Content uploaded by Frederick Menzia
Author content
All content in this area was uploaded by Frederick Menzia on Apr 07, 2021
Content may be subject to copyright.
A preview of the PDF is not available
The solubility of sulfur hexafluoride in water and seawater
Abstract
The concentration of sulfur hexafluoride (SF6) in the atmosphere has been rapidly increasing during the past several decades. This long-lived compound enters the surface ocean by air–sea gas exchange and is potentially a very useful transient tracer for studying ocean circulation and mixing. SF6 has also been directly injected into the ocean at a minimal number of locations as a part of deliberate tracer release experiments to study gas exchange and sub-surface mixing rates. In this study, laboratory measurements of the solubility of SF6 in water and seawater were made over the temperature range of ∼−0.5°C to 40°C. Volumes of water and seawater held at constant temperature in glass chambers were equilibrated with a gas mixture containing SF6 and CFC-12 (CF2Cl2) at parts-per-trillion levels in nitrogen. Small volume water samples were analyzed by electron capture gas chromatography. Using the method of least squares, equations previously used in describing gas solubility as a function of temperature and salinity were fit to the SF6 and CFC-12 measurements. The CFC-12 results were in good agreement with previous work, while substantial differences were found between these SF6 results and those reported in earlier studies. The mean error for the analytical measurements is estimated to be ∼0.5%. Based on errors in the fits and the analytical errors, we estimate the overall accuracy of the SF6 solubility function to be of the order of 2%. The results from this work should be useful in determining equilibrium concentrations for SF6 in ocean observation and modeling studies.
... Transient tracer concentrations in water are often reported as partial pressure values with units in ppt (parts per trillion). This eliminates the effect of changing temperature and salinity on the tracer solubility and allows for direct comparison to atmospheric concentrations (Bullister et al., 2002;Warner & Weiss, 1985). ...
... During the SAS-Oden cruise, the two transient tracers CFC-12 and SF 6 , were measured on-board using a gas chromatograph, electron capture detector system in combination with a purge and trap unit (Bullister et al., 2002;Snoeijs-Leijonmalm & Party, 2022;Tanhua, Bulsiewicz, & Rhein, 2005;Tanhua et al., 2004). Water samples were drawn from Niskin bottles in 250 mL glass syringes and briefly stored in a 0°C water bath to prevent outgassing prior to measurement. ...
The Arctic Ocean plays an important role in the regulation of the earth's climate system, for instance by storing large amounts of carbon dioxide within its interior. It also plays a critical role in the global thermohaline circulation, transporting water entering from the Atlantic Ocean to the interior and initializing the southward transport of deep waters. Currently, the Arctic Ocean is undergoing rapid changes due to climate warming. The resulting consequences on ventilation patterns, however, are scarce. In this study we present transient tracer (CFC‐12 and SF6) measurements, in conjunction with dissolved oxygen concentrations, to asses ventilation and circulation changes in the Eurasian Arctic Ocean over three decades (1991–2021). We constrained transit time distributions of water masses in different areas and quantified temporal variability in ventilation. Specifically, mean ages of intermediate water layers in the Eurasian Arctic Ocean were evaluated, revealing a decrease in ventilation in each of the designated areas from 2005 to 2021. This intermediate layer (250–1,500 m) is dominated by Atlantic Water entering from the Nordic Seas. We also identify a variability in ventilation during the observation period in most regions, as the data from 1991 shows mean ages comparable to those from 2021. Only in the northern Amundsen Basin, where the Arctic Ocean Boundary Current is present at intermediate depths, the ventilation in 1991 is congruent to the one in 2005, increasing thereafter until 2021. This suggests a reduced ventilation and decrease in the strength of the Boundary Current during the last 16 years.
... is an Essential Ocean Variable (EOV) in the Global Ocean Observing System (GOOS) framework. SF6 has been used as a new tracer since mid-1990s complementing the CFCs due to the reduction of CFCs in the atmosphere (Maiss et al., 1996;Bullister et al., 2002). The application of SF6 is generally appropriate for recently formed water masses in the upper ocean (Tanhua et al., 2008), i.e. for well ventilated, or young, waters. ...
... can be estimated; 3) their input functions are known; 4) there are no sources or sinks in the ocean interior. The solubility (F) of transient tracers in the seawater is a function of potential temperature (θ) and practice salinity (S) (Warner and Weiss, 1985;Bullister et al., 2002). ...
The distribution of water masses, and the ventilation rates of these, are of significance to the thermohaline circulation and biogeochemistry of the world oceans. The distribution of the main water masses in the Atlantic Ocean is published in a companion study (Liu and Tanhua, 2021), their ages and ventilation time-scales are reported here by using observations of the transient tracers, CFC-12 and SF6. Two different definitions of water mass ages are presented; the mean-age representing an average age of a water mass, and the mode-age that better represents the advective time-scale. In general, ages increase with pressure and along the pathway of a water mass. The central waters in the upper layer obtain the mean-ages of up to ~100 years and the mode-ages of up to ~30 years. In the intermediate layer, the Antarctic Intermediate Water (AAIW) and the Mediterranean Water (MW) show gradients of water mass ages in the meridional and zonal direction respectively. The AAIW obtains the highest mean-age of ~300 years and mode-age of ~80 years at 30° N, while the MW shows the highest mean-age of ~400 years and mode-age of ~100 years in the equator region. As the dominant water mass in the deep and overflow layer, the North Atlantic Deep Water (NADW) from high northern latitudes obtains the highest mean-age of ~600 years and mode-age of ~100 years in the Antarctic Circumpolar Current (ACC) region at 50° S. In the bottom layer, the Antarctic Bottom Water (AABW) from the Weddell Sea obtains the highest mean-age of ~600 years and mode-age of ~100 years in the equator. As the continuation of AABW, the Northeast Atlantic Bottom Water (NEABW) obtains the highest mean-age of ~800 years and mode-age of ~120 years at 50° N. The mode-age increases with the transport distance from formation area, accompanied by significant differences between the eastern and western basins. The mode-age is used to calculate the oxygen utilization rate (OUR) with apparent oxygen utilization (AOU) during the active transport in water masses. The western basin exhibits lower mode-age with higher oxygen (low AOU) due to the better ventilation. The OUR shows similar distribution to dissolved oxygen (DO), indicating higher oxidation rate in the high oxygen region.
... To determine the relative ages between upper layers of the water column, apparent ages of water-masses were computed using the SF 6 partial pressure (pSF 6 ) dating method as reviewed by Fine (2011). This method uses the measured concentration of SF 6 in a water sample and the empirical SF 6 solubility function (Bullister et al., 2002) to determine the partial pressure of SF 6 in a water-mass at the measured temperature and salinity (Equations 1 and 2). pSF 6 = [SF 6 ]/F (T,S) (1) ...
... T = absolute temperature (K) S = salinity (PSS-78) a i and b i constants are given in Bullister et al. (2002). ...
While modeling efforts have furthered our understanding of marine iron biogeochemistry and its influence on carbon sequestration, observations of dissolved iron (dFe) and its relationship to physical, chemical and biological processes in the ocean are needed to both validate and inform model parameterization. Where iron comes from, how it is transported and recycled, and where iron removal takes place are critical mechanisms that need to be understood to assess the relationship between iron availability and primary production. To this end, hydrographic and trace metal observations across the GO‐SHIP section SR3, south of Tasmania, Australia, have been analyzed in tandem with the novel application of an optimum multiparameter analysis. From the trace‐metal distribution south of Australia, key differences in the drivers of dFe between oceanographic zones of the Southern Ocean were identified. In the subtropical zone, sources of dFe were attributed to waters advected off the continental shelf, and to recirculated modified mode and intermediate water‐masses of the Tasman Outflow. In the subantarctic zone, the seasonal replenishment of dFe in Antarctic surface and mode waters appears to be sustained by iron recycling in the underlying mode and intermediate waters. In the southern zone, the dFe distribution is likely driven by dissolution and scavenging by high concentrations of particles along the Antarctic continental shelf and slope entrained in high salinity shelf water. This approach to trace metal analysis may prove useful in future transects for identifying key mechanisms driving marine dissolved trace metal distributions.
... John Bullister (PMEL) working on the CTD frame before deployment on a GO-SHIP cruise in the Atlantic Ocean (left) and holding a sampling bottle from the same cruise (right). Bullister was an integral part of NOAA's GO-SHIP program and played a key role in developing the methodology for shipboard measurements and calibration of chlorofluorocarbons (CFCs) and other anthropogenic gases (Bullister and Weiss, 1988;Bullister andWisegarver, 1998, 2008;Bullister et al., 2002). He also made crucial measurements of tropospheric concentrations of these gases (Bullister, 2015). ...
The ocean is warming, acidifying, and losing oxygen. The Global Ocean Ship-based Hydrographic Investigations Program (GO-SHIP) carries out repeat hydrographic surveys along specified transects throughout all ocean basins to allow accurate and precise quantification of changes in variables such as temperature, salinity, carbon, oxygen, nutrients, velocity, and anthropogenic tracers, and uses these observations to understand ventilation patterns, deoxygenation, heat uptake, ocean carbon content, and changes in circulation. GO-SHIP provides global, full-depth, gold-standard data for model validation and calibration of autonomous sensors, including Argo. The Pacific Marine Environmental Laboratory, through sustained funding from NOAA, has developed methods to measure several of the variables routinely sampled through GO-SHIP and is a core contributor to these repeat hydrographic cruises.
... Considering the period when significant amounts of these gases had accumulated in the atmosphere, the method would be able to date water recharged since about the 1960s for CFC-12 and the 1970s for SF 6 . The concentrations of these gases in groundwater are governed by Henry's Law and were computed from Henry's coefficients published by Bullister et al. (2002) and by Warner and Weiss (1985) for SF 6 and CFC-12 respectively. Corresponding recharge periods (SF 6 and CFC-12 tracer age) were estimated by comparing the mixing ratios measured in the groundwater samples (corrected for excess air with noble gases data) with North American historical atmospheric records (Washington DC station, USGS 2019). ...
Information about the age distribution of groundwater provides unique and particularly relevant insights about groundwater flow and mixing conditions, as well as about the vulnerability and sustainability of the resource. Constraining flow paths in heterogeneous and fractured aquifers requires spatially distributed and depth-related observations. Where multiple wellbores are not available, depth-specific samples from individual long open-borehole wells may provide useful information. In this study, results from depth-specific sampling of environmental tracers (3H, 85Kr, SF6, CFC-12, noble gases, 13C, 14C) are reported from seven long open-borehole wells (depths ≤100 m below the surface) in the area of the St Lawrence Lowlands (Quebec, Canada). The multitracing approach allowed for the determination of the groundwater mixing conditions within the fractured-rock aquifers. The results indicate that fracture networks induce complex groundwater mixing patterns, without any evidence that the residence times significantly increase with the depth. Under unstressed conditions, the investigated aquifers contain 20–50% “modern” water (less than 60 years old) and 50–80% “old” water with residence times of ~4–11,000 years. The observed groundwater bi-modal age distribution is caused by fast flows through interconnected fractures and slow flows through poorly connected fractures or the primary porosity of the rock. Within the catchment of the wellfields, the pumped groundwater becomes essentially “modern” because of the preferential draining of productive fractures, while the water within the less transmissive volumes cannot be leaked out as fast. Thus, groundwater abstraction increases the aquifer’s intrinsic vulnerability to anthropogenic pollution.
Studying the change in foam viscosity during foam decay, a spontaneous and inevitable process, is of fundamental and practical interest across many applications, ranging from the froth in a cup...
Trace gases have demonstrated their strength for oceanographic studies, with applications ranging from the tracking of glacial meltwater plumes to estimates of the abyssal overturning duration. Yet measurements of such passive tracers in the ice-covered Arctic Ocean are sparse. We here present a unique data set of trace gases collected during the Multidisciplinary drifting Observatory for the Study of Arctic Climate (MOSAiC) expedition, during which R/V Polarstern drifted along with the Arctic sea ice from the Laptev Sea to Fram Strait, from October 2019 to September 2020. During the expedition, trace gases from anthropogenic origin (chlorofluorocarbon 12 (CFC-12), sulfur hexafluoride (SF6), and tritium) along with noble gases (helium and neon) and their isotopes were collected at a weekly or higher temporal resolution throughout the entire water column (and occasionally in the snow) from the ship and from the ice. We describe the sampling procedures along with their challenges, the analysis methods, and the data sets, and we present case studies in the central Arctic Ocean and Fram Strait to illustrate possible usage for the data along with their robustness. Combined with simultaneous hydrographic measurements, these trace gas data sets can be used for process studies and water mass tracing throughout the Arctic in subsequent analyses. The two data sets can be downloaded via PANGAEA: https://doi.org/10.1594/PANGAEA.961729 (Huhn et al., 2023a) and https://doi.org/10.1594/PANGAEA.961738 (Huhn et al., 2023b).
Many atmospheric chemicals occur in the gas phase as well as in liquid cloud droplets and aerosol particles. Therefore, it is necessary to understand their distribution between the phases. According to Henry’s law, the equilibrium ratio between the abundances in the gas phase and in the aqueous phase is constant for a dilute solution. Henry’s law constants of trace gases of potential importance in environmental chemistry have been collected and converted into a uniform format. The compilation contains 46 434 values of Henry’s law constants for 10 173 species, collected from 995 references. It is also available on the internet at https://www.henrys-law.org (last access: October 2023). This article is a living review that supersedes the now obsolete publication by Sander (2015).
Sulphur hexafluoride (SF6) has potential as a transient tracer of recently ventilated water masses, as its atmospheric burden continues to increase. Northern Arabian Sea hydrography was examined using measurements of atmospheric and dissolved SF6, CFC-11, CFC-12 and CFC-113. Persian Gulf Water (PGW) was characterized by its SF6 signal, and the time elapsed since its formation was evaluated by two approaches. Four ventilation age estimates were derived from SF6/CFC-11, SF6/CFC-12, CFC-113/CFC-11 and CFC-113/CFC-12, and their agreement at the oceanic stations confirms the validity of SF6 as a transient tracer. A second approach, of correcting SF6 partial pressure for PGW dilution by an optimal mixing model and referencing to the atmospheric SF6 chronology, provided a relative tracer age. This indicated a PGW flow of 0.016 (+/-0.003) m/s across the northern Arabian Sea, with an associated oxygen consumption of 10.1 mumol/l p.a. that exceeds tracer-derived estimates but confirms rates derived from export flux.
An automated gas chromatographic technique to measure the concentrations of chlorofluorocarbon 113 (CFC-113:CCl2FCClF2) dissolved in seawater has been developed. The method also quantifies chlorofluorocarbons II and 12 (CFC-11:CCl3F and CFC-12:CCl2F2). Seawater collected from Niskin bottles in ground-glass syringes is stripped by a gas stream and concentrated on a cryogenic trap in the manner of Bullister and Weiss (1988) and Gammon et al. (1982). By isolating and heating the trap, the chlorofluorocarbon compounds are reliberated and injected onto a high-resolution capillary gas chromatographic column, followed by electron-capture detection. The analysis time for each sample is less than 15 min. Surface seawater precisions are 2.9%, 2.4%, and 1.2% for CFC-113, CFC-11, and CFC-12, with detection limits of 0.003-0.004, 0.02, and 0.03-0.05 pmol L(-1), respectively. Although these statistics do not compare favorably with other CFC-11 and CFC-12 techniques (precision similar to 1%, detection limit similar to 0.005 pmol L(-1) (Bullister and Weiss, 1988)), the dynamic ranges of the CFC-113:CFC-11 and CFC-113:CFC-12 ''ventilation ages'' are 20:1, better than that of the best CFC-11:CFC-12 age, albeit with inferior precisions. Estimates of the solubility ratios of CFC-113:CFC-11 and CFC-113:CFC-12 are 0.303 and 1.22, disagreeing with the work of Wisegarver and Gammon (1988), whose CFC-113 results are believed to be boosted by coelution with methyl bromide. The optimum tracer ventilation age resolution is +/-0.9 years for both CFC-113:CFC-11 and CFC-113:CFC-12 from a cast considered in the northeastern Atlantic. A plot of CFC-113:CFC-12 ventilation age is presented on an outcropping isopycnal. A strong correlation with pressure and dissolved oxygen concentration is noted and an oxygen utilization rate between 4.2 and 5.5 +/- 0.4 mu mol L(-1) is implied, depending on the choice of CFC-113 atmospheric history.
The vertical distributions of the man-made chlorofluoromethanes CCl3F (F-11) and CCl2F2(F-12) have been measured at two locations in the eastern North Pacific Ocean to depths greater than 500 m. At both sites (46°N, 125°W off the Washington-Oregon coast and 50°N, 140°W in the Gulf of Alaska) the halomethane concentrations were found to fall off exponentially with increasing depth below the mixed layer. For F-11 at 50°N, the surface concentration (3.2×10−12 mol/l) was found to be in saturation equilibrium with the measured atmospheric concentration (190 pptv). The measured chlorofluoromethane profiles have been interpreted in terms of a one-dimensional model for the vertical diffusion/advection of an exponentially driven, conservative tracer into a bottomless ocean. In the appropriate limit of ‘transient steady state’ the projected profiles are simple exponentials described by an advective-diffusive scale depth H, which is a function of the vertical eddy diffusivity Kz, the upwelling velocity w, and the characteristic time τ for the exponential growth of the tracer concentration at the boundary. At the two ocean locations studied the freon thermocline depth scales were in the range H ≃ 120–140 m, with F-12 generally 10–20% deeper than F-11. At 50°N the F-11 and F-12 vertical profiles gave consistent values of vertical diffusivity (Kz ∼ 1.2–1.3 cm2 s−1) and upwelling velocity (w ≃ 12–14 m yr−1). The model also allows a simple scaling from one exponentially driven, transient tracer (freon) to another (fossil fuel CO2), leading to a predicted mean depth of penetration of fossil fuel CO2 of approximately 300 m in the eastern North Pacific.
An improved analytical technique has been developed for the rapid and
accurate shipboard measurement of two anthropogenically produced
chlorofluorocarbons (CFCs), CCl 3F (F-11) and CCl
2F 2 (F-12) in air and seawater. Gas samples (dry
air or standard) are injected into a stream of purified gas and then
concentrated in a low temperature trap. Seawater samples collected in
oceanographic Niskin bottles are transferred into glass syringes for
storage until analysis. An aliquot of approximately 30 cm 3
of seawater is introduced into a glass stripping chamber where the
dissolved gases are purged with purified gas, and the evolved CFCs are
concentrated in the same cold trap. The trap is subsequently isolated
and heated, and the CFCs are automatically transferred by a stream of
carrier gas into a precolumn and then a chromatographic separating
column. The CCl 3F and CCl 2F 2 peaks
are detected by an electron capture detector (ECD) and their areas are
integrated digitally. CFC amounts are calculated using fitted
calibration curves, generated by injection of various multiple aliquots
of gas standard containing known concentrations of CFCs. Preliminary
concentration values for these compounds are printed at the completion
of each analysis. Total analysis time for air and water samples is <
10 min, allowing detailed vertical profiles of the concentrations of
these compounds in the water column and concentrations in the overlying
atmosphere to be determined within a few hours of the completion of a
hydrographic station. Typical relative standard deviations for analyses
of CCl 3F and CCl 2F 2 in near-surface
seawater containing equilibrium levels of these compounds are
approximately 1%. Limits of detection for both compounds in 30 cm
3 seawater samples are about 0.005 × 10 -12
mol kg -1.
Annual mean mixing ratios for the halocarbons CFC-11 (CCl3F),
CFC-12 (CCl2F2), CFC-113
(CClF2CCl2F), and carbon tetrachloride
(CCl4) have been determined from their first year of
industrial production through 1998. From the late 1970s (in the case of
CFC-11 and CFC-12) or early 1980s (in the case of CFC-113 and carbon
tetrachloride) the reported mixing ratios have been determined from
experimental observations made by the Atmospheric Lifetime
Experiment/Global Atmospheric Gases Experiment/Advanced Global
Atmospheric Gases Experiment program. For years prior to these times we
have used estimates of industrial emissions and atmospheric lifetimes to
calculate historic concentrations. The likely error bounds of the annual
mean values are also reported here. Errors in the annual mean mixing
ratio may primarily be a result of incorrect industrial emissions data,
an incorrect atmospheric lifetime, or uncertainty in the ALE/GAGE/AGAGE
observations. Each of these possible sources of error has been
considered separately. These results show that atmospheric
concentrations for each of these compounds have experienced a rapid rise
in the early part of their production. It is only within the past decade
that rise rates have decreased sharply and (except in the case of
CFC-12) in the past few years that atmospheric concentrations have begun
to decrease. The uncertainties in the reconstructed histories are a
similar proportion for each of the chlorofluorocarbons (<4% for most
of the history). However, uncertainty in the history of carbon
tetrachloride is much greater (up to 12%), and this is mainly the result
of poor knowledge of CCl4 emissions.
A fully-automated analysis system for the tracer gas sulphur hexafluoride (SF6) is described. The system could be readily adapted to analyse seawater samples of varying volume and a concentration range exceeding 7 orders of magnitude. Automation facilitated a high precision despite continual analysis for periods of 30 days, and the performance of the instrument over a two year period of intensive use is discussed. Background SF6 profiles from different oceans are compared and the surface concentrations utilised to obtain atmospheric concentrations. These are incorporated with other published data to derive a history for atmospheric SF6 concentration. The data suggest an initial exponential rise in atmospheric SF6 to 1979, followed by a linear increase to late 1993 at a rate of 5.5% p.a. With the sensitive analytical system and documented atmospheric history described in this paper, SF6 has potential application as a transient tracer of recently ventilated waters, as demonstrated by comparison with the analytical parameters of the chlorofluorocarbons.
The first quantitative marine measurements of trichlorotrifluoroethane (F‐113) were made in the North Pacific subarctic gyre in July of 1986, using a modification of the analytical procedure developed for the transient tracers F‐11 and F‐12 (Gammon et al., 1982; Wisegarver and Cline, 1985). The measured mixed layer concentration was 0.29±0.02 pM/L; the corresponding atmospheric mixing ratio was 35±1 pptv (Rasmussen scale).
Below a subsurface maximum both F‐11 and F‐113 were found to fall off exponentially with increasing depth, F‐113 reaching effective blank levels first. The significance of adding F‐113 to the existing suite of measurable oceanic transient tracers rests in the possibility of age‐dating water masses by their (F‐113/F‐11) ratio with near annual resolution for the period since 1977.
We report and evaluate data of the chlorofluorocarbon CFC-113, in comparison with concurrent data for CFC-12, from cruises into the temperate North Atlantic, the tropical western Pacific, the Eastern Mediterranean, and the Weddell Sea. We consistently find CFC-113 deficiencies in the warmer upper waters, which we interpret as CFC-113 depletion at rates of the order of 3% per year, with possibly accelerated rates in the mixed layer or near the surface. These results severely limit a tracer utility of CFC-113 in the warm upper waters of the ocean. Tracer applications in the deep waters of the ocean should be much less affected, since CFC-113 loss in such waters appears to be far slower. This also holds for the Eastern Mediterranean, despite its rather warm deep waters. For the Weddell Sea, the data indicate a CFC-113 deficiency (~20%) in the shelf waters, which converts into a similar deficiency in newly formed deep and bottom water. Such CFC-113 deficiency has to be accounted for in Southern Ocean tracer work. Future work is proposed to study CFC-113 loss in the ocean in additional detail and to elucidate the loss mechanism(s), on which our data give little clue.
Exploratory measurements of a suite of anthropogenic halocarbon compounds (CCl4, CCl2FCClF2(CFC-113), CH3CCl3, CCl3F(CFC-11) were made using a new analytical technique on RV Meteor cruise 15 along 19°S [World Ocean Circulation Experiment (WOCE) Line A9] in the Atlantic Ocean during February-March 1991. A separate analytical system was used to determine CCl2F2 (CFC-12) and CCl3F(CFC-11). A limited number of CFC-113 profiles indicated that it was undetectable below 400-500 m. The CCl4 data indicate that the entire Brazil Basin contains readily measurable levels of CCl4(>0.05 pmol kg-1), whereas the deep Angola Basin contains very low levels (1000 m) had undetectable levels of CFC-11, CFC-12, and CFC-113. Preindustrial levels of CCl4 in the atmosphere were therefore negligible (atmospheric mixing ratio