• unlimited access with print and download
    $ 37 00
  • read full document, no print or download, expires after 72 hours
    $ 4 99
More info
Unlimited access including download and printing, plus availability for reading and annotating in your in your Udini library.
  • Access to this article in your Udini library for 72 hours from purchase.
  • The article will not be available for download or print.
  • Upgrade to the full version of this document at a reduced price.
  • Your trial access payment is credited when purchasing the full version.
Buy
Continue searching

Nitrogen, oxygen and the noble gases as tracers of upper-ocean productivity and air-sea gas fluxes

ProQuest Dissertations and Theses, 2009
Dissertation
Author: David Nicholson
Abstract:
In this study, measurements and modeling of dissolved nitrogen, oxygen and the noble gases are used to investigate biogeochemical fluxes of elements and physical processes of gas exchange to better understand the cycling of carbon within the ocean/atmosphere system. The first aspect of my work involved a year-long deployment of autonomous underwater gliders, called Seagliders, equipped with oxygen sensors to study the upper ocean in the subtropical North Pacific gyre near Hawaii. Using data collected by Seagliders and an oxygen mass balance approach, I was able to constrain the annual net community production of oxygen and organic carbon at my study site. The continuous nature of Seaglider data also allowed me to investigate the role of mesoscale variability in driving primary productivity in the subtropical upper ocean. During Seaglider deployments, four productivity events were observed with elevated oxygen and fluorescence in the deep euphotic zone, each of which corresponded to isopycnal shoaling events induced by passing Rossby waves. Seaglider temperature and salinity measurements were also used to investigate mixing rates in the upper thermocline. In the second main component of this study I made measurements of Kr/Ar and δ 40 Ar in deep ocean water and, along with other inert gas measurements, used a hierarchy of models to better understand ventilation processes during deepwater formation and rates of bubble fluxes. Using box models, I was able to demonstrate that most box models greatly overestimate the amount of ventilation at high latitudes because the surface-areas of high latitude boxes are too large. I used noble gas measurements to constrain an appropriate high latitude surface area of about 3% of total ocean surface area, which when applied in box models, results in gas cycling that is much more consistent with general circulation model results. Inert gas measurements were also used to constrain the flux of bubbles across the air-sea interface and I proposed new parameters to calculate bubble fluxes from 10-meter windspeed.

Table of Contents List of Figures ii List of Tables iv Acknowledgements v Chapter 1 1 1.1 Background 1 1.2 Properties of dissolved gases in the sea 2 1.3 Overview of Research 4 Chapter 2 9 2.1 Introduction 9 2.2 Methods 11 2.3 Results 14 2.4 Discussion 21 Chapter 3 38 3.1 Introduction 38 3.2 Salt fingering and mixing at Station Aloha 40 3.3 Constraining Kz with a salt mass balance 44 Chapter 4 68 4.1 Introduction 68 4.2 Methods 71 4.3 Results •. 79 4.4 Discussion 81 4.5 Conclusions 91 Kr/Ar , 98 Chapter 5 110 5.1 Introduction 110 5.2 Background I l l 5.3 Methods 112 5.4 Results 121 5.5 Discussion 124 5.6 Conclusions 127 Bibliography 134 Appendix A: Oxygen sensor performance on Seagliders 143 A.l Introduction 143 A.2 Oxygen sensors background 144 A.3 Oxygen sensors on Seagliders 145 A.4 Sensor calibration 146 Appendix B: Gas extraction line procedures for noble gas ratio analysis 156 B.l General operating rules 156 B.2 Line Preparation 156 B.3 Sample Procedure 157 B.4 Reference Air Procedure: 159 Curriculum Vitae 161 l

List of Figures Figure 1.1 Physical properties of gases 8 Figure 2.1 Seaglider study area 28 Figure 2.2 Oxygen optode calibration 29 Figure 2.3 Mooring temperature correlation timescale 30 Figure 2.4 Temperature correlation map 31 Figure 2.5 Sea surface height anomaly Hovmoller diagram 32 Figure 2.6 Mapped properties at Station ALOHA 33 Figure 2.7 Spatial maps of oxygen and oxygen saturation 34 Figure 2.8 Isopycnal depth and productivity events 35 Figure 2.9 Upper ocean fluorescence 36 Figure 2.10 Deep euphotic zone oxygen mass balance 37 Figure 3.1 Salinity mass balance domain 55 Figure 3.2 Density ratio Rp parameterization 56 Figure 3.3 Rp&t Station Aloha 57 Figure 3.4 Double diffusion staircase 59 Figure 3.5 Frequency of salt fingering during shoaling events 60 Figure 3.6 Model geostrophic flow velocities ...61 Figure 3.7 Schematic of isopycnal salinity mass balance 62 Figure 3.8 Isopycnal salinity mass balance 63 Figure 3.9 Seaglider depth averaged current 64 Figure 3.10 Geostrophic profiles calculated from Seaglider data 65 Figure 3.11 Depth-coordinate salinity mass balance 66 Figure 3.12 Lateral salinity gradients 67 Figure 4.1 Cooling and bubble sensitivity of gas ratios 98 Figure 4.2 Schematic of three-box ocean model 99 Figure 4.3 Map of sampling locations 100 Figure 4.4 Diagram of vacuum extraction line 101 Figure 4.5 D(Kr/Ar) and 840Ar air standards 102 Figure 4.6 D(Kr/Ar) and 840Ar analytical results 103 Figure 4.7 AAr sensitivity in the three-box model 104 Figure 4.8 Three-box model solution 105 Figure 4.9 Seven-box model solution. 106 Figure 4.10 Sensitivity to model parameters 107 Figure 4.11 High latitude biology in the three-box model 108 Figure 4.12 AH = 3% compared to AH = 15% 109 Figure 5.1 Winter sea-surface forcing 129 Figure 5.2 Mixed layer gas saturation anomalies 130 Figure 5.3 Interior gas saturation anomalies 131 Figure 5.4 Model data gas and gas ratio comparison 132 Figure 5.5 Bubble flux parameterization 133 Figure A.l Physical properties of gases : 150 Figure A.2 Seaglider study area 151 ii

Figure A.3 SBE43 calibration coefficients 152 Figure A.4 SBE43 sensor drift 153 Figure A.5 Optode and SBE43 calibrated profiles 154 Figure A.6 Optode and SBE43 regression after calibration 155 Figure B.l Gas extraction line schematic 160 m

List of Tables Table 2.1: Objective analysis parameters 27 Table 4.1A: Physical properties of gases 93 Table 4.IB: Physical properties of Argon isotopes 93 Table 4.2A: Three-box model default parameters 94 Table 4.2B: Seven-box model default parameters 94 Table 4.3: D(Kr/Ar) and 540Ar measurement statistics 95 Table 4.4: 540Ar and AKr/Ar measurements 96 IV

Acknowledgements My advisor, Steve Emerson, has taught me so much through classes, our discussions and his critical feedback. Working with him has been a pleasure and I greatly appreciate all the time he took to discuss my research, provide insightful scientific advice. Throughout my time at UW, Steve provided me with unparalleled opportunities to pursue exciting research and present my results to the scientific community. My committee members were also immensely supportive. Paul Quay and Jim Murray volunteered many hours to discussing science and serving as members of the reading committee. Craig Lee and Charlie Eriksen provided essential help (and lots of code) for my work interpreting Seaglider measurements. Becky Alexander lent insightful questions and comments and contributed a wonderful broad perspective. None of the technical aspects of my work could have happened without the help of Chuck Stump, who helped out with almost every aspect of both the Seaglider and inert gas portions of my thesis research. My mass spectrometric analysis would not have been possible without the help of Mark Haught. I also am indebted to Taka Ito, UW undergraduates, and the captains and crew of the CCGS John P. Tulley, R/V Thomas G. Thompson, R/V Mirai and R/V Oceanus for aid in collecting samples. I also would like to thank Drs. Curtis Deutsch, Robbie Toggweiler, Michael Bender and Samar Khatiwala for lending technical advice and modeling code. Additionally, I would like to acknowledge all of my fellow graduate students in Oceanography and the Program on Climate Change. In particular I would like to thank Laurie Juranek, Bonnie Chang, John Kirkpatrick, Katie Fagan and Aaron Donohoe. Lastly, thank you to my parents, Paul and Ginnie, my sister, Maika, my grandmother, Nai Nai and my wife, Sutin. Without your support I would never have made it this far. Love you very much. v

1 Chapter 1 INTRODUCTION TO GAS TRACERS OF THE MARINE CARBON CYCLE /./ Background The marine carbon cycle strongly influences atmospheric CO2 concentrations on a whole range of timescales. Ocean mechanisms are generally believed to have caused the changes in atmospheric pC02 associated with the glacial/interglacial cycles. Today, the ocean takes up over a third of the carbon dioxide that we release to the atmosphere by the burning of fossil fuels. The ocean's ability to regulate atmospheric pCCh largely depends on how the interplay of physics and biology transfers carbon from the upper ocean, where it is rapidly exchanged with the atmosphere, to the deep ocean, where CO2 is sequestered from the atmosphere for hundreds or years. The flux of carbon to the deep ocean occurs via several pathways. The measurement of dissolved gases in the sea can help us understand the biological pathways of CO2 uptake, as well as the physically driven fluxes at the air-sea interface. The 'biological carbon pump' is controlled by ocean biology, which takes up CO2 during photosynthesis, and subsequently sinks to the deep ocean where it is remineralized. As water cools, before filling the deep ocean, its ability to hold CO2 increases, because gas solubility is an inverse function of water temperature. This physically driven flux of CO2 into water and subsequent subduction into the deep ocean is called the 'solubility pump' This thesis examines how measurements of gases in the ocean, namely, O2, N2, Ne, Ar and Kr can be used as tracers to better understand and quantify the ocean biological pump, solubility pump and air sea gas exchange. The thesis is divided into two main parts. Part 1, encompassing Chapters 2 and 3 is a study of the biological pump at Station Aloha in the subtropical North Pacific Ocean using oxygen measurements made by autonomous Seagliders. Part 2 (Chapters 4 and 5) uses

2 measurements and modeling of inert gases to investigate the strength of the solubility pump during the formation of deepwater and rates of bubble mediated gas exchange across the air-sea interface. Oxygen measurements are used as a tool to quantify the biological pump because ocean biology produces oxygen in stoichiometric ratio to the amount of carbon dioxide that is converted to organic carbon during photosynthesis. Respiration in the ocean causes the reverse reaction, consuming oxygen and organic carbon and producing CO2. Organic carbon produced in the surface ocean that is not subsequently consumed by respiration eventually sinks to the deep ocean. The net production of oxygen and organic carbon by the upper ocean ecosystem, termed Net Community Production (NCP) is thus, over annual and longer timescales, a measure of the amount of carbon exported by the biological pump. Researchers use annual mass balances of oxygen to quantify the magnitude of biological oxygen production and the biological pump. The strength of the solubility pump is difficult to determine from observations of the carbon system alone, because measurements represent the intertwined signature of biology and physics. The distribution of the inert gases (N2 and the noble gases), on the other hand, is determined solely by physical processes. Observations of these gases can help us understand the physical mechanisms of air-sea gas exchange and deepwater formation that determine the strength of the ocean solubility pump. Each of the inert gases has unique physical properties and thus has unique sensitivities to the processes of interest such as cooling, bubble fluxes, mixing and diffusive gas exchange. Studying a suite of inert gases can elucidate the role of a range of physical processes. Results from the study of inert gases can then be applied to the carbon system. 1.2 Properties of dissolved gases in the sea All of the gases investigated in this dissertation are well mixed and of known composition in the atmosphere. To the first order, the concentration of these gases dissolved in seawater is dictated by Henry's Law, which states

Pcxa(T,S) = [C]M (1.1) where [C]sat is the equilibrium saturated concentration of a gas C, pc is the partial pressure of the gas, and a(T,S) is the temperature and salinity dependent solubility. The solubility of each gas is a non-linear function of temperature and salinity, where solubility of gases decreases with increasing temperature (Fig 1.1 A). Gas solubilities have been determined analytically, and for this study we use the results of Garcia and Gordon for O2 [1992], Hamme and Emerson for Ne, N2 and Ar [2004], and Weiss and Kyser [1978] for Kr. Observed dissolved gas concentrations in the surface ocean are generally within a few percent of the equilibrium condition described in Equation 1.1. Thus, we often refer to gases in terms of the saturation anomaly, AC, rather than concentration, [C], such that AC = 100 x ( [C] . (1.2) V L^ hat J The flux of gases in and out of the ocean, Fc, is regulated by air-sea gas exchange, which can be described as the sum of diffusive gas exchange and bubble fluxes Fc=-kc(icncial)+Bc where kc is the gas transfer coefficient, and Be is the bubble flux. In Chapters 4 and 5 we will further discuss the parameterization of Be- The gas transfer velocity is a function of wind speed, temperature and gas diffusivity. At the air-sea interface, there are three primary processes that contribute to the initial saturation anomaly of a gas; cooling (heating), causes a negative (positive) anomaly by increasing (decreasing) [C]sat, bubble fluxes caused by breaking waves causes a positive anomaly, and the gas transfer velocity, kc, which regulates the magnitude of the saturation anomaly. Due to the nonlinearity of the solubility function, mixing in the thermocline can also induce a positive saturation anomaly. The distinct physical properties of each gas tracer causes each gas to respond differently to each of these processes, and thus provide unique tracer information. The solubility of heavier

4 gases is more sensitive to temperature change than lighter gases (Fig 1.1B), so heavier gases such as Kr are more sensitive to cooling and heating. The lighter gases are less soluble than heavier gases (Fig 1.1 A), causing the injection of air bubbles to have a greater impact on the saturation anomaly of light gases such as neon. Lighter gases diffuse faster than heavier gases and thus have a larger kc for a given windspeed (Fig 1.1C). A larger kc dampens the magnitude of saturation anomaly for lighter gases more than for heavy gases. Heavier gases also have a greater non-linearity to their solubility function, and thus also have a greater mixing-induced saturation anomaly. 1.3 Overview of Research 1.3.1 Seaglider observations ofNet Community Production- Ch. 2 Quantifying NCP using an oxygen mass balance method requires measurements of oxygen throughout the annual cycle. Previous studies have used time-series from the monthly ship based observations of programs such as the Hawaii Ocean Time- series (HOT) and the Bermuda Atlantic Time-series (BATS). Such an approach is limited to a few time-series locations in the world's ocean that are regularly visited by research ships. Additionally, the monthly resolution of these studies may be too coarse to capture the episodic nature of biology in the ocean. Methods of continuous, in situ monitoring are needed to be able to quantify the magnitude of NCP and better understand what processes control the episodic nature of ocean productivity. To this end, Seagliders, deployed through most of 2005 in the subtropical North Pacific gyre, made measurements of temperature, salinity, and dissolved oxygen to quantify net community production (NCP) at Station A Long-Term Oligotrophic Habitat Assessment (ALOHA) of the Hawaii Ocean Time-series (HOT) using an oxygen mass balance approach. A 'bowtie'-shaped pattern, 50 km by 50 km in size was repeatedly traversed at two-week intervals with the goal of observing the influence of Rossby waves and eddies on the productivity of the study region. Rossby waves and eddies in the region cause a vertical displacement of isopycnal depth of about ±50 m at the base of the euphotic zone. Shoaling of isopycnals is demonstrated to drive

productivity in the deep euphotic zone. Four mesoscale shoaling events were observed between February and November in 2005. During each event when isopycnals shoaled, oxygen concentrations on isopycnals increased, fluorescence in the deep euphotic zone was higher, and net community production was elevated. Productivity in the deep euphotic zone was strongly influenced by Rossby waves and eddies, but this influence was not observed to extend into the mixed layer. 1.3.2 Constraining thermocline mixing from Seaglider observations - Ch. 3 One of the greatest uncertainties in the calculating NCP using an oxygen mass balance method is uncertainty in the diapycnal mixing flux of oxygen down into the thermocline. Because of steep vertical oxygen gradients at the base of the euphotic zone, NCP calculations are very sensitive to assumptions about the diapycnal mixing rate [Hamme and Emerson, 2006]. The coefficient of vertical diffusivity, Kz, is uncertain over an order of magnitude, from 0.1 to 1.0 cm2 s. Studies of ocean mixing indicate that mixing rates can very dramatically in space and time. Slow background mixing rates have been observed to be dramatically increased by processes such as flow over rough topography [Rudnick et at, 2003], and double-diffusion induced salt fingering [Schmitt et at, 2005]. Previous attempts to constrain the mixing flux of oxygen using season heat budgets have been unsuccessful due to uncertainty in surface heat and bubble fluxes. Results from the 2005 Seaglider survey provide a wealth of observations that provide insight into mixing processes at Station Aloha. In Chapter 3 I use high- resolution Seaglider temperature and salinity profiles to evaluate mixing at the base of the euphotic zone. First, temperature and salinity profiles are used to calculate when and where salt fingering is likely to enhance mixing. Seaglider observations reveal that while salt fingering does not increase annually average mixing by much, during brief periods of time, salt fingering greatly enhances vertical mixing. Furthermore, these periods correspond to the same periods when isopycnal shoaling events were observed to stimulate NCP. The second part of Chapter 3 demonstrates how a salt budget at the

6 base of the euphotic zone can be used to constrain the average diapycnal mixing rate in 2005. At Station Aloha, a subsurface salinity maximum persists through the year. Seaglider salinity observations are used to create an isopycnal mass balance model to constrain Kz. Using objectively mapped salinity fields and modeled geostrophic advection, the annual average diapycnal mixing rate is constrained to be Kz = 0.7+ 0.3 cm2 s"1. 1.3.3 Inert gas constraints on ventilation rates during deepwater formation - Ch. 4 The concentration of inert gases in the deep ocean records a fingerprint of the physical conditions during water mass formation in high latitude regions. The concentration of each gas varies slightly from the Henry's Law equilibrium saturation concentration. This saturation anomaly is set by the combined influence of low atmospheric pressure, the rate at which water cools, bubble fluxes, and the area and rate of diffusive gas exchange. Because each gas has a unique sensitivity to each process, quantifying the saturation anomaly of a suite of inert gases can constrain the relative contribution of each process. Certain ratios of inert gases are particularly sensitive to certain physical processes. In Chapter 4 I report precise measurements of N2/AJ, 40Ar/36Ar and Kr/Ar from deep profiles in the Atlantic and Pacific Oceans. These results are interpreted using simple box models to constrain the surface area available for gas exchange during deepwater formation, and the bubble injection rate of this area. In this model we constrain the appropriate surface area of the high latitude box to be 2.68% (+ 2.40%, - 1.27%o) of ocean surface area and the rate of high latitude bubble air injection to be 22.3 ±8.8 mol m"2 yr"1. Our results provide geochemical support for previous hypotheses that box models, which typically have high latitude surface areas of 15%, overestimate communication between the deep ocean and the atmosphere. The surface of the high latitude box serves as a small window through which gases dissolved in the ocean can communicate with the deep ocean. Reducing high latitude surface area is

demonstrated to have implications for both the strength of the carbon solubility pump, and the sensitivity of the carbon cycle to changes in high latitude carbon export. 1.3.4 Modeling the distribution of inert gases in the deep ocean - Ch. 5 I extend the box model approach of the previous chapter to interpret gas measurements using offline circulation from a global climate model in Chapter 5. The global distribution of inert gases in the ocean was calculated from a quasi-steady state mixed layer model [Schudlich and Emerson, 1996] and a tracer transport matrix (TMM) method based on circulation from the MITgcm [Khatiwala, 2007]. In this approach, climatological heat flux, atmospheric pressure and windspeed are used to calculate surface gas disequilibria for the inert gases Ne, N2, Ar and Kr. Mixed layer concentrations are then transported using the TMM method to determine gas distributions in the ocean interior. Using the TMM modeling approach, I isolate the relative influence of cooling, atmospheric pressure variations, mixing and bubble fluxes to setting the saturation state of inert gases in the ocean. Measurements of Ne, Ar, Na/Ar and Kr/Ar from this and previous studies are then used in an inverse calculation, to constrain the contribution of cooling and bubble fluxes. I use the inverse method to determine a new parameterization for the air-sea flux of injecting and exchanging bubbles as a function of wind speed. This new parameterization is based on observations of deep ocean water that last outcropped at high latitudes in regions with higher wind speed. My new parameterization therefore expands and compliments previous attempts to paramteterize bubble fluxes based on subtropical observations and lower wind speeds [Hamme and Emerson, 2006; Stanley et al, 2006, Stanley et al. 2009]. The parameterization from this work suggests that bubble injection fluxes are about 40% less than the result of Stanley et al. while the flux of exchanging bubbles is less well constrained, but of similar magnitude. The discrepancy could be explained if bubble fluxes do not increase as quickly with wind speed as cubic relationship implies.

Figure 1.1 8 ? B I ! 10 20 Temperature (°C) 10 20 Temperature (°C) 30 10 20 Temperature (°C) 30 Figure 1.1: Physical properties of gases in seawater. (A.) The solubility of pure gases at 1 atm in seawater. The heavier gases are more soluble than lighter gases (B.) The percentage change in the equilibrium solubility of each gas as a function of temperature relative to 5°C. Solubility of gases is higher at low temperatures and solubility of heavy gases more sensitive to temperature change than for lighter gases. N2 Ar and O2 all have very similar temperature dependence of their solubility functions. (C.) The gas transfer velocity (k) normalized to 5°C for Argon. Gases have a faster gas transfer rate at higher temperatures (for a given wind speed) and lighter gases have faster gas transfer velocities relative to heavier gases.

9 Chapter 2 NET COMMUNITY PRODUCTION IN THE DEEP EUPHOTIC ZONE OF THE SUBTROPICAL NORTH PACIFIC GYRE FROM GLIDER SURVEYS 2.1 Introduction The net community production (NCP) of the ocean is an important aspect of the carbon cycle and a primary control on the partial pressure of carbon dioxide (pCCh) of the atmosphere. The processes that regulate NCP, the amount that gross primary productivity (GPP) exceeds respiration (R), are poorly understood. The net biological production of oxygen (O2) can be used to quantify NCP and is a powerful tracer of biological processes in the surface ocean. Geochemical studies suggest that annually, the euphotic zone of the subtropical gyres have a positive NCP and produce an excess of organic carbon which subsequently is exported to the deep ocean [e.g. Gruber et al, 1998; Homme and Emerson, 2006; Quay and Stutsman, 2003; Spitzer and Jenkins, 1989]. In vivo studies of short term GPP and R using bottle incubations suggest the opposite, that R exceeds GPP [le B. Williams et al, 2004; Duarte and AgustVi, 1998; del Giorgio andDuarte, 2002]. One hypothesis reconciling these observations is that NCP occurs in brief bursts of 'productivity events' [Emerson et al, 2002] and that net heterotrophic conditions exist between events [McAndrew et al, 2007]. Eddies and Rossby waves may be the trigger that drives these events. Because primary productivity in the subtropical gyres of the ocean is probably nitrate and phosphate limited, any physical processes that bring nutrients into the euphotic zone can trigger NCP and organic matter export. Cyclonic eddies can elevate isopycnals, bringing nutrients from deeper waters into the euphotic zone, driving new production [McGillicuddy Jr et al, 2003; Sweeney et al, 2003]. In the subtropical North Pacific, westward propagating Rossby waves cause oscillations in sea surface dynamic height that can be observed in altimetry data [Chiswell, 1994]. Rossby waves

10 and eddies, while displacing sea surface height (SSH) by only a few centimeters, often displace thermocline isopycnals by tens of meters. These oscillations may provide a significant flux of nutrients into the euphotic zone [Sakamoto et ah, 2004] and drive productivity events. The ability to observe these episodic events and their impact on NCP requires continuous observation over many months. The autonomous Seaglider is an optimal platform for such a study. We use Seaglider measurements to determine NCP using the oxygen mass balance approach [Emerson et ah, 1997; Hamme and Emerson, 2006]. Biological production of oxygen in the ocean is stoichiometrically related to carbon uptake by the ratio AC>2:AC = 1.45 during nitrate based photosynthesis [Anderson, 1995; Hedges et al, 2002]. Regenerated production based on NH4+ uptake has a ratio of AC>2:AC =1.1 [Laws, 1991]. We suspect that new production is primarily driven by nitrate and that AC^iAC = 1.45 is an appropriate ratio to convert net biological oxygen production to NCP of carbon. In this paper, we focus on NCP in the deep euphotic zone because this zone is most directly influenced by nutrient injections from below. In the mixed layer, nutrients are more likely supplied by other sources, such as nitrogen fixation. Furthermore, uncertainties in gas exchange and bubble injection processes make it difficult to constrain the mixed layer NCP using only O2 gas measurements. Mixed layer NCP for our study site is calculated in Emerson et al. [Emerson et al, 2008] using in situ mooring measurements of nitrogen and oxygen. Mixed layer NCP was not observed to correlate with the timing of mesoscale shoaling events. Seaglider oxygen measurements centered on Station A Long-Term Oligotrophic Habitat Assessment (ALOHA) (158°W 22°45*N) of the Hawaii Ocean Time-series and the Hawaii Air-sea Logging Experiment (HALE-ALOHA) mooring [Moore et al. unpubl.; 22° 46TM, 158° 5.5'W ] during 2005 are used to investigate controls on NCP in the deep euphotic zone. We use shipboard Winkler titration measurements to calibrate Seabird and Aanderaa oxygen sensors and evaluate their accuracy and precision during Seaglider use. An objective analysis (OA) is used to

11 identify the primary time and length scales of temperature, salinity and oxygen variability and to create a gridded dataset to be used for mass balance calculations. Using OA-mapped Seaglider data we calculate net biological oxygen production for the deep euphotic zone (below the mixed layer) in the vicinity of Station ALOHA during 2005. Finally, we investigate the role four observed mesocale shoaling events in driving NCP in the deep euphotic zone of the subtropical North Pacific gyre. 2.2 Methods 2.2.1 The Seaglider The Seaglider is a small, autonomous sampling platform developed at the University of Washington collaboratively between the School of Oceanography and Applied Physics Laboratory [Eriksen et al, 2001]. Seagliders are battery powered, using buoyancy control and wing lift to propel the glider forward at speed of roughly 0.25 m s"1. They can be deployed for up to six months at a time and have a range of thousands of kilometers because of very low power consumption. Seagliders use global positioning system (GPS) fixes acquired at the sea surface to navigate and the Iridium satellite network to communicate control commands and return data in near real time. Measurements include the concentration of dissolved oxygen ([O2]), temperature (T), salinity (S), fluorescence, optical backscatter, and depth-averaged currents. Seagliders for this project were equipped with a custom Seabird CTD to measure S, T, and pressure, a Seabird SBE 43 dissolved oxygen sensor (SBE43), an Aanderaa optode 3830 dissolved oxygen sensor (optode) and Wetlabs BB2SF fluorescence and optical backscatter. Salinity and temperature sensors are mounted on a sail protruding from the top of the Seaglider, between the wings. The SBE43 and optode were mounted atop and beneath the aft portion of the fairing, respectively. Three Seaglider deployment periods were used in this study to collect data through much of 2005 at Station ALOHA. Deployments of Seaglider 021 spanned 16 February to 22 May and 15 August to 13 November while Seaglider 020 was deployed

Full document contains 175 pages
Abstract: In this study, measurements and modeling of dissolved nitrogen, oxygen and the noble gases are used to investigate biogeochemical fluxes of elements and physical processes of gas exchange to better understand the cycling of carbon within the ocean/atmosphere system. The first aspect of my work involved a year-long deployment of autonomous underwater gliders, called Seagliders, equipped with oxygen sensors to study the upper ocean in the subtropical North Pacific gyre near Hawaii. Using data collected by Seagliders and an oxygen mass balance approach, I was able to constrain the annual net community production of oxygen and organic carbon at my study site. The continuous nature of Seaglider data also allowed me to investigate the role of mesoscale variability in driving primary productivity in the subtropical upper ocean. During Seaglider deployments, four productivity events were observed with elevated oxygen and fluorescence in the deep euphotic zone, each of which corresponded to isopycnal shoaling events induced by passing Rossby waves. Seaglider temperature and salinity measurements were also used to investigate mixing rates in the upper thermocline. In the second main component of this study I made measurements of Kr/Ar and δ 40 Ar in deep ocean water and, along with other inert gas measurements, used a hierarchy of models to better understand ventilation processes during deepwater formation and rates of bubble fluxes. Using box models, I was able to demonstrate that most box models greatly overestimate the amount of ventilation at high latitudes because the surface-areas of high latitude boxes are too large. I used noble gas measurements to constrain an appropriate high latitude surface area of about 3% of total ocean surface area, which when applied in box models, results in gas cycling that is much more consistent with general circulation model results. Inert gas measurements were also used to constrain the flux of bubbles across the air-sea interface and I proposed new parameters to calculate bubble fluxes from 10-meter windspeed.