OS22B-01 INVITED
Southern Ocean Gas Exchange Experiment
The Southern Ocean Gas Exchange Experiment (SO GasEx) is the third in a series of US-led open ocean process studies aimed at quantifying air-sea gas exchange. SO GasEx took place in the Atlantic sector of the Southern Ocean near South Georgia Island from February 29 to April 12, 2008. The main objectives of the experiment are to 1) improve quantification of the gas transfer velocity in a high wind and wave environment; and 2) determine whether there are unique controls on the gas transfer velocity in this significant CO2 sink region. An important goal of these efforts is to improve the quantification of gas transfer velocities on regional scales so that more accurate global air-sea CO2 fluxes can be determined. A systematic approach was followed during the cruise and will be continued after the cruise to accomplish this goal: 1) Perform direct flux measurements to obtain short-term estimates of local gas transfer velocities; 2) Combine integrated measurements of gas transfer velocities using 3He/SF6 dual tracer technique with short-term estimates and water column budgets; 3) Understand the mechanisms controlling ocean mixed layer pCO2 on short time and space scales; 4) Elucidate the forcing functions controlling gas transfer; and 5) Relate forcing functions to parameters that can be detected by remote sensing.
OS22B-02
Parameterization of Gas Exchange from the Southern Ocean Gas Exchange Experiment
Momentum, mass and heat are transferred across the air-sea interface mediating the biogeochemical cycling processes that facilitates life on earth. A particular relevant process is the carbon transfer between atmosphere and ocean via the exchange of CO2. A micrometeorological approach is taken in order to parameterize the gas transfer velocity (k) under a variety of atmospheric and oceanic conditions due to wind speeds ranging up to 20 m/s. The fluxes of CO2, momentum and heat are calculated via the direct covariance method (eddy correlation) and analyzed over a wide range of atmospheric forcing conditions. This process captures the physical variability in the transfer velocity and its dependents on turbulent transport and air-sea exchange. Different atmospheric variables are analyzed in order to capture the atmospheric conditions ruling the gas transfer. Particular attention is given to the relation of (k) and wind speed where we explore whether the exchange is best modeled by a linear, quadratic or cubic wind speed relationship for gas transfer velocity at high winds. GASEX 08 data set is complemented with data from GASEX 98 and GASEX 01, which provides a wind velocity range between 4 and up to 20 m/s, generating a solid data ensemble from which an empirical parameterization for (k) is developed.
OS22B-03
DOGEE-SOLAS: The Role of Surfactants in Air-Sea Gas Exchange
One of the major aims of DOGEE-SOLAS was to improve our understanding of the role of surfactants in air- sea gas exchange. With this in mind we carried out a number of artificial surfactant releases on a research cruise in the North Atlantic (D320), during June-July of 2007. We used oleyl alcohol, a surrogate for natural surfactants which is relatively cheap and easy to obtain (it is used in the manufacture of cosmetics). The main release overlaid a dual tracer "patch" of SF6 and 3He; our aim was to directly compare values of the gas transfer velocity, kw, estimated within the surfactant covered patch with those estimated quasi- simultaneously in a second, surfactant-free patch about 20km away. A second release in conjunction with colleagues from the University of Hawaii had the aim of measuring DMS fluxes by eddy correlation both inside and outside a surfactant slick, and a third was undertaken in the path of one of two 14m ASIS (Air-Sea Interaction Spar) buoys operated by the University of Miami for direct comparison of surfactant effects on the fluxes of CO2, H2O, heat and momentum (eddy correlation) etc. We present here some preliminary findings from the work.
OS22B-04
DMS transfer velocities above 10 m/s
We measured the sea/air flux of DMS by eddy correlation (EC) in two programs that encountered higher winds and seas than our previous cruises. As a part of the Deep Ocean Gas Exchange Experiment, DOGEE, we measured fluxes and sea water concentrations of DMS in the North Atlantic from the UK ship RRS Discovery in June and July of 2007. During the Southern Ocean Gas Exchange Experiment (SO-GasEx), we made similar measurements east of the southern tip of South America aboard the NOAA ship R/V Ronald H. Brown in Feb and March of 2008. Wind speeds during DOGEE ranged from nearly calm to over 15 m/s, while a few samples during SO-GasEx were near 20 m/s. DMS exchange velocities during DOGEE were nearly linear vs wind speed, close to Wanninkhof (92) below 5 m/s and Liss and Merlivat (87) from 9-13 m/s. All the exchange velocities during SO-GasEx were considerably smaller, roughly half the DOGEE values in the 9-13 m/s range. Above 15 m/s Kdms clearly decreased with wind speed. Possible explanations include surface DMS depletion, asymmetric exchange from bubble dissolution, and the reduction of tangential stress in favor of momentum transfer to waves (Soloviev and Schlüssel, 96).
OS22B-05 INVITED
Air-sea fluxes from drifting buoys during two recent open-ocean gas exchange experiments
A summary (progress report) of measurements from Air-Sea Interaction Spar (ASIS) buoys during the 2007 Deep Ocean Gas Exchange Experiment (DOGEE) and 2008 Southern Ocean Gas Exchange experiment (SO Gasex) is presented. Flux instrumentation consisted of a sonic anemometer, measuring temperature and the velocity components, and an open-path gas analyzer, LI-7500, measuring carbon dioxide and water vapor. Exchange-coefficients used in bulk-formulas, such as the Dalton and Stanton number are calculated. SAMI sensors mounted on the buoy measured the partial pressure of CO2 in the water. In combination with the vertical CO2 flux and mean CO2 values measured by the LI-7500 the transfer velocity of CO2 is evaluated. A specially designed outrigger attached to the ASIS structure enabled measurements of small waves. Wave lengths of the order of 20 cm could be resolved. This allows for a unique study of the relation between the transfer velocity and the mean square slope of the short waves.
OS22B-06
Simultaneous Eddy Covariance CO2 and DMS Flux Measurements in the North Atlantic
We present eddy covariance based CO2 and DMS fluxes and gas exchange coefficients (piston velocities) measured in situ during a 10-day cruise in the North Atlantic during summer 2007. Most current parameterizations of air/sea gas exchange utilize a generalized piston velocity, k, which relates the gas flux to the air/sea concentration difference, and incorporates all the physical factors controlling gas transport through both sides of the air/sea interface. Tracer release studies have shown that k increases with wind speed, though the functional form of the wind speed-dependence has not been well constrained due to averaging of flux estimates over varying environmental conditions, and also likely due to processes affecting gas exchange that are not correlated to wind speed. During the past decade, the micrometeorological technique eddy covariance (EC) has been used from ships at sea to measure k on short temporal and spatial scales, allowing for measurement of a wide range of environmental conditions on a single expedition. EC- based studies of CO2 and DMS in the Atlantic and Pacific Oceans have clearly shown wind speed dependency of k; however, a wide range of functional relationships is found among the cruises data sets. These differences may reflect differences in physical forcing of gas exchange in different environments, differing behavior of DMS versus CO2 due to solubility, or perhaps methodological differences. The simultaneous measurement of gas transfer coefficients of CO2 and DMS can provide insight into the sensitivity of gas exchange to gas solubility, and the extent to which gas transfer coefficients can be estimated from similarity relationships.
OS22B-07
Wintertime Air-Sea Gas Transfer Rates and Air Injection Fluxes at Station Papa in the NE Pacific
In recent studies of air-sea fluxes of N2 and O2 in hurricanes, McNeil and D'Asaro (2007) used a simplified model formulation of air-sea gas flux to estimate simultaneous values of gas transfer rate, KT, and air injection flux, VT. The model assumes air-sea gas fluxes at high to extreme wind speeds can be explained by a combination of two processes: 1) air injection, by complete dissolution of small bubbles drawn down into the ocean boundary layer by turbulent currents, and 2) near-surface equilibration processes, such as occurs within whitecaps. This analysis technique relies on air-sea gas flux estimates for two gases, N2 and O2, to solve for the two model parameters, KT and VT. We present preliminary results of similar analysis of time series data collected during winter storms at Station Papa in the NE Pacific during 2003/2004. The data show a clear increase in KT and VT with increasing NCEP derived wind speeds and acoustically measured bubble penetration depth.
OS22B-08
Evaluation of Satellite-Derived Estimates of the Gas Transfer Velocity Using Direct Observations From Research Vessels
Understanding and quantifying the exchange of gases, such as carbon dioxide, between the atmosphere and global oceans is of vital importance to climate community. While in situ observations from research vessels provide the most accurate observations of air-sea gas exchange, ultimately remote sensing techniques will be needed to provide the necessary temporal and spatial sampling to capture seasonal and interannual variations of gas exchange for all oceanic regions. Satellite methods, however, remain problematic due to difficulties in parameterizing the gas/chemical transfer processes and accurately determining and retrieving the essential satellite-derived inputs. The focus of this study is to assess and improve the ability to estimate the gas transfer velocity of carbon dioxide, ozone, and dimethylsulfide from satellite data. The physically- based COARE gas transfer model has been implemented for application with satellite-derived inputs. Global estimates of the gas transfer velocity from COARE will be presented and compared with more simplified parameterizations based solely on wind speed. Uncertainties in the satellite-derived transfer velocity due to errors in the input quantities and their sometimes indirect relationship to satellite observables will be derived and described. Comparisons of the satellite-derived gas transfer velocity to direct observations from the 1998 and 2001 GasEx cruises and 2008 Southern Ocean Experiment will provide initial validation of the satellite-derived estimates.