OS11C-01 INVITED 08:00h
Application of helium isotopes to studies of ocean circulation
Since the discovery of excess He-3 in the ocean by Clarke and Craig in the 1960's helium isotopes have been used in local, regional and global studies of circulation patterns and water mass transformation in the world ocean. From initial pilot studies through systematic exploration of these tracers during the GEOSECS (Geochemical Ocean Sections) program to the recent global survey as part of the WOCE (World Ocean Circulation Experiment) hydrographic program (WHP) we obtained more detailed information on the distribution of helium isotopes, as well as their sources and sinks in the ocean. This information can now be applied to construct global fields of helium isotopes and to extract unique information on the circulation patterns at different depth levels in the ocean, as well as on local and regional processes such as ventilation of water masses in deep water formation regions. Additionally, the data sets are now sufficiently large to be useful for validation of Ocean General Circulation Models (OGCM's). In this contribution we present examples of global helium isotope fields constructed from major programs such as GEOCECS, TTO (Transient Tracers in the Ocean), SAVE (South Atlantic Ventilation Experiment) and WOCE, as well as from individual ocean sections. We use the data to delineate circulation patterns in the major ocean basins at several depth levels, especially mid-depth waters. Additionally, we outline the use of helium isotopes in studies of ocean ventilation. Finally, we compare observed and simulated helium isotope fields to highlight OGCM capabilities and deficiencies to reproduce internal He-3 excesses in the ocean and the related ventilation processes.
OS11C-02 INVITED 08:15h
Net and gross production in the equatorial Pacific determined by measuring O2/Ar ratios and the triple isotope composition of O2
Simultaneous measurements of the dissolved O2/Ar ratio of the mixed layer, and the triple isotope composition of dissolved O2, allow one to constrain the basic metabolic rate processes (photosynthesis, respiration, and net production). Dissolved O2/Ar ratios are indicative of net production, because O2 and Ar share similar solubility properties with respect to physical mechanisms, but only O2 is biologically influenced (work of Craig, Jenkins, Emerson, Quay and colleagues). Photosynthesis attenuates the (mass-independent) 17O anomaly of atmospheric O2, which originates in the stratosphere, providing an index of the fraction of dissolved O2 derived from photosynthesis in situ. From measurements of these O2 properties and a suitable wind speed-gas exchange parameterization, one can calculate net and gross production rates of oceanic ecosystems. One can make extensive measurements of the O2 properties in the oceanic mixed layer, yielding experimental values of net and gross production at larger scales and higher resolution than is possible with traditional approaches. We have achieved a significant advance for the O2/Ar method by building a membrane inlet mass spectrometer that allows continuous shipboard analysis of O2, Ar, N2, and eventually CO2 and other gases. It was successfully deployed for the first time on transects at 110 W and 95 W, 8 S to 8 N, in the Equatorial Pacific in October/November 2003, followed by additional cruises at 125 W, 140 W, 155 W, and 170 W between June and September, 2004. The mass spectrometric measurements were accompanied by discrete O2/Ar measurements and isotopic measurements, along with continuous O2 concentration measurements. The results give gross production values and allow us to compute absolute Ar supersaturations, which may provide insight into the origin of physical supersaturation. On the 2003 cruise, a short-term reproducibility of 0.05 percent was achieved for the O2/Ar ratio, with a sampling frequency greater than twice/minute. Meridional gradients across the equatorial upwelling and small-scale local phenomena were resolved. Net production estimated from wind speed-based gas exchange parameterizations was near zero north of the equatorial upwelling, and about 12 mmol/m2/d south of it. Gross production estimated from 17O measurements ranged up to 150 mmol/m2/d. The corresponding equatorial net to gross O2 production ratios were approximately 0 to the north and 0.07 to the south. We compare these results with an extensive data set from the Southern Ocean.
OS11C-03 08:30h
In Vitro and In Situ Primary Productivity in the North Pacific Subtropical Gyre as Determined by the Triple Isotope Composition of Dissolved O$_{2}$, $^{18}$O$_{2}$ Labeling, and O$_{2}$/Ar Gas Ratios
Marine organic carbon production rates are a fundamental property of marine ecosystems and are essential to an understanding of surface ocean carbon cycling. However, uncertainties inherent in the methodologies commonly used to determine marine primary productivity (PP) (e.g. $^{14}$C-labeling, O$_{2}$ light/dark changes) make interpretations of measured PP difficult. As a result, comparisons among various in vitro and in situ methods are useful. We used a relatively new technique, measurement of the triple isotope composition of dissolved O$_{2}$, to characterize the mixed layer gross primary productivity (GPP) in situ at the Hawaii Ocean Time-series station ALOHA on four cruises during 2002-2003. Winter and summer GPP averaged 90 and 120 mmol O$_{2}$ m$^{-2}$d$^{-1}$, respectively, with uncertainties of $\sim \pm 40%$. Concurrent in vitro measurements of $^{18}$O$_{2}$ enrichment and $^{14}$C-labelling yielded a mixed layer-integrated GPP that was always less than or equal to the in situ GPP ($^{18}$O$_{2}$: 45-70 mmol O$_{2}$ m$^{-2}$d$^{-1}$; $^{14}$C: 35-50 mmol O$_{2}$ m$^{-2}$d$^{-1}$). Ratios of net community productivity (NCP) to GPP, determined by pairing oxygen isotope measurements with dissolved O$_{2}$/Ar gas ratio measurements, averaged 0.1 in the summer, close to the canonical f-ratio for the open ocean. These results suggest that in vitro methodologies may miss infrequent productivity events that contribute to the time-integrated average measured by in situ techniques.
OS11C-04 INVITED 08:45h
A Fast Response Gas Tension Device for Use on Profiling Floats
We present the development and testing phase of a new fast-response gas tension device (GTD) designed specifically for use on an air-deployable, buoyancy adjustable, profiling float. The GTD measures total dissolved air pressure when equilibrated with the local waters through a gas permeable membrane. Dissolved N2 levels are calculated using gas tension, water temperature, salinity and dissolved O2. The new fast-response GTD maintains the existing GTD's high precision and accuracy (0.01 %) but provides a significantly faster response. The new design physically separates the existing GTD into two primary components, a membrane interface and a precision pressure sensor. The two components communicate remotely through a low dead-volume tube. This allows the existing GTD to be minaturized for incorporation on the float and allows use of a novel, water pumped, larger surface area, fast response membrane interface. The equilibration time of the new instrument varies with depth due to compression of the membrane and its associated reduction in gas pernmeability. This effect is found to be significant to a depth of 10-20 m. In addition, the large surface area of the new fast-response interface provides a significant hydro-static response to the GTD due to degassing of the air dissolved in the membrane when it compresses. These responses are however very repeatable allowing the raw GTD signal to be corrected. The new GTD design equilibrates fastest near the sea surface and has a minimum response time (e-folding) of around 2 minutes. We suspect this can be improved further. The response time at 50 m is around 10 minutes. The depth controllable float enables an optimal 'sample-and-hold' approach at specified depths to ensure complete equilibration of the GTD. This GTD-float package allows remote upper ocean dissolved O2/N2 profiles, and time series at specified depths, to be collected for use in air-sea gas exchange and net biological productivity studies.
OS11C-05 INVITED 09:00h
Physical and biogeochemical controls on the air-sea disequilibrium of trace gases
The degree of saturation of trace gases in surface waters drives the flux across the sea surface, and influences interior tracer distributions. We develop a conceptual framework for understanding the mechanisms controlling the disequilibrium of trace gases in the surface ocean, and illustrate it using a global ocean biogeochemistry model. Air-sea heat flux, physical transport and biological sources drive surface concentrations away from equilibrium, and the degree of saturation depends on the relative timescale of these processes with respect to the timescale of air-sea gas exchange. Numerical simulations allow us to separate and quantify the relative importance of these processes. For example, in the region of deep water formation, the intense heat loss drives both surface O$_2$ and CO$_2$ toward undersaturation. However, deep convection and associated entrainment of deep waters bring old, O$_2$-depleted and CO$_2$-enriched waters to the surface, driving surface waters toward undersaturation of O$_2$ and supersaturation of CO$_2$. Furthermore, sea ice cover increases the timescale of air-sea gas exchange, and effectively increases the magnitude of surface disequilibrium. Although air-sea equilibration of oxygen is relatively rapid, "preformed" O$_2$ in deep waters is significantly undersaturated due to the reinforcing effects of these processes. This leads to a significant, systematic bias in estimates of respiration based on Apparent Oxygen Utilization (AOU).
OS11C-06 09:15h
Argon as a Tracer for Vertical Mixing and the Maintenance of the Subtropical Thermocline
We explore the use of Argon as a tracer of mixing in the main thermocline. After much debate, current thermocline theory suggests that the thermocline consists of an upper, adiabatic, ventilated portion and a lower, diffusive portion. Because the degree to which Argon is saturated in seawater is a nonlinear function of temperature, regions of heating due to mixing will also be regions of Argon supersaturation. We run a primitive equation model with Argon as a tracer to look at the nature of this supersaturation. Argon is forced with gas exchange at the surface with a tendency toward saturation, so the effect of bubbles is neglected. It is a passive tracer in the interior, subject to the same diffusivity as the temperature and salinity. In the main subtropical gyre away from the western boundary region, examination of the balance of terms in the temperature equation indicates a region in which diffusion is a dominant term in the balance, the so-called diffusive thermocline region. In this region, there is evidence of argon supersaturation. However, the exact depth and magnitude of the supersaturation maximum depend on the strength of the vertical diffusivity, the speed of the surface gas exchange, and the nature of the parameterized eddy mixing. In the case of the latter, eddy mixing parameterizations tend to mix heat from the warm, diffusive western boundary current region into the interior, reducing the western boundary region supersaturation and increasing the interior supersaturation at certain depths. At the equator, the supersaturation magnitude is larger, in conjunction with the larger total diffusion term in the temperature equation. Comparison with available data shows that the modeled supersaturation of Argon in the gyre is in good agreement with observations at BATS, displaying a small supersaturation indicative of weak vertical diffusivity or slow gas exchange.
OS11C-07 09:30h
Testing ocean models using measurements of interannual variations in atmospheric O2 and CO2 concentrations
Measurements of atmospheric O2 and CO2 concentration over the past 14 years show evidence of interannual variability related to both land and oceanic exchanges. The oceanic component is resolvable by summing the concentrations of O2 and CO2 to compute the tracer, atmospheric potential oxygen (APO), which varies due to the combined exchanges of CO2 and O2 across the air-sea interface but is largely invariant to land exchanges. Observed APO shows variations at stations in both the northern and southern hemispheres that are coherent between stations within each hemisphere and therefore are apparently caused by large-scale extra-tropical air-sea exchanges of CO2 and/or O2. These signals provide a window to study the response of ocean biogeochemistry to decadal climate variability. Here we compare the observed APO variability with predictions from a suite of biological ocean models driven by observed recent climate variability. The modeled air-sea O2 and CO2 fluxes are then fed into an atmospheric tracer transport model to predict APO variations. The comparison between predicted and observed interannual APO variations provides a check on the validity of the coupled models response to climate variations as well as insight into the origin of observed signals.
OS11C-08 09:45h
Measurements and Models of Oceanic $O_2$ and $CO_2$ Fluxes
Measurements of atmospheric $O_2$ and $CO_2$ can be combined to create a ``tracer'' referred to as Atmospheric Potential Oxygen (APO) [Stephens {\em et al.} 1998]. APO (defined as $O_2 + 1.1 \times CO_2$) is unaltered by terrestrial photosynthesis and respiration, and influenced only slightly by fossil fuel combustion. Consequently, measurements of $O_2$ and $CO_2$ have the potential to provide unique constraints on models of oceanic circulation and biology, and air-sea gas exchange. We present a 7-year record of $O_2$ and $CO_2$ measurements, combining data from the Princeton University and Scripps Institute of Oceanography $O_2$ flask-sampling networks. These values include both regular land-based observations and shipboard samples collected in the tropical Pacific. We find that the annually averaged North-South gradient in APO exhibits the gross features we expect from the combined influences of ocean processes, fossil fuel combustion and atmospheric transport: an interhemispheric gradient that is slightly higher in the South. Furthermore, the data show a substantial APO peak in the tropics (the so-called ``equatorial bulge'') resulting from large fluxes of both $O_2$ and $CO_2$ from the ocean to the atmosphere due to equatorial upwelling. We then use inversely estimated fluxes of $O_2$ and $N_2$ [Gruber {\em et al.}, 2001, Gloor {\em et al.}, 2001] along with observationally based $CO_2$ fluxes [Takahashi {\em et al.}, 1999] and the TM3 model of atmospheric transport to predict APO values at the locations of our observations. In contrast to earlier work [Stephens {\em et al.}, 1998, Gruber {\em et al.}, 2001], we find that the model generally agrees well with the data, reproducing the pole-to-pole gradient within a few per meg. Furthermore, the observed equatorial bulge is actually somewhat larger than predicted by the model. We find that our comparison metric for APO data-model agreement is particularly sensitive to the choice of atmospheric transport model.