UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Dissolved aluminum in the northeast Pacific and the western Arctic Ocean Cain, Amy Nicole 2014

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

Item Metadata

Download

Media
24-ubc_2014_september_cain_amy.pdf [ 1.77MB ]
Metadata
JSON: 24-1.0167483.json
JSON-LD: 24-1.0167483-ld.json
RDF/XML (Pretty): 24-1.0167483-rdf.xml
RDF/JSON: 24-1.0167483-rdf.json
Turtle: 24-1.0167483-turtle.txt
N-Triples: 24-1.0167483-rdf-ntriples.txt
Original Record: 24-1.0167483-source.json
Full Text
24-1.0167483-fulltext.txt
Citation
24-1.0167483.ris

Full Text

Dissolved Aluminum in the Northeast Pacific and the Western Arctic Ocean by Amy Nicole Cain  B.Sc., Southwestern Oklahoma State University, 2009  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Chemistry)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)   June 2014  © Amy Nicole Cain, 2014   ii  Abstract Aluminum dynamics in the ocean are unique given its rapid removal from the surface through adsorption onto sinking particles and its high abundance in the earth's crust.  We optimized a flow injection analysis method to allow detection of the extremely low concentrations of dissolved aluminum found in the study region. Samples from the northeast Pacific have the lowest concentrations of dissolved Al found anywhere in the world’s ocean.   We present dissolved aluminum data from a transect along Line P in the northeast Pacific in 2010 and 2011.  Stations extend from near coastal waters, going through areas of potential eddy influence, and terminate at the historic Ocean Station Papa. Open ocean data shows concentrations an order of magnitude lower than extensive work performed in the Atlantic but overall similar trends with higher aluminum in the surface layer. The concentrations can go as low as 0.06 nmol/kg at depth and 0.36 nmol/kg in the surface.  Coastal stations show elevated aluminum levels relative to the open ocean, and eddy influence inputs high aluminum concentrations from coastal shelf water.  In the western Arctic, we present a transect from the 2009 Canadian GEOTRACES program. Here we see concentrations that increase with depth, from below 1.0 nmol/kg at the surface up to 11.6 nmol/kg at depth. This is inconsistent with previous measurements in the region, which showed an increase in the surface, proposed to be due to sea ice melt supplying dissolved aluminum to the surface. The high concentration of dissolved aluminum at depth indicate a bottom source of dissolved aluminum in this region. This is also observed in the north Atlantic iii  near Labrador.   Lastly, we present dust input data from both of these regions and compare them to previous modelled data, showing that models in the Pacific over-represent the amount of dust input. Arctic data is relatively less well known for dust.   iv  Preface This research work is done in collaboration with the Department of Fisheries and Oceans of Canada and included work done on research vessels deployed by the Government of Canada. I was responsible for all aluminum analyses done in this study; my lab mates, other trace metal oceanographers from the University of Victoria, and scientists from the Department of Fisheries and Oceans helped with the collection of the samples. The equipment used for the research is co-owned by our trace metal lab as well as one at the University of Victoria. Research work was funded by NSERC.  2010 and 2011 Line P cruises were on the John P. Tully and the 2009 Arctic cruise was on the CCGS Amundsen.  v  Table of Contents Abstract .......................................................................................................................................... ii Preface ........................................................................................................................................... iv Table of Contents ...........................................................................................................................v List of Tables ................................................................................................................................. ix List of Figures .................................................................................................................................x Acknowledgements ..................................................................................................................... xii Dedication ................................................................................................................................... xiii Chapter 1: Introduction ................................................................................................................1     1.1 Main Goals of Research .......................................................................................................1     1.2 General Geochemistry and Background Information ......................................................2        1.2.1 Trace Metals in the Ocean .............................................................................................2        1.2.2 The Major Profile Types ................................................................................................3           1.2.2.1 Conservative Profiles ...............................................................................................4           1.2.2.2 Nutrient Profiles .......................................................................................................5           1.2.2.3 Scavenged Profiles ...................................................................................................6           1.2.2.4 Iron - A Type of Hybird Profile  ..............................................................................8     1.3 The Biogeochemistry of Aluminum ....................................................................................8        1.3.1 Sources of Aluminum in the Ocean ..............................................................................9        1.3.2 Alumimum Concentrations around the World ......................................................... 11        1.3.3 Silicon and Aluminum .................................................................................................13     1.4 Method of Analysis .............................................................................................................13     1.5 Research Objectives ...........................................................................................................14 vi  Chapter 2: Method for Analyzing Dissolved Aluminum in Seawater .....................................15     2.1 The Method .........................................................................................................................15     2.2 Analytical Procedures ........................................................................................................18        2.2.1 Blanks ............................................................................................................................18        2.2.2 Challenges .....................................................................................................................19     2.3 Sampling Methods .............................................................................................................20     2.4 Conclusion ..........................................................................................................................22 Chapter 3: Measurements of Dissolved Aluminum in the Northeastern Pacific Ocean .......23     3.1 The Subarctic Northeastern Pacific .................................................................................23        3.1.1. Information about the Study Area ............................................................................23        3.1.2 Eddies in the Gulf of Alaska ........................................................................................25        3.1.3 El Niño-Southern Oscillation ......................................................................................26     3.2 Aluminum in the Pacific -- A Region of Low Dust Input................................................26        3.2.1 Previous Data for Dissolved Aluminum in the Pacific ..............................................27          3.2.1.1 Aluminum in the Eastern Subtropical Gyre .........................................................27          3.2.1.2 Aluminum in the Subtropical and Subarctic Northwestern Pacific ...................28          3.2.1.3 Aluminum in the Northeastern Subarctic Pacifc .................................................29          3.2.1.4 Iron in the Northeastern Subarctic Pacific ...........................................................29        3.2.2 Sources of Aluminum (and Iron) in the Northeast Subarctic Pacific ......................30     3.3 Results -- Distribution of Dissolved Aluminum in the Northeastern Pacific Ocean ....32        3.3.1 Sampling Protocol ........................................................................................................32        3.3.2 Results for Dissolved Aluminum along Line P ..........................................................33        3.3.3 Vertical Profiles for Dissolved Aluminum along Line P ...........................................34 vii      3.4 Discussion............................................................................................................................36        3.4.1 P26 -- Open Ocean .......................................................................................................36        3.4.2. P12 and P16 -- Offshore -- Coastal Input and Eddies .............................................36        3.4.3 P4 -- Coastal Input .......................................................................................................39     3.5 Closing Remarks about the Northeast Pacific.................................................................41 Chapter 4: Measurements of Dissolved Aluminum in the Western Arctic Ocean .................43     4.1 The Western Arctic Ocean.................................................................................................43        4.1.1 Information about the Study Area .............................................................................43        4.1.2 The Mackenzie River ...................................................................................................45        4.1.3 Seasonal Glacial Melting .............................................................................................46     4.2 Aluminum in the Arctic -- A Polar Basin .........................................................................47        4.2.1 Aluminum in the Arctic Ocean ...................................................................................48          4.2.1.1 2007 Study of Aluminum in the Eurasian Arctic Ocean .....................................48          4.2.1.2 2007 Study of Aluminum in the Canadian Arctic Ocean ....................................49          4.2.1.3 Iron in the Western Arctic Ocean ..........................................................................50        4.2.2. Sources of Aluminum (and Iron) in the Western Arctic Ocean ..............................50          4.2.2.1 Dirty Sea Ice or Shelf Input? .................................................................................51          4.2.2.2 Other Inputs ............................................................................................................52     4.3 Results -- Distribution of Dissolved Aluminum in the Western Arctic Ocean ..............53        4.3.1 Sampling Protocol ........................................................................................................53        4.3.2 Results for Dissolved Aluminum in the Western Arctic Ocean................................54        4.3.3 Vertical Profiles for Dissolved Aluminum in the Western Arctic Ocean .................56     4.4 Discussion............................................................................................................................58 viii         4.4.1 Beyond the Shelf ...........................................................................................................58        4.4.2 S4 Sampling -- Near the Shelf .....................................................................................60     4.5 Closing Remarks about the Western Arctic ....................................................................62 Chapter 5: The Role of Aluminum as a Tracer of Dust Input .................................................64     5.1 The Origins of Studying Dust Deposition ........................................................................64     5.2 Why Study Aeolian Input?  ...............................................................................................64        5.2.1 Iron and Dust Deposition ............................................................................................65        5.2.2 Aeolian Input's Influence on the Past and Present Climate .....................................67     5.3 Aluminum's Role in Determining Aeolian Input ............................................................69        5.3.1 Aluminum as a Tracer of Aeolian Input .....................................................................70        5.3.2 Quantifying Aeolian Input ..........................................................................................71        5.3.3 Models Using Dissolved Aluminum as a Tracer of Aeolian Input ...........................72     5.4 Calculating Dust Deposition in the North Pacific and Western Arctic  ........................74        5.4.1 Regional Dust Input Based on Models .......................................................................74        5.4.2 Limitations of the Dust Model ....................................................................................76     5.5 Closing Remarks about Dust Deposition .........................................................................77 Chapter 6: Conclusions ...............................................................................................................79 References .....................................................................................................................................82 Appendix A ...................................................................................................................................96  ix  List of Tables Table 5-1 Dust deposition in the North Pacific based on surface dissolved aluminum concentrations ................................................................................................................................75 Table 5-2: Dust deposition in the Western Arctic based on surface dissolved aluminum concentrations ................................................................................................................................76    x  List of Figures Figure 1-1: Dissolved molybdenum in the North Pacific, 44°40'N, 177°00'W  (Sohrin et al. 1987) ..........................................................................................................................4 Figure 1-2: Vertical distribution of Zn in the Atlantic (47°N, 20°W) and Pacific (32°41'N, 144°59’W) (Bruland 1980, Martin et al. 1993) ...............................................................................6 Figure 1-3: Vertical profile of aluminum in the North Pacific, 155°07'W, 28°15'N (Orians and Bruland 1986) ..................................................................................................................................7 Figure 1-4: The summer mineral deposition as estimated by modelling (Tegen and Fung 1994) ..................................................................................................................10 Figure 2-1: The structure of lumogallion, from TCI America website. The aluminum forms a complex with one N and two O- ....................................................................................................16 Figure 2-2: The FIA system. Adapted from Brown (2008) and Giesbrecht (2013) .......................17 Figure 3-1: The Line P time series (from the UBC PCIGR website) ............................................24 Figure 3-2: Dissolved aluminum concentrations at (a) P26, 50°00'N, 145°00'W and VERTEX VII-T7 (Orians and Bruland 1988) (1987) (b) P16, 49°17' N, 134°40’W .....................................34 Figure 3-3: Comparison between 2010 and 2011 for dissolved Al down to 400 m from (a) P12, 48°58' N, 130°40'W and (b) P16, 49°17' N, 134°40'W .................................................................35 Figure 3-4: Dissolved aluminum concentrations from P4, 48°39' N, 126°40'W. (a) to 400 m and (b) to 2000 m ..................................................................................................................................35 Figure 3-5: Sea surface height measurements at Line P (from NOAA’s website) for 2010 (top) and 2011 (bottom). Scale is cm above sea level ............................................................................38 Figure 3-6: Dissolved aluminum concentrations at P4 in 2011 compared to values off the coast of California (36°06' N, 122°38'W) ...................................................................................................41 xi  Figure 4-1: The location of the stations relative to Canada ...........................................................45 Figure 4-2: Study area of the Eurasian Arctic in 2007 (Roeske et al. 2012) .................................48 Figure 4-3: The stations sampled in this study ..............................................................................54 Figure 4-4: Dissolved aluminum at (a) L1 (71°06’ N, 139°18’ W); (b) L1.1  (72°30’ N, 135°35’ W) (c) Station L1.5 (73°19’ N, 139°23’ W), (d) Station L2 (74°36’ N, 137°07’ W) to 3000 m ...................................................................................................................56 Figure 4-5: Dissolved aluminum at (a) L1 (71°06’ N, 139°18’ W); (b) L1.1  (72°30’ N, 135°35’ W) (c) Station L1.5 (73°19’ N, 139°23’ W), (d) Station L2 (74°36’ N, 137°07’ W) to 1600 m ...................................................................................................................57 Figure 4-6: Dissolved aluminum concentrations at S4, 71°11’ N, 132°56’ W .............................58 Figure 4-7: L1.1 (72°30’ N, 135°35’ W) and KC2000 (71°73’ N, 135°50’ W) (Giesbrecht et al. 2013), down to 600 m ....................................................................................................................59 Figure 4-8: S4 (71°11’ N, 132°56’ W) and KC200 (70°97’ N, 135°20’ W) down to 200 m ........................................................................................................................................................62 Figure 5-1: Modelled dust deposition using dissolved Al, in g·m-2·yr-1 (Han et al. 2008) ............74 Figure 5-2: Modelled residence time of the surface ocean using dissolved Al (Han et al. 2008) .77        . xii  Acknowledgements I thank my research supervisor, Kristin Orians, for her help in writing this thesis as well as her help with my research work. I thank the members of my committee, Roger Francois and David Chen, and the Chair, Alexander Wang, for your participation in my defense. A special mention goes to my labmate and good friend Jason McAlister for his unwavering faith in me and his time and energy spent in helping me solve problems. Also, I thank my other labmates Jeffrey Charters, Nari Sim, and Ania Posacka (and honorary labmate Francis Choi) for being around to bounce ideas off of and for generally being great people. I’d also like to thank Maureen Soon and Milan Coschizza for giving me advice in the lab, Keith Johnson for teaching me how to sample on a boat, Marie Robert for being so helpful on the Line P cruises, and the variety of people on the cruises for making those days away from home a little brighter.  I’d like to thank my parents, my brother, and my grandmother as well as my boyfriend’s family for being supportive through my work, my best friend Erek for having enlightened discussions about marine chemistry and biology with me, and my friend Sherry for all of the fun times we had going out together. xiii  Dedication For David, who makes my life complete1  Chapter 1: Introduction to Aluminum Biogeochemistry and Aluminum Analysis Method in Seawater 1.1 Main Goals of Research One active area of research into future climate is dust input, with the goal of determining if iron brought in by dust into the ocean is changed by changes in wind patterns and biota caused by increased CO2 (Mahowald et al. 2006). Iron has an impact on phytoplankton productivity and limits growth in many parts of the ocean (Boyd et al. 2000, Boyd et al. 1998, Jickells et al. 2005), so dust that brings iron to the ocean is a key part of the carbon cycle. Because dust deposition is poorly quantified even in the present, a long-term investigation of dust deposition would help bolster knowledge about future events. This leads to an interest in investigating dissolved aluminum as a proxy for dust deposition.   We use a flow injection analysis system to analyze dissolved aluminum in two regions of interest. The first is the northeast Pacific Ocean, a place of low dust input, including a large high-nutrient low-chlorophyll (HNLC) part of the subarctic gyre. We also study continental shelf input into the coastal region of the northeast Pacific. The second is the western Arctic Ocean, a region with heavy influence from water masses from both the Pacific and Atlantic Oceans. The western Arctic Ocean has more contact with shelves than other regions due to being a small basin surrounded by land masses. We calculate dust input using surface concentrations in both of these regions.   2  1.2 General Geochemistry and Background Information 1.2.1  Trace Metals in the Ocean Although all naturally occurring elements are found in the ocean, the concentrations of the elements vary immensely, from chloride concentrations of 0.5 M (Wright and Colling 1995) to osmium concentrations of 0.03 pM (Woodhouse et al. 1999). Trace metals exist at low concentrations in the ocean, from nanomolar to sub-picomolar depending on the element and the location (Bruland 1983). Biology plays an important role in trace element distribution in a few ways; iron and zinc (among others) are directly involved in the metabolic process enabling phytoplankton growth, manganese is reduced in low oxygen conditions caused by remineralization (the breakdown of organic matter with oxygen by bacteria for energy), and aluminum and other particle-reactive elements interact with the surfaces of dead phytoplankton, lowering their concentration.   Work in trace metal chemistry became viable in the 1970’s as methods for contamination-free sample collecting and processing were developed (Bruland et al. 1979). Prior to clean method development, contamination from ships and collection bottles dominated the trace metal concentration values reported rather than the seawater’s actual concentrations. Clean handling techniques, such as the use of Kelvar lines and Teflon-lined bottles for collection rather than metal hydrowire and uncoated bottles improved results (Hunter et al. 1996). The determination of optimal bottle types to be used for sample storage has also contributed to more accurate values for sensitive trace metals such as aluminum and titanium (Brown and Bruland 2008), as well as improved bottle washing techniques. Analysis is also done in clean rooms and in laminar flow 3  benches to try to reduce contamination (Nolting and de Jong 1994).   This attention to cleanliness coupled with the evolution of analytical instrumentation has allowed values to be obtained at even sub-nanomolar concentrations. These analytical methods include the method used in this study, molecular fluorescence (Resing and Measures 1994), as well as absorption spectroscopy (Sturgeon et al. 1980), inductively coupled plasma mass spectrometry (Hansen 2005) and chemiluminescence (Zamzow et al. 1998). These methods can be used to accurately assess trace metal concentrations, and are often coupled with a preconcentration step (which also serves to separate the analyte from the seawater matrix) or isotope dilution to achieve the sub-nanomolar concentrations needed for trace metal detection.  1.2.2  The Major Profile Types  The vertical distribution of elements in the ocean can be used to infer the processes which they are involved in. These include surface processes such as atmospheric input, phytoplankton growth, and fluvial input, as well as deeper ocean processes such as remineralization and processes which occur throughout the water column, such as scavenging.  Lateral advection can complicate matters by introducing a horizontal input. Vertical profiles are divided primarily into three types based on steady-state behaviour through the water column: conservative profiles, nutrient-type profiles, and scavenged profiles (Bruland and Lohan 2003). The type of profile expected for an element is largely based on its residence time, the average amount of time it spends in the ocean assuming a steady state. By studying vertical distributions, we can infer sources, internal cycling, and removal mechanisms.  4  1.2.2.1  Conservative Profiles Distributions of conservative elements are controlled only by physical processes and remain essentially constant relative to salinity. These physical processes include evaporation/precipitation, fluvial input, sea spray, and ice melt/formation; some conservative elements can be used in biological processes but the usage is small relative to the total concentration. A high concentration cation such as Na+ is an example of a conservative element, but among trace elements, uranium (U) and molybdenum (Mo) are two examples. These elements have a long residence time due to a slow removal mechanisms. For example, molybdenum's residence time is 8x105 yr (Emerson and Huested 1991). Figure 1-1 shows a nearly constant concentration throughout the water column for Mo.   Figure 1-1: Dissolved molybdenum in the North Pacific, 44°40'N, 177°00'W (Sohrin et al. 1987) 01000200030004000500060000 50 100 150depth(m)Dissolved Mo(nmol/kg)5  1.2.2.2  Nutrient Profiles Nutrient-type elements are depleted at the surface relative to their average water column concentration due to their role in phytoplankton productivity. Macronutrients such as phosphate (“PO4”), nitrate (NO3-), and silicic acid (H4SiO4) are used as building blocks for proteins and other organic molecules and for structural material such as shells. Trace elements such as cadmium and zinc can be important in enzymes and other cofactors (Lohan, Statham & Crawford 2002, Price and Morel 1990), resulting in distributions similar to the major nutrients. Nutrient elements increase in concentration below the surface where phytoplankton are remineralized and remain in the dissolved phase due to lack of abiotic removal mechanisms.   The residence times of nutrient elements are shorter than those of conservative elements due to biological uptake, some of which is transported to the sediments and therefore removed from the oceans. During remineralization at depth, however, most of the nutrients are released into the water column. As water travels in the deep ocean, nutrient concentrations will increase. Older water masses in the North Pacific, therefore, have elevated levels of nutrients relative to younger water masses in the deep Atlantic Ocean. The profile of zinc (Figure 1-2) is an example of this inter-ocean fractionation. These trace metals can be brought from the deep to the surface via upwelling, influencing primary production. 6   Figure 1-2: Vertical distribution of Zn in the Atlantic (47°N, 20°W) and Pacific (32°41'N, 144°59’W) (Bruland 1980, Martin et al. 1993)  1.2.2.3  Scavenged Profiles Scavenged elements have concentrations which are highest at the surface, decrease in the middle of the water column, and sometimes are elevated at the sediment interface. Scavenging refers to the passive adsorption of elements onto particle surfaces, most of which are anionic in the marine environment (Hydes 1979). Adsorption occurs onto biogenic material such as phytoplankton or onto sinking clay minerals.   The rapid adsorption leads to short residence times for scavenged elements. In the surface ocean removal times can vary from a few weeks to a few years, with residence times of a few hundred 05001000150020002500300035004000450050000 5 10depth(m)Dissolved Zn(nmol/kg)7  years in deeper waters where there are fewer particles. The residence times of scavenged elements are shorter than the circulation time of the ocean, so older water masses are depleted in these elements relative to younger waters (Orians and Bruland 1986). Concentrations of scavenged elements are typically controlled by their external inputs.   Aluminum is an example of a scavenged element with external inputs to the surface ocean from atmospheric dust and a potential input to the bottom waters from the sediments. The distribution of dissolved aluminum in the central North Pacific (Figure 1-3) shows elevated concentrations at the surface and at the bottom with low concentrations away from the sources.  Figure 1-3: Vertical profile of aluminum in the North Pacific, 155°07'W, 28°15'N (Orians and Bruland 1986)  01000200030004000500060000 2 4 6depth(m)Dissolved Al(nmol/kg)8  1.2.2.4   Iron – A Type of Hybrid Profile With elements such as iron, the distinction between nutrient and scavenged elements is less obvious because it displays properties of both. It is depleted at the surface due to phytoplankton growth but it is also scavenged onto particles, resulting in a lower residence time. Iron is often efficiently recycled in surface waters due to high demand and low supply in regions of low dust supply.  At times of maximum stratification and low scavenging (due to low productivity), such as in the North Pacific Gyre in the summer, iron and other scavenged trace elements can accumulate in the surface (Bruland, Orians & Cowen 1994). Overall, the distribution reflects a complicated balance between sources, biological uptake, scavenging, and remineralization (Johnson et al. 2003)  High-nutrient low-chlorophyll (HNLC) zones are often characterized by high amounts of the macronutrients “PO4” and NO3- but will still have low amounts of phytoplankton growth, also referred to as primary productivity. Iron deficiency is typically the cause, affecting 20-40% of the ocean (Greene et al. 1994). The iron limitation in the northeast Pacific will be discussed in more detail later.  1.3 The biogeochemistry of aluminum  Aluminum is abundant in the Earth’s crust; at 8.2% of the total, it is the third most common element after oxygen and silicon (Wedepohl 1995), but it is often at nanomolar concentrations in the ocean. Aluminum has a low solubility in seawater, existing as the insoluble Al(OH)30 or the somewhat more soluble Al(OH)4-, with the ratio depending on the pH (Mackin and Aller 1984, Wesolowski 1992). At the pH of seawater, Al(OH)4- will be the dominant form, although there is 9  still significant amounts of Al(OH)30. The concentration of dissolved aluminum is further limited by scavenging in the water column, both passively onto biogenic matter such as opal and organic coatings (Hydes 1989) and actively into siliceous shells (Gehlen et al. 2003). Aluminum is in the +3 oxidation state in the environment in all natural conditions and is monoisotopic as 27Al.  1.3.1  Sources of Aluminum to the Ocean It was originally believed that the primary source of aluminum to the ocean was fluvial because river concentrations are 2-3 orders of magnitude higher than the concentration in the ocean (Hydes and Liss 1977, Stoffyn and Mackenzie 1982). This, however, does not necessarily translate to a transfer of aluminum from rivers to the ocean because of processes occurring in the estuary. Both the changing ionic strength and the changing pH (from slightly acidic in river water to slightly basic in seawater) cause flocculation to occur, removing most of the aluminum via adsorption onto particle surfaces and precipitation (Morris, Howland & Bale 1986, Brown and Bruland 2009). A study on the coast of southern Alaska demonstrates that even 20 km offshore from a glacial river deposit with high dissolved aluminum shows a nearly 100-fold decrease in concentration (Brown, Lippiatt & Bruland 2010). Further offshore will see little influence from fluvial aluminum.  Having excluded riverine input as a significant aluminum source to the open ocean, another mechanism must explain the input of aluminum into the ocean. This leaves the primary path of entry as aeolian dust because dust particles contain large quantities of aluminum. Some of the highest ocean concentrations of aluminum in the world are found off the coast of the Africa near the Sahara Desert (Measures and Vink 2000) and in the Arabian Sea near the Middle Eastern 10  deserts (Tindale and Pease 1999) because of significant dust mobilization leading to increased aeolian input in these regions. In areas both north and south of the large Saharan plume, advection of equatorial Atlantic water increases dissolved aluminum concentrations.   Early estimates of mineral dust flux indicate high dust input in a few specific areas (Tegen and Fung 1994) (Fig 1-4), including off the western coasts of Africa and Australia and in the Arabian Sea. There is low dust input in almost the entire Pacific Ocean; the only regions that see much dust deposition based on this model are places near Asia and off the coast of southern North America. Both the north Pacific and the south Pacific are very low in dust input. Other later studies show a plume of dust from China and small inputs of dust from North America into the north Pacific, although still lower than most places in the world (Han et al. 2008).           Figure 1-4: The summer mineral deposition as estimated by modelling (Tegen and Fung 1994)  11  Another potential source of aluminum is from sediments, both on the continental shelf and in the deep ocean. A benthic flux due to pore water diffusion could provide a source of dissolved aluminum. Another explanation is passive reversible exchange with particles at the surface of the sediments (Moore and Millward 1984). Either of these mechanisms could provide dissolved aluminum.   The shelf break is at about 200 m; after the shelf break, the ocean quickly deepens down to 2000 m or deeper. Vertical distributions of aluminum from the California coast show a subsurface maximum at 100- 150 m, as well as a higher concentration of suspended particulate aluminum at this depth. A coastally derived shelf input is suggested to explain the aluminum profile in this region (Orians and Bruland 1986). In addition, while generally dismissed in open ocean profiles, aluminum from fluvial input can be relevant near the river mouth and can carry dissolved aluminum up to 20 km away from the river mouth (Brown, Lippiatt & Bruland 2010).  1.3.2  Aluminum Concentrations around the World A significant amount of the early work in the open ocean for dissolved aluminum was done the Atlantic. In the western Atlantic, the element shows a C-shape in its vertical profile, varying from 40 nmol/kg in the surface to 20 nmol/kg in the middle of the water column and an increase back to 40 nmol/kg below 3500 m (Hydes 1979, Yeats, Dalziel & Moran 1992). The recent GEOTRACES cruise in the western Atlantic shows surface concentrations above 30 nmol/kg at 0–30°N, but surface concentrations of 1-2 nmol at 60°N. There is also a large increase in dissolved aluminum below 4000 m around 40-60°N. This increase indicates an input at depth, originally proposed to be from sediments but may be from hydrothermal vents or an unknown 12  source.  High dissolved aluminum concentrations are seen in the surface in the equatorial eastern Atlantic, with values above 25 nmol/kg (Measures et al. 2008), although more the most recent GEOTRACES survey shows no surface maximum in this region (M. Hatta, eGEOTRACES, unpublished data), so this feature is evidently variable. The input to the bottom is less pronounced in the eastern Atlantic than the western Atlantic.   In the Pacific, the concentrations are around an order of magnitude lower than in the Atlantic; at a station in the middle of the subtropical North Pacific Gyre, the surface concentration was around 5 nmol/kg, followed by a decrease to 0.6 nmol/kg at mid-depth (Orians and Bruland 1986). Much like in the Atlantic, there is an increase near the sediments, but to a much lesser extent.  In the Southern Ocean, there are low concentrations at the surface, ranging from 0.5-1.0 nmol/kg, and increasing to 3-6 nmol/kg at depth (Middag et al. 2011). The surface has a high amount of variation based on location; stations closer to the Antarctic ice sheet tend to have more variable dissolved aluminum concentrations than those in the Weddell Sea near Antarctica. In the Arctic Ocean, the highest concentrations are found at the bottom (Moore 1981, Middag et al. 2009), suggesting a different source for aluminum than in the Pacific and Atlantic. The concentration reported in this region is from 0.3 nmol/kg near the surface to 10 nmol/kg at depth in the western Arctic and 1 nmol/kg to 25 nmol/kg in the eastern Arctic. Work in the Mediterranean Sea shows a profile that increases with depth, with the surface around 40 nmol/kg, elevating to 120 nmol/kg in the deepest waters (Mackenzie, Stoffyn & Wollast 1978, Chou and Wollast 1997).  13  1.3.3  Silicon and Aluminum Like aluminum, silicon is a common component of the Earth’s crust, accounting for 27% of the composition by weight (Taylor 1964, Wedepohl 1995). Since both elements are primarily derived from the Earth’s crust, the expectation of a correlation between silicon and aluminum is not an unreasonable one. The crustal ratio of the two elements is about 3:1 (Gehlen et al. 2003, Wedepohl 1995), but the elements are not generally correlated in the marine environment since silicon does not flocculate out in estuaries and is taken up by diatoms for their shells. In some ocean basins, however, there is a correlation between Si(OH)4 concentrations and aluminum concentrations, such as in the Mediterranean Sea and the eastern basin of the Arctic Ocean (Middag et al. 2009). A relationship is most likely to form in regions with high diatom growth and low aluminum and silica input, where both vertical profiles resemble nutrient profiles. Whether the link between silica and dissolved aluminum is a geochemical feature or a coincidence is still an active debate.  1.4 Method of Analysis Methods such as atomic spectroscopy (Orians and Bruland 1986) and electron capture detection gas chromatography (Measures and Edmond 1989) were historically used to measure dissolved aluminum concentrations in natural waters. However, due to on-ship portability, sensitivity, lower volume requirements, and increased sample throughput, flow injection analysis (FIA) using fluorometry is the preferred method today (Middag et al. 2009, Brown 2009, Giesbrecht et al. 2013, Measures, Brown & Vink 2005, Resing and Measures 1994). This method can be applied to river water or seawater, but requires a preconcentration step in seawater because of low dissolved aluminum concentrations in the oceans.  14  1.5 Research Objectives The main objective of this research was to characterize the distribution of dissolved aluminum in the northeast Pacific Ocean from the coast to the subarctic gyre and in the Canadian Basin of the western Arctic Ocean with better spatial and temporal resolution for these under-sampled regions. From these distributions, we aim to better understand the sources and sinks of dissolved aluminum in these regions.   In the northeast Pacific, we investigated the relative importance of continental shelf input, atmospheric dust input, and the influence of mesoscale eddies in transporting aluminum from the coast to the subarctic gyre. In the Arctic, we investigated the relative importance of sea ice, continental shelf input, advection, and influence from the Mackenzie River. We used the dissolved aluminum concentrations in the surface to infer dust deposition in the two regions.   The last objective was to optimize an on-ship analytical method that allows for lower sample contamination, low detection limits, and the opportunity to analyze dissolved aluminum at sea in real time.    15  Chapter 2: Method for Analyzing Dissolved Aluminum in Seawater To analyze dissolved aluminum in seawater, a sensitive method that would not be hindered by matrix effects was needed. Flow injection analysis (with a preconcentration column) is one of the easiest ways to do this. All steps are in-line, lowering the risk of contamination, and the method is fast enough to run samples in between sampling times on a ship.   2.1  The Method Flow injection analysis involves a column preconcentration step to enable detection of aluminum. Using a Toyopearl AF-Chelate 650M resin with a hydroxylated methacrylic polymer base, aluminum is retained by the column. The pH range necessary to facilitate a high yield in the preconcentration step ranges from 5.0-7.4, which can be obtained with an in-line ammonium acetate buffer. The column containing the resin is 2 cm long and has an internal volume of 27 uL, with non-metal frits to prevent contamination. The concentration factor ranges from 60:1 to 450:1. After the preconcentration step, a valve is switched and the concentrated aluminum is carried away with an eluent. The eluent used for the carrier is 0.1 M HCl, since that concentration is sufficient for a high analytical yield (Brown and Bruland 2008).   Aluminum is not a fluorescent compound in its natural state, so a reaction with the organic ligand lumogallion [3-(2,4-dihydroxyphenylazo)-2-hydroxy-5-chlorobenzenesulphonic acid] (Figure 1-5) is required to detect the dissolved metal. The aluminum-lumogallion complex emits a fluorescent signal at 550 nm when excited by a 485 nm wavelength excitation (Giesbrecht et al. 2013). Under the conditions of this method, only aluminum in the free ionic state or in loosely bound complexes, is detected. The first known work done using lumogallion to detect aluminum 16  showed that the optimum pH range for the analysis was around 5 for aluminum (Nishikawa et al. 1967), so a second ammonium acetate buffer (4M) is used to stabilize the pH of the system. For a faster, more complete reaction between aluminum and lumogallion, an 8m coil reaction chamber is heated to 55°C. Figure 2-1: The structure of lumogallion, from TCI America website. The aluminum forms a complex with one nitrogen and two oxygens.  After the aluminum reacts with lumogallion, another step is added to increase sensitivity. A non-ionic surfactant is added to the aluminum-lumogallion complex, increasing the complex’s fluorescence by 5 times. It is believed that the micelle structure of the surfactant offers increased rigidity and thus enhances the quantum yield (Howard et al. 1986). A study on surfactant efficacy showed that a surfactant called Brig-35 is most effective at enhancing the signal (Resing and Measures 1994). The surfactant must be added after the aluminum-lumogallion complex is formed or else it will serve to interfere with the formation rather than increasing the signal.   Finally, the micelle-enhanced aluminum-lumogallion complex is sent through the flow cell and detected by the fluorometer (Shimadzu RF-535). The 0.1 M HCl, the lumogallion solution, and the Brig-35 solution are the components of the fluorescent baseline. Since the reaction between the detected aluminum and lumogallion happens in a heated cell, Teflon tubing within the fluorometer itself is not necessary. All tubing used before the fluorometer is Teflon and cleaned 17  with acid. A peristaltic pump with perfluoroalkoxy alkane (PFA) tubing is used to move all solutions through the system. Figure 2-2: The FIA system. Adapted from Brown (2008) and Giesbrecht (2013)  Potential improvements to the method include using a column conditioning step to change the pH back to 6 before the beginning of each analysis, which causes sample retention to be slightly more quantitative (Brown and Bruland 2008). After trying this with a few analyses, this step was deemed unnecessary due to the small change in the sensitivity relative to the increased complexity of the system and the introduction of a potential source of contamination. Another proposed improvement is the removal of iron and fluoride, which are both known to interfere modestly with the signal of aluminum by adding a second resin which removes the iron and 18  fluoride or a liquid-liquid extraction (Ren et al. 2001).   2.2  Analytical Procedures 2.2.1  Blanks Sample blanks can be obtained by either running MQ water or pre-cleaned seawater through the column for 300s to find the contribution of the ammonium acetate buffer followed by running a sample for 15 seconds to find the matrix effect sans the preconcentration. It was found that the blank changes by less than 5% when using MQ water as opposed to a pre-cleaned seawater sample (Brown and Bruland 2008). Middag (2010) uses column-cleaned seawater whereas Brown (2008) and Resing (1994) both use MQ water for blank determination. Synthetic seawater cannot be used because it is high in aluminum relative to natural seawater. Since North Pacific seawater contains some of the lowest aluminum concentrations in the world, an aluminum-free blank is critical. Concern over the matrix difference between seawater and MQ ultimately led us to using standard addition in seawater. The detection limit of the method was 0.03 nM based on the longest-ran samples in the open ocean of the North Pacific, and the blank was determined by running a single low concentration sample through the column for 30, 60, 90, and 120 seconds and extrapolating to 0 sample volume.   For analysis, a standard curve was created at the beginning of the each set of analyses to test linearity. After that, (one point) standard additions were performed using high purity standards after dilution. The standards were made in 0.5% HNO3. The small amount of acid added in from the standard did not have a significant effect on the pH of the system nor does the addition of 19  NO3- affect the matrix.  An advantage of the flow injection analysis system is that it is easily adaptable to a variety of concentrations. Because samples in the northeast Pacific contain extremely low levels of dissolved aluminum, the loading time for the seawater was 300 s-1800 s (5-30 min) for the Arctic samples, 240 s–600 s (4-10 min) was more appropriate. The load times can be adjusted based on the concentration of aluminum found. In the presence of high concentration of aluminum (>85 nM), the pre-concentration step can be eliminated and the column is replaced with a sample loop. The linear dynamic range of the system is at least up to 200 nM; that is the maximum concentration of standard studied. By adjusting the preconcentration step, we analyzed concentrations from 0.06 nmol/kg – 11.6 nmol/kg.  2.2.2  Challenges The cleanliness of the ammonium acetate buffer involved in the preconcentration step is the most important part of obtaining accurate results for low dissolved aluminum concentrations. Methods for cleaning it include running impure ammonium acetate solution through a resin at an appropriate pH, bubbling ultra-pure ammonia gas into MQ (Brown 2009), or recrystallizing impure ammonium acetate to omit impurities. Attempts to purify ammonium acetate by recrystallization or by chelating resin did not yield sufficiently low level blanks. Bubbling ultra-pure ammonia gas into MQ water was impractical in the facilities available at the time of this study. Ultimately, we used high purity acetic acid and ammonia purchased from Seastar Chemicals Inc. to make the ammonium acetate solutions rather than using any of the cleaning 20  methods.   Another challenge with FIA systems is to ensure that the tubing does not become clogged.  A shaker was used to fully dissolve the surfactant and to dissolve the lumogallion complex in the second ammonium acetate buffer. An in-line filter can also be used to ensure that particles do not enter the system and disrupt the flow. 0.1 M HCl and MQ water were used to clean the system daily to prevent particles from accumulating in the Teflon tubing and the flow cell. The flow cell can also accumulate bubbles over time, which was fixed by using a syringe to insert isopropyl alcohol into the flow cell.  2.3  Sampling Methods In preparation for sample collection, low-density polyethylene (LDPE) bottles were filled with 5% Extran (a cleaning surfactant) solution for one week, 6 M reagent grade HCl for one month, and 1% environmental grade HNO3 for one month. Bottles were cleaned with MQ between steps, and after the last step, a small amount of HNO3 was left in the bottom of the bottle to continue to preserve the cleanliness of the bottle. Prior to sample collection, the bottles were washed with in-situ seawater three times to remove any remaining acid residue.   As determined during the GEOTRACES intercalibration study (Henderson et al. 2007) and confirmed in a comparison of different bottle types (Brown and Bruland 2008), low-density polyethylene (LDPE) bottles are preferred for aluminum sample storage.  In the manufacturing 21  of LDPE bottles the aluminum catalyst used for other bottle types, such as high-density polyethylene and polypropylene, is not used.  These other bottle types, therefore, lead to variable contamination for dissolved aluminum at the low levels found in the oceans. Another, more expensive alternative is using fluorinated ethylene propylene bottles.   The ships used to obtain the seawater samples are largely made of metal, so great care during collection methods is taken to avoid contamination. Go-Flo (General Oceanics) 12 L Teflon-lined bottles are used to collect the samples; the Teflon bottles are triggered to close at a specific depth in order to obtain samples. Kelvar lines are used rather than metal hydrowire. All handling of seawater on the ship is performed in an isolated environment to avoid contamination by airborne rust. Gloves must be worn at all times during bottle handling, labelling, and collection. All samples are filtered and then acidified using Instrument Quality (IQ) Seastar HCl in a HEPA filtered flow bench.  Once back on land, samples were stored for up to 5 years before analysis. Bottles were all subsampled in a Class 100 flowbench within a Class 1000 clean room. The subsequent analysis was carried out in an Airclean® 600 PCR Workstation Class 100 laminar flow hood in a standard laboratory. Contamination is minimized by the laminar flow hood and the closed system nature of the FIA.    22  2.4  Conclusion Flow injection analysis has the potential to provide shipboard analysis with accuracy and speed, allowing for real-time data collection. This is beneficial for contamination-prone elements to identify problems while still in the field. The methods can also be used on acidified stored samples, as was done in this study. All steps are in-line as part of the flow injection analysis system, reducing potential contamination. However, at the low dissolved aluminum levels found in the northeast Pacific Ocean, contamination from reagents is a significant risk, but careful reagent preparation coupled with the use of ultra-pure MQ water and pure reagents can mitigate this problem.    23  Chapter 3: Measurements of Dissolved Aluminum in the Northeastern Pacific Ocean 3.1  The Northeastern Subarctic Pacific  3.1.1  Information about the Study Area In the subarctic northeast Pacific, the overarching feature is the cyclonic Alaska gyre which brings cool, nutrient rich water from depth to the surface. At the southern edge of the gyre, the eastern flowing North Pacific Current splits into northward flow into the subarctic gyre and southward flow forming the beginning of the California Current. This area is in a transition between the warm, saline subtropical north Pacific gyre and the cold subarctic water of the Alaska gyre (Dodimead 1963).   Ocean Station PAPA (aka OSP or P26) was originally a permanent weather station in the North Pacific Ocean. In 1981, the permanent weather station was removed and ships travelled to its original location 2-5 times a year, sampling 27 stations en route (Crawford, Galbraith & Bolingbroke 2007). This transect is from 125ºW to 145ºW at around 50ºN.  Using data collected on the Line P transect, researchers have learned much about the ecosystem in this subarctic HNLC region. Data has also been used to monitor the overall health of the ocean, including a study of the oxygen depletion of the subarctic North Pacific (Whitney, Freeland & Robert 2007). Some of the features measured include temperature, salinity, and oxygen levels, as well as studies of various trace metals starting in the early 1990’s, focused largely on the August/September cruises. There are 27 stations on this transect, but five of them, shown here (P4, P12, P16, P20, and P26), are sampled in more detail (Fig 2-1).  P4 is the closest to the shore 24  and P26 is the farthest off-shore. The ship used for sampling was the John P. Tully, a Canadian research vessel, and the cruise was facilitated by the Department of Oceans and Fisheries.    Figure 3-1: The Line P time series (from the UBC PCIGR website)  The Line P transect covers nearly 1500 km from east to west, from the coast of British Columbia to the open ocean. The coastal region (sampled by P4 and P12) has spring and summer phytoplankton blooms due to upwelling (Thomson 1981), which supplies nutrients. When the wind direction is from north to south in the summer, it brings high nutrient water from depth to the surface to replace water moved offshore due to Ekman transport. Shelf break stations such as P4 tend to be macronutrient limited with respect to phytoplankton growth because of an adequate supply of iron. The subarctic gyre (sampled at P26 and P20) features a high nutrient-low chlorophyll region, a part of a much larger HNLC North Pacific (Harris et al. 2009).    25  3.1.2  Eddies in the Gulf of Alaska Mesoscale eddies form off the coast of the northeast Pacific, bringing in water from the coast into the open ocean. There are several places which eddies originate in the eastern subarctic Pacific; Sitka and Yakutat eddies come from the coast of southeast Alaska, while Haida eddies form off the Queen Charlotte Islands in British Columbia (also known as “Haida Gwaii”) (Whitney and Welch 2002). These are all anticyclonic eddies, often referred to as downwelling or warm core eddies because Ekman transport pushes warm water from the surface into their core, causing an increase sea surface height.  They form in the winter, due to the Aleutian Low, from water moving up the coast with cyclonic vorticity (Crawford 2002).  The strength and the direction of the eddy depends on the strength of the flow that produces it (Rodionov, Overland & Bond 2005).  This feature can be seen by sea surface height anomalies, using satellite data collected by NOAA. With the use of satellite altimetry to track eddy movement using sea surface height measurements, we can identify the trajectory of eddy movement (Meyers and Basu 1999). Drifters can also be used to track eddy movement.    Depending on the strength and directional movement of the eddy, they can sometimes cross paths with the Line P transect and significantly alter the temperature and salinity readings in the area with the eddy (Crawford 2002). Since the water contained inside Haida eddies comes from the fresher water in Hecate Strait, the density of the water inside of the eddy is very low compared to ambient water. There is evidence that the water inside of a warm core eddy moves nutrients along isopycnal surfaces (surfaces of same density) into shallower regions near the 26  eddy (Peterson, Whitney & Harrison 2005).   Typically, a Haida eddy will only cross the Line P transect is if there is an El Niño during the winter before the summer cruise. El Niños strengthens the northward flow onto the island chain that causes eddies to form, and as a result they take a more equatorward flow (Crawford, Galbraith & Bolingbroke 2007).   3.1.3  El Niño-Southern Oscillation The El Niño-Southern Oscillation (ENSO) has an important impact on oceanographic conditions in the Pacific. El Niño is the positive phase of the ENSO and La Niña is the negative phase. In the North Pacific, El Niño (warming) events are often characterized by a strengthened Aleutian Low, a deep mixed layer relative to normal conditions, and decreased upwelling on the coast of British Columbia during the summer (Wong et al. 2002). The ENSO can affect the growth of phytoplankton. A strong El Niño regime tends toward lower phytoplankton growth and productivity due to suppression of upwelling (Harris et al. 2009), which may affect the rate of particle scavenging (Wong et al. 1998). The two years sampled were in a transition period between a strong El Niño and a strong La Niña in 2010 and before a weak La Niña in 2011 (Null 2012).  3.2  Aluminum in the Pacific –A Region of Low Dust Input The majority of the studies on dissolved aluminum have been done in the Atlantic, partially due to the higher concentrations of dissolved aluminum found there. However, there are previous reports on dissolved aluminum in the Pacific. The Pacific Ocean is a much larger basin with on 27  average much lower dissolved aluminum (Orians and Bruland 1985), both due to lower dust deposition and a smaller percentage of ocean in contact with the continental shelf (Anschutz et al. 1998).   3.2.1  Previous Data for Dissolved Aluminum in the Pacific 3.2.1.1  Aluminum in the Eastern Subtropical Gyre Dissolved aluminum has been measured in several distinct regions of the Pacific Ocean; the most analyzed region is in the subtropical North Pacific Gyre, from the California coast to the Hawaiian Islands. Dust deposition has also been studied on the Hawaiian Islands using a filter to collect dust.  In the 1980’s, a series of cruises called VERTEX collected samples for dissolved aluminum. VERTEX 5 crossed from the coast of California to the Hawaiian Islands (Orians and Bruland 1986). A station inside of the California Current (VERTEX 5-C) shows concentrations of dissolved aluminum at the surface of around 0.5 nmol/kg, which is significantly lower than expected for the coastal station based on previous stations in the Atlantic. The high productivity derived from upwelling in the California Current is likely the cause of such low surface concentrations (Huyer 1983, Parrish, Nelson & Bakun 1981); scavenging rates from phytoplankton growth at the surface increases near the shore and adsorption removes dissolved aluminum from solution. The subsurface maximum seen in this profile was proposed to be due to shelf input from the California coast (Orians and Bruland 1986).  Higher concentrations of aluminum are seen farther from the coast of California at a station 28  (VERTEX 5-A) about halfway between California and Hawaii. This location has a vertical profile with a surface maximum of dissolved aluminum of about 1.7-1.8 nmol/kg, indicating a small but notable surface input. Below that, a minimum of about 0.3-0.5 nmol/kg at around 450-3000 m followed by an increase in concentration in the deepest waters, likely due to benthic flux and/or sediment surface remineralization (Orians and Bruland 1986).  North of Hawaii (VERTEX 4), dissolved aluminum in the surface reaches nearly 5 nmol/kg, while at depth the concentrations are similar to those near the coast at around 0.1-0.5 nmol/kg. Another study done closer to Hawaii during the 2002 Intergovernmental Oceanographic Commission (IOC) cruise showed that the concentration at depth was 2.5 nmol/kg, which is a significant increase in deeper waters from the station 100 km north (Measures, Brown & Vink 2005). This was speculated to be due to decreased scavenging (caused by lower phytoplankton productivity, as indicated by the intensity of the O2 minimum) near Hawaii relative to the offshore station. A study of the aerosol activity off the coast of Hawaii showed that aerosol dust input increases in the spring relative to the winter and is higher than March and April than it is in May (Measures et al. 2010). VERTEX 4 was sampled in July and the 2002 IOC cruise was sampled in June, so the peak of the dust input may not be captured during either cruise.   3.2.1.2  Aluminum in the Northwestern Pacific – Subtropical and Subarctic  Off the coast of Japan, there are two distinct areas; south of Japan is the western subtropical North Pacific Gyre and the Kuroshio Current, and north of Japan is the western subarctic gyre. The highest aluminum concentrations in this region are found southeast of the coast of Honshu (Measures, Brown & Vink 2005); this is to be expected, because there is both a source from 29  Japan itself and the Asiatic dust source is within proximity of the region. Concentrations of 4–6 nmol/kg are seen in the surface in the subtropical gyre, and the subsurface minimum does not go below 2 nmol/kg. This is higher than any of the eastern subtropical gyre stations sampled. On the other hand, in the western subarctic gyre, north of Japan, the surface concentrations of aluminum are around 2 nmol/kg and decrease to around 1.5 nmol/kg at depth.  From this data, we see that the stations south of the Japanese coast generally have aluminum concentrations that are 3 times higher than those to the north, indicating that the western subarctic gyre does not receive a significant amount of Asiatic dust (Buck et al. 2006). The rate of scavenging in this region is likely lower than in the eastern subtropical gyre because the profile does not change dramatically with depth.  3.2.1.3  Aluminum in the Northeastern Subarctic Pacific There has only been one independent depth profile for aluminum on the Line P transect at Ocean Station PAPA.  Another VERTEX study, VERTEX 7-T7, studied dissolved aluminum in the northeastern subarctic Pacific at station P26 and showed, at mid-depth, some of the lowest concentrations of aluminum ever studied. Between 300-700m, concentrations were below 0.06 nm/kg and surface samples in this region were measured at around 0.6 nmol/kg.  These values are exceptionally low compared to the other values measured during the VERTEX study in the subtropical gyre, where mid-depth values at around 0.3-1.0 nmol/kg.   3.2.1.4  Iron in the Northeast Subarctic Pacific Aluminum and iron are both terrestrial elements with similar oceanic inputs. In the late 1980’s, 30  interest in iron’s role as a limiting nutrient in high-nutrient low-chlorophyll regions came to the forefront, and as a result, it has been better studied than aluminum. Martin (1988) describes an experiment at VERTEX 7-T7 (at P26) where he spiked seawater samples with iron and noticed a marked difference between the unspiked samples and the spiked samples in the same light conditions (Martin et al. 1989). The effect was particularly pronounced at the high nutrient P26. The role that iron plays in biological productivity has been suspected since the 1930’s to contribute to the productivity of phytoplankton (Gran 1931); it is required for electron transport and reduction of macronutrients in the cell during photosynthesis (Wells, Price & Bruland 1995).  Continued research work has been done on iron limitation which shows that up to 40% of the world’s oceans are limited in productivity by iron, in areas that are not near high dust deposition (the Southern Ocean, for example) (Schiermeier 2004).  An in-situ iron enrichment study in the Gulf of Alaska called SERIES (Boyd et al. 2004), was performed to see if productivity could be increased by the addition of iron and to see how long-lasting the induced phytoplankton bloom was. The study showed an increase in phytoplankton productivity with an increase in dissolved iron.  3.2.2  Sources of Aluminum (and Iron) in the Northeast Subarctic Pacific The supply of aluminum and iron to northeast subarctic Pacific comes from a few primary sources; atmospheric sources, which are a relatively sporadic source to the open ocean (Jickells et al. 2005), eddies (Johnson et al. 2005), which bring coastal water offshore, shelf input, which is generally constrained to the coast baring periodic advection (Lam et al. 2006), and volcanic emissions (Hamme et al. 2010), which can fuel sporadic phytoplankton blooms in the North 31  Pacific region from the Aleutian Islands.   The dust sources to the subarctic North Pacific are from the North American mainland, the Alaskan shore, the Aleutian Islands (Boyd, Berges & Harrison 1998), and Asia. According to dust models, North American dust does not move very far offshore and is likely not a large source of dust to the HNLC. An Asiatic source to P26 can be seen from lead data (McAlister, personal comm.), although only a small amount of the Asiatic dust reaches the subarctic gyre.  In the northeast Pacific, Haida eddies bring coastal water from the Hecate Strait into the open ocean, including high concentrations of dissolved trace metals. A study found that iron in an eddy can be 2 orders of magnitude higher than the ambient levels (Johnson et al. 2005). As a result of iron input, phytoplankton growth often increases near eddies, even though anticyclonic eddies are usually less productive than the water around them due to downwelling and a deep mixed layer in the core. A survey of the chlorophyll-a present in the Gulf of Alaska showed that around half of the productivity on the surface of the gulf is within 150 km of the center of an eddy (this value can go up to 80% in the spring) (Ueno, Crawford & Onishi 2010, Crawford et al. 2005). Since these mesoscale eddies only comprise about 10% of the total area of the Gulf of Alaska at any given time, this productivity level is disproportionately high. Based on the iron study and the relationship between dissolved iron and dissolved aluminum, dissolved aluminum may also be higher inside of an eddy.   Continental shelf input is influential at stations closest to shore, bringing dissolved aluminum to the subsurface via advection. Advection from the shelf is limited at offshore stations (Whitney 32  and Welch 2002). Volcanic eruptions are interesting, but rare, and not likely to play a significant role in the north Pacific’s geochemistry on most timescales.   All previous studies indicate that the northeast Pacific, especially P26, should contain low concentrations of dissolved aluminum relative to other regions of the Pacific as well as the Atlantic, although events such as eddies and volcanic eruptions in the region may increase the aluminum input.  3.3  Results - Distribution of Dissolved Aluminum in the Northeastern Pacific Ocean 3.3.1  Sampling Protocol On the 2010 Line P cruise, trace metal sampling was done with a trace metal clean pump down to 40 m; below that, samples were collected with 12 L GO-FLO bottles manually attached to a Kevlar line and deployed off the side of the boat. In addition, surface samples were collected on a Zodiac boat by hand. These samples were all filtered using a 0.1 µm Acropak™ 500 filter with a Supor® membrane, and then the samples were acidified to pH = 1.7 (using 1ml Seastar 6N HCl per 500ml bottle) in a laminar flow hood on the ship. The deepest sample we sampled in 2010 was 800 m.  On the 2011 cruise, two major changes were made to this protocol. Samples were collected with a trace metal rosette with a powder-coated frame, modeled after (Measures et al. 2008), allowing for 12 bottles to be sampled per cast. This allowed for an increase the sampling resolution at each 33  station. Also, samples were acidified in the shore based lab, one to two weeks after the collection and filtration. The deepest sample we sampled in 2011 was 2000 m. Five stations were sampled on both of these cruises; P4, P12, P16, P20, and P26. Due to contamination issues, P20 results are included in the table but not included in the discussion for either of the years and P26 vertical profile is only presented for 2011.  3.3.2  Results for Dissolved Aluminum along Line P Starting at the historical station P26, we see concentrations in the surface layer (0-40 m) of 0.36-0.56 nmol/kg. Below that, we see a decline from 0.36 nmol/kg at 100 m to values of 0.06-0.10 nmol/kg below 200 m. Plotted below (Figure 3-2a) is a comparison to the previous study done at P26; the high resolution sampling in 2011 is consistent with the results of the VERTEX VII T-7 study (Orians and Bruland 1988).   At P16 in 2011 (Figure 3-2b), the vertical profiles shows us a similar pattern, but the concentrations are generally higher at all depths. The surface layer (0-40 m) has dissolved aluminum concentrations of 0.69-0.92 nmol/kg, with concentrations below the surface of 0.51-0.55 nmol/kg and steadily declining to 0.12-0.24 nmol/kg below 200 m. At P16 (Figure 3-3a), the surface values are around 0.68-0.91 nmol/kg; below that, we see an increase to a maximum at 75 m at 2.09 nmol/kg followed by a decline to 2011 levels at 300 and 400 m.  At P12 in 2011, we see a fairly steady concentration of 1.0-1.8 nmol/kg down to 300 m, in sharp contrast to the previous profiles discussed (Figure 3-3b). Little vertical feature is seen in this profile; there is no surface maximum, and there is a small increase with depth.  At both P12 and 34  P16 in 2010 we see very different profiles than in 2011. At P12 (Figure3-3b) in the surface, we see a concentration of 3.6 nmol/kg, a decrease in concentration around 10-25 m to 1.6-1.7 nmol/kg, and an increase in concentration from 50 to 150 m reaching a maximum of 3.6 nmol/kg. Below 150 m, the concentration decreases again to 2.1 nmol/kg at 300 m.  At P4 in 2011 (Figure 3-4a), we see minimum concentrations in the surface of 1.7 nmol/kg. We then see a subsurface maximum of 3.3 nmol/kg at 100 m followed by a decrease to 1.5-2.3 nmol/kg below that (Figure 3-4b). In 2010, there were only five samples taken at the surface, and they range from 0.7 nmol/kg to 1.7 nmol/kg.  3.3.3  Vertical Profiles for Dissolved Aluminum along Line P   Figure 3-2: Dissolved aluminum concentrations at (a) P26, 50°00'N, 145°00'W and VERTEX VII-T7 (Orians and Bruland 1988) (1987) (b) P16, 49°17' N, 134°40'W  05001000150020000 0.5 1depth( (m)Dissolved Al(nmol/kg)P26(a)0 0.5 10500100015002000Dissolved Al(nmol/kg)P16(b)35    Figure 3-3: Comparison between 2010 and 2011 for dissolved Al down to 400 m from (a) P12, 48°58' N, 130°40'W and (b) P16, 49°17' N, 134°40'W   Figure 3-4: Dissolved aluminum concentrations from P4, 48°39' N, 126°40'W. (a) to 400 m and (b) to 2000 m. 0501001502002503003504000 1 2 3 4Dissolved Al(nmol/kg)P16(a)0501001502002503003504000 1 2 3 4depth (m)Dissolved Al(nmol/kg)P12(b)0501001502002503003504000 1 2 3 4depth (m)Dissolved Al (nmol/kg)P4(a)0200400600800100012000 1 2 3 4depth (m)Dissolved Al (nmol/kg)P4(b)36  3.4  Discussion  3.4.1  Ocean Station PAPA – Open Ocean  P26 is a part of the larger HNLC northeast Pacific Ocean, a region that lacks significant continental shelf input and has a low amount of dust input (Han et al. 2008). Consistent with the data from 1987, we observe extremely low dissolved aluminum concentrations. Both profiles from the northeast Pacific shown have concentrations down to 0.06 nmol/kg in the deeper waters, which, to our knowledge, are lower than previously measured work anywhere in the world. This is also consistent with previous dissolved iron studies in this region, which show extremely low iron concentrations (Boyd, Berges & Harrison 1998).  The profile shows a small dust input in the surface. If we observe rain and wind data from those regions from the days before the sampling, there are both heavy winds coming from the northern coast and large amounts of rainfall, which may have delivered dissolved Al to the surface ocean. The surface concentrations are likely to be variable depending on the timing of sampling compared to input events, due to the short residence time of dissolved aluminum in the surface, and we do see the greatest difference between the two years in the uppermost samples.  3.4.2  P12 and P16 – Offshore – Coastal Input and Eddies At P16 in 2011, we see consistently higher concentrations than at P26 at all depths, but the dissolved aluminum is still very low relative to other locations in the Pacific. P12 in 2011 does not have a surface maximum, but concentrations are significantly higher than those seen at P16. A dust source or a continental shelf source could be responsible for supplying aluminum at P12 and P16. We do not see a subsurface maximum at either of these stations, unlike at P4 in 2011, 37  but there is less of a subsurface minimum at these stations compared to P26, so there could be an advective subsurface source.  During the 2010 cruise, the influence of an eddy was seen along Line P, with the core of the eddy at P13. It was observed onboard the ship and confirmed by sea surface height measurements on the map below (Figure 3-5). The sea surface height near P12 is much higher than in 2011. This eddy was not particularly large and had a small sea surface height anomaly compared to larger eddies, but it was generated recently from the coast. Since North Pacific eddies can extend up to 240 km (Crawford et al. 2005), we would expect the influence of the eddy to be seen at multiple stations.  The influence could also be extended by transfer along isopyncal surfaces.    38     Figure 3-5: Sea surface height measurements at Line P (from NOAA’s website) for 2010 (top) and 2011 (bottom). Scale is cm above sea level.  From the vertical profiles at P12, we observe a large increase in aluminum from 50–150 m in 2010 compared to 2011, a slightly elevated concentration at 200 m, and a convergence at 300 m. The high surface concentration could be from the eddy or from dust input. There was a wind and rain event in the two days before the sampling, so that may contribute to the increased concentration. At the time of the 2011 sampling, there does not appear to have been a significant 39  dust input.   At P16 in 2010, we also see the influence of the eddy, with concentrations to 200 m which are elevated relative to 2011; the profile looks very similar to that of P12 but with a lower maximum concentration of aluminum. The samples below the influence of the eddy at 300 – 400 m are nearly identical to the 2011 samples. Unlike at P12, we see a maximum dissolved aluminum concentration at 75 m rather than 150 m.   Previous studies show that both the core and the edge of coastally-derived eddies have elevated levels of dissolved iron (Johnson et al. 2005). Isopycnal transport between the core of the eddy (which has a much lower density than non-eddy water at the same depth) and the edge of the eddy (which will have a density closer to ambient density) will cause an increase in dissolved iron concentrations along the edges of an eddy.   We believe that this eddy transferred aluminum-rich coastal water offshore.  If aluminum is transferred through isopyncal surfaces, similar to dissolved iron (Johnson et al. 2005), we would expect the maximum concentration of dissolved aluminum to be at a shallower depth near the edge of the eddy than near the core. Our observation of a shallower depth for the P16 maximum concentration than the P12 concentration is consistent with this.  3.4.3  P4 – Coastal Input P4 is around 120 km offshore near the shelf break. The water has a substantial amount of influence from the poleward California Undercurrent during the summer at 100-200 m depth. It 40  is warmer and saltier (also known as “spicy”) than water offshore due to upwelling and a supply from the California Undercurrent (Hickey 1979, Thomson and Krassovski 2010). Upwelling can introduce a notable sedimentary input from suspended solids (Whitney, Crawford & Harrison 2005).    Aluminum concentrations in this region are higher than those further offshore because of a supply from the coast of British Columbia from dust and/or the continental shelf. The lack of iron limitation is also evidence for a larger input of terrigenous material to this region. A higher concentration of aluminum is seen in the surface waters in 2011 than in 2010, possibly due to increased dust input. Plotted below is the 2011 vertical profile at P4 compared to a vertical profile in a coastal location off California. They are a similar distance from the shore and both near the continental shelf break. A subsurface maximum is seen in both locations, although concentrations are higher at all depths off the coast of British Columbia. Due to the high productivity of the California Current, dissolved aluminum is rapidly scavenged, so the input is offset by a high removal rate (Orians and Bruland 1986).  41   Figure 3-6: Dissolved aluminum concentrations at P4 in 2011 compared to values off the coast of California (36°06' N, 122°38'W)  3.5  Closing Remarks about the Northeast Pacific We see a dissolved aluminum concentration gradient between the coastal stations and the offshore stations along the Line P transect. At mid-depth, concentrations were above 1.5 nmol/kg at P4 and below 0.1 nmol/kg at P26. Two sources, continental shelf and aeolian input, are proposed to supply dissolved aluminum to this region. Surface maxima are more common offshore, but are spatially and temporally variable, presumably due to sporadic dust input events.  The coastal station (P4) from this study shows vertical structure similar to a station off the California coast, but with higher values throughout the water column. Both show a subsurface maximum, which is believed to be from the continental shelf.  The station in the subarctic gyre (P26) is similar in shape to results from VERTEX 7-T7 at the same location in 1986, showing a 0200400600800100012000 1 2 3 4Dissolved Al(nmol/kg)P442  mid-depth concentration lower than previous measurements for dissolved Al anywhere in the world. The surface concentrations were higher in 1986 with a strong gradient in the upper 50 m. Although sampling depths were not identical, the differences likely reflect differences in dust input. Further inshore at P16, the surface maximum was comparable to V7-T7, but concentrations are higher at mid-depth.  An eddy was observed in the middle of the Line P transect in 2010. An increase in dissolved aluminum concentration is consistent with a source from the coast of British Columbia. Both the station near the core of the eddy (P12) and the station at the edge of the eddy (P16) had elevated dissolved aluminum concentrations in subsurface waters down to 200 m. This source of aluminum is also a source of iron, and is one of the main non-dust sources of iron to the northeast Pacific HNLC region.   43  Chapter 4: Measurements of Dissolved Aluminum in the Western Arctic Ocean 4.1 The Western Arctic Ocean 4.1.1  Information about the Study Area The Arctic Ocean is at a high risk for impact from climate change because of changing sea ice cover (Aagaard and Coachman 1975). Therefore, research into the geochemistry of the Arctic Ocean is timely and may be helpful to discovering both past processes from glacial eras and to monitor the changes from climate change as the world moves forward with increasing global pCO2 concentrations.  The western Arctic is the basin adjacent to Alaska and the Canadian territories Yukon and the Northwest Territories (McLaughlin et al. 1996). Near the Mackenzie River, the bottom is shallow, going only down to 200 m, whereas the open Arctic Ocean it goes below 4000 m. Organic material and nutrients from the Mackenzie River can influence phytoplankton growth; glacial melting and influx from the Pacific and Atlantic Oceans can affect the physical movement of water as well as supplying nutrients to the Arctic Ocean.  The large continental shelves are a supply of particles to mid-depth in the basin. Low light levels due to ice cover in the winter and early spring leads to sparse phytoplankton growth; the main growth is ice algae, which are an important part of the food chain in the Arctic (Carmack, Macdonald & Jasper 2004). In the summer, there are often phytoplankton blooms, leading to low nutrient concentrations; growth is limited by nitrate in the open ocean and phosphate near the 44  Mackenzie River (Emmerton, Lesack & Vincent 2008, Macdonald, Wong & Erickson 1987).  Much like the North Pacific, there are eddies in the Arctic, and they are even more plentiful.  During a 1974 survey of Arctic dynamics, there were 127 distinct eddies found over a 14-month period (Newton, Aagaard & Coachman 1974, D'Asaro 1988). . Most of them are anticyclonic and responsible for the large amount of kinetic energy found in the region (Manley and Hunkins 1985). Unlike the Haida eddies in the North Pacific, there is little evidence to show an increase in phytoplankton productivity in these eddies; this region may represent a more typical regime of decreased nutrients in areas with anticyclonic eddies. However, depending on the origin of the eddy’s water, nutrients from the Pacific might be supplied, specifically from the Chukchi Sea to the northeast of Alaska (Mathis 2006).  The region’s water masses can be subdivided by their origin – the surface mixed layer from 0 to 50 m, is fresher and often influenced by ice melt. The upper halocline is cold water of medium salinity at 50–150 m from the Bering Strait and shallow Pacific water. This feature is only present in the western Arctic (McLaughlin et al. 1996, Aagaard, Coachman & Carmack 1981). Atlantic Ocean water is seen from 200-1000 m, with a higher salinity than Pacific water. Below that, the water is simply classified Canadian Basin Deep Water and is very salty due to brine rejection incorporated in sinking shelf water. The Canadian Basin, much like the Eurasian Basin, has seven layers of water below 1700 m which correspond with a different salinity, each with a different entrainment rate from shelf water and different 14C distributions (a measurement of the time since the water mass has had contact with the surface) (Jones, Rudels & Anderson 1995). The deeper waters of the Canadian Basin come from the Lomonosov Ridge, bringing in water 45  from the Eurasian source.        Figure 4-1: The location of the stations relative to Canada   4.1.2  The Mackenzie River The Mackenzie River supplies freshwater into the Canadian Basin, especially during May-September when ice on the continent melts (Emmerton, Lesack & Vincent 2008). Earlier in the year, thick sea ice restricts the movement of the Mackenzie River water at low flow. Freshwater input causes a large amount of brackish water on the surface of the Arctic, forming a 2m thick layer of low salinity water in the summer (Carmack and Macdonald 2002). There is no meaningful amount of precipitation to change the freshwater budget in the Arctic Ocean (about 10 cm/yr) (Macdonald et al. 1989).   In the spring, melting ice leads to more river discharge, releasing warm and turbid water into the Beaufort Sea. The timing of the runoff of Mackenzie River has recently changed due to the effects of climate change, coming earlier than in the past (Rouse et al. 1997). The timing of land ice melt and the increased melting of the tundra permafrost in northern Canada has increased the 46  amount of water the Mackenzie River delivers to the Arctic Ocean (Moritz, Bitz & Steig 2002). This may increase nutrient input and alter the nutrient distribution of the Arctic Ocean (Macdonald, Harner & Fyfe 2005). For the foreseeable future, the Arctic Ocean will not be steady state relative to nutrients, and changing phytoplankton distribution is expected.   4.1.3  Seasonal Glacial Melting and the Effects of Climate Change The second major supply of freshwater in the Arctic is sea ice melting, which increases in magnitude as the temperature increases with elevated atmospheric CO2 (Moritz, Bitz & Steig 2002). Air temperature plays a vital role in controlling the seasonal variation in ice cover in the Arctic Ocean; it changes both when the sea-ice retracts during the summer and to what extent it retracts (Macdonald et al. 2000). During the winter, both land-based glaciers and sea ice dominate the Beaufort Sea, extending from the surface to the 20 m isobath (Carmack and Macdonald 2002). Both seasonal light limitation and albedo from the ice allow little light to come into the Arctic in the winter. In the late spring, warm water from rivers flows out across the ice, supplying fresh water to the Arctic Ocean. Biological production is inhibited due to light limitation caused by the albedo of the ice and the snow that still inhabits the area.   In the summer there is continued stratification due to the supply of water from the river (Macdonald et al. 1989) which, along with the sea-ice melt, causes a surface layer to form in the top 5-10 m. During the later summer, there is little ice in the Beaufort Sea and phytoplankton can grow. New production in this region is estimated at around 3-12 g C m-2yr-1 (Carmack, Macdonald & Jasper 2004). The silicate depletion relative to winter is around 20 mmol/m3 (normalized to salinty) near the shelf of the Canadian Basin, indicating uptake by diatoms 47  (Parsons et al. 1988). On the middle and outer shelf and into the open ocean in the Beaufort Sea, the silicate supply is lower, and smaller creatures dominate the biology instead (Hsiao 1976).   Climate change will likely have a profound effect on the seasonal distribution of nutrients in the Arctic Ocean and the levels of productivity. The extent of ice cover has been decreasing during the summer in the last twenty years (Rothrock, Yu & Maykut 1999), as well as the thickness of the ice, and it is believed that in the next century we could see an ice-free Arctic (Wang and Overland 2009). This will likely lead to an increase in precipitation, lowered albedo, increased river discharge (Burn 2008), and changes in the hydrological cycle which are hard to predict (Carmack and Macdonald 2002).   4.2 Aluminum in the Arctic -- A Polar Basin The Arctic was the third location measured for dissolved aluminum, with concentrations an order of magnitude lower in the surface than the Atlantic. Aluminum profiles were obtained for the Central Arctic in the early 1980's showing concentrations ranging from 5 nmol/kg at the surface to 20 nmol/kg at around 3000 m, with an increase in concentration with depth (Moore 1981). Both sides of the Lomonosov Ridge, the dividing line between the Canadian Arctic and the Eurasian Arctic, showed similar depth distributions. These profiles are a contrast to the profiles of zinc, cadmium, and copper in the Arctic, which behave more like traditional nutrients. It was previously believed that aluminum was involved directly in biological processes, but based on this early study it clearly had a different profile from the other nutrients and thus had a different role to play in ocean dynamics. 48  The Arctic Ocean is a combination of different water masses; in the Nansen Basin and the Amundsen Basin, Atlantic water is the primary influence. In the Makarov Basin, the boundary between the Eurasian and the Canadian Arctic, we see an input of both Atlantic and Pacific water, with Pacific water entering at a shallower depth due to its lower density (Coachman 1969). The influence of Atlantic water at depth is seen even in the Canadian Arctic, although now it competes with both Pacific winter water and Pacific summer water, which form two distinct layers (Carmack and Macdonald 2002).  4.2.1  Aluminum in the Arctic Ocean 4.2.1.1  2007 Study of Aluminum in the Eurasian Arctic Ocean There was a comprehensive study in the Eurasian Arctic Ocean in 2007, sampling much of the eastern coast and close to the North Pole, including the Laptev Sea near Siberia, the Barents Sea between Norway and Russia, and a large transect going from the Kara Sea to the Makarov Basin, the body of water separating the Canadian Basin and the Eurasian Basin.   Figure 4-2: Study area of the Eurasian Arctic in 2007 (Roeske et al. 2012)  49  Low aluminum concentrations were found at the surface in the entire region, from 0.5-2.0 nmol/kg, with a gradual increase with depth. In the Nansen Basin, closest to the Eurasian coast, there is a sharper increase in dissolved aluminum from the low surface concentrations; the concentration is above 10 nmol/kg at 1000 m and above 20 nmol/kg at 2000 m. In stations near the Eurasian Continental Slope, the dissolved aluminum concentration is over 5 nmol/kg at 200 m. Further from the coast into the Makarov Basin, which is more influenced by Pacific water, the increase in dissolved aluminum is more gradual, with dissolved aluminum concentrations reaching 5 nmol/kg at 1000 m and 10 nmol/kg at 2000 m. The concentrations increase at all depths below the surface.  A correlation between aluminum and silica was found in this region, believed to be caused by adsorption of aluminum onto silica shells of diatoms. The highest concentration of aluminum is at depth, where salty Atlantic water flows in. In regions with greater Pacific water influence and more water mixing, such as the Makarov Basin, the relationship between aluminum and silica is not present.  4.2.1.2  2007 Study of Aluminum in the Canadian Arctic Ocean Dissolved aluminum concentrations were also measured in the Canadian Basin, in the Beaufort Sea in 2007, showing somewhat elevated surface concentrations, a subsurface minimum and a maximum at depth (Giesbrecht et al. 2013), a contrast to the Eurasian study in the same year. At the surface, the concentration is from 2.5 nmol/kg at the station furthest from the shore, and 4 nmol/kg at the second furthest station. There is a concentration range from below 0.3 nmol/kg in the subsurface to 6 nmol/kg at depth, with one station having a concentration of 10 nmol/kg at 50  400 m. The deepest waters sampled on this transect is 1000 m, which does not provide any insight into the deeper water column of the western Arctic. In the western Arctic, there is no correlation between dissolved aluminum and silica. The relationship seems to be specific based on region.  4.2.1.3  Iron in the Western Arctic Ocean Sampling from the Chukchi Sea off the coast of Alaska, a few hundred kilometers west of the Mackenzie River delta, to the western Beaufort Sea (Aguilar-Islas et al. 2013), showed that shelf activity supplied dissolved iron from the Chukchi Sea.  Sources of dissolved iron and dissolved aluminum are often linked, so a source of Al from this region is also expected.  There is relatively high iron in the Beaufort Sea, but there is low productivity due to limited macronutrients. This source of iron is believed to come from the shelf and melting sea ice. The iron concentration is highest at around 100 m rather than in the surface due to biological processes. For aluminum, the surface values are expected to be less influenced by biology. Closer to the shore is a supply from the Alaskan Coastal Current, which entrains iron containing sediments as well as high nutrients; these features are delivered through canyon upwelling and supply iron to coastal waters.  4.2.2  Sources of Aluminum (and Iron) in the Western Arctic Ocean Considering that the Arctic Ocean has multiple water masses of different salinity, temperature, and place of origin, sources of dissolved aluminum and iron are hard to differentiate. Advection from the shelf region, water from the Atlantic and Pacific Oceans, ice transport and melting, dust input, and fluvial input are all possible sources. There is currently debate over what the most 51  important source of dissolved aluminum is – the two best candidates are sea ice input and continental shelf input.  4.2.2.1  Dirty Sea Ice or Shelf Input? A study of surface samples from the Chukchi Sea to the coast of Siberia was done in 1994. Due to finding unusually high concentrations of reactive aluminum at the surface, it was speculated that dirty sea ice was a source for dissolved aluminum. This proposed 'dirty ice' entrained particles from the shelf, which was then digested and released as dissolved aluminum inside of the ice (Measures 1999). Iron studies in the Arctic in 2008 also put forth the hypothesis of dirty ice's influence on the distribution of iron from the Eurasian Basin to the Makarov Basin (Klunder et al. 2012).   The concept of ‘dirty sea ice’ was disputed due to the absence of an increase of aluminum at the surface in the study from the Eurasian Arctic (Middag et al. 2009). A shelf or suspended sediment source, not dirty sea ice input, is proposed to be the source of dissolved aluminum in this region. The dissolution of silica produced by diatom blooms in the Arctic releases aluminum to the dissolved phase near the sediment surface (Roeske et al. 2012), so this could explain the input of dissolved aluminum. The Arctic Ocean has about 30% of its surface area in contact with the continental shelf (Carmack and Chapman 2003), increasing the likelihood of benthic exchange.  Rejected brine water accumulates high dissolved aluminum from the shelf before sinking, and this is proposed to be a reason for the deep source of dissolved aluminum (Measures and Edmond 1992). Surface water from the Atlantic also has a large influence on the deeper water in this region, due to an advective supply over shelf sediments. 52  4.2.2.2  Other Inputs Early research regarding dust input into the Arctic suggested that the primary source of dust is from Eurasia rather than from the Northern Hemisphere (Rahn 1981).  Most of the early dust modelling work did not include Arctic Ocean at all, aside from some discussion about organic pollutants (Duce et al. 1991). Dust deposition in the Arctic Ocean shows high interannual variability (Werner et al. 2002). It is hard to quantify how much aluminum and iron is reaching the Arctic via aeolian input, but the dust input is likely reduced because of ice obstruction, but this could later be delivered by melting sea ice.  Reduced ice cover on land, as a result of climate change, can contribute to an increase in erosion and wind; both due to the increase in surface area not restricted by ice, allowing contact with the wind. Increased Arctic precipitation results in more rainwater scavenging of dust particles from the air, which also increases dust input into the ocean. These factors may contribute to increased aluminum and iron concentrations over time in the Arctic Ocean (Jorgenson and Brown 2005, Rouse et al. 1997, Burn 2008). There is insufficient data to determine if there have been changes over time to date.  Fluvial input is generally considered unimportant in the biogeochemistry of aluminum in the global ocean, but in the Arctic Ocean, river input is a much larger part of the water system. Despite only being 1% of the total ocean by volume, 10% of the world’s river input flows into the Arctic Ocean (Dittmar and Kattner 2003). Near a river mouth, it is possible to see a high signal of dissolved aluminum before it is scavenged and precipitates out into the riverbed (Brown and Bruland 2009). Based on results from other regions, such as the Columbia River and the 53  Alaska gyre (Brown, Lippiatt & Bruland 2010), even stations 40 km away see at least an order of magnitude decrease in aluminum concentration relative to river water. However, the highly stratified freshwater surface layer could affect the amount of dissolved aluminum that is transported from the Mackenzie River to the open ocean; the lower pH, lower salinity water may decrease the rate in which flocculation occurs (Hydes 1989).  Advection may play an unusually large role in dissolved aluminum transport in this region due to the low scavenging rates. In the deep part of the western Arctic, the Lomonosov Ridge may provide a significant source of dissolved aluminum due advection of waters in contact with sedimentary matter (Jones, Rudels & Anderson 1995).  4.3  Results - Distribution of Dissolved Aluminum in the Western Arctic Ocean 4.3.1  Sampling Protocol Samples were collected with a trace metal rosette with a powder-coated frame (modelled after Measures et al. 2008), allowing for 12 bottles to be sampled per cast. These samples were filtered using a 0.1 µm Acropak™ 500 filter with a Supor® membrane and acidified to pH = 1.7 (using 1ml Seastar 6N HCl per 500ml bottle) onboard the ship. The lowest depth sampled was 3000 m. Five stations were analyzed from this cruise – L1, L1.1, L1.5, L2 and the coastal S4 (Figure 4-3). 54     Figure 4-3: The stations sampled in this study  4.3.2  Results for Dissolved Aluminum in the Western Arctic Ocean At station L1 (Figure 4-4a, Figure 4-5a), the lowest concentrations are at the surface, typically ranging from 0.5-0.8 nmol/kg, except at 15 m, which is 2.5 nmol/kg. The concentration increases with depth, approaching 7.0 nmol/kg at 1800 m. L1.1 (Figure 4-4b, Figure 4-5b) shows more scatter but the concentration of dissolved Al still generally increases with depth. The surface concentration is from 1.3-1.9 nmol/kg, 0.9 nmol/kg is seen at 50 m, and higher values around 4-5 nmol/kg seen below 1600 m.   L1.5 (Figure 4-4c, Figure 4-5c), further offshore, has the same general pattern as the previous two. The surface has the lowest concentration (0.6 nmol/kg) and the concentration increases with depth to 3.3-3.8 nmol/kg at 800-1000 m, the lowest depths sampled. L2 (Figure 4-4d, Figure 4-5d), the station furthest offshore, has lower concentrations than those observed at the other stations in the upper 1000 m. The concentration at the surface is 0.3-0.6 nmol/kg and a gradual increase up to 1700 m, where the concentration is 5.3 nmol/kg. Below this level, scatter increases 55  with values from 8.1-11.6 nmol/kg.   Lastly, S4 (Figure 4-6) is a coastal station sampled down to 600 m. The surface samples range from 0.9-1.2 nmol/kg and decrease to a minimum at 50 m. Below 50 m, the concentrations increase again; at 600 m, the concentration is 1.9 nmol/kg.    56   4.3.3  Vertical Profiles for Dissolved Aluminum in the Western Arctic Ocean   Figure 4-4: Dissolved aluminum at (a) L1 (71°06’ N, 139°18’ W); (b) L1.1 (72°30’ N, 135°35’ W) (c) Station L1.5 (73°19’ N, 139°23’ W), (d) Station L2 (74°36’ N, 137°07’ W) to 3000 m. 200 m sample at L1.1 is suspected to be contaminated.  0500100015002000250030000 2 4 6 8 10 12depth(m)Dissolved Al (nmol/kg)L1(a)0500100015002000250030000 2 4 6 8 10 12Dissolved Al(nmol/kg)L1.1( )(b)0 2 4 6 8 10 12050010001500200025003000Dissolved Al(nmol/kg)depth(m)L1.5(c)0500100015002000250030000 2 4 6 8 10 12Dissolved Al(nmol/kg)L2(d)57    Figure 4-5: Dissolved aluminum at (a) L1 (71°06’ N, 139°18’ W); (b) L1.1 (72°30’ N, 135°35’ W)(c) Station L1.5 (73°19’ N, 139°23’ W), (d) Station L2 (74°36’ N, 137°07’ W) to 1600 m  020040060080010001200140016000 2 4 6depth(m)Dissolved Al (nmol/kg)L1(a)020040060080010001200140016000 2 4 6Dissolved Al(nmol/kg)L1.1(b)0 2 4 602004006008001000120014001600Dissolved Al(nmol/kg)depth(m)L1.5(c)020040060080010001200140016000 2 4 6Dissolved Al(nmol/kg)L2(d)58    Figure 4-6: Dissolved aluminum concentrations at S4, 71°11’ N, 132°56’ W  4.4  Discussion 4.4.1  Beyond the Shelf Station L1, the closest to the Mackenzie river mouth, shows no signs of river water influence. A slightly elevated dissolved aluminum is observed at 15 m; it could be a source of sea ice melt or river water, but contamination from sampling or long-term storage is the more likely explanation. No other sample from this station supports this as a feature and the increase is not observed in the uppermost sample. Otherwise, we see a consistent increase with depth all the way to 2000 m.  L1.1 is closest to Station KC2000 on the 2007 cruise (Figure 4-7, (Giesbrecht et al. 2013)). The concentration we observe at the surface in 2009 is half the concentration reported from 2007. The aluminum concentrations in the 2007 profiles in this region are also higher below 250 m, though they agree from 50-150 m. Changes in gyre circulation affecting the river plume 01002003004005006000 1 2 3 4depth (m)Dissolved Al(nmol/kg)S459  trajectory and differences in sea ice melt from year to year are likely reasons for the variability in the surface values. Studies on barium in 2007 and 2009 season show a very different distribution of sea ice melt. In 2007, a strong surface layer signal of sea ice melt was observed, whereas in 2009 most of the freshwater in the surface is derived from Eurasian rivers (Sim, IPY GEOTRACES poster)    Figure 4-7: L1.1 (72°30’ N, 135°35’ W) and KC2000 (71°73’ N, 135°50’ W) (Giesbrecht et al. 2013), down to 600 m. 200 m sample at L1.1 is suspected to be contaminated.  At Station L1.5, we see no notable surface input in these samples, although the lowest point is below the surface at 150 m. Surface concentrations at L2 are very low for this region and consistent with low sea ice and low dust input at the surface. Shelf input to the surface will also likely decrease as you move further into the basin, especially in the western Arctic.  Both Station L1.5 and L2 show elevated dissolved aluminum at 800 m, but it is difficult to determine if this is 0501001502002503003504004505000 2 4 6 8 10depth(m)Dissolved Al(nmol/kg)L1.1 vs. KC2000( )60  a real feature. It is possible that there is an advective source of dissolved aluminum at this depth, possibly from the Eurasian Basin.  Scatter and elevated values are observed below 2000 m at L2 (the only station sampled multiple times below this depth). Scatter below 2000 m was also observed in Moore (1981) for dissolved aluminum, and other studies showed scatter below 2000 m for other elements such as lead (Charters, thesis). Multiple water masses in the open Arctic mix together in the Makarov Basin in the central Eurasian Arctic as shown by Middag (2009).  Other studies in deep waters indicated an unpredictable pattern below 1500 m in the Arctic, likely due to interaction with shelf sediments and complicated distributions of water masses (Jones, Rudels & Anderson 1995).  The increase in scatter in the dissolved Al values in the deeper samples at this station could be due to this complex situation. Contamination is possible but with concentrations of 10-12 nmol/kg, that seems unlikely.  4.4.2  S4 Sampling -- Near the Shelf S4 is close to the KC200 station on the 2007 transect (Figure 4-8, (Giesbrecht et al. 2013)). Station S4 is likely more influenced by the river plume water than L1, even though it is further away from the river mouth, due to the trajectory of the plume. The route the Mackenzie River plume takes varies from year to year based on changes in upwelling in the region. In 2007, there was a strong upwelling regime which likely transported the river plume farther offshore, but there seems to be no equivalent event in 2009.  The most notable difference between the two profiles is the value at 175-180 m; in 2007, this 61  depth had dissolved aluminum up to 3.5 nmol/kg whereas it was below 1.0 nmol/kg in 2009. This could be contamination or an actual feature; unfortunately, the KC200 station is shallower, and was only sampled to 175 m, leaving this inconclusive. From the 2009 data there does not seem to be a major input of aluminum at depth, down to 600 m.   No major fluvial input signal is seen in this profile.  Either the plume trajectory did not bring sufficient river water to this region, or the distance from the river mouth allowed adequate time for aluminum to be scavenged. This is true both in 2007 and 2009; during the 2007 sampling there was a higher amount of Mackenzie River in the region, due to upwelling, but there still was little influence on the dissolved Al concentrations at this location. This is consistent with previous studies of aluminum flocculation, which show that a salinity of 25 is adequate for flocculating dissolved aluminum (Mackin and Aller 1984). It seems that the Arctic’s surface salinity does not affect the flocculation rate enough to see additional aluminum 160 km from the river plume.  62   Figure 4-8: S4 (71°11’ N, 132°56’ W) and KC200 (70°97’ N, 135°20’ W) down to 200 m  4.5  Closing Remarks about the Western Arctic  We did not observe a surface maximum at any of the stations sampled in 2009, in contrast to the results of the 2007 study, with the possible exception of S4. The debate over the importance of a sea ice source for dissolved Al may be due to real variability based on extent of melt as well as wind patterns and upwelling cycles. Despite going to 74°N into the open Arctic, the data obtained during this study suggests no significant aluminum input from the entrainment of sea ice. This agrees with Middag (2009) and Moore (1981) and is inconsistent with the results of Measures (1999) and Giesbrecht (2013). The difference between results from the same region in different years is consistent with estimates of the relative importance of ice melt in this region, using barium as a tracer, between 2007 and 2009. In addition, the lowest sea ice extent, and therefore highest amount of sea ice melt ever recorded was in 2007, while in 2009 there was significantly less melting of new sea ice (National Snow and Ice Data Center). 0204060801001201401601802000 1 2 3 4depth(m)Dissolved Al(nmol/kg)S4 vs. KC20063   Concentrations are higher at depth than in the surface at all stations, and we observe little surface or subsurface river signal in the stations closer to the river. This is not surprising considering the distance from the river mouth and is consistent with the results found in 2007. Stations L1, L1.1, L1.5, and L2 have similar depth distributions, going from below 0.5 nmol/kg at the surface to 4 nmol/kg at around 1000 m.  The deep layers of the Arctic Ocean are comprised of Atlantic water running through the Fram Strait, and this water holds more aluminum than the corresponding Pacific water from the subsurface. Mixing between these two layers of water may explain the increase. At Station L2, where the deepest sampling occurred, there is significant scatter in the concentrations at the five deepest stations. We do not believe that this is due to contamination, but rather to the complicated water mass structure in this region.  The origin of this water is likely from the deep Eurasian Basin.   In conclusion, the input of dissolved aluminum in the Arctic seems to be derived primarily from shelf water and from the Atlantic Ocean, with periodic sea ice input. Scavenging is weak in the Arctic due to low phytoplankton growth, so subsurface concentrations are much higher than seen in the north Pacific.  64  Chapter 5: The Role of Aluminum as a Tracer of Dust Input  5.1  The Origins of Studying Dust Deposition Dust deposition was initially considered an unimportant part of the geochemistry of the ocean because of its lack of solubility (Hodge, Johnson & Goldberg 1978). However, studies in the 1980's showed a more soluble component to dust than previously believed (Prospero, Nees & Uematsu 1987). The study assessed the ratio of dissolved crustal components to particulate crustal components in rain samples in Florida and found a 3-5% dissolution rate. Another study confirmed this effect in seawater, showing a 1-5% dissolution rate (Maring and Duce 1987). To utilize this knowledge in an oceanic environment, a method to estimate dust deposition rates for open ocean areas was needed. A few methods were considered, most of them presenting major challenges. Collection of shipboard dust samples is limited by the inconsistency of dust input, and using sediment traps to estimate the vertical flux of terrigenous matter from the atmospheric as it leaves the surface ocean is complicated by a number of factors.  Turbulence and sheer can lead to undertrapping, and horizontal advection can lead to either over- or under-trapping, depending on the region. Alterations in the surface water can also affect estimates from traps (Buesseler 1991).  5.2  Why Study Aeolian Input? Aeolian input is a major pathway of entry for many elements into the ocean. The Earth is made up primarily of silicon, aluminum, and iron which are deposited onto the ocean by a combination of wind and precipitation. In addition, aerosols can provide ammonium, nitrate, and phosphate to areas that might otherwise be depleted of nutrients (Uematsu, Duce & Prospero 1985). Although 65  natural deposition will be the primary discussion of this study, there is a great deal of anthropogenic deposition of lead and other polluted elements from the atmosphere into the ocean, often from emissions of smelting and other industrial work.   Aeolian input is both seasonal and episodic (Tsunogai et al. 1985), a contrast to riverine input, which is variable but continuous throughout the year. Aeolian inputs are separated into two major categories; dry deposition, where deposition comes directly from the source carried by wind, and wet deposition, where rain scavenges dust particles in the air and precipitates them onto the surface of the ocean. The majority of soluble particles pass through several cycles of evaporation and condensation during dust cloud formation. The dust particles are exposed to low pH environments, which can break up mineral lattices (Junge 1963). The ocean is less acidic than rainwater, so dust directly deposited onto the ocean's surface is less soluble than dust scavenged by rain.  As a result, the past chemical history of the mineral lattice, i.e. how many times dust goes through this cycle, is important in determining dust solubility (Spokes, Jickells & Lim 1994). The rate of dust deposition may be changing over time due to anthropogenic SO2 and NOx lowering the pH of rainwater (Quinn et al. 1990).  5.2.1  Iron and Dust Deposition Deposition impacts the uptake of CO2 and thus the carbon cycle, which is mediated by the life and death of phytoplankton. Interest in iron's importance as part of the carbon cycle began with the discovery of a correlation between the high-nutrient low-chlorophyll regions of the ocean and iron deficiency (Martin and Fitzwater 1988). Low deposition to the north Pacific, equatorial Pacific, and the Southern Ocean results in iron deficiency and lower productivity in the ocean 66  than could be supported by the other required nutrients. More confirmation of the importance of iron in the carbon cycle came from analyzing Vostok ice cores, which showed that an increase in atmospheric deposition coincided with a decrease in atmospheric CO2 in the last four glacial cycles (Petit et al. 1999). Using both evidence from ice cores and from sedimentary records, the Fe deposition flux is estimated to be 5-50 times higher during glacial times than it is today.  As a result, CO2 was consumed at a much faster pace during glacial periods (Kumar et al. 1995, De Angelis, Barkov & Petrov 1987). Preindustrial era CO2 was around 280 ppm, while the CO2 at the last glacial maximum was around 180 ppm (Indermühle et al. 1999).  Iron deficiency slows the biological pump and inhibits phytoplankton productivity. It plays an important role in a variety of metabolic processes for phytoplankton due to its ability to catalyze many common reactions used for cell function and growth. It facilitates nitrate utilization and chlorophyll creation. There have been several fertilization experiments in high-nutrient low-chlorophyll regions to demonstrate iron limitation; the Southern Ocean, the equatorial Pacific, and the north Pacific all had increases in phytoplankton growth with the addition of iron in-situ (Boyd et al. 2000, Boyd et al. 1998, Coale et al. 1996). Iron fertilization has been proposed as a way to use geoengineering to mitigate climate change, although both the feasibility (Zeebe and Archer 2005) and the effectiveness (De Baar et al. 2005) have been brought into question.   Atmospheric input is an important source for trace metal nutrients in the ocean. Iron is primarily supplied by aeolian sources from land, so regional differences in aeolian transport and input can change biological processes in a region (Mahowald et al. 2005). For example, in the subtropical north Atlantic, the nutrients nitrate and phosphate are very low and the region is N-limited, so 67  iron obtained from the high aeolian dust input provided by the Sahara desert can be used to catalyze nitrogen fixation. This is a process that captures nitrogen from the atmosphere for use by phytoplankton (Brzezinski et al. 2011). This allows for more primary productivity than anticipated from the nutrient levels and leaves phosphate as the single limiting nutrient.  5.2.2  Aeolian Input’s Influence on the Past and Present Climate Is dust deposition a mechanism that changes the regime from interglacial to glacial or is it a consequence of the change? Recent evidence suggests dust deposition during glacial periods causes phytoplankton blooms to increase in the Southern Ocean (Martínez-Garcia et al. 2009). The drawdown from the Southern Ocean is believed to account for half of the total change in CO2 from interglacial to glacial regimes. This evidence suggests that dust deposition has a role in large-scale climate change and any comprehensive consideration of the effects of climate change on the future climate should consider its role.  The Northern Hemisphere dust load increased more dramatically than the Southern Hemisphere because most of the glaciers were located at latitudes above 50ºN. Antarctic glaciers extended north through the Andes in South America, but nowhere else, and since there is less land in the southern hemisphere, less could be eroded by glaciers. In addition, many of the enhanced non-glaciogenic dust sources north of the equator in Africa and Asia. However, due to the extent of iron limitation in the Southern Ocean, iron addition in that region was disproportionately important to the decrease in CO2 concentrations in the atmosphere (Ridgwell and Watson 2002).  68  There are two major causes for increased dust deposition during glacial periods. The first is glaciers accelerating dust mobilization by rapidly eroding any land it comes in contact with, called glaciogenic sources. Glaciers covered about one-third of the Earth's surface during glacial periods as compared to minimal land coverage in the current era, so glaciogenic sources are important in dust transport and erosion (Mahowald et al. 2006). Secondly, non-glaciogenic sources such as deserts increased in aridity during glacial periods due to a decrease in precipitation. The decrease in terrestrial life forms, particularly plants, causes wind to have less drag and causes a higher percentage of the surface area of dirt and eroded rocks to be exposed to wind and swept away. Stronger winds, which occur during glacial maxima due to increased local temperature gradients, also increase erosion (Werner et al. 2002, McGee, Broecker & Winckler 2010).  Aerosols are also important in other areas of science. They are used to study desertification of low vegetation regions such in the Sahel region of Africa and to measure changing ecosystem health in this region and other high population areas (Nicholson, Tucker & Ba 1998). As high population regions continue to be depleted of vegetation, an increase in dust activity is anticipated. Another effect of dust's presence in the atmosphere is its effect on radiative forcing and thus the global radiation budget (Tegen, Lacis & Fung 1996). Aerosols can scatter light and absorb radiation. The smaller the particles in the aerosols, the higher the albedo, which decreases radiation received globally. Changes in aerosol concentration over time may affect global wind patterns.  69  However, it is hard to couple aerosol input to human activity. In the last 40 years, the equatorial Atlantic near Africa has had dust input variability by a factor of four. This could be either human induced, naturally induced, or a combination of both, but it is difficult to distinguish between human-induced increases and natural increases because of the high variability (Prospero and Lamb 2003). Still, the disappearance of forests and potential desertification is likely accelerating the dust activity of the region. Human disturbance of soil causes increased erodibility, which increases dust circulation, but the magnitude of that change is poorly quantified and variable based on the study area (Tegen and Fung 1995, Prospero et al. 2002). Looking into the future, models indicate that the Earth might be less dusty due to anthropogenic effects depending on how aerosols are modulated (Mahowald and Luo 2003).  An increase in CO2 in the atmosphere may increase vegetation due to carbon dioxide fertilization and the increased wind drag may lower the amount of dust deposition. The change in heat distribution from the poles to the equator may lower the strength of wind, limiting dust transportation into the ocean.   5.3  Aluminum's Role in Determining Aeolian Input Iron is one of the most important and limited nutrients. However, due to iron uptake by phytoplankton, the source of the input cannot be accurately obtained by iron measurements. Iron is also recycled in the euphotic zone efficiently, further contributing to its complicated geochemistry. Measuring iron concentrations is important for understanding its biogeochemistry and the importance of primary productivity in a region, but it cannot inform us of the aeolian input or other input which provided the iron.  70  5.3.1  Aluminum as a Tracer of Aeolian Input Iron and aluminum are both derived terrestrially from the Earth’s crust, but aluminum is much less biologically active. Dissolved aluminum can be used as a proxy for determining the rate of dust entry into the ocean and can help us understand dust’s effect on the ocean biological pump (Measures et al. 2008). It has both a high crustal abundance (8.1%) that overcomes its low solubility in water (Vink and Measures 2001) and low variance in crustal abundance based on location. As a result, we can use it to trace where terrestrial input is important.  Dust solubility is hard to measure directly, so aerosol experiments at different pHs and in different regions can be used to predict the expected dissolved concentrations (Cornell et al. 2003).   Dissolved aluminum is scavenged readily by particles, resulting in aluminum’s short residence time in the surface ocean of about 5 years (Orians and Bruland 1985), which prevents aluminum from travelling laterally from the coast to the open ocean. This residence time can decrease in regions with a large amount of biological activity since it is first-order dependent on the particle concentration (Moran and Moore 1992). Otherwise, lateral transport could misrepresent the amount of aeolian input into these remote regions. In some specific locations, such as the equatorial north Atlantic, advection can distort the dust deposition estimates due to the large amount of dust deposited onto the region, but this only occurs in regions of high input and will not apply to the levels of aluminum present in the Pacific Ocean (Han et al. 2008)  The ratio of aluminum to iron is fairly constant throughout the Earth’s crust (Wedepohl 1995), but due to the different solubility rates of the two metals, the ratios are different in the ocean than on land (Han et al. 2008).  One factor in the changing ratios between aluminum and iron 71  deposition is the scavenging rate from the air, since aluminum and iron react differently to changes in pH in the rain. Sporadic, time-variable fluxes of dust makes a unified view of dust input from year to year impossible, further complicating the overall estimation of aeolian input (Duce et al. 1980). There is no consistent dust input, no consistent solubility of aluminum and iron, and no consistent ratios of aluminum and iron. Each of these factors must be calculated based on the specific region.  5.3.2  Quantifying Aeolian Input Northern Africa and Asia have the largest deserts in the world. The majority of the sources can be traced to draining of water from highlands, which erode the land below, creating dust (Prospero et al. 2002). Asia also has a high volume of dust come from dust storms. Specific areas have high levels of dust sources due to high winds and erosion; in North Africa, years of erosion from glacial eras increased the amount of mobile dust. In addition, due to North Africa’s increased land-use and low vegetation, the wind has little resistance. These two areas are responsible for much of the ocean’s dust supply.  Traditional methods of quantifying atmospheric flux have been limited to land-based measurements (Duce et al. 1991), but more recent studies have aimed to combine land-based measurements and ocean-based measurements by using seawater analysis and collection of rain droplets shipboard. Since dust generation is variable, it is hard to predict when a ship should be sent to collect dust samples. The soil must be dry in the region and large particles that can be swept away must be present (Mahowald et al. 2005).   72  Values for deposition are complicated by the variability of the fractional solubility of aluminum in seawater (Measures et al. 2010). In general, an inverse relationship between the amount deposited and the solubility was observed. Locations close to dust sources have lower solubility (Baker and Jickells 2006); the lowest solubility rates in the world are seen in the Sahara desert, where dust is the newest. The highest solubility rates are distant from sources because several cycles of acidification causes rainwater to erode and break down the minerals in solution (Vink et al. 2000).   Solubility is hard to determine because different areas have different mineral structures in particles and soils, so tests done in one location may not necessarily represent the spectrum of dissolution that can occur and incorrect assumptions for solubility can misrepresent the input significantly. One of the biggest sources of error in these calculations is the solubility component since it varies from 1-85% of the aluminum in the mineral, although the average is around 1.5-5% (Prospero, Nees & Uematsu 1987, Baker and Jickells 2006).  5.3.3  Models Using Dissolved Aluminum as a Tracer for Aeolian Input Models of aluminum deposition utilize limited dissolved aluminum measurements in order to study the relationship between aluminum and total dust input (Gehlen et al. 2003). A model called the Measurement of Aluminum for Dust Calculation in Oceanic Waters (MADCOW) was developed to use the dissolved aluminum measured from collected samples to calculate dust deposition (Brown 1996). Many simplifications were originally made in the model, including a consistent removal rate from the mixed layer of 5 years, a constant mixed layer of 30 m, and a consistent aluminum composition in dust of 8.1%. The solubility used in this study ranged from 73  1.5% to 5% solubility, which alone introduces an uncertainty factor of 3.3. We used this model to calculate dust deposition. This original model developed has been reasonably consistent with observational studies (Duce et al. 1991). The assumptions about solubility account for most of the variation in the north Atlantic and equatorial Pacific, but the modelling has unresolved results that are systematically too high in the non-equatorial Pacific.  Later studies have worked to refine the parameters used in the original model. The Biogeochemical Elemental Cycling (BEC) model includes phytoplankton and zooplankton abundance into the model, allowing for variable scavenging rates (Moore, Doney & Lindsay 2004). It also predicts the mixed layer depth with reasonable accuracy. A dust input field that simulates different sizes based on distance from the original source is used in this model, called the Dust Entrainment and Deposition (DEAD) input field (Zender, Bian & Newman 2003). In this new model, the Pacific dust deposition is split between overrepresentation north of 30°N and underrepresentation south of 30°N (Figure 5-1) (Han et al. 2008). Still, observations are within a factor of two compared to the model, which is an improvement over previous models. In 2012, subsequent work was done to integrate a diffusion model which addresses the size dependence on solubility and how the size affects diffusion (Han et al. 2012). Additional information will improve the quality and resolution of data to input into the models. Overall, dust deposition models need more dissolved aluminum data to better define the global dust cycle.  74   Figure 5-1: Modelled dust deposition using dissolved Al, in g·m-2·yr-1 (Han et al. 2008)  5.4  Calculating Dust Deposition in the North Pacific and the Western Arctic 5.4.1  Regional Dust Input Based on Models Earlier iterations of the MADCOW model used this equation (Measures and Brown 1996): SDAMyrLmG 13 2.01000  Where: )/(][)( 12kgnmolAlAyrmgdustGdiss  M = mixed layer depth (m) S = fractional solubility D = molar concentration of Al in dust (nmol/g) 0.2 = fractional scavenging rate  In this calculation, we assumed D = 3·106 nmol/g (based on the 8.1% fraction in the Earth’s crust) (Taylor 1964), the mixed layer depth to be 25 m based on plotting the density of the surface waters at each station, and fractional solubility is assumed to range from 1.5-5%. Both 1.5% and 5% are reported and further investigation of dust cycles would need to be performed to 75  figure out the values in this region. We also assume that the fractional scavenging rate of aluminum lost is 20%, i.e. a residence time of 5 years in the surface ocean.  Dissolved Al (nmol/kg) Station (Line P) Lower limit (5%) g·m-2·yr-1 Higher limit (1.5%) g·m-2·yr-1 1.65 P4 (2010) 0.055 0.183 1.74 P4 (2011) 0.058 0.193 3.57 P12 (2010) 0.119 0.397 1.01 P12 (2011) 0.034 0.112 0.92 P16 (2010) 0.031 0.102 0.91 P16 (2011) 0.030 0.101 0.62 P26 (2011) 0.021 0.069 Table 5-1: Dust deposition in the North Pacific based on surface dissolved aluminum concentrations  Aside from the anomalous concentrations seen at the surface of the eddy in 2010 at P12, the stations are reasonably consistent in predicted dust input. We can assume that the eddy activity may have changed the concentration of the water at P12 in 2010. The most recent biogeochemical models for the dust input into the North Pacific estimate the dust input to be  from 0.35 – 0.7 g·m-2·yr-1 (Han et al. 2012). This seems to be an overestimate based on the dissolved aluminum concentrations observed in this region. This model had no dissolved aluminum data incorporated from anywhere in the subarctic northeast Pacific, only extrapolated previously mentioned data from Hawaii and Asia. The only of the data points that shows aluminum deposition near this level is the one which we know to be influenced by eddy water. Previous models show signs of overestimation of dust input into very low dust input regions; this is explained in the south Pacific by the deepening of the mixed layer in the winter (Measures and Vink 2000), but in the north Pacific is likely just poor extrapolation.  76  Dissolved Al (nmol/kg) Station (Arctic) Lower limit(5%) g·m-2·yr-1 Higher limit (1.5%) g·m-2·yr-1 0.6 L1 0.020 0.067 0.6 L1.5 0.019 0.067 2.0 L1.1 0.067 0.222 0.6 L2 0.020 0.067 1.2 S4 0.040 0.133 Table 5-2: Dust deposition in the Western Arctic based on surface dissolved aluminum concentrations  These Arctic numbers show low dust input and we believe that it may even be overestimated. The scavenging rate assumed in the calculation is 20% of the total aluminum per year, but in the Arctic, little scavenging occurs due to low phytoplankton productivity. By factoring in a different scavenging rate, a calculation of the dust input into the Arctic can be done. However, shelf input, sedimentary input, and possible sea ice input at the surface are likely to be more important inputs to the Arctic Ocean due to sea ice cover obstructing dust input.   5.4.2  Limitations of the Dust Model In our dust modelling, we have assumed the dissolution rate was 5% maximum, but in some regions the indication is that that solubility of aluminum in seawater could be as high as 86% (Prospero, Nees & Uematsu 1987, Gehlen et al. 2003). More work needs to be done to quantify this number for each specific region, but it is challenging to obtain. Utilization of a clean rainwater collection system is the easiest way to find the rate of dissolution, but the system is hard to keep clean on a ship made of metal. In addition, the scavenging rate is assumed to remove 20% of deposited dissolved aluminum per year (Orians and Bruland 1986).   77  The more recent modelling techniques (BEC) account for aluminum taken up by siliceous shells, the remineralization of aluminum containing species in water, and advection of dissolved aluminum from other locations. The equation below is part of their model:   SCAVPRODCIRCREMINDISSdtAld diss ][ This model incorporates more realistic scavenging numbers that vary by location (Figure 5-2), but these seem to only be well-defined for the Atlantic Ocean and are outright incorrect in the Arctic Ocean because the model did not incorporate phytoplankton growth under sea ice. The figures for residence time are believed to be overestimated, but the factor in which they are overestimated is unknown.   Figure 5-2: Modelled residence time of the surface ocean using dissolved Al (Han et al. 2008).   5.5  Closing Remarks about Dust Deposition Dust deposition, while important to understanding the broader geochemistry of the ocean, has not been very well defined in locations outside of the Atlantic Ocean. Surface aluminum 78  concentration studies allow us to calculate a rate of dust deposition in a particular region, but measurements are sparingly done in the Pacific. We have used previous calculations from models to find the dust deposition rate in the North Pacific. The result indicates that models overestimate the dissolved aluminum and dust deposition rate based on this study.  We have also calculated dust deposition in the Arctic and concluded that the method is not useful in this region based on the given assumptions and is not a reasonable assessment of the input of dissolved aluminum. A study of shelf, sediment, or sea ice input would be more appropriate to find the primary sources of dissolved aluminum in this basin.   79  Chapter 6: Conclusion In this study we used flow injection analysis to obtained data from two regions for dissolved aluminum; in the North Pacific, we examined a transect over two different years, documenting a difference between the eddy-influenced year 2010 and the non-eddy year 2011. In addition, we measured some of the lowest steady state concentrations of aluminum ever measured in the open ocean of the Northeast Pacific and compared those results with data from the 1980’s, which showed similar patterns of dissolved aluminum concentrations. These profiles are highest in the surface and decrease with depth. Lastly, we observe, both off the coast of British Columbia in 2011 and off the coast of California in 1985, a subsurface maximum in the dissolved aluminum profile, likely from shelf water advecting into both regions. Aluminum is higher concentration off the coast of British Columbia due to the heavy scavenging in the productive California Current region.   In the Arctic we find profiles which generally increase with depth, indicating a distribution of dissolved aluminum independent of input at the surface. This contradicted previous work in the Arctic two years prior, where surface maxima were observed. During the historic 2007 melting season, aluminum-containing particles entrained in the ice may have released into the surface water, but in 2009 the sea ice melt did not reach the Beaufort Sea. This surface signal is inconsistent, observed in some studies but not others, and could possibly be due to something besides sea ice or could be the change in weather patterns. We believe that shelf input, Atlantic Deep Water, and water from across the Makarov Basin are the primary influences of Arctic dissolved aluminum distributions. Riverine input was not shown to be a significant factor close 80  to the Mackenzie River, further confirming the low impact of fluvial input of aluminum into the ocean.  Finally, we used the surface dissolved aluminum concentration data together to try to estimate dust input into the northeastern Pacific and the Arctic Ocean. The northeastern Pacific calculations show an inconsistency between models and the actual dissolved concentrations; possibly due to underestimating the solubility of the particles that reach the region or overestimation of the influence of Asiatic dust on the dust deposition. The Arctic data was compiled but is flawed for a variety of reasons; it is unclear if dust is the source of dissolved aluminum in the surface waters of the Arctic and the scavenging rate is highly overestimated in early models and highly underestimated in later models.  Shortcomings of the work include contamination reducing the amount of stations analyzed, sensitivity problems during later analyses, and that the ability to run duplicates was impossible for a few sets of samples because we had to run them multiple times due to contamination issues. Multiple years of Arctic Ocean sampling would have also been helpful to shed more light onto the sea ice problem.   Future work would be to allow for more spatial coverage of the Pacific Ocean now that methods have been developed to analyze low enough concentrations of dissolved aluminum. This would include work in between Japan and British Columbia as well as in the South Pacific Ocean. Learning about how dust moves through the world is important in understanding high-nutrient low-chlorophyll regions and predicting the impact climate change will have on the iron 81  distribution of the world’s oceans. Quantifying dust input via clean rainwater collection systems would allow us to define the solubility ratio which would give a better idea of the input rate.  In addition, more Arctic Ocean data needs to be collected; in a time of climate change, understanding how iron and dust interacts with the basin is important in understanding how CO2 uptake will be affected with the decreased sea ice coverage since the Arctic is a potential sink for CO2 in the future. Understanding the mechanism for which iron is brought into the basin will show the potential of this sink as a long-term or short-term climate change mitigation factor. Year-to-year variability, as observed in this study, is likely to continue and should be investigated. Finding accurate scavenging and production rates for the region will help us determine the sinks and sources of aluminum in this region.  We have added key dissolved aluminum analyses to a well-studied transect in the Pacific which had previously not been incorporated into any dust models and we found results in the Arctic which were different than expected. A more comprehensive study of these areas will need to be done to answer more questions.   82  References Aagaard, K., and Coachman, L. (1975). Toward an ice-free Arctic Ocean. Eos, Transactions    American Geophysical Union 56, 484-486. Aagaard, K., Coachman, L., and Carmack, E. (1981). On the halocline of the Arctic Ocean. Deep Sea Research Part A.Oceanographic Research Papers 28, 529-545. Aguilar-Islas, A.M., Rember, R., Nishino, S., Kikuchi, T., and Itoh, M. (2013). Partitioning and lateral transport of iron to the Canada Basin. Polar Science 7, 82-99. Anschutz, P., Zhong, S., Sundby, B., Mucci, A., and Gobeil, C. (1998). Burial efficiency of phosphorus and the geochemistry of iron in continental margin sediments. Limnology and Oceanography 43, 53-64. Baker, A.R., and Jickells, T.D. (2006). Mineral particle size as a control on aerosol iron solubil-ity. Geophysical Research Letters 33, L17608. Boyd, P.W., Watson, A.J., Law, C.S., Abraham, E.R., Trull, T., Murdoch, R., Bakker, D.C., Bowie, A.R., Buesseler, K., and Chang, H. (2000). A mesoscale phytoplankton bloom in the polar Southern Ocean stimulated by iron fertilization. Nature 407, 695-702. Boyd, P., Berges, J.A., and Harrison, P.J. (1998). In vitro iron enrichment experiments at iron-rich and -poor sites in the NE subarctic Pacific. Journal of experimental marine biology and ecology 227, 133-151. Boyd, P.W., Law, C.S., Wong, C.S., Nojiri, Y., Tsuda, A., Levasseur, M., Takeda, S., Rivkin, R., Harrison, P.J., Strzepek, R., Gower, J., McKay, R.M., Abraham, E., Arychuk, M., Barwell-Clarke, J., Crawford, W., Crawford, D., Hale, M., Harada, K., Johnson, K., Kiyosawa, H., Kudo, I., Marchetti, A., Miller, W., Needoba, J., Nishioka, J., Ogawa, H., Page, J., Robert, M., Saito, H., Sastri, A., Sherry, N., Soutar, T., Sutherland, N., Taira, Y., Whitney, F., Wong, S.K.E., and Yoshimura, T. (2004). The decline and fate of an iron-induced subarctic phytoplankton bloom. Nature 428, 549-553. Boyd, P.W., Wong, C.S., Merrill, J., Whitney, F., Snow, J., Harrison, P.J., and 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. Brown, E. (1996). Estimating Dust Input to the Atlantic Ocean Using Surface Water Aluminium Concentrations. In ' The Impact of Desert Dust Across the Mediterranean '. pp. 301-311. (Springer:. 83  Brown, M. T. (2009). Aluminum and iron distributions in the Central and Western North Pacific Ocean: results from the 2002 Intergovernmental Oceanographic Commission Cruise. Uni-versity of Hawaii at Manoa.  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., and Bruland, K.W. (2008). An improved flow-injection analysis method for the determination of dissolved aluminum in seawater. Limnology and Oceanography-Methods 6. Brown, M.T., Lippiatt, S.M., and Bruland, K.W. (2010). Dissolved aluminum, particulate alumi-num, and silicic acid in northern Gulf of Alaska coastal waters: Glacial/riverine inputs and extreme reactivity. Marine Chemistry 122, 160-175. Bruland, K.W. (1983). Trace elements in seawater.  Bruland, K.W., and Lohan, M.C. (2003). 6.02 - Controls of Trace Metals in Seawater. In ' Trea-tise on Geochemistry '. (Eds Editors-in-Chief:  Heinrich D. Holland and Karl K. Turekian.) pp. 23-47. (Pergamon:Oxford). Bruland, K.W. (1980). Oceanographic distributions of cadmium, zinc, nickel, and copper in the North Pacific. Earth and Planetary Science Letters 47, 176-198. Bruland, K.W., Orians, K.J., and Cowen, J.P. (1994). Reactive trace metals in the stratified cen-tral North Pacific. Geochimica et Cosmochimica Acta 58, 3171-3182. Bruland, K., Franks, R., Knauer, G., and Martin, J. (1979). Sampling and Analytical Methods for the Determination of Copper, Cadmium, Zinc, and Nickel at the Nanogram Per Liter Level in Sea-Water. Analytica Chimica Acta 105, 233-245. Brzezinski, M.A., Krause, J.W., Church, M.J., Karl, D.M., Li, B., Jones, J.L., and Updyke, B. (2011). The annual silica cycle of the North Pacific subtropical gyre. Deep-Sea Research Part I-Oceanographic Research Papers 58, 988-1001. Buck, C., Landing, W., Resing, J., and Lebon, G. (2006). Aerosol iron and aluminum solubility in the northwest Pacific Ocean: Results from the 2002 IOC cruise. Geochemistry Geophys-ics Geosystems 7, Q04M07. Buesseler, K.O. (1991). Do upper-ocean sediment traps provide an accurate record of particle flux? Nature 353, 420-423. Burn, D.H. (2008). Climatic influences on streamflow timing in the headwaters of the Mackenzie River Basin. Journal of Hydrology 352, 225-238. 84  Carmack, E.C., and Macdonald, R.W. (2002). Oceanography of the Canadian Shelf of the Beau-fort Sea: a setting for marine life. Arctic , 29-45. Carmack, E.C., Macdonald, R.W., and Jasper, S. (2004). Phytoplankton productivity on the Ca-nadian Shelf of the Beaufort Sea. Marine ecology.Progress series 277, 37-50. Carmack, E., and Chapman, D.C. (2003). Wind‐driven shelf/basin exchange on an Arctic shelf: The joint roles of ice cover extent and shelf‐break bathymetry. Geophysical Research Let-ters 30. Chou, L., and Wollast, R. (1997). Biogeochemical behavior and mass balance of dissolved alu-minum in the western Mediterranean Sea. Deep Sea Research Part II: Topical Studies in Oceanography 44, 741-768. Coachman, L.K. (1969). Physical oceanography in the Arctic Ocean: 1968. Arctic 22, 214-224. Coale, K.H., Fitzwater, S.E., Gordon, R.M., Johnson, K.S., and Barber, R.T. (1996). Control of community growth and export production by upwelled iron in the equatorial Pacific Ocean. , 621-624. Cornell, S., Jickells, T., Cape, J., Rowland, A., and Duce, R. (2003). Organic nitrogen deposition on land and coastal environments: a review of methods and data. Atmospheric Environment 37, 2173-2191. Crawford, W. (2002). Physical characteristics of Haida eddies. Journal of Oceanography 58, 703-713. Crawford, W.R., Brickley, P.J., Peterson, T.D., and Thomas, A.C. (2005). Impact of Haida Ed-dies on chlorophyll distribution in the Eastern Gulf of Alaska. Deep Sea Research Part II: Topical Studies in Oceanography 52, 975-989. Crawford, W., Galbraith, J., and Bolingbroke, N. (2007). Line P ocean temperature and salinity, 1956-2005. Progress in Oceanography 75, 161-178. D'Asaro, E.A. (1988). Generation of submesoscale vortices: A new mechanism. Journal of Geo-physical Research 93, 6685-6693. De Angelis, M., Barkov, N., and Petrov, V. (1987). Aerosol concentrations over the last climatic cycle (160 kyr) from an Antarctic ice core. Nature 325, 318-321. De Baar, H.J., Boyd, P.W., Coale, K.H., Landry, M.R., Tsuda, A., Assmy, P., Bakker, D.C., Bo-zec, Y., Barber, R.T., and Brzezinski, M.A. (2005). Synthesis of iron fertilization experi-ments: From the Iron Age in the Age of Enlightenment. Journal of Geophysical Research: Oceans (1978–2012) 110. 85  Dittmar, T., and Kattner, G. (2003). The biogeochemistry of the river and shelf ecosystem of the Arctic Ocean: A review. Marine Chemistry 83, 103-120. Dodimead, A. (1963). Review of the oceanography of the Subarctic Pacific Region. Int.North Pac.Fish.Comm.Bull. 13, 1-195. Duce, R., Liss, P., Merrill, J., Atlas, E., Buat-Menard, P., Hicks, B., Miller, J., Prospero, J., Arimoto, R., and Church, T. (1991). The atmospheric input of trace species to the world ocean. Global Biogeochemical Cycles 5, 193-259. Duce, R., Unni, C., Ray, B., Prospero, J., and Merrill, J. (1980). Long-range atmospheric transport of soil dust from Asia to the tropical North Pacific- Temporal variability. Science 209, 1522-1524. Emerson, S., and Huested, S. (1991). Ocean anoxia and the concentrations of molybdenum and vanadium in seawater. Marine Chemistry 34, 177-196. Emmerton, C.A., Lesack, L.F., and Vincent, W.F. (2008). Mackenzie River nutrient delivery to the Arctic Ocean and effects of the Mackenzie Delta during open water conditions. Global Biogeochemical Cycles 22, GB1024. Emmerton, C.A., Lesack, L.F., and Vincent, W.F. (2008). Nutrient and organic matter patterns across the Mackenzie River, estuary and shelf during the seasonal recession of sea-ice. Jour-nal of Marine Systems 74, 741-755. Gehlen, M., Heinze, C., Maier-Reimer, E., and Measures, C.I. (2003). Coupled Al-Si geochemis-try in an ocean general circulation model: A tool for the validation of oceanic dust deposi-tion fields? Global Biogeochemical Cycles 17, 1028. Giesbrecht, T., Sim, N., Orians, K.J., and 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 . Gran, H. (1931). On the conditions for the production of plankton in the sea. Rapp.Proc.Verb.Cons.Int.Explor.Mer 75, 37-46. Greene, R., Falkowskill, P., Chisholm, S., Hoge, F., and Swift, R. (1994). Testing the iron hy-pothesis in ecosystems of the equatorial Pacific Ocean. Nature 371, 8. 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., and Lockwood, D. (2010). Volcanic ash fuels anomalous plankton bloom in subarctic northeast Pacific. Geophysical Research Let-ters 37, L19604. 86  Han, Q., Moore, J.K., Zender, C., Measures, C., and Hydes, D. (2008). Constraining oceanic dust deposition using surface ocean dissolved Al. Global Biogeochemical Cycles 22, GB2003. Han, Q., Zender, C.S., Moore, J.K., Buck, C.S., Chen, Y., Johansen, A., and Measures, C.I. (2012). Global estimates of mineral dust aerosol iron and aluminum solubility that account for particle size using diffusion-controlled and surface-area-controlled approximations. Global Biogeochemical Cycles 26, GB2038. Hansen, E. (2005). Use of flow injection and sequential injection analysis schemes for the deter-mination of trace-level concentrations of metals in complex matrices by ETAAS and ICPMS. Journal of Environmental Science and Health Part A-Toxic/hazardous Substances & Environmental Engineering 40, 1507-1524. Harris, S.L., Varela, D.E., Whitney, F.W., and Harrison, P.J. (2009). Nutrient and phytoplankton dynamics off the west coast of Vancouver Island during the 1997/98 ENSO event. Deep-Sea Research Part Ii-Topical Studies in Oceanography 56, 2487-2502. 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., and SCOR Working Grp (2007). GEOTRACES - An interna-tional study of the global marine biogeochemical cycles of trace elements and their isotopes. Chemie Der Erde-Geochemistry 67. Hickey, B.M. (1979). The California Current system—hypotheses and facts. Progress in Ocean-ography 8, 191-279. Hodge, V., Johnson, S.R., and Goldberg, E.D. (1978). Influence of atmospherically transported aerosols on surface ocean water composition. Geochem.J 12, 7-20. Howard, A.G., Coxhead, A.J., Potter, I.A., and Watt, A.P. (1986). Determination of Dissolved Aluminum by the Micelle-Enhanced Fluorescence of its Lumogallion Complex. Analyst 111, 1379-1382. Hsiao, S. I. (1976). 'Biological productivity of the southern Beaufort Sea: phytoplankton and sea-weed studies.' (Beaufort Sea Project:. Hunter, C., Gordon, R., Fitzwater, S., and Coale, K. (1996). A rosette system for the collection of trace metal clean seawater. Limnology and Oceanography 41, 1367-1372. Huyer, A. (1983). Coastal upwelling in the California Current system. Progress in Oceanography 12, 259-284. 87  Hydes, D.J. (1989). Seasonal-Variation in Dissolved Aluminum Concentrations in Coastal Wa-ters and Biological Limitation of the Export of the Riverine Input of Aluminum to the Deep-Sea. Continental Shelf Research 9, 919-929. Hydes, D.J. (1979). Aluminum in Seawater - Control by Inorganic Processes. Science 205, 1260-1262. Hydes, D., and Liss, P. (1977). Behavior of Dissolved Aluminum in Estuarine and Coastal Wa-ters. Estuarine and Coastal Marine Science 5, 755-769. Indermühle, A., Stauffer, B., Stocker, T.F., Raynaud, D., and Barnola, J. (1999). Early Holocene atmospheric CO2 concentrations. Science 286, 1815-1815. Jickells, T.D., An, Z.S., Andersen, K.K., Baker, A.R., Bergametti, G., Brooks, N., Cao, J.J., Boyd, P.W., Duce, R.A., Hunter, K.A., Kawahata, H., Kubilay, N., laRoche, J., Liss, P.S., Mahowald, N., Prospero, J.M., Ridgwell, A.J., Tegen, I., and Torres, R. (2005). Global iron connections between desert dust, ocean biogeochemistry, and climate. Science 308, 67. Johnson, K., Miller, L., Sutherland, N., and Wong, C. (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., Elrod, V., Fitzwater, S., Plant, J., Chavez, F., Tanner, S., Gordon, R., Westphal, D., Perry, K., Wu, J., and Karl, D. (2003). Surface ocean-lower atmosphere interactions in the Northeast Pacific Ocean Gyre: Aerosols, iron, and the ecosystem response. Global Biogeo-chemical Cycles 17, 1063. Jones, E., Rudels, B., and Anderson, L. (1995). Deep waters of the Arctic Ocean: origins and cir-culation. Deep Sea Research Part I: Oceanographic Research Papers 42, 737-760. Jorgenson, M., and Brown, J. (2005). Classification of the Alaskan Beaufort Sea Coast and esti-mation of carbon and sediment inputs from coastal erosion. Geo-Marine Letters 25, 69-80. Junge, C.E. 1963, Air chemistry and radioactivity, New York and London. Academic Press. Klunder, M.B., Bauch, D., Laan, P., Baar, H., Heuven, S., and Ober, S. (2012). Dissolved iron in the Arctic shelf seas and surface waters of the central Arctic Ocean: Impact of Arctic river water and ice‐melt. Journal of Geophysical Research: Oceans (1978–2012) 117, C01027. Kumar, N., Anderson, R., Mortlock, R., Froelich, P., Kubik, P., Dittrich-Hannen, B., and Suter, M. (1995). Increased biological productivity and export production in the glacial Southern Ocean. Nature 378, 675-680. 88  Lam, P.J., Bishop, J.K., Henning, C.C., Marcus, M.A., Waychunas, G.A., and Fung, I.Y. (2006). Wintertime phytoplankton bloom in the subarctic Pacific supported by continental margin iron. Global Biogeochemical Cycles 20, GB1006. Lohan, M.C., Statham, P.J., and Crawford, D.W. (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, PII S0967-0645(02)00215-1. Macdonald, R.W., Harner, T., and Fyfe, J. (2005). Recent climate change in the Arctic and its impact on contaminant pathways and interpretation of temporal trend data. Science of the total environment 342, 5-86. Macdonald, R., Barrie, L., Bidleman, T., Diamond, M., Gregor, D., Semkin, R., Strachan, W., Li, Y., Wania, F., and Alaee, M. (2000). Contaminants in the Canadian Arctic: 5 years of pro-gress in understanding sources, occurrence and pathways. Science of the Total Environment 254, 93-234. Macdonald, R., Carmack, E., McLaughlin, F., Iseki, K., Macdonald, D., and O'Brien, M. (1989). Composition and modification of water masses in the Mackenzie Shelf Estuary. Journal of Geophysical Research 94, 18057-18,070. Macdonald, R., Wong, C., and Erickson, P. (1987). The distribution of nutrients in the southeast-ern Beaufort Sea: Implications for water circulation and primary production. Journal of Ge-ophysical Research 92, 2939-2952. Mackenzie, F.T., Stoffyn, M., and Wollast, R. (1978). Aluminum in Seawater - Control by Bio-logical-Activity. Science 199, 680-682. Mackin, J.E., and Aller, R.C. (1984). Diagenesis of Dissolved Aluminum in Organic-Rich Estua-rine Sediments. Geochimica et Cosmochimica Acta 48, 299-313. Mackin, J.E., and Aller, R.C. (1984). Processes Affecting the Behavior of Dissolved Aluminum in Estuarine Waters. Marine Chemistry 14, 213-232. Mahowald, N.M., Baker, A.R., Bergametti, G., Brooks, N., Duce, R.A., Jickells, T.D., Kubilay, N., Prospero, J.M., and Tegen, I. (2005). Atmospheric global dust cycle and iron inputs to the ocean. Global Biogeochemical Cycles 19, GB4025. Mahowald, N.M., and Luo, C. (2003). A less dusty future? Geophysical Research Letters 30, 1903. Mahowald, N.M., Muhs, D.R., Levis, S., Rasch, P.J., Yoshioka, M., Zender, C.S., and Luo, C. (2006). Change in atmospheric mineral aerosols in response to climate: Last glacial period, preindustrial, modern, and doubled carbon dioxide climates. USGS Staff--Published Re-search , 166. 89  Mahowald, N., Muhs, D., Levis, S., Rasch, P., Yoshioka, M., Zender, C., and Luo, C. (2006). Change in atmospheric mineral aerosols in response to climate: Last glacial period, prein-dustrial, modern, and doubled carbon dioxide climates. Journal of Geophysical Research-Atmospheres 111, D10202. Manley, T., and Hunkins, K. (1985). Mesoscale eddies of the Arctic Ocean. Journal of Geophysi-cal Research 90, 4911-4930. Maring, H.B., and Duce, R.A. (1987). The Impact of Atmospheric Aerosols on Trace-Metal Chemistry in Open Ocean Surface Seawater .1. Aluminum. Earth and Planetary Science Letters 84, 381-392. Martin, J.H., and Fitzwater, S.E. (1988). Iron deficiency limits phytoplankton growth in the north-east Pacific subarctic. Paleoceanography 5, 1-13. Martin, J.H., Fitzwater, S.E., Gordon, R.M., Hunter, C.N., and Tanner, S.J. (1993). Iron, Primary Production and Carbon Nitrogen Flux Studies during the Jgofs North-Atlantic Bloom Ex-periment. Deep-Sea Research Part Ii-Topical Studies in Oceanography 40, 113-134. Martin, J.H., Gordon, R.M., Fitzwater, S., and Broenkow, W.W. (1989). Vertex - Phytoplankton Iron Studies in the Gulf of Alaska. Deep-Sea Research Part A-Oceanographic Research Pa-pers 36. Martínez-Garcia, A., Rosell-Melé, A., Geibert, W., Gersonde, R., Masqué, P., Gaspari, V., and Barbante, C. (2009). Links between iron supply, marine productivity, sea surface tempera-ture, and CO2 over the last 1.1 Ma. Paleoceanography 24, PA1207. Mathis, J. T. (2006). Cycling and fate of carbon in the western Arctic Ocean during the Shelf-Basin Interactions Project. University of Miami.  McGee, D., Broecker, W.S., and Winckler, G. (2010). Gustiness: The driver of glacial dustiness? Quaternary Science Reviews 29, 2340-2350. McLaughlin, F.A., Carmack, E.C., Macdonald, R.W., and Bishop, J.K. (1996). Physical and geo-chemical properties across the Atlantic/Pacific water mass front in the southern Canadian Basin. Journal of Geophysical Research 101, 1183-1197. Measures, C., and Edmond, J. (1989). Shipboard Determination of Aluminum in Seawater at the Nanomolar Level by Electron-Capture Detection Gas-Chromatography. Analytical Chemis-try 61, 544-547. Measures, C.I. (1999). The role of entrained sediments in sea ice in the distribution of aluminium and iron in the surface waters of the Arctic Ocean. Marine Chemistry 68, 59-70. 90  Measures, C.I., and Brown, E. (1996).    Estimating Dust Input to the Atlantic Ocean Using Sur-face Water Aluminum Concentrations. The Impact of Desert Dust Across the Mediterranean 11, 301-311. Measures, C.I., Brown, M.T., and Vink, S. (2005). Dust deposition to the surface waters of the western and central North Pacific inferred from surface water dissolved aluminum concen-trations. Geochemistry Geophysics Geosystems 6, Q09M03. 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.I., Landing, W.M., Brown, M.T., and Buck, C.S. (2008). High-resolution Al and Fe data from the Atlantic Ocean CLIVAR-CO(2) repeat hydrography A16N transect: Extensive linkages between atmospheric dust and upper ocean geochemistry. Global Biogeochemical Cycles 22, GB1005. Measures, C.I., Sato, T., Vink, S., Howell, S., and Li, Y.H. (2010). The fractional solubility of aluminium from mineral aerosols collected in Hawaii and implications for atmospheric dep-osition of biogeochemically important trace elements. Marine Chemistry 120, 144-153. Measures, C.I., and Vink, S. (2000). On the use of dissolved aluminum in surface waters to esti-mate dust deposition to the ocean. Global Biogeochemical Cycles 14, 317-327. Measures, C., Landing, W., Brown, M., and Clifton, S. (2008). A commercially available rosette system for trace metal–clean sampling. Limnol.Oceanogr.: Methods 6, 384-394. Meyers, S.D., and Basu, S. (1999). Eddies in the eastern Gulf of Alaska from TOPEX/POSEI-DON altimetry. Journal of geophysical research 104, 13333-13,343. Middag, R., de Baar, H.J.W., Laan, P., and Bakker, K. (2009). Dissolved aluminium and the sili-con cycle in the Arctic Ocean. Marine Chemistry 115, 176-195. Middag, R., van Slooten, C., de Baar, H.J.W., and Laan, P. (2011). Dissolved aluminium in the Southern Ocean. Deep-Sea Research Part Ii-Topical Studies in Oceanography 58, 2647-2660. Moore, J.K., Doney, S.C., and Lindsay, K. (2004). Upper ocean ecosystem dynamics and iron cycling in a global three-dimensional model. 18, GB4028. Moore, R.M. (1981). Oceanographic Distributions of Zinc, Cadmium, Copper and Aluminum in Waters of the Central Arctic. Geochimica et Cosmochimica Acta 45, 2475-2482. Moore, R.M., and Millward, G.E. (1984). Dissolved-Particulate Interactions of Aluminum in Ocean Waters. Geochimica et Cosmochimica Acta 48, 235-241. 91  Moran, S.B., and Moore, R.M. (1992). Kinetics of the Removal of Dissolved Aluminum by Dia-toms in Seawater - a Comparison with Thorium. Geochimica et Cosmochimica Acta 56, 3365-3374. Moritz, R.E., Bitz, C.M., and Steig, E.J. (2002). Dynamics of recent climate change in the Arctic. Science 297, 1497-1502. Morris, A., Howland, R., and Bale, A. (1986). Dissolved aluminium in the Tamar Estuary, south-west England. Geochimica et Cosmochimica Acta 50, 189-197. Newton, J., Aagaard, K., and Coachman, L. K. (1974). Baroclinic eddies in the Arctic Ocean. In 'Deep Sea Research and Oceanographic Abstracts'. p. 707. Elsevier:  . Nicholson, S., Tucker, C., and Ba, M. (1998). Desertification, drought, and surface vegetation: An example from the West African Sahel. Bulletin of the American Meteorological Society 79, 815-829. Nishikawa, Y., Hiraki, K., Morishige, K., and Shigematsu, T. (1967). Fluorophotometric deter-mination of aluminum and gallium with lumogallion. Japan Analyst 16, 692-697. Nolting, R., and de Jong, J. (1994). Sampling and Analytical Methods for the Determination of Trace-Metals in Surface Seawater. International journal of environmental analytical chemis-try 57, 189-196. Null, J. 2012, El Nino and La Nina Years and Intensities, ggweather.com/enso/oni.html edn. Orians, K.J., and Bruland, K.W. (1988). The Marine Geochemistry of Dissolved Gallium - a Comparison with Dissolved Aluminum. Geochimica et Cosmochimica Acta 52. 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. Na-ture 316. Parrish, R.H., Nelson, C.S., and Bakun, A. (1981). Transport mechanisms and reproductive suc-cess of fishes in the California Current. Biological Oceanography 1, 175-203. Parsons, T., Webb, D., Dovey, H., Haigh, R., Lawrence, M., and Hopky, G. (1988). Production studies in the Mackenzie River—Beaufort Sea estuary. Polar Biology 8, 235-239. Peterson, T.D., Whitney, F.A., and Harrison, P.J. (2005). Macronutrient dynamics in an anticy-clonic mesoscale eddy in the Gulf of Alaska. Deep-Sea Research Part Ii-Topical Studies in Oceanography 52, 909-932. 92  Petit, J.R., Jouzel, J., Raynaud, D., Barkov, N., Barnola, J., Basile, I., Bender, M., Chappellaz, J., Davis, M., and Delaygue, G. (1999). Climate and atmospheric history of the past 420,000 years from the Vostok ice core, Antarctica. Nature 399, 429-436. Price, N., and Morel, F. (1990). Cadmium and cobalt substitution for zinc in a marine diatom. Nature 344, 658-660. Prospero, J.M., Ginoux, P., Torres, O., Nicholson, S.E., and Gill, T.E. (2002). Environmental characterization of global sources of atmospheric soil dust identified with the Nimbus 7 To-tal Ozone Mapping Spectrometer (TOMS) absorbing aerosol product. Reviews of Geophys-ics 40, 2-1-2-31. Prospero, J.M., and Lamb, P.J. (2003). African droughts and dust transport to the Caribbean: Cli-mate change implications. Science 302, 1024-1027. Prospero, J.M., Nees, R.T., and Uematsu, M. (1987). Deposition Rate of Particulate and Dis-solved Aluminum Derived from Saharan Dust in Precipitation at Miami, Florida. Journal of Geophysical Research-Atmospheres 92, 14723-13731. Quinn, P.K., Bates, T.S., Johnson, J.E., Covert, D.S., and Charlson, R.J. (1990). Interactions be-tween the sulfur and reduced nitrogen cycles over the central Pacific Ocean. Journal of Geo-physical Research 95, 16405-16416. Rahn, K.A. (1981). Relative importances of North America and Eurasia as sources of Arctic aer-osol. Atmospheric Environment (1967) 15, 1447-1455. Ren, J.L., Zhang, J., Luo, J.Q., Pei, X.K., and Jiang, Z.X. (2001). Improved fluorimetric determi-nation of dissolved aluminium by micelle-enhanced lumogallion complex in natural waters. Analyst 126, 698-702. Resing, J.A., and Measures, C. (1994). Fluorometric determination of Al in seawater by flow in-jection analysis with in-line preconcentration. Analytical Chemistry 66, 4105-4111. Ridgwell, A., and Watson, A.J. (2002). Feedback between aeolian dust, climate, and atmospheric CO2 in glacial time. Paleoceanography 17, 1059. Rodionov, S., Overland, J., and Bond, N. (2005). The Aleutian Low and Winter Climatic Condi-tions in the Bering Sea. Part I: Classification*. Journal of Climate 18, 160-177. Roeske, T., van der Loeff, M.R., Middag, R., and Bakker, K. (2012). Deep water circulation and composition in the Arctic Ocean by dissolved barium, aluminium and silicate. Marine Chemistry 132, 56-67. Rothrock, D.A., Yu, Y., and Maykut, G.A. (1999). Thinning of the Arctic sea-ice cover. Geo-physical Research Letters 26, 3469-3472. 93  Rouse, W.R., Douglas, M.S., Hecky, R.E., Hershey, A.E., Kling, G.W., Lesack, L., Marsh, P., McDonald, M., Nicholson, B.J., and Roulet, N.T. (1997). Effects of climate change on the freshwaters of arctic and subarctic North America. Hydrological Processes 11, 873-902. Schiermeier, Q. (2004). Iron seeding creates fleeting carbon sink in Southern Ocean. Nature 428, 788. Sohrin, Y., Isshiki, K., Kuwamoto, T., and Nakayama, E. (1987). Tungsten in North Pacific Wa-ters. Marine Chemistry 22, 95-103. Spokes, L.J., Jickells, T.D., and Lim, B. (1994). Solubilisation of aerosol trace metals by cloud processing: A laboratory study. Geochimica et Cosmochimica Acta 58, 3281-3287. Stoffyn, M., and Mackenzie, F.T. (1982). Fate of dissolved aluminum in the oceans. Marine Chemistry 11, 105-127. Sturgeon, R.E., Berman, S.S., Desaulniers, J., Mykytiuk, A.P., McLaren, J., and Russell, D.S. (1980). Comparison of Methods for the Determination of Trace-Elements in Seawater. Ana-lytical Chemistry 52, 1585-1588. Taylor, S. (1964). Abundance of chemical elements in the continental crust: a new table. Geo-chimica et Cosmochimica Acta 28, 1273-1285. Tegen, I., and Fung, I. (1995). Contribution to the atmospheric mineral aerosol load from land surface modification. Journal of Geophysical Research 100, 18707-18,726. Tegen, I., and Fung, I. (1994). Modeling of Mineral Dust in the Atmosphere - Sources, Transport, and Optical-Thickness. Journal of Geophysical Research-Atmospheres 99, 22897-22914. Tegen, I., Lacis, A., and Fung, I. (1996). The influence on climate forcing of mineral aerosols from disturbed soils. Nature 380, 419-422. Thomson, R. E. (1981). 'Oceanography of the British Columbia coast.' (Department of Fisheries and Oceans Sidney, BC:. Thomson, R.E., and Krassovski, M.V. (2010). Poleward reach of the California Undercurrent ex-tension. Journal of Geophysical Research 115, C09027. Tindale, N., and Pease, P. (1999). Aerosols over the Arabian Sea: Atmospheric transport path-ways and concentrations of dust and sea salt. Deep Sea Research Part II: Topical Studies in Oceanography 46, 1577-1595. 94  Tsunogai, S., Suzuki, T., Kurata, T., and Uematsu, M. (1985). Seasonal and areal variation of continental aerosol in the surface air over the western North Pacific region. Journal of Oceanography 41, 427-434. Uematsu, M., Duce, R.A., and Prospero, J.M. (1985). Deposition of atmospheric mineral parti-cles in the North Pacific Ocean. Journal of Atmospheric Chemistry 3, 123-138. Ueno, H., Crawford, W.R., and Onishi, H. (2010). Impact of Alaskan Stream Eddies on Chloro-phyll Distribution in the North Pacific. Journal of Oceanography 66, 319-328. Vink, S., Boyle, E.A., Measures, C.I., and Yuan, J. (2000). Automated high resolution determi-nation of the trace elements iron and aluminium in the surface ocean using a towed Fish coupled to flow injection analysis. Deep-Sea Research Part I-Oceanographic Research Pa-pers 47, 1141-1156. Vink, S., and Measures, C.I. (2001). The role of dust deposition in determining surface water dis-tributions of Al and Fe in the South West Atlantic. Deep-Sea Research Part Ii-Topical Stud-ies in Oceanography 48, 2787-2809. Wang, M., and Overland, J.E. (2009). A sea ice free summer Arctic within 30 years? Geophysi-cal Research Letters 36, L07502. Wedepohl, H.K. (1995). The composition of the continental crust. Geochimica et Cosmochimica Acta 59, 1217-1232. Wells, M.L., Price, N.M., and Bruland, K.W. (1995). Iron chemistry in seawater and its relation-ship to phytoplankton: a workshop report. Marine Chemistry 48, 157-182. Werner, M., Tegen, I., Harrison, S., Kohfeld, K., Prentice, I., Balkanski, Y., Rodhe, H., and Roe-landt, C. (2002). Seasonal and interannual variability of the mineral dust cycle under present and glacial climate conditions. Journal of Geophysical Research 107, 4744. Wesolowski, D.J. (1992). Aluminum speciation and equilibria in aqueous solution: I. The solu-bility of gibbsite in the system Na-K-Cl-OH-Al (OH)4 from 0 to 100° C. Geochimica et Cosmochimica Acta 56, 1065-1091. Whitney, F.A., Crawford, W.R., and Harrison, P. (2005). Physical processes that enhance nutri-ent transport and primary productivity in the coastal and open ocean of the subarctic NE Pa-cific. Deep-Sea Research Part Ii-Topical Studies in Oceanography 52, 681-706. Whitney, F., and Welch, D. (2002). Impact of the 1997-1998 El Nino and 1999 La Nina on nutri-ent supply in the Gulf of Alaska. Progress in Oceanography 54, 405-421. Whitney, F.A., Freeland, H.J., and Robert, M. (2007). Persistently declining oxygen levels in the interior waters of the eastern subarctic Pacific. Progress in Oceanography 75, 179-199. 95  Wong, C.S., Waser, N.A.D., Nojiri, Y., Johnson, W.K., Whitney, F.A., Page, J.S.C., and Zeng, J. (2002). Seasonal and interannual variability in the distribution of surface nutrients and dis-solved inorganic carbon in the northern North Pacific: Influence of El Nino. Journal of Oceanography 58. Wong, C.S., Whitney, F.A., Matear, R.J., and Iseki, K. (1998). Enhancement of new production in the northeast subarctic Pacific Ocean during negative North Pacific index events. Limnol-ogy and Oceanography 43. Woodhouse, O., Ravizza, G., Kenison Falkner, K., Statham, P., and Peucker-Ehrenbrink, B. (1999). Osmium in seawater: vertical profiles of concentration and isotopic composition in the eastern Pacific Ocean. Earth and Planetary Science Letters 173, 223-233. Wright, J., and Colling, A. (1995). 'Seawater: its composition, properties, and behaviour.' (Per-gamon press:. Yeats, P.A., Dalziel, J.A., and Moran, S.B. (1992). A Comparison of Dissolved and Particulate Mn and Al Distributions in the Western North-Atlantic. Oceanologica Acta 15, 609-619. Zamzow, H., Coale, K.H., Johnson, K.S., and Sakamoto, C.M. (1998). Determination of copper complexation in seawater using flow injection analysis with chemiluminescence detection. Analytica Chimica Acta 377, 133-144. Zeebe, R., and Archer, D. (2005). Feasibility of ocean fertilization and its impact on future at-mospheric CO2 levels. Geophysical Research Letters 32, L09703. Zender, C.S., Bian, H.S., and Newman, D. (2003). Mineral Dust Entrainment and Deposition (DEAD) model: Description and 1990s dust climatology. Journal of Geophysical Research-Atmospheres 108, 4416.     96  Appendix I  Line P data (2010 and 2011) Station Latitude Longitude Depth Dissolved [Al], nM P4 (2010) 48°39' N 126°40'W 0 1.6, 1.7    5 0.7, 0.7    10 1.5, 1.5    25 1.4, 1.3    40 1.2, 1.2      P12 (2010) 48°58' N 130°40'W 0 3.6, 3.6    5 3.3, 3.3    10 1.6, 1.4    25 1.7, 1.7    40 2.7, 2.7    75 3.1, 3.1    100 3.2, 3.1    150 3.7, 3.6    200 2.4, 2.4    300 2.1, 2.0      P16 (2010) 49°17' N 134°40'W 0 0.68    5 0.91    10 1.23, 1.21    25 1.00    40 1.09, 1.10    75 2.09    100 1.59    200 1.18    300 0.27, 0.28    400 0.23      P20 (2010)* 49°34' N 138°40'W 5 0.98, 1.00    10 1.08, 1.05    25 0.95, 0.95    40 0.75, 0.73    75 1.13, 1.01    100 1.03, 1.08    150 0.85, 0.95    200 1.00, 1.00    300 0.72, 0.74    400 0.82, 0.87   97  Station Latitude Longitude Depth Dissolved [Al], nM P26 (2010)* 50°00'N 145°00'W 5 0.19    10 0.17    25 0.19    40 0.24    75 0.27    100 0.28    200 0.20    400 0.22    600 0.21    800 0.20      P4 (2011) 48°39' N 126°40'W 10 1.6, 1.8    40 1.8, 1.7    50 2.6, 2.7    75 1.6, 1.7    100 3.3, 3.4    200 2.1, 2.1    400 1.5, 1.6    600 2.2, 2.1    800 2.3, 2.3    1000 2.2, 2.2    1200 2.5, 2.5      P12 (2011) 48°58' N 130°40'W 10 1.0    25 1.2, 1.2    40 1.2    50 1.3    100 1.4    150 1.5, 1.4    200 1.4    250 1.3    300 1.8    400 1.8, 1.8      P16 (2011) 49°17' N 134°40'W 10 0.90, 0.92    25 0.68, 0.69    50 0.62, 0.64    75 0.60, 0.61    100 0.51, 0.51    150 0.59, 0.55    200 0.39, 0.37    300 0.24, 0.24    400 0.30, 0.28    600 0.29, 0.28 98  Station Latitude Longitude Depth Dissolved [Al], nM P16 (2011) 49°17' N 134°40'W 1100 0.27, 0.27    1200 0.31, 0.32    1400 0.26, 0.29    1800 0.15, 0.15    2000 0.11, 0.13 P26 (2011) 50°00'N 145°00'W 10 0.45, 0.41    25 0.41, 0.41    40 0.64, 0.59    50 0.25, 0.23    75 0.32, 0.35    100 0.36, 0.36    150 0.32, 0.32    200 0.16, 0.11    300 0.15, 0.17    400 0.12, 0.12    600 0.10, 0.12    800 0.07, 0.10    1000 0.12, 0.15    1200 0.10, 0.15    1400 0.10, 0.11    1600 0.10, 0.10    1800 0.07, 0.06    2000 0.07, 0.07      *P20 and P26 2010 were not included in the discussion. 99  Arctic Ocean data (2009) Station Latitude Longitude Depth Dissolved [Al], nM L1 71°06’ N 139°18’ W 10 0.6, 0.7    15 2.5, 2.3    50 0.5, 0.5    70 0.6, 0.5    90 0.7, 0.8    120 1.3, 1.4    150 1.1, 1.6    200 1.8, 2.1    250 2.3, 2.1    300 2.5, 2.8    400 3.0, 3.2    500 4.1, 4.4    600 4.4, 4.6    750 4.8, 5.0    1000 5.3, 5.3    1250 5.5, 5.7    1500 6.0, 6.2    1800 6.9, 7.1      L1.1 72°30’ N 135°35’ W 10 1.9, 2.0    30 1.4, 1.3    50 0.9, 0.9    70 0.9, 1.0    90 1.1, 1.1    120 1.0, 1.1    150 1.2, 1.1    200 7.8, 7.7    250 2.4, 2.3    300 1.9, 1.9    350 2.0, 2.0    400 2.4, 2.4    600 1.9, 2.0    750 3.0, 3.1    1000 3.7, 3.9    1300 4.0, 3.9    1600 5.1, 5.2    1900 3.3, 3.4    2500 4.6, 4.5      L1.5 73°19’ N 139°23’ W 10 0.6    25 0.9    40 1.0   100  Station Latitude Longitude Depth Dissolved [Al], nM L1.5 73°19’ N 139°23’ W 90 1.1    140 0.5    190 0.9    280 1.3    380 1.9    450 2.0    600 2.3    800 3.8    1000 3.3      L2 74°36’ N 137°07’ W 10 0.6, 0.6    20 0.3, 0.3    40 0.4, 0.4    80 0.3, 0.4    120 0.7, 0.8    180 0.5, 0.5    270 1.1, 1.3    360 1.4, 1.6    400 1.8, 1.7    440 3.2, 3.2    550 2.3, 2.2    700 4.1, 4.1    900 3.0, 3.0    1100 3.5, 3.5    1300 3.7, 3.6    1500 4.5, 4.4    1700 5.3, 5.5    2100 8.9, 8.9    2300 11.2, 11.5    2500 8.1, 7.8    2750 11.6, 11.7    2950 10.1, 9.9      S4 71°11’ N 132°56’ W 15 1.2, 1.2    30 0.9, 0.9    50 0.5, 0.5    70 1.0, 1.1    90 0.6, 0.6    120 0.7, 0.8    180 0.7, 0.7    250 0.8, 0.9    360 1.1, 1.2    420 1.4, 1.4  

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.24.1-0167483/manifest

Comment

Related Items