Trends of pH decrease in the Mediterranean Sea through high frequency observational data: indication of ocean acidification in the basin


A significant fraction of anthropogenic carbon dioxide (CO2) released to the atmosphere is absorbed by the oceans, leading to a range of chemical changes and causing ocean acidification (OA). Assessing the impact of OA on marine ecosystems requires the accurate detection of the rate of seawater pH change. This work reports the results of nearly 3 years of continuous pH measurements in the Mediterranean Sea at the Strait of Gibraltar GIFT time series station. We document a remarkable decreasing annual trend of −0.0044 ± 0.00006 in the Mediterranean pH, which can be interpreted as an indicator of acidification in the basin based on high frequency records. Modeling pH data of the Mediterranean outflow allowed to discriminate between the pH values of its two main constituent water masses, the Levantine Intermediate Water (LIW) and the Western Mediterranean Deep Water (WMDW). Both water masses also exhibited a decline in pH with time, particularly the WMDW, which can be related to their different biogeochemical nature and processes occurring during transit time from formation sites to the Strait of Gibraltar.


CO2 emissions from fossil fuels burning and land use change since the industrial revolution have caused a considerable increase in atmospheric CO2 concentrations1. Recent investigations have estimated that cumulative emissions of CO2 have reached in the period from 1870 to 2013 about 535 ± 55 GtC2. However, a significant CO2 amount has been captured from the atmosphere by natural sinks, such as the terrestrial biosphere and the ocean. In particular, the global oceans have absorbed about 30% of the anthropogenic carbon emissions over the past 200 years3.

The withdrawal of CO2 by the oceans has, however, drastic consequences for the marine environment, as it originates a rise in average surface ocean concentration of H+ that leads to a pH decrease in seawater and a range of chemical changes known collectively as “the other CO2 problem” or the ocean acidification (OA) phenomenon4,5. The impact of OA on marine biogeochemical cycles and biota has been well documented by laboratory studies and already observed to occur in certain ocean areas6,7,8,9. It has been suggested that the Mediterranean Sea (MS) represents one of the world’s most sensitive ocean regions to increasing atmospheric CO2 and the subsequent OA10,11. Nevertheless, contradicting observations of pH changes with time have been reported in the basin. A recent study affirms that the MS is already acidified, although distinct OA rates are provided depending on the degree of anthropogenic carbon accumulation by a particular sub-basin, with regional pH decreases oscillating between −0.055 to −0.156 pH units with respect to the preindustrial levels12. On the other hand, another work points out to pH reductions between 0.005 to 0.06 pH units due to the anthropogenic carbon storage in the basin during the same period13. Moreover, it has been proposed that the pH decline would be amplified in the MS due to its higher capacity for CO2 absorption in relation to open ocean regions14,15 and the relatively short ventilation times of its water masses14, statements that were challenged recently by a modelling approach13 which indicates that the average anthropogenic change in surface pH does not differ significantly from the global-ocean average.

Considerable efforts have been made over the last decade to characterize the carbonate system in Mediterranean water masses16,17,18, to explore how much anthropogenic CO2 has been taken up by this semi-enclosed sea and to assess the corresponding pH diminution13,14,19,20. Most of these studies rely on discrete data sets that extend over a specific period of time or cover a particular Mediterranean sub-region. Thus, sometimes, these studies lead to discrepant conclusions when comparisons between data acquired in different periods, using distinct techniques and/or in distant locations are made.

The assessment of the marine ecosystems responses to the oceanic CO2 uptake requires then sustained observations that provide the needed high frequency data about changes in ocean chemistry. At present, this type of measurements is being collected in a few ocean time-series21. The continuous monitoring of the carbon system parameters (and other tracers for hydrography and biochemistry) in these key sites have supplied relevant information on the ocean CO2 sink and the derived pH changes in different regions, confirming a general OA trend over the past two decades.

Sustained time-series observations also started in the Strait of Gibraltar (SG) a decade ago through the establishment of the GIFT (Gibraltar Fixed Time Series) observatory, since this region represents a privileged site to observe the evolution of the Mediterranean waters over time. This narrow channel (14 km wide in its narrowest section) is the only connection of the MS with the North Atlantic, thereby playing a major role in the global circulation and biogeochemistry of the basin22,23.

The circulation pattern in the SG has been traditionally described as a two-layer system, with surface Atlantic water (AW) flowing eastwards to the MS and the Mediterranean Outflow Water (MOW) moving westward to the Atlantic Ocean underneath. The AW enters the MS and flows clockwise in the Alboran Sea (AS, Fig. S1). Subsequent surface circulation patterns at the basin level are influenced by deep and intermediate water formation driven by strong winds, which is in turn affected and amplified by topography. Deep and intermediate waters are formed in four major areas: the Levantine Basin (LB, Fig. S1), the source of the Levantine Intermediate Water (LIW); the Gulf of Lions (GL, Fig. S1) where the Western Mediterranean Deep Water (WMDW) is formed; and two adjacent regions, the Adriatic and the Aegean Seas (AdS, AeS, respectively in Fig. S1), which together merge to form the Eastern Mediterranean Deep Water (EMDW). The MOW that leaves the basin through the Strait of Gibraltar is then a mixture of these intermediate and deep waters, fundamentally LIW, which flows across the Strait of Sicily (SS, Fig. S1) into the Western Mediterranean basin and the WMDW, which occupies the bottom layer24. The contribution of the AW that penetrates into the MS in surface to the final outflow exiting the basin is negligible25. By monitoring the hydrography and biogeochemistry of the MOW in the SG, the history and evolution of the main intermediate and deep Mediterranean water masses can be examined.

The exchange of waters of different ages carrying diverse concentrations of biogeochemical properties in the SG also influences global inventories in the two neighbour regions26,27. Regarding the marine carbon cycle, a net transport of anthropogenic carbon from the Atlantic towards the Mediterranean has been identified19,20,26, which has been indeed responsible for 25% of the basin-wide CO2 uptake over the last 200 years13.

In this work, we use pH measurements taken at the GIFT at a high sampling rate to assess temporal trends of pH change in Mediterranean waters. Data were obtained by autonomous sensors installed in a mooring line deployed at the Espartel Sill (ES, Fig. S1), which has been proven to be the most suitable section in the SG for monitoring the MOW20,24,25,27,28,29. Results presented here correspond to the first continuous pH records at a high temporal resolution registered in the channel from August 2012 to June 2015. Our work provides the first rates of pH decrease in the MOW and in its forming water masses separately, which can be considered indicators of OA in the basin, confirming previous evidence12,13,15. In addition, by using a simple model, we present a tool for tracking pH and its temporal variability in the MS.

Results and Discussion

Water masses and pH

From August 2012 to June 2015, the data collected at the mooring site fluctuated within small ranges of values. Potential temperature () oscillated from 13.01 to 13.63 °C, salinity from 38.01 to 38.48 and pH in total scale at a reference temperature of 25 °C (pHT25) from 7.8618 to 7.9370 (Fig. 1). The pHT25 mean value was 7.8934 ± 0.0076 (n = 15937), which is in good agreement with the average pH in the MOW obtained from sustained spectrophotometric pHT25 measurements taken periodically within this layer at the GIFT stations from 2005 to 2014 (n = 102) and equivalent to 7.8875 ± 0.0124.

Figure 1

(a) Potential temperature (), (b) Salinity obtained with the CT and (c) SAMI-pH data from August 2012 to June 2015.

Water masses fractions variability in the MOW, discriminated with an OMP (Optimum MultiParameter) analysis30, (see SI text for more details), clearly describes seasonal and interannual fluctuations (Fig. 2). During the monitoring period, the LIW appeared to dominate the outflow most of the time, with a mean fraction of 0.55 ± 0.1 (Fig. 2a) whereas the WMDW showed a fraction of 0.36 ± 0.1 (Fig. 2b). Because of the sampling depth, the presence of the AW (Fig. 2c) was almost negligible within the MOW, with an average fraction of 0.09 ± 0.03, confirming historic observations25.

Figure 2

Fractions of the water masses forming the MOW during the monitoring period, according to the OMP analysis (see text): (a) LIW, (b) WDMW and (c) AW.

As previously described24,31, WMDW ventilation through the SG is modulated by several physical processes among which the WMDW formation events in the Gulf of Lions and the intensification of the Western Alboran Gyre (WAG, red arrow in Fig. S1) in the Alboran Sea are probably the most relevant. The WAG provides additional energy necessary to uplift deep waters from the Alboran Sea, facilitating the arrival of WMDW to the eastern entrance of the SG31. On the contrary, the absence or the relaxation of the WAG propitiates a major drainage of the LIW. Both processes can be traced in the monitoring station by gradual drops of potential temperature to values below 13.1°C24. During the monitoring period, such fall was clearly registered in January 2013 (Fig. 1a), which was accompanied by a concomitant rise in the WMDW fraction and a decrease in the LIW fraction (Fig. 2a,b). A similar pattern was also observed in February 2014 and February 2015 (Figs 1a and 2a,b). In a previous study that also used potential temperature as the parameter to trace water masses in the area24, WDMW formation events left a noticeable signature around March. Following the procedure described by these authors, the densest sample recorded in every semidiurnal tidal cycle was then extracted from our database to obtain a new subseries with semidiurnal sampling interval (not shown). In this new subseries, the major diminutions were found to commence by the middle of January 2013 and beginning of February 2014, with a gradual drop from ~13.2°C to values below 13.05 °C taking place in around 9 days in year 2013 and 13 days in 2014. The original temperature was recovered a month later approximately, a fact that may be partially attributed to the relaxation of the WAG.

Modelled data

As shown in Fig. 3a, the pH data modelled by a MLR (MultiLinear Regression, see SI text) faithfully reproduced in situ values (averaged to 84 h for comparison) and discrepancies between observational data and modelled outputs were in the order of ±0.005 pH units. Averaging such discrepancies to a 6 h period (Fig. 3b), residuals followed the tidal cycle pattern, thereby confirming that tidal variability was excluded in our estimations. However, higher residuals could still be detected during some periods. For instance, a sharp decline was evident in June 2013 (Fig. 3b), when pH data were recalculated from AT and pCO2 measurements. This signal coincides with a rise in the pHT25 values (Fig. 1c) and it could be then attributed to the SAMI-CO2 instrument stabilization time. A second period of higher residual values was observed in October 2013 lasting until April 2014 (Fig. 3b), which may be attributable in this case to the change in the SAMI-pH sampling interval from hourly to bi-hourly. pCO2 data measured uninterruptedly in the MOW from June 2013 to December 2014 by a SAMI-CO2 device were also modelled by the MLR (see Fig. S3), with very low residuals being obtained, which supports the model robustness.

Figure 3

(a) pHT25 obtained with the SAMI device averaged to 84 h (red line) and modelled pHT25 (black line), (b) Residuals between observed values and modelled outputs.

As each water mass is characterized by a distinctive salinity and potential temperature, pH can also act as a tracer to define water masses. The OMP and MLR analysis allowed determining that LIW and WDMW in the SG were characterized by average pHT25 and standard error values of 7.8897 ± 0.0003 and 7.9077 ± 0.0004, respectively (see Fig. 4). Those values faithfully correspond to recently reported measures18 equal to ~7.89 for the LIW in the SG and 7.9–7.91 for the WMDW in the Western Mediterranean basin. Differences in pH values betwen both water masses can be explained on the basis of the transit times from their respective formation sites to the SG; the LIW takes around 8 years to complete the distance from the Levantine basin (LB, Fig. S1) to the Strait of Sicile (SS, Fig. S1)32. An active remineralization of organic matter can take place during such period, which implies the rise of dissolved inorganic carbon concentration, the decrease of dissolved oxygen and the consequent pH decrease16,18. In contrast, the WMDW takes roughly 1.8 years to travel from the Gulf of Lions to the Alboran Sea (GL and AS respectively in Fig. S1), a much shorter transit time33. Although the ventilation time of LIW and WMDW is an issue that has not been solved yet and there are still large discrepancies regarding the age of each water mass34,35,36,37, estimated ages are around 80–120 ± 20 yr and 20–40 ± 40 yr for the bottom waters of the eastern and western basins, respectively, when they arrive at the SG35. Therefore, considering the circulation pattern in the MS38 and regardless of a particular transit time or age for each water mass, when they both arrive at the SG the WMDW is much younger than the LIW formed in the distant Levantine Basin (LB, Fig. S1). Accordingly, the LIW is more stable than the WMDW from a biogechemical point of view, as the former lost contact with the atmosphere a longer ago. Therefore, in the SG, the WMDW,which has been exposed to the atmosphere more recently, exhibits higher pH values than the eastern-originated LIW (Fig. 4). This is confirmed when the monthly variability of pH in both water masses is analysed (see Fig. S2). Although a little seasonality can be detected, over the summer months, the two water masses showed stable pH values whereas winter conditions resulted in the highest pH variability, which was especially remarkable in February. During this month, pH in the WMDW presented noticeable oscillations, ranging from 7.7789 to 8.0003 whereas the pH of the LIW varied slightly, with values changing from 7.8195 to 7.9634. This seasonal variability could be again attributed to the ventilation pattern of the WMDW. A previous work25 demostrated that a lag of few weeks can be found between the events of WMDW fomation in winter in the Gulf of Lions and the detection of cold pulses of old WMDW at the monitoring site in Espartel Sill. Renovation of the old resident WMDW in the SG depends on the volume of water formed during winter by deep convection in the Gulf of Lions and the intensification of the WAG that subsequently uplifts the ancient WMDW at the SG. During our study period, the appearance of colder (and older) WDMW pulses in February 2013, 2014 and 2015 (Figs 1a and 2b) possibly resulted in the lowest pH values detected in the WMDW during the months of February (see Fig. S2b), as the older WMDW residing in the eastern side of the Strait will be characterized by lower pH values, due to the active remineralization processes occurring in the Alboran basin39.

Figure 4

Linear fitting of pH with time (see SI text) of the MOW and its forming water masses during the monitoring period: (a) MOW, (b) LIW and (c) WMDW. Blue and red lines represent the 95% confidence and prediction bands, respectively. Equations are shown in the SI text. Note the different scales for “y” axes in figures (a–c).

Trends of pH decrease in Mediterranean waters

The pH data series clearly depicted a negative trend (Fig. 1) although the data gap after the first period (April-June 2013) seems to visually break the tendency. From June 2013 onwards, the decreasing trend was even clearer. Taking into account the whole period, a pH-time linear regression was calculated (see SI text for calculation details) and plotted in Fig. 4. The regression statistics were significant and resulted in a ΔpH of −0.0044 units per year in the MOW. This rate of pH decline is of the same order of magnitude than acidification rates reported by sustained observations of different carbon system parameters collected in several oceanic seasurface time series21, although it is two or three fold higher than those (depending on the site). Nevertheless, it still falls within the range of pH change in Mediterranean deep waters estimated recently through a modelling approach (−0.005 to −0.06 pH units13).

The consistency of our pH decreasing pattern is supported by the complementary pCO2 measurements taken in the SG at the same sampling frequency. When the ΔpCO2 in the MOW obtained during the monitoring period is calculated, an increase in pCO2 of 5.1 μatm y−1 is obtained (see SI text and Fig. S4 for calculation details). This CO2 rise is coherent with (and matches) the rate of pH decline observed in the MOW.

Performing a separate analysis for each of the two main water masses forming the MOW, annual decline rates of −0.0006 and −0.0165 pH units per year are calculated for the LIW and WMDW, respectively (Fig. 4b,c). The regression descriptors were statistically significant (see SI text). These pH changes correspond to pCO2 variations of −0.6 and 18.5 μatm in each water mass respectively (see Fig. S4), according to the CO2 data in the SG.

The low ΔpHLIW may be well attributable to its age (in the order of 100 years upon arrival at the SG) and stability, as this intermediate water mass has not been affected by the present atmopheric CO2 concentrations. On the other hand, the more recently produced WMDW showed a considerable pH decreasing trend and a noticeable pCO2 rise. As previously noted, this may be a consequence of several processes. First, the impact of higher atmospheric CO2 concentration this water mass was exposed to when formed, secondly, the faster penetration of CO2 in the water column brought about by the high alkalinity of this water mass40 and finally the sinking of labile and easily oxidizable organic matter facilitated by the formation events. Also, both LIW and WMDW reflect the gradient of trophic conditions from the Western to the Eastern Mediterranean basin. With a eastward decreasing gradient pattern of primary productivity41, the Alboran Sea represents one of the most productive Mediterranean areas41 involving a greater deep degradation potential of organic inputs coming from the photic zone42. Therefore, the WMDW, support high rates of organic carbon degradation42,43 and subsequently an increase on the CO2 content while residing in the deep Alboran Sea.

Observations gathered in other time series have shown that ocean regions characterized by high rates of increase in pCO2 (Irminger Sea and CARIACO, for instance) depict the highest rates of decrease in surface seawater pH (around −0.0025 pH units y−1)21.

Even though the time period of observations used in our work (roughly 3 years) can be considered relatively short to asses long term changes in the pH of the Mediterranean Sea, it still provides a reasonable wealth of data with sufficient sensitivity and accuracy to establish seasonal and interannual pH variations in the water masses of the Mediterranean Sea that reflect changes in both the natural carbon cycle and anthropogenic perturbation. It is worthy to point out that, in addition to the uptake of anthropogenic CO2 from the atmosphere, the Mediterranean basin receives continuously a considerable amount of anthropogenic carbon from the North Atlantic Ocean through the SG20. Monitoring the magnitude of the resulting pH decline in the basin requires sustained high-accuracy observations. The data presented here may well serve as a base line to assess sensitiviy and evolution of the Mediterranean waters to the impact of increasing CO2 emmissions.


pH measurements shown in this study were collected from August 2012 to June 2015 at the GIFT time series in the Strait of Gibraltar (see SI text for mooring line details, Fig. S1). Water masses fractions within the MOW were obtained by an Optimum MultiParameter (OMP) analysis and modelled pH was constructed by performing a Multiple Linear Regression (MLR) least square fitting. In order to smooth out the short-scale variability and turbulence originated in the supercritical-to-subcritical flow transitions44 as well as the fluctuations associated with the tidally-generated short internal waves in the SG, which result mostly in the rapid mixing between water masses, data were averaged in periods of 84 hours (half week), as our work was meant to focus on long-term variability related to basin scale processes, which is relevant for OA signals. Details regarding the CO2 system calculations, data sets used and the statistical treatment may be found in SI (see SI text).

Additional Information

How to cite this article: Flecha, S. et al. Trends of pH decrease in the Mediterranean Sea through high frequency observational data: indication of ocean acidification in the basin. Sci. Rep. 5, 16770; doi: 10.1038/srep16770 (2015).


  1. Cubasch, U. et al. in Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change (eds T. F. Stocker et al. ) Ch. 1, 119–158 (Cambridge University Press, 2013).

  2. Le Quéré, C. et al. Global carbon budget 2013. Earth System Science Data 6, 235–263, doi: 10.5194/essd-6-235-2014 (2014).

    ADS  Article  Google Scholar 

  3. Le Quéré, C., Takahashi, T., Buitenhuis, E. T., Rödenbeck, C. & Sutherland, S. C. Impact of climate change and variability on the global oceanic sink of CO2 . Glob. Biogeochem. Cycle 24, GB4007, doi: 10.1029/2009GB003599 (2010).

    CAS  ADS  Article  Google Scholar 

  4. Gattuso, J.-P., Mach, K. & Morgan, G. Ocean acidification and its impacts: an expert survey. Climatic Change 117, 725–738, doi: 10.1007/s10584-012-0591-5 (2013).

    CAS  ADS  Article  Google Scholar 

  5. Doney, S. C., Fabry, V. J., Feely, R. A. & Kleypas, J. A. Ocean acidification: the other CO2 problem. Annual Review of Marine Science 1, 169–192, doi: 10.1146/annurev.marine.010908.163834 (2009).

    ADS  Article  Google Scholar 

  6. Fabry, V. J., Seibel, B. A., Feely, R. A. & Orr, J. C. Impacts of ocean acidification on marine fauna and ecosystem processes. ICES Journal of Marine Science: Journal du Conseil 65, 414–432, doi: 10.1093/icesjms/fsn048 (2008).

    CAS  Article  Google Scholar 

  7. Orr, J. C. et al. Anthropogenic ocean acidification over the twenty-first century and its impact on calcifying organisms. Nature 437, 681–686, doi: 10.1038/nature04095 (2005).

    CAS  ADS  Article  Google Scholar 

  8. Feely, R. A., Doney, S. C. & Cooley, S. R. Ocean acidification: present conditions and future changes in a high-CO2 world. Oceanography 22, 36–47 (2009).

    Article  Google Scholar 

  9. Ríos, A. F. et al. Decadal acidification in the water masses of the Atlantic Ocean. Proceedings of the National Academy of Sciences, doi: 10.1073/pnas.1504613112 (2015).

  10. Bramanti, L. et al. Detrimental effects of ocean acidification on the economically important Mediterranean red coral (Corallium rubrum). Global Change Biology 19, 1897–1908, doi: 10.1111/gcb.12171 (2013).

    CAS  ADS  Article  PubMed  Google Scholar 

  11. Calvo, E. et al. Effects of climate change on Mediterranean marine ecosystems: the case of the Catalan Sea. Climate Research 50, 1–29 (2012).

    ADS  Article  Google Scholar 

  12. Hassoun, A. E. R. et al. Acidification of the Mediterranean Sea from anthropogenic carbon penetration. Deep Sea Research Part I: Oceanographic Research Papers 102, 1–15 (2015).

    CAS  ADS  Article  Google Scholar 

  13. Palmiéri, J. et al. Simulated anthropogenic CO2 storage and acidification of the Mediterranean Sea. Biogeosciences 12, 781–802, doi: 10.5194/bg-12-781-2015 (2015).

    CAS  ADS  Article  Google Scholar 

  14. Touratier, F. & Goyet, C. Impact of the eastern mediterranean transient on the distribution of anthropogenic CO2 and first estimate of acidification for the Mediterranean Sea. Deep Sea Research Part I: Oceanographic Research Papers 58, 1–15 (2011).

    CAS  ADS  Article  Google Scholar 

  15. Touratier, F. & Goyet, C. Decadal evolution of anthropogenic CO2 in the northwestern Mediterranean Sea from the mid-1990s to the mid-2000s. Deep-Sea Research Part I-Oceanographic Research Papers 56, 1708–1716, doi: 10.1016/j.dsr.2009.05.015 (2009).

    CAS  ADS  Article  Google Scholar 

  16. Rivaro, P., Messa, R., Massolo, S. & Frache, R. Distributions of carbonate properties along the water column in the Mediterranean Sea: Spatial and temporal variations. Marine Chemistry 121, 236–245 (2010).

    CAS  Article  Google Scholar 

  17. Touratier, F. et al. Distributions of the carbonate system properties, anthropogenic CO2 and acidification during the 2008 BOUM cruise (Mediterranean Sea). Biogeosciences Discussions 9, 2709–2753 (2012).

    ADS  Article  Google Scholar 

  18. Álvarez, M. et al. The CO2 system in the Mediterranean Sea: a basin wide perspective. Ocean Sci. 10, 69–92, doi: 10.5194/os-10-69-2014 (2014).

    CAS  ADS  Article  Google Scholar 

  19. Schneider, A., Tanhua, T., Körtzinger, A. & Wallace, D. W. High anthropogenic carbon content in the eastern Mediterranean. Journal of Geophysical Research: Oceans (1978–2012) 115, C12050, doi: 10.1029/2010JC006171 (2010).

    CAS  ADS  Article  Google Scholar 

  20. Huertas, I. E. et al. Anthropogenic and natural CO2 exchange through the Strait of Gibraltar. Biogeosciences 6, 647–662 (2009).

    CAS  ADS  Article  Google Scholar 

  21. Bates, N. R. et al. A time-series view of changing ocean chemistry due to ocean uptake of anthropogenic CO2 and ocean acidification. Oceanography 27, 126–141, doi: 10.5670/oceanog.2014.16 (2014).

    Article  Google Scholar 

  22. Millot, C. Circulation in the western Mediterranean Sea. Journal of Marine Systems 20, 423–442 (1999).

    ADS  Article  Google Scholar 

  23. Millot, C. & Taupier-Letage, I. in The Mediterranean Sea 29–66 (Springer, 2005).

  24. García‐Lafuente, J. et al. Interannual variability of the Mediterranean outflow observed in Espartel sill, western Strait of Gibraltar. Journal of Geophysical Research: Oceans (1978–2012) 114, C10018, doi: 10.1029/2009JC005496 (2009).

    Article  Google Scholar 

  25. García Lafuente, J., Sánchez Román, A., Díaz del Río, G., Sannino, G. & Sánchez Garrido, J. Recent observations of seasonal variability of the mediterranean outflow in the Strait of Gibraltar. Journal of Geophysical Research: Oceans (1978–2012) 112, C10005, doi: 10.1029/2006JC003992 (2007).

    ADS  Article  Google Scholar 

  26. Flecha, S. et al. Anthropogenic carbon inventory in the Gulf of Cádiz. Journal of Marine Systems 92, 67–75, doi: 10.1016/j.jmarsys.2011.10.010 (2012).

    ADS  Article  Google Scholar 

  27. Huertas, I. E. et al. Atlantic forcing of the Mediterranean oligotrophy. Global Biogeochem. Cycles 26, GB2022, doi: 10.1029/2011gb004167 (2012).

    ADS  Article  Google Scholar 

  28. Sánchez-Román, A., Sannino, G., García-Lafuente, J., Carillo, A. & Criado-Aldeanueva, F. Transport estimates at the western section of the Strait of Gibraltar: A combined experimental and numerical modeling study. Journal of Geophysical Research: Oceans 114, C06002, doi: 10.1029/2008JC005023 (2009).

    ADS  Article  Google Scholar 

  29. García‐Lafuente, J., Sánchez‐Román, A., Naranjo, C. & Sánchez‐Garrido, J. C. The very first transformation of the Mediterranean outflow in the Strait of Gibraltar. Journal of Geophysical Research: Oceans (1978–2012), C07010, 116, doi: 10.1029/2011JC006967 (2011).

    Article  Google Scholar 

  30. Tomczak, M. & Large, D. G. B. Optimum multiparameter analysis of mixing in the thermocline of the eastern Indian Ocean. Journal of Geophysical Research: Oceans 94, 16141–16149, doi: 10.1029/JC094iC11p16141 (1989).

    Article  Google Scholar 

  31. Naranjo, C., García Lafuente, J., Sánchez Garrido, J. C., Sánchez Román, A. & Delgado Cabello, J. The western alboran gyre helps ventilate the western mediterranean deep water through Gibraltar. Deep Sea Research Part I: Oceanographic Research Papers 63, 157–163 (2012).

    ADS  Article  Google Scholar 

  32. Roether, W., Klein, B., Beitzel, V. & Manca, B. B. Property distributions and transient-tracer ages in Levantine Intermediate Water in the Eastern Mediterranean. Journal of Marine Systems 18, 71–87 (1998).

    ADS  Article  Google Scholar 

  33. Schmidt, S. & Reyss, J.-L. Radium as internal tracer of Mediterranean Outflow Water. Journal of Geophysical Research: Oceans 101, 3589–3596, doi: 10.1029/95JC03308 (1996).

    CAS  Article  Google Scholar 

  34. Andrie, C. & Merlivat, L. Tritium in the western Mediterranean Sea during 1981 Phycemed cruise. Deep Sea Research Part A. Oceanographic Research Papers 35, 247–267, doi: 10.1016/0198-0149(88)90039-8 (1988).

    CAS  ADS  Article  Google Scholar 

  35. Stratford, K., Williams, R. G. & Drakopoulos, P. G. Estimating climatological age from a model-derived oxygen–age relationship in the Mediterranean. Journal of Marine Systems 18, 215–226, doi: 10.1016/S0924-7963(98)00013-X (1998).

    ADS  Article  Google Scholar 

  36. Stratford, K. & Williams, R. G. A tracer study of the formation, dispersal and renewal of Levantine Intermediate Water. Journal of Geophysical Research: Oceans 102, 12539–12549, doi: 10.1029/97JC00019 (1997).

    Article  Google Scholar 

  37. Rhein, M. Deep water formation in the western Mediterranean. Journal of Geophysical Research: Oceans 100, 6943–6959, doi: 10.1029/94JC03198 (1995).

    CAS  Article  Google Scholar 

  38. Millot, C. Another description of the Mediterranean Sea outflow. Progress in Oceanography 82, 101–124 (2009).

    ADS  Article  Google Scholar 

  39. Minas, H. J., Coste, B., Le Corre, P., Minas, M. & Raimbault, P. Biological and geochemical signatures associated with the water circulation through the Strait of Gibraltar and in the western Alboran Sea. Journal of Geophysical Research: Oceans 96, 8755–8771, doi: 10.1029/91JC00360 (1991).

    Article  Google Scholar 

  40. Copin-Montégut, C. & Bégovic, M. Distributions of carbonate properties and oxygen along the water column (0–2000m) in the central part of the NW Mediterranean Sea (Dyfamed site): influence of winter vertical mixing on air–sea CO2 and O2 exchanges. Deep Sea Research Part II: Topical Studies in Oceanography 49, 2049–2066 (2002).

    ADS  Article  Google Scholar 

  41. Skliris, N. & Beckers, J.-M. Modelling the Gibraltar Strait/Western Alboran Sea ecohydrodynamics. Ocean Dynamics 59, 489–508 (2009).

    ADS  Article  Google Scholar 

  42. Sanchez-Vidal, A., Calafat, A., Canals, M., Frigola, J. & Fabres, J. Particle fluxes and organic carbon balance across the Eastern Alboran Sea (SW Mediterranean Sea). Continental Shelf Research 25, 609–628, doi: 10.1016/j.csr.2004.11.004 (2005).

    ADS  Article  Google Scholar 

  43. Luna, G. M. et al. The dark portion of the Mediterranean Sea is a bioreactor of organic matter cycling. Glob. Biogeochem. Cycle 26, n/a-n/a, doi: 10.1029/2011GB004168 (2012).

  44. Sánchez-Garrido, J. C., Sannino, G., Liberti, L., García Lafuente, J. & Pratt, L. Numerical modeling of three-dimensional stratified tidal flow over Camarinal Sill, Strait of Gibraltar. Journal of Geophysical Research: Oceans 116, C12026, doi: 10.1029/2011JC007093 (2011).

    ADS  Article  Google Scholar 

Download references


The cooperation of the captains and crews of the research vessels “Cornide de Saavedra”, “Angeles Alvariño”, “Ramon Margalef” and “Sarmiento de Gamboa” is greatly appreciated. We also thank María Ferrer, Antonio Moreno, David Roque and Cristina Naranjo for assistance during mooring operations. Funding for this work was provided by the European Commission through the projects CARBOCHANGE (FP7-264879) and PERSEUS (FP7-287600) and by the Spanish Ministry of Economy and Competitiveness (CTM2010-21229). SF was supported by a JAE PREDOCTORAL scholarship, part-funded by the European Commission (European Social Fund, ESF2007-2013) and the Spanish Ministry for Economy and Competitiveness.

Author information




S.F. and I.E.H. wrote the manuscript text and supporting information. S.F. prepared all the figures. S.F., F.F.P. and S.S. analized the data. I.E.H. and J.G.-L. have contributed to the conception and design of the study. F.F.P., J.G.-L. and A.F.R. have contributed in the results discussion. All authors reviewed the manuscript and supporting information. S.F. and I.E.H. made equal contributions.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Electronic supplementary material

Rights and permissions

This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Flecha, S., Pérez, F., García-Lafuente, J. et al. Trends of pH decrease in the Mediterranean Sea through high frequency observational data: indication of ocean acidification in the basin. Sci Rep 5, 16770 (2015).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.