ALBERT

Description: The ‘International Intercomparison Exercise of fCO2 Systems’ was carried out in 1996 during the R/V Meteor Cruise 36/1 from Bermuda/UK to Gran Canaria/Spain. Nine groups from six countries (Australia, Denmark, France, Germany, Japan, USA) participated in this exercise, bringing together 15 participants with seven underway fugacity of carbon dioxide (fCO2) systems, one discrete fCO2 system, and two underway pH systems, as well as systems for discrete measurement of total alkalinity and total dissolved inorganic carbon. Here, we compare surface seawater fCO2 measured synchronously by all participating instruments. A common infrastructure (seawater and calibration gas supply), different quality checks (performance of calibration procedures for CO2, temperature measurements) and a common procedure for calculation of final fCO2 were provided to reduce the largest possible amount of controllable sources of error. The results show that under such conditions underway measurements of the fCO2 in surface seawater and overlying air can be made to a high degree of agreement (±1 μatm) with a variety of possible equilibrator and system designs. Also, discrete fCO2 measurements can be made in good agreement (±3 μatm) with underway fCO2 data sets. However, even well-designed systems, which are operated without any obvious sign of malfunction, can show significant differences of the order of 10 μatm. Based on our results, no “best choice” for the type of the equilibrator nor specifics on its dimensions and flow rates of seawater and air can be made in regard to the achievable accuracy of the fCO2 system. Measurements of equilibrator temperature do not seem to be made with the required accuracy resulting in significant errors in fCO2 results. Calculation of fCO2 from high-quality total dissolved inorganic carbon (CT) and total alkalinity (AT) measurements does not yield results comparable in accuracy and precision to fCO2 measurements.

Description: We show the distribution of nutrients, oxygen, total dissolved inorganic carbon (CT) and total alkalinity (AT) along three sections close to the Canary Islands, between 18°W and the African coast during Meteor 37/2 cruise (January 1997). From the thermohaline properties of Eastern North Atlantic Central Water (ENACW), Mediterranean Water (MW), Antarctic Intermediate Water (AAIW) and North Atlantic Deep Water (NADW), a mixing model has been established based on the water mass description. It can explain most of the variabilities found in the distribution of the chemical variables, including the carbon system, and it is validated through the use of conservative chemical variables like ‘NO.’ From nutrients, oxygen, AT and CT, the chemical characterisation of the water masses was performed by calculating the concentration of these variables in the previously defined thermohaline end-members. The relative variation of nutrient concentrations, resulting from the regeneration of organic matter, was estimated. Close to the African shelf-break, a poleward undercurrent conveying as much as a 11% of AAIW was observed only in the southern section (28.5°N). From the chemical and thermohaline properties of the end-members, a comparison with data from other oceanic regions was made in respect to conservative chemical variables (‘NO’). In addition, a north–south gradient in the ventilation pattern of water masses is observed from the residuals of the model.

Description: Two newly designed underway systems for the measurement of CO2 partial pressure (pCO2) in seawater and the atmosphere are described. Results of an intercomparison experiment carried out in the North Sea are presented. A remarkable agreement between the two simultaneously measured (pCO2) data sets was observed even though the spatial variability in surface pCO2 was high. The average difference of all l -min averages of the seawater pCO2 was as low as 0.15 μatm with a standard deviation of 1.2 μatm indicating that no systematic difference is present. A closer examination of the profiles shows that differences tend to be highest during maxima of the pCO2 gradient (up to 14 μatm/min). The time constants of both systems were estimated from laboratory experiments to 45 s, respectively, 75 s thus quantitatively underlining their capability of a fast response to pCO2 changes

Description: Highlights: • The distribution of particulate matter was studied using Argo float measurements. • Its spatio-temporal properties were analyzed in the eastern tropical North Atlantic. • Surface, subsurface, intermediate and bottom nepheloid layers were considered. • High correlations between particulate matter and phytoplankton were verified. • The depth of subsurface particle maxima correlated to the distance to shore. Abstract: The spatial and temporal distribution of particulate matter in the water column of the eastern tropical North Atlantic between 16.9–22.9°N and 16.6–29.3°W was investigated using optical measurements from transmissometers mounted on Argo floats. The corresponding profiles of beam attenuation coefficients measured from February 2008 to May 2009 were used to study particulate matter in different layers such as the surface nepheloid layer (SNL), subsurface nepheloid layer (SSNL), intermediate nepheloid layer (INL) and bottom nepheloid layer (BNL) as well as to investigate sinking particles (SP). The SNL were down to about 60 m water depth at thicknesses between 20 and 60 m. Our analyses verified high correlation between particulate matter and phytoplankton in the SNL. High offshore SNL extension of up to 750 km was found in the area of Cape Blanc filaments in January 2009. Their typical widths ranged from 11 to 72 km. Furthermore, float-borne observations even resolved atmospheric dust deposition into the surface water layer during a strong Saharan dust event in October 2008. The observed dust concentration in the mixed water layer was found to vary between 0.0021 and 0.0168 g m−3 depending on applied assumptions. An abrupt change from a SNL to a SSNL regime over distances of only 80 to 90 km was observed. The particulate matter in the SSNL showed lateral extensions from 420 to 1020 km offshore. A statistically significant correlation between the depth of subsurface particle maxima and the distance to shore was found. An averaged diameter of 30 km was determined for the sharply isolated patches of INL which was consistent with model simulations of other studies. The lateral transport of particulate matter in these INL features in the area of the giant Cape Blanc filaments was found to be more pronounced than reported in earlier studies. The distribution of particulate matter within the INL filaments reached up to 610 km off the shelf edge. The frequency of INL decreased with increasing distance to shore. The sinking velocity of particulate matter of one long-term observed INL was approximately 1.3 m day−1. Highly concentrated BNLs with beam attenuation coefficients of up to 4.530 m−1 were observed in the continental slope region. INLs appeared more frequently than SP events which lead to the conclusion that the lateral transport of particulate matter in INL features in the study area was more important than their passive vertical sinking.

Description: Total alkalinity (TA) is one of the few measurable quantities that can be used together with other quantities to calculate concentrations of species of the carbonate system (CO2, HCO3 −, CO32−, H+, OH−). TA and dissolved inorganic carbon (DIC) are conservative quantities with respect to mixing and changes in temperature and pressure and are, therefore, used in oceanic carbon cycle models. Thus it is important to understand the changes of TA due to various biogeochemical processes such as formation and remineralization of organic matter by microalgae, precipitation and dissolution of calcium carbonate. Unfortunately deriving such changes from the common expression for TA in terms of concentrations of on-conservative chemical species (HCO3 −, CO3 2 −, B(OH)4 −, H+, OH−, etc.) is rarely obvious. Here an expression for TA (TAec) in terms of the total concentrations of certain major ions (Na+, Cl−, Ca2+ etc.) and the total concentrations of various acid-base species (total phosphate etc.) is derived from Dickson's original definition of TA under the constraint of electroneutrality. Changes of TA by various biogeochemical processes are easy to derive from this so-called explicit conservative expression for TA because each term in this expression is independent of changes of temperature or pressure within the ranges normally encountered in the ocean and obeys a linear mixing relation. Further, the constrains of electroneutrality for nutrient uptake by microalgae and photoautotrophs are discussed. A so-called nutrient-H+-compensation principle is proposed. This principle in combination with TAec allows one to make predictions for changes in TA due to uptake of nutrients that are consistent with observations. A new prediction based on this principle is the change in TA due to nitrogen fixation followed by remineralization of organic matter and subsequent nitrification of ammonia which implies a significant sink of TA in tropical and subtropical regions where most of the nitrogen fixation takes place.

Description: The subpolar North Atlantic (NA) plays a key role in the oceanic uptake of anthropogenic CO2. The availability of a historical high quality data set from the Transient Tracers in the Ocean North Atlantic Study (TTO-NAS) in 1981, together with data from recent studies in 1997 and 1999, makes it possible to assess the temporal increase of anthropogenic CO2 (View the MathML sourceCTant) in the region. We introduce an extension of a previous published empirical approach for estimating temporal increases of View the MathML sourceCTant, which is known as multiple linear regression approach (MLR). The method is based on a multiple linear-regression model employing hydrographic and chemical parameters. The accuracy of the extended MLR calculation (eMLR) proposed here is estimated to be ±3 μmol/kg for a parameterization based on potential temperature, total alkalinity, silicate, and phosphate. Calculated increases of View the MathML sourceCTant (View the MathML sourceΔCTant(PO4)) for the time period 1981–1997 are 1–20 μmol/kg at depths greater than 100 m. The distribution corresponds well to silicate and CFC-12 distributions. Open ocean profiles show a relative minimum between 300 and 1000 m, which is not apparent in profiles of the total View the MathML sourceCTant concentration. The View the MathML sourceΔCTant(PO4) inventory calculation for the northern NA region (40–65°N) yields a change in anthropogenic CO2 storage of 4.2 (±1) pg C over the 16-yr time period 1981–1997. This is equivalent to a mean annual View the MathML sourceCTant increase of 0.27 (±0.06) pg C/yr or more than 10% of the global ocean uptake for this period.

Description: A coulometrically-based SOMMA system for the determination of total dissolved carbon dioxide (TCO2) in a continuous mode was designed and tested at sea. The new continuous technique approached the same high accuracy and reliability associated with prior discrete TCO2 measurements. During three cruises encompassing more than 19 weeks and 6000 continuous TCO2 measurements none of the three different systems tested exhibited any hardware-related failures. We found that coulometer cell lifetimes can greatly exceed prior expectations with many of the titration cells in the continuous mode remaining accurate for up to 72 h at carbon ages exceeding 50 mg C. We suggest a practical definition based on the CRM analyses for changing coulometer cells in the continuous mode. Systematic deviations of the SOMMA pipette volume from a theoretical temperature dependence were identified both from field data comparisons and pipette calibrations. Hence pipettes should be kept at constant temperature or they must be gravimetrically calibrated over the expected temperature range. Comparison of the continuous TCO2 data together with simultaneously measured additional CO2 system parameters showed that the refitted “Mehrbach” dissociation constants for carbonic acid best-represent fCO2 when calculated from TCO2 and alkalinity over a wide range of sea-surface temperatures and salinities. Some remaining systematic differences of calculated–measured fCO2 of up to 9 μatm likely reflect uncertainty in the temperature-dependence of the “Mehrbach” constants as well as possible uncertainty in the alkalinity–salinity relationship used to estimate alkalinity in the consistency checks.

Description: Data on the carbonate system of the Northwestern Indian Ocean obtained on a cruise of F.S. Meteor during SW monsoon in July/August 1995 were compared with those of George et al. [George, M.D., Kumar, M.D., Naqvi, S.W.A., Banerjee, S., Narvekar, P.V., de Sousa, S.N., Jayakumar, D.A., 1994. A study of the carbon dioxide system in the northern Indian Ocean during premonsoon. Mar. Chem. 47, 243–254] collected during intermonsoon. In general, deep water values agreed well between the two expeditions. Surface waters, however, showed a substantial increase in dissolved inorganic carbon (CT) in the coastal regions due to strong upwelling in the SW monsoon. This was also accompanied by very high CO2 partial pressures in surface waters. The north–south gradients in vertical profiles of the measured parameters in the Arabian Sea are discussed by comparing profiles from the oligotrophic equatorial region with those from the highly productive central Arabian Sea. The effect of denitrification on regenerated CT and AT is minor, with contributions of 〈9 and 〈8 μmol kg−1, respectively, to the total amount regenerated also utilizing oxygen. The dissolution of biogenic carbonates is discussed; different approaches to define the depth, where the dissolution starts (lysocline(s), carbonate critical depth (CCrD)), are compared together with the calculation of saturation depth from carbonate concentrations. It is shown, that small differences in measured CT and AT (found between our data and those measured during GEOSECS) and different calculation approaches to the CO2 system (different dissociation constants for species involved and taking into account phosphate and silicate concentrations) can produce pronounced differences in the calculated saturation depths. However, CT and AT data suggest substantial dissolution of biogenic carbonate in the water column even above the calcite lysocline, irrespective of the procedures followed to calculate this horizon.

Description: The transit time distribution method was applied to dichlorodifluoromethane and sulfur hexafluoride measurements from four cruises to the tropical North Atlantic between 2006 and 2009 in order to estimate anthropogenic carbon (C-ant) concentrations. By assuming an Inverse Gaussian distribution of the transit time distribution the best fit to the data was achieved with the ratio of mean age to width equals 1. Significant differences in the mean age and C-ant concentrations between the equatorial belt (5 degrees S-5 degrees N) and the Guinea dome area (5 degrees-15 degrees N) was found. Mean ages are higher and C-ant concentrations are lower in the Guinea dome area than at same depths, or densities, in the equatorial belt. The mean column inventories in the upper 1200 m are higher by about 3 mol m(-2) in the equatorial belt compared to the Guinea dome area. The mean column inventory of C-ant, for the whole water column, in the tropical Atlantic is 32.2 mol m(-2) (error range: 30.6-45.2 mol m(-2)), which is significantly lower than the previous estimates. The total C-ant inventory in the eastern tropical Atlantic is 2.5 Pg (error range: 2.3-3.5 Pg) for an area of 6 x 10(6) km(2), comprising the Guinea dome region and the equatorial belt. The equatorial belt has 40% higher storage of C-ant compared to the Guinea dome area which reflects the occurrence of relatively young deep waters at the equator, being high in anthropogenic carbon. Our tracer based C-ant estimates were compared to C-ant concentrations calculated with the TrOCA method applied to measurements conducted in 1999. The TrOCA based estimates are significantly higher than our tracer based C-ant estimates. Comparison between tracer measurements in 1999 and the 2006-2009 time-frame revealed possible speed-up of ventilation in the upper water column, increasing the C-ant concentration in this depth range at a faster rate and a C-ant increase of 12.1 mu mol kg(-1) in the tropical surface water was found

Description: A climatological mean distribution for the surface water pCO2 over the global oceans in non-El Niño conditions has been constructed with spatial resolution of 4° (latitude) ×5° (longitude) for a reference year 2000 based upon about 3 million measurements of surface water pCO2 obtained from 1970 to 2007. The database used for this study is about 3 times larger than the 0.94 million used for our earlier paper [Takahashi et al., 2002. Global sea–air CO2 flux based on climatological surface ocean pCO2, and seasonal biological and temperature effects. Deep-Sea Res. II, 49, 1601–1622]. A time-trend analysis using deseasonalized surface water pCO2 data in portions of the North Atlantic, North and South Pacific and Southern Oceans (which cover about 27% of the global ocean areas) indicates that the surface water pCO2 over these oceanic areas has increased on average at a mean rate of 1.5 μatm y−1 with basin-specific rates varying between 1.2±0.5 and 2.1±0.4 μatm y−1. A global ocean database for a single reference year 2000 is assembled using this mean rate for correcting observations made in different years to the reference year. The observations made during El Niño periods in the equatorial Pacific and those made in coastal zones are excluded from the database. Seasonal changes in the surface water pCO2 and the sea-air pCO2 difference over four climatic zones in the Atlantic, Pacific, Indian and Southern Oceans are presented. Over the Southern Ocean seasonal ice zone, the seasonality is complex. Although it cannot be thoroughly documented due to the limited extent of observations, seasonal changes in pCO2 are approximated by using the data for under-ice waters during austral winter and those for the marginal ice and ice-free zones. The net air–sea CO2 flux is estimated using the sea–air pCO2 difference and the air–sea gas transfer rate that is parameterized as a function of (wind speed)2 with a scaling factor of 0.26. This is estimated by inverting the bomb 14C data using Ocean General Circulation models and the 1979–2005 NCEP-DOE AMIP-II Reanalysis (R-2) wind speed data. The equatorial Pacific (14°N–14°S) is the major source for atmospheric CO2, emitting about +0.48 Pg-C y−1, and the temperate oceans between 14° and 50° in the both hemispheres are the major sink zones with an uptake flux of −0.70 Pg-C y−1 for the northern and −1.05 Pg-C y−1 for the southern zone. The high-latitude North Atlantic, including the Nordic Seas and portion of the Arctic Sea, is the most intense CO2 sink area on the basis of per unit area, with a mean of −2.5 tons-C month−1 km−2. This is due to the combination of the low pCO2 in seawater and high gas exchange rates. In the ice-free zone of the Southern Ocean (50°–62°S), the mean annual flux is small (−0.06 Pg-C y−1) because of a cancellation of the summer uptake CO2 flux with the winter release of CO2 caused by deepwater upwelling. The annual mean for the contemporary net CO2 uptake flux over the global oceans is estimated to be −1.6±0.9 Pg-C y−1, which includes an undersampling correction to the direct estimate of −1.4±0.7 Pg-C y−1. Taking the pre-industrial steady-state ocean source of 0.4±0.2 Pg-C y−1 into account, the total ocean uptake flux including the anthropogenic CO2 is estimated to be −2.0±1.0 Pg-C y−1 in 2000.