Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Biogeochemistry of dissolved gallium and lead isotopes in the northeast Pacific and western Arctic Oceans McAlister, Jason 2015

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

Item Metadata

Download

Media
24-ubc_2015_september_mcalister_jason.pdf [ 4.89MB ]
Metadata
JSON: 24-1.0166278.json
JSON-LD: 24-1.0166278-ld.json
RDF/XML (Pretty): 24-1.0166278-rdf.xml
RDF/JSON: 24-1.0166278-rdf.json
Turtle: 24-1.0166278-turtle.txt
N-Triples: 24-1.0166278-rdf-ntriples.txt
Original Record: 24-1.0166278-source.json
Full Text
24-1.0166278-fulltext.txt
Citation
24-1.0166278.ris

Full Text

BIOGEOCHEMISTRY OF DISSOLVED GALLIUM AND LEAD ISOTOPES IN THE NORTHEAST PACIFIC AND WESTERN ARCTIC OCEANS by  Jason McAlister  B.Sc., San Diego State University, 2003 M.Sc., University of Nebraska – Lincoln, 2006  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Oceanography)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  May 2015 © Jason McAlister, 2015   ii Abstract  This thesis presents novel applications of dissolved Ga and Pb isotope ratios as oceanographic tracers of biogeochemical cycling and trace metal input mechanisms in the northeast Pacific Ocean, the Columbia River plume, and the Western Arctic Ocean.  Conclusions of this thesis are based on the first multi-year transect of dissolved Ga in the northeast Pacific Ocean along Line P, the first dissolved Ga concentrations reported in the Arctic Ocean, and the first Line P transect of Pb isotope ratios.  This thesis introduces 5 new tracer applications contributing to oceanographic research, dissolved Ga traces: 1) advective inputs of trace metals in the northeast Pacific, 2) Pacific and Atlantic source waters in the western Arctic, and 3) plume transport of the Columbia River, Ga/Pb traces: 4) fluvial freshwater inputs, and Pb isotope ratios trace: 5) geographically and temporally distinct trace metal inputs to the northeast Pacific.  Dissolved Ga is positively correlated with the variable spice, interpreted to trace advective trace metal inputs from the California Undercurrent and North Pacific Intermediate Waters bisecting the Line P transect.  This thesis identifies the Ga/Pb ratio as a tracer of an interface along Line P of fluvial trace metal inputs.  Isotopic ratios of Pb trace eolian Asiatic and fluvial North American sources, identifying a front of differential trace metal inputs along Line P.  Dissolved Ga is identified as a conservative tracer of Pacific waters in the Arctic and applies the conservative nature of dissolved Ga to interpretation of calculated rates of nitrogen fixation within shallow waters of the Beaufort Gyre, impacting phosphate concentrations and ecosystems in both the Arctic and North Atlantic Oceans.  Finally, dissolved Ga traces conservative and non-conservative behaviour under upwelling and downwelling conditions associated with the dynamic Columbia River plume.  This thesis contributes the five tracer applications described above to interpretation of trace metal inputs and advection in the northeast Pacific Ocean and source waters and nitrogen and phosphate coupling in the western Arctic Ocean.      iii Preface All of the work presented in this thesis was conducted at the University of British Columbia.  I was the lead investigator, responsible for all major areas of concept formation, data collection and analysis, as well as manuscript composition.  Chapter 3 and 4 of this thesis are prepared for submission to peer reviewed journals.  Chapter 5 has been submitted for publication and is currently in review (McAlister, J.A. and Orians, K.J. Dissolved gallium in the Beaufort Sea of the Western Arctic Ocean: A GEOTRACES cruise in the International Polar Year).  Chapter 6 has been published (McAlister, J.A.. and Orians, K.J., 2012. Calculation of river-seawater endmembers and differential trace metal scavenging in the Columbia River plume. Estuarine Coastal and Shelf Science. 99, 31-41).  Orians. K.J. was the supervisory author on this project and was involved throughout the project and manuscript edits.      iv Table of Contents  Abstract .......................................................................................................................................... ii Preface ........................................................................................................................................... iii Table of Contents ......................................................................................................................... iv List of Tables ............................................................................................................................... xii List of Figures ............................................................................................................................. xiii List of Symbols ...........................................................................................................................xxv List of Abbreviations .............................................................................................................. xxvii Acknowledgements .................................................................................................................. xxix Dedication ................................................................................................................................. xxxi Chapter 1: Introduction ................................................................................................................1 1.1 Trace metal oceanography .............................................................................................. 1 1.2 Trace metal profiles ........................................................................................................ 3 1.2.1 Conservative profiles .............................................................................................. 4 1.2.2 Nutrient type trace metals ....................................................................................... 4 1.2.3 Scavenged-type trace metals ................................................................................... 5 1.3 Trace metals as tracers .................................................................................................... 5 1.4 Physical and biological controls on trace metals ............................................................ 7 1.5 Trace metal ratios ............................................................................................................ 8 1.5.1 Hydroxide speciation .............................................................................................. 9 1.5.2 Trace metal ratios: Example ................................................................................... 9   v 1.6 Gallium (Ga) ................................................................................................................. 11 1.6.1 Dissolved Ga: Sources and sinks .......................................................................... 11 1.6.2 Dissolved Ga: Pacific Ocean ................................................................................ 12 1.6.3 Dissolved Ga: Atlantic Ocean ............................................................................... 12 1.6.4 Dissolved Ga: Arctic Ocean.................................................................................. 13 1.6.5 Dissolved Ga: Freshwater ..................................................................................... 13 1.6.6 Dissolved Ga: Advantages .................................................................................... 14 1.7 Lead (Pb) isotopes......................................................................................................... 15 1.7.1 Radiogenic Pb isotopes ......................................................................................... 16 1.7.2 Dissolved Pb in the ocean: Sources and sinks ...................................................... 17 1.7.3 Pb isotopes: Atlantic Ocean .................................................................................. 17 1.7.4 Pb isotopes: Pacific Ocean .................................................................................... 18 1.7.5 Pb isotopes: Advantages ....................................................................................... 18 1.8 Thesis study areas and tracer application...................................................................... 19 1.8.1 Northeast Pacific – Line P and Ocean Station Papa (OSP) .................................. 19 1.8.1.1 Study area.......................................................................................................... 19 1.8.1.2 Line P hydrography........................................................................................... 20 1.8.1.3 HNLC ................................................................................................................ 22 1.8.1.4 Tracer applications to Line P: Ga, Ga/Pb and Pb isotopes ............................... 22 1.8.2 Arctic Ocean – Beaufort Sea ................................................................................. 23 1.8.2.1 Study area.......................................................................................................... 23 1.8.2.2 Arctic hydrography ........................................................................................... 24   vi 1.8.2.3 Nitrogen cycle ................................................................................................... 25 1.8.2.4 Tracer application to Arctic: Ga ....................................................................... 26 1.8.3 Columbia River ..................................................................................................... 26 1.8.3.1 Study area.......................................................................................................... 26 1.8.3.2 Columbia River hydrography ........................................................................... 27 1.8.3.3 Conservative and non-conservative mixing ...................................................... 28 1.8.3.4 Tracer application to Columbia River plume: Ga ............................................. 28 1.8.3.5 Tracer application to Columbia River plume: Zr .............................................. 28 1.9 Thesis aims and contributions:...................................................................................... 29 1.9.1 Aim 1: Describe biogeochemical cycling of dissolved Ga in the northeast Pacific Ocean along Line P (Chapter 3)............................................................................................ 29 1.9.1.1 Aim 1: Contribution .......................................................................................... 29 1.9.2 Aim 2: Investigate the Ga/Pb ratio as an oceanographic tracer (Chapter 3) ......... 30 1.9.2.1 Aim 2: Contribution .......................................................................................... 30 1.9.3 Aim 3: Apply Pb isotopic ratios to identify sources along Line P (Chapter 4) .... 30 1.9.3.1 Aim 3: Contribution .......................................................................................... 30 1.9.4 Aim 4: Report the first dissolved Ga profiles in the Arctic Ocean (Chapter 5) .... 30 1.9.4.1 Aim 4: Contribution .......................................................................................... 30 1.9.5 Aim 5: Identify sources of elevated trace metal concentrations within the Columbia River plume (Chapter 6)....................................................................................... 31 1.9.5.1 Aim 5: Contribution .......................................................................................... 31 1.9.6 Chapter 7: Thesis conclusions and synthesis ........................................................ 31   vii Chapter 2: Methods .....................................................................................................................33 2.1 Sampling ....................................................................................................................... 33 2.1.1 Clean sample collection ........................................................................................ 33 2.1.2 Sampling depth selection ...................................................................................... 34 2.2 Sample analysis ............................................................................................................. 35 2.2.1 Trace metal clean laboratories .............................................................................. 35 2.2.2 Sample processing ................................................................................................ 35 2.2.3 Instrumental analysis ............................................................................................ 36 2.2.4 Data Analysis ........................................................................................................ 37 2.3 Methods: Ga .................................................................................................................. 39 2.3.1 Mg coprecipitation: Ga ......................................................................................... 39 2.3.2 ICP-MS: Ga .......................................................................................................... 39 2.3.3 Isotope dilution: Ga............................................................................................... 40 2.3.4 Method modifications ........................................................................................... 40 2.3.5 Method evaluation: Ga .......................................................................................... 42 2.4 Methods: Pb isotopes .................................................................................................... 45 2.4.1 Sample processing: Pb isotopes ............................................................................ 45 2.4.2 Pb isotopes: ICP-MS ............................................................................................. 46 2.4.3 Data handling: Pb isotopes.................................................................................... 46 2.4.4 Method evaluation: Pb isotopes ............................................................................ 47 2.5 Methods conclusions ..................................................................................................... 50 Chapter 3: Dissolved Gallium in the Northeast Pacific Ocean ................................................52   viii 3.1 Synopsis ........................................................................................................................ 52 3.2 Introduction ................................................................................................................... 52 3.3 Methods......................................................................................................................... 56 3.4 Results ........................................................................................................................... 58 3.5 Discussion ..................................................................................................................... 64 3.5.1 Dissolved Ga: Shallow .......................................................................................... 64 3.5.2 Dissolved Ga: 150m – 200m local maxima .......................................................... 65 3.5.3 Geographic spice variability: North Pacific spice interface ................................. 70 3.5.4 Temporal spice variability: Line P ........................................................................ 71 3.5.5 Dissolved Ga: Deep north Pacific ......................................................................... 75 3.5.6 Ga/Pb ratio ............................................................................................................ 76 3.5.7 Dissolved Ga: Eddy influence .............................................................................. 79 3.6 Conclusion .................................................................................................................... 80 Chapter 4: Geographic and temporal variability of Pb isotopes in the northeast Pacific Ocean .............................................................................................................................................81 4.1 Synopsis ........................................................................................................................ 81 4.2 Introduction ................................................................................................................... 82 4.3 Methods......................................................................................................................... 84 4.4 Results ........................................................................................................................... 85 4.5 Discussion ..................................................................................................................... 90 4.5.1 Temporal differentiation ....................................................................................... 90 4.5.2 Geographic source differentiation......................................................................... 93   ix 4.5.2.1 Pb sources from Asia ........................................................................................ 95 4.5.2.2 Pb sources from western North America .......................................................... 97 4.5.3 Physical inputs of Pb: Eolian and fluvial ............................................................ 103 4.5.4 Implications......................................................................................................... 106 4.6 Conclusion .................................................................................................................. 109 Chapter 5: Dissolved Ga in the Beaufort Sea of the western Arctic Ocean .........................111 5.1 Synopsis ...................................................................................................................... 111 5.2 Introduction ................................................................................................................. 112 5.3 Methods....................................................................................................................... 114 5.4 Results and discussion ................................................................................................ 116 5.4.1 Dissolved Ga in the Beaufort Sea: Canadian Basin ............................................ 118 5.4.2 Dissolved Ga in the Beaufort Sea: Shelf............................................................. 119 5.4.3 Dissolved Ga: Conservative Pacific tracer ......................................................... 121 5.4.4 Dissolved Ga: Application to nitrogen fixation in the Beaufort Gyre ................ 122 5.4.5 Dissolved Ga: Comparison with dissolved Al .................................................... 127 5.5 Conclusion .................................................................................................................. 129 Chapter 6: Calculation of river-seawater endmembers and differential trace metal scavenging in the Columbia River plume ................................................................................131 6.1 Synopsis ...................................................................................................................... 131 6.2 Introduction ................................................................................................................. 132 6.3 Methods....................................................................................................................... 134 6.4 Results ......................................................................................................................... 135   x 6.4.1 Trace metal concentrations ................................................................................. 135 6.4.2 Hydrography ....................................................................................................... 136 6.4.3 Model development ............................................................................................ 138 6.4.4 Model application ............................................................................................... 141 6.4.5 Model results ....................................................................................................... 144 6.4.6 Model results: Ga ................................................................................................ 144 6.4.7 Model results: Zr ................................................................................................. 149 6.5 Discussion ................................................................................................................... 152 6.6 Conclusion .................................................................................................................. 158 Chapter 7: Thesis conclusion ....................................................................................................160 7.1 Trace metal sources: Line P synthesis ........................................................................ 160 7.1.1 Input 1: Fluvial .................................................................................................... 160 7.1.2 Input 2: North American source ......................................................................... 161 7.1.3 Input 3: Asian source .......................................................................................... 163 7.1.4 Input 4: Advective– California Undercurrent ..................................................... 164 7.1.5 Input 5: Advective– North Pacific Intermediate Water ...................................... 164 7.1.6 Trace metal sources: Future research .................................................................. 165 7.2 Pacific and Atlantic sources: Western Arctic ............................................................. 166 7.2.1 Pacific and Atlantic sources: Future research ..................................................... 167 7.3 Nitrogen fixation: Western Arctic .............................................................................. 167 7.3.1 Nitrogen fixation: Future research ...................................................................... 168 7.4 River plume sources and sinks: Columbia River ........................................................ 169   xi 7.4.1 River plume sources and sinks: Future research ................................................. 169 7.5 Thesis conclusions ...................................................................................................... 170 References ...................................................................................................................................171    xii List of Tables  Table 2.1. Replicate samples of dissolved Ga from seawater: Northeast Pacific Ocean (Line P) 44 Table 2.2. Replicate samples of dissolved Ga from seawater: Arctic Ocean (Beaufort Sea) ....... 45 Table 2.3. Isotopic Pb ratios of the GDI reference standard ......................................................... 47 Table 2.4. Isotopic Pb ratios of the GSI reference standard ......................................................... 47 Table 3.1. Concentrations of dissolved Ga (pmol kg-1) measured along Line P in August 2010. 59 Table 3.2. Concentrations of dissolved Ga (pmol kg-1) measured along Line P in August 2011 and P26 in 2012.  Selection of sample depths represent both historical consistency as well as targeted sampling of features such as isopycnals, resulting in sampling depths specific to individual stations across the transect. .......................................................................................... 60 Table 4.1. 206Pb/207Pb ratios and standard deviation (SD) at stations P26 – P4 across Line P ..... 87 Table 4.2. 208Pb/206Pb ratios and standard deviation (SD) at stations P26 – P4 across Line P ..... 87 Table 5.1. Dissolved Ga (pmol kg-1) measured in the Beaufort Sea of the western Arctic Ocean..................................................................................................................................................... 117 Table 6.1 Temperatures and salinity and Ga and Zr concentrations. .......................................... 135 Table 6.2 Coefficients failing criteria per 1000 bootstrap replicates .......................................... 144 Table 6.3 Regression values and calculated endmember results.  Concentrations of Ga (pmol kg-1) and Zr (pmol kg-1) represented by two significant figures. ..................................................... 146    xiii List of Figures  Figure 2.1: Profiles of dissolved Ga at OSP (50N, 145W) sampled in 1983 (Orians and Bruland 1988b) and 2012 (this work) demonstrate excellent agreement, supporting the accuracy of the method of dissolved Ga analysis utilized in this thesis................................................................. 43 Figure 2.2 a) Graph demonstrating the statistical resolution provided by the standard deviation of any two samples.  Contours indicate m, the difference between two mean values that can be statistically resolved given the standard deviation of samples i and j, indicated on the axis.  Plotted are the respective standard deviations of all sample pairs, indicating the difference in means that can be resolved between the two samples.  Blue symbols represent sample pairs that, if differing by the target resolution of m = 0.01, will be statistically resolved based on their respective SD values.  Red symbols represent sample pairs that, if differing by the target resolution of m = 0.01, would not be statistically resolved based on their respective SD values b). Histogram displaying the number of sample comparisons plotted in panel a) corresponding to each resolution bin, >98% of all possible sample comparisons can be resolved at the target resolution of m = 0.01. ............................................................................................................... 49 Figure 3.1: Map of the northeast Pacific Ocean indicating stations sampled for dissolved Ga, arrow indicates source and direction of an eddy present in August 2010 .................................... 56 Figure 3.2: Profiles of dissolved Ga at stations P4 – P26 across Line P in 2010 (triangles) and 2011 (circles).  Note extended depth scale to 2000m in 2011.  Inter-annual comparison of profiles in the upper 400m is shown in Figure 3.3 ....................................................................... 61   xiv Figure 3.3: Annual comparison of dissolved Ga profiles in 2010 (triangles) and 2011 (circles) as a function of both depth and density.  To provide a reference in the density plots, the local dissolved Ga maxima are shown as a filled symbol. .................................................................... 62 Figure 3.4: Comparison of dissolved Ga profiles along Line P in a. 2010 (triangle) and b. 2011 (circle).  Symbol colours representing stations are consistent between the two plots, symbol shape indicating year is consistent with previous plots. ............................................................... 63 Figure 3.5: Full depth profile of dissolved Ga at P26 sampled in 2012 (this work) and 1983 (Orians and Bruland 1988b).......................................................................................................... 64 Figure 3.6: Spice calculated along Line P from CTD data in a. 2010 and b. 2011, contours indicate density ().  Filled symbols reference depths of dissolved Ga maxima. ....................... 66 Figure 3.7: Dissolved Ga and spice potted at stations P16, P20, and P26 along Line P at 150m and 200m in 2010 and 2011, showing the decrease in spice and dissolved Ga at P26 relative to stations P20 and P16. .................................................................................................................... 68 Figure 3.8: a. Correlation of dissolved Ga and spice at P16, P20, and P26 at 150m (filled symbols) and 200m (open symbols), in 2010 (triangles) and 2011 (circles), linear regression results in a slope of 11.5, intercept of 13.2, r2 = 0.84, p < 0.001.  b. Note: regression line in panel a. is presented in panel b. for reference and extrapolated (dashed line).  Concentrations of dissolved Ga at P4 and P12 150m are lower than predicted by the correlation in a., potentially due to increased scavenging at this more productive location.  At station P12 at 200m in 2010 (open green triangle) anonymously spicy waters could result from the presence of the observed surface eddy. ................................................................................................................................. 69   xv Figure 3.9: Spice calculated at 50°N across the north Pacific from Argo float data.  Plotting as a function of density reveals differential temperature and salinity characteristics of isopycnal surfaces across the north Pacific.  Triangles represent density of the local maximum concentrations of dissolved Ga at 150m at stations P4 – P26 demonstrating the interface in spice between P26 and P20 and the extension of spice values observed at P26 across the north western Pacific. .......................................................................................................................................... 71 Figure 3.10: Spice calculated along Line P from CTD data in a) 2009, b) 2010, c) 2011, and d) 2012.  Horizontal dashed line at  = 26.5 kg m-3 and depth contours provide reference showing dynamic spice as a function of longitude across Line P as well as interannual variation. ........... 73 Figure 3.11: Spice values along Line P at  = 26.5 kg m-3, years 2010 and 2011 in black show higher spice values at P16 and P20 relative to 2009 and 2012 in grey.  All years show a decrease in spice approaching P26. ............................................................................................................. 74 Figure 3.12: Ga / Pb ratio at stations P4 – P26 across Line P.  Note log depth scale to emphasize shallow dynamics associated with the 30m mixed layer depth. ................................................... 77 Figure 3.13: Ga/Pb ratio plotted as a function of salinity, elevated Ga/Pb ratios at low salinity support Ga/Pb as a tracer of fluvial freshwater sources ................................................................ 78 Figure 3.14: Concentrations of dissolved Ga at P12 in 2010 and 2011, note log depth scale to emphasize shallow water dynamics.  Elevated concentrations in 2010 of dissolved Ga as a function of a. depth and b. density are interpreted to result from the presence of an eddy. ......... 79 Figure 4.1. Pb isotope results reported from Stations P4 – P26 along Line P in the northeast Pacific Ocean.  Additionally, sample sites referenced in the Discussion of reported Pb isotope   xvi results from the north and south of Vancouver Island, a coastal fjord, and the Nass River are indicated for reference. ................................................................................................................. 84 Figure 4.2. Profiles of 206Pb/207Pb at stations P4 – P26 across the Line P transect and isotope - isotope plots of 206Pb/207Pb vs. 208Pb/206Pb.  Horizontal dashed lines in profiles represent summer mixed layer depth at P4 – P20, and winter mixed layer at P26.  Dashed lines in 206Pb/207Pb vs. 208Pb/206Pb plots provide reference to orientate data.  Triangles represent samples associated with the summer mixed layer at stations P4 – P20 and the winter mixed layer at P26. ....................... 88 Figure 4.3. Contoured section of d / dz along Line P in August 2010, warm colours indicating a density gradient represent the base of the summer mixed layer ................................................ 91 Figure 4.4.  a. Contoured section of d / dz along Line P in February 2010, warm colours indicating a density gradient represent the base of the winter mixed layer, b. profile of sq at 50°N 145°W (OSP) in February 2010 from Argo float data indicates a winter mixed layer depth of 75 – 100m. ......................................................................................................................................... 92 Figure 4.5. Comparison of isotopic ratios of Pb measured in the summer mixed layer (P4 – P20) and winter mixed layer (P26).  Higher 206Pb/207Pb values associated with the upper left quadrant are from stations P4 – P20, and represent more radiogenic dissolved Pb ratios relative to P26 in the lower right quadrant.  Note that symbol colour identifying stations here are consistent with Figure 4.2.  Interpretation of Figure 4.6, Figure 4.7, and Figure 4.8 is aided by application of a colour scheme specific to those figures. ....................................................................................... 94 Figure 4.6. a. Isotopic ratios of Pb collected from filtered air sampled from locations in Asia at >40°N (red circles, top half filled) (see also map in b.), and <40°N (red circles, bottom half filled) (Bollhofer and Rosman 2001), and enclosed by the red ellipse, a time series study   xvii demonstrating temporal variability (Wang et al. 2006) is represented by red dashes.  Also plotted are isotopic ratios of dissolved Pb within the surface ocean (Gallon et al. 2010) (see dashed blue ellipse in a. and map in b.), indicated by red circles outlined in blue, and samples from 50, 100, 200m depths at 30°N 140°W (red diamond outlined in blue) (Wu et al. 2010), ocean samples are indicated by the dashed blue ellipse.  P26 samples measured in this work (red inverted triangles in a. and b.) are indicated by the black ellipse in a. and are enclosed within the red ellipse, supporting an Asian source of Pb at P26.  Green circles along the coast of North America plot proximal to stations P4 – P20 (green triangles) in the upper left quadrant (green) and will be discussed in section 4.5.2.2. .......................................................................................................... 96 Figure 4.7. a. Isotopic ratios of Pb measured in lichen (Simonetti, et al. 2003) across western North America (see map in b.) are indicated by squares and plot within both the upper left and lower right quadrant.  P4 – P20 samples measured in this work (green triangles) identified by the black ellipse in the upper left quadrant are enclosed by the green ellipse indicating lichen samples plotting in the upper left quadrant.  See text for additional interpretation and discussion........................................................................................................................................................ 99 Figure 4.8. a. Pb isotope ratios of filtered air samples (Bollhofer and Rosman 2002) and dust collected from the air (Preciado, et al. 2007) within coastal lower BC and oysters (Shiel, et al. 2012) are indicated on the map in b.  All but two of these samples from lower BC plot within the green ellipse containing stations P4 – P12.  The remaining two samples suggest anthropogenic sources from Victoria.  c. Isotopic ratios of terrestrial samples: rock samples from Vancouver Island (diamonds) (Greene et al. 2009) and coastal BC waters draining the terrestrial watershed (circles) (Stukas, et al. 1999) (see also map in b).  Terrestrially derived samples are more   xviii radiogenic than values measured along Line P, thus relatively small amounts of terrestrial Pb would be required to influence Pb isotope ratios at P4 and P12 (green triangles). .................... 102 Figure 4.9. Temperature-salinity of the summer mixed layer at stations across Line P in August 2010 reveal four distinct regions identified by cool, saline waters indicative of upwelling (blue), warmer lower salinity waters (green) containing P12, followed by P16 and P20 within stations of increasing salinity (red), and P26 within waters of the decreasing temperature (black). ........... 103 Figure 4.10. Temperature and salinity of the mixed layer along the Line P transect in August 2010 demonstrates the longitudinal extent of the four regions identified in the TS plot, upwelling is clearly demonstrated along the coast (blue), followed by low salinity waters along the remainder of the eastern portion of the transect (green), a sharp salinity gradient occurs approaching the edge of the upwelling Alaska Gyre (gyre), and a decreasing temperature gradient upon entering the Alaska Gyre (black). ........................................................................ 104 Figure 4.11. Mixed layer 206Pb/207Pb ratios measured across Line P.  Small grey circles show all samples from within the mixed layer.  Coloured circles corresponding to regions described in Figure 4.10 and Figure 4.11 and represent the average and the cumulative error of all measurements within the mixed layer at each station.  Note the 206Pb/207Pb scale is reversed to provide consistency with Figure 4.14. ........................................................................................ 105 Figure 4.12. Fluvial sources to Line P traced by Ga/Pb within the mixed layer demonstrate a gradient between P12 and P16, and transition zone along P16 and P20 towards lower values at P26.  Note log scale on y axis to assist visualization of full data range. .................................... 106 Figure 4.13. Concentrations of NO3 (triangles),Chl a (circles), and NO3 / Chl a ratio (squares) combine with colours based on T-S characteristics (Figure 4.9 and Figure 4.10) to identify   xix regions of NO3 utilization (blue), NO3 limitation (green), followed by increasing concentrations of underutilized NO3 resulting in HNLC conditions approaching P26 (black) with transition toward HNLC conditions at P16 and P20 (red). ......................................................................... 107 Figure 4.14. Values of the NO3/Chl a ratio, representing a measure of HNLC character, demonstrate similar patterns observed for 206Pb/207Pb, non-HNLC conditions are coincident with North American fluvial sources, increasing HNLC character at P16 and P20 corresponds to a shift in 206Pb/207Pb approaching the edge of the Alaska Gyre, and rapidly increasing HNLC character entering the Alaska Gyre is matched by the gradient in 206Pb/207Pb. .......................... 108 Figure 4.15. HNLC character, indicated by NO3 / Chl a, correlated with mixed layer 206Pb/207Pb resulting in R2 = 0.93, p < 0.001. ................................................................................................ 109 Figure 5.1  Arctic Study Area: Beaufort Sea a) Study area in the Beaufort Sea of the Arctic Ocean, circles indicate sampling stations b) trace metal sampling performed at stations L1 – L3 and S4, stations S1 – S2 provide transmissometry data and water mass identification utilized in interpreting results. ..................................................................................................................... 114 Figure 5.2  Profiles of temperature, salinity, and O2 at station L1.  Open circles represent trace metal sampling depths, T and O2 minima indicate PWW and the thermocline delineates cool Pacific waters (<~150m) from relatively warm Atlantic waters (>~350m).  Note log scale for depth to enhance visualization of Pacific water structure. .......................................................... 115 Figure 5.3  Profiles of dissolved Ga at stations L1 – L3, log depth scale show lower concentrations through Pacific waters, transitioning to higher concentrations associated with Atlantic waters. ........................................................................................................................... 118   xx Figure 5.4  Dissolved Ga concentrations at L1 – L3 demonstrate consistent profiles across the five basin stations.  Dashed lines indicate ranges of dissolved Ga in the Pacific (Orians and Bruland 1988b) and dash-dot-dot lines represent concentrations of dissolved Ga reported in the north Atlantic (Shiller 1998). ...................................................................................................... 119 Figure 5.5  a. Temperature and dissolved Ga profiles at L1 and S4, L1 represents the similar profiles of the L stations.  Dissolved Ga and temperature are higher at S4 relative to L1 in the upper 25m.  Temperatures are similar through the thermocline at S4 and L1, yet increasing Ga concentrations are not observed at S4. b. Transmittance at shelf stations S1, S1.1, S1.2, and S2 (Fig. 1) is lower and more variable than the nearly vertical profile at L1. ................................. 120 Figure 5.6  Dissolved Ga, temperature, and N* at L1.  Dissolved Ga is conservative through Pacific waters while temperature and N* show minima associated with Pacific Winter Waters...................................................................................................................................................... 122 Figure 5.7  Nitrate (a.) and phosphate (b.) profiles, green represents nitrate deficient waters, blue PWW, and red Atlantic waters; c) NO3 vs. PO4, PWW are described by a linear fit to data indicated by blue symbols (R2 = 0.98), dashed line represents Atlantic sourced waters (Jones, et al. 1998), d) positive N** values within nitrate deficient waters suggest N fixation. ................ 124 Figure 5.8  a) Calculated rates of nitrogen fixation (mmol N m-3 yr-1) at basin stations, b) integrated rates of nitrogen fixation (mmol N m-2 yr-1) from 7.5m – 40m, representing sampling depths to ~1% light level. ........................................................................................................... 125 Figure 5.9  Profiles of dissolved Ga (station L1) and Al, a) as a linear function of depth, b) plotted as a log depth axis.  Al Central Arctic (Moore. 1989) , Al Beaufort Sea (Giesbrecht, et al. 2013), Al Eastern Arctic (Middag et al. 2009). .......................................................................... 128   xxi Figure 6.1  Hydrography. a. Map indicating stations along transects T7, T8, and T9; location of NOAA buoy DESW1 is indicated. b. T-S diagram indicating hydrographic conditions along the sampling transects, trace metal sampling stations indicated by t:s, where t and s indicate transect and station number, respectively. c. Salinity along transects as a function of distance (km) from the coast, trace metal sampling stations along transects are indicated. d. Temperature along transects T7, T8, and T9. ............................................................................................................ 136 Figure 6.2 Demonstration of the model determining the conservative mixing line based on sampling within the mixing zone.  Regression of elemental concentrations as a function of salinity produces a function for calculation of the river and seawater endmembers, resulting in bounding of the mixing zone and construction of the conservative mixing line, see text for details. ......................................................................................................................................... 140 Figure 6.3 a. Regression of Ga concentrations as a function of salinity, stations 6 and 7 (T9) are indicated as filled symbols.  b. Residual values calculated from regression in panel a.  c. Conservative mixing line constructed based on endmembers calculated from the model, long and short dashed lines indicate 68% and 95% confidence intervals, respectively, based on bootstrap analysis.  d. full range of mixing zone. ....................................................................................... 146 Figure 6.4 a. Relative percent contributions of river and seawater endmembers to dissolved Ga concentrations as a function of salinity, given scavenged removal. b. concentrations of Ga from river and seawater endmembers and total dissolved Ga, note log scale. c. comparison of river-sourced Ga as a function of salinity during conservative mixing and scavenged removal, dotted lines indicate salinity at which 50% of Ga contributed by the river remains. d. percent Ga removed by scavenged removal relative to dilution by conservative mixing. ............................ 148   xxii Figure 6.5 . a. Regression of Zr concentrations as a function of salinity.  Stations 1 and 2 (T8) (black squares) indicate upwelled sources and Station 9 (T8) (grey square) demonstrates concentrations suggestive of the open ocean, outside the defined mixing zone. b. Residual values calculated from regressive fit. c. Conservative mixing line constructed from model, long and short dashed lines indicate 68% and 95% confidence intervals based on bootstrap analysis. Stations 6 and 7 (T9) are consistent with the conservative mixing line.  d. full range of mixing zone. ............................................................................................................................................ 150 Figure 6.6 a. Relative contributions of river and seawater endmembers to dissolved Zr concentrations as a function of salinity based on scavenged removal; b. concentrations of Zr from river and seawater endmembers and total dissolved Zr, note log scale; c. comparison of river-sourced Zr as a function of salinity during conservative mixing and scavenged removal, dotted lines indicate salinity at which 50% of Zr contributed by the river remains; d. percent Zr removed by scavenged removal relative to dilution by conservative mixing. ............................ 152 Figure 6.7 Stick vectors indicating wind direction and magnitude (NOAA National Data Buoy Center, Destruction Island buoy, location indicated in Figure 6.2).  Downwelling winds point up, blowing from the south toward the north.  Upwelling winds point down, blowing from north towards the south.  Sampling along transects 7, 8, and 9 are indicated by T7, T8, and T9. ....... 154 Figure 6.8 Summary figure indicated processes associated with sampling of new and aged plume waters along T9.  a. Dotted arrows indicate northward wind direction and plume transport during the interval June 26-28, encompassed in panel d by the dotted lines, solid arrow indicates Ekman transport.  b. Dashed arrows correspond to plume transport during June 29 – July 5, indicated in panel d. by the dashed box.  c. temperature – salinity along transect 9, stations numbered as in   xxiii Figure 6.2.  d. detail of winds presented in Figure 9, T7, T8, and T9 indicate respective transect sampling dates. ............................................................................................................................ 155 Figure 6.9 a. Si(OH)4 and b. NO3- concentrations along Transect 9 as a function of salinity.  Station numbers where trace metal samples were collected are indicated along the top of the plot...................................................................................................................................................... 157 Figure 7.1: Line P schematic indicating regions of fluvial inputs based on Ga/Pb identified in this thesis.  Note log base10 depth scale to emphasize mixed layer dynamics across the transect.  Fluvial inputs present at P4 and P12 within the summer mixed layer (30m) extend to 100m at P12 resulting from eddy transport............................................................................................... 161 Figure 7.2: Isotopic ratios of Pb identify sources from North America at P4 – P20 within the summer mixed layer (30m).  Fluvial inputs are identified within the summer mixed layer at P4 and P12 by Ga/Pb, with advection within the summer mixed layer to stations P16 and P20 resulting in a transition zone across the transect (see Figure 4.11 and Figure 4.12). ................. 162 Figure 7.3: Eolian inputs from Asia are identified at P26 based on Pb isotopes and associated with winter mixed layer depths to 75m, contrasting with inputs from North American at P4 – P20 confined to the summer mixed layer. .......................................................................................... 163 Figure 7.4: Spicy water at 150 – 200m at P20 and P16 correlate with local maxima of dissolved Ga concentrations, lower concentrations of dissolved Ga at P26 associated with lower spice NPIW.  Construction of this figure summarizes the trace metal inputs identified from results of this thesis and tracer applications of dissolved Ga and Pb isotopes developed in this thesis. .... 165   xxiv Figure 7.5: Dissolved Ga provides a tracer of Pacific source waters (blue) to the Arctic, with concentrations increasing through the thermocline (green), and higher concentrations observed in Atlantic source waters to the Arctic. ........................................................................................... 167 Figure 7.6: Intensity of red colour indicates decreasing concentrations of excess phosphate as a result of in-situ nitrogen fixation during transport of nitrate deficient waters across the Arctic and returning to the north Atlantic, providing a schematic summary representation of implications of nitrogen fixation identified in this thesis. ................................................................................... 168 Figure 7.7: Graphical representation of results from development of endmember mixing model and interpretation of upwelling and downwelling favouring winds associated with the Columbia River plume, and resulting concentrations of trace metals, based on dissolved Ga, exhibiting conservative mixing and scavenged behaviour. .......................................................................... 169    xxv List of Symbols Al: Aluminum Ba: Barium Cd: Cadmium Cu: Copper Fe: Iron Ga: Gallium GF-AAS: Graphite Furnace Atomic Absorption Spectroscopy  HCl: Hydrochloric acid Hf: Hafnium HNO3: Nitric acid m: Atomic mass  Mn: Manganese Mo: Molybdenum N2: Nitrogen gas NO3: Nitrate O2: Oxygen gas Pb: Lead PO4: Phosphate Th: Thorium U: Uranium z: Ion charge   xxvi Zn: Zinc Zr: Zirconium : Potential density – 1000kg/m3    xxvii List of Abbreviations mol: micro mol AAIW: Antarctic Intermediate Water AOU: Apparent Oxygen Utilization C: Celsius cps: counts per second CTD: Conductivity, Temperature, Depth instrument  CUC: California Under Current ENSO: El Niño Southern Oscillation fmol: femto mol GDI: GEOTRACES Deep Isotope (Reference Standard) GSI: GEOTRACES Shallow Isotope (Reference Standard) HEPA: High Efficiency Particulate Air (filter) HNLC: High Nutrient Low Chlorophyll ICP-MS: Inductively Coupled Plasma Mass Spectrometer ICP: Inductively Coupled Plasma kg: kilogram LDPE: Low Density Polyethylene LR: Low Resolution m: meter mmol: milli mol MR: Medium Resolution   xxviii MS: Mass Spectrometer NADW: North Atlantic Deep Water nmol: nano mol NPDW: North Pacific Deep Water NPIW: North Pacific Intermediate Water OSP: Ocean Station Papa (equivalent to station P26) pmol: pico mol PSW: Pacific Summer Water PWW: Pacific Winter Water r: radius RSD: Relative Standard Deviation S: Salinity SD: Standard Deviation T: Temperaure    xxix Acknowledgements My experience as a graduate student at UBC has been rewarding beyond expectation and is made possible by many people in my personal and professional life.     My advisor Dr. Kristin Orians made the research in this thesis possible by allowing me to explore new research directions, always being available for fruitful discussions, and providing the perfect balance of autonomy and mentorship. Thank you so much for providing opportunities to present my research at conferences, allowing me to become a part of the international research community.    I would like to thank my committee for their support and interest in my research during my program.  Dr. Maite Maldonado, your passion for research is infectious and I am beyond appreciative for the time you spent to improve the writing and clarity of this thesis.  Dr. Roger Francois, your thoughtful consideration of ideas is absolutely inspiring and your candour is always refreshing and appreciated.  Dr. Jay Cullen, thank you for wonderful training opportunities and field experience, your trust was very motivating.   Special thanks also to Dr. Dominique Weis for providing the opportunity to work on state of the art instruments and for being a wonderful example of how to be a successful leader.  Thank you also to Maureen Soon, an invaluable resource and always available to assist with the absolute most joyful of demeanours.    Thank you again to Dr. Kristin Orians for making your research group a family, and a part of your family.  Thank you to all members of the Orians research group, you have all assisted in important ways to my research.  Special thanks to Jeffrey Charters for always being available for great discussion of research ideas, and also for being my liaison to Canadian culture.  Thank you also to Ania Posacka for always being excited to share research results and for your authentic enthusiasm.      xxx Thank you to Dr. David Semeniuk for excellent scientific conversations and sharing common perspectives.  Thank you to Brett Gilley for illuminating an entirely new path of interest in teaching and learning in the sciences.    I would like to thank the Captains and crew of the CCGS Amundsen and CCGS John P. Tully for all their work to help make the research in this thesis possible.  Thank you also to Marie Robert and all the amazing researchers at the Institute of Ocean Science for maintaining the Line P program and for great camaraderie and hospitality at sea and in Sidney.    Thank you to the entire Department of Earth, Ocean, and Atmospheric Sciences faculty and staff for a wonderful graduate education experience and all the fellow graduate students for great friendships.  Special thanks to the department for providing a myriad of opportunities for involvement and professional development.    I would like to express sincere gratitude to the University of British Columbia, the Department of Earth, Ocean, and Atmospheric Sciences, and NSERC for financial support.    To my wife Jennifer Geddes-McAlister, you are wonderful and wise, you are the most fabulous partner and scientist, and I learn from you everyday.  Thank you for everything you do, we make ourselves an amazing life together.  Thank you to my parents for providing the perfect foundation for me to build an amazing life.  Finally, to our daughter Hazel, you take our family to a whole new level of amazing!     xxxi Dedication My Family    1 Chapter 1: Introduction 1.1 Trace metal oceanography   Representing an area covering ~70% of the Earth’s surface, to an average depth of 4000m, studying the oceans is quite simply an enormous task.  Early chemical analysis of seawater identified principal ions composing ocean salinity at mmol kg-1 (10-3mol kg-1) concentrations.  Biological study revealed that macroscopic fish and whales are dependent on an ecologic web based on microscopic photosynthetic phytoplankton, with phytoplankton growth dependent on the foundational elements of nitrogen and phosphorus, present at mol kg-1 (10-6mol kg-1) concentrations.  Until ~30 years ago, however, the majority of the elements in the periodic table were unknown in the ocean, present at nmol kg-1 (10-9mol kg-1), pmol kg-1 (10-12mol kg-1), and even fmol kg-1 (10-15mol kg-1) concentrations.  While trace metal oceanography represents a comparatively young field, recognition of the importance of trace metals to understanding global biogeochemical oceanic processes is demonstrated by the massive international multi-decadal research efforts of the GEOTRACES program (Henderson, et al. 2007) studying trace metals in the world’s oceans.  This thesis presents new applications of the trace metal gallium (Ga) in the northeast Pacific Ocean, western Arctic Ocean, and the Columbia River plume, and isotopic ratios of the trace metal lead (Pb) in the northeast Pacific, advancing both the study of trace metals in the ocean and the field of oceanography.    Oceanography combines physical, chemical, and biological data to study ocean circulation (Whitney and Freeland 1999; McLaughlin, et al. 2004; Hickey, et al. 2010), biogeochemical cycling of elements (Morel and Price 2003; Saito, et al. 2003), and investigate climate change and paleoclimate (Doney, et al. 2014; Henderson 2002).  While temperature and salinity observations provided early insight into ocean circulation, and nitrate and phosphate nutrient concentrations and oxygen measurements assisted in interpreting controls on biological production, trace metals and isotopes provide key new parameters useful in the integrative study of biogeochemical cycling in the ocean.  Trace metals have proven to be vital to elucidating enigmatic regions of the ocean wherein an abundance of nitrate is not utilized by phytoplankton,   2 resulting in low chlorophyll concentrations.  Trace metal research demonstrated the control of Fe on phytoplankton productivity in such High Nutrient Low Chlorophyll (HNLC) regions in the ocean sourced from eolian deposition and upwelling (Martin 1990; de Baar, et al. 2005).  In addition, trace metals have been shown to provide proxies of these source processes of Fe, such as Al for eolian sources (Orians and Bruland 1986; Measures and Vink 2000) and Mn for sedimentary sources (Jones and Murray 1985).    In addition to biological requirements, trace metals provide important applications as proxies and tracers, as illustrated by Al and Mn above.  Utility of an oceanographic tracer is dependent on spatial and temporal heterogeneity.  Trace metal concentrations in the ocean are by definition low, potentially one billion times lower than common ions such as sodium and chloride. However, the heterogeneous distribution of trace elements, in contrast to the nearly uniform conservative distribution of the major ions as a function of salinity, trace oceanographic sources, sinks, physical processes, and biogeochemical dynamics.  Utilization of multiple tracers, including trace metals, is essential for identification and explanation of oceanographic processes.  Results of this thesis present new tracer applications of dissolved Ga and Pb isotopes in the ocean, summarized below.    This thesis interprets measurements of dissolved Ga concentrations and radiogenic Pb isotopic ratios in the Northeast Pacific Ocean along Line P, the Western Arctic Ocean, and the Columbia River Plume.  This thesis presents the first reports of dissolved Ga in the Arctic Ocean, providing a conservative tracer of Pacific waters to the Arctic Ocean.  Dissolved Ga concentrations in the Northeast Pacific Ocean are correlated with the variable spice to identify advective sources associated with North Pacific Intermediate Waters and the California Undercurrent to stations along Line P.  Additionally, dissolved Ga is combined with dissolved Pb to introduce the Ga/Pb ratio as a proxy of fluvial sources to Line P.  Radiogenic Pb isotopes are interpreted to indicate eolian sources to Line P, with Asiatic sources to the western terminus Ocean Station Papa (OSP), while isotopic Pb ratios consistent with coastal Canadian sources are observed along the remainder of the transect.  Finally, this thesis presents a plume mixing model demonstrating both conservative mixing and scavenged removal of dissolved Ga within the Columbia River plume.    3 Results of this thesis present opportunities for future research investigating controls on HNLC conditions in the northeast Pacific, tracing source waters in the Arctic Ocean, and investigating trace metal sources from rivers.    In the following sections, introduction to trace metals as tracers begins with description of characteristic trace metal profiles, leading to discussion of physical and biological controls on trace metal profiles, and concluding with the pairing of trace elements and the additional utility that trace metal ratios serve as oceanographic tracers.  Next, the two elements studied in this thesis, Ga and Pb, will be introduced, followed by introduction of study areas of this thesis, and the application of selected elements specific to each study area.  Finally, the aims of this thesis will be presented along with brief accompanying statements of the contributions of this thesis to the field of oceanography.  1.2 Trace metal profiles  Trace metal oceanography required two primary advances: analytical methods and instrumentation, and clean sample collection and processing.  Low concentrations of trace metals in seawater require sensitive analytical methods, and while advances in analytical chemistry provided opportunities to quantify low concentrations of trace metals found in the ocean, clean sampling techniques and sample processing was necessary to ensure that results accurately reflected the ocean sample and not simply contamination.  Analytical methods and sampling techniques employed in this thesis will be described in Chapter 2.    Following the advancements of clean sampling and analytical techniques, a new era in oceanography was born as previously unknown profiles of elements throughout the periodic table began to be reported.  Profiles of trace metal in the ocean were shown to exhibit a variety of characteristics based on the element and sampling environment considered.  Synthesis of trace metal studies in the oceans has allowed for classification of trace metals into three common categories: conservative, nutrient-type, and scavenged.  While elements may be grouped into   4 categories, based on observed profiles and known chemical and biological interactions, archetypal elemental profiles are overprinted by the local physical and biological source and sink characteristics of a given area, resulting in the heterogeneous distributions benefitting tracer applications.   1.2.1 Conservative profiles  Elements possessing a long oceanographic residence time produce profiles of elemental concentration as a function of depth that are conservative, typically varying only as a function of salinity.  Long residence times of conservative elements result from slow removal processes from the water column.  While conservative elements comprise a bulk of ions resulting in the saline character of the ocean, trace elements may also demonstrate conservative profiles.  Given that the conservative nature of an element is dependent on residence time and not concentration, the element molybdenum remains at trace concentrations due to a low crustal abundance, and therefore a low input of Mo to the ocean, yet slow removal of Mo from the ocean results in a long residence time and therefore a largely conservative profile.  While Mo represents an example of a conservative trace metal, many trace metals demonstrate heterogeneous distributions in the ocean, described by nutrient type and scavenged type profiles.    1.2.2 Nutrient type trace metals    Profiles of nutrient-type trace metals result from assimilation by phytoplankton and release back to the water column following consumption and respiration.  Profiles of nutrient type trace metals possess behaviour analogous to phosphate and nitrate profiles, as demonstrated for Cd (Bruland, et al. 1978a), or to silicic acid as demonstrated by Zn (Bruland, et al. 1978b), and similarly concentrations increase from the Atlantic to Pacific Oceans during Meridional Overturning Circulation.  Beyond strictly biological requirements by phytoplankton, nutrient type trace metals can serve as tracers as well, such as the cadmium (Cd) phosphorous (P) ratio   5 (Cd/P) proving insight into past nutrient regimes and Fe availability (Cullen, et al. 2003; Elderfield and Rickaby 2000).    1.2.3 Scavenged-type trace metals  Scavenged removal on particle surfaces influences trace metal concentrations.  For example, Al concentrations are higher in the Atlantic than in the north Pacific due to greater atmospheric deposition of Al to the Atlantic.  Therefore, in contrast to the increasing concentration of nutrient type trace metals along thermohaline circulation, distributions of scavenged type trace metals are more dependent on local sources and sinks.  Surface sources include eolian or fluvial freshwater sources, while bottom sources can result from sediment flux and shelf processes, while low concentrations through the remainder of the water column are maintained by chemical scavenging reactions.  Scavenged type trace metals, including Ga and Pb studied in this thesis, are useful as tracers due to heterogeneous distributions, a function of particle reactivity resulting in short oceanic residence times with respect to ocean circulation time scales of 1000 years.  Scavenged-type trace metal profiles in the ocean are classically illustrated by dissolved aluminum, wherein despite the high crustal abundance of Al, low concentrations of dissolved Al in the oceans are maintained by particulate scavenging, while sources are identified by concentration gradients at the ocean surface and approaching the bottom sediments (Orians and Bruland 1985).    1.3 Trace metals as tracers  Ocean temperature and salinity provide tracers of physical processes of upwelling, gyre circulation, and water mass advection and are combined to produce plots of temperature vs salinity (T-S plots), assisting in differentiating ocean waters.  Chemical tracers in the ocean include both the individual measured concentrations of constituents such as O2, NO3, and PO4, as well as combinations of multiple chemical and/or physical properties creating additional chemical tracers such as Apparent Oxygen Utilization (AOU), preformed nitrate (Abell, et al.   6 2005), N* (N* = NO3 – 16xPO4 + 2.9) (Gruber and Sarmiento 1997), and NO (NO = 9xNO3 + O2) (Broecker 1974).    Trace metals are additional chemical tracers of oceanographic processes.  Trace metal profiles reflect local source and sink functions in the ocean.  For example, eolian inputs to the ocean are identified by Al (Orians and Bruland 1985; Han, et al. 2008; Measures and Vink 2000), inputs of trace metals following sedimentary reduction are traced by Mn (Jones and Murray 1985; Lam and Bishop 2008; Aguilar-Islas, et al. 2007), and freshwater sources are tracked by Ba (Guay, et al. 2009).  Isotopes provide additional opportunities for identification of source terms, as hydrothermal sources are identified utilizing primordial 3He (Lupton 1998; Ruth, et al. 2000), groundwater inputs with 226Ra (Moore 1996), and continental sources and margin exchange suggested by Nd isotopes (Jeandel, et al. 2007; Lacan and Jeandel 2005).  Sinks are identified by scavenged type trace metal profiles, including quantification of flux based on depletion of the scavenged elemental isotope 234Th and 230Th relative to the conservative distribution of the parent isotopes 238U and 234U (Bacon and Anderson 1982).  Chemical and physical tracers work in concert to elucidate oceanographic processes.  For example, an organic matter flux calculated by 234Th resulting from a phytoplankton bloom initiated by upwelled nutrients, identified by temperature and salinity, could be fuelled by Fe from the sediment, traced by Mn.  Application of trace metals as complimentary tracers of oceanographic processes arises from global and local physical controls on trace metal profiles, such as global thermohaline circulation, gyre structure, and eddy formation, and biologic influence of both trace metal utilization and scavenging.  Physical and biological effects on trace metal profiles will each be discussed followed by an example demonstrating the tandem effect of these two processes.      7 1.4 Physical and biological controls on trace metals   Trace metals receive much deserved attention for their role in mediating biology in the ocean.  Physical processes also govern trace metal distributions, resulting in applications of trace metals as tracers of physical processes such as water mass source, structure, and advection.   Distributions of heat, nutrients, and observed trace metal distributions are dependent on physical processes that govern water transport.  For example, physical processes associated with North Atlantic Deep Water (NADW) formation and advection influence profiles of dissolved Al in the Atlantic Ocean.  Advection of NADW results in elevated concentrations of Al at depth in the Atlantic as eolian sources of Al in surface waters are entrained during NADW formation (Measures and Edmond 1992).  In contrast to the north Atlantic, deep water formation in the Southern Ocean is low in dissolved Al due to reduced eolian transport in the southern hemisphere (Shiller 1998).  Additionally, removal of Al occurs on particulate surfaces during thermohaline transport.  Fractionation of dissolved Al between the Atlantic and Pacific is therefore a product of physical processes of water mass formation and transport, local sources of Al, and chemical properties of Al with particle scavenging.  Combination of these factors gives Al a short residence time with respect to thermohaline circulation and concentrations of Al are therefore 1-2 orders of magnitude lower in the north Pacific than in the north Atlantic.    Additional physical controls on dissolved Al in the ocean include eddies. Within the low Al concentration waters of the North Pacific, elevated Al concentrations are observed associated with eddies (Brown, et al. 2012) as coastal waters of relatively high Al concentration are transported offshore.  Eddies also transport Fe to offshore areas (Johnson, et al. 2005) potentially enhancing phytoplankton production, leading to increased scavenging rates of Al.  Concentrations of dissolved Al within eddies relative to surrounding waters therefore illustrate a balance between local sources and sinks imparted by physical and biologic controls.     Trace metals are key in understanding ocean biology, as trace metals are utilized within catalytic enzymatic centers essential to phytoplankton growth in the ocean.  While iron (Fe) has been   8 famously demonstrated to allow nutrient utilization under HNLC conditions, additional trace metals have been shown to elicit control on phytoplankton, including Cu (Semeniuk, et al. 2009), Cd (Lane and Morel 2000), and Zn (Jakuba, et al. 2012).  Biological influence on observed trace metal profiles and dynamics occur for both nutrient-type trace metals, as a consequence of active biological uptake, and scavenged-type trace metals as a function of particulate surface area in the water column, associated with phytoplankton bloom conditions (Moran and Moore 1988).    A summative example of linked physical and biological controls on scavenged-type trace metal distributions is provided by surface concentrations of dissolved Ga along a transect extending from the subtropical to subpolar North Pacific (Orians and Bruland 1988b).  Suppression of nutrient availability imposed by the anticyclonic subtropical gyre impedes biological production as the gyre retains a shallow mixed layer of warm water. Stratification and inhibition of nutrient sources results in reduced rates of Ga scavenging, allowing scavenged-type trace metals to accumulate within the shallow mixed layer (Bruland, et al. 1994).  Alternatively, cyclonic rotation of the subpolar gyre supports biological production, leading to decreased dissolved Ga concentrations due to increased scavenging on biotic particulates. Dissolved gallium concentrations therefore change as a function of latitude from 28°N – 55°N in the northeast Pacific reflecting physical, biological, and chemical controls on this scavenged-type trace metal.    1.5 Trace metal ratios  Ratios of two trace metal concentrations can provide tracer utility beyond that of each individual element.  Changes in the elemental ratio results from physical and biogeochemical processes leading to differential scavenging.  Hydroxide speciation influences differential scavenging of the elemental pairs Al and Ga, and zirconium (Zr) and hafnium (Hf).  Examples of Al/Ga and Zr/Hf ratio applications are followed by introduction of the Ga/Pb ratio presented in this thesis.       9 1.5.1 Hydroxide speciation  Removal of trace metals from the ocean via scavenging is influenced by the ability of an element to be hydrolyzed, which is a function of their electron attracting ability as determined by the charge to radius (z2/r) ratio (Turner, et al. 1981).  Increasing charge and/or decreasing radius results in a higher polarizing power of the element and a correspondingly greater ability to attract and pull electrons from potential ligand donors.  Water acts as a ligand and surrounds cationic elements within hydration spheres as the negative dipole of the oxygen atom within water molecules orientate themselves toward the positive charge of the cation.  Given a sufficiently high z2/r value, elemental hydration will be followed by hydrolysis, resulting in hydroxide speciation of the element (ie. Ga(OH)4).  With increasing polarizing power, full hydrolysis occurs, resulting in formation of oxyacids, as in molybdenum, MoO42-.  Low electron binding energy precludes hydrolysis (Li 1991), resulting in the presence of free hydrated ions in solution, as observed for sodium, Na+(H2O)n.    Hydroxide speciation of an element affects elemental reactivity and scavenging onto particulates.  Reaction and scavenging of an element on a negatively charged particulate surface occurs to a greater extent for a neutrally charged hydroxide species than for a negatively charged species.  For example, Ga and Al are chemically similar elements, residing in the same column of the periodic table.  Speciation of Ga is dominated by the anionic form, Ga(OH)4-, while Al speciation results in ~1:1 ratio of the neutral, Al(OH)30, and anionic Al(OH)4- species (Orians and Bruland 1988a).  Greater proportion of the neutral species of Al results in increased scavenging relative to Ga and a shorter residence time of dissolved Al in the ocean relative to dissolved Ga.    1.5.2 Trace metal ratios: Example  Ratios of trace metals highlight differential residence times of two elements over time scales ranging from global ocean circulation to phytoplankton blooms.  Evolution of elemental ratios as   10 a function of time, depth, and location, therefore lends insight into geochemical behaviour of these elements in the ocean and provides utilization of elemental ratios as oceanographic tracers.    As discussed above, hydroxide speciation leads to a shorter residence time of Al relative to Ga, resulting in a positive correlation of Ga/Al ratios and chlorophyll concentrations (Shiller and Bairamadgi 2006), as the more reactive Al is removed preferentially to Ga onto biogenic particles.  Similar to Ga and Al, zirconium (Zr) and hafnium (Hf) are an elemental pair displaying a slightly greater relative abundance of the more reactive neutral species Hf(OH)50 relative to Zr(OH)50, providing a potential mechanism for Zr/Hf fractionation (McKelvey 1994).  Correspondingly, terrestrial Zr/Hf ratios of 60-120 fractionate to ratios of 180 – 250 in the Atlantic Ocean and evolve to ratios of 400 in the north Pacific during thermohaline ocean circulation on 1000 year time scales (McKelvey 1994), as the longer residence time of dissolved Zr results in an increasing Zr/Hf ratio with time as dissolved Hf is scavenged more rapidly from the ocean.  Fractionation of Zr/Hf on shorter time scales and associated with freshwater sources to the ocean is observed along the Hudson River as terrestrial Zr/Hf values of ~80 increase to 140 as a function of preferential removal of dissolved Hf within regions of elevated particle abundance within the river and as salinity increases towards the ocean (Godfrey, et al. 2008).  In addition to scavenged removal, fractionation of trace metal ratios can occur as a function of differential weathering, modifying the Ga/Al ratio during freshwater transport (Shiller and Frilot 1996).    This work introduces the Ga/Pb ratio as an oceanographic tracer.  Both Ga and Pb are sourced to the ocean by atmospheric deposition and rivers.  Fluvial freshwater inputs of trace metals to the oceans are altered within plume environments (Boyle, et al. 1977).  Fluvial inputs of trace metals specify freshwaters from rivers.  While both Ga and Pb are scavenged during transport from freshwater rivers to the ocean, high rates of Pb removed occur with estuarine environment upon metal oxides (Nelson, et al. 1999).  This thesis presents the Ga/Pb ratio as a tracer of freshwater fluvial inputs to the ocean, wherein an elevated Ga/Pb ratio results from preferential removal of Pb during fluvial transport.      11 1.6 Gallium (Ga)  This thesis presents the following advancements in utilization of dissolved Ga as an oceanographic tracer: 1) the first dissolved Ga transect along Line P identifies correlation of dissolved Ga concentrations with spice, identifiying distinct water sources as a function of temperature and salinity along surfaces of equal density and interpreted to represent advective sources of trace metals along Line P, 2) the first dissolved Ga profiles in the Arctic Ocean provide a conservative tracer of Pacific source waters, 3) introduction of the Ga/Pb ratio provides a proxy of fluvial freshwater inputs to the ocean, and 4) development of a river plume endmember mixing model demonstrates local wind influence on observed concentrations of dissolved Ga associated with the Columbia River plume.  Relevant to these studies, the ocean biogeochemistry of Ga will be introduced, beginning with discussion of sources and sinks, describing known behaviour of dissolved Ga in the Pacific Ocean, Atlantic Ocean, Arctic Ocean, and freshwaters, and then concluding with a summary of advantages of utilizing dissolved Ga as an oceanographic tracer.    1.6.1 Dissolved Ga: Sources and sinks  Profiles of dissolved Ga are consistent with scavenged type behaviour analogous to Al (Orians and Bruland 1988a).  Sources of Ga to the ocean include atmospheric deposition (Orians and Bruland 1988a; Shiller 1998) and fluvial freshwaters (Shiller and Frilot 1996, McAlister and Orians 2012).  Eolian deposition represents a common source mechanism of trace metals to the ocean, such as Al and Fe.  Eolian inputs represent atmospheric sources of trace metals and may include both wet and dry deposition. Gallium, however, provides a complementary tracer to the more rapidly scavenged Al and Fe, which is biologically utilized, redox sensitive, and scavenged.  Similarly, while freshwaters represent a source of both Al and Ga to the oceans, greater particle reactivity of Al results in high rates of scavenging of Al within river plumes during transport to the ocean.  While scavenging of Ga also occurs through river plumes,   12 elevated concentrations of Ga representing a fluvial freshwater source are observed at increasing salinities relative to Al (Lanthier 1999).    1.6.2 Dissolved Ga: Pacific Ocean  Dissolved Ga in the north Pacific Ocean displays a range of concentrations.  Regionally dynamic concentrations within shallow waters are observed, with concentrations of ~10 – 15 pmol kg-1 in the central subtropical gyre and only 3 – 5 pmol kg-1 in the eastern and western north Pacific associated with the sub polar gyre (Shiller and Bairamadgi 2006; Orians and Bruland 1988b). Scavenging of dissolved Ga with depth results in concentration minima (Orians and Bruland 1988b), relative to higher concentrations at shallower depths associated with surface sources.  At depths greater than the concentration minima, concentrations of dissolved Ga increase steadily toward the ocean bottom, reaching concentrations of 20 – 30 pmol kg-1 (Shiller and Bairamadgi 2006; Orians and Bruland 1988b).  Increasing concentrations can be the result of dissolved Ga sources from the sediments (Orians and Bruland 1988b) or from the Southern Ocean, transported by thermohaline circulation  (Shiller and Bairamadgi 2006).    Measurements of dissolved Ga at OSP (50°N, 145°W) identified a local concentration maxima at ~150m (Orians and Bruland 1988b). Similar local maxima in dissolved Ga concentrations across the extent of Line P are reported in this thesis.  Description of the source mechanism of this feature in Chapter 3 identifies fronts across the north Pacific potentially impacting additional trace metals. Additionally, results of this thesis differentiate influence of fluvial and eolian sources of Ga to the north Pacific, providing a proxy for additional trace metal sources to the north Pacific.    1.6.3 Dissolved Ga: Atlantic Ocean  Atlantic Ocean profiles of dissolved Ga demonstrate unique characteristics relative to the Pacific.  Surface concentrations of dissolved Ga in the north Atlantic are 20 – 30 pmol kg-1 (Shiller 1998),   13 4 – 10 times higher than surface concentrations in the north Pacific.  Higher concentrations of Ga in the surface represent greater influence of eolian sources to the north Atlantic (Orians and Bruland 1988a; Shiller 1998; Measures and Vink 2000).  Contrasting concentrations in the north Atlantic and north Pacific demonstrate the ability of dissolved Ga to act as a tracer of regional processes.    In contrast to high eolian deposition to the north Atlantic, eolian deposition to the south Atlantic is lower given less relative land mass. Correspondingly, local minima in dissolved Ga concentration in the south Atlantic Ocean are interpreted to represent Antarctic Intermediate Water (AAIW) (Shiller and Bairamadgi 2006), therefore demonstrating dissolved Ga as a tracer of advected water masses.  This thesis identifies dissolved Ga as a tracer of advected water masses of Pacific and Atlantic origin to the western Arctic Ocean and interprets dissolved Ga concentrations as a tracer of advectively sourced trace metals to the northeast Pacific Ocean.    1.6.4 Dissolved Ga: Arctic Ocean  Profiles of dissolved Ga have not previously been reported in the Arctic Ocean.  Differential characteristics of dissolved Ga in the Pacific and Atlantic Oceans, however, elicits interest in dissolved Ga profiles in the Arctic given the presence of Pacific and Atlantic source waters to the Arctic Ocean.  Additional description of the unique oceanography of the Arctic Ocean will be discussed in section 1.8.2.2 below.  Opportunity therefore exists for dissolved Ga to provide a tracer of Pacific and Atlantic waters in the Arctic based on higher concentrations associated with Atlantic water inputs and lower concentrations of dissolved Ga in Pacific source waters.  This thesis presents the first profiles of dissolved Ga in the Arctic Ocean (Chapter 5).    1.6.5 Dissolved Ga: Freshwater   Dissolved Ga in terrestrial freshwaters transported to the ocean originates from weathering of rocks and soils.  Concentrations of dissolved Ga in freshwater vary widely, from 68 – 250 pmol   14 kg-1 in U.S. rivers (Shiller 1998), 1 – 78 pmol kg-1 in California streams (Shiller and Frilot 1996), and 11 – 7785 pmol kg-1 in the Amazon (Gaillardet et al. 2003).  Chemical behaviour of Ga influences dissolved Ga concentrations in freshwaters, and specifically differentiation is observed relative to Al.  For instance, while both Ga and Al are sourced to freshwaters from weathering reactions, the observed ratio of Ga/Al dissolved in river waters is higher than the Ga/Al of rocks (Shiller and Frilot 1996; Shiller 1988).  Both preferential dissolution of Ga to the dissolved phase (Shiller and Frilot 1996) and/or less relative scavenging of Ga from the dissolved phase (Shiller 1998) can result in the elevated Ga/Al ratios.    In addition to differential chemical control mechanisms within the confines of freshwater rivers and streams, mixing of freshwaters to the ocean within plume environments impact concentrations of trace metals (Shiller and Boyle 1987; Boyle, et al. 1977).  This thesis utilizes concentrations of dissolved Ga within the Columbia River plume to construct a river-ocean endmember mixing model.    1.6.6 Dissolved Ga: Advantages  Gallium possesses chemical characteristics that both promote investigation as an oceanographic tracer and are advantageous to the analytical measurement of dissolved Ga in seawater.  First, the residence time of dissolved Ga in the ocean, due to its scavenged type behaviour, is appropriate to identifying regional and local source and sink mechanisms.  Dissolved Ga can complement the more rapidly removed dissolved Al, providing a longer lived tracer of advective transport of trace metal sources to the ocean.  Second, the chemical behaviour of Ga allows application of the analytical method of magnesium coprecipitation, an advantageous method for measuring dissolved Ga in seawater based on minimal sample volume and time requirements.  Third, Ga has two naturally occurring isotopes and therefore Ga concentrations can be calculated by isotope dilution following measurement by Inductively Coupled Plasma Mass Spectrometry (ICP-MS).  Isotope dilution is advantageous as it provides an ideal internal standard during sample processing, as will be described in the methods chapter of this thesis (Chapter 2).  Fourth,   15 the low crustal abundance of Ga, relative to Al for instance, presents a lower risk of contamination for Ga during seawater sampling, analytical processing, and analysis.  Finally, while a solid foundational knowledge of dissolved Ga in the ocean is known, much remains to be discovered.  For example, in addition to presenting the first reports of dissolved Ga in the Arctic Ocean, this thesis provides new applications of dissolved Ga to oceanography resulting from the first interannual transect of dissolved Ga and the first high resolution shallow transect, including 5 measurements in the upper 40m of the water column.  Specific applications of dissolved Ga relevant to each study area considered in this thesis will be discussed in sections 1.8.1.4, 1.8.2.4, and 1.8.3.4.    1.7 Lead (Pb) isotopes   Stable radiogenic isotope ratios of Pb trace sources and inputs of Pb to the environment (Komarek, et al. 2008) and Pb isotopes are applied to monitoring industrial processing (Shiel, et al. 2010) and remediation (Gulson, et al. 2012).  Additionally, Pb isotopes have been used to study and monitor anthropogenic Pb within the ocean (Gallon, et al. 2011; Wu and Boyle 1997a).  In addition to contaminant concerns of anthropogenic Pb to the ocean, Pb isotopes are applied within oceanography as water mass tracers (Veron, et al. 1999; Alleman, et al. 2001; Flegal, et al. 1989).  Determination of Pb isotopes in the ocean has been reported for depth profiles and surface transects in the Atlantic and Pacific Oceans (Wu, et al. 2010; Gallon, et al. 2011).  This work reports the first transect of Pb isotope depth profiles, along the Line P transect in the northeast Pacific Ocean.  Introduction of the radiogenic series producing stable Pb isotopes will be followed below by biogeochemical cycling of Pb in the oceans, describing sources and sinks of Pb within the ocean, applications of Pb isotopes within the Atlantic and Pacific Oceans, and concluding with advantages of Pb isotope analysis presented in this thesis.        16 1.7.1 Radiogenic Pb isotopes  Individual parent isotopes produce the radiogenic stable Pb isotopes 206Pb, 207Pb, and 208Pb.  Each decay series occurs over a range of time scales, discussed below.  Additionally, relative composition of the parent isotopes varies within different sources.  Unique isotopic ratios of 206Pb/207Pb and 208Pb/206Pb are therefore produced given the geographic, geologic, and temporal history of a sample.  Isotopic ratios of Pb can thus be applied analogous to a ‘fingerprint’, providing a diagnostic tool to identify Pb sources and input mechanisms.    While both 206Pb and 207Pb are derived from uranium, 238U and 235U, respectively, the time scale of the two radiogenic series is very different.  Alpha decay of 235U to 231Th with a half-life of 704 million years initiates production of 207Pb.  While 704 million years is a very long time, ~6.5 half lives have passed in the 4.6 billion year history of the earth, and thus only ~1% of 235U exists today.  Conversely, 206Pb begins with decay of 238U to 234Th, a reaction with a half-life of 4.5 billion years.  Given the 4.6 billion year age of the earth, only approximately half of the 238U has decayed to 206Pb and therefore production of 206Pb continues.  While decay of U leads to production of 206Pb and 207Pb, 208Pb results from the radioactive decay of 232Th.  Initiation of the 208Pb sequence, the alpha decay of 232Th to 228Ra, has a half-life of 14 billion years, and therefore only ~15% of 232Th has decayed over 4.6 billion years.  As 206Pb and 208Pb are the daughters of 238U and 232Th, respectively, the U/Th ratio of a host material will influence the 208Pb/206Pb ratio.    Isotopic ratios of Pb measured in samples can therefore trace the geographic source of the Pb input. For instance, comparatively high or low 206Pb/207Pb ratios can be used to differentiate Pb samples.  Higher 206Pb/207Pb ratios are associated with younger, more geologically recent sources as 238U remains in the sample, while little 235U will be available for decay.  The amount of 207Pb will therefore remain nearly constant while 238U decay to 206Pb results in higher 206Pb/207Pb ratios.  In contrast, in an older geologic setting more 235U has decayed to 207Pb resulting in a comparatively low 206Pb/207Pb ratio.  In this thesis, Pb isotope ratios measured in seawater will be compared to reported ratios from Asia and North America.  Geographic sources of Pb inputs   17 are identified by comparatively low 206Pb/207Pb representing inputs of Pb from Asia, while higher 206Pb/207Pb indicate Pb from North America.    1.7.2 Dissolved Pb in the ocean: Sources and sinks  Inputs of Pb to the ocean by atmospheric deposition and fluvial freshwaters are common to other trace metals, and therefore Pb provides a proxy of trace metal inputs to the ocean.  Pb in the ocean represents both natural and anthropogenic sources, the latter originating from industrial use including both Pb added to gasoline as well as emissions from coal power plants.  Isotopic ratios of Pb can be applied to determining the source material (ie. gasoline, coal) or the source location (ie. Asia, North America) of Pb inputs.  This thesis focuses on identification of source location, as the intent is to differentiate Asian and North American Pb inputs along the Line P transect in the northeast Pacific Ocean.  Sinks of dissolved Pb from the ocean include particle reactivity and therefore Pb distributions reflect local source and sink processes, similar to dissolved Ga discussed above.  Scavenging of dissolved Pb occurs on the surface of particulate matter in the ocean and therefore scavenging increases during phytoplankton blooms.  Additionally, scavenging occurs upon the surfaces of oxide minerals (Tebo, et al. 2004; Nelson, et al. 1999), observed as freshwater inputs of Pb to the ocean are modified by removal within plume environments.    1.7.3 Pb isotopes: Atlantic Ocean  Following the phaseout of Pb in gasoline, Pb isotopes traced the decrease in anthropogenic Pb inputs to the Atlantic as a function of time (Weiss, et al. 2003; Wu and Boyle 1997a).  Isotopes of Pb measured in the Atlantic have also been utilized to show isotopic equilibrium between dissolved and particulate Pb (Sherrell and Boyle 1992).  Isotopic signatures of dissolved Pb from the surface ocean can be transported vertically on the surface of sinking particles (Wu, et al. 2010).  Atlantic Pb isotope data also show the advective influence on Pb isotopes (Boyle, et al. 2014).  Finally, the mixed layer in the Sargasso Sea of the Atlantic Ocean has been shown to   18 capture distinct Pb isotopes signatures, relative to ratios at depths below the mixed layer (Veron, et al. 1993).  This thesis identifies unique Pb isotope signatures associated with the summer mixed layer, and previous winter mixed layer, along Line P in the northeast Pacific.    1.7.4 Pb isotopes: Pacific Ocean  Isotopic ratios of Pb reported in the Pacific Ocean have also been utilized to identify anthropogenic sources to the surface ocean.  Emissions from Asia contribute Pb to the western north Pacific surface ocean (Gallon, et al. 2011) and urban sources are identified in fluvial inputs to the Gulf of California (Soto-Jimenez and Flegal 2009).  Anthropogenic Pb ratios in the deep waters of the central north Pacific (Wu, et al. 2010) confirm vertical transport of Pb from surface sources, as thermohaline circulation is not rapid enough to source anthropogenic Pb to deep Pacific waters.  Physical oceanographic processes in the Pacific have also been traced by isotopic ratios of Pb, such as upwelling (Flegal, et al. 1989).  This thesis utilizes Pb isotopic signatures to differentiate eolian sources of Pb from Asia and fluvial freshwater sources from North America to the Line P transect in the northeast Pacific.    1.7.5 Pb isotopes: Advantages  Isotopic ratios of Pb provide an excellent environmental tracer based on distinct decay series radiogenic production of the stable Pb isotopes 206Pb, 207Pb, and 208Pb.  This thesis utilizes the analytical advantage provided by the Nu AttoM, discussed in section 2.4.2 to measure Pb isotope ratios in 50mL seawater samples, providing logistical advantages to sample collection and processing relative to previous methods utilizing sample volumes 2 – 50 times larger, as discussed in section 4.3.  Isotopic ratios of Pb measured in this thesis on the Nu AttoM ICP-MS are shown in section 2.4.4 to provide sufficient precision to resolve Pb isotopic ratios relevant to identification and interpretation of oceanographic features and processes.  An additional advantage of Pb isotopes is the wealth of existing reports of Pb isotope ratios from a wide variety   19 of samples from locations around the world, assisting identification of the geographic source of Pb isotopes to the northeast Pacific Ocean studied in this thesis.    1.8 Thesis study areas and tracer application   The study areas in this thesis, the northeast Pacific along Line P, the Beaufort Sea of the Arctic Ocean, and the Columbia River plume, each possess unique properties and research questions.  Dissolved Ga, the Ga/Pb ratio, and Pb isotopes are beneficial to different study areas and research questions given the physical, chemical, and biological controls on their distributions.  Hydrography and topics specific to each study area are discussed followed by specific tracer application of Ga, Ga/Pb, and Pb isotopes within each research area, leading to the development of the aims of this thesis presented in section 1.9.      1.8.1 Northeast Pacific – Line P and Ocean Station Papa (OSP) 1.8.1.1 Study area  Ocean Station Papa (OSP) and Line P represent one of the longest oceanographic time series (Freeland 2007).  Extending ~1500km off the coast of Vancouver Island, Line P is currently comprised of 26 stations, with station P26 equivalent to OSP (50°N 145°W).  Observations of decadal change in the northeast Pacific Ocean include decreases in O2 levels (Whitney, et al. 2007), atmospheric oscillations influencing temperature increases (Crawford, et al. 2007b), and shifts in nutrient availability (Pena and Varela. 2007).  Additionally, OSP, as a station within the VERTEX oceanographic program, assisted in establishing Fe as a key control in High Nutrient Low Chlorophyll (HNLC) regions (Martin and Fitzwater 1988).  Therefore inputs of Fe along Line P are important and have been identified from coastal inputs (Cullen, et al. 2009), eddies (Johnson, et al. 2005), volcanic (Hamme, et al. 2010), and atmospheric dust (Mahowald, et al. 2005) sources.      20 Trace metal time series data, however, is limited for Line P cruises.  Iron measurements conducted for the last decade by the Institute of Ocean Science allow comparison of multiple years of data (Nishioka, et al. 2001).  Recently additional trace metal concentrations have been determined along the full Line P transect at increasingly frequent intervals, including Cu (Semeniuk 2014, Posacka, et al. 2012), Cd  (Janssen, et al. 2014), Zn (Janssen, et al. 2014), Fe(II) (Schallenberg, et al. 2014), Pb (Charters 2011), Mn (Sim and Orians 2014), and Al (Cain 2014).   This thesis reports the first transect of dissolved Ga along Line P, including interannual comparison between 2010 and 2011, and interdecadal time series comparison at OSP based on reported data from sampling in 1983 (Orians and Bruland 1988b).  This thesis also reports the first transect of dissolved Pb isotopes along Line P.   1.8.1.2 Line P hydrography  Line P transects a dynamic oceanographic region of the north Pacific providing a natural laboratory to study HNLC regions of the ocean.  While HNLC areas also exist in the equatorial Pacific and southern Ocean, the Line P transect provides the most convenient geographic and logistical access to an HNLC region.  Along the Line P transect HNLC conditions vary temporally and geographically as regional processes influence nutrient availability.  Four processes influencing hydrographic conditions along Line P will be discussed below, the Alaskan gyre, upwelling along the coast, freshwater inputs, and transport of coastal waters into the northeast Pacific by eddies.    Counterclockwise rotation of the Alaskan gyre produces upwelling through divergence, creating a source of available nutrients.  Utilization of these nutrients by phytoplankton is hindered by low Fe supplies creating the anomalous HNLC conditions in this region.  The Alaska Gyre is situated along the western portion of the Line P transect and the greatest influence of the Alaskan gyre along Line P is seen at station P26.  Therefore as the Line P transect enters into the Alaska Gyre, concentrations of nitrate increase due to supply from the upwelling gyre.  Low concentrations of chlorophyll however indicate that phytoplankton are not able to utilize the   21 available nitrate.  A gradient in HNLC conditions is therefore established along Line P associated with the Alaska Gyre.    Upwelling occurs seasonally and episodically along the coastal portion of Line P when winds blow from north to south along the coast.  Nutrient sources are provided by Ekman pumping as surface waters transported offshore are replaced by upwelled waters.  Upwelling along the coast and upwelling associated with the Alaska Gyre both provide nutrient sources.  Therefore in the summer Line P can be bookended by regions of nutrient availability in the east and west.  High chlorophyll concentrations, however, are only observed along the eastern coastal portion of Line P, while iron limitation exists in the west associated with the Alaska Gyre.    Freshwater inputs along the eastern portion of the Line P transect produce a ‘Dilute Domain’ (Favorite, et al. 1976).  Freshwater inputs vary seasonally, with Vancouver Island rivers contributing freshwater from winter rainfall (Freeland, et al. 1984).  Additionally, summer freshwater inputs from the Fraser River to the Strait of Georgia are amplified by a 2 – 3 times greater outflow through the Strait of Juan de Fuca in the summer (Hickey, et al. 1991).  Additionally, freshwater is stored along the margin of the Juan de Fuca eddy (MacFadyen, et al. 2008), extending along the shelf break to the eastern portion of Line P.  Nitrate and phosphate inputs from the Fraser River are known to influence phytoplankton dynamics (Ware and Thompson 2005).  Trace metal inputs from fluvial freshwater sources will be investigated in Chapter 4 of this thesis.     Eddies also influence the northeast Pacific and hydrographic conditions along the Line P transect.  While ephemeral in nature, eddies are ubiquitous in both space and time (Crawford, et al. 2007a).  Eddies in the northeast Pacific form as anticyclonic, clockwise, rotating water masses resulting in convergence and downwelling toward their core.  Eddies are typically transported toward the east and south into the northeast Pacific.  Coastal waters trapped during eddy formation are therefore transported offshore, producing microcosms within the northeast Pacific of coastally derived nutrients potentially intersecting the Line P transect.  Satellite and hydrographic data identify the presence of an eddy along Line P during August 2010,   22 implications of the eddy on observed dissolved Ga concentrations are discussed in sections 3.5.6 and 3.5.7 of this thesis.    1.8.1.3 HNLC  Phytoplankton blooms require a nutrient source of phosphate (PO4) and nitrate (NO3), however, paradoxically large areas of the ocean possess an abundance of unutilized NO3 and PO4, resulting in regions described as High Nutrient Low Chlorophyll (HNLC).  Regions of the ocean displaying HNLC characteristics include the equatorial Pacific, the Southern Ocean, and the northeast Pacific.  These seemingly disparate regions share commonality in representing areas of upwelling within large regions of the open oceans.  Divergence created by opposing Corriolis forces across the equator, along sub Antarctic fronts, and cyclonic gyre rotation of the Alaskan gyre provide sources of upwelled nutrients.  Remote sensing, however, indicates that these large regions of the ocean chronically display low chlorophyll concentrations.  Given the presence of nutrients, the lack of accompanying phytoplankton growth encouraged iron addition experiments within the three HNLC regions, confirming that Fe results in phytoplankton blooms (de Baar, et al. 2005).  Given that phytoplankton growth in the HNLC northeast Pacific is mediated by trace metal availability, this thesis seeks to identify sources of trace metals to this region.    1.8.1.4 Tracer applications to Line P: Ga, Ga/Pb and Pb isotopes   Three tracers are applied in this thesis to the Line P study area, each possessing characteristics unique to the investigation of trace metal sources along Line P.  First, Dissolved Ga is applied as a tracer of advective trace metal sources.  Selection of Ga provides the opportunity to identify sources potentially masked by the rapid removal of dissolved Al, the biological uptake of Fe, and offers decoupling of the multiple sources of redox sensitive trace metals such as Mn.  Dissolved Ga distributions are compared with the oceanographic variable spice to investigate advective inputs of trace metals along Line P.  Spice identifies different temperature and salinity characteristics among waters of the same density (Flament 2002).  Along a given isopycnal   23 comparatively warm salty water is differentiated by a higher spice value from cooler fresher water along the same isopycnal.  Therefore advection of water masses from different locations can be identified by spice and correlation with dissolved Ga is demonstrated in Chapter 3.  Second, application of dissolved Ga and Pb as a combined tracer is investigated based on differential removal processes within freshwater sources to the ocean, providing a tracer of fluvial freshwater sources of trace metals along Line P.  Finally, isotopic ratios of dissolved Pb are applied in Chapter 4 to the identification of geographic sources of trace metals along Line P.  Dissolved Ga, the Ga/Pb ratio, and Pb isotopes are therefore applied in this thesis to an integrated study of advective, fluvial freshwater, and geographic sources of trace metals along Line P and the northeast Pacific Ocean.    1.8.2 Arctic Ocean – Beaufort Sea  1.8.2.1 Study area  Logistical requirements associated with Arctic sampling have caused the Arctic Ocean to be undersampled with respect to trace metals.  Dissolved Ga concentrations have not been reported in the Arctic Ocean and profiles of the chemically similar element Al in the Arctic (Middag, et al. 2009, Moore 1989) are different relative to observations in the Pacific (Orians and Bruland 1988b), Atlantic (Measures, et al. 1986), or Mediterranean (Hydes, et al. 1988) Oceans.  Relative to these other oceans, unique traits of the Beaufort Sea in the western Arctic Ocean include sea ice formation and melt, water sources from both the Pacific and Atlantic Oceans, and oscillating cycles of the Beaufort Gyre that store and release freshwater from the gyre.    This thesis identifies dissolved Ga as a tracer of Pacific and Atlantic waters in the Arctic Ocean.  Additionally, interpretation of dissolved Ga as a tracer of shelf waters suggests that calculations of nitrogen fixation presented in this thesis represent a feature of the Beaufort Gyre, with subsequent cyclical release of stored water from within the Gyre delivering a pulse of PO4 deficient water to the North Atlantic Ocean.     24 1.8.2.2 Arctic hydrography  While sea ice represents a defining characteristic of the Arctic Ocean, sea ice melt and formation is impacted by climate change, initiating change from ecosystems to economies (Barber, et al. 2008a).  Lower annual ice volume results from both earlier seasonal ice melt and later ice formation (Comiso, et al. 2008; Stroeve, et al. 2007).  Inputs to the Arctic from both Pacific (Shimada et al. 2006) and Atlantic waters threaten ice melt (Rudels, et al. 2004),  Additionally, warming Atlantic source waters (McLaughlin, et al. 2009) infiltrating further west into the Beaufort Sea (McLaughlin, et al. 2004) can upwell into shallow shelf waters of the Western Arctic (Woodgate, et al. 2005). Sea ice melt increases stratification and influences phytoplankton dynamics.  For example a limited nutrient regime results in decreased phytoplankton size (Li, et al. 2009).  Additionally, ice melt will influence carbon cycling in the Arctic (McGuire, et al. 2009).   Source inputs from both the Pacific and Atlantic Oceans to the Arctic Ocean link the world’s two largest oceans as transport of Pacific waters through the Arctic Ocean represents a return of fresh water lost from the Atlantic to the Pacific across Central America back to the Atlantic (Carmack 2007).  Sources of Pacific and Atlantic waters to the Arctic Ocean and Beaufort Sea demonstrate temporal and spatial variability, including decadal shifts in the intrusion of warm Atlantic waters (Carmack, et al. 1997; McLaughlin, et al. 2004) and dependency of Pacific water transport on the Arctic Oscillation (Steele, et al. 2004) and inflow along the Alaskan current (Woodgate, et al. 2006).  Atlantic inputs to the Arctic represent comparatively warmer and saltier waters relative to colder, fresher, and less dense Pacific waters.  Strong thermo- and haloclinic structure established by the divergent properties of Pacific and Atlantic source waters impede heat stored in the Atlantic layer of the Arctic from warming the surface and accelerating Arctic sea ice melt (Shimada, et al. 2005).  Tracing Pacific and Atlantic waters is therefore required as these sources from opposite sides of the globe meet within the Arctic Ocean.       25 Characterization of Atlantic and Pacific water contributions in the eastern Arctic have been determined given NO3 and PO4 relationships (Jones, et al. 1998).  Calculation of Pacific source contributions based on nutrient composition, however, can be obscured by denitrification (Bauch, et al. 2011). Seasonal denitrification is particularly relevant given the large shelf area of the Bering and Chukchi seas, therefore influencing NO3 dynamics within Pacific waters entering the Beaufort Sea.  Therefore, within the unique environment of the Arctic, exhibiting inter- and intra- annual variability in temperature and NO3, Ga will be investigated as a complimentary tracer of Pacific and Atlantic waters.    Pacific waters entering the western Arctic become trapped within the anticyclonic Beaufort Gyre (Proshutinsky, et al. 2002).  Storage of waters in the Beaufort Gyre continues until a weakening of the gyre circulation releases waters from the gyre resulting in a pulse of Pacific freshwater to the Atlantic (McPhee, et al. 2009).  Additionally, effects of nitrogen cycling identified in this thesis, introduced in the next section, accumulate during the anticyclonic residence time of the gyre and therefore represent potential shifts in nutrient sources to the north Atlantic when waters stored in the Beaufort Gyre are released.    1.8.2.3 Nitrogen cycle  Climate change in the Arctic, impacting earlier ice melt and later ice formation, modify the phytoplankton growing season (Tremblay, et al. 2006; Pabi, et al. 2008). While high concentrations of nitrate are present within the western Arctic Ocean at depths of ~120m, availability of these nutrients to phytoplankton in the sunlit surface is hindered by density gradients within the stratified Arctic (Tremblay, et al. 2008).  Given stratified hindrance to nitrate sources, are additional nitrogen sources available to allow utilization of increased light provided by reduced ice cover?   Nitrogen within the ocean cycles between forms of nitrogen as N2, reduced NH4 and fully oxidized NO3, with intermediates of N2O and NO2-, during processes of nitrogen fixation and respiration.  Within areas of low oxygen, nitrate is utilized as a respiratory oxidant during   26 denitrification, representing a loss of fixed nitrogen from the ocean cycle as nitrate is returned to N2.  Denitrification consumes NO3 while producing a source of PO4, therefore resulting in an excess of available PO4 relative to NO3.  Globally, the Arctic Ocean is a vital component in balancing the nitrogen cycle.  Pacific waters returning to the Atlantic through the Arctic Ocean bring excess phosphate resulting from cumulative effects of net global denitrification.  Nitrogen fixation in the north Atlantic utilizes the excess phosphate returned from the Pacific Ocean through the Arctic Ocean, balancing the global nitrogen cycle.    This thesis interprets an observed phosphate deficiency as representing nitrogen fixation within the Beaufort Gyre of the western Arctic Ocean and calculates rates of nitrogen fixation. Dissolved Ga concentrations in areas of nitrogen fixation are consistent with Pacific sources, supporting interpretation of nitrogen fixation occurring within Pacific waters contained within the Beaufort Gyre, as opposed to advective sources from the shelf.  In addition to representing a nitrogen source to the western Arctic Ocean, utilization of excess PO4 by nitrogen fixation in the Arctic Ocean potentially impacts north Atlantic ecosystems as less excess PO4 is returned to the north Atlantic.    1.8.2.4 Tracer application to Arctic: Ga  This thesis will determine the utility of Ga as a tracer of water masses in the Beaufort Sea of the Arctic Ocean.  Dissolved Ga represents an advantageous tracer of Pacific and Atlantic water sources in the Arctic Ocean given the divergent concentrations of Ga within the Atlantic and Pacific Oceans.    1.8.3 Columbia River 1.8.3.1 Study area  Rivers provide a source of trace metals to coastal ecosystems.  Coastal systems, while comprising a small portion of the global ocean, combine upwelling and fluvial nutrient inputs   27 resulting in highly productive ecosystems.  While considering the northeast Pacific Ocean and Line P in this thesis, it is therefore relevant to include the Columbia River, the largest freshwater source along the west coast of North America.  Indeed, the plume waters of the Columbia River can reach as far north as the Strait of Juan de Fuca (MacFadyen, et al. 2005) influencing the near-coast region of the Line P transect.    1.8.3.2 Columbia River hydrography  Dynamic Columbia River plume transport results from frequently shifting winds.  Therefore, spatially and temporally unique plume environments are produced during upwelling and downwelling conditions (Banas, et al. 2009; Hickey, et al. 2010; Berdeal, et al. 2002).  Columbia River water entering the Pacific may be transported to the south and offshore with winds from the north or transported north and alongshore with winds from the south (Hickey, et al. 2009).  Often, seasonal cycles may be considered wherein winds favouring upwelling occur in the summer and downwelling in the winter.  However, winds impacting upwelling and downwelling conditions shift on daily to hourly time scales. The Columbia River plume therefore represents a dynamic environment with frequent northward and southward shifts in surface waters (Hickey, et al. 2005).    Trace metals supplied from the Columbia River (Lippiatt, et al. 2010; Aguilar-Islas and Bruland 2006) may therefore help fuel phytoplankton blooms when upwelling conditions provide nutrients.  Alternatively, downwelling conditions preclude upwelled nutrients, however supply of trace metals by the Columbia River continues.  An effectively new source of trace metals to the coastal ocean may thus be considered as utilization of fluvial trace metals inputs is curtailed by downwelling conditions.  This thesis presents a new application of dissolved Ga to trace fluvial inputs of trace metals from the Columbia River during downwelling conditions.       28 1.8.3.3 Conservative and non-conservative mixing  River waters mixing with ocean waters may be predicted to result in conservative mixing of salinity.  Scavenged trace metal concentrations, however, decrease non-linearly along estuarine transects, resulting in observed concentrations below predictions based on conservative mixing as a function of salinity (Boyle, et al. 1977; Shiller and Boyle 1987).  Trace metal sources from the Columbia River are therefore dependent on particle dynamics influencing scavenging during mixing of the Columbia River and seawater endmembers.  Oceanographic sampling logistics do not always provide the opportunity to directly sample both a river endmember and a representative seawater endmember.  Can endmembers therefore be identified in the absence of direct sampling?  Chapter 6 of thesis presents a method of differentiated parabolic regression to calculate seawater and river endmembers.    1.8.3.4 Tracer application to Columbia River plume: Ga  Dissolved Ga is applied as a tracer of advected Columbia River plume waters during upwelling and downwelling conditions.  Scavenged type behaviour of Ga allows observation of non-conservative mixing between river and seawater endmembers while also possessing a residence time sufficient to identify conservative mixing.  Utilization of dissolved Ga as a tracer within the Columbia River plume allows development of the river-seawater endmember mixing model developed in Chapter 6.    1.8.3.5 Tracer application to Columbia River plume: Zr  While the focus of this thesis is on dissolved Ga, dissolved zirconium (Zr) is included in the published manuscript (McAlister and Orians 2012) comprising Chapter 6 of this thesis.  Following development of the river-seawater endmember mixing model based on interpretation of dissolved Ga concentrations, the method was applied to concentrations of dissolved Zr as well.  While Zr is not the focus of this thesis, results in Section 6.4.7: Model results: Zr, provides   29 a demonstration of application of the river-seawater endmember mixing model developed in Chapter 6.    1.9 Thesis aims and contributions:  Aims of the research conducted in this thesis are presented below and provide the foundation of the remaining chapters.  Additionally, the contributions of this thesis to the oceanographic research field associated with each Aim are summarized.  Results of this thesis provide new applications of Ga and Pb isotopes as oceanographic tracers and therefore promote inclusion of dissolved Ga and isotopic ratios of Pb as increasingly routine oceanographic trace metal measurements.    1.9.1 Aim 1: Describe biogeochemical cycling of dissolved Ga in the northeast Pacific Ocean along Line P (Chapter 3) 1.9.1.1 Aim 1: Contribution  Dissolved Ga concentrations along Line P exhibit local concentration maxima at ~150 – 200m, demonstrated in this thesis to correlate with the variable spice.  Lower spice waters along the western terminus of the transect, indicative of North Pacific Intermediate Waters, are coincident with lower concentrations of dissolved Ga, while higher spice waters representing the influence of the California Under Current possess higher concentrations of dissolved Ga.  This work therefore identifies dissolved Ga as a tracer of advective source mechanisms of trace metals to the northeast Pacific.  Geographic and temporal variability in spice along the Line P transect suggest differential trace metal sources as a function of space and time across Line P and providing a potential control on transitional HNLC regions along Line P.      30 1.9.2 Aim 2: Investigate the Ga/Pb ratio as an oceanographic tracer (Chapter 3)  1.9.2.1 Aim 2: Contribution   Ratios of Ga/Pb are interpreted in this thesis to represent a novel tracer of fluvial freshwater inputs of trace metals.  Elevated Ga/Pb ratios are observed along Line P at stations along the shelf break and associated with the presence of an eddy, as coastal waters of high Ga/Pb are transported offshore.  Application of the Ga/Pb ratio to future research includes the opportunity to trace fluvial sources to the north Pacific and along Line P during intra-annual cycles of river discharge as a function of snow accumulation and melt and interannual precipitation changes such as effects of El Niño.    1.9.3 Aim 3: Apply Pb isotopic ratios to identify sources along Line P (Chapter 4) 1.9.3.1 Aim 3: Contribution   This thesis utilizes stable Pb isotopes to identify Asian and North American sources to the northeast Pacific along Line P.  In addition to geographic source variability, temporal source variation is also identified as ratios of Pb consistent with freshwater fluvial sources from the west coast of North America are confined to the summer mixed layer.  Finally, correlation is demonstrated between isotopic ratios of Pb and HNLC conditions across Line P.  These contributions reveal influence of geographic, temporal, and source mechanism controls on trace metal sources along the Line P HNLC gradient.   1.9.4 Aim 4: Report the first dissolved Ga profiles in the Arctic Ocean (Chapter 5) 1.9.4.1 Aim 4: Contribution   This thesis provides the first investigation of dissolved Ga in the Arctic, resulting in unique profiles relative to other oceans.  Dissolved Ga in the Arctic is shown to provide a   31 complimentary tracer of Pacific waters in the Arctic.  Conservative behaviour of dissolved Ga in Pacific waters supports interpretation of calculated nitrogen fixation rates in this thesis as representing in-situ processes within the Beaufort Gyre, as opposed to influence from freshwater sources.  Additionally, profiles of dissolved Ga in the Arctic reported here provide insight into dissolved Al profiles in the Arctic.  Results of this thesis provide a baseline of dissolved Ga concentrations in the western Arctic, providing comparison with future measurements of dissolved Ga through the Arctic archipelago and the eastern Arctic.  Finally, this work promotes investigation of impacts of nitrogen fixation in the Beaufort Gyre on phosphate delivery to the north Atlantic.    1.9.5 Aim 5: Identify sources of elevated trace metal concentrations within the Columbia River plume (Chapter 6)  1.9.5.1 Aim 5: Contribution   This thesis develops a method of calculating estuarine plume endmembers.  Dissolved Ga is shown to trace dynamic shifts in advective plume waters, identifying spatially and temporally distinct plume structures, and demonstrating differential trace metal sources during periods of upwelling and downwelling.  This work promotes future research on differential trace metal sources within river plumes utilizing autonomous real time measurements of ocean temperature, salinity, and wind data to allow targeted sampling of upwelling and downwelling events.   1.9.6 Chapter 7: Thesis conclusions and synthesis   Conclusions of this thesis are summarized and synthesized in Chapter 7 and figures are developed providing a graphical representation of conclusions presented in this thesis. This thesis provides the first reports of dissolved Ga in the Arctic Ocean, the first transect and interannual study of dissolved Ga along Line P, and the first reports of isotopic Pb ratios along the Line P transect.  This thesis introduces 5 new tracer applications contributing to   32 oceanographic research, dissolved Ga traces: 1) advective sources of trace metals in the northeast Pacific, 2) Pacific and Atlantic source waters in the western Arctic, and 3) plume transport of the Columbia River, Ga/Pb traces: 4) fluvial freshwater sources, and Pb isotope ratios trace: 5) geographically and temporally distinct trace metal sources to the northeast Pacific.  Conclusions presented in this thesis contribute to the fields of trace metal oceanography and ocean biogeochemical cycling and encourage increasingly routine measurement and application of dissolved Ga and Pb isotopes as tracers of physical and chemical processes in the ocean.      33 Chapter 2: Methods  Measuring low concentrations of trace metals in seawater requires sensitive analytical methods and sample preconcentration.  In addition, clean sampling techniques and sample processing is necessary to ensure that results accurately reflected the ocean sample and not secondary contamination. Advantages and disadvantages of methods for preconcentration of Ga and Pb from seawater and detection by different analytical instruments are discussed. Specifics related to each method and each study are contained within the Methods section of each thesis chapter.    2.1 Sampling  2.1.1 Clean sample collection  Breakthroughs in trace metal oceanography required the development of sampling methods to collect ocean water samples without trace metal contamination (Bruland, et al. 1978a).  Teflon lined sample bottles attached to a Kevlar line and tripped with Teflon messengers provide a classic method of oceanographic trace metal sampling.  Pumps provide a method of trace metal sampling to shallow depths.  Surface samples can be collected into clean bottles by hand from a small boat.  Surface samples can also be continuously pumped through a ‘fish’ suspended off the bow of a research vessel.  Modern trace metal rosette systems provide a platform for 12 GoFlo bottles and include a Conductivity, Temperature, Depth (CTD) instrument housed in a titanium frame.  Conducting Kevlar line allows oceanographic properties, such as salinity, temperature, oxygen, and fluorescence, to be monitored in real time, therefore allowing for targeted selection of sample depths.  Studies included in this thesis utilize samples obtained by trace metal rosette and GoFlo bottles, Niskin X bottles attached to a Kevlar line and Teflon messengers, surface sampling by hand from a small boat, and a surface sampling fish system.    Acquisition of trace metal samples to be stored and analyzed in the lab requires careful sampling and handling techniques to avoid contamination.  Samples were collected into Low Density   34 Polyethylene (LDPE) bottles, rigorously cleaned with alkaline detergent and HCl and HNO3 acids.  Acidification of samples was performed using trace metal clean HCl obtained from Seastar Chemicals Inc (Sidney, BC, Canada) within a hood providing laminar flow following passage through a HEPA filter.  Trace metal clean procedures are continued during sample processing and analysis within class 1000 clean rooms and class 100 laminar flow benches, described in section 2.2.1 below.    2.1.2 Sampling depth selection  Interpretation of trace metal profiles in the ocean and identification of new oceanographic applications is dependent on selection of sample depths specific to a given study site and hydrographic observations within individual profiles.  Real time Conductivity, Temperature, Depth (CTD) data provide a unique opportunity to select sampling depths targeting physical and chemical characteristics of the water column.  For example, sample depths may be selected based on local maxima or minima in oxygen, representing photosynthesis and respiration processes.  Isopycnals can be targeted across a transect, assisting in interpretation of advected water masses.  Temperature and salinity provide opportunities to select sampling depths corresponding to the mixed layer or upwelling events.  Sampling at depths above and below a gradient observed in a profile can elucidate processes associated with property interfaces in the water column.  Targeted sampling in this thesis was performed in the Beaufort Sea of the Arctic Ocean and in 2011 along Line P.    Selection of sampling depths based on oceanographic properties is of particular importance in the highly stratified Arctic Ocean.  Complex Arctic hydrography demands high resolution sampling with appropriate selection of sampling depths.  Temperature, salinity, and oxygen profiles were therefore monitored on the downward rosette cast, allowing for selection of sample depths based on hydrographic parameters.  Primary profile characteristics identified for sampling included the temperature minimum and O2 minimum indicative of Pacific Winter Waters (PWW) and temperature maximum associated with Pacific Summer Waters (PSW) (see Figure 5.2).    35 Specific water mass sampling assists interpretation of dissolved Ga as a tracer in the Arctic Ocean.    Trace metal sampling along Line P in August 2010 provided an additional opportunity, use of a trace metal clean pump to collect shallow depth samples at high resolution. Trace metal sampling at high resolution (0, 5, 10, 25, 40, 75, 100m) provided detail essential for the identification and application of the Ga/Pb ratio and the interpretation of geographic and temporal variability of Pb isotopes.  Dynamic trace metal concentrations and isotopic ratios in the upper 75m across the Line P transect reported in this thesis emphasize heterogeneity within the upper water column and associated with mixed layer depths, encouraging future high resolution sampling.   2.2 Sample analysis 2.2.1 Trace metal clean laboratories  Following trace metal clean sample collection analysis of the trace metals in seawater occurs in trace metal clean laboratories.  To prevent contamination during sample processing, procedures are conducted in laminar flow fume hoods with air supplied through HEPA filters.  Laminar flow fume hoods provide a Class 100 workspace, indicating the maximum number of particles (100) of a given size class (5m) per ft3.  Trace metal clean laboratories represent class 1000 space and are maintained in an overpressure environment by HEPA filtered air.  Entrants to trace metal labs wear full body clean suits, a bonnet over the head, and gloves.  Work surfaces within the clean lab are wiped down before and after use and care is exercised handling ocean samples and chemical reagents to prevent contamination.    2.2.2 Sample processing  Sample processing prior to instrumental analysis is often required to preconcentrate elements of interest, remove matrix interference, and prepare the sample in a solution appropriate for analysis.  Instrumental sensitivity is insufficient to detect most trace metals at ambient seawater   36 concentrations, therefore preconcentration of ocean water samples is required prior to instrumental analysis.    Preconcentration by liquid-liquid extraction uses an added chelator in the form of an organic molecule to bind trace metals of interest, followed by separation of an organic phase from the aqueous matrix of ocean water.  Preparation of the sample for analysis requires the element to be returned to an aqueous phase, either by back extraction by acid addition or evaporation of the organic phase.  Resins of various types have also been utilized to retain elements of interest from seawater, including recent utilization of the NOBIAS resin (Sohrin, et al. 2008; Biller and Bruland 2012).  Resins can be loaded into columns offering both online as well as offline detection capability.  Selective retention of elements as a function of pH allows the advantage of retaining multiple trace elements, while excluding the major ions of the seawater matrix, such as Na+, K+, and Ca2+.    Magnesium coprecipitation represents an additional method for concentrating trace metals from seawater while also removing unwanted ions from the seawater matrix.  Simple addition of NH4(OH) to a seawater sample initiates formation of Mg(OH)2 and scavenges trace metals from the seawater.  Centrifugation results in a pellet containing the precipitated trace metals and the seawater matrix is decanted.  Acidification returns the precipitate to the dissolved phase and the sample is ready for analysis.  Magnesium coprecipitation has been previously utilized for measurement of trace metals in seawater including Fe and Mn (Saito and Schneider 2006), Ga (Shiller and Bairamadgi 2006) and Pb (Wu and Boyle 1997b).  Given a variety of preconcentration methods, advantages and disadvantages of each method are evaluated in Sections 2.3.1 and 2.4.1 for Ga and Pb.  2.2.3 Instrumental analysis   Early analytical advancements in trace metal oceanography included application of Graphite Furnace Atomic Absorption Spectroscopy (GF-AAS).  Atomization of the sample deposited   37 within a graphite tube is achieved by high temperatures within a furnace.  Detection by AAS provides elemental specificity, as only a single wavelength specific to the element of interest is available for absorption.  Sensitivity of GF-AAS often requires preconcentration of a given trace metal from seawater, employing methods such as those described in section 2.2.2 above.    Mass Spectrometry (MS) allows detection of multiple elements or isotopes concurrently during sample analysis, based on the mass to charge ratio of the ion.  Sector field instruments such as the Element 2 employ double focusing capabilities of a magnetic sector, steering ions based on their energy and mass, and an electrostatic sector focusing ions based on energy alone.  Selection of ions of different mass to charge ratios is achieved by scanning the magnet.    As mass spectrometry separates and detects ions based on their mass to charge ratio, analytes of interest within a sample must be ionized.  While a variety of ionization methods exist, an Inductively Coupled Plasma (ICP) provides effective ionization of trace metals.  Preconcentration is still required for ICP-MS analysis of trace metals at pmol kg-1 concentrations in the ocean.  Additionally, sample processing for the removal of major salt ions in seawater is required for ICP-MS as precipitation of salts during sample delivery to the MS will result in fewer ions entering the MS and therefore decreased sensitivity.    While ICP-MS and GF-AAS are most commonly utilized in land based laboratories following offline sample preconcentration, fluorescence detection combined with flow through sample preconcentration systems provide opportunity for trace metal analysis at sea.  Each detection method has merits specific to a given application.  Selection of instruments for detection of dissolved Ga and Pb isotopes in this work will be discussed in Sections 2.3.2 and 2.4.2.    2.2.4 Data Analysis  Following elemental detection, quantification methods are required to calculate elemental concentrations in the original seawater sample.  Standard curves provide a classic means of   38 quantification.  Standards, however, are rarely able to be prepared in a matrix exactly matching that of sample solutions, and therefore the matrix can influence quantification given differential matrix effects in standard and sample solutions.  Standard addition provides a means of matrix matching, with addition of an elemental spike to each sample.  While standard addition removes the requirement of measuring a standard curve prepared in a separate matrix, a standard curve must still be produced.  While standard curves could be produced for every sample, an assumption of sample matrix similarity is often employed, allowing one standard curve to be used to quantify multiple samples.    Elements possessing multiple isotopes provide the benefit of isotope dilution methods.  Addition of a stable isotope at the beginning of a sample preparation analysis provides an ideal internal standard, normalizing for variation in element recovery and inherent instrumental inconsistencies.  Given the known natural abundance ratio of two isotopes of an element, and the known amount of enriched isotope added to a sample, the amount of non-enriched isotope present in the sample is calculated.  Isotope dilution has the advantage that any processes affecting the concentration or detection of ions in a sample will not impact the quantification of an element calculated by isotope dilution, as only the ratio of isotopes is needed for quantification.    Similarly, interest in isotopic ratios of Pb is not dependent on the absolute amount of ions present, but rather only on the relative amounts of different Pb isotopes.  While measured isotopic ratios are not quantified with respect to a standard curve or isotope dilution, ratios are corrected relative to a reference material.  Correction to a reference standard allows isotopic ratio comparison intraday, interday, inter-instrument, and inter-laboratory.  Processing of data for calculation of Ga concentrations and Pb isotopic ratios will be discussed in the corresponding sections 2.3.3 and 2.4.3 below.      39 2.3 Methods: Ga 2.3.1 Mg coprecipitation: Ga  Various methods to preconcentration Ga from seawater have advantages and disadvantages.  While liquid-liquid extraction provides optimal removal of major salts from the sample, the method is time consuming and organic reagents utilized in the procedure can be harmful.  Resins can be packed into columns and provide the ability to perform multi-element analysis.  Additionally, automated systems can be developed, increasing sample throughput.  However, cleaning and maintaining a system with columns, tubing, connectors, resin, and multiple reagents can lead to inconsistencies during operation.  For instance, columns packed with resin can form channels and peristaltic pump tubing continuously changes rigidity with use, resulting in improper proportions of reagent mixing.    For this thesis, therefore, magnesium coprecipitation was selected for preconcentrating Ga from seawater.  This method is less time consuming, requires minimal addition of reagents (reducing contamination risk) and no specialized equipment is required, thus reducing inconsistencies during sample processing.  This method, however, is not without faults.  Differential precipitation may occur based on the properties of individual samples potentially impacting element recovery. The use of isotope dilution methods described in section 2.3.3 alleviates this latter concern.  Additionally, this method precipitates Mg within the seawater matrix. To manage the amount of Mg introduced to the ICP-MS a second precipitation step is performed, described in section 2.3.4.    2.3.2 ICP-MS: Ga  GF-AAS (described in section 2.2.3 above) provides a quick and inexpensive method of measuring concentrations of Ga preconcentrated from seawater.  Additionally, given the specificity of AAS, additional elements in the matrix are less likely to produce interference, and therefore specific matrix removal is of less concern.  ICP-MS, however, offers a distinct   40 advantage relative to GF-AAS: the ability to measure isotopes of an element and perform quantification by isotope dilution.  Analysis of Ga in this thesis was therefore performed by ICP-MS.  Results presented in Chapter 3 and Chapter 5 were measured on a Thermo Scientific Element 2 sector field ICP-MS operated in Medium Resolution mode, as described in section 2.3.4 below.    2.3.3 Isotope dilution: Ga  Gallium possess two naturally occurring isotopes 69Ga and 71Ga, occurring at a natural abundance of 60% 69Ga and 40% 71Ga.  Isotope dilution adds an enriched solution of 71Ga (99.8% 71Ga), and thereafter only the isotopic ratio of 71Ga / 69Ga of the sample is necessary for quantification.  Concentration of the enriched 71Ga solution is determined by reverse isotope dilution, relative to a standard solution of natural abundance Ga.  Given the known concentration of the enriched solution, addition of a known volume of this solution to each sample allows quantification solely on the analysis of each individual sample.  Relative to quantification based on a standard curve, isotope dilution minimizes effects of inherent fluctuations in instrument performance and differences in matrix between samples and standards.    Dissolved Ga reported in this thesis from Line P in the northeast Pacific and the western Arctic Ocean is therefore determined by a method of preconcentration by Mg coprecipitation, analysis by ICP-MS, and quantification by isotope dilution.  Evaluation of the method will be presented in section 2.3.5, following description of method modifications developed as a part of this thesis.    2.3.4 Method modifications  This work introduces modest modifications to sample preparation and analytical procedures.  Medium Resolution (MR) mode was utilized on the ICP-MS to resolve interference of 138Ba2+ with 69Ga.  While this interference has been mitigated through washing of the precipitate (Shiller and Bairamadgi 2006), attempts at washing in this work were not successful in consistently   41 removing sufficient amounts of Ba from the sample.  Removal of Ba by washing is more advantageous for samples collected from the Atlantic Ocean, as concentrations of Ba are lower, while Ga concentrations are higher, relative to the Pacific.  Conversely, Ba concentrations are higher in the North Pacific as Ba concentrations increase along thermohaline circulation, while concentrations of Ga are lower as discussed previously, therefore exacerbating the problem of Ba interfering with Ga in samples from the North Pacific.    Provided that washing of the precipitate removes sufficient Ba, Ga can be measured by ICP-MS in Low Resolution (LR) mode, including in Pacific samples (Shiller and Bairamadgi 2006).  While a correction can be performed by analyzing a Ba standard and taking the signal at m/z = 69 to represent 138Ba2+ and applying a correction factor, inconsistency in the washing did not provide confidence in application of a single correction factor to all samples.  Given insufficient removal of Ba by washing in this work, MR mode was investigated as a means of removing the interference of 138Ba2+.  Analysis of a Ba standard in LR resulted in 8-10% counts/sec (cps) of the 138Ba+ signal detected as 138Ba2+, interfering with measurement of 69Ga.  While in MR only 0.002% of the 138Ba+ signal is detected as 138Ba2+, removing the interference of Ba with measurement of 69Ga.  While absolute signal is lower in MR compared to LR, the signal/noise ratio was higher in MR than LR by a factor of 1.5 – 2.    Preconcentration of Ga in this work beginning with a 50mL seawater sample and final solution volume of 0.5mL provides a 100X concentration of Ga from seawater for analysis by ICP-MS. For Mg coprecipitation, an appropriate volume of NH4(OH) must be selected to initiate precipitation while minimizing introduction of salts to the ICP-MS interface.  Multiple factors result in differential NH4(OH) volume requirements among samples.  Given salinity stratification within the water column, particularly in the shallow Beaufort Sea of the Arctic Ocean, the volume of NH4(OH) required to initiate precipitation will vary.  Samples of similar salinity can differ in the amount of NH4(OH) required as sample pH can vary based on the exact volume of the seawater sample and HCl added, given logistical constraints of sampling at sea.  Additionally, the total alkalinity and buffering capacity of the sample can influence the effects of HCl and NH4(OH) addition.  Therefore, prior to precipitation of 50mL seawater sample volumes,   42 1.3mL sample volumes were tested with a range of NH4(OH) concentrations and the optimal NH4(OH) volume was then scaled up to precipitate the 50mL sample.   This work also introduced a second precipitation step to minimize Mg introduced to the MS.  Following addition of NH4(OH) to a 50mL seawater sample, centrifugation, and decanting, the precipitated Mg pellet, and associated scavenged trace elements, are dissolved in 1mL 0.1% HCl and transferred to a 1.3mL microcentrifuge tube.  This solution is then precipitated with a small volume (10 – 20 L) of NH4(OH), sufficient to produce only a very small precipitate, approximately 1 – 10% the size of the original precipitate.  While less precipitate is formed, concentrations of Ga are orders of magnitude lower than Mg and therefore Mg precipitation is sufficient to scavenge Ga from the solution.  Additionally, calculation of element concentrations by isotope dilution is not dependent on absolute recovery, but rather is performed based on the relative ratio of elemental isotopes.    2.3.5 Method evaluation: Ga  Oceanographic consistency (Boyle, et al. 1977) is a concept employed to evaluate the quality of oceanographic data by comparison of relative concentrations within a profile, i.e. that the profile is smooth, with gradients present as expected based on hydrographic and biological influence. This work reports smooth oceanographically consistent profiles of dissolved Ga (see for example Figure 3.2, Figure 5.3).  Due to logistical constraints of oceanographic and trace metal sampling, replicate samples are often not available, and therefore consistency with previous observations is an additional method to evaluate data.  Additionally, deep ocean waters provide a quasi-standard, as little modification is expected on decadal time scales.  Accuracy of the method is therefore supported by excellent agreement between profiles at P26 sampled in 1983 (Orians and Bruland 1988b) and this work (Figure 2.1).        43   Figure 2.1: Profiles of dissolved Ga at OSP (50N, 145W) sampled in 1983 (Orians and Bruland 1988b) and 2012 (this work) demonstrate excellent agreement, supporting the accuracy of the method of dissolved Ga analysis utilized in this thesis.  Sample precision is evaluated by performing method replicates on approximately 1 in every 5 samples of dissolved Ga measured in this thesis along Line P in the northeast Pacific and the western Arctic Ocean.  Method replicates consist of replicate samples from the same sample bottle, representing evaluation of method precision.  Precision based on percent relative standard deviation (%RSD) of n = 32 replicate sample preparations (n = 2) for samples along Line P (Table 2.1) results in a mean = 3.2% RSD, median = 3.3% RSD, and range = 0 – 7.4% RSD.  Precision of Arctic samples (Table 2.2), n = 22 replicate samples (n = 2), were mean = 3.5% RSD, median = 2.8% RSD, and range = 0 – 12% RSD.          44  Table 2.1. Replicate samples of dissolved Ga from seawater: Northeast Pacific Ocean (Line P) Year Station Depth (m) Replicate Mean SD RSD (%) 1 2 2010 P4 10 4.1 3.8 4.0 0.2 5.4 2010 P12 150 7.0 7.0 7.0 0.0 0.0 2010 P16 300 6.8 6.7 6.8 0.1 1.0 2010 P20 0.5 3.6 3.5 3.6 0.1 2.0 2010 P20 40 3.9 3.7 3.8 0.1 3.7 2010 P26 150 6.9 7.1 7.0 0.1 2.0 2010 P26 800 4.6 4.8 4.7 0.1 3.0 2011 P4 35 4.1 3.7 3.9 0.3 7.3 2011 P4 200 8.2 8.1 8.2 0.1 0.9 2011 P4 600 5.9 5.9 5.9 0.0 0.0 2011 P4 900 5.5 5.3 5.4 0.1 2.6 2011 P4 1100 5.8 5.8 5.8 0.0 0.0 2011 P12 800 5.0 5.5 5.3 0.4 6.7 2011 P16 130 9.5 9.5 9.5 0.0 0.0 2011 P16 150 9.3 10.0 9.7 0.5 5.1 2011 P20 10 3.5 3.7 3.6 0.1 3.9 2011 P20 25 4.0 3.6 3.8 0.3 7.4 2011 P20 40 3.6 3.6 3.6 0.0 0.0 2011 P20 50 3.8 3.5 3.7 0.2 5.8 2011 P20 75 5.0 5.4 5.2 0.3 5.4 2011 P20 100 6.6 6.1 6.4 0.4 5.6 2011 P20 130 7.8 7.7 7.8 0.1 0.9 2011 P20 150 9.5 9.0 9.3 0.4 3.8 2011 P20 240 6.7 6.6 6.7 0.1 1.1 2011 P20 300 6.3 6.3 6.3 0.0 0.0 2011 P20 600 4.9 5.0 5.0 0.1 1.4 2011 P20 1400 4.9 5.4 5.2 0.4 6.9 2011 P20 1600 6.0 5.7 5.9 0.2 3.6 2011 P26 75 3.3 3.5 3.4 0.1 4.2 2011 P26 185 5.4 5.7 5.6 0.2 3.8 2011 P26 2000 6.5 6.3 6.4 0.1 2.2 2012 P26 150 5.0 4.5 4.8 0.4 7.4    45  Table 2.2. Replicate samples of dissolved Ga from seawater: Arctic Ocean (Beaufort Sea) Year Station Depth (m) Replicate Mean SD RSD (%) 1 2 2009 L1 30 4.3 5.1 4.7 0.6 12.0 2009 L1 250 17.5 17.5 17.5 0.0 0.0 2009 L1 400 24.7 26.1 25.4 1.0 3.9 2009 L1 1250 27.9 29.0 28.5 0.8 2.7 2009 L1 1500 25.5 23.1 24.3 1.7 7.0 2009 L1.1 20 6.8 6.2 6.5 0.4 6.5 2009 L1.1 60 4.6 4.7 4.7 0.1 1.5 2009 L1.1 110 5.2 5.3 5.3 0.1 1.3 2009 L1.1 360 24.7 26.0 25.4 0.9 3.6 2009 L1.1 400 26.7 25.3 26.0 1.0 3.8 2009 L1.1 425 26.0 25.6 25.8 0.3 1.1 2009 L1.1 500 28.1 26.9 27.5 0.8 3.1 2009 L1.1 600 27.3 27.8 27.6 0.4 1.3 2009 L1.1 1250 29.6 30.6 30.1 0.7 2.3 2009 L1.1 1500 29.1 31.3 30.2 1.6 5.2 2009 L2 8 6.3 5.8 6.1 0.4 5.8 2009 L2 120 6.0 6.1 6.1 0.1 1.2 2009 L2 900 26.6 26.4 26.5 0.1 0.5 2009 L2 2100 29.0 29.0 29.0 0.0 0.0 2009 L3 10 7.6 7.4 7.5 0.1 1.9 2009 L3 180 8.1 9.2 8.7 0.8 9.0 2009 S4 17 12.7 12.2 12.5 0.4 2.8  2.4 Methods: Pb isotopes 2.4.1 Sample processing: Pb isotopes  Preconcentration of Pb from seawater is performed in an analogous procedure to Ga (Wu and Boyle 1997b).  Following preconcentration of Pb from a 50mL seawater sample by Mg coprecipitation, a secondary resin procedure is performed to eliminate additional trace metal and matrix ions (Shiel, et al. 2010).  Retention of Pb on an anion exchange column occurs in the presence of Br- anions, excluding transition metals and alkali earth elements within the matrix.  Lead is then eluted from the column following the addition of Cl- anions. The resulting solution is then dried down and reconstituted in 0.5mL 1% HNO3 for analysis by ICP-MS.  Sample processing therefore results in a 100X concentration of Pb from seawater for analysis of stable radiogenic Pb isotopes.     46  2.4.2 Pb isotopes: ICP-MS  Isotopic Pb ratios of seawater samples were measured on the Nu AttoM ICP-MS.  While the Nu AttoM is a single collector sector field MS, similar to the Element 2 described above, an advantage of the Nu AttoM is the introduction of deflector plates at the entrance and exit of the magnet to focus the ion beam, as opposed to scanning the magnet, therefore increasing magnet stability and optimizing precision.  Increased instrumental precision is beneficial as the volume of seawater to be processed for analysis is dependent on the amount of Pb required to provide the level of precision required for a given application.  Precision requirements specific to oceanographic applications and this method will be evaluated in section 2.4.4.  Lower seawater volume requirements of 50mL applied in this thesis provide logistical advantages for both field sampling and laboratory sample preparation, promoting increasingly routine analysis of Pb isotopes during oceanographic research cruises.    2.4.3 Data handling: Pb isotopes  Reference standard NBS-981 brackets all sample analyses and linear correction is performed for each sample.  Comparison of Pb isotope ratios to NBS-981 provides internal consistency and accuracy of measured isotopic ratios of Pb relative to the NBS-981 standard.  Correction to NBS-981 provides the ability to compare both intra-day and inter-day ratios of Pb isotopes measured on a given instrument, as well as allowing comparison with results from other laboratories.   Regarding instrumental precision, slight variations in instrumental performance invariably and continuously occur, potentially impacting absolute counts of individual isotopes.  For instance, variation in the plasma influences ionization and therefore variation in the measured counts of an individual isotope.  Ratios of isotopes measured on each individual scan, performed by the ICP-MS on millisecond time scales, display greater consistency, as factors impacting the counts of one isotope are likely to impact other isotopes as well.  Consideration of isotopic ratios therefore   47 provides the advantage of reducing the impact of inherent instrumental variation.  Accuracy and precision of the Pb isotope method presented in this thesis are discussed further in the following section.    2.4.4 Method evaluation: Pb isotopes  Reference standards for Pb isotopes in the ocean are being evaluated as part of the global trace metal program GEOTRACES (Boyle, et al. 2012).  Accuracy of Pb isotope ratios measured by the method in this thesis on the Nu AttoM ICP-MS is evaluated by comparison to reference standards available from GEOTRACES: GDI, collected at 2000m at the Bermuda Atlantic Time Series (BATS) and GSI, collected at ~7m at BATS (Boyle et al. 2012).  Isotopic Pb ratios of reference standards measured in this work compare favourably (p < 0.05) with reported values from laboratories at the Massachusetts Institute of Technology (MIT) and the University of California Santa Cruz (UCSC) (Boyle et al. 2012).    Table 2.3. Isotopic Pb ratios of the GDI reference standard   Reference Standard: GDI   206Pb/207Pb SD 208Pb/206Pb SD This work 1.180 0.0025 2.074 0.0064 MIT 1.184 0.0017 2.070 0.0029 UCSC 1.181 0.0003 2.073 0.0016  Table 2.4. Isotopic Pb ratios of the GSI reference standard   Reference Standard: GSI   206Pb/207Pb SD 208Pb/206Pb SD This work 1.174 0.0031 2.078 0.0030 MIT 1.175 0.0003 2.082 0.0008 UCSC 1.178 0.0005 2.078 0.0029     48 Standard deviation values of Pb isotope ratios measured in this work are greater than reported values from the other two labs.  Precision of the method presented in this thesis is therefore evaluated.  Precision requirements for interpretation of a given set of data is dependent on the data resolution required, and therefore must be viewed as context specific.  Oceanographic 206Pb/207Pb isotope ratio values exhibit ranges differentiated by 206Pb/207Pb ≥ 0.01.  For example, 206Pb/207Pb ratios of 1.16 are observed in the upper 500m of the central north Pacific, increasing to 1.18 at 1500m, and contrasting with 206Pb/207Pb ratios of 1.20-1.21 within the underlying sediments (Wu, et al. 2010).  Similarly, water masses in the south Atlantic are differentiated based on 206Pb/207Pb ratios of 1.14 – 1.19 (Alleman, et al. 2001).   Statistical comparison of data may be performed utilizing a Student’s t-test, a function of the mean difference between two data points and a reciprocal function of data precision.  Precision dictates the resolution at which two data can be discriminated.  Data means differing by a value greater than the resolution provided by the precision of the data are therefore statistically different.  Therefore, what precision values are required for a given resolution requirement?  Presented here is a graphical representation of achievable resolution as a function of precision (Figure 2.2).               49  Figure 2.2 a) Graph demonstrating the statistical resolution provided by the standard deviation of any two samples.  Contours indicate m, the difference between two mean values that can be statistically resolved given the standard deviation of samples i and j, indicated on the axis.  Plotted are the respective standard deviations of all sample pairs, indicating the difference in means that can be resolved between the two samples.  Blue symbols represent sample pairs that, if differing by the target resolution of m = 0.01, will be statistically resolved based on their respective SD values.  Red symbols represent sample pairs that, if differing by the target resolution of m = 0.01, would not be statistically resolved based on their respective SD values b). Histogram displaying the number of sample comparisons plotted in panel a) corresponding to each resolution bin, >98% of all possible sample comparisons can be resolved at the target resolution of m = 0.01.      50 Graphical representation of statistical resolution provides an approach to determining resolution provided by the precision of two data to be compared.  Standard deviations of two data to be compared are plotted and contours represent the minimum mean difference required for statistical discrimination.  As the targeted resolution of the method is 206Pb/207Pb ≥ 0.01, all sample pairs of precision sufficient to provide resolution of m ≤ 0.01, represented as blue symbols in Figure 2.2 meet the requirement of providing statistical resolution of 206Pb/207Pb ≥ 0.01.    For instance, given sample i with a SD value of 0.004, comparison with another sample j of SD = 0.004 provides statistical resolution of ~0.006, therefore if 206Pb/207Pb ratio of the two samples differed by more than 0.006, the samples could be statistically distinguished, if the 206Pb/207Pb ratio of the two samples differed by <0.006, a null hypothesis that the two samples are different would be rejected.  Alternatively, if the same sample i, with SD = 0.004, was compared with a different sample j of SD=0.007, then the difference, m that could be statistically resolved would be ~0.008.  Both of these examples meet the targeted resolution of 206Pb/207Pb ≥ 0.01.    Standard deviation of Pb isotope ratios reported in Chapter 4 of this thesis range from SD = 0.0007 – 0.008 (Table 4.1, Table 4.2), with mean SD = 0.0034 and median SD = 0.0030.  Comparison of 206Pb/207Pb ratios for all stations and all depths result in a mean m = 0.0053, a median m = 0.0052, and > 98.5% of all possible sample comparisons provide resolution at m ≤ 0.01.  Therefore, results of this method provide sufficient resolution for interpretation of Pb isotopic ratios presented in Chapter 4 of this thesis and application as tracers of geographic and temporal resolution of Pb sources along Line P.    2.5 Methods conclusions  This thesis utilizes Mg coprecipitation to preconcentrate Ga and Pb from seawater.  While both elements are measured by ICP-MS, the Nu AttoM provides increased precision allowing for analysis of Pb isotopes from 50mL sample volumes.  Quantification of Ga is performed by isotope dilution and Pb isotope ratios are corrected by comparison with NBS-981.  Profiles of   51 dissolved Ga demonstrate oceanographic consistency and concentrations measured in deep north Pacific waters in this work show excellent comparison with previous reports (Orians and Bruland 1988b).  Isotopic ratios of Pb of reference standards measured in this work compare well with published results.  Precision of Pb isotope measurements provided by the Nu AttoM is shown to be sufficient to provide oceanographically relevant resolution.  Methods introduced here are utilized to measure dissolved Ga along the Line P transect in the northeast Pacific ocean (Chapter 3), Pb isotopes along Line P (Chapter 4), and dissolved Ga in the western Arctic Ocean (Chapter 5).      52 Chapter 3: Dissolved Gallium in the Northeast Pacific Ocean  3.1 Synopsis  Results of this work demonstrate novel applications of dissolved Ga in the ocean as a tracer of advective and fluvial sources.  Dissolved Ga concentrations in the northeast Pacific Ocean measured in this work represent the first depth profile transects of dissolved gallium (Ga) measured in consecutive years.  While comparison of the two years reveals similar features within dissolved Ga profiles along the transect, geographically and temporally distinct characteristics are observed.  Specifically, variation is observed in the magnitude of local dissolved Ga concentration maxima occurring at 150 – 200m along Line P.  These local dissolved Ga maxima are shown to correlate positively with the variable spice, an indicator of temperature and salinity characteristics of waters along an isopycnal.  Dissolved Ga is interpreted in this work to represent a tracer of advective sources of trace metals to the northeast Pacific Ocean originating from the California Undercurrent (CUC) and North Pacific Intermediate Water (NPIW), based on correlation of higher concentrations of Ga with higher spice waters of the CUC and lower Ga with lower spice NPIW.  Given the correlation identified in this work interannual longitudinal variability in spice shown here across Line P indicates dynamic trace metal advective sources.  Additional tracer application of dissolved Ga is identified here paired to concentrations of dissolved Pb (Charters 2012).  Elevated concentrations of the resulting Ga/Pb ratio are interpreted to represent fluvial freshwater sources of trace metals.  Finally, this work indicates elevated concentrations of dissolved Ga associated with an eddy, transporting coastal waters offshore and intersecting the Line P transect.   3.2 Introduction  Line P represents one of the longest running sampling programs in oceanography, extending more than 50 years (Freeland 2007).  Line P consists of a series of stations extending from   53 coastal waters off the coast of Vancouver Island to open ocean stations more than 1600km offshore into the northeast Pacific Ocean.  Multi-decadal time series of temperature and salinity along Line P indicate warming trends (Crawford, et al. 2007b), declining oxygen (Whitney, et al. 2007), nutrient dynamics (Pena and Varela 2007), decadal regime shifts (Wong, et al. 2007), and ENSO cycles (Schwing, et al. 2002).  Physical and chemical oceanographic conditions along the transect elicit biologic controls on phytoplankton dynamics, ultimately impacting fisheries and local economies along coastal British Columbia.   The Alaska Gyre represents a dominant physical oceanographic presence within the northeast Pacific Ocean. Ekman pumping associated with the cyclonic Alaska Gyre provides an upwelled source of nutrients to the western extent of Line P including the terminus station Ocean Station Papa (OSP, equivalent to P26).  Upwelling along the coast in the summer when winds blow from north to south also provide nutrient inputs to the near coastal portion of Line P.  While physical processes provide nutrient availability along Line P, phytoplankton have additional trace metal requirements to utilize available nutrients (Morel and Price 2003).  Nutrients within the Alaska Gyre are underutilized by Fe-limited phytoplankton, creating a High Nutrient Low Chlorophyll (HNLC) region.  In contrast, along the eastern extent of Line P phytoplankton are able to utilize available nutrinets. Line P therefore provides the opportunity to study inputs of trace metals influencing phytoplankton blooms along the coast to HNLC conditions associated with the Alaska Gyre.  Trace metals can be sourced from upwelling along the continental shelf (Lohan and Bruland 2008).  Fluvial waters also provide inputs of trace metals (Gaillardet et al. 2003), influencing nutrient utilization along the eastern portion of the Line P (Ware and Thompson 2005).  Extending beyond the shelf break along Line P is the Dilute Domain (Favorite et al. 1976) representing local fluvial inputs providing nutrients to Line P.  In the summer the Dilute Domain is influenced by snowmelt fuelling the Fraser River freshet and inputs in the winter include Vancouver Island rivers fed by winter rainfall (Freeland et al. 1984).      54 Eddies provide additional nutrient inputs to Line P (Whitney and Robert 2002).  Surface eddies are a ubiquitous feature of the northeast Pacific (Crawford, et al. 2007a) and represent a source of trace metal (Johnson, et al. 2005).  Additionally, eddies can occur within the subsurface originating from the California Under Current (CUC) and have been called cuddies (Garfield, et al. 1999).  Cuddies can result in advection of CUC waters offshore within the north Pacific (Collins, et al. 2013) and have been observed in the region of Line P and may provide Fe sources (Pelland, et al. 2013).  Waters of the CUC bring warm salty waters from the south to the cool fresher waters of the northeast Pacific.  Mixing of these differentially sourced waters along isopycnal surfaces can be identified by the oceanographic variable spice (Flament 2002). Spice differentiates waters of equivalent density based on their temperature and salinity characteristics.  In addition to cuddies, advection from the shelf can occur along isopycnal surfaces influenced by Ekman transport (Cullen, et al. 2009).  Volcanoes can also supply episodic inputs of Fe, fuelling phytoplankton blooms along Line P (Melancon, et al. 2014; Hamme, et al. 2010).  Given the spectrum of trace metal limitation that exists along Line P trace metal distributions along Line P are required to understand phytoplankton dynamics.  Trace metal data along the Line P transect, however, is limited.  Iron (Fe) represents the most extensively studied trace metal along Line P.  Paradigm shifting linkage of Fe to HNLC areas was established at OPS (Martin, et al. 1991; Boyd, et al. 1996).  Additionally, multi-annual time series of Fe across Line P have been maintained (Ross 2014) and recently investigation of reduced Fe(II) has been added (Schallenberg, et al. 2014).  Additional trace metal profiles have been reported at individual stations, particularly OSP (50oN, 145oW), including Al and Ga (Orians and Bruland 1988b) and Zn (Lohan, et al. 2002) and biological influence of Cu availability has been shown along Line P (Semeniuk, et al. 2009, Semeniuk 2014)   Trace metal sampling along Line P in August 2010 and 2011 produced the first Line P transects reported for dissolved Pb (Charters 2012), Al (Cain 2014), Cu (Posacka et al. 2012), Mn (Sim and Orians 2014), Cd (Janssen, et al. 2014), and Pb isotopes (this thesis, Chapter 4).  This chapter presents the first reports of dissolved Ga along the Line P transect.      55 Dissolved Ga is a scavenged type trace metal, with concentrations of 30-40 pmol kg-1 in the north Atlantic, 20-30 pmol kg-1 in the south Atlantic (Shiller 1998), declining to 10-15 pmol kg-1 in the North Pacific gyre, and 3-4 pmol kg-1 in the northeast Pacific at OSP (Orians and Bruland 1988b).  Scavenging of dissolved Ga in the ocean results in heterogeneous distribution and profiles.  Utility of dissolved Ga as an oceanographic tracer results from this observed spatial and temporal heterogeneity as demonstrated in the north Pacific (Orians and Bruland 1988b; Shiller and Bairamadgi 2006), south Atlantic (Shiller 1998), and Arctic (this thesis, Chapter 5) Oceans, as well as freshwater sources and fluvial ocean inputs (Shiller and Frilot 1996; McAlister and Orians 2012).    Advantages of Ga as a tracer include a residence time short enough to identify local source and sink processes, yet longer than the rapidly scavenged Al (Orians and Bruland 1988b).  Second, Ga is not redox sensitive and thus inputs are not complicated by photo reduction or reduction associated with suboxic environments.  Third, while Ga can be bound by siderophores (Emery 1986), the complex would not be taken up by phytoplankton using reductive transport mechanisms (Maldonado and Price 1999).  Additionally, Ga provides analytical advantages as contamination risks are less than for Al and Fe, and concentrations can be calculated using methods of isotope dilution.  Finally, an excellent foundational knowledge of Ga in the ocean exists, and yet relatively few reports have been published in the last 20 years.  Therefore new applications identified here promote the increasingly routine measurement of dissolved Ga on oceanographic cruises.    This work identifies two novel tracer applications of dissolved Ga.  Correlation of dissolved Ga concentrations with the variable spice is interpreted to indicate dissolved Ga as a tracer of advective trace metal sources to the northeast Pacific intersecting Line P.  Additionally, this work presents the Ga/Pb ratio as a tracer of fluvial inputs, including eddy transport of coastal waters to offshore stations along Line P.  These new tracers identify features and interpretations of processes necessary to the understanding of biogeochemical cycling along Line P and regionally in the north east Pacific Ocean, an area important to study of HNLC regions (Boyd   56 and Harrison 1999), balancing the global nitrogen budget (Yamamoto-Kawai, et al. 2006), carbon flux (Boyd, et al. 1998), and fisheries (Hutchings, et al. 2012).    3.3 Methods  Dissolved Ga concentrations were sampled along Line P (Figure 3.1) in the northeast Pacific during three cruises in August 2010, 2011, 2012.  Sampling details specific to each cruise are described below.  All three cruises were aboard the CCGS Tully in conjunction with the Line P program at the Institute of Ocean Sciences and Fisheries Canada.  Trace metal sampling was performed at the five stations indicated in Figure 3.1.  Station P4 represents eutrophic conditions at the edge of the continental shelf with nutrients supplied by summer upwelling supporting phytoplankton blooms.  Phytoplankton at P12 are nutrient limited, while at station P26 available nutrients are underutilized, representing a High Nutrient Low Chlorophyll (HNLC) region.  Stations P16 and P20 represent a transition zone of increasing HNLC characteristics approaching P26.  Station P12 was influenced by the presence of an eddy in August 2010, observed by satellite altimetry (Colorado Center for Astrodynamics Research) to originate along the path indicated by the arrow in Figure 3.1.  Eddies are shown to be a consistent (Crawford and Greisman 1987), although temporally and geographically variable, feature in the northeast Pacific including intersecting Line P (Crawford, et al. 2007a).    Figure 3.1: Map of the northeast Pacific Ocean indicating stations sampled for dissolved Ga, arrow indicates source and direction of an eddy present in August 2010   57  Trace metal sampling performed on August 20 – August 26, 2010 was performed with Niskin X bottles for depths below 40m, Teflon pump for samples at 5m – 40m, and hand sampling at 0.5m.  Niskin X bottles were suspended from a Kevlar line and tripped with Teflon messengers.  Pump samples were collected from a Teflon hose attached to Kevlar line and pumped into a HEPA laminar flow hood for sampling.  Hand sampling was performed by dipping a sample bottle 0.5m below the surface with a gloved hand and plastic coated sleeve over the side of a Zodiac inflatable boat hundreds of meters away from the ship.  Sampling performed on August 19 - August 26, 2011 utilized a trace metal rosette (Giesbrecht, et al. 2013) consisting of 12 GoFlo bottles attached to a powder coated frame and a CTD housed in a titanium frame providing real time T, S, and O2 data transmitted by a conducting Kevlar line.  GoFlo bottles were transported from the rosette into a trace metal clean container for sampling.  Sampling at P26 to 4000m on August 25, 2012 was performed by trace metal rosette at depths of 25m – 1800m, and Niskin X bottles on Kevlar line for depths 2100m – 4000m.  All samples were collected into LDPE bottles washed prior to the cruise with a surfactant, 6M HCl (Environmental Grade), and 0.7M HNO3 (Seastar Chemicals).  All sample bottles were then rinsed three times with the ocean water sample prior to sample collection.  All samples were filtered to 0.45m and acidified at sea, 1mL 12M HCl (Seastar Chemicals) to 1L seawater sample, in a HEPA filter laminar flow hood.    Analysis of Ga in the filtered seawater samples was performed by concentrating a 50mL seawater sample 100X by magnesium coprecipitation induced by addition of ammonium hydroxide (Seastar Chemicals).  Similar methods have been utilized for concentration of Fe and Mn (Saito and Schneider 2006), Pb, Cu, and Cd (Wu and Boyle 1997b), and Ga (Shiller and Bairamadgi 2006).  Quantitation was performed by isotope dilution, performed by 50L addition of an enriched solution of 71Ga (99.8%), and analysis on an Element 2 inductively coupled plasma mass spectrometer (ICP-MS).  Detection was performed in medium resolution (MR) to resolve interference of doubly charged 138Ba2+ with 69Ga+.  Blank subtraction was performed on samples prior to calculation of sample concentration by isotope dilution and blank signals represented < 1% of sample signal.  Blank solutions of 1% HNO3 were analyzed and ammonium   58 hydroxide blanks were prepared by evaporation of ammonium hydroxide in a Teflon beaker and reconstitution in 1% HNO3 for analysis.    Results of the method are supported by oceanographically consistent (Boyle, et al. 1977) profiles and corroboration of dissolved Ga concentrations within deep waters of the north Pacific (Orians and Bruland 1988b).  Error bars in plots represent standard deviation of samples measured in duplicate (Table 2.1), 31 samples were analyzed in duplicate (~1 in 5), the n=32 duplicate pairs produce a mean percent relative standard deviation = 3.2% RSD and median = 3.3% RSD, with a range of 0 – 7.4%.    While trace metal sampling was performed at 5 stations across the transect, calculation and plotting of isopycnals and the variable spice included in the discussion section are performed utilizing CTD data from all available stations along the transect, available from Line P archives maintained by the Institute of Ocean Science (www.waterproperties.ca/linep).    3.4 Results  Dissolved Ga measured at 5 stations across Line P demonstrates both broad similarity with depth and consistency with time, while also identifying unique features demonstrating both geographic and temporal heterogeneity across the transect.  Following comparison in the results section of annual and decadal profile and description of profile features across the transect, proposal of processes producing observed features will be presented in the Discussion.    Dissolved Ga profiles sampled across Line P in 2010 and 2011 (Table 3.1 and Table 3.2) are presented in Figure 3.2 and interannual comparison of the upper 400m is provided in Figure 3.3.  Broad profile characteristics are revealed in Figure 3.2, including low concentrations of dissolved Ga in the upper 50m increasing to a maximum at 150m – 200m.  Sampling in 2011 and 2012 reveals that the maximum at 150 – 200m represents a local maxima within a profile, as concentration increase approaching 2000m shown in 2011, including continued increasing to   59 4000m (P26, 2012).  While local maxima at 150 – 200m represent similar features across the transect, maximum concentrations of dissolved Ga, as well as the depth and density of local maxima vary with longitude across Line P (Figure 3.3).    Table 3.1. Concentrations of dissolved Ga (pmol kg-1) measured along Line P in August 2010  Depth (m) 2010 P4 P12 P16 P20 P26 0.5 3.4 4.0 2.7 3.6 - 5 4.3 4.5 3.3 4.3 3.7 10 4.0 5.1 3.1 3.3 3.2 25 4.4 5.2 2.4 3.8 3.8 40 4.3 5.1 4.1 3.8 3.8 75 - 4.8 4.9 4.9 4.3 100 - 5.3 7.9 6.0 5.3 150 - 7.0 9.5 8.8 7.0 200 - 7.6 8.0 7.8 5.5 300 - 6.8 6.8 5.2 5.5 400 - - 7.6 6.1 4.6 600 - - - - 4.6 800 - - - - 4.7    60  Table 3.2. Concentrations of dissolved Ga (pmol kg-1) measured along Line P in August 2011 and P26 in 2012.  Selection of sample depths represent both historical consistency as well as targeted sampling of features such as isopycnals, resulting in sampling depths specific to individual stations across the transect.   Depth (m) 2011 2012   P4 P12 P16 P20 P26 P26 10 3.6 3.5 3.4 3.6 2.5 - 25 4.1 3.9 4.0 3.8 - 2.4 35 3.9 - - - - - 40 3.9 3.9 4.0 3.6 2.2 - 50 3.3 4.0 4.2 3.6 2.4 3.0 75 4.4 4.7 5.2 5.2 3.4 3.3 100 5.3 7.2 8.3 6.3 3.8 3.2 130 - - 9.5 7.7 - - 150 6.8 7.6 9.7 9.2 5.1 4.7 185 - - - - 5.6 - 200 8.1 8.3 7.3 6.6 5.1 4.5 240 - - - 6.7 - - 250 7.9 7.0 - - - - 275 - - 6.5 - - - 300 7.0 6.7 7.2 6.3 5.2 4.5 360 7.2 - - - - - 400 7.5 6.1 5.9 5.9 4.9 4.1 600 - 5.4 6.2 5.0 5.0 - 675 - 6.0 - - - - 700 6.2 5.6 - - - - 800 5.2 5.3 5.6 5.3 4.5 4.5 900 5.4 - - - - - 1000 6.1 5.2 - 5.0 4.6 - 1100 5.8 - 5.1 4.7 5.2 5.1 1200 6.6 5.9 5.2 5.3 - - 1400 - 6.2 5.0 5.2 5.7 5.3 1600 - 5.9 5.3 5.9 5.6 - 1800 - 7.0 6.7 6.2 5.8 5.8 2000 - 8.3 7.2 7.0 6.4 - 2100 - - - - - 6.7 2500 - - - - - 8.4 3000 - - - - - 11.1 3500 - - - - - 13.8 4000 - - - - - 17.5    61       Figure 3.2: Profiles of dissolved Ga at stations P4 – P26 across Line P in 2010 (triangles) and 2011 (circles).  Note extended depth scale to 2000m in 2011.  Inter-annual comparison of profiles in the upper 400m is shown in Figure 3.3    62   Figure 3.3: Annual comparison of dissolved Ga profiles in 2010 (triangles) and 2011 (circles) as a function of both depth and density.  To provide a reference in the density plots, the local dissolved Ga maxima are shown as a filled symbol.    Prominent local maxima in 2010 and 2011 at P16 and P20 are both centered at 150m, corresponding to  = 26.5 – 26.6 (Figure 3.3).  In contrast, at P26  = 26.5 occurs at 100 – 150m and sampling at 150m in 2010 and 185m in 2011 represent local maxima correspond to  = 26.7 – 26.8, .  Stations P12 and P4 both demonstrate local maxima at 200m, however corresponding densities differ: P4 2011  = 26.55, P12 2010  = 26.60, and P12 2011  = 26.68.  In addition to varying depth and density of local maxima across Line P, the magnitude of the maxima illustrates differences across the transect (Figure 3.4).  Highest concentrations of dissolved Ga within the local maxima are observed at Stations P16 and P20 in both 2010 and 2011.  Station P26 has the lowest concentration at the local maxima.  Presence of a dissolved Ga subsurface maxima at P26 was observed in 1983 (Orians and Bruland 1988) and confirmed in   63 this work by similar features at P26 in 2010, 2011, and 2012.  Local maxima in dissolved Ga are reported here for the first time at additional stations along the Line P transect.  Source of this subsurface Ga maximum and opportunity for use as a tracer is presented in the Discussion section below.                     a.                                            b. Figure 3.4: Comparison of dissolved Ga profiles along Line P in a. 2010 (triangle) and b. 2011 (circle).  Symbol colours representing stations are consistent between the two plots, symbol shape indicating year is consistent with previous plots.  Finally, results presented here provide confidence in the analytical technique and sampling methodologies employed in this work.  Interdecadal comparison at P26 to 4000m (Figure 3.5) indicates excellent agreement between samples from 2012 (this thesis) and the profile sampled in 1983 (Orians and Bruland 1988b).  This comparison provides a cross validation of the isotope dilution Mg coprecipitation method utilized in this work with the method of ion exchange column preconcentration and graphite furnace Atomic Absorption Spectroscopy (Orians and Bruland 1988b).      64  Figure 3.5: Full depth profile of dissolved Ga at P26 sampled in 2012 (this work) and 1983 (Orians and Bruland 1988b).      3.5 Discussion 3.5.1 Dissolved Ga: Shallow  Low concentrations dissolved Ga of 2 – 4 pmol kg-1 are associated with the summer mixed layer, with a depth of ~30m across the transect, and the fluorescence maximum at 30 – 50m, consistent with the scavenged behavior of Ga.  Similar concentrations of 3 – 4 pmol kg-1 are observed within shallow waters of both the northeast Pacific at OSP (50°N 145°W) (Orians and Bruland 1988b) and northwest Pacific at 50°N 167°E (Shiller and Bairamadgi 2006).  Along the edge of the continental shelf at P4, concentrations of 3 – 4 pmol kg-1 in the upper 40m are similar to coastal waters off central California (Orians and Bruland 1988b).  Elevated concentrations of ~5 pmol kg-1 are observed at P12 in 2010, discussed in section 3.5.7 below associated with the presence of an eddy.       65 3.5.2 Dissolved Ga: 150m – 200m local maxima   Profiles of dissolved Ga display local maxima at depths of 150 – 200m, corresponding to  = 26.5 – 26.8 (Figure 3.3).  Similar maxima were observed at P26 in 1983, as well as in the central and eastern coastal north Pacific (Orians and Bruland 1988b).  Less prominent local subsurface maxima were observed in the central and western north Pacific (Shiller and Bairamadgi 2006).  What accounts for these local maxima?  Regeneration of dissolved Ga from particulate matter and subsequent scavenging (Orians and Bruland 1988b) is consistent with observation of maxima crossing isopycnal surfaces.  Alternatively, given the absence of similar features of dissolved Ga local maxima in the Atlantic, advective sources were suggested in the Pacific (Shiller and Bairamadgi 2006).  Additionally, within the northeast Pacific proximal to Line P, concentrations of Fe have been shown to be elevated along  = 26.5 – 26.8 associated with sedimentary reduction (Cullen, et al. 2009).  While Fe is a key element in biogeochemical redox cycling within low oxygen environments, Ga is not known to be associated with biogeochemical cycling of Fe reduction, and influence of the oxygen minimum zone along Line P related to concentrations of dissolved Ga is not observed.    Waters along  = 26.5 – 26.8 kg/m3 have been identified as originating within the California Under Current (CUC) (Pierce, et al. 2000).  Waters of the CUC can be identified by spice, allowing waters of the same density to be differentiated based on their relative temperature and salinity characteristics (Flament 2002).  Warmer, saltier waters of a given density will have greater spice than cooler, fresher waters of the same density.  Spice values of -0.1 – -0.3 are associated with the CUC at  = 26.6 – 26.8 (Pierce, et al. 2000).  Offshore transport of CUC is observed within subsurface anticyclonic eddies originating in the CUC and termed ‘cuddies’ (Garfield, et al. 1999), defined as lasting > 70 days, and observed to last as long as 520 days and travel 1650km (Collins, et al. 2013).  Cuddies observed just south of Line P are estimated to transport offshore an average of 44% of the heat and salt lost from the CUC (Pelland, et al. 2013).      66 Contoured sections of spice along the Line P transect (Figure 3.6) indicate relatively spicy waters within the warm 30m summer mixed layer overlaying lower spice waters of the winter mixed layer.  Below the winter mixed layer, a dynamic pattern of spice is revealed across the transect.  While low spice waters of the winter mixed layer follow isopycnal contours, below the winter mixed layer isopycnals penetrate regions of varying spice, indicating differential temperature and salinity signatures within these waters of the same density.    b. a.  Figure 3.6: Spice calculated along Line P from CTD data in a. 2010 and b. 2011, contours indicate density ().  Filled symbols reference depths of dissolved Ga maxima.      67 Local maxima of Ga concentrations are coincident with higher spice values, particularly at stations P16 and P20 in 2010, and similarly in 2011 (Figure 3.6).  Lower concentrations of the dissolved Ga maxima at P26 relative to P16 and P20 (Figure 3.4, Figure 3.7) correspond as well to lower spice values at P26 compared to P16 and P20 (Figure 3.6, Figure 3.7).  Lower spice values at P26, associated with the presence of North Pacific Intermediate Waters (NPIW), and increasing Ga concentrations with higher spice values associated with the CUC, suggests a source of dissolved Ga provided by the CUC.       68  Figure 3.7: Dissolved Ga and spice potted at stations P16, P20, and P26 along Line P at 150m and 200m in 2010 and 2011, showing the decrease in spice and dissolved Ga at P26 relative to stations P20 and P16.        69 Correlation of dissolved Ga with spice at stations P16 – P26 at 150m – 200m (Figure 3.8), resulting in R2 = 0.86 and significance at P < 0.001, supports the interpretation of the CUC providing a source of dissolved Ga along Line P.  Spice suggests advective mixing of CUC and NPIW waters along isopycnal surfaces, and correlation of dissolved Ga with spice is interpreted to represent dissolved Ga as a tracer of advective trace metal sources in the northeast Pacific.               a.                                                                     b. Figure 3.8: a. Correlation of dissolved Ga and spice at P16, P20, and P26 at 150m (filled symbols) and 200m (open symbols), in 2010 (triangles) and 2011 (circles), linear regression results in a slope of 11.5, intercept of 13.2, r2 = 0.84, p < 0.001.  b. Note: regression line in panel a. is presented in panel b. for reference and extrapolated (dashed line).  Concentrations of dissolved Ga at P4 and P12 150m are lower than predicted by the correlation in a., potentially due to increased scavenging at this more productive location.  At station P12 at 200m in 2010 (open green triangle) anonymously spicy waters could result from the presence of the observed surface eddy.    Based on the correlation of dissolved Ga and spice, high spice waters at P4 are suspected to provide a trace metal source from the CUC.  Lower concentrations of dissolved Ga are measured at P4, however, than predicted by the spice correlation (Figure 3.8b).  Dissolved Ga is susceptible to increased scavenging at P4 along the continental margin as a result of sediment transport, nepheloid layers, and/or increased coastal phytoplankton abundance.  Similar observations of increased scavenging of Ga along continental margins are observed along the   70 California coast (Orians and Bruland 1988b) and along the shelf in the western Arctic Ocean (this thesis, Chapter 5).  While correlation is similar for P12 150m 2010 (Figure 3.8b), anonymously spicy waters at 200m in 2010 (open green triangle) could result from the impact of an eddy at P12 in 2010.  At P12 150m in 2011 scavenging may decrease Ga concentrations, as suggested at P4 in 2011.  Biological and physical processes may therefore impact the correlation of dissolved Ga and spice at stations P4 and P12.  However, this does not preclude spicy waters of the CUC as an input of trace metals at P4 and P12.    3.5.3 Geographic spice variability: North Pacific spice interface  Low spice waters of ~ -0.65 in the western portion of the transect along  = 26.7 – 26.9 are more prominent in 2010, extending to ~137.5°W, compared to confinement of these waters to ~144°W in 2011 (Figure 3.6).  Calculation of spice based on temperature and salinity measurements available from Argo floats allows a spice section to be plotted extending across the north Pacific along 50°N (Figure 3.9).  Line P is shown to transect an interface in spice at ~140°W, representing distinct sources of waters along  = 26.8 at P26 relative to P4 – P20. North Pacific waters at  = 26.8 are consistent with North Pacific Intermediate Water (NPIW) (Talley 1993).  Alternatively, these waters could represent inputs from the Alaska coast (Lam et al. 2006), sourced from recirculation of the Alaska Gyre (Whitney et al. 2005). Given the correlation of spice with dissolved Ga along the Line P transect (Figure 3.8), and the interannual variability in the infiltration of higher spice waters from the east to the western most stations along Line P (Figure 3.6), the spice interface along Line P represents an area of future research regarding trace metal availability and ecosystem structure of the northeast Pacific Ocean.    71  Figure 3.9: Spice calculated at 50°N across the north Pacific from Argo float data.  Plotting as a function of density reveals differential temperature and salinity characteristics of isopycnal surfaces across the north Pacific.  Triangles represent density of the local maximum concentrations of dissolved Ga at 150m at stations P4 – P26 demonstrating the interface in spice between P26 and P20 and the extension of spice values observed at P26 across the north western Pacific.    3.5.4 Temporal spice variability: Line P  The correlation of dissolved Ga with spice suggests a CUC source of Ga.  Thus, variation in spice as a function of depth, density, longitude, and time may suggest an influence in trace metal inputs along Line P.  Sections of spice plotted as a function of density highlight differential source waters across Line P in August of each year 2009 – 2012 (Figure 3.10).  While spice values at P4 and P12 are consistently high across the four years from 2009 – 2012, dynamic shifts in the influence of CUC at stations P16 and P20 occur at  = 26.3 – 26.6 (Figure 3.10).  Cuddies are estimated to form at rates of 4 – 8 per year (Pelland, et al. 2013), therefore frequent enough to be regularly observed features, and yet infrequent enough to yield heterogeneous distributions and inter-annual variability.  Longitudinal variation in low spice waters of NPIW from the west observed at P26 display interannual variation as well, providing an additional influence on spice variability across Line P.  Additionally, depth contours in Figure 3.10 demonstrate that trace metal sampling at depths of 100m, 150, and 200m intersect very different isopycnal surfaces and spice characteristics inter-annually and across Line P.     72     a. b.    73 c. d.  Figure 3.10: Spice calculated along Line P from CTD data in a) 2009, b) 2010, c) 2011, and d) 2012.  Horizontal dashed line at  = 26.5 kg m-3 and depth contours provide reference showing dynamic spice as a function of longitude across Line P as well as interannual variation.   Considering interannual variability in spice values along the Line P transect at  = 26.5 (Figure 3.11), indicated as a horizontal dashed line in Figure 3.10, relatively little variation in spice is observed at P4 in 2009, 2011, and 2012, while lower spice values in 2010 indicate upwelling (Figure 3.11).  Spice values of -0.2 to -0.3 support a California Undercurrent source and concentrations at the dissolved Ga maximum at P4 in 2011 of 8.0 pmol kg-1 are in agreement with observations of 7.3 pmol kg-1 off central California (Orians and Bruland 1988b).  In contrast to large variations in spice at P16 and P20 in 2009 and 2012 relative to 2010 and 2011, spice values converge across all four years approaching P26, supporting the greater influence of NPIW at P26, and therefore displaying less sensitivity due to variation in spicy waters from the south.  Dissolved Ga concentrations in the area of NPIW formation reported in the western north Pacific   74 at 50°N, 167°E at depths consistent with NPIW were 4.2 – 7.2 pmol kg-1 (Shiller and Bairamadgi 2006) are consistent with concentrations of 4.5 – 5.1 pmol kg-1 measured at P26 in 2011 and 2012 in this work, and concentrations of 7.1 pmol kg-1 in 2010 and 6.5 pmol kg-1 in 1983 (Orians and Bruland 1988b).  Advection from the margins along isopycnals is not indicated as a source of the dissolved Ga maximum observed within the north Pacific subtropical gyre (Orians and Bruland 1988b).  Transport by cuddies can cross isopycnals (Pelland, et al. 2013) and therefore advection from the margins by cuddies may influence dissolved Ga in the northeast Pacific.     Figure 3.11: Spice values along Line P at  = 26.5 kg m-3, years 2010 and 2011 in black show higher spice values at P16 and P20 relative to 2009 and 2012 in grey.  All years show a decrease in spice approaching P26.    While spice values at  = 26.5 decrease nearly linearly along Line P from P12 to P26 in 2009 and 2012, spice values remain comparatively elevated across stations P16 and P20 in 2010 and 2011 (Figure 3.11).  High spice values observed at P16 and P20 in 2010 and 2011 at 150m (Figure 3.6) are associated with positive salinity anomalies identified in analysis available from the Institute of Ocean Science (www.waterproperties.ca/linep).  Alternatively, negative salinity anomalies are observed at these locations in 2009 and spice values are lower in 2009 at  = 26.3 – 26.6 at P16 and P20 compared to 2010 and 2011 (Figure 3.10).  El Niño conditions have been shown to result in increasing transport of waters from the south intersecting Line P (Schwing et   75 al. 2002), consistent with observed positive salinity anomalies and spicier waters in 2010 and 2011, relative to 2009.  Increased new production and maxima in particulate flux are observed along Line P in El Niño years (Wong, et al. 2002; Wong, et al. 1999).  Correlations of dissolved Ga with spice and interannual variations in spice associated with El Niño therefore suggest advective trace metal inputs may influence phytoplankton dynamics in the northeast Pacific.   Fronts in spice bisect Line P between P20 and P26 (Figure 3.10) and are coincident with HNLC gradients along Line P.  Trace metals associated with local Ga maxima at 150m may be sourced to the surface during winter mixing to the halocline.  Mixed layer depth is controlled by salinity in the winter, as opposed to shallow summer mixed layer depths dictated by temperature (Kara, et al. 2000).  Future research is required to investigate if correlations between dissolved Ga and spice impact availability of additional trace metals along Line P and ultimately phytoplankton growth.  3.5.5 Dissolved Ga: Deep north Pacific  Dissolved Ga increases at depths below 1500m in the northeast, northwest, and north central Pacific (Orians and Bruland 1988b; Shiller and Bairamadgi 2006).  Increasing concentrations of dissolved Ga within North Pacific Deep Waters (NPDW) have been suggested to result from sediment flux (Orians and Bruland 1988b).  While a sediment source of Ga to North Atlantic Deep Water (NADW) was proposed by Shiller (1998), Shiller and Bairamadgi (2006) suggest that dissolved Ga concentrations in the deep north Pacific (>3500m) are consistent with Antarctic Bottom Waters (AABW) and imply that a sediment source of dissolved Ga would be minimal.  Alternatively Shiller and Bairamadgi (2006) suggest that the increasing dissolved Ga concentrations observed in the central north Pacific result from a mixing of local dissolved Ga minima, suggested to be of North Pacific Intermediate Water (NPIW) origin, with AABW.    This work presents the first profiles of dissolved Ga to 2000m at stations P20, P16, and P12 along Line P.  Comparison of profiles indicates increasing concentrations of dissolved Ga at   76 2000m at P20, P16, and P12 (Figure 3.2), similar to observed increases at P26 (Figure 3.5). While this work is unable to assess the mechanisms leading to the increase of dissolved Ga with depth, sampling to 4000m depths along Line P is encouraged to evaluate controls on NPDW trace element distributions.    3.5.6 Ga/Pb ratio  Local maxima in trace element concentrations along Line P, as observed here for dissolved Ga are also observed for dissolved lead (Pb) (Charters 2012).  Considering the Ga/Pb ratio as a means to evaluate the comparative behaviour of these two elements, very similar Ga/Pb ratios are observed at P16, P20, and P26 below the 30m summer mixed layer (Figure 3.12).  Within the mixed layer, some variability is observed at stations P16 – P26.  However at P4 and P12 much higher Ga/Pb ratios are observed relative to P16 – P26 (Figure 3.12).  Below the mixed layer, the Ga/Pb ratio at 40m at P4 converges with ratios measured at P16 – P26.  Conversely, station P12 retains the Ga/Pb ratio of the mixed layer at 40m, and the ratio remains slightly elevated relative to P16 – P26 at 75m and 100m before converging at 150m and 200m with Ga/Pb ratios observed along the remainder of the transect.      77  Figure 3.12: Ga / Pb ratio at stations P4 – P26 across Line P.  Note log depth scale to emphasize shallow dynamics associated with the 30m mixed layer depth.    Both Ga and Pb may be sourced to the ocean by atmospheric deposition and fluvial freshwater sources.  Elevated ratios of Ga/Pb within the mixed layer at stations P4 and P12 may result from different inputs at these two locations relative to the remainder of the transect.  Multiple opportunities exist for relative fractionation of Ga and Pb sourced from rivers.  Differential weathering of terrestrial sources influence trace metal sources in rivers, as demonstrated for Ga (Shiller and Frilot 1996, Shiller 1988).  Therefore Ga/Pb ratios may be fractionated within freshwater river sources.  For example, mean Ga/Pb x 10 = 10.4 with median 8.0, and range 0.9 – 34.0 for n=8 rivers (Gaillardet, et al. 2003).  Fractionation of the Ga/Pb ratios can also occur during fluvial freshwater inputs of Ga and Pb to the ocean.  Both elements will be scavenged during mixing with seawater within river plumes. Additionally, removal of Pb occurs upon metal oxide surfaces (Nelson, et al. 1999; Tebo, et al. 2004).    Therefore, Ga/Pb ratios are expected to be higher within fluvial sources given differential weathering during river transport and the additional removal mechanism of dissolved Pb to metal oxide surfaces during mixing of fluvial freshwaters with seawater.  Indeed, elevated Ga/Pb ratios are associated with lower salinities, whereas lower Ga/Pb ratios observed at P16 – P26 occur at   78 higher salinities (Figure 3.13).  Higher ratios of Ga/Pb at stations P4 and P12 are therefore interpreted here as a tracer of freshwater fluvial sources to stations P4 and P12.    Figure 3.13: Ga/Pb ratio plotted as a function of salinity, elevated Ga/Pb ratios at low salinity support Ga/Pb as a tracer of fluvial freshwater sources  While station P4 is along the continental slope, P12 is westward of the continental margin.  Therefore freshwater fluvial sources may not be expected at P12.  However Haida eddies are know to influence Line P (Whitney and Robert 2002) and an eddy was in the area of station P12 at the time of sampling (visualized by sea surface altimetry from the Colorado Center for Astrodynamics Research).  Convergence of P12 Ga/Pb ratios at 150m (Figure 3.12) suggests that the distributions of Ga and Pb at 150m are controlled by the influence of advectively sourced high spice waters below the low spice of the winter mixed layer.  Elevated Ga/Pb, established by fluvial freshwaters at the source of the eddy, are observed to penetrate the summer mixed layer to 40m and are diluted during transport within the eddy.  Eddy influence indicated by the Ga/Pb ratio demonstrates the ability of eddies to transport coastal fluvial water far offshore within the northeast Pacific, including intersecting Line P.  This thesis presents the Ga/Pb ratio as a tracer   79 of fluvial freshwater inputs, an important global source of trace metals and nutrients to the ocean (Gaillardet, et al. 2003).  3.5.7 Dissolved Ga: Eddy influence  Eddies intersecting Line P have been shown to supply Fe along Line P (Johnson, et al. 2005), while concentrations of dissolved Pb have been shown to be lower within eddies relative to surrounding waters (Charters 2012).  Transport of dissolved Al from coastal freshwater sources occurs within eddies of the northeast Pacific (Brown, et al. 2012).  Consistent with the eddy at P12 in 2010, elevated concentrations of dissolved Al were observed at P12 in 2010 relative to 2011 (Cain 2014).  Higher concentrations of dissolved Ga are also present at P12 in 2010 relative to 2011 (Figure 3.14), representing the first description of dissolved Ga associated with eddies.                   a.                                            b. Figure 3.14: Concentrations of dissolved Ga at P12 in 2010 and 2011, note log depth scale to emphasize shallow water dynamics.  Elevated concentrations in 2010 of dissolved Ga as a function of a. depth and b. density are interpreted to result from the presence of an eddy.        80 3.6 Conclusion  This work reports the first transect of dissolved Ga across Line P in the northeast Pacific Ocean and identifies new applications of dissolved Ga as a tracer of ocean biogeochemical cycling.  Dissolved Ga profiles reported here show that local maxima concentrations of dissolved Ga at 150m – 200m correlate positively with the variable spice.  This correlation is interpreted to represent the California Under Current as an advective source of trace metals along the eastern portion of Line P, extending to stations P16 and P20.  Station P26 (OSP) is associated with lower spice and thus lower concentrations of dissolved Ga indicating advected North Pacific Intermediate Waters as a lower trace metal source.  Fronts in spice bisecting Line P demonstrate inter-annual variability.  Results therefore encourage future work investigating the impact of additional trace metal sources and controls on annual and longitudinal shifts in phytoplankton production along Line P.  A second tracer application is identified combining dissolved Ga and Pb and demonstrating the Ga/Pb ratio as a potential tracer of fluvial freshwater inputs to the ocean.  This ratio was indeed used to identify offshore transport of fluvial waters within an eddy.  Given annual variation in river runoff, and the ephemeral nature of eddies, fluvial inputs represent an additional area of future research regarding controls on phytoplankton dynamics and fisheries in the northeast Pacific Ocean.  Finally, this work provides an interdecadal comparison for dissolved Ga at OSP, a future GEOTRACES crossover station.      81 Chapter 4: Geographic and temporal variability of Pb isotopes in the northeast Pacific Ocean  4.1 Synopsis  This thesis presents the first profiles of radiogenic stable isotopic ratios of lead (Pb) along the Line P transect in the north east Pacific Ocean including Ocean Station Papa (OSP), a future GEOTRACES crossover station.  Line P intersects the Alaska Gyre along the eastern extent of the transect, wherein nitrate sources are often underutilized due to iron (Fe) limitation.  A gradient of High Nutrient Low Chlorophyll (HNLC) conditions therefore exists along Line P.  Isotopic Pb ratios can identify the origin of Pb inputs to the ocean, informing potential Fe and trace metal sources to this HNLC region of the ocean.  Isotopic ratios of Pb reported here reveal geographically and temporally distinct sources of Pb along the Line P transect.  Isotopic ratios of 206Pb/207Pb and 208Pb/206Pb at OSP, the westernmost station of the transect, indicate eolian sources from Asia.  In contrast, isotopic Pb ratios at stations east of OSP are consistent with mixing of local coastal terrestrial ratios.  Isotopic ratios of Pb and Ga/Pb together suggest that these terrestrial signatures are representative of fluvial freshwater sources.  Terrestrial Pb inputs are restricted to the summer mixed layer.  In contrast, isotopic ratios of Pb indicating Asian sources at OSP extend to the winter mixed layer.  This thesis therefore identifies geographic and temporal fronts of Pb sources along Line P.  Additionally, correlation is identified between Pb isotope ratios and HNLC conditions.  Isotopic Pb ratios along Line P transition from eolian Asian inputs in the west to coastal fluvial freshwaters.  These distinct geographic sources of Pb correspond to the gradient in HNLC characteristics along Line P.  Isotopic ratios of Pb reported here are obtained from 50mL seawater samples analyzed on the Nu AttoM single collector ICP-MS.  Reference standard results for 206Pb/207Pb and 208Pb/206Pb compare favourably (p < 0.05) with published values.  Smaller sample volume requirements reduce logistical barriers and promote more routine analysis of Pb isotopes in seawater.      82 4.2 Introduction  Radiogenic stable lead (Pb) isotope ratios of 206Pb/207Pb and 208Pb/206Pb provide a geochemical tracer of Pb sources to the environment (Cheng and Hu 2010; Komarek, et al. 2008).  Characteristic Pb ratios are a function of source location and age (Cowart and Burnett 1994), originating from the production of 206Pb, 207Pb, 208Pb from decay series initiated by 238U, 235U, and 232Th, respectively.   Application of Pb isotope ratios has demonstrated anthropogenic release of Pb to the environment during the industrial revolution (Graney, et al. 1995; Weiss, et al. 1999a) including the introduction of leaded gasoline (Schaule and Patterson 1981; Weiss, et al. 2003).  Pb isotopic ratios are now tracking the decline in Pb following the phase out of Pb in gasoline (Wu and Boyle 1997a; Boyle, et al. 2014, Wang et al. 2006).  However, atmospheric sources of Pb remain, such as coal combustion (Chen, et al. 2008) and mining (Csavina, et al. 2012).  Pb isotopes have also been utilized to trace Pb sources in the past (Weiss, et al. 1999b), from records contained in lake sediments (Outridge, et al. 2002) and ice (Shotyk, et al. 2005).    The ocean behaves as an additional reservoir providing an archive of anthropogenic Pb.  Atmospheric deposition of anthropogenic Pb to the ocean is captured by isotopic ratios of Pb within banded corals (Inoue and Tanimizu 2008) and time series profiles in the ocean (Boyle, et al. 2014).  Differential Pb sources to the surface ocean are evidenced by Pb isotope ratios within the northwest Pacific (Gallon, et al. 2011).  In addition, fluvial inputs to the ocean can be identified by isotopic ratios of Pb (Alleman, et al. 2000).  While Pb isotope ratios within the ocean provide a tracer of pollution sources, Pb isotope ratios can also be used to trace oceanic physical and chemical processes.  Pb isotopic ratios can delineate and identify water masses (Alleman, et al. 2001), upwelling events (Flegal, et al. 1989), paleo-ocanographic circulation (Banner 2004; Frank 2002), and demonstrate reversible exchange   83 of Pb between the dissolved and particulate phase (Sherrell and Boyle 1992; Wu, et al. 2010; Nagaoka, et al. 2010).    This work presents Pb isotope ratios measured in seawater analyzed with the Nu AttoM single collector Inductively Coupled Plasma Mass Spectrometer (ICP-MS), utilizing sample volumes of 50ml.  The Nu AttoM provides increased sensitivity and precision among single collector ICPMS instruments.  Control of ion trajectory by deflector plates at the entrance and exit to the magnet allows the magnet to remain stable.  Sample volume requirements of only 50mL reduce logistical barriers to sample collection and analysis and facilitates the determination of Pb isotopes along Line P, an oceanographic transect in the northeast Pacific Ocean.    Line P represents one of the longest oceanographic time series, initiated in 1956 (Freeland 2007).  Line P begins near the coast of Vancouver Island and continues ~1500km into the northeast Pacific Ocean.  Upwelling along the coast provides nutrients such as nitrate and support phytoplankton blooms.  In contrast, upwelling within the Alaska Gyre along the western portion of Line P is characterized by low phytoplankton abundance, despite abundant nitrate sources.  This enigmatic High Nitrate Low Chlorophyll (HNLC) area along Line P is Fe limited (Boyd, et al. 1996).  Inputs of Fe along Line P include eolian (Bishop, et al. 2002), volcanic (Hamme, et al. 2010; Melancon, et al. 2014), coastal upwelling and transport (Cullen, et al. 2009), and eddy sources (Johnson, et al. 2005).   This work applies Pb isotopic ratios to investigate trace metal inputs to Line P.  Geographic sources from Asia and North America are delineated.  Additionally, inputs are differentiated temporally between summer and winter.  Finally, inputs identified here are shown to correlate with observed HNLC conditions along Line P.        84 4.3 Methods  Seawater samples were collected for trace metal analysis at 5 stations in the northeast Pacific along the Line P transect (Figure 4.1) between August 20 – 27, 2010.  Seawater samples at depths greater than 40m were collected in Niskin X bottles mounted on Kevlar coated rope.  Samples collected at 5, 10, 25, and 40m were collected utilizing a Teflon pump and Teflon tubing attached to Kevlar coated rope.  Surface samples were collected from a Zodiac boat, hundreds of meters from the ship, directly into clean sample bottles by dipping ~0.5m below the surface with gloved hands and clean plastic protected forearms.  Samples were collected into Bel Art LDPE bottles and filtered to 0.45m.  Samples were acidified at sea, 1mL 12M HCl (Seastar Chemicals) to 1L seawater sample, within a HEPA flow hood.  All sample bottles were cleaned prior to the cruise with dilute surfactant solution, 6M HCl, (Environmental Grade) and then 0.7M HNO3 (Seastar Chemicals).  Bottles were thoroughly rinsed with seawater prior to sample collection.     Figure 4.1. Pb isotope results reported from Stations P4 – P26 along Line P in the northeast Pacific Ocean.  Additionally, sample sites referenced in the Discussion of reported Pb isotope results from the north and south of Vancouver Island, a coastal fjord, and the Nass River are indicated for reference.      85 Analysis of Pb isotopes in seawater has been performed following extraction of Pb from seawater sample volumes of 4L and 2L (Flegal, et al. 1986; Sanudo-Wilhelmy and Flegal 1994; Flegal, et al. 1989), 0.1L – 1L (Gallon, et al. 2011), and 0.25L – 0.8L (Wu, et al. 2010) and 29mL (Weiss, et al. 2000).  Recent demonstration of Pb isotopic analysis utilizing < 0.2ng Pb (Newman and Georg. 2012) would allow analysis of 50mL seawater sample volumes given Pb concentrations > 20 pmol kg-1.  Concentrations of Pb measured from the surface to 400m along Line P range from 30-80 pmol kg-1 (Charters 2012).  Therefore this work will utilize 50mL seawater samples for analysis of Pb isotopes.    Preconcentration of Pb from seawater samples is performed by established methods of magnesium coprecipitation with ammonium hydroxide (Wu and Boyle 1997b), followed by additional matrix removal by ion exchange (described in Shiel, et al. 2010).  Mass bias effects were corrected for by standard bracketing with NBS-981 and application of a linear correction factor based on established ratios of NBS-981 of 206Pb/207Pb=1.0932 and 208Pb/206Pb=2.1677 (Abouchami, et al. 1999).  Relative standard deviation (RSD) of 206Pb/207Pb for NBS-981 is 0.1% (n=86), comparable to results of Newman and Georg (2012).  Total Pb concentrated from seawater samples and analyzed was 0.2 – 0.4ng Pb.  Blank subtraction resulted in a standard deviation of ≤ 0.0006 between blank corrected and uncorrected ratios which is within instrumental precision.  GEOTRACES reference standards were analyzed and resulted in 206Pb/207Pb = 1.174 +/- 0.0031 and 1.180 +/- 0.0025 for samples GSI and GDI, respectively.  These values compare well (p < 0.05) to those previously reported (Boyle, et al. 2012).    4.4 Results  Isotopic ratios of dissolved Pb (Table 4.1, Table 4.2) plotted in Figure 4.2a, b display unique characteristics both within individual depth profiles, and longitudinally across the Line P transect.  Profiles of 206Pb/207Pb (Figure 4.2a) and 208Pb/206Pb (Figure 4.2b) are plotted with a log base10 depth axis to aid visualization of high resolution shallow sampling depths of 0.5, 5, 10, 25, and 40m.  Combination of the three isotopes 206Pb, 207Pb, and 208Pb in plots of 206Pb/207Pb vs.   86 208Pb/206Pb (Figure 4.2c) further differentiates samples.  Dotted lines in 206Pb/207Pb vs. 208Pb/206Pb plots are provided as a reference for comparison between plots.  Deeper samples, plotted as circles, are centered near the intersection of the two lines and are similar across the transect. Triangles are isotopically differentiated (Figure 4.2c) and are used to identify Pb sources to Line P.       87 Table 4.1. 206Pb/207Pb ratios and standard deviation (SD) at stations P26 – P4 across Line P z (m) P26 P20 P16 P12 P4 206Pb/207Pb SD 206Pb/207Pb SD 206Pb/207Pb SD 206Pb/207Pb SD 206Pb/207Pb SD 0     1.166 0.005 1.173 0.004 1.173 0.007 1.177 0.001 5 1.138 0.003 1.164 0.006 1.169 0.007 1.177 0.006 1.181 0.001 10 1.145 0.005 1.172 0.001 1.161 0.004 1.181 0.005 1.176 0.003 25 1.144 0.003 1.172 0.002 1.171 0.006 1.180 0.002 1.175 0.002 40 1.147 0.001 1.159 0.002 1.168 0.001 1.162 0.004 1.166 0.002 75 1.144 0.001 1.162 0.003 1.166 0.004 1.165 0.002   100 1.163 0.004 1.163 0.003 1.164 0.002 1.167 0.003   150 1.170 0.001 1.162 0.002 1.165 0.002 1.161 0.001   200 1.161 0.004 1.165 0.003 1.164 0.002 1.161 0.004   300 1.164 0.003 1.161 0.002 1.171 0.003 1.168 0.002   400 1.167 0.004 1.163 0.002 1.170 0.002           Table 4.2. 208Pb/206Pb ratios and standard deviation (SD) at stations P26 – P4 across Line P z (m) P26 P20 P16 P12 P4 208Pb/206Pb SD 208Pb/206Pb SD 208Pb/206Pb SD 208Pb/206Pb SD 208Pb/206Pb SD 0   2.083 0.006 2.070 0.006 2.067 0.010 2.077 0.003 5 2.111 0.005 2.084 0.006 2.084 0.011 2.068 0.004 2.073 0.003 10 2.112 0.002 2.083 0.006 2.090 0.012 2.056 0.006 2.079 0.003 25 2.109 0.008 2.079 0.004 2.077 0.005 2.070 0.004 2.079 0.002 40 2.111 0.004 2.098 0.003 2.099 0.002 2.097 0.005 2.099 0.003 75 2.122 0.001 2.092 0.004 2.096 0.003 2.096 0.003   100 2.094 0.006 2.096 0.002 2.100 0.003 2.095 0.002   150 2.095 0.002 2.099 0.003 2.099 0.003 2.115 0.002   200 2.087 0.007 2.100 0.003 2.096 0.002 2.100 0.007   300 2.093 0.005 2.092 0.001 2.095 0.004 2.097 0.003   400 2.088 0.003 2.097 0.006 2.094 0.003             88   Figure 4.2. Profiles of 206Pb/207Pb at stations P4 – P26 across the Line P transect and isotope - isotope plots of 206Pb/207Pb vs. 208Pb/206Pb.  Horizontal dashed lines in profiles represent summer mixed layer depth at P4 – P20, and winter mixed layer at P26.  Dashed lines in 206Pb/207Pb vs. 208Pb/206Pb plots provide reference to orientate data.  Triangles represent samples associated with the summer mixed layer at stations P4 – P20 and the winter mixed layer at P26.   Beginning at station P4, samples collected at 0.5, 5, 10, and 25m display 206Pb/207Pb ≥ 1.175 (Range = 1.175 – 1.181) (Table 3.1) and are all significantly higher (p < 0.05) than 206Pb/207Pb = 1.166 at 40m (Figure 4.2a).  Similar statistically significant (p < 0.05) differences are observed for 208Pb/206Pb ratios at P4 as 0 – 25m depths display 208Pb/206Pb ≤ 2.079 (Range = 2.073 – 2.079), contrasting with 208Pb/206Pb = 2.099 at 40m (Figure 4.2b).  Isotopic distinction is further   89 visualized in 206Pb/207Pb vs. 208Pb/206Pb space (Figure 4.2c) as samples at depths 0 – 25m (triangles) plot in the upper left quadrant of the figure, exhibiting higher ratios of 206Pb/207Pb and lower of 208Pb/206Pb, relative to the 40m sample (circle).    Continuing to P12, samples at 0 – 25m display 206Pb/207Pb ≥ 1.173 (Range = 1.173 – 1.181), significantly higher (p < 0.05) than the 1.162 value at 40m.  The values at 0 – 25m for 208Pb/206Pb ≤ 2.070 (Range = 2.056 – 2.070) and are statistically lower (p < 0.05) than the 208Pb/206Pb = 2.097 at 40m.  Plots in 206Pb/207Pb vs. 208Pb/206Pb space (Figure 4.2c) differentiate samples at 0 – 25m (triangles) from the remainder of the profile (circles), similar to observations at P4. This work therefore identifies a depth isotopic front along the western extent of Line P at stations P4 and P12.  Divergent patterns of 206Pb/207Pb and 208Pb/206Pb ratios are observed at P26 relative to P4 and P12 (Figure 4.2).  Isotopic ratios at P4 and P12 are 206Pb/207Pb = 1.173 – 1.181, contrasting with 206Pb/207Pb = 1.138 – 1.144 at P26.  Similar to P4 and P12, an isotopic interface is observed as a function of depth, as all 206Pb/207Pb ratios from 5m – 75m at P26 are statistically (p < 0.05) different from 206Pb/207Pb = 1.163 at 100m.  However while the depth interface at P4 and P12 is between 25 – 40m, the interface at P26 is located between 75 – 100m.  This same isotopic interface between 75m and 100m at P26 is observed for 208Pb/206Pb (Figure 4.2b), and triangles representing depths of 5 – 75m at P26 in 206Pb/207Pb vs. 208Pb/206Pb space plot opposite of the trend observed at P4 and P12 (Figure 4.2c).  Distinct depth interfaces at P26 relative to P4 and P12 will be investigated in the Discussion.   Plots of 206Pb/207Pb vs. 208Pb/206Pb at P16 and P20 reveal slight isotopic shifts toward the upper left quadrant, similar to trends observed at P4 and P12, although less pronounced.  However, shallow samples are not consistent with the observation that P26 plots in the lower right quadrant.  Similarly, while the isotopic shifts at P16 and P20 are not as clearly defined by a shallow interface as observations at P4 and P12, a deep interface analogous to P26 is not observed.      90 Two distinct patterns are therefore observed along Line P.  First, while isotopic ratios below 25m at stations P4 – P20 and below 75m at P26 are very similar – plotting near the center of the 206Pb/207Pb vs. 208Pb/206Pb plots – Pb isotope ratios above these depth interfaces are distinct.  Second, ratios at depths shallower than the interface vary longitudinally along the Line P transect.  Isotopic shifts toward more radiogenic ratios at stations P4 – P20 plot in the upper left quadrant and are confined to depths < 40m.  Isotopic ratios at P26 plot in the lower right quadrant, indicating lower 206Pb/207Pb and higher 208Pb/206Pb, with an interface at 75 – 100m.  Two Pb isotopic regimes are therefore identified, bisecting the transect between P20 and P26.  Origin of these differential Pb isotope ratios will be explored in the Discussion section.    4.5 Discussion 4.5.1 Temporal differentiation  Depth interfaces in Pb isotopic ratios at stations P4 – P20 suggest an input of Pb to the upper 25m that does not penetrate to 40m.  This depth interface is interpreted to represent a summer input of Pb to the mixed layer, as Pb inputs to the surface during the summer would only be mixed down to the summer mixed layer depth (~30m).  Does the summer mixed layer correspond to the Pb isotope interface?  Here, the mixed layer is identified by d/dz, where  represents potential density (kg m-3) and z = depth (m).   Contouring a section of d/dz across Line P allows visualization of the mixed layer (Figure 4.3).  Warm colours identify the density gradient associated with the base of the mixed layer, extending across the transect from P26 to P16 at ~25 – 30m.  Mixing to depths of 40m would therefore be prevented by the density gradient established by the summer mixed layer.  Weakening of the density gradient at P4 results from upwelling observed at the time of sampling; a summer mixed layer depth similar to the remainder of the transect could be expected during periods of less upwelling.         91  Figure 4.3. Contoured section of d / dz along Line P in August 2010, warm colours indicating a density gradient represent the base of the summer mixed layer  Confinement of the distinct Pb isotopic ratios at stations P4 – P20 to the summer mixed layer (Figure 4.2) suggests a seasonal source of Pb to the surface. Concentrations of Pb increase with depth through the winter mixed layer (Charters 2012).  Therefore winter mixing at stations P4 – P20 would overprint the summer mixed layer signal and remove the interface of Pb isotope ratios observed between 25 – 40m.  Additionally, the interface could be removed by the process of reversible exchange of Pb on sinking particles, resulting in the vertical transport of surface source Pb ratios to greater depths in the water column (Wu, et al. 2010; Sherrell and Boyle 1992). Particle sinking rates of 50 – 100m per year (Yoon, et al. 2001) and reversible exchange of particulate Pb would result in the penetration of this source signal to depths > 25m over annual cycles, supporting the Pb isotopic interface between 25 – 40m as a seasonal event.    In contrast, station P26 does not exhibit a Pb isotopic interface associated with the summer mixed layer, rather the interface occurs between 75 – 100m.  Calculation of d/dz during the February 2010 Line P cruise from archived data (www.waterproperties.ca/linep) reveals winter mixed layer depths of 75 – 100m at stations P4 – P20 (Figure 4.4a).  Unfortunately stations west of P20 were not sampled due to inclement weather.  However, a density profile for P26 in February 2010 can be calculated based on Argo float data (Figure 4.4b) and reveals a similar   92 mixed layer depth of 75 – 100m.  Therefore, the Pb isotope interface observed at P26 in August 2010 is coincident with the 2010 winter mixed layer, suggesting that the isotopically distinct Pb source observed associated with the summer mixed later at P4 – P20 is not present at P26.  `              a.              b.  Figure 4.4.  a. Contoured section of d / dz along Line P in February 2010, warm colours indicating a density gradient represent the base of the winter mixed layer, b. profile of sq at 50°N 145°W (OSP) in February 2010 from Argo float data indicates a winter mixed layer depth of 75 – 100m.    93 This work therefore identifies temporally distinct sources of Pb to the northeast Pacific Ocean along Line P.  Observed 206Pb / 207Pb and 208Pb / 206Pb interfaces as a function of depth (Figure 4.2) result from distinct ratios within the summer mixed layer at stations P4 – 20 (Figure 4.3), and the winter mixed layer at P26 (Figure 4.4).  Sources of Pb to P4 – P20 in the summer are not present at P26.  While summer inputs to P26 are not precluded, an isotopically distinct signature of Pb associated exclusively with the summer mixed layer is not evident at P26. Additionally, isotopic ratios associated with the winter mixed layer at P26 are not observed at P4 – P20 at winter mixed layer depths of 40 – 75m.  Following temporal differentiation of Pb sources across the Line P transect established here, geographic sources of this Pb are investigated in the next section.     4.5.2 Geographic source differentiation  In addition to a depth interface in Pb isotope ratios, a longitudinal interface is observed between ratios at P26 and the remainder of the transect (Figure 4.2).  Samples associated with the summer mixed layer at stations P4 – P20 and the winter mixed layer for P26 are compared in 206Pb / 207Pb vs. 208Pb / 206Pb space (Figure 4.5).  Ratios associated with stations P4 and P12 are more radiogenic, owing to greater relative amounts of 206Pb than at P26.  Stations P16 and P20 represent a transition in Pb isotope ratios between stations P4 and P12 in the east and P26 in the west (Figure 4.5).  Concentrations of Pb within the mixed layer increase along Line P (Charters 2012), with concentrations of 10 – 20 pmol kg-1 at stations P4 and P12 increasing to 30 – 40 pmol kg-1 at stations P16 and P20, and 40 – 50 pmol kg-1 at P26.  It is possible therefore that inputs of Pb from North America may not influence the Pb isotope signal at P26.  However, given the higher concentrations of Pb at P26, the interface of Pb isotope ratios across Line P is also supported by ratios from P26 not overprinting the ratios along the eastern portion of the transect.       94  Figure 4.5. Comparison of isotopic ratios of Pb measured in the summer mixed layer (P4 – P20) and winter mixed layer (P26).  Higher 206Pb/207Pb values associated with the upper left quadrant are from stations P4 – P20, and represent more radiogenic dissolved Pb ratios relative to P26 in the lower right quadrant.  Note that symbol colour identifying stations here are consistent with Figure 4.2.  Interpretation of Figure 4.6, Figure 4.7, and Figure 4.8 is aided by application of a colour scheme specific to those figures.    What are the origins of the isotopic ratios of the eastern and western endmembers of the Line P transect?  Isotopic Pb ratios reported in the literature will be compared to ratios measured along Line P to provide suggestions of source origin.  Literature values represent Pb isotopic ratios of a wide range of sample types.  Filtered air samples are compared here to represent Pb isotopic ratios of potential eolian inputs to Line P.  Additionally, terrestrial Pb sources located proximal to Line P are included to evaluate potential fluvial sources to Line P.    While comparison of Pb isotope ratios from the literature can help elucidate Pb sources measured in this work, limitations exist, particularly regarding comparison of ratios from samples collected years apart.  However, given that ratio stability is observed (Bollhofer and Rosman 2002) the focus will be on the primary isotopic signature of a source region.  Reference lines plotted in Figure 4.2c and Figure 4.5 will be retained throughout all following plots, easing orientation as axis scales change to accommodate additional literature data.  In addition to these reference lines,   95 colours facilitate interpretation and discussion of Pb as samples plotting in the lower right quadrant will be red, upper left quadrant will be green, and the lower left quadrant will be purple.  The red and green quadrants are the primary focus of this discussion.  Data points in purple can simply be considered as a linear transition between data plotting in the red and green quadrants.   4.5.2.1 Pb sources from Asia  Eolian sources from Asia are deposited into the north Pacific, providing a potential source of Pb (Gross, et al. 2012) as well as Fe (Grand, et al. 2014).  Reports of Pb isotope ratios from filtered air samples across Asia are compared here to ratios measured along Line P.  Geographic sources from Asia are grouped into two broad categories based on interpretation of 206Pb / 207Pb vs. 208Pb / 206Pb plots, locations in Asia at latitudes > 40oN and < 40oN.    Isotopic ratios for samples from > 40oN (Bollhofer and Rosman 2001) are plotted as circles with the upper half filled red (Figure 4.6a), with sample locations mapped for reference (Figure 4.6b).  Samples from > 40oN plot in the lower right quadrant, similar to ratios observed in the winter mixed layer at P26 and represent less radiogenic Pb.  Filtered air samples from < 40oN (Bollhofer and Rosman 2001) are plotted as circles with the lower half filled red (Figure 4.6a, b).  While the majority of the samples from < 40oN plot in the lower right quadrant, 206Pb / 207Pb and 208Pb / 206Pb values are generally higher than samples at > 40oN.  Higher relative 208Pb / 206Pb values in samples from < 40oN are consistent with Pb sourced from China (Cheng and Hu 2010). Similar Pb isotope ratios observed in sources from filtered air in Shanghai are associated with anthropogenic inputs from coal (Chen, et al. 2008).  Sources of Pb in air masses from Asia therefore exhibit isotopic Pb ratios within the range enclosed by the red ellipse in Figure 4.6.          96  a. b.  Figure 4.6. a. Isotopic ratios of Pb collected from filtered air sampled from locations in Asia at >40°N (red circles, top half filled) (see also map in b.), and <40°N (red circles, bottom half filled) (Bollhofer and Rosman 2001), and enclosed by the red ellipse, a time series study demonstrating temporal variability (Wang et al. 2006) is represented by red dashes.  Also plotted are isotopic ratios of dissolved Pb within the surface ocean (Gallon et al. 2010) (see dashed blue ellipse in a. and map in b.), indicated by red circles outlined in blue, and samples from 50, 100, 200m depths at 30°N 140°W (red diamond outlined in blue) (Wu et al. 2010), ocean samples are indicated by the dashed blue ellipse.  P26 samples measured in this work (red inverted triangles in a. and b.) are indicated by the black ellipse in a. and are enclosed within the red ellipse, supporting an Asian source of Pb at P26.  Green circles along the coast of North America plot proximal to stations P4 – P20 (green triangles) in the upper left quadrant (green) and will be discussed in section 4.5.2.2.   97  Samples from the P26 winter mixed layer, enclosed by the black ellipse (Figure 4.6), plot within the field of Asian airborne Pb.  Therefore Pb isotopes within the winter mixed layer at P26 are interpreted to be of Asian origin.  Isotopic ratios of dissolved Pb along a surface transect from the western to the central north Pacific (Gallon, et al. 2011) plot within the dashed blue ellipse (Figure 4.6), also enclosed within the red ellipse of eolian inputs of Asian origin.  Additionally, isotopic ratios of dissolved Pb at 50m, 100m, 200m sampled at 30oN, 150oW (Wu, et al. 2010) plot within the same dashed blue ellipse field (Figure 4.6) describing the Gallon et al. 2010 surface transect.  While Asian sources of Pb to P26 reported are consistent with previous studies, ratios at P26 are less radiogenic, suggesting sources of airborne Pb from > 40oN latitude.  Pb isotope ratios observed at P26 in August 2010 from 5m – 75m are interpreted to represent an eolian input of trace metals from Asia.   4.5.2.2 Pb sources from western North America  While P26 Pb ratios plot in the lower right quadrant and indicate Pb sources from Asia, ratios at stations P4 – P20 plot in the upper left quadrant and do not overlap with Asian sources.  Filtered air samples of Pb from central California (Bollhofer and Rosman 2002), indicated as green circles (Figure 4.6a), have Pb isotopic ratios similar to stations P4 – P20 (green triangles).  Analogous to filtered air samples, isotopic ratios of Pb in lichen represent a record of atmospheric Pb (Simonetti, et al. 2003).  Isotopic ratios in lichen across western Canada display a range of values (Figure 4.7), including the upper left quadrant (green) corresponding to ratios observed at stations P4 – P20.  Three conclusions are drawn from these results.  First, Pb ratios plotting in the green quadrant observed in North American samples (Figure 4.7a) are not observed in samples from Asia (Figure 4.6a).  Second, Pb isotope ratios from lichens sampled in western Canada that plot in the red quadrant are primarily on the eastern side of the Rocky Mountains (Figure 4.7b) and are near the cities of Calgary and Edmonton, suggesting an anthropogenic influence.  Third, while ratios plotting in the red quadrant are similar to ratios   98 measured at P26, these ratios are not observed at stations P4 – P20.  Additionally, the majority of lichen samples plotting in the red quadrant are more radiogenic than samples at P26.    Pb isotope ratios at stations P4 – P20 plot in the green quadrant and similar ratios are also observed within some lichen from North American sources.  Asian sources identified in section 4.5.2.1 are not consistent with Pb isotope ratios observed at stations P4 – P20.  Additional samples associated with North American sources are discussed below to further investigate North American source of Pb to stations P4 – P20.     99 a. b.  Figure 4.7. a. Isotopic ratios of Pb measured in lichen (Simonetti, et al. 2003) across western North America (see map in b.) are indicated by squares and plot within both the upper left and lower right quadrant.  P4 – P20 samples measured in this work (green triangles) identified by the black ellipse in the upper left quadrant are enclosed by the green ellipse indicating lichen samples plotting in the upper left quadrant.  See text for additional interpretation and discussion.    100 Pb isotopes reported from the area enclosed by the box in Figure 4.7b are plotted in Figure 4.8a,b and mapped in Figure 4.8c.  Isotopic ratios of airborne Pb are reported from Victoria and Burnaby (near Vancouver) British Columbia (Bollhofer and Rosman 2002), revealing opposing ratios as Victoria plots near ratios observed at P26, while Pb isotope ratios from Burnaby plot within the proximity of ratios measured in this work at stations P4 – P20.  While results from Victoria are consistent with ratios measured at P26, this influence is not observed at stations P4 – P20 and therefore may not be a significant source to the coastal ocean relative to other sources.  Samples representing Vancouver, BC road dust (Preciado, et al. 2007) plot largely within the green quadrant, while two road dust samples plot just outside and are coloured purple.  Airborne Pb samples from Vancouver and Burnaby therefore support a North American source of Pb to stations P4 – P20.  Dissolved Pb in the western and central north Pacific ocean was shown above to support interpretation of an Asian source of dissolved Pb measured at P26 (Figure 4.6).  Here, Pb isotopic ratio of oysters sampled along the west coast of Vancouver Island (Shiel, et al. 2012) provide a proxy for dissolved Pb in the ocean as filtering of ocean waters by oysters represents an application analogous to lichen sampling the air.  Three oyster samples plot within the green quadrant and one within the red quadrant, although at less radiogenic values than observed at P26 (Figure 4.8a).  While it was suggested above that Pb sources from Victoria would be unlikely to influence Pb isotope ratios at P26 without impacting stations P4 – P20 along Line P, Pb sources from Victoria could potentially impact oysters along the coast of Vancouver Island, and therefore influencing the isotopic ratio of the oyster sample plotting in the red quadrant.  Oyster samples therefore both support isotopic signatures observed at P4 – P20, while also being consistent with possible influence from less radiogenic isotopic ratios measured in air samples from Victoria.      101              a.                  b.   102               c. Figure 4.8. a. Pb isotope ratios of filtered air samples (Bollhofer and Rosman 2002) and dust collected from the air (Preciado, et al. 2007) within coastal lower BC and oysters (Shiel, et al. 2012) are indicated on the map in b.  All but two of these samples from lower BC plot within the green ellipse containing stations P4 – P12.  The remaining two samples suggest anthropogenic sources from Victoria.  c. Isotopic ratios of terrestrial samples: rock samples from Vancouver Island (diamonds) (Greene et al. 2009) and coastal BC waters draining the terrestrial watershed (circles) (Stukas, et al. 1999) (see also map in b).  Terrestrially derived samples are more radiogenic than values measured along Line P, thus relatively small amounts of terrestrial Pb would be required to influence Pb isotope ratios at P4 and P12 (green triangles).     Finally, Pb ratios of potential terrestrial fluvial freshwater inputs to Line P are much more radiogenic than Line P samples (Figure 4.8c).  Terrestrial Pb sources from rocks on the northern and southern end of Vancouver Island possess radiogenic Pb isotope ratios of 206Pb/207Pb of 1.20 – 1.25 (Greene, et al. 2009), compared to 206Pb/207Pb 1.16 – 1.18 for stations P4 – P20.  Dissolved Pb measured in the Nass River (Stukas, et al. 1999) demonstrates similarly elevated 206Pb/207Pb ratios, suggesting that radiogenic Pb could be supplied to the ocean via fluvial freshwater sources.  Ocean water sampled from a coastal fjord following a spring freshet (Stukas, et al. 1999) results in the highest 206Pb/207Pb ratios, suggesting that the watershed drains radiogenic terrestrial Pb.  Following winter renewal of the fjord with ocean waters, 206Pb/207Pb ratios (Stukas, et al. 1999) decrease to values similar those at P4 – P20, suggesting the dilution of   103 the more radiogenic ratios associated with the spring freshet.  Terrestrial sources of Pb transported by fluvial freshwaters could therefore influence the Pb isotopic ratio observed at stations P4 – P20.  Comparison and interpretation of samples representing potential sources of Pb measured at stations P4 – P20 reveals that the most probable source of Pb resulting in the depth interface associated with the summer mixed layer at P4 – P20 is from North America.    4.5.3 Physical inputs of Pb: Eolian and fluvial   Sources of Pb to the ocean may be eolian or fluvial and will be discussed here to evaluate the temporal and geographic differentiation of Pb isotopes identified in this work.  Temperature-salinity plots identify 4 distinct groupings of stations across the Line P transect (Figure 4.9), assisting interpretation of Pb sources.     Figure 4.9. Temperature-salinity of the summer mixed layer at stations across Line P in August 2010 reveal four distinct regions identified by cool, saline waters indicative of upwelling (blue), warmer lower salinity waters (green) containing P12, followed by P16 and P20 within stations of increasing salinity (red), and P26 within waters of the decreasing temperature (black).    While upwelling is observed at stations P2 – P5 (blue) (Figure 4.9, Figure 4.10), lower salinity waters are observed extending from P6 all the way to P13 (green), suggesting potential   104 freshwater influence.  While a salinity gradient represents a sharp transition between stations P13 and P14 (Figure 4.10b), a temperature gradient delineates stations P22 – P26 in T-S space (Figure 4.9).  Decreasing temperatures from P22 – P26 (black) are interpreted to represent the transect entering into the Alaska Gyre.  Stations P14 – P21 (red) represent a transition to the Alaska Gyre, from lower salinity waters at stations P6 – P13 to the cooler waters of the Alaska Gyre.     Figure 4.10. Temperature and salinity of the mixed layer along the Line P transect in August 2010 demonstrates the longitudinal extent of the four regions identified in the TS plot, upwelling is clearly demonstrated along the coast (blue), followed by low salinity waters along the remainder of the eastern portion of the transect (green), a sharp salinity gradient occurs approaching the edge of the upwelling Alaska Gyre (gyre), and a decreasing temperature gradient upon entering the Alaska Gyre (black).  These four distinct regions exhibit isotopic shifts along Line P (Figure 4.11).  Transition in 206Pb/207Pb values from P12 to P16 is consistent with the transition identified by the salinity gradient (green to red) (Figure 4.10).  Similar 206Pb/207Pb values at P16 and P20 (red) are   105 corroborated by the position of these two stations within the same region of T-S space (Figure 4.9).  Ratios at P26, associated with the Alaska Gyre based on T-S interpretation (black), represent Asian Pb sources, as identified in 4.5.2.1 above.  Transition toward the Alaska Gyre is therefore accompanied by a shift from high 206Pb/207Pb ratios associated with North American sources at P4 and P12, followed by intermediate ratios across the salinity gradient toward P16 and P20, to lower ratios at P26 upon entering the Alaska Gyre.     Figure 4.11. Mixed layer 206Pb/207Pb ratios measured across Line P.  Small grey circles show all samples from within the mixed layer.  Coloured circles corresponding to regions described in Figure 4.10 and Figure 4.11 and represent the average and the cumulative error of all measurements within the mixed layer at each station.  Note the 206Pb/207Pb scale is reversed to provide consistency with Figure 4.14.    Delineation of fluvial and eolian input process assists in interpreting patterns of Pb isotopes across Line P.  Fluvial freshwater sources are identified using the Ga/Pb tracer (Chapter 3) and are associated with low salinity waters at P4 and P12.  The observed pattern of Pb isotopes across Line P (Figure 4.11) is similar to Ga/Pb within the mixed layer (Figure 4.12).  The influence of fluvial sources of Pb on Pb isotope ratios is demonstrated within coastal fjord samples (Stukas, et al. 1999).  Indeed, a Ga/Pb interface is consistent with the salinity gradient observed between P12 and P16.  Additionally, Ga/Pb at P20 and P16 represents a transition along Line P towards the lowest and least variable Ga/Pb ratios at P26 (Figure 4.12).  Given the   106 confinement of the terrestrial Pb isotope signature to the summer mixed layer, fluvial sources may be advected to offshore regions within the shallow mixed layer.  Additionally, fluvial freshwater inputs can also be transported offshore by eddies.  Fluvial inputs of radiogenic Pb are therefore interpreted to influence observed 206Pb/207Pb at stations P4 – P20.  Eolian deposition provides sources of Pb from Asia to the north Pacific, evidenced by isotopic ratios within ocean samples (Wu, et al. 2010; Gallon, et al. 2011) as well as Alaskan ice cores (Gross, et al. 2012).  Sources of Pb from Asia to P26 are therefore considered to result from eolian inputs.  This work therefore concludes eolian sources at P26 bounded by the Alaskan Gyre and the influence of fluvial sources of Pb to the remainder of the Line P transect.     Figure 4.12. Fluvial sources to Line P traced by Ga/Pb within the mixed layer demonstrate a gradient between P12 and P16, and transition zone along P16 and P20 towards lower values at P26.  Note log scale on y axis to assist visualization of full data range.    4.5.4 Implications  Trace metals such as Fe, Cd, Zn, and Mn are important for phytoplankton growth and are similarly sourced by eolian and fluvial mechanisms.  Interpretations in this thesis suggest that fluvial inputs in the summer may provide increased availability of bioactive metals to phytoplankton.  Furthermore, in August 2010, this source appears to be restricted to stations P4 –   107 P20, outside the Alaska Gyre.  Isotopic ratios of Pb are therefore used as a tracer of potential trace metal sources to the HNLC waters across Line P.  Here, HNLC conditions are quantified by NO3 / Chl a (Figure 4.13), where NO3 is used to denote nitrate plus nitrite.  High values of NO3 / Chl a indicate underutilized nutrients and therefore HNLC conditions, and low values indicating non-HNLC conditions.  Ratios of NO3 / Chl a across Line P in August 2010 indicate non-HNLC conditions along the eastern portion of the transect including stations P4 and P12 (Figure 4.13).  Following the salinity interface (Figure 4.10), HNLC conditions begins to be observed and are present at P16 and P20 (Figure 4.13).  Finally, and associated with the transition to the Alaska Gyre (Figure 4.9), NO3 / Chl a ratios increase along a steep gradient approaching station P26 (Figure 4.13).  HNLC conditions and Fe-limitation are therefore indicated at P26, with a transitional region at P16 and P20.     Figure 4.13. Concentrations of NO3 (triangles),Chl a (circles), and NO3 / Chl a ratio (squares) combine with colours based on T-S characteristics (Figure 4.9 and Figure 4.10) to identify regions of NO3 utilization (blue), NO3 limitation (green), followed by increasing concentrations of underutilized NO3 resulting in HNLC conditions approaching P26 (black) with transition toward HNLC conditions at P16 and P20 (red).    A strong correlation between the NO3 / Chl a ratio (a proxy for Fe-limitation) and the 206Pb / 207Pb ratio is observed along Line P (Figure 4.13).  At P4 high Chl a indicating utilization of available NO3 is coincident with higher 206Pb / 207Pb (Figure 4.14) associated with summer fluvial freshwater sources from North America.  Increasing NO3 / Chl a ratios at P16   108 corresponds to decreasing influence of fluvial sources (Figure 4.12, Figure 4.10) and a lower North American Pb isotope signal (Figure 4.14).  While HNLC character increases at stations P17 and 18, Pb isotope ratios and HNLC conditions at P20 are similar to observations at P16 (Figure 4.14).  Increasing HNLC character at P26 (Figure 4.13) is coincident with Pb isotopes (Figure 4.14) indicative of eolian Asian sources to the Alaska Gyre.     Figure 4.14. Values of the NO3/Chl a ratio, representing a measure of HNLC character, demonstrate similar patterns observed for 206Pb/207Pb, non-HNLC conditions are coincident with North American fluvial sources, increasing HNLC character at P16 and P20 corresponds to a shift in 206Pb/207Pb approaching the edge of the Alaska Gyre, and rapidly increasing HNLC character entering the Alaska Gyre is matched by the gradient in 206Pb/207Pb.    The strong correlation between NO3/Chl a, and 206Pb/207Pb values (Figure 4.15) (R2 = 0.93, p < 0.001) suggests that fluvial North American summer sources of dissolved Pb and trace metals may influence HNLC conditions along Line P.  While correlation certainly does not imply causation, these results support future research of biologically important trace metals associated with North American fluvial inputs.  For instance, transport of Fe within eddies formed in coastal waters support phytoplankton production in the northeast Pacific (Fiechter and Moore 2012).     109 While sources from Asia can also provide Fe for phytoplankton (Meskhidze, et al. 2005), Asian Pb isotope signatures at P26 in August 2010 are associated with Fe-limited HNLC conditions.  This does not imply that Fe was not delivered with the Asian input.  Rather, the Fe may have already have been utilized by phytoplankton.  Volcanic ash can also provide Fe to the northeast Pacific (Hamme, et al. 2010) and can have radiogenic Pb isotope ratios consistent with North American sources (Johnson, et al. 1996).  While volcanoes produce spectacular episodic sources of Fe, fluvial sources provide a more consistent input to the northeast Pacific Ocean.    Figure 4.15. HNLC character, indicated by NO3 / Chl a, correlated with mixed layer 206Pb/207Pb resulting in R2 = 0.93, p < 0.001.    4.6 Conclusion  This thesis presents the first measurement of Pb isotopes across the Line P transect.  Isotopic ratios of Pb vary spatially and temporally along the transect.  Isotopic ratios of Pb are interpreted to represent Asian sources at P26, the western terminus of the transect, while terrestrial North American sources are observed along the remainder of the transect.  Asian sources at P26 are interpreted to be of eolian deposition, while terrestrial North American sources are interpreted to   110 be of fluvial origin.  Isotopic ratios of fluvial origin are confined to the summer mixed layer, while isotopic ratios indicative of Asian sources extend to the depth of the winter mixed layer.  Incomplete nutrient utilization, indicative of HNLC conditions, is coincident with Asian sourced Pb within the Alaskan Gyre at stations approaching P26.  Transition from non-HNLC to HNLC conditions along stations P16 and P20 is aligned with transition in Pb isotope ratios from terrestrial fluvial summer sources to less radiogenic Asian eolian sources associated with the winter mixed layer.  This thesis therefore identifies 206Pb/207Pb and 208Pb/206Pb as tracers of eolian Asian and fluvial North American sources of trace metals correlating with HNLC conditions along Line P within the northeast Pacific Ocean.       111 Chapter 5: Dissolved Ga in the Beaufort Sea of the western Arctic Ocean  5.1 Synopsis  Dissolved gallium (Ga) concentrations are presented from the Beaufort Sea, representing the first reported values of Ga in the Arctic Ocean.  Profiles of dissolved Ga in this region reflect Pacific and Atlantic Oceans source waters.  Dissolved Ga concentrations of 4 – 6 pmol kg-1 at depths < ~150m are characteristic of dissolved Ga in the North Pacific and at depths > ~350m, concentrations of 25 – 28 pmol kg-1 are indicative of Atlantic waters.  A smooth gradient of increasing dissolved Ga concentrations is observed through the thermocline, consistent with mixing of cool Pacific waters of low Ga concentration and relatively warm Atlantic waters of higher Ga concentration.  Dissolved Ga is shown here to represent a complimentary tracer of Pacific inputs to the Western Arctic Ocean, as seasonal processes influencing temperature and NO3 : PO4 of Pacific Water entering the Arctic do not influence dissolved Ga profiles.  Observed PO4 deficiencies within NO3 deplete waters in the upper ~70m is interpreted to represent nitrogen fixation in Pacific waters stored within the anticyclonic Beaufort Gyre.  Nitrogen fixation rates calculated in this work are interpreted to represent the Beaufort Gyre, as opposed to freshwater sources associated with the Mackenzie River or sea ice melt, based on conservative behaviour of dissolved Ga within Pacific waters.  Storage of PO4 deficient Pacific waters in the Beaufort Gyre followed by release and transport to the north Atlantic would result in pulses of PO4 deplete waters from the Arctic, impacting nitrogen fixation in the Atlantic Ocean.  Sampling at one shelf station revealed local impacts on dissolved Ga concentrations. At the shelf station, high concentrations at shallow depths are interpreted to represent fresh waters inputs from the Mackenzie River or sea ice melt.  Additionally, contrary to the remainder of the stations, an increase in Ga concentrations associated with the thermocline is not observed, suggested to result from increased scavenging rates within shelf environments due to increased particle abundance.  Finally, first reports here of dissolved Ga in the Arctic allow comparison with profiles of dissolved Al. Distinct profiles of these two similar elements in the Arctic Ocean are interpreted   112 to result from differential scavenging rates and vertical transport of dissolved Ga and Al within Atlantic sourced waters and deep basin waters.   5.2 Introduction  Trace metal research in the Arctic Ocean has benefitted from the GEOTRACES program and the International Polar Year (Henderson, et al. 2007; Melling, et al. 2012), including observation of hydrothermal inputs of Mn (Middag, et al. 2011) and Fe (Klunder, et al. 2012), confirmation of dissolved Al concentration profiles unique to the Arctic (Moore. 1989; (Middag, et al. 2009; Giesbrecht, et al. 2013), and classic Fe limitation of phytoplankton (de Baar, et al. 2005) is shown to be paired with light limitation in the Arctic (Taylor, et al. 2013).  However, dissolved Ga has not been reported in the Arctic Ocean.   Gallium is a favourable element for oceanographic analysis and interpretation given three key characteristics.  First, as a scavenged type trace metal, dissolved Ga concentrations vary in the world oceans as a function of both depth as well as geographic location (Orians and Bruland 1988a; Orians and Bruland 1988b; Shiller 1998; Shiller and Bairamadgi 2006).  Second, dissolved Ga provides utility as a tracer given a residence time longer than dissolved Al, yet short enough to display regional characteristics (Orians and Bruland 1988a).  Third, although dissolved Ga concentrations in the ocean are very low, 2-40pmol kg-1, less contamination risk exists during sampling and analysis relative to more contamination prone elements, such as Al, Fe, and Zn, due to the low natural abundance of Ga in the earth’s crust and relatively less use in industrial and manufacturing processes.    Gallium concentrations measured in the ocean have been utilized to identify eolian inputs (Orians and Bruland 1988b; Shiller 1998).  Comparison of dissolved Ga concentrations with dissolved Al provided insight into differential scavenging as Ga is less particle reactive than Al (Orians and Bruland 1988a), and differential scavenging of Ga and Al results in a positive correlation between the dissolved Ga/Al ratio and Chl a (Shiller and Bairamadgi 2006).    113 Additionally, dissolved Ga is shown to be a tracer of temporally distinct plumes of the Columbia River, based on development of an endmember mixing model (McAlister and Orians 2012).  This work here presents dissolved Ga as a tracer Pacific and Atlantic sourced waters to the Arctic Ocean.  Dissolved Ga concentrations therefore compliment ratios of dissolved NO3 : PO4 (Jones, et al. 2003) as tracers of Pacific freshwater in the Arctic.    Multiple tracers are required to understand unique Arctic Ocean water mass properties and stratification, particularly as climate change influences Arctic Ocean circulation dynamics, timing of ice melt and formation, and biogeochemical cycling (McLaughlin, et al. 2011).  Climate impacts in the Arctic exhibit local and global geophysical, biological, and social impacts (Barber, et al. 2008; Budikova 2009; Carmack and Wassmann 2006).  Recent years have seen shrinking Arctic sea ice (Zhang, et al. 2008), catalyzed by inputs from the Pacific and coupled ocean-atmosphere-sea ice dynamics (Shimada, et al. 2006).  Sea ice impacts climate via changes to albedo, modifies the stratified structure of the upper Arctic leading to changes to phytoplankton community production and composition (Li, et al. 2009), and influences seasonal inputs of Pacific waters to the Arctic (McLaughlin, et al. 2004)     Pacific waters transported across the Arctic return freshwater to the Atlantic, completing global ocean cycling (Wijffels, et al. 1992).  Pacific freshwater inventory exceeds freshwater from rivers across the Canada Basin in the Western Arctic (Jones, et al. 2008) and storage of freshwater in the Beaufort Gyre results in episodic release of freshwater to the North Atlantic (Proshutinsky, et al. 2002).  Changes to Pacific water flow into the Arctic (Woodgate and Aagaard 2005) and the Arctic Oscillation (Steele, et al. 2004) further influence the extent of Pacific water transport across the Arctic.  Therefore tracing Pacific inputs of fresh water from the Arctic to the North Atlantic (Sutherland, et al. 2009; Aagaard and Carmack 1989) is critical given freshwater impacts on North Atlantic Deep Water formation (Keigwin, et al. 1991).    Temperature and salinity provide identification of Arctic water masses, and have indicated recent and rapid changes in Arctic structure and circulation (McLaughlin, et al. 2004; Carmack, et al. 1997; Shimada, et al. 2006; Jackson, et al. 2010).  Chemical tracers, including nitrate, phosphate,   114 and O2, provide additional markers of physical and biogeochemical sources and sinks in the Arctic (Falkner, et al. 2005; Jones, et al. 1998).  While utilization of temperature and N:P relationships is important for identifying Pacific waters in the Arctic, trace metals provide an additional opportunity for identification of complimentary tracers of physical and biogeochemical Arctic Ocean dynamics.  This work presents the first profiles of dissolved Ga concentrations in the Arctic Ocean, demonstrating utility as a complimentary conservative tracer of Pacific waters, highlighting differential source and sink processes in the Beaufort Gyre and along the shelf, application to interpretation of nitrogen fixation and available phosphate in the Beaufort Gyre, and suggesting differential scavenging rates and vertical transport producing distinct profiles of dissolved Ga and Al in the Arctic.    5.3 Methods  Dissolved Ga is reported from six profiles in the Beaufort Sea (Figure 5.1) sampled aboard the CCGS Amundsen in September 2009, a GEOTRACES cruise supporting the International Polar Year (Melling, et al. 2012).  Seawater was collected with GoFlo bottles attached to a trace metal clean rosette supported by a conducting Kevlar line (Giesbrecht, et al. 2013), filtered into acid cleaned LDPE bottles utilizing trace metal clean practices, and acidified onboard the ship in a laminar flow hood.    Figure 5.1  Arctic Study Area: Beaufort Sea a) Study area in the Beaufort Sea of the Arctic Ocean, circles indicate sampling stations b) trace metal sampling performed at stations L1 – L3 and S4, stations S1 – S2 provide transmissometry data and water mass identification utilized in interpreting results.     115  Sample depths within the Pacific waters and Atlantic thermocline were selected based on real time CTD data.  Utilization of temperature, salinity, and oxygen data to select sample depths allowed specific targeting of oceanographically relevant features of Atlantic sources and seasonably variable inputs from the Pacific including unique features of the Western Arctic (Figure 5.2).  For instance, local temperature and oxygen minima were targeted as indicators of Pacific Winter Water (PWW).     Figure 5.2  Profiles of temperature, salinity, and O2 at station L1.  Open circles represent trace metal sampling depths, T and O2 minima indicate PWW and the thermocline delineates cool Pacific waters (<~150m) from relatively warm Atlantic waters (>~350m).  Note log scale for depth to enhance visualization of Pacific water structure.    Gallium was analyzed by Inductively Coupled Plasma Mass Spectrometry (ICP-MS) following 100X preconcentration from 50mL seawater samples utilizing Mg coprecipitation and quantified by isotope dilution (Shiller and Bairamadgi 2006).  Following addition of an isotopically   116 enriched solution of 71Ga, samples were precipitated with trace metal clean ammonium hydroxide (Seastar Chemicals), centrifuged, decanted, and the precipitate dissolved in 1% HNO3.  Sample preparation was performed within a Class 100 laminar flow fume hood in a Class 1000 clean room.  Operation of the ICP-MS was performed in medium resolution to resolve 69Ga and doubly charged 138Ba.    Results here representing the first reports of dissolved Ga in the Arctic are oceanographically consistent, displaying smooth profiles and concentrations consistent with north Pacific (Orians and Bruland 1988b) and north Atlantic (Shiller 1998) source waters.  Duplicate analysis of samples (n=17 duplicates) resulted in a mean RSD of 3.2%, a range of 0.1% – 12.3% RSD and a median RSD of 2.1%.   5.4 Results and discussion  Results and Discussion will be presented and organized into five sections.  First, profiles of dissolved Ga from five stations (L1, L1.1, L1.5, L2, and L3: Figure 5.1) in the Canadian Basin are presented, showing concentrations consistent with Pacific and Atlantic sources.  Next, distinct Ga concentrations at station S4 on the shelf are discussed relative to the Basin stations.  Third, application of dissolved Ga as a conservative tracer of Pacific waters, absent of seasonal influence of Pacific Winter Water will be demonstrated.  Fourth, conservative behaviour of dissolved Ga within nitrate deficient waters is used to interpret calculated nitrogen fixation as occurring within Pacific source waters of the Beaufort Gyre.  Finally, profiles of dissolved Ga and Al in the Arctic are shown to be similar within Pacific waters, while differences within the Atlantic portion of the profile are considered to result from the effects of differential scavenging and vertical transport of these two related elements.      117 Table 5.1. Dissolved Ga (pmol kg-1) measured in the Beaufort Sea of the western Arctic Ocean L1 L1.1 L1.5 L2 L3 S4 Depth Ga Depth Ga Depth Ga Depth Ga Depth Ga Depth Ga 8.5 3.9 7.5 6.3 10 7.5 8 6.0 10 7.5 7 22.7 15 5.7 20 6.5 19 15.6 20 5.4 32 4.6 8 18.8 30 4.7 40 5.2 40 4.4 40 4.9 55 4.5 17 12.5 50 4.5 60 4.6 90 5.7 80 5.0 140 6.2 25 6.7 70 4.4 75 4.8 140 8.5 120 6.0 180 8.7 50 6.3 90 5.1 90 5.1 190 6.7 180 10.2 260 18.4 70 6.0 120 4.8 110 5.2 280 21.3 270 22.5 350 24.9 90 4.4 150 6.2 130 4.8 380 25.2 360 24.1 440 26.9 120 4.2 175 7.9 170 6.7 450 26.2 400 28.7 600 28.5 150 4.7 200 15.6 210 12.5 600 27.3 440 27.3 800 27.8 200 3.9 250 17.5 270 27.0 800 27.5 550 26.5 1000 28.1 240 5.1 350 22.4 360 25.3 1000 29.0 700 28.1 1200 27.9 275 4.5 400 25.4 400 26.0   900 26.5     500 24.8 425 25.8   1300 28.4     600 24.0 500 27.5   1500 27.8     750 25.2 600 27.6   1700 27.5     1000 25.7 800 28.6   1900 27.1     1250 28.5 1000 28.9   2100 29.0     1500 24.3 1250 30.1   2300 27.7     1800 25.5 1500 30.2   2500 26.5       1750 28.4   2700 26.7       2250 27.5   2950 25.8         2400 27.0                     118  5.4.1 Dissolved Ga in the Beaufort Sea: Canadian Basin   Dissolved Ga in the Beaufort Sea at stations L1 – L3 shows smooth, oceanographically consistent profiles (Figure 5.3).  Concentrations of dissolved Ga are uniform with depth through Pacific waters (0 – ~150m) within each individual profile and across the transect (Figure 5.4).  Similarly, waters sourced from the Atlantic (> ~350m) exhibit a narrow range of dissolved Ga concentrations (Figure 5.4).  Transition from Pacific to Atlantic waters results in a smooth increase in dissolved Ga concentrations associated with the thermocline.     Figure 5.3  Profiles of dissolved Ga at stations L1 – L3, log depth scale show lower concentrations through Pacific waters, transitioning to higher concentrations associated with Atlantic waters.      119  Figure 5.4  Dissolved Ga concentrations at L1 – L3 demonstrate consistent profiles across the five basin stations.  Dashed lines indicate ranges of dissolved Ga in the Pacific (Orians and Bruland 1988b) and dash-dot-dot lines represent concentrations of dissolved Ga reported in the north Atlantic (Shiller 1998).   5.4.2 Dissolved Ga in the Beaufort Sea: Shelf  Results of dissolved Ga at the shelf break, at station S4, are distinct from stations L1 – L3.  While concentrations of dissolved Ga at S4 are similar to those of L1 at depths of 25 – 120m (Figure 5.5a), deviations exist at shallow depths and at depths extending through the thermocline.  Higher concentrations of Ga are observed at shallow depths, associated with comparatively warmer waters at S4 (Figure 5.5a).  Similar thermocline structure at depths > 150m at S4 and L1 suggest the presence of Atlantic waters at both stations (Figure 5.5a).  However dissolved Ga concentrations at S4 do not increase through the thermocline as observed in the basin stations L1 – L3.        120  Figure 5.5  a. Temperature and dissolved Ga profiles at L1 and S4, L1 represents the similar profiles of the L stations.  Dissolved Ga and temperature are higher at S4 relative to L1 in the upper 25m.  Temperatures are similar through the thermocline at S4 and L1, yet increasing Ga concentrations are not observed at S4. b. Transmittance at shelf stations S1, S1.1, S1.2, and S2 (Fig. 1) is lower and more variable than the nearly vertical profile at L1.  Elevated concentrations of dissolved Ga at shallow depths at S4 may be associated with freshwaters sources from rivers (McAlister and Orians 2012; Shiller 1996) or sea ice melt, as suggested for aluminum (Measures 1999).  Delineation of water masses along the S1, S2, L1 transect (Figure 5.1), calculates a Mackenzie River water contribution of 10 – 15% at 0 – 15m at S1 and S2, and ~5% at L1 (Alfonso Mucci, pers. comm.).  Sea ice melt is also suggested to contribute 10 – 20% of waters at 0 – 15m at S1 and S2.  However, a similar amount of sea ice melt is determined at L1 (Alfonso Mucci, pers. comm.).  Sea ice formation along the shelf may include greater amounts of entrained sediment.  Therefore, while sea ice melt is similar along the shelf and in the basin, shelf derived ice may provide a greater source of Ga.  Elevated dissolved Ga at S4 may therefore result from freshwater input, in the form of Mackenzie water and/or sea ice melt.    Another contrast at S4 relative to the L stations is the lack of increasing concentrations of dissolved Ga through the thermocline at depths > 150m (Figure 5.5a).  Greater particle abundances associated with the shelf, resulting in increased scavenging of dissolved Ga, could   121 remove the Ga transported in Atlantic sourced waters at this depth.  Lower concentrations of dissolved Ga are also observed approaching the coast off central California (Orians and Bruland 1988b), and the residence time of dissolved Ga near the coast is an order of magnitude lower than the open ocean (Orians and Bruland 1988b).  Although transmittance data are not available at station S4, four shelf stations S1 – S2 (see Figure 5.1) suggest a greater presence of particles, including suspended sediments, which are expected to increase scavenging of dissolved Ga compared to the offshore station L1 (Figure 5b).    Given the extensive shelf area of the Arctic Ocean, observations implying freshwater sources of dissolved Ga to the shallow surface and a sink due to scavenging within the Atlantic thermocline along the shelf will require additional research.     5.4.3 Dissolved Ga: Conservative Pacific tracer  Dissolved Ga is demonstrated above to be a proxy of Pacific waters.  Temperature and N:P are also particularly well suited to trace Pacific Winter Waters (PWW).  Temperature minima identify PWW due to sea ice formation in the winter.  Additionally, PWW is influenced by denitrification in the sediments of the Bering Sea and Chukchi Shelf during transport from the Pacific into the Arctic Ocean.  Denitrification identifies PWW based on lower N:P relative to Atlantic waters (Jones et al. 1998) and is demonstrated by the N* minimum (Figure 5.6), where N* = NO3 – 16xPO4 + 2.9 (Gruber and Sarmiento 1997).  While temperature and N* reflect seasonal dynamics influencing Pacific water sources, concentrations of dissolved Ga remain essentially consistent through the Pacific waters (Figure 5.6), providing a complimentary conservative tracer of Pacific waters in the Arctic.         122  Figure 5.6  Dissolved Ga, temperature, and N* at L1.  Dissolved Ga is conservative through Pacific waters while temperature and N* show minima associated with Pacific Winter Waters.  Climate change, including an earlier loss of sea ice (Tremblay, et al. 2006) and sea level rise (Addison, et al. 2012), influence primary production and denitrification in Pacific waters entering the Arctic.  Therefore, calculation of the amount of Pacific water based on N:P could be obscured due to the influence of climate change on denitrification (Jones, et al. 1998).  Further research is therefore warranted investigating dissolved Ga as a complimentary tracer of Pacific waters, including sampling dissolved Ga in the Bering and Chukchi Seas, through the Canadian archipelago, and in the Eastern Arctic.    5.4.4 Dissolved Ga: Application to nitrogen fixation in the Beaufort Gyre  In addition to differentiation of Pacific and Atlantic sourced waters utilizing NO3 vs. PO4 plots, a third region of the water column is identified in the upper 70m, wherein concentrations of NO3   123 approach zero while PO4 remains abundant (Figure 5.7a,b).  Plots of NO3 vs. PO4 (Figure 5.7c) reveal that PO4 concentrations within these NO3 deficient waters range from 0.5M – 0.75M, in contrast to intercept values of 0.8M predicted by linear regression of Pacific Waters.  Similar observations of net community production exceeding NO3 based new production in the Beaufort Sea (Tremblay et al. 2008) could result from freshwater inputs of nutrients, allochotonous inputs of organic nitrogen (Simpson et al. 2008), or identification of nitrogen fixation within cold waters of the Beaufort Sea (Blais, et al. 2012).  Can concentrations of dissolved Ga help delineate processes leading to drawdown of PO4 in NO3 deficient waters?    Profiles of dissolved Ga within these nitrate deficient waters (< 70m) do not display systematic variation across these stations (Figure 5.4).  Conservative behaviour of dissolved Ga suggests that the drawdown of PO4 in nitrate deplete waters is specifically associated with Pacific source waters within the Beaufort Gyre as opposed to freshwater sources from outside the Gyre, which are likely to contain elevated concentrations of dissolved Ga.  Elevated concentrations of dissolved Ga at S4 do not display the same conservative behaviour as dissolved Ga within the Beaufort Gyre (Figure 5.5a).  Freshwater sources from river inputs or melting of land fast sea ice could provide different N:P inputs at S4.  Therefore, relationships between NO3 and PO4 are considered here at the L stations, displaying conservative behaviour of dissolved Ga within Pacific waters.   Decoupling of dissolved NO3 and PO4 in NO3 deficient waters (< 70m) represents ~35% of the water column identified as Pacific (~200m) based on N:P plots (Figure 5.7).  What is controlling N:P dynamics in these nitrate deficient waters?  Calculation of the N* parameter provides differentiation of nitrogen cycling processes and identifies areas of the oceans impacted by nitrogen fixation or denitrification, indicated by positive or negative N* values, respectively.  Globally averaged biogeochemical cycling of NO3 and PO4 results in N* = NO3 – 16xPO4 + 2.9, representing a Redfield ratio of N:P = 16 and an intercept of 2.9 accounting for global average denitrification.  Given the unique environment of the Beaufort Sea and nitrogen dynamics associated with Pacific Winter Waters, however, an N** function is defined here, specific to the Pacific waters sampled in this work which do not exhibit nitrate depletion (NO3 > 0.2 mol kg-1):   124 N** = NO3 – 13.6xPO4 + 11 (Figure 5.7c).  Profiles of this N** parameter reveal positive values within nitrate depleted waters, indicative of nitrogen fixation (Figure 5.7d).  Therefore, while denitrification typifies Pacific Winter Waters, nitrogen fixation allows utilization of available PO4 within NO3 deficient waters.     Figure 5.7  Nitrate (a.) and phosphate (b.) profiles, green represents nitrate deficient waters, blue PWW, and red Atlantic waters; c) NO3 vs. PO4, PWW are described by a linear fit to data indicated by blue symbols (R2 = 0.98), dashed line represents Atlantic sourced waters (Jones, et al. 1998), d) positive N** values within nitrate deficient waters suggest N fixation.    Nitrogen fixation rates within NO3 deplete waters are calculated based on the PO4 deficit, and are influenced by the residence time of the gyre (gyre) and the N:P ratio.  While PO4 uptake by nitrogen fixers occurs at N:P = 45 (Karl et al. 2002), decomposition results in a nitrogen input to the system that is utilized by non nitrogen fixing organisms at the Redfield ratio of N:P = 16.  Therefore utilization of every 1 PO4 by N fixers fuels the use of ~3 (45/16) PO4 by non N fixers, and N fixers have utilized 16/45 of the PO4 deficiency.  Some N fixers sink before decomposition, thus removing available N, taken to be represented by N:P ratio of 23.3 in sinking matter (see Yamamoto-Kawai, et al. 2006).  Calculations of nitrogen fixation in this work utilize N:P values of 16 – 23.3, indicated by range bars in Figure 5.8a.  Residence time of the Beaufort Gyre ranges from ~ 5-10 years since 1950, alternating between anticyclonic modes   125 of freshwater storage within the gyre followed by cyclonic release of freshwater through the Canadian Archipelago and Fram Strait (Proshutinsky, et al. 2002).  Freshwater in the Beaufort Gyre was ~50% greater than the climatologically mean in 2008 (McPhee, et al. 2009), as the last transition to an anticyclonic mode occurred in 1997 (Proshutinsky, et al. 2002).  Residence time of the gyre for calculation of nitrogen fixation rates in this work from samples collected in 2009 is therefore set at 12 years.     Figure 5.8  a) Calculated rates of nitrogen fixation (mmol N m-3 yr-1) at basin stations, b) integrated rates of nitrogen fixation (mmol N m-2 yr-1) from 7.5m – 40m, representing sampling depths to ~1% light level.    Nitrogen fixation rates are similar among the L stations within the Beaufort Gyre (Figure 5.8a). Calculated nitrogen fixation rates are integrated to 40m, approximating the 1% light level (Figure 5.8b).  Integrated rates of N fixation of 10 – 15 mmol N m-2 yr-1 (or 25 – 35 mol N m-2 day-1) compare to ranges of 24 – 66 mol N m-2 day-1 at station ALOHA (Dore, et al. 2002, Montoya, et al. 2004).  Nitrogen fixation rates in the upper 40m of the Beaufort Gyre (Figure 5.8a) of ~0.25 – 0.5 mmol m-3 yr-1 (or 0.7 – 1.4 nmol N kg-1 day-1) compare to incubation experiments in   126 the Arctic of 0.4 – 2.0 nmol N kg-1 day-1 associated with the Mackenzie River estuary (Blais, et al. 2012).  However, the rates calculated here are an order of magnitude greater than incubation rates of 0.04 – 0.16 nmol N kg-1 day-1 measured at marine stations proximal to the Beaufort Sea shelf (Blais, et al. 2012).    Higher rates of nitrogen fixation calculated here within the Beaufort Gyre, although still only representing ~2 – 3% of observed Arctic denitrification rates (Devol, et al. 1997), may be the result of greater nitrogen deficiency as basin stations sampled in this study are less influenced by upwelled nitrogen sources along the shelf.  Additionally, nitrogen fixation rates calculated here are sensitive to selected values of N:P and gyre, as discussed above.  Further, storage within the Beaufort Gyre over a residence time of 12 years will reflect seasonal and annual transitions in nitrogen fixation within Pacific source waters to the Beaufort Gyre.  Finally, nitrogen fixation can require Fe and an increased availability of dissolved Fe is reported in the Beaufort Gyre at station L1 (Taylor et al. 2013).  A PO4 deficiency could also result from the utilization of an allochotonous nitrogen source, such as a dissolved organic nitrogen (DON) source in excess of a Redfield ratio of dissolved organic phosphate (DOP).  Elevated DON/DOP ratios have been observed in the surface of the Beaufort Sea (Simpson, et al. 2008).  Calculation of nitrogen fixation here considers specifically the Beaufort Gyre, based on conservative behaviour of dissolved Ga in Pacific waters, as rivers may deliver inputs of different N:P and/or carry a signal of upstream nitrogen fixation.  This study therefore represents nitrogen fixation rates in nitrogen deplete waters within the Beaufort Gyre over a 12 year period in the presence of an available source of Fe.     Identification in this work of a phosphate deficiency resulting from nitrogen fixation has important implications for nitrogen cycling in both the Arctic and Atlantic Oceans.  Balance of the global nitrogen cycle is dependent on nitrogen fixation replacing nitrogen lost during denitrification.  Pacific waters returned through the Arctic to the north Atlantic provide PO4, with 90% of PO4 at 47°N in the north Atlantic being supplied by Arctic export (Torres-Valdes, et al. 2013), and this return of PO4 to the Atlantic fuels nitrogen fixation (Yamamoto-Kawai, et al. 2006).  Phosphate deficiencies of 0.15 – 0.3 mol kg-1 used to calculate nitrogen fixation in this   127 work represent ~15 – 35% of expected excess PO4 available for transport to the north Atlantic.  Release of freshwater stored in the Beaufort Gyre influencing formation of North Atlantic Deep Water (Peterson, et al. 2006) would therefore be coincident with an influx of PO4 deplete water to the Atlantic.  Climate change impacts in the Arctic, including early sea ice retreat and changes to stratification, will influence nitrogen fixation and therefore potentially impact ecosystems in the Arctic and Atlantic Oceans and the global nitrogen cycle.    5.4.5 Dissolved Ga: Comparison with dissolved Al  Finally, first reports here of dissolved Ga in the Arctic will be compared to profiles of dissolved Al in the Arctic.  Both Ga and Al are scavenged type trace metals exhibiting hydroxide speciation (Turner, et al. 1981).  However profiles of dissolved Ga in the Arctic exhibit different characteristics than dissolved aluminum (Al), particularly when plotted as a linear function of depth (Figure 5.9a).  Increasing concentrations of dissolved Al in the Arctic with depth may result from reversible exchange (Giesbrecht, et al. 2013) or links with the silica cycle (Middag, et al. 2009).  Increasing trace metal concentrations with depth, similar to Al in the Arctic, are also observed for Th (Bacon and Anderson 1982) and rare earth elements (Nozaki and Alibo 2003).  Models consistent with these observations couple reversible exchange and decreasing particulate abundance with depth (Siddall, et al. 2008).  Profiles of Al in the Arctic may therefore be a function of particulate abundance and reversible exchange, or links to the silica cycle; do these processes assist in explanation of dissolved Ga in the Arctic?      128  Figure 5.9  Profiles of dissolved Ga (station L1) and Al, a) as a linear function of depth, b) plotted as a log depth axis.  Al Central Arctic (Moore. 1989) , Al Beaufort Sea (Giesbrecht, et al. 2013), Al Eastern Arctic (Middag et al. 2009).  Whereas linear depth profiles of Ga and Al appear rather different (Figure 5.9a), plotting depth on a log axis (Figure 5.9b) demonstrates that both dissolved Ga and Al exhibit little change through Pacific waters to 150m.  Differences arise through the thermocline and deeper waters.  Dissolved Ga concentrations increase rapidly through the thermocline to the Atlantic water temperature maxima at ~350m, then exhibit minimal variation through the remainder of the profile.  Similar to Ga, dissolved Al also begins to increase at the thermocline, albeit more gradually (Figure 5.7b).  In contrast to Ga, concentrations of dissolved Al continue to increase throughout the remainder of profile.    Differential scavenging rates may help explain the variation observed in the shape of the dissolved Ga and Al profiles.  Ga exhibits a longer residence time given a decreased propensity to scavenging relative to Al (Orians and Bruland 1988a), and therefore retains the signature of the Pacific and Atlantic inputs to the Arctic.  Alternatively, given a more rapid removal rate, dissolved Al concentrations are attenuated via scavenging during transport from the Atlantic and   129 Al is vertically transported on sinking particles.  Given reversible scavenging and decreasing particulate abundance with depth, removal rates of dissolved Al are slowed, resulting in increasing Al concentrations with depth.  Divergent profiles of dissolved Ga and Al in the Arctic may therefore be a function of differential particle reactivity of these two elements (Orians and Bruland 1988a).  Below 1500m, concentrations of dissolved Al increase more rapidly with depth, while dissolved Ga concentrations show a slight decrease (Figure 5.9b).  Waters below ~1500m represent Canadian Basin Deep Waters (CBDW) dated at 500 years (Macdonald and Carmack 1991, Macdonald and Carmack 1993, Schlosser, et al. 1997) and continued renewal is unlikely (Timmermans, et al. 2003, Timmermans and Garrett 2006).  Deep waters in the Eurasian Basin in the eastern Arctic have surface renewal times of ~ 200 – 700 years (Bonisch and Schlosser 1995).  Reversible exchange of vertically transported Al and the decreased scavenging rates due to low particulate abundance could thus represent a storage of dissolved Al within Arctic deep waters.  In contrast, the slight decrease in dissolved Ga concentrations in CBDW could result from a net scavenged removal during the long residence time of CBDW coupled to a decreased input by vertical transport compared to Al.  Marked differences in Arctic profiles and biogeochemical cycling between Ga and Al, two chemically similar elements, illustrates the importance of investigating multiple trace element proxies (Henderson, et al. 2007).  5.5 Conclusion  Dissolved Ga concentrations are reported for the first time in the Arctic Ocean.  Smooth profiles are observed, transitioning from concentrations of 4 – 6 pmol kg-1, consistent with Pacific inputs, to 25 – 28 pmol kg-1 at depths > ~350m, indicative of Atlantic water.  Profiles of dissolved Ga are consistent within the Beaufort Sea basin, representing a conservative tracer of Pacific water inputs to the Western Arctic, and complimenting N:P ratios as indicators of Pacific waters.  Dissolved Ga concentrations at the shelf station are distinct from the basin stations; higher concentrations in the upper 25m are interpreted to result from river and/or sea ice melt   130 freshwater inputs.  In addition, at the shelf station, increased scavenging of dissolved Ga through the thermocline is suggested to be due to greater particle abundance.  Differential profiles of dissolved Ga and Al in the Arctic suggest increased scavenging and vertical transport of Al to old waters of the deep basin.    Observed PO4 depletion within NO3 deficient waters implies that nitrogen fixation is occurring within Pacific source waters within the Beaufort Gyre, as opposed to freshwater sources.  This interpretation is supported by the conservative behaviour of dissolved Ga within the Beaufort Gyre.  Based on calculated nitrogen fixation rates, up to 30% of the potential PO4 supply to the Atlantic could be removed, impacting nitrogen fixation in the North Atlantic.    This first report of dissolved Ga in the Arctic Ocean encourages future research in the utilization of dissolved Ga to describe Arctic Ocean dynamics.  Measurement of dissolved Ga profiles in the Bering Sea, Eastern Arctic, and Canadian archipelago will test the use of Ga in tracing of Pacific Waters through the Arctic to the North Atlantic. Sampling of shelf environments in the Arctic will investigate freshwater sources and scavenged removal processes of dissolved Ga.  Pairing of dissolved Ga and Al may provide estimates of particle abundance within the deep waters of the Arctic Ocean, given the differential scavenging rates of these two similar elements.  Finally, investigation of nitrogen fixation and resulting impacts of PO4 delivery to the north Atlantic will need to be included in predictions of Arctic response to climate change.      131 Chapter 6: Calculation of river-seawater endmembers and differential trace metal scavenging in the Columbia River plume  6.1 Synopsis  A simple model is presented for calculating river and seawater mixing zone endmembers.  This model allows construction of a conservative mixing line with respect to trace metals scavenged from surface waters, in the absence of direct endmember sampling.  Results of the model suggest a transition from non-conservative to conservative behaviour of the trace metals gallium (Ga) and zirconium (Zr) along a transect extending offshore of the Columbia River, indicating environmental controls on differential scavenging.  Mechanistic explanation of observed differential scavenging is supported by wind data indicating northward transport of Columbia River plume waters, followed by southwest, offshore, transport.  Columbia River plume waters transported to the southwest at the time of sample collection were therefore flanked offshore by aged plume waters returning from the north.  While upwelling conditions provide nutrients during southerly transport of plume waters, northward plume transport is associated with downwelling, low nutrient conditions.  Low phytoplankton abundance during periods of northward plume transport reduces trace metal scavenging, therefore dilution of river waters with ocean surface waters of comparatively low Ga and Zr concentrations results in near-conservative mixing of Ga and Zr in offshore, aged, plume waters.  Results of this work suggest dynamic modification of trace metal residence time within close geographic proximity established by physical and biologic controls, dictated by local winds.  Lower trace metal scavenging rates, resulting in a longer residence time, are suggested during northward transport of Columbia River plume waters, compared to higher scavenging rates, and thus shorter residence times, during southerly transport of plume waters.  Gallium, demonstrating high analytical precision, reduced contamination risk, and decreased complications of upwelled sources relative to Zr, is proposed to provide identification of aged Columbia River plume waters.     132 6.2 Introduction  Trace metal sources to the ocean from rivers are altered within the mixing zone between the freshwater and seawater endmembers as dilution and removal processes are initiated within dynamic estuarine environments (Boyle, et al. 1977; Sholkovitz 1978; Godfrey, et al. 2008).  Modification of trace metal concentrations continues beyond the estuary within plume waters of the mixing zone as a function of chemical, physical, and biological processes during mixing with the seawater endmember.    Physical processes control mixing within estuaries and as river plumes are transported offshore.  Local winds result in the dynamic transport of plume waters (Hickey, et al. 2009) and influence upwelling, establishing nutrient availability for phytoplankton growth.  Phytoplankton contribute to trace metal removal through both cellular uptake and sorption onto cell surfaces.  Scavenging of trace metals within the surface and photic zone is therefore enhanced by increased particulate surface area provided by phytoplankton.    Upon sinking of particulate matter through the water column, trace metals removed from the surface may exhibit characteristics associated with scavenged or nutrient type profiles.  Elements demonstrating nutrient type profiles are remineralized during heterotrophic respiration, returning to the dissolved phase, and therefore accumulating as a function of increasing water mass age.  Alternatively, elements maintaining association with the particulate phase are removed by settling through the water column, resulting in scavenged type profiles.    Aluminum represents the archetypal scavenged element (Orians and Bruland 1985), demonstrating rapid scavenged removal from the water column, including scavenging of Al by diatoms (Moran and Moore 1988, Gehlen, et al. 2002; Middag, et al. 2009).  Gallium (Ga) exhibits similar behavior, however, scavenging occurs more slowly relative to the rapid loss of Al from the water column (Orians and Bruland 1988a).  Gallium therefore represents an analogue of Al, providing a greater geographic and temporal range from a common elemental   133 point source.  Zirconium (Zr) concentrations increase with depth and exhibit enrichment along thermohaline transport (McKelvey and Orians, 1993; Godfrey, et al. 1996; Firdaus et al. 2011).  Dilution of trace metal concentrations sourced from a river results in conservative mixing with respect to salinity.  Conservative mixing is exemplified by linear elemental concentration as a function of salinity, given a two endmember steady state system of fresh river water mixing with seawater.  Deviation from linearity indicates source or removal processes.  Positive deviations from a linear conservative mixing line indicate a source during mixing.  Conversely, negative deflections from linearity are indicative of trace metal removals during mixing of river and seawater endmembers (Boyle, et al. 1974, Boyle, et al. 1977, Officer 1979, Rattray and Officer 1979, Liss 1976).   Chemical properties of trace metals determine the extent of source, removal, or conservative behavior during transition from river to seawater as a function of differential chemical reactivities.  Definition of the mixing zone will therefore be unique to individual trace metals, and to the environment sampled.  Behaviour of trace metals within the mixing zone may be defined given sampling along a transect extending from a riverine endmember, S = 0, across the mixing zone, to a salinity associated with a mixing zone endmember, allowing for connection of a conservative mixing line (Holiday and Liss 1976).  However, mixing zones covering a broad extent, extending 50 – 100km offshore (Brown and Bruland 2009), may hinder acquisition of samples explicitly defining the two endmembers and rather samples acquired may all be from within the mixing zone, thus complicating identification of the endmembers.    This work therefore presents a method for the definition of river and seawater endmembers based on samples acquired from within the mixing zone.  Further, connection of these endmembers produces a conservative mixing line, allowing for evaluation of trace metal removal behaviour from a riverine source during mixing of river and seawater endmembers.    Concentration data for Ga and Zr from three transects associated with the Columbia River plume will be utilized in this work.  Ocean sampling and analysis is described in Section 6.3, Methods.    134 Trace metal concentrations are reported in the Results Section 6.4.1, followed by hydrography (Section 6.4.2).  Model development is a direct result of interpreting trace metal concentrations, and is therefore presented in the Results, Section 6.4.3.  Constraints governing application of the model are outlined in Section 6.4.4.  Results of the model applied to Ga and Zr concentrations associated with the Columbia River plume are described in Section 6.4.6 and 6.4.7 for Ga and Zr, respectively.  Description of a mechanism supporting observed trace metal concentrations as predicted by the model, based on winds associated with the sampling period is presented the Discussion (Section 6.5).  Finally, results and discussion are summarized in the Conclusion (Section 6.6).    6.3 Methods  Oceanic sampling for trace metals was conducted aboard the R/V Point Sur in June and July 1997.  Samples for trace metal analysis were acquired via a Teflon pump from a ‘fish’ towed at a depth of 1 – 2m, samples were collected with established trace metal clean sampling techniques and acidified with 2mL HCl.   Analysis of Ga and Zr were conducted by ICP-MS following concentration on Chelex-100 resin.  Samples were adjusted to pH = 2 and 1000mL were loaded onto 1.5mL Chelex-100 at a flow rate of 0.2mL/min, a method optimized for Zr by McKelvey and Orians (1998).  Gallium concentrations analyzed by this method utilizing Chelex-100, at pH = 2, and calculated based on an internal standard (Rh), are supported based on agreement with results from the method of Orians and Boyle 1993 utilizing 8-hydroxyquinoline resin at pH = 4 calculated via standard addition.  Precision for Ga and Zr are 5% and 13% and limits of detection are 0.24 and 6 pmol/kg, for Ga and Zr respectively (Lanthier, 1999).    Hydrographic properties, temperature and salinity, acquired continuously along each transect were provided by Ken Bruland’s group at University of California Santa Cruz.  Wind data,   135 presented in the Discussion, was accessed from the National Oceanic and Atmospheric Administration (NOAA) National Data Buoy Center.  6.4 Results 6.4.1 Trace metal concentrations  Measured trace metal concentrations of Ga and Zr (Lanthier, 1999) are presented in Table 1.  Table 6.1 Temperatures and salinity and Ga and Zr concentrations. Transect Station S T Ga Zr oC pmol kg-1 pmol kg-1 T9 1 25.413 16.78 28 48 2 24.669 16.55 36 63 3 23.791 16.45 42 81 4 22.624 16.5 49 86 5 28.564 17.9 16 29 6 28.31 19.36 72 65 7 29.277 18.88 43 54 T8 1 31.605 13.92 10.7 53 2 31.277 15.17 11.8 39 3 30.112 15.77 14.7 42 4 30.142 15.76 12.7 35 5 30.121 16.17 11.6 28 6 28.372 18.3 20 42 7 27.364 18.71 23 34 8 29.38 18.73 16 29 9 31.647 17.85 7.5 16 T7 1 27.888 16.57 19 45 2 26.953 17.66 24 46 3 26.469 17.89 27 47 4 25.903 18.04 26 39 5 25.751 18.07 31 48 6 26.058 18.38 29 57 7 26.482 18.1 26 45 8 25.507 18.35 33 46 9 25.989 18.28 30 52 10 26.293 18.29 27 49 11 26.98 18.14 25 45    136 6.4.2 Hydrography  Brief interpretation of hydrographic data along the cruise transects is instructive in interpreting model results and application of the endmember calculation method presented in Section 3.3.  Transect 9 (T9) extended from the mouth of the Columbia River, while transects 8 (T8) and 7 (T7) are south along the Oregon coast (Figure 6.1a).  Stations at which trace metal samples were acquired are numbered based on increasing distance from the coast.    Figure 6.1  Hydrography. a. Map indicating stations along transects T7, T8, and T9; location of NOAA buoy DESW1 is indicated. b. T-S diagram indicating hydrographic conditions along the sampling transects, trace metal sampling stations indicated by t:s, where t and s indicate transect and station number, respectively. c. Salinity along transects as a function of distance (km) from the coast, trace metal sampling stations along transects are indicated. d. Temperature along transects T7, T8, and T9.    137 Temperature and salinity data collected continuously along transects indicate distinct characteristics among the three transects in T-S space (Figure 6.1b).  Trace metal stations are indicated along each transect and may be referenced to Figure 6.1a.  Transect T9 demonstrates the influence of the Columbia River plume, as stations 1 – 4 occupy the lowest salinity values among the three transects (Figure 6.1b).  Salinities decrease from stations 1 to 4, and the lowest salinities of the core waters of the plume occur ~30km offshore (Figure 6.1c).  Following a rapid increase in salinity to station 5, a local salinity minimum is observed at Station 6 (Figure 6.1c), Station 7 is associated with increasing salinity, leading to a return to high salinity values at the termination of the transect (Figure 6.1b, c).  Stations 6 and 7 exhibit the highest temperatures along the three transects (Figure 6.1b, d), temperatures increase steadily to a maximum at station 6, followed by a slight decrease to Station 7 (Figure 6.1b, d).    Coastal surface waters at T8 Stations 1-2 are > 5oC cooler than any other waters sampled (Figure 6.1d) and are associated with the highest salinity values (Figure 6.1b, c).  High salinity, low temperature waters (Figure 6.1b) are indicative of upwelling along the coast as a result of Ekman transport arising from northerly winds.  Stations 1 and 2 exhibit low temperatures associated with upwelling, temperatures increase steadily to Stations 7 and 8, before decreasing to Station 9 (Fig 1d).  Salinity values indicate a local minimum associated with Station 7 followed by rapidly increasing salinity from Station 8 to Station 9 (Figure 6.1c).  Salinity values at Station 9 are the highest measured offshore and Station 9 may indicate a transitional boundary of the plume-influenced waters.  Temperatures and salinities along T7 demonstrate the narrowest range among the three transects (Figure 6.1b).  Station 1 has the highest salinity and lowest temperatures along the transect, suggesting residual upwelling.  However, although T-S trends are directed towards the upwelling signatures associated with Stations 1 and 2 at T8, temperatures are higher and salinities lower at T7 Station 1 than even Stations 3 – 5 along T8.  Salinity values across the entirety of T7 do not exceed 28 (Figure 6.1c).      138 6.4.3 Model development  Conservative mixing lines may be drawn connecting elemental concentrations of samples collected at minimum and maximum salinity values, provided sufficient sampling coverage across the mixing zone.  However, minimum salinity values in the Columbia River plume dataset are only 22, potentially yielding an artificially low intercept value.  Therefore, a method is developed here allowing for calculation of mixing zone endmembers and construction of a conservative mixing line in the absence of low salinity sampling.    Scavenged removal of trace metals from the river-seawater mixing zone may be identified by parabolic characteristics of trace metal concentrations as a function of salinity (Boyle, et al. 1974; Boyle, et al. 1977).  Initiation of the model presented here therefore fits trace metal concentrations (M) as a function of salinity (S) to second-degree polynomials, given MR > MSW, (R: river, SW: seawater):   cbSaSSfM  2)(        (1)  Trace metal removal as a function of salinity from a river endmember source occurs provided (Figure 6.2):  02  baSdSdM         (2)  As fresh river waters of MR > MSW are mixed with seawater, dilution and scavenging results in an approach to steady state.  Similarly, open ocean steady state is assumed for the scavenged element Al (Measures and Vink 2000), supporting the assumption of steady state for elements with residence times longer than the comparatively short residence time of Al (Orians and Bruland 1988b).  Salinity of the seawater endmember (SSW) of the river-seawater mixing zone is   139 therefore calculated as the salinity at which open ocean steady state is approached, a salinity approaching the limit of the river influence.    0lim dSdM (Fig 2): 0lim  dSdMsw SS         (3) abS sw 2          (4)  Salinity of the riverine endmember equals zero, therefore determining the corresponding trace metal concentration (MS=0) at the ordinate intercept:   0RS          (5) cMs 0          (6)  Definition of the mixing zone is therefore anchored between two endmembers, initiated at river salinities of zero and proceeding across salinity values over which trace metal concentrations decrease with increasing salinity, 0dSdM , and terminating at 0lim dSdM.   Prediction of conservative mixing is achieved by linear extrapolation connecting the two mixing zone endmembers (Figure 5.2).  Conservative mixing is therefore calculated to occur along the line connecting the river (R) endmember: ),0(),( 0 cMS sR           (7) and the seawater (SW) endmember: ))2()2(,2(),(2 cabbabaabMS SWSW      (8)       140  Figure 6.2 Demonstration of the model determining the conservative mixing line based on sampling within the mixing zone.  Regression of elemental concentrations as a function of salinity produces a function for calculation of the river and seawater endmembers, resulting in bounding of the mixing zone and construction of the conservative mixing line, see text for details.  Dependent on the trace metal investigated and the environment sampled, SSW (eq. 3) may represent an inflection point, or initiate a stepwise function.  Piecewise functionality of the method allows separation of the mixing zone (S < SSW), termination of the mixing zone (SSW), and post-termination of the mixing zone (S > SSW).  Trace metal concentrations at S > SSW may remain unchanged, demonstrating steady state, or increase, indicating additional sources to the system such as upwelling.  Examples are presented in Figure 2, wherein triangles indicate the mixing zone, circles demonstrate a steady state condition, and upwelling sources of trace metals are indicated by squares.  Upwelling conditions result in increasing trace metal concentrations at S > SSW.  For instance, Zr concentrations increase with depth and during transport along thermohaline circulation (McKelvey and Orians 1993, Godfrey et al. 1996), alternatively, profiles of Ga (Orians and Bruland 1988b) demonstrate smaller concentration gradients with depth relative to Zr (McKelvey and Orians 1993).  Upwelling along the coast injects an   141 additional endmember to the two endmember system, as upwelling provides a potential source of trace metals, for example Cd and Zn (Shiller 1996).  Upwelling may be considered to represent a potential source for Zr, yet Ga is expected to exhibit an absent or minimal source from upwelling.     Application of this method is necessarily applied within areas of river-seawater mixing zones.  Mixing of fresh and saline waters results in shallow surface layers, 2-20m for the Columbia River plume (Bruland, et al. 2008); therefore, surface samples are required for investigation of mixing zone processes.  Requirement of surface sampling represents potential opportunities, as high resolution surface transects may be available during oceanographic research cruises via underway sampling systems.    6.4.4 Model application  Application of any model must be made appropriately and assumptions considered.  Assumptions included here include a steady state two endmember system with higher elemental concentrations in the river than the ocean endmember.  Observed non-linearity may also be a function of variability in river end member concentrations (Loder and Reichard 1981); flushing time of the Columbia River estuary is rapid (Jay and Smith 1990), dampening effects of river periodicity.  Evaluation of data for applicability with the model presented is performed based on a set of boundary conditions.  Data not applicable to the second degree polynomial fit will fail the following criteria.  Three criteria must be met, associated with the fundamental requirement that all calculated endmembers must be positive:   Criteria 1: 0SWS abS sw 2 (eq 4), therefore 02  ab           (9)   142 Criteria 2: 0SWM 0)2()2(2  cabbaba        (10) Criteria 3: 00 SM 0c            (11)  Constraints may therefore be placed upon coefficients a, b, and c of the second degree polynomial fit, cbSaSSfM  2)( For Criteria 1 above (eq 9): either b < 0 and a > 0, or b > 0 and a < 0 However, considering Criteria 2 above (eq 10):  acb 4           (12) acb 4           (13)  To prevent imaginary numbers, either a > 0 and c > 0, or a < 0 and c < 0 Then, considering Criteria 3 above (eq 11): given c > 0, then a > 0 based on Criteria 2 (eqs 12 and 13),  therefore b < 0 by Criteria 1 (eq 9).  Finally, a minimum bound must be must be placed on coefficient a such that the polynomial fit is sufficiently non-linear to describe a scavenging model, as a linear fit indicates conservative mixing: cbSSMa  )(lim0         (14) Based on Criteria 2) above (eq 10): cba 42          (15)  Consideration of the full suite of above constraints ensures that relevant regressions are obtained.  For example, while values of coefficient a < 0 and b > 0 will produce a curve wherein criteria 1   143 and 2 above are met, coefficient a values < 0 require coefficient b values to be imaginary numbers (eqs 12, 13).  However, criteria 3 rectifies this situation, as given c > 0, coefficient a values < 0 will fail cba 42 (eq 15).    Establishment of coefficient constraints allows applicability of the model to be evaluated for a given data set.  Demonstration of this evaluation technique is presented through use of bootstrap sampling.  Additionally, bootstrap sampling is performed to provide confidence intervals on data regressions.    Sample sets consisting of 1000 bootstrap replicates containing an equal number of data points as utilized for regression were sampled with replacement (Ga: 27 data points, Zr: 24 data points), and evaluated for coefficient failures.  Least squares second degree polynomial fits are calculated for each replicate and resulting coefficients are evaluated based on the three acceptance criteria above (Eq 9 – 11).  Each sample set therefore determines the number of coefficient failures observed for a given set of 1000 bootstrap replicates.  For example, one sample set of 1000 bootstrap replicates resulted in 25 Ga regression coefficient failures.  Each sample set of 1000 bootstrap replicates is unique and thus likely to result in a unique number of coefficient failures, therefore, 100 sample sets were evaluated to provide bootstrap statistics on a sample size of n=100.    Bootstrap results for Ga regressions indicate an average (n=100) of 26 ± 0.5 coefficient failures (Table 6.2); therefore 2.6% of every 1000 bootstrap replicates failed coefficient criteria.  Among the 2.6% of Ga regression failures, 99.3% were due to coefficient a failures.  Coefficient a values that are not sufficiently large result in functions tending toward linearity (equations 14, 15).  Coefficient failures illustrate the utility of establishing coefficient constraints, providing a mechanism to evaluate data set applicability to the model.       144 Regressions of bootstrapped Zr data resulted in an average (n=100) of 47 ± 0.4 coefficient failures per 1000 bootstrap replicates, or 4.7%.  Coefficient a failures comprised 30% of the total failures, 40% of the failures were due to coefficient b, and coefficient c resulted in 30% of the 4.7% of failures (Table 6.2).    Table 6.2 Coefficients failing criteria per 1000 bootstrap replicates Element Average (n=100) failures per 1000 bootstrap replicates Coefficient a Coefficient b Coefficient c Ga 26 ± 0.5 0.2 ± 0.1 0.01 ± 0.1 Zr 14 ± 0.3 19 ± 0.4 14 ± 0.4  6.4.5 Model results  Application of the model will be demonstrated utilizing Ga and Zr concentrations measured along transects influenced by the Columbia River plume.  Regression of Ga and Zr concentrations as a function of salinity is performed with the exclusion of Stations 6 and 7 along transect 9 (T9) as they exhibit elevated concentrations as a function of S, suggesting alternative control mechanisms compared to other samples within the mixing zone.  Results of this model initiate description of a mechanism rectifying observed concentrations at stations 6 and 7, presented in the Discussion (Section 6.5).    6.4.6 Model results: Ga  Concentrations of Ga as a function of salinity are well fit to a second degree polynomial (Figure 6.3a, Table 6.3).  Minimal scatter is indicate around zero on the residual plot (Figure 6.3b) and the elevated concentrations at Stations 6 and 7 along T9 are highlighted; as described above in Section 3.5, these stations are not included in regression analysis.  Calculation of the seawater endmember results in SSW (lim dGa/dS  0) at S = 34.3 (Table 6.3).  Calculated Ga concentrations of 7.5pmol kg-1 at S = 34.3 (Table 6.3) compare well to Ga concentrations of 6.0 pmol kg-1 (S = 33.0), and 8.5 pmol kg-1 (S = 34.2) observed at 39.6°N 140.8°W and 33.0°N   145 139.0°W, respectively, in the Eastern North Pacific (Orians and Bruland 1988b).  As the calculated SSW value is greater than observed salinity values along the sampled transects, Ga concentrations are expected to decrease as salinity increases due to dilution with the seawater endmember and continued removal processes.  Incremental changes in Ga concentrations, however, are minimal as dGa/dS  0.  Connection of the river and seawater mixing zone endmembers (Table 6.3) for Ga results in the calculated conservative mixing line for Ga (Figure 6.3c) during the conditions observed during sampling.  Confidence intervals are presented at 68% and 95%, based on 100 bootstrap samples.  Concentrations of Ga measured at Stations 6 and 7 along T9 plot proximal to the conservative mixing line calculated based on the model and not drawn with regard to Station 6 and 7 (Figure 6.3c, d).  Therefore, near conservative mixing of Ga is suggested at Stations 6 and 7 along T9, in contrast to non-conservative removals at Stations 1 – 5 along T9.  Samples within close geographic proximity along T9 therefore illustrate differential control mechanisms on removal rates and elemental residence times.  Mechanisms for this observed fractionation are presented in the Discussion.                146  Figure 6.3 a. Regression of Ga concentrations as a function of salinity, stations 6 and 7 (T9) are indicated as filled symbols.  b. Residual values calculated from regression in panel a.  c. Conservative mixing line constructed based on endmembers calculated from the model, long and short dashed lines indicate 68% and 95% confidence intervals, respectively, based on bootstrap analysis.  d. full range of mixing zone.  Table 6.3 Regression values and calculated endmember results.  Concentrations of Ga (pmol kg-1) and Zr (pmol kg-1) represented by two significant figures. Element M = f(S) r2 Endmembers (S, M) Seawater River Ga 0.30S2 - 20.6S + 361.1 0.98 34.3, 7.5 0, 360 Zr 1.1S2 - 64.7S + 993.8 0.83 29.4, 42 0, 990    147 Measured concentrations of Ga at a given salinity are composed of Ga sourced from the river endmember, diluted with increasing salinity by lower Ga concentrations from the seawater endmember.  Additionally, scavenged removal processes reduce the concentration of Ga derived from the river to a greater extent than conservative mixing (dilution) alone.  During endmember mixing, and considering the inclusion of scavenged removal, Ga sourced from the river remains > 80% of the total dissolved Ga concentration at S < 25, while at S > 32 Ga sourced from the seawater endmember comprises > 80% of total dissolved Ga (Figure 6.4a).  Equal Ga concentrations contributed by the river and seawater endmembers occurs at a salinity of 30 (Figure 6.4b), indicating that total dissolved Ga concentrations are dominated by river inputs over much of the salinity range.  Comparison of conservative mixing and scavenged removal of Ga sourced from the river indicates that while 50% dilution of Ga sourced from the river occurs at S = 17 based on conservative mixing, inclusion of scavenged removal results in Ga from the river comprising 50% of the measured Ga at S = 10.3 (Figure 6.4c).  Finally, the percent Ga removed via scavenging, in contrast to dilution by conservative mixing, reaches a maximum at S=17, and approaches zero towards the endmembers (Figure 6.4d).  Definition of an endmember requires that the same endpoint is reached, regardless of the mechanism, however the panels in Figure 6.4 illustrate that evolution of trace metal concentrations within the plume vary given scavenged removal acting in concert with conservative mixing.                 148   Figure 6.4 a. Relative percent contributions of river and seawater endmembers to dissolved Ga concentrations as a function of salinity, given scavenged removal. b. concentrations of Ga from river and seawater endmembers and total dissolved Ga, note log scale. c. comparison of river-sourced Ga as a function of salinity during conservative mixing and scavenged removal, dotted lines indicate salinity at which 50% of Ga contributed by the river remains. d. percent Ga removed by scavenged removal relative to dilution by conservative mixing.  Gallium concentrations fit here to a second degree polynomial results in an r2 value of 0.98, however an exponential fit results in a similar r2 value of 0.96.  Consideration of the calculated intercept value lends support to application of the second degree polynomial fit of the presented   149 model.  Applying the polynomial fit, Ga concentrations for the Columbia River source at S = 0 are calculated at the intercept value of 361.1pmol kg-1, or 360pmol kg-1 given the model extrapolation.  Alternatively, exponential fit of Ga concentrations (Ga = 3524 e-0.19S) produces an intercept value at S = 0 of > 3500pmol kg-1.  River concentrations of Ga range from low levels in California streams (<80pmol kg-1) (Shiller and Frilot 1996) to >1500pmol kg-1 in the Mengong River (Viers et al. 1997), less than half that predicted based on the exponential fit.  Additionally, as the derivative of an exponential function approaches zero only asymptotically, selection of a threshold value considered to be zero is required.  Selection of such a value injects a measure of subjectivity avoided by the method presented here utilizing a second degree polynomial.    6.4.7 Model results: Zr  Hydrographic conditions described in Section 3.2 indicate upwelling at stations 1 and 2 along T8.  Zirconium concentrations at Station 1 and 2 along T8 are therefore not included in regressive analysis of the mixing zone, as upwelling provides an additional source for Zr at these stations, outside the river source considered in this model.  Concentrations of Zr at Station 9 along T8 (Zr = 16pmol kg-1) are the lowest observed, and correspond with the highest salinity values.  Additionally, Zr concentrations at this station, 16pmol kg-1, are similar to Pacific Ocean surface values of 15.7 – 25.1pmol kg-1 (McKelvey 1994).  Therefore Zr concentrations at Station 9 (T8) are interpreted to lie outside the mixing zone and are not utilized in regressive analysis, yet this point is included in figures for reference.  Concentrations of Zr demonstrate strong gradients approaching the Western and Eastern margins of the Pacific Ocean, a phenomenon not observed for Ga (Merrin and Orians 2001), therefore supporting inclusion of Station T8:9 for model application to Ga concentrations.    Zirconium concentrations as a function of salinity exhibit greater scatter relative to Ga (Figure 6.5a, b, Table 6.3), however, general trends and results are preserved.  Anchoring of the seawater endmember is achieved at SSW = 29.4, at a Zr concentration of 34.2pmol kg-1.  Unique SSW values are expected for individual elements, given biogeochemical cycling specific to each   150 element.  Three T8 stations at salinities of ~30.1 extend beyond the calculated mixing zone.  Stations 3-5 likely have been influenced by upwelling during formation and T, S conditions (Figure 6.2b) have thus evolved following incorporated into the plume.  Inclusion and exclusion of Stations T8:3-5 results in equivalent r2 values of 0.83, and functions of 1.1S2 – 64.7S + 994 and 0.99S2 – 59.6S + 929 for inclusion and exclusion of Stations T8:3-5, respectively.  Calculated conservative mixing lines are consistent and support the equivalent interpretation presented in the Discussion section.   Figure 6.5 . a. Regression of Zr concentrations as a function of salinity.  Stations 1 and 2 (T8) (black squares) indicate upwelled sources and Station 9 (T8) (grey square) demonstrates concentrations suggestive of the open ocean, outside the defined mixing zone. b. Residual values calculated from regressive fit. c. Conservative mixing line constructed from model, long and short dashed lines indicate 68% and 95% confidence intervals based on bootstrap analysis. Stations 6 and 7 (T9) are consistent with the conservative mixing line.  d. full range of mixing zone.   151  Consistent with observations for Ga, conservative mixing accounts for elevated Zr concentrations at Stations 6 and 7 along T9 (Figure 6.5c).  Calculated Zr concentrations of 42.4pmol kg-1 at SSW are within the range of surface Zr concentrations extending along Line P in the North East Pacific, decreasing from 192pmol kg-1 nearshore (48°N, 126°W) to 64.7pmol kg-1 at 49°N, 134°W, to 24.7pmol kg-1 at 50°N, 145°W (McKelvey 1994).  Similar to Ga, concentrations of dissolved Zr are dominated by river inputs at low salinities (Figure 6.6a, b).  Given conservative mixing, 50% of the Zr sourced from the river is diluted at a salinity of 15 (Figure 6.6c), while with inclusion of scavenged removal, 50% of the Zr contributed by the river remains at a salinity of only 9.  Removal of Zr contributed by scavenging, in addition to dilution by conservative mixing, reaches a maximum at S=15 and decreases to zero at the calculated endmember, as required by definition.      152  Figure 6.6 a. Relative contributions of river and seawater endmembers to dissolved Zr concentrations as a function of salinity based on scavenged removal; b. concentrations of Zr from river and seawater endmembers and total dissolved Zr, note log scale; c. comparison of river-sourced Zr as a function of salinity during conservative mixing and scavenged removal, dotted lines indicate salinity at which 50% of Zr contributed by the river remains; d. percent Zr removed by scavenged removal relative to dilution by conservative mixing.  6.5 Discussion  Elevated concentrations of Ga and Zr observed at Stations T9:6-7 are consistent with near conservative mixing.  Alternatively, given persistent concerns of contamination during trace metal sampling and analysis, these elevated data points could be suspected to be a result of   153 contamination.  However, sample analysis was repeated and consistent values obtained, therefore lending support to the analytical results.  While contamination during sampling remains a possibility, TS data (Figure 6.2b) suggests that waters at Stations T9:6-7 are distinct.  Further, Ga exhibits a lesser contamination risk than ubiquitous elements such as Al and Fe, for example.  Therefore, given elevated concentrations at T9:6-7 suggestive of conservative mixing in contrast to scavenged removal observed at other Stations; do mechanisms exist to reconcile these divergent removal processes occurring within close geographic proximity?    Fresh waters of the Columbia River plume exhibit dynamic fluxes north and south of the Columbia River mouth, as a function of the prevailing wind stress (Berdeal, et al. 2002; Hickey, et al. 2005).  Southerly winds, blowing from the south toward the north, transport plume waters to the north, producing downwelling conditions and transporting offshore ocean surface waters along the coast as a result of onshore Ekman transport.  Conversely, northerly winds, blowing toward the south from the north, transport plume waters to the south, with the Ekman component resulting in upwelling along the coast and transport of Columbia River freshwaters offshore towards the southwest.    Controls on transport of Columbia River plume waters imparted by the winds produce the observed river-ocean mixing zone and therefore require consideration of local winds associated with the sampling period.  Frequent shifting of the prevailing winds within the study area is indicated in Figure 6.7, demonstrating multiple reversals of prevailing wind patterns during the month preceding sampling along transects T7, T8, and T9.      154  Figure 6.7 Stick vectors indicating wind direction and magnitude (NOAA National Data Buoy Center, Destruction Island buoy, location indicated in Figure 6.2).  Downwelling winds point up, blowing from the south toward the north.  Upwelling winds point down, blowing from north towards the south.  Sampling along transects 7, 8, and 9 are indicated by T7, T8, and T9.  Detailed winds and resulting proposed plume transport are summarized in Figure 8.  Coupled transport of plume waters to the north (June 25 – 28) followed by wind reversal on June 29 (Figure 6.8d), results in a return of northerly advected waters to the south.  Columbia River plume waters returning from north of the river mouth back to the south have been termed Aged Plume waters (Hickey, et al. 2005; Hickey, et al. 2009).  Northward onshore transport of plume waters (Figure 6.8a), followed by southerly offshore flow (Figure 6.8b) therefore results in new Columbia River plume waters near the coast, with aged waters along the periphery of offshore new plume waters (Figure 6.8b).  Linear conservative mixing of temperature and salinity is indicated along T9 offshore toward station 5 (Figure 6.8c), however, stations 6 and 7 are warmer at similar salinities, indicative of warm surface waters transported north under downwelling conditions.  Station 6 (T9) demonstrates the highest temperatures along the transect, coincident with Ga concentrations of closest proximity to conservative mixing (Figure 6.3c).      155   Figure 6.8 Summary figure indicated processes associated with sampling of new and aged plume waters along T9.  a. Dotted arrows indicate northward wind direction and plume transport during the interval June 26-28, encompassed in panel d by the dotted lines, solid arrow indicates Ekman transport.  b. Dashed arrows correspond to plume transport during June 29 – July 5, indicated in panel d. by the dashed box.  c. temperature – salinity along transect 9, stations numbered as in Figure 6.2.  d. detail of winds presented in Figure 9, T7, T8, and T9 indicate respective transect sampling dates.  What leads to conservative behaviour of Ga and Zr within aged plume waters, while scavenged removal is observed within new plume waters?  Upwelling and downwelling associated with advective plume transport influence nitrate and iron availability, as remineralized upwelling sources augment contributions from the Columbia River (Bruland, et al. 2008).  Nutrient controls on phytoplankton abundance and the resulting scavenging of Ga and Zr as a result of phytoplankton surface area is therefore proposed to provide for the observed contrasting concentrations of Ga and Zr indicative of scavenged removal in the new plume and conservative mixing within aged plume waters.     156  Plumes formed under upwelling and downwelling conditions are therefore unique as nitrate and iron are sourced from upwelling (Bruland, et al. 2008), while silicic acid is sourced from the Columbia River.  Low nitrate, low Fe surface waters pushed onshore by Ekman transport during downwelling plumes are therefore expected to result in lower primary productivity and reduced scavenging rates relative to upwelling plumes.  In addition to a reduction in scavenging rates, downwelling conditions entrain offshore surface waters low in both nutrients and dissolved Ga and Zr concentrations into the northward travelling plume.  Mixing with low concentration Ga and Zr surface waters under low primary production conditions results in linear conservative mixing of Ga and Zr added by the river.  Addition of scavenged removal mechanisms under upwelling conditions leads to the curvature related to non-conservative removal processes.    Conservative mixing of new plume waters is suggested by linear Si(OH)4 plots, as observed along T9at stations 1-4 (Figure 6.9a).  Low Si(OH)4 concentrations observed at stations 5, 6, and 7 are representative of surface ocean concentrations, indicating that the Si(OH)4 sourced from the river has been diluted sufficiently to be consistent with surface seawater in this region.  Similarly, low NO3- concentrations are observed within waters transported north (Figure 6.9b), as low NO3- surface waters are pushed onshore.  In contrast, sources of nitrate and iron during upwelling of waters sampled at Stations 1-5 promote phytoplankton growth and observed scavenging of Ga and Zr.    157  Figure 6.9 a. Si(OH)4 and b. NO3- concentrations along Transect 9 as a function of salinity.  Station numbers where trace metal samples were collected are indicated along the top of the plot.  Upwelling conditions and resulting increases in scavenging have been shown to decrease the residence time of Al (Orians and Bruland 1988b).  Gallium exhibits scavenging similar to its analogue Al, as both occupy the same column of the periodic table, thus sharing the same valence electron configuration.  Gallium, however, exhibits a lesser proportion of neutral speciation, Ga(OH)30, compared to Al (Orians and Bruland 1988a), and is therefore less rapidly removed from the water column via scavenging and thus acts as a longer-lived tracer.  Zirconium demonstrates hydroxylated speciation (Turner, et al. 1981), similar to Al and Ga, therefore expected to exhibit similar scavenged surface removal.    Nutrient controls on trace metal scavenging has been demonstrated for Al, as residence times (Al) in low productivity gyres of 3 – 4 years decrease to 100 days with greater productivity in the California current and in the presence of increased upwelling,Al values of only 35 days are observed (Orians and Bruland 1988b).  Similarly, therefore, lower scavenging rates are expected under low nutrient downwelling conditions associated with northward Columbia River plume transport, versus accelerated scavenging during upwelling conditions.  Downwelling conditions associated with the Columbia River plume have been shown to result in Al concentrations lying along a conservative mixing line at S = 13 – 25, while during upwelling conditions Al values fall   158 below the conservative mixing line, indicative of scavenged removal (Brown and Bruland 2009).  Therefore scavenged removal is suggested under nitrate-replete upwelling conditions with decreased removal under downwelling conditions.    In summary, Transect T9 along the mouth of the Columbia River is considered to sample new plume waters along Stations 1 – 5, and aged plume waters at stations 6 and 7 (Figure 6.8b).  Evaluation of winds prior to sampling (Figure 6.8d) supports intersection of T9 with aged plume waters offshore of new plume waters.  Transport of Columbia River waters toward the north under downwelling conditions with onshore Ekman transport results in warm, nutrient-poor, low particulate waters capping the surface.  High temperatures within northward transported Columbia River plume waters (Brown and Bruland 2009) are consistent with elevated temperatures at Station 6 and 7 along T9 (Figure 6.8c).  Nutrient-poor aged plume waters, initiated under downwelling conditions, are therefore anticipated to exhibit diminished particle scavenging, resulting in dilution of river inputs of Ga and Zr and demonstrating near conservative mixing on a timescale of days to weeks.  Conversely, new plume waters south of the Columbia River, influenced by enhancing productivity fuelled by upwelling of nitrate and Fe, exhibit scavenged removal.     6.6 Conclusion  A method is presented for the calculation of river and seawater endmembers, allowing for the determination of a conservative mixing line, in the absence of distinctly sampled endmembers.  Samples collected at S > 20 in this work demonstrate application of the method.  Results from the model suggest that elevated concentrations of Ga and Zr measured in samples 70 – 80km offshore result from near conservative mixing of the Columbia River plume, whereas scavenged removal is indicated along the remainder of the transect.    Evidence from wind data supports the presence of aged plume waters, formed under downwelling conditions, offshore of new plume waters influenced by upwelling.  Low nutrient   159 and phytoplankton abundance associated with downwelling northward plume transport results in reduced scavenging rates of Ga and Zr, whereas higher scavenging rates as a result of increased productivity resulting from nitrate and iron availability occurs during upwelling conditions.    Gallium and Zr demonstrate residence times appropriate to investigate scavenged removal from the plume.  Relative to Zr, Ga provides superior analytical precision and a reduced influence of the upwelling source observed for Zr.  Gallium is therefore proposed to provide identification of aged Columbia River plume waters.  Results of the presented method suggest the opportunity to calculate and apply a conservative mixing line to riverine-sourced trace metal concentrations wherein salinity values approaching zero are not available.      160 Chapter 7: Thesis conclusion  Results and conclusions presented in this thesis are synthesized here, including building a schematic view of trace metal inputs to Line P.  Brief summaries contained here are paired with summary figures.  Opportunities for future research resulting from this thesis are discussed and a final conclusion summarizes the novel tracer applications developed and presented in this thesis.   7.1 Trace metal sources: Line P synthesis  This thesis identifies 5 trace metal inputs along Line P, each described below and added sequentially to the Line P transect, building a summary figure of Line P.  Trace metal inputs along Line P identified in this thesis are shown to exhibit geographic and temporal variability based on interpretations of profiles of dissolved Ga, correlation with the variable spice, introduction of the Ga/Pb ratio, and interpretation of Pb isotopic ratios.   7.1.1 Input 1: Fluvial   Elevated Ga/Pb ratios are identified in this thesis to represent a marker of fluvial freshwater inputs (Chapter 3).  Ratios of Ga/Pb within the summer mixed layer at P4 and P12 in August 2010 are higher than Ga/Pb ratios observed within the mixed layer at P16 – P26, and coincide with lower salinity values.  Ratios of Ga/Pb are interpreted to be higher within freshwater inputs due to a combination of higher Ga concentrations associated with riverine inputs and a decrease in Pb concentrations due to scavenging of Pb with metal oxides during mixing of fresh and saline waters.   At P4 Ga/Pb ratios below the summer mixed layer converge to values measured at P16 – P26.  At P12 ratios of Ga/Pb remain high below the mixed layer at depths of 40m, 75m, and 100m, before converging at 150m with values observed at P16 – P26.  Elevated Ga/Pb ratios at depths of 40m, 75m and 100m at P12 relative to P16 – P26 are associated with the presence of an eddy   161 during sampling in August 2010.  Eddies are associated with elevated concentrations of dissolved Ga and decreased concentrations of dissolved Pb from fluvial inputs.  Eddy transport of coastal fluvial waters is therefore identified by the Ga/Pb ratio.    Fluvial inputs identified by Ga/Pb at P4 and P12 are illustrated in Figure 7.1, indicating that while fluvial inputs are confined to the summer mixed layer at P4, the presence of fluvial freshwaters transported offshore by the downwelling eddy are observed to depths of 100m at P12.  Additional inputs will be added in subsequent sections.     Figure 7.1: Line P schematic indicating regions of fluvial inputs based on Ga/Pb identified in this thesis.  Note log base10 depth scale to emphasize mixed layer dynamics across the transect.  Fluvial inputs present at P4 and P12 within the summer mixed layer (30m) extend to 100m at P12 resulting from eddy transport.    7.1.2 Input 2: North American source  Within the summer mixed layer at P4 and P12, 206Pb/207Pb and 208Pb/206Pb ratios are consistent with inputs from North American sources (Chapter 4).  Isotopic ratios observed in the summer mixed layer at P16 and P20 suggest similar sources as to these at P4 and P12.  However, signals   162 at P16 and P20 are not as prominent as P4 and P12, suggesting a reduced influence from North America with increasing distance west along Line P, approaching the edge of the Alaska Gyre.    Isotopic ratios of dissolved Pb measured in this thesis work in concert with the Ga/Pb ratio to identify trace metal inputs along Line P.  Based on Ga/Pb, North American sources to the summer mixed layer at stations P4 and P12 are interpreted to be fluvial.  Given confinement of North American source Pb isotope ratios at P16 and P20 to the summer mixed layer, a fluvial input, consistent with observations at P4 and P12, is interpreted to be advected and mixed within the summer mixed layer along Line P.  Fluvial North American sources are added in Figure 7.2 to the previously described fluvial inputs (Figure 7.1), extending within the summer mixed layer to the transition zone stations P16 and P20.  Cross-hatching indicates the presence of both fluvial inputs specifically identified by Ga/Pb and Pb isotope ratios of North American sources to the mixed layer at P4 and P12.     Figure 7.2: Isotopic ratios of Pb identify sources from North America at P4 – P20 within the summer mixed layer (30m).  Fluvial inputs are identified within the summer mixed layer at P4 and P12 by Ga/Pb, with advection within the summer mixed layer to stations P16 and P20 resulting in a transition zone across the transect (see Figure 4.11 and Figure 4.12).   163  7.1.3 Input 3: Asian source  Isotopic ratios of Pb at P26 trend the opposite direction as observations at P4 – P20 and are consistent with sources from Asia (Chapter 4).  In addition to geographic source differentiation at P26, temporal variability is observed.  While Asian Pb isotope signatures extend to depths of the previous winter mixed layer, North American ratios are confined to the summer mixed layer.  Inputs of atmospheric deposition from Asia to P26 are added to Figure 7.3, contrasting with inputs from Vancouver Island at station P20 and the remainder of the transect.  Summer and winter mixed layer depths are labeled, emphasizing the seasonally temporal differentiation of trace metal sources along Line P.     Figure 7.3: Eolian inputs from Asia are identified at P26 based on Pb isotopes and associated with winter mixed layer depths to 75m, contrasting with inputs from North American at P4 – P20 confined to the summer mixed layer.      164 7.1.4 Input 4: Advective– California Undercurrent   Positive correlation is demonstrated in this thesis between the oceanographic variable spice and concentrations of dissolved Ga at 150m – 200m at stations P12 – P26 (Chapter 3).  Advective transport from the south associated with the CUC therefore represents an additional trace metal input to Line P.  Temporal variation is observed with respect to spice along Line P, coincident with El Niño conditions and positive salinity anomalies in 2010 and 2011.  Negative salinity anomalies observed in 2009 and 2012 are coincident with lower spice waters, indicating interannual variability in the spice variable.  Increased advection of high spice water during El Niño could therefore increase trace metal availability, influencing primary production and observed shifts in distribution and abundance of commercial fish populations.   7.1.5 Input 5: Advective– North Pacific Intermediate Water   Relative to stations P16 and P20, dissolved Ga concentrations at P26 are lower within the local maxima at 150 – 200m, coinciding with lower spice values associated with NPIW.  A spice front is therefore identified across the north Pacific bisecting the Line P transect between stations P20 and P26.  Additionally, while variation in spice associated with El Niño is observed at stations P16 and P20, little variation in spice values are observed at P26.  Therefore representing a greater influence of NPIW advected from the east to P26.   Advective inputs from the CUC and NPIW are added to Figure 7.4.  This thesis therefore identifies multiple fronts of trace metal inputs along Line P between stations P26 and P20 including Asian and North American geographic sources, summer and winter temporal differentiation, and fluvial and eolian input mechanisms.         165  Figure 7.4: Spicy water at 150 – 200m at P20 and P16 correlate with local maxima of dissolved Ga concentrations, lower concentrations of dissolved Ga at P26 associated with lower spice NPIW.  Construction of this figure summarizes the trace metal inputs identified from results of this thesis and tracer applications of dissolved Ga and Pb isotopes developed in this thesis.  7.1.6 Trace metal sources: Future research  This work extends the delineation of ocean regimes along Line P, identifying trace metal inputs that vary in space and time across the transect and promoting future research of biogeochemical cycling in the northeast Pacific Ocean.  For instance, fluvial inputs indentified in this work encourage research given inter-seasonal and inter-annual variation in fluvial trace metal inputs to the ocean.  Additionally, climate change impacts on precipitation, timing of snow melt, and terrestrial weathering represent areas for future research to determine the impacts to fluvial inputs.  Eolian inputs to the Alaska Gyre and P26 could be impacted by climate change and shifts in timing, areal extent, or volume of atmospheric deposition as a function of changing wind or precipitation patterns.  Advective inputs of trace metals from the south could be impacted by   166 changes in timing or intensity of ENSO events.  Additionally, increased frequency or intensity of winter storms resulting in deep winter mixing could increase availability of advected trace metal inputs to the surface.  Each of these topics represents opportunity for future research, including collaboration with physical oceanographers and atmospheric scientists.  Including employing atmospheric remote sensing of shifting wind patterns and eolian deposition, and autonomous sampling devices such as ARGO floats to monitor and study spice distributions and advective inputs.  Delineation of these 5 input mechanisms to Line P therefore advances the field of oceanography, informing potential controls on HNLC conditions observed in the northeast Pacific Ocean, with implications to other HNLC areas of the World’s oceans in the present, past and future.  7.2 Pacific and Atlantic sources: Western Arctic  This work reports the first profiles of dissolved Ga in the Arctic Ocean (Chapter 5) and establishes dissolved Ga as a tracer of Pacific and Atlantic water masses in the Beaufort Sea of the Western Arctic Ocean (Figure 7.5).  Pacific sources represented by low concentration of dissolved Ga, 4 – 6 pmol kg-1, transition smoothly through the thermocline to warmer Atlantic waters of 25 – 30 pmol kg-1 dissolved Ga.  Dissolved Ga represents a conservative tracer of Pacific waters, providing a compliment tracer to N:P and temperature.      167  Figure 7.5: Dissolved Ga provides a tracer of Pacific source waters (blue) to the Arctic, with concentrations increasing through the thermocline (green), and higher concentrations observed in Atlantic source waters to the Arctic.    7.2.1 Pacific and Atlantic sources: Future research  Future measurement of dissolved Ga concentrations in the Bering Sea and Chukchi shelf will constrain seasonal modification of dissolved Ga, providing additional evaluation of dissolved Ga as a conservative tracer of Pacific waters and the global balance of freshwater in the ocean.  Additionally, elevated concentrations of dissolved Ga could be applied as a tracer of Atlantic waters infiltrating the western Arctic.  7.3 Nitrogen fixation: Western Arctic  Application of the conservative behaviour of dissolved Ga in Pacific waters supports the identification and calculation of in-situ nitrogen fixation within nitrate deplete waters of the Beaufort Sea basin (Chapter 5).  Denitrification results in excess PO4 transported from the Pacific, through the Arctic, and fuelling nitrogen fixation in the Atlantic.  Therefore, in-situ   168 nitrogen fixation in the Arctic will decrease the amount of available PO4 to be delivered to the Atlantic (Figure 7.6).     Figure 7.6: Intensity of red colour indicates decreasing concentrations of excess phosphate as a result of in-situ nitrogen fixation during transport of nitrate deficient waters across the Arctic and returning to the north Atlantic, providing a schematic summary representation of implications of nitrogen fixation identified in this thesis.    7.3.1 Nitrogen fixation: Future research  This thesis provides a baseline estimate of nitrogen fixation at locations within the Beaufort Sea.  Given rapid climate change in the Arctic and modification of sea ice conditions and stratification, increases in nitrogen fixation in the Beaufort Sea could lead to decreasing nitrogen fixation in the Atlantic.  While a shift in nitrogen fixation from the Atlantic to the Arctic may allow global nitrogen budgets to remain in balance, local ecosystems in both the Atlantic and Arctic would be impacted.  Continued research of nitrogen fixation within nitrate deficient waters during transport across the Arctic Ocean is required to evaluate impacts of excess phosphate delivery to the Atlantic.     169  7.4 River plume sources and sinks: Columbia River  Rivers provide both a source of trace metals to the ocean, as well as a sink, as plume waters mix with the ocean and promote scavenging.  This work develops a model to describe dissolved Ga dynamics associated with the Columbia River plume (Chapter 6).  Northward plume transport along the coast under downwelling conditions results in dilution of dissolved Ga concentrations consistent with conservative mixing between freshwater and ocean water (Figure 7.7).  Contrastingly, plume transport offshore to the southwest occurs during upwelling conditions resulting in scavenged removal of dissolved Ga, in addition to dilution (Figure 7.7).  Therefore the spatial distribution of dissolved Ga within the Columbia River plume is shown to result from temporal variability in local winds.     Figure 7.7: Graphical representation of results from development of endmember mixing model and interpretation of upwelling and downwelling favouring winds associated with the Columbia River plume, and resulting concentrations of trace metals, based on dissolved Ga, exhibiting conservative mixing and scavenged behaviour.    7.4.1 River plume sources and sinks: Future research  Rivers represent a source of many trace metals to the ocean.  Higher concentrations of trace metals sourced from rivers during downwelling conditions potentially influence local shelf and   170 offshore ecosystems. Autonomous in-situ instruments and real-time data can be used to inform sampling opportunities to evaluate trace metal sources from rivers to coastal ecosystems during differential plume conditions.    7.5 Thesis conclusions  This thesis describes biogeochemical cycling and input mechanisms for trace metals in the northeast Pacific along Line P, in the northwest coast of North America influenced by the Columbia River plume, and in the Beaufort Sea of the Western Arctic Ocean.  Conclusions of this thesis are based on measurement and interpretation of concentrations of dissolved gallium (Ga), ratios of dissolved Ga and Pb (Ga/Pb), and isotopic ratios of dissolved Pb.   This thesis introduces 5 new tracer applications contributing to oceanographic research.  Dissolved Ga traces: 1) advective sources of trace metals in the northeast Pacific, 2) Pacific and Atlantic source waters in the western Arctic, and 3) plume transport of the Columbia River.  Ga/Pb traces: 4) fluvial freshwater sources; and Pb isotope ratios trace: 5) geographically and temporally distinct trace metal sources to the northeast Pacific.  Results, applications, and conclusions of this thesis encourage increasingly routine measurement of dissolved Ga and Pb isotopes in oceanography.      171 References  Aagaard, K. and Carmack, E. C., 1989. The role of sea ice and other fresh-water in the Arctic circulation. Journal of Geophysical Research-Oceans. 94, 14485-14498. Abell, J., Emerson, S., Keil, R., 2005. Using preformed nitrate to infer decadal changes in DOM remineralization in the subtropical North Pacific. Global Biogeochemical Cycles. 19, GB1008. Abouchami, W., Galer, S., Koschinsky, A., 1999. Pb and Nd isotopes in NE Atlantic Fe-Mn crusts: Proxies for trace metal paleosources and paleocean circulation. Geochimica Et Cosmochimica Acta. 63, 1489-1505. Addison, J. A., Finney, B. P., Dean, W. E., Davies, M. H., Mix, A. C., Stoner, J. S., Jaeger, J. M., 2012. Productivity and sedimentary delta N-15 variability for the last 17,000 years along the northern Gulf of Alaska continental slope. Paleoceanography. 27, PA1206. Aguilar-Islas, A. M. and Bruland, K. W., 2006. Dissolved manganese and silicic acid in the Columbia River plume: A major source to the California current and coastal waters off Washington and Oregon. Marine Chemistry. 101, 233-247. Aguilar-Islas, A. M., Hurst, M. P., Buck, K. N., Sohst, B., Smith, G. J., Lohan, M. C., Bruland, K. W., 2007. Micro- and macronutrients in the southeastern Bering Sea: Insight into iron-replete and iron-depleted regimes. Progress in Oceanography. 73, 99-126. Alleman, L. Y., Church, T. M., Ganguli, P., Veron, A. J., Hamelin, B., Flegal, A. R., 2001. Role of oceanic circulation on contaminant lead distribution in the South Atlantic. Deep-Sea Research Part Ii-Topical Studies in Oceanography. 48, 2855-2876. Alleman, L. Y., Hamelin, B., Veron, A. J., Miquel, J. C., Heussner, S., 2000. Lead sources and transfer in the coastal Mediterranean: evidence from stable lead isotopes in marine particles. Deep-Sea Research Part Ii-Topical Studies in Oceanography. 47, 2257-2279. Bacon, M. P. and Anderson, R. F., 1982. Distribution of Thorium Isotopes between Dissolved and Particulate Forms in the Deep-Sea. Journal of Geophysical Research-Oceans and Atmospheres. 87, 2045-2056. Banas, N. S., MacCready, P., Hickey, B. M., 2009. The Columbia River plume as cross-shelf exporter and along-coast barrier. Continental Shelf Research. 29, 292-301. Banner, J., 2004. Radiogenic isotopes: systematics and applications to earth surface processes and chemical stratigraphy. Earth-Science Reviews. 65, 141-194.   172 Barber, D. G., Lukovich, J. V., Keogak, J., Baryluk, S., Fortier, L., Henry, G. H. R., 2008. The changing climate of the Arctic. Arctic. 61, 7-26. Bauch, D., van der Loeff, M. R., Andersen, N., Torres-Valdes, S., Bakker, K., Abrahamsen, E. P., 2011. Origin of freshwater and polynya water in the Arctic Ocean halocline in summer 2007. Progress in Oceanography. 91, 482-495. Berdeal, I. G., Hickey, B. M., Kawase, M., 2002. Influence of wind stress and ambient flow on a high discharge river plume. Journal of Geophysical Research-Oceans. 107, 3130. Biller, D. V. and Bruland, K. W., 2012. Analysis of Mn, Fe, Co, Ni, Cu, Zn, Cd, and Pb in seawater using the Nobias-chelate PA1 resin and magnetic sector inductively coupled plasma mass spectrometry (ICP-MS). Marine Chemistry. 130, 12-20. Bishop, J. K. B., Davis, R. E., Sherman, J. T., 2002. Robotic observations of dust storm enhancement of carbon biomass in the North Pacific. Science. 298, 817-821. Blais, M., Tremblay, J.E., Jungblut, A.D., Gagnon, J., Martin, J., Thaler, M., Lovejoy, C., 2012. Nitrogen fixation and identification of potential diazatrophs in the Canadian Arctic. Global Biogeochemical Cycles. 26, GB3022.  Bollhofer, A. and Rosman, K., 2002. The temporal stability in lead isotopic signatures at selected sites in the Southern and Northern Hemispheres. Geochimica Et Cosmochimica Acta. 66, 1375-1386. Bollhofer, A. and Rosman, K., 2001. Isotopic source signatures for atmospheric lead: The Northern Hemisphere. Geochimica Et Cosmochimica Acta. 65, 1727-1740. Bonisch, G. and Schlosser, P., 1995. Deep water formation and exchange rates in the Greenland/Norwegian Seas and the Eurasian Basin of the Arctic Ocean derived from tracer balances. Progress in Oceanography. 35, 29-52.  Boyd, P. and Harrison, P. J., 1999. Phytoplankton dynamics in the NE subarctic Pacific. Deep-Sea Research Part Ii-Topical Studies in Oceanography. 46, 2405-2432. Boyd, P. W., Wong, C. S., Merrill, J., Whitney, F., Snow, J., Harrison, P. J., Gower, J., 1998. Atmospheric iron supply and enhanced vertical carbon flux in the NE subarctic Pacific: Is there a connection? Global Biogeochemical Cycles. 12, 429-441. Boyd, P., Muggli, D., Varela, D., Goldblatt, R., Chretien, R., Orians, K., Harrison, P., 1996. In vitro iron enrichment experiments in the NE subarctic Pacific. Marine Ecology Progress Series. 136, 179-193.   173 Boyle, E., Collier, R., Dengler, A. T., Edmond, J. M., Ng, A. C., Stallard, R. F., 1974. Chemical mass-balance in estuaries. Geochimica Et Cosmochimica Acta. 38, 1719-1728. Boyle, E. A., Edmond, J. M., Sholkovitz, E. R., 1977. Mechanism of iron removal in estuaries. Geochimica Et Cosmochimica Acta. 41, 1313-1324. Boyle, E., Sclater, F., Edmond, J., 1977. Distribution of dissolved copper in Pacific. Earth and Planetary Science Letters. 37, 38-54. Boyle, E. A., John, S., Abouchami, W., Adkins, J. F., Echegoyen-Sanz, Y., Ellwood, M., Flegal, A. R., Fornace, K., Gallon, C., Galer, S., Gault-Ringold, M., Lacan, F., Radic, A., Rehkamper, M., Rouxel, O., Sohrin, Y., Stirling, C., Thompson, C., Vance, D., Xue, Z., Zhao, Y., 2012. GEOTRACES IC1 (BATS) contamination-prone trace element isotopes Cd, Fe, Pb, Zn, Cu, and Mo intercalibration. Limnology and Oceanography-Methods. 10, 653-665. Boyle, E. A., Lee, J., Echegoyen, Y., Noble, A., Moos, S., Carrasco, G., Zhao, N., Kayser, R., Zhang, J., Gamo, T., Obata, H., Norisuye, K., 2014. Anthropogenic lead emission in the ocean the evolving global experiment. Oceanography. 27, 69-75. Broecker, W. S., 1974. No a conservative water-mass tracer. Earth and Planetary Science Letters. 23, 100-107. Brown, M. T. and Bruland, K. W., 2009. Dissolved and particulate aluminum in the Columbia River and coastal waters of Oregon and Washington: Behavior in near-field and far-field plumes. Estuarine Coastal and Shelf Science. 84, 171-185. Brown, M. T., Lippiatt, S. M., Lohan, M. C., Bruland, K. W., 2012. Trace metal distributions within a Sitka eddy in the northern Gulf of Alaska. Limnology and Oceanography. 57, 503-518. Bruland, K.W. and Lohan, M.C., 2003. Controls of trace metals in seawater. Treatise on Geochemistry, 6: 23-47.    Bruland, K.W. and Lohan, M.C., 2008. Elevated Fe(II) and dissolved Fe in hypoxic shelf water off Oregon and Washington: An enhanced source of iron to coastal upwelling regimes. Environmental Science & Technology. 42. 6462-6468.  Bruland, K. W., Knauer, G. A., Martin, J. H., 1978a. Cadmium in northeast Pacific waters. Limnology and Oceanography. 23, 618-625. Bruland, K. W., Knauer, G. A., Martin, J. H., 1978b. Zinc in northeast Pacific water. Nature. 271, 741-743. Bruland, K. W., Orians, K. J., Cowen, J. P., 1994. Reactive trace-metals in the stratified central north Pacific. Geochimica Et Cosmochimica Acta. 58, 3171-3182.   174 Bruland, K. W., Lohan, M. C., Aguilar-Islas, A. M., Smith, G. J., Sohst, B., Baptista, A., 2008. Factors influencing the chemistry of the near-field Columbia River plume: Nitrate, silicic acid, dissolved Fe, and dissolved Mn. Journal of Geophysical Research-Oceans. 113, C00B02. Budikova, D., 2009. Role of Arctic sea ice in global atmospheric circulation: A review. Global and Planetary Change. 68, 149-163. Cain, A. N., 2014. Dissolved aluminum in the northeast Pacific and the western Arctic Ocean. MSc Thesis, University of British Columbia.    Carmack, E. C., Aagaard, K., Swift, J. H., MacDonald, R. W., McLaughlin, F. A., Jones, E. P., Perkin, R. G., Smith, J. N., Ellis, K. M., Killius, L. R., 1997. Changes in temperature and tracer distributions within the Arctic Ocean: Results from the 1994 Arctic Ocean section. Deep-Sea Research Part Ii-Topical Studies in Oceanography. 44, 1487. Carmack, E. C., 2007. The alpha/beta ocean distinction: A perspective on freshwater fluxes, convection, nutrients and productivity in high-latitude seas. Deep-Sea Research Part Ii-Topical Studies in Oceanography. 54, 2578-2598. Carmack, E. and Wassmann, P., 2006. Food webs and physical-biological coupling on pan-Arctic shelves: Unifying concepts and comprehensive perspectives. Progress in Oceanography. 71, 446-477. Charters, J. W., 2012. Determination and biogeochemistry of Pb in the Northeastern Subarctic Pacific Ocean and the Beaufort Sea of the Arctic Ocean. MSc Thesis, University of British Columbia. Chen, J., Tan, M., Li, Y., Zheng, J., Zhang, Y., Shan, Z., Zhang, G., Li, Y., 2008. Characteristics of trace elements and lead isotope ratios in PM2.5 from four sites in Shanghai. Journal of Hazardous Materials. 156, 36-43. Cheng, H. and Hu, Y., 2010. Lead (Pb) isotopic fingerprinting and its applications in lead pollution studies in China: A review. Environmental Pollution. 158, 1134-1146. Collins, C.A., Margolina, T., Rago, T.A., Ivanov, L., 2013. Looping RAFOS floats in the California Current system. Deep-Sea Research II. 85, 42-61.  Comiso, J. C., Parkinson, C. L., Gersten, R., Stock, L., 2008. Accelerated decline in the Arctic Sea ice cover. Geophysical Research Letters. 35, L01703. Cowart, J. and Burnett, W., 1994. The distribution of uranium and thorium decay-series radionuclides in the environment - a Review. Journal of Environmental Quality. 23, 651-662.   175 Crawford, W. R. and Greisman, P., 1987. Investigation of permanent eddies in Dixon Entrance, British-Columbia. Continental Shelf Research. 7, 851-870. Crawford, W. R., Brickley, P. J., Thomas, A. C., 2007a. Mesoscale eddies dominate surface phytoplankton in northern Gulf of Alaska. Progress in Oceanography. 75, 287-303. Crawford, W., Galbraith, J., Bolingbroke, N., 2007b. Line P ocean temperature and salinity, 1956-2005. Progress in Oceanography. 75, 161-178. Csavina, J., Field, J., Taylor, M. P., Gao, S., Landazuri, A., Betterton, E. A., Saez, A. E., 2012. A review on the importance of metals and metalloids in atmospheric dust and aerosol from mining operations. Science of the Total Environment. 433, 58-73. Cullen, J. T., Chase, Z., Coale, K. H., Fitzwater, S. E., Sherrell, R. M., 2003. Effect of iron limitation on the cadmium to phosphorus ratio of natural phytoplankton assemblages from the Southern Ocean. Limnology and Oceanography. 48, 1079-1087. Cullen, J. T., Chong, M., Ianson, D., 2009. British Columbian continental shelf as a source of dissolved iron to the subarctic northeast Pacific Ocean. Global Biogeochemical Cycles. 23, GB4012. de Baar, H., Boyd, P., Coale, K., Landry, M., Tsuda, A., Assmy, P., Bakker, D., Bozec, Y., Barber, R., Brzezinski, M., Buesseler, K., Boye, M., Croot, P., Gervais, F., Gorbunov, M., Harrison, P., Hiscock, W., Laan, P., Lancelot, C., Law, C., Levasseur, M., Marchetti, A., Millero, F., Nishioka, J., Nojiri, Y., van Oijen, T., Riebesell, U., Rijkenberg, M., Saito, H., Takeda, S., Timmermans, K., Veldhuis, M., Waite, A., Wong, C., 2005. Synthesis of iron fertilization experiments: From the iron age in the age of enlightenment. Journal of Geophysical Research-Oceans. 110, C09S16. Devol, A.H., Codispoti, L.A., Christensen, J.P., 1997. Summer and winter denitrification rates in western Arctic shelf sediments. Continental Shelf Research. 17, 1029-1050. Doney, S. C., Bopp, L., Long, M. C., 2014. Historical and future trends in ocean climate and biogeochemistry. Oceanography. 27, 108-119. Dore, J.E., Brum, J.R., Tupas, L.M., Karl, D.M., 2002. Seasonal and interannual variability in sources of nitrogen supporting export in the oligotrophic north Pacific Ocean. Limnology and Oceanography. 47, 1595-1607.  Elderfield, H. and Rickaby, R., 2000. Oceanic Cd/P ratio and nutrient utilization in the glacial Southern Ocean. Nature. 405, 305-310. Emery, T., 1986. Exchange of iron by gallium in siderophores. Biochemistry. 25, 4629-4633.   176 Falkner, K. K., Steele, M., Woodgate, R. A., Swift, J. H., Aagaard, K., Morison, J., 2005. Dissolved oxygen extrema in the Arctic Ocean halocline from the North Pole to the Lincoln Sea. Deep-Sea Research Part I: Oceanographic Research Papers. 52, 1138-1154. Favorite, F., Dodimead, A.J., Nasu, K., 1976. Oceanography of the subarctic Pacific region. International North Pacific Fisheries Commission. Bulletin Number 33.  Fiechter, J. and Moore, A.M., 2012. Iron limitation impact on eddy-induced ecosystem variability in the coastal Gulf of Alaska. Journal of Marine Systems. 92, 1-15.  Firdaus, M.L., Minami, T., Norisuye, K., Sohrin, Y., 2011. Strong element fractionation of Zr-Hf and Nb-Ta across the Pacific Ocean. Nature Geoscience 4, 227-230. Flament, P., 2002. A state variable for characterizing water masses and their diffusive stability: spiciness. Progress in Oceanography. 54, 493-501. Flegal, A. R., Duda, T. F., Niemeyer, S., 1989. High gradients of lead isotopic composition in northeast Pacific upwelling filaments. Nature. 339, 458-460. Flegal, A. R., Itoh, K., Patterson, C. C., Wong, C. S., 1986. Vertical profile of lead isotopic compositions in the northeast Pacific. Nature. 321, 689-690. Frank, M., 2002. Radiogenic isotopes: Tracers of past ocean circulation and erosional input. Reviews of Geophysics. 40, 1001. Feeland, H.J., Crawford, W.R., Thompson, R.E., 1984. Currents along the Pacific coast of Canada. Atmosphere-Ocean. 22, 151-172.  Freeland, H., 2007. A short history of ocean station papa and Line P. Progress in Oceanography. 75, 120-125. Gaillardet, J., Viers, J. and Dupre, B., 2003. Trace elements in river waters. Treatise on Geochemistry, 5: 225-272. Gallon, C., Ranville, M. A., Conaway, C. H., Landing, W. M., Buck, C. S., Morton, P. L., Flegal, A. R., 2011. Asian industrial lead inputs to the north Pacific evidenced by lead concentrations and isotopic compositions in surface waters and aerosols. Environmental Science & Technology. 45, 9874-9882. Garfield, N., Collins, C.A., Paquette, R.G., Carter, E., 1999. Langrangian exploration of the California Undercurrent, 1992-1995. Journal of Physical Oceanography. 29, 560-583.    177 Gehlen, M., Beck, L., Calas, G., Flank, A. M., Van Bennekom, A. J., Van Beusekom, J. E. E., 2002. Unraveling the atomic structure of biogenic silica: Evidence of the structural association of Al and Si in diatom frustules. Geochimica Et Cosmochimica Acta. 66, 1601-1609. Giesbrecht, T., Sim, N., Orians, K. J., Cullen, J. T., 2013. The distribution of dissolved and total dissolvable aluminum in the Beaufort Sea and Canada Basin region of the Arctic Ocean. Journal of Geophysical Research-Oceans. 118, 6824-6837. Godfrey, L. V., Field, M. P., Sherrell, R. M., 2008. Estuarine distributions of Zr, Hf, and Ag in the Hudson River and the implications for their continental and anthropogenic sources to seawater. Geochemistry Geophysics Geosystems. 9, Q12007. Godfrey, L. V., White, W. M., Salters, V. J. M., 1996. Dissolved zirconium and hafnium distributions across a shelf break in the northeastern Atlantic Ocean. Geochimica Et Cosmochimica Acta. 60, 3995-4006. Grand, M. M., Buck, C. S., Landing, W. M., Measures, C. I., Hatta, M., Hiscock, W. T., Brown, M., Resing, J. A., 2014. Quantifying the impact of atmospheric deposition on the biogeochemistry of Fe and Al in the upper ocean A decade of collaboration with the US CLIVAR-CO2 Repeat Hydrography Program. Oceanography. 27, 62-65. Graney, J., Halliday, A., Keeler, G., Nriagu, J., Robbins, J., Norton, S., 1995. Isotopic record of lead pollution in lake-sediments from the northeastern United-States. Geochimica Et Cosmochimica Acta. 59, 1715-1728. Greene, A. R., Scoates, J. S., Weis, D., Nixon, G. T., Kieffer, B., 2009. Melting history and magmatic evolution of basalts and picrites from the accreted Wrangellia Oceanic Plateau, Vancouver Island, Canada. Journal of Petrology. 50, 467-505. Gross, B. H., Kreutz, K. J., Osterberg, E. C., McConnell, J. R., Handley, M., Wake, C. P., Yalcin, K., 2012. Constraining recent lead pollution sources in the North Pacific using ice core stable lead isotopes. Journal of Geophysical Research-Atmospheres. 117, D16307. Gruber, N. and Sarmiento, J., 1997. Global patterns of marine nitrogen fixation and denitrification. Global Biogeochemical Cycles. 11, 235-266. Guay, C. K. H., McLaughlin, F. A., Yamamoto-Kawai, M., 2009. Differentiating fluvial components of upper Canada Basin waters on the basis of measurements of dissolved barium combined with other physical and chemical tracers. Journal of Geophysical Research-Oceans. 114, C00A09. Gulson, B., Korsch, M., Winchester, W., Devenish, M., Hobbs, T., Main, C., Smith, G., Rosman, K., Howearth, L., Burn-Nunes, L., Seow, J., Oxford, C., Yun, G., Gillam, L., Crisp, M., 2012.   178 Successful application of lead isotopes in source apportionment, legal proceedings, remediation and monitoring. Environmental Research. 112, 100-110. Hamme, R. C., Webley, P. W., Crawford, W. R., Whitney, F. A., DeGrandpre, M. D., Emerson, S. R., Eriksen, C. C., Giesbrecht, K. E., Gower, J. F. R., Kavanaugh, M. T., Pena, M. A., Sabine, C. L., Batten, S. D., Coogan, L. A., Grundle, D. S., Lockwood, D., 2010. Volcanic ash fuels anomalous plankton bloom in subarctic northeast Pacific. Geophysical Research Letters. 37, L19604. Han, Q., Moore, J. K., Zender, C., Measures, C., Hydes, D., 2008. Constraining oceanic dust deposition using surface ocean dissolved Al. Global Biogeochemical Cycles. 22, GB2003. Henderson, G. M., 2002. New oceanic proxies for paleoclimate. Earth and Planetary Science Letters. 203, 1-13. Henderson, G. M., Anderson, R. F., Adkins, J., Andersson, P., Boyle, E. A., Cutter, G., de Baar, H., Eisenhauer, A., Frank, M., Francois, R., Orians, K., Gamo, T., German, C., Jenkins, W., Moffett, J., Jeandel, C., Jickells, T., Krishnaswami, S., Mackey, D., Measures, C. I., Moore, J. K., Oschlies, A., Pollard, R., van der Loeff, M. R., Schlitzer, R., Sharma, M., von Damm, K., Zhang, J., Masque, P., SCOR Working Grp, 2007. GEOTRACES - An international study of the global marine biogeochemical cycles of trace elements and their isotopes. Chemie Der Erde-Geochemistry. 67, 85-131. Hickey, B., Geier, S., Kachel, N., MacFadyen, A. F., 2005. A bi-directional river plume: The Columbia in summer. Continental Shelf Research. 25, 1631-1656. Hickey, B., McCabe, R., Geier, S., Dever, E., Kachel, N., 2009. Three interacting freshwater plumes in the northern California Current System. Journal of Geophysical Research-Oceans. 114, C00B03. Hickey, B. M., Kudela, R. M., Nash, J. D., Bruland, K. W., Peterson, W. T., MacCready, P., Lessard, E. J., Jay, D. A., Banas, N. S., Baptista, A. M., Dever, E. P., Kosro, P. M., Kilcher, L. K., Horner-Devine, A. R., Zaron, E. D., McCabe, R. M., Peterson, J. O., Orton, P. M., Pan, J., Lohan, M. C., 2010. River influences on shelf ecosystems: Introduction and synthesis. Journal of Geophysical Research-Oceans. 115, C00B17. Hickey, B.M., Thomson, R.E., Yih, H., LeBlond, P.H., 1991. Velocity and temperature fluctuations in a buoyancy-driven current off Vancouver Island. Journal of Geophysical Research. 96, 10507-10538.  Holiday, L.M., Liss, P.S,, 1976. The behaviour of dissolved iron, manganese and zinc in the Beaulieu Estuary, S. England. Estuarine and Coastal Marine Science 4, 349-353.   179 Hutchings, J. A., Cote, I. M., Dodson, J. J., Fleming, I. A., Jennings, S., Mantua, N. J., Peterman, R. M., Riddell, B. E., Weaver, A. J., 2012. Climate change, fisheries, and aquaculture: trends and consequences for Canadian marine biodiversity. Environmental Reviews. 20, 220-311. Hydes, D. J., Delange, G. J., Debaar, H. J. W., 1988. Dissolved aluminum in the mediterranean. Geochimica Et Cosmochimica Acta. 52, 2107-2114. Inoue, M. and Tanimizu, M., 2008. Anthropogenic lead inputs to the western Pacific during the 20th century. Science of the Total Environment. 406, 123-130. Jackson, J. M., Carmack, E. C., McLaughlin, F. A., Allen, S. E., Ingram, R. G., 2010. Identification, characterization, and change of the near-surface temperature maximum in the Canada Basin, 1993-2008. Journal of Geophysical Research-Oceans. 115, C05021. Jay, D.A., Smith, J.D., 1990. Circulation, density distribution and neap-spring transistions in the Columbia River estuary. Progress in Oceanography. 25, 81-112. Jakuba, R. W., Saito, M. A., Moffett, J. W., Xu, Y., 2012. Dissolved zinc in the subarctic North Pacific and Bering Sea: Its distribution, speciation, and importance to primary producers. Global Biogeochemical Cycles. 26, GB2015. Janssen, D. J., Conway, T. M., John, S. G., Christian, J. R., Kramer, D. I., Pedersen, T. F., Cullen, J. T., 2014. Undocumented water column sink for cadmium in open ocean oxygen-deficient zones. Proceedings of the National Academy of Sciences of the United States of America. 111, 6888-6893. Janssen, D.J. and Cullen, J., T., 2014. Improvements to a fluorescence-based flow-injection method for shipboard determination of dissolved zinc. Poster Presentation, Ocean Sciences Conference. Jeandel, C., Arsouze, T., Lacan, F., Techine, P., Dutay, J. -., 2007. Isotopic Nd compositions and concentrations of the lithogenic inputs into the ocean: A compilation, with an emphasis on the margins. Chemical Geology. 239, 156-164. Johnson, W. K., Miller, L. A., Sutherland, N. E., Wong, C. S., 2005. Iron transport by mesoscale Haida eddies in the Gulf of Alaska. Deep-Sea Research Part Ii-Topical Studies in Oceanography. 52, 933-953. Johnson, K.E., Harmon, R.S., Richardson, J.M., Moorbath, S., Strong, D.F., 1996. Isotope and trace element geochemmistry of Augustine Volcano, Alaska: Implications for magmatic evolution. Journal of Petrology. 37, 95-115.  Jones, C. J. and Murray, J. W., 1985. The geochemistry of Manganese in the northeast Pacific-Ocean Off Washington. Limnology and Oceanography. 30, 81-92.   180 Jones, E. P., Anderson, L. G., Swift, J. H., 1998. Distribution of Atlantic and Pacific waters in the upper Arctic Ocean: Implications for circulation. Geophysical Research Letters. 25, 765-768. Jones, E. P., Swift, J. H., Anderson, L. G., Lipizer, M., Civitarese, G., Falkner, K. K., Kattner, G., McLaughlin, F., 2003. Tracing Pacific water in the north Atlantic Ocean. Journal of Geophysical Research-Oceans. 108, 3116. Jones, E. P., Anderson, L. G., Jutterstrom, S., Mintrop, L., Swift, J. H., 2008. Pacific freshwater, river water and sea ice meltwater across Arctic Ocean basins: Results from the 2005 Beringia Expedition. Journal of Geophysical Research-Oceans. 113, C08012. Kara, A., Rochford, P., Hurlburt, H., 2000. Mixed layer depth variability and barrier layer formation over the North Pacific Ocean. Journal of Geophysical Research-Oceans. 105, 16783-16801. Karl D., Michales, A., Bergman, B., Capone, D., Carpenter, E., Letelier, R., Lipschultz, F., Paerl, H., Sigman, D., Stal, L., 2002. Dinitrogen fixation in the World's oceans. Biogeochemistry, 57, 47-98. Keigwin, L., Jones, G., Lehman, S., Boyle, E., 1991. Deglacial meltwater discharge, North-Atlantic deep circulation, and abrupt climate change. Journal of Geophysical Research-Oceans. 96, 16811-16826. Klunder, M. B., Laan, P., Middag, R., de Baar, H. J. W., Bakker, K., 2012. Dissolved iron in the Arctic Ocean: Important role of hydrothermal sources, shelf input and scavenging removal. Journal of Geophysical Research-Oceans. 117, C04014. Komarek, M., Ettler, V., Chrastny, V., Mihaljevic, M., 2008. Lead isotopes in environmental sciences: A review. Environment International. 34, 562-577. Lacan, F. and Jeandel, C., 2005. Neodymium isotopes as a new tool for quantifying exchange fluxes at the continent-ocean interface. Earth and Planetary Science Letters. 232, 245-257. Lanthier, K., 1999. Sources of Ga, In, Zr, and Hf to the coastal waters of the California Current system. MSc Thesis, University of British Columbia.  Lam, P.J., Bishop, J. K. B., Henning, C. C., Marcus, M. A., Waychunas, G. A., Fung, I. Y., 2006. Wintertime phytoplankton bloom in the subarctic Pacific supported by continental margin iron. Global Biogeochemical cycles. 20, GB1006.  Lam, P. J. and Bishop, J. K. B., 2008. The continental margin is a key source of iron to the HNLC North Pacific Ocean. Geophysical Research Letters. 35, L07608.   181 Lane, T. W. and Morel, F. M. M., 2000. A biological function for cadmium in marine diatoms. Proceedings of the National Academy of Sciences of the United States of America. 97, 4627-4631. Li, W. K. W., McLaughlin, F. A., Lovejoy, C., Carmack, E. C., 2009. Smallest algae thrive as the Arctic Ocean freshens. Science. 326, 539-539. Li, Y. H., 1991. Distribution patterns of the elements in the Ocean - a Synthesis. Geochimica Et Cosmochimica Acta. 55, 3223-3240. Lippiatt, S. M., Brown, M. T., Lohan, M. C., Berger, C. J. M., Bruland, K. W., 2010. Leachable particulate iron in the Columbia River, estuary, and near-field plume. Estuarine Coastal and Shelf Science. 87, 33-42. Liss, P.S. 1976.  Conservative and non-conservative behaviour of dissolved constituents during estuarine mixing. Estuarine Chemistry, editors J.D. Burton and P.S. Liss, Academic Press, New York, 93-130. Loder, T.C., and Reichard R.P., 1981.  The dynamics of conservative mixing in estuaries. Estuaries 4, 64-69. Lohan, M.C., and Bruland, K.W., 2006. Importance of vertical mixing for additional source of nitrate and iron to surface waters of the Columbia River plume: Implications for biology. Marine Chemistry 98, 260 - 273. Lohan, M., Statham, P., Crawford, D., 2002. Total dissolved zinc in the upper water column of the subarctic North East Pacific. Deep-Sea Research Part Ii-Topical Studies in Oceanography. 49, 5793-5808. Lupton, J., 1998. Hydrothermal helium plumes in the Pacific Ocean. Journal of Geophysical Research-Oceans. 103, 15853-15868. Macdonald, R.W. and Carmack, E.C., 1993. Tritium and radiocarbon dating of Canada Basin deep waters. Science. 259, 103-104. Macdonald, R.W. and Carmack, E.C., 1991. Age of Canada Basin deep waters: A way to estimate primary production for the Arctic Ocean. Science. 254, 1348-1350.  Macfadyen, A., Hickey, B.M., Cochlan, W.P., 2008. Influences of the Juan de Fuca eddy on circulation, nutrients, and phytoplankton production in the norther California Current systen. Journal of Geophysical Research. 113, C08008.  MacFadyen, A., Hickey, B., Foreman, M., 2005. Transport of surface waters from the Juan de Fuca eddy region to the Washington coast. Continental Shelf Research. 25, 2008-2021.   182 Mahowald, N., Baker, A., Bergametti, G., Brooks, N., Duce, R., Jickells, T., Kubilay, N., Prospero, J., Tegen, I., 2005. Atmospheric global dust cycle and iron inputs to the ocean. Global Biogeochemical Cycles. 19, GB4025. Maldonado, M. T. and Price, N.M., 1999. Utilization of Fe bound to strong organic ligands by plankton communities in the subarctic Pacific Ocean. Deep-Sea Research II. 46, 2447-2473.   Martin, J. H., Gordon, R. M., Fitzwater, S. E., 1991. The case for iron. Limnology and Oceanography. 36, 1793-1802. Martin, J. and Fitzwater, S., 1988. Iron-deficiency limits phytoplankton growth in the northeast Pacific subarctic. Nature. 331, 341-343. Martin, J., Gordon, R., Fitzwater, S., Broenkow, W., 1989. Vertex - phytoplankton iron studies in the Gulf of Alaska. Deep-Sea Research Part A-Oceanographic Research Papers. 36, 649-&. Martin, J. H., 1990. Glacial-interglacial CO2 change: the iron hypothesis. Paleoceanography. 5, 1-13. McAlister, J. and Orians, K., 2012. Calculation of river-seawater endmembers and differential trace metal scavenging in the Columbia River plume. Estuarine Coastal and Shelf Science. 99, 31-41. McGuire, A. D., Anderson, L. G., Christensen, T. R., Dallimore, S., Guo, L., Hayes, D. J., Heimann, M., Lorenson, T. D., Macdonald, R. W., Roulet, N., 2009. Sensitivity of the carbon cycle in the Arctic to climate change. Ecological Monographs. 79, 523-555. Mckelvey, B. A. and Orians, K. J., 1993. Dissolved zirconium in the north Pacific Ocean. Geochimica Et Cosmochimica Acta. 57, 3801-3805. McKelvey, B. A., 1994. The marine geochemistry of zirconium and hafnium. PhD Thesis, University of British Columbia.  McKelvey, B.A. and Orians, K.J., 1998. The determination of deissolved zirconium and hafnium from seawater using isotope dilution inductively coupled plasma mass spectrometry. Marine Chemistry 60, 245-255. McLaughlin, F. A., Carmack, E. C., Macdonald, R. W., Melling, H., Swift, J. H., Wheeler, P. A., Sherr, B. F., Sherr, E. B., 2004. The joint roles of Pacific and Atlantic-origin waters in the Canada Basin, 1997-1998. Deep-Sea Research Part I-Oceanographic Research Papers. 51, 107-128.   183 McLaughlin, F., Carmack, E., Proshutinsky, A., Krishfield, R. A., Guay, C., Yamamoto-Kawai, M., Jackson, J. M., Williams, B., 2011. The rapid response of the Canada Basin to climate forcing from bellwether To alarm bells. Oceanography. 24, 146-159. McPhee, M.G., Proshutinsky, A., Morison, J.H., Steele, M., Alkire, M.B., 2009. Rapid change in freshwater content of the Arctic Ocean. Geophysical Research Letters. 36, L10602.  Measures, C. I. and Edmond, J. M., 1992. The distribution of aluminum in the Greenland Sea and its relationship to ventilation processes. Journal of Geophysical Research-Oceans. 97, 17787-17800. Measures, C., 1999. The role of entrained sediments in sea ice in the distribution of aluminum and iron in the surface waters of the Arctic Ocean. Marine Chemistry. 68, 59-70. Measures, C. I. and Vink, S., 2000. On the use of dissolved aluminum in surface waters to estimate dust deposition to the ocean. Global Biogeochemical Cycles. 14, 317-327. Measures, C., Edmond, J., Jickells, T., 1986. Aluminum in the northwest Atlantic. Geochimica Et Cosmochimica Acta. 50, 1423-1429. Melancon, J., Levasseur, M., Lizotte, M., Delmelle, P., Cullen, J., Hamme, R. C., Pena, A., Simpson, K. G., Scarratt, M., Tremblay, J., Zhou, J., Johnson, K., Sutherland, N., Arychuk, M., Nemcek, N., Robert, M., 2014. Early response of the northeast subarctic Pacific plankton assemblage to volcanic ash fertilization. Limnology and Oceanography. 59, 55-67. Melling, H., Francois, R., Myers, P. G., Perrie, W., Rochon, A., Taylor, R. L., 2012. The Arctic Ocean-a Canadian perspective from IPY. Climatic Change. 115, 89-113. Meskhidze, N., Chameides, W. L., Nenes, A., 2005. Dust and pollution: A recipe for enhanced ocean fertilization? Journal of Geophysical Research-Atmospheres. 110, D03301. Middag, R., de Baar, H. J. W., Laan, P., Bakker, K., 2009. Dissolved aluminium and the silicon cycle in the Arctic Ocean. Marine Chemistry. 115, 176-195. Middag, R., de Baar, H. J. W., Laan, P., Klunder, M. B., 2011. Fluvial and hydrothermal input of manganese into the Arctic Ocean. Geochimica Et Cosmochimica Acta. 75, 2393-2408. Montoya, J.P., Holl, C.M., Zehr, J.P., Hansen, A., Villareal, T.A., Capone, D.G., 2004. High rates of N2 fixation by unicellular diazotrophs in the oligotrophic Pacific Ocean. Nature. 430, 1027-1030. Moore, R. M., 1989. The relationship between distributions of dissolved cadmium, iron and aluminium and hydrography in the central Arctic Ocean; Selected LOREX contributions; limited   184 facsimile edition of publications resulting from the 1979 Lomonosov Ridge Experiment with a summary of scientific results. Open-File Report - Geological Survey of Canada. 2009, 251-262. Moore, W., 1996. Large groundwater inputs to coastal waters revealed by Ra-226 enrichments. Nature. 380, 612-614. Moran, S. B. and Moore, R. M., 1988. Temporal variations in dissolved and particulate aluminum during a spring bloom. Estuarine Coastal and Shelf Science. 27, 205-215. Morel, F. and Price, N., 2003. The biogeochemical cycles of trace metals in the oceans. Science. 300, 944-947. Nagaoka, D., Shigemitsu, M., Minagawa, M., Noriki, S., 2010. Anthropogenic Pb in settling particulate matter in the northwestern Pacific examined using stable isotopes of Pb. Journal of Oceanography. 66, 117-132. Nelson, Y., Lion, L., Shuler, M., Ghiorse, W., 1999. Lead binding to metal oxide and organic phases of natural aquatic biofilms. Limnology and Oceanography. 44, 1715-1729. Newman, K. and Georg, R. B., 2012. The measurement of Pb isotope ratios in sub-ng quantities by fast scanning single collector sector field-ICP-MS. Chemical Geology. 304, 151-157. Nishioka, J., Takeda, S., Wong, C., Johnson, W., 2001. Size-fractionated iron concentrations in the northeast Pacific Ocean: distribution of soluble and small colloidal iron. Marine Chemistry. 74, 157-179. Nozaki, Y. and Alibo, D. S., 2003. Importance of vertical geochemical processes in controlling the oceanic profiles of dissolved rare earth elements in the northeastern Indian Ocean. Earth and Planetary Science Letters. 205, 155-172. Officer, C.B., 1979. Discussion of the behaviour of nonconservative dissolved constituents in estuaries. Estuarine and Coastal Marine Science 9, 91-94. Orians, K.J. and Boyle, E.A., 1993. Determination of picomolar concentrations of titanium, gallium and indium in sea water by inductviely coupled plasma mass spectrometry following an 8-hydroxyquinoline chelating resin preconcentration. Analytica Chimica Acta 282, 63-74. Orians, K. J. and Bruland, K. W., 1988a. Dissolved gallium in the open ocean. Nature. 332, 717-719. Orians, K. J. and Bruland, K. W., 1988b. The marine geochemistry of dissolved gallium - a comparison with dissolved aluminum. Geochimica Et Cosmochimica Acta. 52, 2955-2962.   185 Orians, K. J. and Bruland, K. W., 1986. The biogeochemistry of aluminum in the Pacific-Ocean. Earth and Planetary Science Letters. 78, 397-410. Orians, K. J. and Bruland, K. W., 1985. Dissolved aluminum in the central north Pacific. Nature. 316, 427-429. Orians, K.J. and Merrin, C.L., 2001. Refractory metals. Encyclopedia of Ocean Sciences Vol 4, editors Thorpe, S.A. and Turekain, K.K., Academic Press, San Diego, 2387-2399.  Outridge, P. M., Hermanson, M. H., Lockhart, W. L., 2002. Regional variations in atmospheric deposition and sources of anthropogenic lead in lake sediments across the Canadian Arctic. Geochimica Et Cosmochimica Acta. 66, 3521-3531. Pabi, S., van Dijken, G. L., Arrigo, K. R., 2008. Primary production in the Arctic Ocean, 1998-2006. Journal of Geophysical Research-Oceans. 113, C08005. Pelland, N.A., Eriksen, C.C., Lee, C.M., 2013. Subthermocline eddies over the Washington continental slope as observed by seagliders, 2003-2009. Journal of Physical Oceanography. 43, 2025-2053.  Pena, M. A. and Varela, D. E., 2007. Seasonal and interannual variability in phytoplankton and nutrient dynamics along Line P in the NE subarctic Pacific. Progress in Oceanography. 75, 200-222. Peterson, B.J., McClelland, J., Curry, R., Holmes, R.M., Walsh, J.E., Aagaard, K., 2006. Trajectory shifts in the Arctic and subarctic Freshwater Cycle. Science. 313, 1061-1066.  Pierce, S., Smith, R., Kosro, P., Barth, J., Wilson, C., 2000. Continuity of the poleward undercurrent along the eastern boundary of the mid-latitude north Pacific. Deep-Sea Research Part Ii-Topical Studies in Oceanography. 47, 811-829. Posacka, A. M.; Semeniuk, D. S.; Maldonado, M. T.; Cullen, J. T.; Orians, K.; 2012, Distribution of the total dissolved copper along the Line P transect of the northeast subarctic Pacific Ocean.  Poster Presentation, Ocean Sciences Conference.   Preciado, H. F., Li, L. Y., Weis, D., 2007. Investigation of past and present multi-metal input along two highways of British Columbia, Canada, using lead isotopic signatures. Water Air and Soil Pollution. 184, 127-139. Proshutinsky, A., Bourke, R.H., McLaughlin, F.A., 2002. The role of the Beaufort Gyre in Arctic climate variability: Seasonal to decadal climate scales. Geophysical Research Letters. 29, L015847.   186 Rattray, M., Officer, C.B., 1979. Distribution of a non-conservative constituent in an estuary with application to the numberical simulation of dissolved solica in the San Francisco Bay. Estuarine and Coastal Marine Science 8, 489-494. Ross, A.R., 2014. The Line P Iron Program: Exploring the Biogeochemistry of trace elements and isotopes in the north-east Pacific. Poster Presentations, Ocean Sciences Conference.  Rudels, B., Jones, E.P., Schauer, U., Eriksson, P., 2004. Atlantic sources of the Arctic Ocean surface and halocline waters. Polar Research. 23, 181-208.  Ruth, C., Well, R., Roether, W., 2000. Primordial He-3 in South Atlantic deep waters from sources on the Mid-Atlantic Ridge. Deep-Sea Research Part I-Oceanographic Research Papers. 47, 1059-1075. Saito, M. and Schneider, D., 2006. Examination of precipitation chemistry and improvements in precision using the Mg(OH)(2) preconcentration inductively coupled plasma mass spectrometry (ICP-MS) method for high-throughput analysis of open-ocean Fe and Mn in seawater. Analytica Chimica Acta. 565, 222-233. Saito, M., Sigman, D., Morel, F., 2003. The bioinorganic chemistry of the ancient ocean: the co-evolution of cyanobacterial metal requirements and biogeochemical cycles at the Archean-Proterozoic boundary? Inorganica Chimica Acta. 356, 308-318. Sanudo-Wilhelmy, S. A. and Flegal, A. R., 1994. Temporal variations in lead concentrations and isotopic composition in the southern California Bight. Geochimica Et Cosmochimica Acta. 58, 3315-3320. Schallenberb, C., Davidson, A. B., Cullen, J.T., 2014. Iron(II) variability in the northeast subarctic Pacific Ocean. Poster Presentation, Ocean Sciences Conference.  Schaule, B. and Patterson, C., 1981. Lead concentrations in the northeast Pacific - Evidence for global anthropogenic perturbations. Earth and Planetary Science Letters. 54, 97-116. Schwing, F. B., Murphree, T., deWitt, L., Green, P. M., 2002. The evolution of oceanic and atmospheric anomalies in the northeast Pacific during the El Niño and La Nina events of 1995-2001. Progress in Oceanography. 54, 459-491. Semeniuk, D. M., Cullen, J. T., Johnson, W. K., Gagnon, K., Ruth, T. J., Maldonado, M. T., 2009. Plankton copper requirements and uptake in the subarctic Northeast Pacific Ocean. Deep-Sea Research Part I-Oceanographic Research Papers. 56, 1130-1142. Semeniuk, D. M., 2014. Copper nutrition and transport mechanisms in plankton communities in the northeast Pacific Ocean. PhD Thesis, University of British Columbia.    187 Sherrell, R. M. and Boyle, E. A., 1992. Isotopic equilibration between dissolved and suspended particulate lead in the Atlantic-Ocean - Evidence from Pb-210 and stable Pb isotopes. Journal of Geophysical Research-Oceans. 97, 11257-11268. Shiel, A. E., Weis, D., Orians, K. J., 2012. Tracing cadmium, zinc and lead sources in bivalves from the coasts of western Canada and the USA using isotopes. Geochimica Et Cosmochimica Acta. 76, 175-190. Shiel, A. E., Weis, D., Orians, K. J., 2010. Evaluation of zinc, cadmium and lead isotope fractionation during smelting and refining. Science of the Total Environment. 408, 2357-2368. Shiller, A. M., 1998. Dissolved gallium in the Atlantic Ocean. Marine Chemistry. 61, 87-99. Shiller, A. M., 1988. Enrichment of dissolved gallium relative to aluminum in natural-waters. Geochimica Et Cosmochimica Acta. 52, 1879-1882. Shiller, A. M. and Boyle, E. A., 1987. Variability of dissolved trace-metals in the Mississippi River. Geochimica Et Cosmochimica Acta. 51, 3273-3277. Shiller, A.M., 1996. The effect of recycling traps and upwelling on estuarine chemical flux estimates. Geochimica et Cosmpchimica Acta 60, 3177-3185.  Shiller, A. M. and Frilot, D. M., 1996. The geochemistry of gallium relative to aluminum in Californian streams. Geochimica Et Cosmochimica Acta. 60, 1323-1328. Shiller, A. M. and Bairamadgi, G. R., 2006. Dissolved gallium in the northwest Pacific and the south and central Atlantic Oceans: Implications for aeolian Fe input and a reconsideration of profiles. Geochemistry Geophysics Geosystems. 7, Q08M09. Shimada, K., Kamoshida, T., Itoh, M., Nishino, S., Carmack, E., McLaughlin, F., Zimmermann, S., Proshutinsky, A., 2006. Pacific Ocean inflow: Influence on catastrophic reduction of sea ice cover in the Arctic Ocean. Geophysical Research Letters. 33, L08605.  Shimada, K., Itoh, M., Nishino, S., McLaughlin, F., Carmack, E., Proshutinsky, A., 2005. Halocline structure in the Canada Basin of the Arctic Ocean. Geophysical Research Letters. 32, L03605. Sholkovitz, E. R., 1978. Flocculation of dissolved Fe, Mn, Al, Cu, Ni, Co and Cd during estuarine mixing. Earth and Planetary Science Letters. 41, 77-86. Shotyk, W., Zheng, J. C., Krachler, M., Zdanowicz, C., Koerner, R., Fisher, D., 2005. Predominance of industrial Pb in recent snow (1994-2004) and ice (1842-1996) from Devon Island, Arctic Canada. Geophysical Research Letters. 32, L21814.   188 Siddall, M., Khatiwala, S., van de Flierdt, T., Jones, K., Goldstein, S. L., Hemming, S., Anderson, R. F., 2008. Towards explaining the Nd paradox using reversible scavenging in an ocean general circulation model. Earth and Planetary Science Letters. 274, 448-461. Sim, N. and Orians, K.J., 2014. Dissolved manganese in the northeast Pacific : Line P 2011 – 2013. Poster Presentation, Ocean Sciences Conference.  Simonetti, A., Gariepy, C., Carignan, J., 2003. Tracing sources of atmospheric pollution in Western Canada using the Pb isotopic composition and heavy metal abundances of epiphytic lichens. Atmospheric Environment. 37, 2853-2865. Simpson, K.G., Tremblay, J.E., Gratton, Y., Price, N.M., 2008. An annual study of inorganic and organic nitrogen and phosphorus and silicic acid in the southeastern Beaufort Sea. Journal of Geophysical Research. 113, C07016. Sohrin, Y., Urushihara, S., Nakatsuka, S., Kono, T., Higo, E., Minami, T., Norisuye, K., Umetani, S., 2008. Multielemental determination of GEOTRACES key trace metals in seawater by ICPMS after preconcentration using an ethylenediaminetriacetic acid chelating resin. Analytical Chemistry. 80, 6267-6273. Soto-Jimenez, M. F. and Flegal, A. R., 2009. Origin of lead in the Gulf of California Ecoregion using stable isotope analysis. Journal of Geochemical Exploration. 101, 209-217. Steele, M., Morison, J., Ermold, W., Rigor, I., Ortmeyer, M., Shimada, K., 2004. Circulation of summer Pacific halocline water in the Arctic Ocean. Journal of Geophysical Research-Oceans. 109, C02027. Stroeve, J., Holland, M. M., Meier, W., Scambos, T., Serreze, M., 2007. Arctic sea ice decline: Faster than forecast. Geophysical Research Letters. 34, L09501. Stukas, V., Wong, C. S., Johnson, W. K., 1999. Sub-part per trillion levels of lead and isotopic profiles in a fjord, using an ultra-clean pumping system. Marine Chemistry. 68, 133-143. Sutherland, D. A., Pickart, R. S., Jones, E. P., Azetsu-Scott, K., Eert, A. J., Olafsson, J., 2009. Freshwater composition of the waters off southeast Greenland and their link to the Arctic Ocean. Journal of Geophysical Research-Oceans. 114, C05020. Talley, L., 1993. Distribution and formation of North Pacific Intermediate Water. Journal of Physical Oceanography. 23, 517-537. Taylor, R. L., Semeniuk, D. M., Payne, C. D., Zhou, J., Tremblay, J., Cullen, J. T., Maldonado, M. T., 2013. Colimitation by light, nitrate, and iron in the Beaufort Sea in late summer. Journal of Geophysical Research-Oceans. 118, 3260-3277.   189 Tebo, B., Bargar, J., Clement, B., Dick, G., Murray, K., Parker, D., Verity, R., Webb, S., 2004. Biogenic manganese oxides: Properties and mechanisms of formation. Annual Review of Earth and Planetary Sciences. 32, 287-328. Timmermans, M.-L. and Garrett, C. 2006. Evolution of the deep water in the Canadian Basin in the Arctic Ocean. Journal of Physical Oceanography. 36, 866-874.  Timmermans, M.-L., Garrett, C., Carmack, E. 2003. The thermohaline structure and evolution of the deep waters in the Canada Basin, Arctic Ocean. Deep-Sea Research I. 50, 1305-1321.  Torres-Valdes, S., Tsubouchi, T., Bacon, S., Naveira-Garabato, A.C., Sanders, R., McLaughlin, F.A., Petrie, B., Gerhard, K., Azetsu-Scott, K., Whitledge, T.E. 2013. Export of nutrients from the Arctic Ocean. Journal of Geophysical Research: Oceans. 118, 1625-1644.  Tremblay, J. E., Michel, C., Hobson, K. A., Gosselin, M., Price, N. M., 2006. Bloom dynamics in early opening waters of the Arctic Ocean. Limnology and Oceanography. 51, 900-912. Tremblay, J., Simpson, K. G., Martin, J., Miller, L., Gratton, Y., Barber, D., Price, N. M., 2008. Vertical stability and the annual dynamics of nutrients and chlorophyll fluorescence in the coastal, southeast Beaufort Sea. Journal of Geophysical Research-Oceans. 113, C07S90. Turner, D. R., Whitfield, M., Dickson, A. G., 1981. The equilibrium speciation of dissolved components in fesh-water and seawater at 25-Degrees-C and 1 Atm Pressure. Geochimica Et Cosmochimica Acta. 45, 855-881. Veron, A. J., Church, T. M., Flegal, A. R., Patterson, C. C., Erel, Y., 1993. Response of lead cycling in the surface Sargasso Sea to changes in tropospheric input. Journal of Geophysical Research-Oceans. 98, 18269-18276. Veron, A. J., Church, T. M., Rivera-Duarte, I., Flegal, A. R., 1999. Stable lead isotopic ratios trace thermohaline circulation in the subarctic North Atlantic. Deep-Sea Research Part Ii-Topical Studies in Oceanography. 46, 919-935. Viers, J., Dupre, B., Polve, M., Schott, J., Dandurand, J-L., Braun, J-J., 1997. Chemical weathering in the drainage basin of a tropical watershed (Nsimi-Zoetele site, Cameroon): comparison between organic-poor and organic-rich waters. Chemical Geology 140, 181-206. Wang, W., Liu, X. D., Zhao, L. W., Guo, D. F., Tian, X. D., Adams, F., 2006. Effectiveness of leaded petrol phase-out in Tianjin, China based on the aerosol lead concentration and isotope abundance ratio. Science of the Total Environment. 364, 175-187. Ware, D.M. and Thomson, R.E., 2005. Bottom-up ecosystem trophic dynamics determine fish production in the northeast Pacific. Science. 308, 1280-1284.    190 Weiss, D., Boyle, E., Chavagnac, V., Herwegh, M., Wu, J., 2000. Determination of lead isotope ratios in seawater by quadrupole inductively coupled plasma mass spectrometry after Mg(OH)(2) co-precipitation. Spectrochimica Acta Part B-Atomic Spectroscopy. 55, 363-374. Weiss, D., Boyle, E., Wu, J., Chavagnac, V., Michel, A., Reuer, M., 2003. Spatial and temporal evolution of lead isotope ratios in the North Atlantic Ocean between 1981 and 1989. Journal of Geophysical Research-Oceans. 108, 3306. Weiss, D., Shotyk, W., Appleby, P., Kramers, I., Cheburkin, A., 1999a. Atmospheric Ph deposition since the industrial revolution recorded by five Swiss peat profiles: Enrichment factors, fluxes, isotopic composition, and sources. Environmental Science & Technology. 33, 1340-1352. Weiss, D., Shotyk, W., Kempf, O., 1999b. Archives of atmospheric lead pollution. Naturwissenschaften. 86, 262-275. Whitney, F. and Robert, M., 2002. Structure of Haida eddies and their transport of nutrient from coastal margins into the NE Pacific Ocean. Journal of Oceanography. 58, 715-723. Whitney, F. and Freeland, H., 1999. Variability in upper-ocean water properties in the NE Pacific Ocean. Deep-Sea Research Part Ii-Topical Studies in Oceanography. 46, 2351-2370. Whitney, F. A., Crawford, W. R., Harrison, P. J., 2005. Physical processes that enhance nutrient transport and primary productivity in the coastal and open ocean of the subarctic NE Pacific. Deep-Sea Research II. 52, 681-706.  Whitney, F. A., Freeland, H. J., Robert, M., 2007. Persistently declining oxygen levels in the interior waters of the eastern subarctic Pacific. Progress in Oceanography. 75, 179-199. Wijffels, S., Schmitt, R., Bryden, H., Stigebrandt, A., 1992. Transport of Fresh-Water by the Oceans. Journal of Physical Oceanography. 22, 155-162. Wong, C. S., Waser, N. A. D., Nojiri, Y., Johnson, W. K., Whitney, F. A., Page, J. S. C., Zeng, J., 2002. Seasonal and interannual variability in the distribution of surface nutrients and dissolved inorganic carbon in the northern North Pacific: Influence of El Niño. Journal of Oceanography. 58, 227-243. Wong, C. S., Xie, L., Hsieh, W. W., 2007. Variations in nutrients, carbon and other hydrographic parameters related to the 1976/77 and 1988/89 regime shifts in the sub-arctic Northeast Pacific. Progress in Oceanography. 75, 326-342. Wong, C., Whitney, F., Crawford, D., Iseki, K., Matear, R., Johnson, W., Page, J., 1999. Seasonal and interannual variability in particle fluxes of carbon, nitrogen and silicon from time series of sediment traps at Ocean Station P, 1982-1993: relationship to changes in subarctic   191 primary productivity. Deep-Sea Research Part Ii-Topical Studies in Oceanography. 46, 2735-2760. Woodgate, R.A., Aagaard, K., Weingartner, T.J., 2006. Interannual changes in the Bering Strait fluxes of volume, heat, and freshwater between 1991 and 2004. Geophysical Research Letters. 33, L15609.  Woodgate, R. A., Aagaard, K., Swift, J. H., Swift, J.H., Falkner, K. K., Smethie, W. M., Jr. 2005. Pacific ventilation of the Arctic Ocean's lower halocline by upwelling and diapycnal mixing over the continental margin. Geophysical Research Letters. 32, L18609.   Woodgate, R. A. and Aagaard, K., 2005. Revising the Bering Strait freshwater flux into the Arctic Ocean. Geophysical Research Letters. 32, L02602. Wu, J. and Boyle, E., 1997a. Lead in the western North Atlantic Ocean: Completed response to leaded gasoline phaseout. Geochimica Et Cosmochimica Acta. 61, 3279-3283. Wu, J. and Boyle, E., 1997b. Low blank preconcentration technique for the determination of lead, copper, and cadmium in small-volume seawater samples by isotope dilution ICPMS. Analytical Chemistry. 69, 2464-2470. Wu, J., Rember, R., Jin, M., Boyle, E. A., Flegal, A. R., 2010. Isotopic evidence for the source of lead in the North Pacific abyssal water. Geochimica Et Cosmochimica Acta. 74, 4629-4638. Yamamoto-Kawai, M., Carmack, E., McLaughlin, F., 2006. Nitrogen balance and Arctic throughflow. Nature. 443, 43-43. Yoon, W., Kim, S., Han, K., 2001. Morphology and sinking velocities of fecal pellets of copepod, molluscan, euphausiid, and salp taxa in the northeastern tropical Atlantic. Marine Biology. 139, 923-928. Zhang, J., Lindsay, R., Steele, M., Schweiger, A., 2008. What drove the dramatic retreat of arctic sea ice during summer 2007? Geophysical Research Letters. 35, L11505.    

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:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0166278/manifest

Comment

Related Items