Zn, Ni, Cu, Pb, Mn and Co concentrations in the western NorthPacific, along the surface transect from Japan to Hawaii, decrease by a factor of 2 to 4 fromthe coastal region to the oligotrophic central gyre. Upwelling can be seen by enrichmentsof Cd, Zn, Cu, Ni, elements with surface minima, and depletion of Mn, which has asurface maximum. Eddy transport, in the form of a cold-core ring, which brings coastalwater to the open ocean, results in higher concentrations of all elements. Both theseadvective features lead to reduced salinity and temperature, and are difficult to distinguishon this basis. The use of multiple trace metals, each with different depth dependentdistributions, helps to elucidate the origin of the low temperature and salinity waters inthese situations. The high degree of surface variability reflects advective transport, thevarying amount of primary production, and the rate of atmospheric input. This surfacetransect shows a lack of a strong gradient in surface water trace metal concentrations fromthe increased eolian inputs closer to Asian sources. Much of the expected input appears tobe balanced by increased scavenging in the coastal region.Dissolved Mn and Co have very high concentrations (0.024-3.22 gmol/kg for Mn, and0.31-7.13 nmol/kg) in coastal inlet waters in Howe Sound, B. C., compared to the coastaland open ocean (0.097-1.95 nmol/kg for Mn, and 6-47 pmol/kg for Co). Dissolved Co96shows behavior similar to that of dissolved Mn in this coastal inlet, as both appear to becontrolled primarily by the degree of oxygenation, which controls the reducing potential ofthe environment.97REFERENCES:Bender, M. L., Klinkhammer, G. P. and Spencer, D. W., 1977. Manganese in seawaterand the marine manganese balance. Deep sea Res., 24, 799-812.Berman, S. S., Mclaren, J. W. and Willie, S. N., 1980. Simultaneous determination offive trace metals in seawater by inductively coupled plasma atomic emissionspectrometry with ultrasonic nebulization. Anal. Chem., 52, 488-492.Boyle, E. A., Sclater, F. and Edmond, J. M., 1976. On the marine geochemistry ofcadmium. Nature, 263, 42-44.Boyle, E. A., Sclater, F. R. and Edmond, J. M., 1977. The distribution of dissolvedcopper in the Pacific. Earth Planet. Sci. Lett., 37, 38-54.Boyle, E. A., Chapmick, S. D., and Shen, G. T., 1986, Temporal variability of lead in thewestern North Atlantic. J. Geophys. Res., 91, C7, 8573-8593.Boyle, E. A., Handy, B. and Geen, A, 1987. Cobalt Determination in natural watersusing cation exchange liquid chromatography with luminol chemiluminescencedetection. Anal. Chem., 59, 1499-1503.Bruland, K. W., Franks, R. P., Knauer, G. A. and Martin, J. H., 1979. Sampling andanalytical methods for the determination of copper, cadmium, zinc and nickel at thenanogram per liter level in seawater. Anal. Chim. Acta, 105, 233-245.Bruland, K.W., Knauer, G. A. and Martin, J. H, 1980. Zinc in Northeast Pacific waters.Limnol. Oceanogr., 23, 618-625.Bruland, K. W., 1980. Oceanographic distributions of cadmium, zinc, nickel and copperin the North Pacific. Earth Planet. Sci. Lett., 47, 176-198.Bruland, K. W.,1983. Trace elements in seawater. In: Chemical oceanography. Riley, J.P. and Chester, R. eds. Academic Press, London, pp 157-220.Bruland, K. W. and Franks, R., 1983. Mn, Ni, Cu, Zn, and Cd in the western NorthAtlantic. In: Trace metals in seawater. Wong, C. S., Boyle, E., Bruland, K. W.,Burton, J. D. and Goldberg, E. B. eds. Plenum Press, pp 395-415.98Bruland, K. W., Orians, K. J. and Cowen, J. P., 1993. Reactive trace metals in thestratified central North Pacific. Submitted to Geochim. Cosmochim. Acta.Burton, J. D., Maher, W. A. and Statham, P. J., 1983. Some recent measurements oftrace metals in Atlantic Ocean waters. In: Trace metals in seawater. Wong, C. S.,Boyle, E., Bruland, K. W., Burton, J. D. and Goldberg, E. B. eds. Plenum Press, etal eds. Plenum Press, pp 415-426.Carritt, D. E. and Carpenter, J. H., 1966. Comparison and evaluation of currentlyemployed modifications of the Winkler method for determining dissolved oxygen inseawater. J. Mar. Res., 24, 21-26.Date, A. R. and Gray, A. L., 1989. In: Application of inductively coupled plasma massspectrometry. Chapman and Hall, a division of Routledge, Chapman and Hall Inc.,New York, pp 1001-2291.Danielsson, L. G., Magnusson, B. and Westerlund, S., 1985. Cadmium, copper, iron,nickel and zinc in the North East Atlantic Ocean. Mar. Chem., 17, 23-41.de Barr, H. J. W., Bacon, M. P., Brewer, P. G. and Bruland, K. W., 1985. Rare earthelements in the Pacific and Atlantic oceans. Geochim. Cosmochim. Acta, 49, 1943-1959.Duce, R.A., Liss, P. S., Merrill, J. T., Atlas, E. L., Buat-Menard, P., Hicks, B. B.,Miller, J. M., Prospero, J. M., Arimoto, R., Church, T. M., Ellis, W., Galloway, J.N., Hansen, L., Jickells, T. D., Knap, A. H., Reinhardt, K. H., Schneider, J. K.,Soudine, A., Tokos, J. J., Tsunogai, S., Wollast, R. and Zhou, M., 1991. TheAtmospheric input of trace species to world ocean. Global Biogeochemical Cycles, 5,No3, 193-259.Falkner, K. K. and Edmond, J. M., 1990. Gold in seawater. Earth Planet. Sci. Lett., 98,208-220.Flegal, A. R. and Patterson, C. C., 1983. Vertical concentration profiles of lead in thecentral Pacific at 15°N and 20°S. Earth and Planet. Sci. lett., 64, 19-32.99German, C. R. and Elderfield, H., 1990. Rare earth elements in the NW Indian Ocean.Geochim. Cosmochim. Acta, 54, 1929-1940.Gordeyev, V. V. and Atnashev, V. B., 1990. Dissolved manganese in the NortheasternAtlantic Ocean. Oceanology, 30, No. 1, 58-63.Halbach, P., Segl, M., Puteanus, D. and Mangini, A., 1983. Co-fluxes and growth ratesin ferromanganese deposits from central Pacific seamount areas. Nature, 304, 25, 716-719.Henshaw, J. M., Heithmar, E. M. and Hinners, T. A., 1989. Inductively coupled plasmamass spectrometry determination of trace elements in surface waters subject to acidicdeposition. Anal. Chem., 61, 335-342.Hodge, V. , Johnson, S. R. and Goldberg, E. D., 1978. Influence of atmosphericallytransported aerosols on surface ocean water composition. Geochem. J., 12, 7-20.Holscher, H. H., 1971. In: Simplified statistical analysis. Cahners Books, division ofCahners Publishing Co. pp 109-132.Hydes, D. J., 1979. Aluminum in seawater: Control by inorganic processes. Science,205, 1260-1262.Jickell, T. D. and Burton, J. D., 1988. Cobalt, copper, manganese and nickel in theSargasso sea. Mar. chem., 23, 131-144.Kingston, H. M., Barnes, I. L., Brady, T. J. and Rains, T. C., 1978. Separation of eighttransition elements from alkali and alkaline earth elements in estuarine and seawaterwith chelating resin and their determination by graphite furnace atomic absorptionspectrometry. Anal. Chem., 50, 2064-2070.Klinkhammer, G. P. and Bender, M. L., 1980. The distribution of manganese in thePacific Ocean. Earth Planet Sci. Lett., 46, No. 3, 361-384.Klinlchammer, G. P. and Chan, L. H., 1990. Determination of barium in marine watersby isotope dilution inductively coupled plasma mass spectrometry. Analyt. Chim.Acta, 232, 323-329.100Knauer, G. A. and Martin, J., 1980. Phosphorus-cadmium cycling in North East Pacificwaters. J. Mar. Res., 39, 65-76.Knauer, G. A., Martin, J. H. and Gordon, R. M., 1982. Cobalt in the North-East PacificOcean. Nature, 297, 49-51.Kremling, K. and Hydes, D., 1988. Summer distribution of dissolved Al, Cd, Co, Cu,Mn and Ni In surface waters around the British Isles. Continental Shelf Res., 8, No.1, 89-106.Landing, W. M. and Bruland, K. W., 1980. Manganese in the North Pacific. EarthPlanet Sci. Lett., 49, No. 1, 45-56.Landing, W. M. and Bruland, K. W, 1987. The contrasting biogeochemistry of iron andmanganese in the Pacific Ocean. Geochim. Cosmochim. Acta, 51, 29-43.Manahan, S. E., 1991. In: Environmental chemistry. Lewis Publisher, pp 79-83.Martin, J. H. and Knauer, F. A., 1984. Vertex: Manganese transport through oxygenminima. Earth Planet. Sci. lett., 67, 63-147.Martin, J. H. and Knauer, G. A., 1980. Manganese cycling in Northeast Pacific waters.Earth Planet Sci. Let., 51, No. 2, 266-274.Measures, C. I., Grant, B., Khadem, M., Lee, D. S. and Edmond, J. M., 1984.Distribution of Be, Al, Se and Bi in the surface waters of the western North Atlanticand Caribean. Earth Planet Sci. Lett., 71, 1-12.McLaren, J. W., Mykytiuk, A. P., Willie, S. N. and Berman, S. S., 1985. Determinationof trace metals in seawater by inductively coupled plasma mass spectrometry withpreconcentration on silica-immobilized 8-Hydroxyquinoline. Anal. Chem., 57, 2907-2911.Mottl, M. J. and McConachy, T. F., 1990. Chemical processes in buoyant hydrothermalplumes on the East Pacific Rise Near 21 0N. Geochim. Cosmochim. Acta, 54, 1911-1927.101Orians, K. J. and Boyle, E. A., 1993. The determination of picomolar concentrations oftitanium, gallium and indium in seawater by inductively coupled plasma massspectrometry (ICP-MS) following an 8-Hydroxyquinoline chelating resinpreconcentration. Submitted to Anal. Chim. Acta.Orians, K. J., and Bruland, K. W., 1988. The marine geochemistry of dissolved gallium:a comparison eith dissolved aluminum. Geochim. Cosmochim. Acta, 52, 2955-2962.Orians, K. J. and Bruland, K. W., 1986. The biogeochemistry of aluminum in the PacificOcean. Earth and Planet. Sci. Lett., 78, 397-410.Orians, K. J. and Bruland, K. W., 1985. Dissolved aluminum in the central NorthPacific. Nature, 316, 427-429.Pai, S. C, Whung, P. Y. and Lai, R. L., 1988. Pre-concentration efficiency of Chelex-100 resin for heavy metals in seawater. Partl , Effects of pH and salts on thedistribution of heavy metals. Analyt. Chim. Acta, 211, 257-270.Pai, S. C., 1988. Pre-concentration efficiency of Chelex-100 resin for heavy metals inseawater. Part 2. Distribution of heavy metals on a Chelex-100 column andoptimization of the column efficiency by a plate simulation method. Analyt. Chim.Acta, 211, 271-280.Paulson, A. J., 1986. Effects of flow rate and pretreatment on the extraction of tracemetals from estuarine and coastal seawater by Chelex-100. Anal. Chem., 58, 183-187.Pickard, G. L. and Emery, W. I., 1990. In: Descriptive physical oceanography. Fifth ed.Pergamon Press, Member of Maxwell Macmillan Pergamon Publishing Corporation,pp 252-257.Price, N. M. and Morel, F. M. M., 1990. Cadmium and cobalt substitution for zinc in Amarine diatom. Nature, 344, 12, 658-660.Reid, J. L. and Lynn, R. J., 1971. On the influence of the Norwegian-Greenland andWeddel seas upon the bottom waters of the Indian and Pacific oceans. Deep Sea Res.,18, 1063-1088.102Resing, J. A. and Mottl, M. J., 1992. Determination of manganese in seawater using flowinjection analysis with on-line preconcentration and spectrophotometric detection.Anal. Chem., 64, 2682-2687.Rudels, B., 1987. On the exchange of water masses between the polar-ocean and theNorth Atlantic. Ph. D. Dissertation. University in Goteborg, Sweden.Saager, P. M., De Barr, H. J. W. and Burkill, P. H. 1989. Manganese and iron in IndianOcean waters. Geochim. Cosmochim. Acta, 53, 2259-2267.Sakamoto-Arnold, C. M. and Johnson, D. S., 1987. Determination of picomolar levels ofcobalt in seawater by flow injection analysis with chemiluminescence detection. Anal.Chem., 59, 1789-1794.Shabani, M. B., Akagi, T., Shimizu, H. and Masuda, A., 1990. Determination of traceLanthanides and yttrium in seawater by inductively coupled plasma mass spectrometryafter preconcentration with solvent extraction and back extraction. Anal. Chem., 62,2709-2714.Shaw, T. J. and Francois, R., 1991. A fast and sensitive ICP-MS assay for thedetermination of 230Th in marine sediments. Geochim. Cosmochim. Acta, 55, 2075-2078.Shiller, A. M., 1988. Enrichment of dissolved gallium relative to aluminum in naturalwater. Geochim. Cosmochim. Acta, 52, 1879-1882.Statham, P. J. and Burton, J. D., 1986. Dissolved manganese in the North AtlanticOcean, 0-350 N. Earth Planet Sci. Lett., 79, No. 1, 55-65.Sturgeon, R. E., Berman, S. S., Desaulniers, A. H. and Russell, D. S., 1980.Preconcentration of trace metals from seawater for determination by graphite funanceatomic absorption spectrometry. Talanta, 27, 85-94.Sunda, G. W. and Huntsman, S. A., 1988. Effect of sunlite on redox cycles ofmanganese in the southwestern Sargasso sea. Deep Sea Res., 35, 8, 1297-1317.Taylor, S. R., 1964. Abundance of chemical elements in the continental crust: A new103table. Geochim. Cosmochim. Acta, 28, 1273-1285.Van, Berkel, W. W., Overbosch, A. W., Feenstra, G. and Maessent, F. J. M. J., 1988.Enrichment of artificial seawater of critical examination of Chelex-100 for group-wiseanalyte pre-concentration and matrix separation. J. Analyt. Atomic spectrometry, 3,249-257.Westerlund, S. and Ohman, P., 1991. Cadmium, copper, cobalt, nickel, lead, and zinc inthe water column of the Weddell sea, Antarctica. Geochim. Cosmochim. Acta, 55,2127-2146.Whitehead, J. A. and Worthington, L. V., 1982. The flux and mixing rates of AntarcticBottom Water within the North Atlantic. J. Geophys. Res., 87, 7903-7924.Whitfield, M. and Turner, D. R., 1987. In: Aquatic surface chemistry. Wiley, New York,pp 459-494.Wilhelmy, S. A. S. and Flegal, A. R., 1991. Trace element distributions in coastal watersalong the US-Mexican boundary: Relative contributions of natural processes vs.anthropogenic inputs. Mar. Chem., 33, 371-392.Yeats, P. A. and Bewers, J. M., 1985. Manganese in the western North Atlantic Ocean.Mar. Chem., 17, No. 2, 255-263.Yeats, P. A. and Campbell, J., 1983. Nickel, copper, cadmium and zinc in the north westAtlantic Ocean. Mar. Chem., 12, 43-58.Yeats, P., 1988. Nickel, copper, cadmium and zinc distribution at the fram 3 and ceasarice camps. Ocean. Acta, 11, 383-388.104

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0061143/manifest

Comment

Related Items