Repeat hydrography in the Mediterranean Sea , data from the Meteor cruise 84 / 3 in 2011

Here we report on data from an oceanographic cruise on the German research vessel Meteorcovering large parts of the Mediterranean Sea during spring of 2011. The main objective of this cruise was to conduct measurements of physical, chemical and biological variables on a section across the Mediterranean Sea with the goal of producing a synoptic picture of the distribution of relevant physical and biogeochemical properties, in order to compare those to historic data sets. During the cruise, a comprehensive data set of relevant variables following the guide lines for repeat hydrography outlined by the GO-SHIP group ( htt ://www.go-ship.org/) was collected. The measurements include salinity and temperature (CTD), an over-determined carbonate system, inorganic nutrients, oxygen, transient tracers (CFC-12, SF 6), helium isotopes and tritium, and carbon isotopes. The cruise sampled all major basins of the Mediterranean Sea following roughly an east-to-west section from the coast of Lebanon through to the Strait of Gibraltar, and to the coast of Portugal. Also a southto-north section from the Ionian Sea to the Adriatic Sea was carried out. Additionally, sampling in the Aegean, Adriatic and Tyrrhenian Seas were carried out. The sections roughly followed lines and positions that have been sampled previously during other programs, thus providing the opportunity for comparative investigations of the temporal development of various parameters. Data coverage and parameters measured Repository-Reference: doi:10.3334/CDIAC/OTG.CLIVAR_06MT20110405 Available at: Bottle data: http://cdiac.ornl.gov/oceans /Coastal /Meteor_Med_Sea.html CTD continuous profiles: http://cchdo.ucsd.edu /cruise/06MT20110405 Coverage: 33–42 ◦ N; 36◦ E–8◦W Location Name: The Mediterranean Sea Date/Time Start: 5 April 2011 Date/Time End: 28 April 2011


Introduction
The Mediterranean Sea is a relatively small, land-locked ocean basin with an active deep and shallow overturning and complex upper layer circulation.The last few decades has seen dramatic changes in the circulation of the Mediterranean Sea.This is manifested, amongst others, as a shift of the deep water formation area from the Adriatic to the Aegean Sea in the Eastern Mediterranean Sea, and an intense deep water formation event in the mid-2000s in the Western Mediterranean Sea (e.g.Roether et al., 1996;Schroeder et al., 2008).The deep water formed from these two sources Published by Copernicus Publications.Schneider et al., 2010).The column inventories of C ant in the Mediterranean is among the highest found anywhere in the world oceans; the Mediterranean Sea thus stores a significant portion of the global anthropogenic emissions of CO 2 despite its relatively small volume.However, few inorganic carbon data exist in the Mediterranean Sea, and knowledge about how recent changes in circulation are affecting the storage rate of C ant is fundamental.
Here we report on a data set obtained during spring 2011 on the German research vessel Meteor, cruise M84/3 (EX-POCODE: 06MT20110405).The cruise was set-up to follow the demands and requirements of repeat hydrography as specified by the GO-SHIP group (http://www.go-ship.org/),i.e. with a comprehensive set of physical and chemical parameters measured to the highest standards.The principal scientific objectives for M84/3 had two closely linked components: understanding and documenting the large-scale Mediterranean water property distributions, their changes and the drivers of those changes.These data will support understanding and addressing questions of a future Mediterranean Sea that will experience increasing concentrations of dissolved inorganic carbon, might become more stratified and experience changes in circulation and ventilation processes.These objectives were achieved by measurements of physical parameters with CTD (including oxygen) and by on-board measurements of discrete water samples for oxygen, nutrients (nitrate, nitrite, phosphate and silicate), dissolved inorganic carbon (DIC), total alkalinity, pH, and the transient tracers SF 6 and CFC-12.In addition, samples were taken for the determination of 3 He and tritium, as well as for the carbon isotopes 14 C and 13 C, for later shore-based analyses.Here we present a short description of the M84/3 cruise that started in Istanbul (Turkey) on 4 April and ended in Vigo (Spain) on 28 April 2011.The cruise track is shown in Fig. 1.For a sub-set of the CTD stations, no chemistry was performed due to time limitations, and for another subset of stations a comprehensive tracer program with sampling for helium (including tritium) and carbon isotopes were performed; these sub-sets are marked with different colors in Fig. 1.A more comprehensive report from the observing program during cruise M84/3 can be found in the cruise report (Tanhua, 2013).

CTD
Altogether, 61 standard hydrographic stations were occupied during the cruise, employing a SeaBird SBE911 plus CTD-O 2 probe, attached to a SeaBird carousel 24 bottle watersampler.At most stations water samples were taken from 24 depth levels within the water column from the surface to the bottom.At least three of those were analyzed for salinity onboard using certified reference seawater 38H11 with a K15factor of 1.07631 (24 • C).The CTD data were post-processed by applying Seabird software and MATLAB routines.At this stage, spikes were removed, 1 dbar average calculated, and temperature, salinity and oxygen were corrected with a regression analysis which fits the downcast profiles with the temperature and salinity.Since corrections to these parameters were small, the data quality was excellent.Overall accuracies are within expected ranges: 0.002 • C for temperature, 0.003 for salinity and approximately 2 µmol kg −1 for oxygen.
Samples for chemical analysis of discrete water samples were drawn from 10 L Niskin bottles.The sampling order were always: helium, CFC12/SF 6 , oxygen, DIC/Carbon isotopes, pH, alkalinity, nutrients, tritium, salinity and other samples.Below follows a short description of the various onboard measurements.

Dissolved oxygen
Dissolved oxygen was measured at all chemistry stations for all depths following the Winkler potentiometric method modified after Langdon (2010).Two independent reference materials for the iodate standard were used to calibrate the thiosulfate solution: commercial potassium iodate solutions of 0.01 N (20 • C) provided by OSIL (UK) and Wako (Japan).Three reproducibility exercises were performed along the cruise, resulting in an estimated precision of ±0.6 µmol kg −1 or better.

The carbonate system
All the stations and depths were sampled for total alkalinity (TA), dissolved inorganic carbon (DIC) and pH, in order to have an over-determined CO 2 system.The DIC content was determined using a SOMMA instrument.Samples were collected in borosilicate bottles according to standard operation protocol.A small headspace (< 1 %) was adjusted to prevent pressure build-up and loss of CO 2 during storage.Samples were not poisoned and normally measured within 12 h; the samples were kept at 20 (±0.2) • C during the analysis.An aliquot of approximately 26 mL was transferred to the instrument where it was acidified with phosphoric acid and the outgassed CO 2 was measured coulometrically.The precision of the analysis was determined to ±0.6 µmol kg −1 by titration of several bottles filled from the same Niskin bottles.The accuracy was determined to be 2.5 µmol kg −1 by analyzing a total of 42 bottles of certified reference material (CRM, Andrew Dickson, Scripps, CA, USA, batch 108); the DIC of this batch is certified at 2022.70 ± 0.7 µmol kg −1 .Measurements of the CRMs were also used to daily correct the temporal drift in the coulometer cell; this correction was never larger than 3 µmol kg −1 .Spectrophotometric pH in seawater was measured at every station and depth using a double-wavelength spectrophotometric procedure (Clayton and Byrne, 1993), and is reported at 25 • C and on the total scale; i.e. pH25T.Samples were collected in cylindrical optical glass 10 cm path-length cells, which were filled to overflowing and immediately closed.All the absorbance measurements were obtained in the thermostated cell at 25±0.2 • C. The absorbance was measured at three different fixed wavelengths (434, 578 and 730 nm) and calculated using the formula in Clayton and Byrne (1993).A correction was applied to compensate for the injection of the indicator into the seawater.The magnitude of this correction over our range of pH is small, ranging from 0.003 to 0.0007 pH units.The accuracy of the pH measurements were determined by analyzing ten to fifteen samples per week of CO 2 reference material (CRM, batch 108).Since the pH in these CRMs are not given, the pH of this batch was calculated to 7.8782 using the dissociation constants from Mehrbach et al. (1973) refitted by Dickson and Millero (1987).Our pH measurements on the CRMs were lower than this theoretical value but according to an internal consistency analysis detailed in Álvarez et al. (2013) no correction was applied.On three occasions several samples collected from the same Niskin bottle were analyzed, giving the reproducibility of the pH measurements to be 0.0012 units.
Samples for total alkalinity (TA) determination were collected in 600 mL borosilicate bottles and were analyzed within a day after sampling.The TA samples were analyzed following a double end point potentiometric technique by Pérez and Fraga (1987) and Pérez et al. (2000).Measurements of CRM were performed in order to control the accuracy of the TA measurements; the pH as measured by the electrode was corrected to obtain the closest mean TA to the CRM; the pH correction was never larger than 0.05.The precision of the TA measurements was better than 0.6 µmol kg −1 determined from duplicate Niskin bottles, and the accuracy of the TA measurements (i.e.including sampling errors) is determined to be 1 µmol kg −1 using CRMs measurements.Additionally, the daily drift of the system was determined by regular measurements of a substandard (aged surface seawater); this drift was always very low.Substandards of aged surface seawater were used instead of CRMs in order to minimize use of expensive CRMs; several tests confirmed that this procedure produces as accurate results as using CRMs.

SF 6 and CFC-12
During the cruise, two gas chromatograph/purge-and-trap (GC/PT) systems (PT2 and PT3) were used for the measurements of the transient tracers CFC-12 and SF 6 , similar to the set-up described by Bullister and Wisegarver (2008).For PT3, samples were collected in 250 mL ground glass syringes, and an aliquot of about 200 mL was injected into the analytical system.For PT2, samples were collected in 350 mL glass ampoules, and an aliquot of about 250 mL was injected into the system through a vacuum-sparge technique, similar to that described by Law et al. (1994).Measurements for the first half of the cruise were conducted on instrument "PT2", and later, due to a defect Electron Capture Detector (ECD), conducted on instrument "PT3".During the transition between the instruments, samples were flame-sealed in ∼ 350 mL ampoules.The system "PT3" was not able to measure SF 6 , so that only CFC-12 could be measured on board.However, samples from selected stations in the Western Mediterranean Sea were flame-sealed for the measurement of SF 6 ; all flame-sealed ampoules were measured onshore at the lab at IFM-GEOMAR in Kiel after the cruise.The CFC-12 data are reported on the SIO98 scale and SF 6 on the NOAA-2000 scale.Calibration curves were measured every few days to characterize the non-linearity of the system, and point calibrations were performed regularly to determine the short-term drift in the detector.Replicate measurements were taken on a few stations and the reproducibility was determined to be 1.0 %/0.65 % for CFC-12/SF 6 measurements on PT2, and 0.34 % for CFC-12 measured on PT3, whereas the values are approximately twice as large for the offline measurements of flame-sealed ampoules.A detailed account for the CFC-12 and SF 6 measurements can be found in Stöven (2011).

Helium and tritium
During the cruise, samples for helium isotope were taken for measurements by two different groups: samples were taken at 10 stations along the whole depth profile for measurement of 3 He and tritium at Institute of Environmental Physics at the University of Bremen, Germany.At 13 other stations, samples were taken for the analysis of 3 He at the Laboratoire de Sciences du Climat in LSCE-CEA, Saclay, France.The volume of the helium samples were either 40 or 6.4 mL for the two groups, respectively, and sampled in copper tubes.Samples for tritium analysis were taken in 1 L plastic bottles and Earth Syst.Sci.Data, 5, 289-294, 2013 www.earth-syst-sci-data.net/5/289/2013/ stored into zip-locked plastic bags.At station 301, a brine lake was sampled that contained very high concentrations of radiogenic helium, which lead to contamination of the Niskin bottles so that values for 3 He had to be flagged as questionable for stations 301, 303 and 305.

Carbon isotopes
Samples were collected for post-cruise measurement of carbon isotopes in the dissolved inorganic carbon.A total of 256 results were reported for 13 C and 263 for 14 C.After rinsing, the samples were collected in 500 ml bottles, immediately poisoned with 100 µL saturated HgCl 2 , then sealed according to standard procedures (NOSAMS protocol).The samples were shipped to the National Ocean Sciences AMS Facility at Woods Hole Oceanographic Institution, USA; http://www.whoi.edu/nosams/home.The propagated counting and background uncertainty of the ∆ 14 C analyses averaged about 0.5 ‰, but the total uncertainty based on replicate analyses is about 3.5 ‰ (Elder et al., 1998).This data set represents the first large-scale survey of the carbon-isotope measurements in the Mediterranean Sea.

Auxiliary data
Several additional and complementary measurement and sampling programs were carried out during the cruise, although the data are currently not available in the data repository at CDIAC due to long processing times.These measurements include: surface distribution of persistent organic pollutants, measurements of Polyfluorooctansulfonate, dissolved barium, large volume sample collections for the determination of Ra isotopes, determination of microbiological community structure (Mapelli et al., 2013), determination of the isotopic composition, abundance and size of coccolithophores, aerosol sampling with daily resolution, incubation experiments for nitrogen fixation (Rahav et al., 2013), determination of methyl-mercury, and determination of neodymium isotopes.Information on access to these data can be found at CDIAC; http://cdiac.ornl.gov/oceans/Coastal/Meteor_Med_Sea.html.

Dataset
The data set from Meteor cruise M84/3 is composed of two main components; the continuous profiles from the CTD, and the discrete data from the water samples.The data set contains information from 61 hydrographic stations.Of these, 47 were CTD-cast with chemistry samples, 6 stations were devoted to large volume sampling of Radium isotopes, and 8 stations were only for CTD measurements.
The CTD data are reported as 1 dbar for temperature, salinity and oxygen.
The bottle data are reported in standard World Hydrographic Program (WHP) Exchange format (Swift, 2008).
This format is commonly used for reporting of bottle data, and is the format used by CDIAC and CCHDO, for instance.

Data access
The hydrographic data set is published at two different repositories due to the different scope of the repositories.The bottle data is available at the Carbon Dioxide Information Analysis Center (CDIAC), Oak Ridge, TN, USA, (Tanhua et al., 2012); http://cdiac.ornl.gov/oceans/Coastal/Meteor_Med_Sea.html.
The CTD profile data are available at the CLIVAR and Carbon Hydrographic Data Office (CCHDO), UCSD Scripps Institution of Oceanography, San Diego, CA, USA; http:// cchdo.ucsd.edu/cruise/06MT20110405.

Figure 1 .
Figure 1.Cruise track of M84/3 with CTD stations marked in different colors based on the extent of measurements carried out.
*The shipboard personnel for these measurements are mentioned in the cruise report.hasdifferent properties of salinity and temperature and different biogeochemical signature.The Mediterranean Sea is obviously not in steady state and is potentially sensitive to climatic changes.The characteristic of the Mediterranean Sea is further such that it has the potential to sequester large amounts of anthropogenic CO 2 , C ant , i.e. the Mediterranean Sea has high alkalinity and temperature, which can rapidly be transported to depth by the overturning circulation (e.g.
Tanhua et al.: Repeat hydrography in the Mediterranean Sea