BGBiogeosciencesBGBiogeosciences1726-4189Copernicus PublicationsGöttingen, Germany10.5194/bg-14-1197-2017High-resolution regional modelling of natural and anthropogenic radiocarbon in the Mediterranean SeaAyacheMohamedmohamed.ayache@lsce.ipsl.frhttps://orcid.org/0000-0002-2965-3377DutayJean-Claudehttps://orcid.org/0000-0003-3306-9015MouchetAnnehttps://orcid.org/0000-0002-8846-3063Tisnérat-LabordeNadinehttps://orcid.org/0000-0002-7697-3605MontagnaPaoloTanhuaTosteSianiGiuseppeJean-BaptistePhilippeLaboratoire des Sciences du Climat et de l'Environnement LSCE/IPSL,
CEA-CNRS-UVSQ, Université Paris-Saclay, 91191 Gif-sur-Yvette, FranceIstituto di Scienze Marine, CNR, Bologna, ItalyGEOMAR Helmholtz Centre for Ocean Research Kiel, Dusternbrooker Weg 20, 24105 Kiel, GermanyGEOPS Geoscience Paris Sud UMR 8148, Orsay, FranceMohamed Ayache (mohamed.ayache@lsce.ipsl.fr)13March20171451197121317October201625October201615February201717February2017This work is licensed under a Creative Commons Attribution 3.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by/3.0/This article is available from https://bg.copernicus.org/articles/14/1197/2017/bg-14-1197-2017.htmlThe full text article is available as a PDF file from https://bg.copernicus.org/articles/14/1197/2017/bg-14-1197-2017.pdf
A high-resolution dynamical model (Nucleus for European Modelling of the
Ocean, Mediterranean configuration – NEMO-MED12) was used to give the first
simulation of the distribution of radiocarbon (14C) across the whole
Mediterranean Sea. The simulation provides a descriptive overview of both the
natural pre-bomb 14C and the entire anthropogenic radiocarbon transient
generated by the atmospheric bomb tests performed in the 1950s and early
1960s. The simulation was run until 2011 to give the post-bomb distribution.
The results are compared to available in situ measurements and proxy-based
reconstructions. The radiocarbon simulation allows an additional and
independent test of the dynamical model, NEMO-MED12, and its performance to
produce the thermohaline circulation and deep-water ventilation. The model
produces a generally realistic distribution of radiocarbon when compared with
available in situ data. The results demonstrate the major influence of the
flux of Atlantic water through the Strait of Gibraltar on the inter-basin
natural radiocarbon distribution and characterize the ventilation of
intermediate and deep water especially through the propagation of the
anthropogenic radiocarbon signal. We explored the impact of the interannual
variability on the radiocarbon distribution during the Eastern Mediterranean
Transient (EMT) event. It reveals a significant increase in 14C
concentration (by more than 60 ‰) in the Aegean deep water and at an
intermediate level (value up to 10 ‰) in the western basin. The
model shows that the EMT makes a major contribution to the accumulation of
radiocarbon in the eastern Mediterranean deep waters.
Introduction
The Mediterranean region has been identified as a hot spot for future
climatic changes
. Because
this midlatitude almost enclosed sea is surrounded by countries with high
population growth to the south and highly industrialized countries to the
north, it is under strong anthropogenic pressures. This stress is expected to
intensify due to factors such as warming and substantial precipitation
decrease . In the context of global change
we need to improve our understanding of how changes in the climate and
circulation of the Mediterranean Sea interact with the biogeochemical
processes that define its functioning.
The Mediterranean Sea can be considered as a “miniature ocean”, where
global change can be studied on smaller/shorter spatial and temporal scales,
e.g. warming at intermediate water depths in the Mediterranean Sea is about
10 times larger than trends reported in literature
e.g.. The Mediterranean Sea has a
well-defined overturning circulation with distinct surface, intermediate, and
deep-water masses circulating in the western and the eastern basins and
varying on interannual timescales. This makes it an excellent test bed for
studying basic processes that will also affect the global thermohaline
circulation.
The Mediterranean is a concentration basin in which evaporation exceeds
precipitation and river runoff. Warmer, fresher water enters at the surface
from the Atlantic (Atlantic water – AW) through Gibraltar and colder saline water leaves
below. Relatively fresh waters of Atlantic origin circulating in the
Mediterranean increase in density and then form new water masses via
convection events driven by intense local cooling from winter storms. The
Levantine Intermediate Water (LIW) represents one of the main water masses of
the Mediterranean Sea. It spreads throughout the entire Mediterranean basin
at intermediate depths (between ∼ 150 and 700 m) and is the major contributor to the Mediterranean outflow into the North
Atlantic (Bryden and Stommel, 1984). Furthermore, the LIW participates in the
deep convection processes of the western Mediterranean deep water (WMDW)
occurring in the Gulf of Lion and in the Adriatic sub-basin for the eastern
Mediterranean deep water (EMDW) . The formation of deep
water in the Mediterranean Sea is also characterized by interannual/decadal
variability such as the Eastern Mediterranean Transient (EMT) event, known to
create a major shift in deep-water formation in the east Mediterranean Sea
(EMed) at the beginning of the 1990s . The EMT
describes a change in the formation site for EMDW, when it temporarily
switched from the Adriatic to the Aegean sub-basin.
In many respects, the most useful diagnostics of the ventilation of the
ocean's interior come from geochemical tracers characterized by simple
boundary conditions at the ocean's surface and conservation in deep water
. In particular, the
passive transient tracers (CFC and tritium) do not affect the water-mass
densities (as opposed to active tracers such as temperature and salinity).
Radiocarbon (14C) is an ideal tracer for studying air–sea gas exchange and for assessing the ventilation rate of the deep-water masses on very long
timescales . Although 14C is
affected by biological processes, especially remineralization of organic
matter, this effect can be considered minimal for the present simulation.
Radiocarbon (14C) is naturally formed by the reaction between nitrogen
atoms in the atmosphere and slow-moving neutrons produced by a whole cascade
of nuclear reactions between cosmic radiation and molecules in the upper
atmosphere. Radiocarbon is not produced in the ocean's interior. All 14C
enters the ocean from the atmosphere through gas exchange with surface water
with an equilibration time of 7–10 years .
Radioactive decay of 14C (the half-life being 5730 years) reduces its
concentration over time in the water column. Over the last 150 years, the
natural distribution of radiocarbon has been disturbed by (i) the dilution of
atmospheric 14C by the release of fossil fuel CO2, depleted in
14C (the Suess Effect; Suess, 1955), and (ii) the production of bomb
14C by thermonuclear weapon testing in the late 1950s and early 1960s.
The latter strongly increased the 14C levels in the atmosphere (Rafter
and Fergusson, 1957) and consequently the gradient between surface and
subsurface waters e.g..
Knowledge of the timescale of the thermohaline circulation is of central
importance in the debate on the sequestration of anthropogenic carbon in the
deep ocean. Unlike the other tracers (e.g. CFC and Tritium), the radiocarbon
concentration in the oceanic water masses is an invaluable tool allowing us
to study the thermohaline circulation from the seasonal cycle, i.e. the
near-surface circulation, vertical transport, and mixing , on decadal and centennial timescales
e.g.. Radiocarbon plays a crucial role in
carbon cycle investigations allowing us to assess the carbon fluxes between
reservoirs e.g. and the description of the air–sea gas
exchange process e.g..
Understanding the spatiotemporal variation of radiocarbon
allows us to determine the ages of different water
masses and to establish the overturning timescale and water-mass renewal time
for individual basins and the global ocean e.g..
Unlike other tracers, such as tritium, the 14C in ocean surface water is
not in equilibrium with atmosphere; this means that the surface ocean does
not have the same 14C age as the atmosphere (i.e. not zero age). This
difference, also known as the “radiocarbon reservoir age”, is caused both by
the delay in exchange rates between atmospheric CO2 and the carbonate
system and the dilution effect due to the mixing of
surface waters with intermediate or deep waters depleted in 14C during
seasonal vertical convection or upwelling, respectively. Indeed, when surface
waters are isolated from the atmosphere, the radiocarbon clock begins to tick
and 14C content of water gradually decays.
Radiocarbon observations have played a crucial role as an experimental tool
revealing the spatial and temporal variability of carbon sources and sinks
. Observational programmes (e.g. GEOSECS, WOCE and TTO)
have provided snapshots of the large-scale distribution of radiocarbon in the
world's oceans. However, few 14C measurements have been made in the
Mediterranean. provided the first characterization of the
natural radiocarbon in the surface and intermediate waters of the whole
Mediterranean Sea from in situ observations. More studies tried to determine
the sea-surface radiocarbon reservoir ages of the Mediterranean, which are
mainly affected by the Atlantic surface waters entering at Gibraltar and/or
by local factors related to freshwater input from rivers .
The first 14C reservoir age (360 ± 80 yr) was calculated by
using the pre-bomb shells collected along the Algerian
continental shelf. Later, Delibrias (1985) obtained an average 14C
reservoir age of 350 ± 35 yr through the analysis of pre-bomb mollusc
shells from the French and Algerian shelves. The average marine reservoir age
for the whole Mediterranean Sea was estimated by to be some
390 ± 85 yr.
Finally, mollusc shells were also used to yield a more significant dataset
for the Mediterranean Sea with a mean sea-surface reservoir 14C age of
400 ± 16 yr . More recently, the first
annually resolved sea-surface 14C record was obtained from a 50-year-old
shallow-water coral (Cladocora caespitosa) from the western
Mediterranean Sea, covering the pre- and post-bomb period
. However, all these observations are discrete in
time and/or space; they cannot give a clear description of radiocarbon
evolution between the past and the actual situation now on either the
regional or the global scale.
Although box models have been extensively used to quantify the radiocarbon
inventory e.g., their application in
deriving the oceanic distribution of radiocarbon is limited due to their very
simple parameterization. On the other hand, numerical modelling gives us a
clear 4-D description of the water column, which provides an additional
opportunity to better understand the 14C distribution in seawater.
Several different ocean models have previously been used to study the global
radiocarbon distribution e.g.. However, these studies used coarse-resolution models
which could not satisfactorily represent the critical spatial and temporal
scales of circulation in the Mediterranean Sea.
Here, we used a high-resolution regional model (Nucleus for European Modelling of the Ocean, Mediterranean configuration – NEMO-MED12; horizontal
resolution 1/12∘; ∼ 7 km) of the entire Mediterranean Sea
. This model has been used previously for
biogeochemical studies
and
dynamical application . Here, we use the model to provide the first
simulation of radiocarbon distribution and the related reservoir age. The
simulation covers the different states of 14C from the steady natural
distribution to the Suess effect, the 14C bomb peak in the 1960s, and
the post-bomb distribution until 2011. Our model results are compared to
available 14C measurements of seawater and marine carbonates reported by
, , , and
and to a 50-year high-resolution 14C record obtained
from a shallow-water coral specimen .
Our work highlights the impact of anthropogenic perturbation (14C bomb
peak and the Suess effect) on the radiocarbon distribution across the whole
Mediterranean Sea, as well as the regional response across the different
sub-basins. In addition, the simulation provides (i) constraints on the
14C air–sea transfer; (ii) a descriptive overview of the Mediterranean
14C distribution, which gives an additional improvement of in situ data
interpretation; and (iii) more perspectives on the impact of the interannual
variability of the Mediterranean thermohaline circulation (e.g. EMT event)
on the modelled 14C distribution. The present radiocarbon simulation
aims at improving the knowledge of the natural distribution of 14C in
the Mediterranean Sea and implementing a geochemical tracer with a longer
timescale allowing more paleo-oriented applications.
Furthermore, this study is part of the work under way to assess the
robustness of the NEMO-MED12 model and its use in studying the thermohaline
circulation and the biogeochemical cycles in the Mediterranean Sea. The
overarching objective of this work is to predict the future evolution of this
basin under the increasing anthropogenic pressure.
MethodCirculation model
The NEMO is a free
surface-ocean circulation model . Here, it is used in its
Mediterranean configuration called NEMO-MED12 with a
horizontal resolution 1/12∘ (∼ 7 km) and 50 vertical
z coordinates ranging from 1 m at the surface to 450 m at depth with
partial-step formulation.
NEMO-MED12 covers the whole Mediterranean Sea and includes part of the near
Atlantic Ocean (buffer zone) from 11∘ W to 36∘ E and from
30 to 47∘ N. The exchange with the Atlantic Ocean occurs through
this buffer zone, where 3-D salinity and temperature fields are relaxed to
the observed climatology . The sea-surface height (SSH)
is restored in the buffer zone from the GLORYS1 reanalysis
in order to conserve the Mediterranean Sea water volume. The Black Sea is not
explicitly represented in NEMO-MED12 configuration; exchanges with the Black
Sea consist of a two-layer flow corresponding to the Dardanelles' net budget
estimates of .
The atmospheric forcing of NEMO-MED12 is provided by daily mean fields of
momentum, freshwater, and heat fluxes from the high-resolution atmospheric
model (ARPERA) over the period 1958–2013 .
The sea-surface temperature (SST) and water-flux correction term are applied
using ERA-40 . River runoff is derived from the
interannual dataset of and .
The initial conditions (temperature, salinity) are prescribed from the
MedAtlas-II climatology weighted by a
low-pass filter with a time window of 10 years between 1955 and 1965
. For the buffer zone (west of the Strait of Gibraltar)
the initial state is based on the World Ocean Atlas 2005 .
This model correctly simulates the main structures of the thermohaline
circulation of the Mediterranean Sea, with mechanisms having a realistic
timescale compared to observations . In particular, tritium
and helium isotope simulations have
shown that the EMT signal from the Aegean sub-basin is realistically
simulated during early 1995. However, some aspects of the model still need to
be improved: for instance the too weak formation of Adriatic Deep Water
(AdDW), followed by a low contribution to the EMDW in the Ionian sub-basin.
In the western basin, the production of WMDW is reliable, but the spreading
of the recently ventilated deep water to the south of the basin is too weak.
Full details of the model and its parameterizations are reported by
, , and .
The tracer model
The 14C distribution in the ocean is often expressed as a delta notation
relative to the 14C / C ratio of the atmosphere (Δ14C =
(14R / Rref -1) × 1000; 14R is the
14C / C ratio of the ocean, and for the purpose of ocean ventilation
studies Rref is set to 1 .
Here we use the approach of in which
the ratio 14R is transported by the model rather than the individual
concentrations of C and 14C. Several model studies adopted the
simplified formulation of to describe the transport of
14C in the ocean .
This approach is based on two main assumptions: (i) the dissolved inorganic
carbon (DIC) field is constant and homogeneous and (ii) the air–sea
fractionation processes and biological activity could be ignored
. The first assumption reduces the
capacity of the model to estimate the 14C inventory and the ocean
bomb-14C uptake but does not much affect the
equilibrium 14C distribution in the ocean . Modelled and observed 14C may be directly
compared since the observed 14R ratios are corrected for the isotopic
fractionation once converted to the standard Δ14C notation
.
This simplified approach is commonly used in model evaluation to critically
examine the dynamics of the model (i.e. circulation and ventilation) against
in situ observation because (i) many oceanic 14C data were obtained
either by measuring 14C in dissolved inorganic carbon in seawater or in
corals and mollusc shells and (ii) it can be implemented in the ocean
circulation models at relatively low computational cost allowing many
sensitivity tests e.g..
Radiocarbon is implemented in the model as a passive conservative tracer,
which does not affect ocean circulation. Hence, its movement can be tracked
in an offline mode using the pre-computed transport daily fields (U, V, W)
of the NEMO-MED12 dynamical model . A time step of 20 min is
applied. The same approach was used to simulate the εNd
(neodymium) distribution in the Mediterranean Sea and the
mantle and crustal helium isotope signature as well as
to model the anthropogenic tritium invasion and CFC and
anthropogenic carbon storage .
Passive tracers are transported in the Mediterranean using a classical
advection–diffusion equation, including the sources and sinks. The equation
governing the transport of the dissolved inorganic carbon 14R in the
ocean is
δδt14R=-∇×(μ14R-K×∇14R)-λ14R,
where λ is the radiocarbon decay rate, u the 3-D velocity field,
and K the diffusivity tensor. Since radiocarbon is not produced in the
ocean, all 14C enters the surface water through gas exchange. The
radiocarbon flux through the sea–air boundary conditions is proportional to
the difference in the ratios between the ocean and the atmosphere
and given as
F=κR(14R-14Ra),
where F is the flux out of the ocean and
14Ra is the atmospheric 14C / C ratio. The
transfer velocity κR for the radiocarbon ratio in Eq. (2) is
computed as
κR=κCO2K0CT‾PaCO2,
with κCO2 being the carbon dioxide transfer velocity, K0
the solubility of CO2 in seawater taken from Weiss (1974),
Pa CO2 the atmospheric CO2 pressure, and
CT‾ the average sea-surface dissolved inorganic
carbon concentration, classically set to 2 mol m-3.
The CO2 transfer velocity is computed with the help of surface-level
wind speeds, w (m s-1), using the ARPERA forcing
following the
formulation:
κCO2=kw×w2660/Sc,
where Sc is the Schmidt number computed with the model S and T
fields.
The value of the empirical coefficient kw depends on the wind
field . Sensitivity tests
were performed to determine the value of the empirical coefficient
kw among the available values in the literature. Sensitivity
tests were performed to determine the value of the empirical coefficient
kw among the available values in the literature. Accordingly,
we have chosen a kw=0.25× (0.01/3600) s m-1 for
our radiocarbon simulation, which produces the best agreement between model
outputs and in situ data for the pre-bomb period.
Model initialization and forcing
The natural radiocarbon distribution was first simulated using the
atmospheric 14Ra =1; the ocean 14R is initially set to a
constant value of 0.85 Δ14C =-150 ‰,
appropriate for the deep ocean;. An atmospheric CO2 of
280 ppm is prescribed for this steady-state simulation. These simulations
were integrated for 700 years using a 10-year interval of NEMO-MED12
circulation fields between 1965 and 1974 continuously repeated until they
reached a quasi-steady state (i.e. the globally averaged drift was less than
0.001 ‰ per year). This forcing period was selected because it does
not include any intense interannual variability, such as the event of the
Eastern Mediterranean Transient EMT,.
Starting from the end of the pre-industrial equilibrium run, the model was
integrated from 1765 to 2011 covering the Suess effect , the
entire radiocarbon (14R) transient generated by the atmospheric nuclear
weapon tests performed in the 1950s and early 1960s, and the anthropogenic
CO2 increase. The 14R level in the atmosphere (Fig. )
is taken from and references cited therein and the atmospheric
CO2 from . Unfortunately, there is no time series data
of 14C concentration around the Strait of Gibraltar. Hence, simulated
14C levels in the model's AW are determined by damping to global model
estimates. The radiocarbon values in the buffer zone are prescribed from a
global simulation of radiocarbon by A. Mouchet, personal communication, 2016.
We have made two simulations with different boundary conditions at Gibraltar
(see Supplement); the time series calculated from the larger box between 35
and 55∘ N and from 0 to 46∘ W improves the radiocarbon
simulation a lot, and the results are more realistic compared to some in situ
data (Tisnérat-Laborde et al., 2013; Tisnerat-Laborde, personal
communication, 2016). So we have used this time series as a boundary
condition at Gibraltar to simulate 14C in the Mediterranean Sea (see
Supplement).
Atmospheric Δ14C in ‰ (orange) and atmospheric
CO2 in ppm (blue) from and references cited therein.
Model output for March 1956 showing the pre-bomb situation. Upper
panel: mean Δ14C (in ‰) in surface waters (0 to 200 m),
intermediate waters (200 to 600 m), and deep waters (600 to 3500 m). Lower
panel: Δ14C along E–W section in (d) WMed and
(e) EMed, where colour-filled dots represent in situ observations
. Both model and data are reported with the same colour
scale.
We also performed a sensitivity test on the impact of the EMT events on the
radiocarbon distribution in the Mediterranean Sea. Two separate simulations
were run for the period between 1990 and 2010 (i.e. covering the EMT event
that occurred at the beginning of the 1990s). The NoEMT run was performed
using the classical atmospheric forcing from ARPERA, as described in
Sect. 2.1.
To improve dense-water fluxes through the Cretan Arc during the EMT
(1992–1993), the ARPERA forcings were modified over the Aegean sub-basin
by increasing mean values as done by
for the Gulf of Lion. More specifically, from November
to March for the winters 1991–1992 and 1992–1993, daily surface heat loss
was increased by 40 W m-2, daily water loss by 1.5 mm and the daily
wind stress modulus by 0.02 N m-2. These changes accelerate the
transfer of surface temperature and salinity perturbations into intermediate
and deep layers of the Aegean sub-basin and improve the dense-water
formation in the Aegean sub-basin during the EMT, with more intense mixing
from winter convection.
ResultsSteady-state pre-bomb distribution
The 14C model results of the radiocarbon natural distribution for March
1956 are expressed in Δ14C (Fig. ) and in surface
radiocarbon reservoir age (Fig. ). They provide a descriptive
overview of the basin-wide distribution of radiocarbon before the
anthropogenic perturbation. Figure a, b, and c present the
horizontal 14C distribution of surface waters (between the surface and
200 m depth), intermediate (between 200 and 600 m), and deep waters (between
600 and 3500 m), respectively. Figure d and e show the radiocarbon
distribution over the whole water column in the Mediterranean along a
longitudinal transect for both the eastern and western basins together with
in situ observations from . Figure compares
model results of reservoir ages and several marine reservoir 14C age
data available for the surface water of the Mediterranean; these data were
obtained from pre-bomb calcareous marine shells between 1867 and 1948 and
coral Cladocora caespitosa.
Average radiocarbon age (years) in the upper 50 m as computed with
the model for 1940. Circles represent reservoir ages derived from
measurements of the composition of shells and
from corals .
Model–data comparison of Δ14C vertical profiles for
(a) the pre-bomb distribution as a composite of seawater
observations from different locations measured by and
(b) total radiocarbon distributions (natural + bomb) in the
eastern basin measured by at 18 ∘E. Model results
are in blue, while black indicates the in situ data.
As illustrated in Figs. a and , there is a significant
geographic heterogeneity in surface water for each sub-basin for “natural”
(or pre-bomb) 14C obtained both from model results and data. Table 1
shows that overall, the average Δ14C values are generally lower in
the WMed corresponding to older reservoir 14C age (402 ± 27)
compared to the EMed (349 ± 14), the Adriatic (373 ± 29), and the
Aegean (349 ± 32) sub-basins that show younger reservoir 14C ages
than the data of . These figures clearly show that the surface
inflow of Atlantic waters through the Strait of Gibraltar were progressively
enriched during their spreading into the EMed, leading to a relatively higher
Δ14C level in the EMed surface water closer to -44 ‰.
For both western and eastern surface water, the model simulates 14C
concentrations slightly higher than the in situ observations
. A careful
comparison between model outputs and seawater observations (1959) reveals a
more pronounced disagreement, especially in the EMed surface water where the
model overestimates the Δ14C values by more than 10 ‰
(Fig. a). However, the lack of more in situ pre-bomb values greatly
limits the comparison between model results and observations.
Regional means of radiocarbon reservoir age before 1950 AD. Column
2 gives the observations from , column 3 the model values in
1940 AD. The uncertainty in the mean is the larger of the standard
deviations based on counting statistics and the “standard deviation”, which is the
square root of the variance.
The model also simulates the rapid decrease in Δ14C values with
depth in the eastern basin, marking a significant vertical gradient and the
most negative values of deep-water Δ14C over the entire
Mediterranean Sea (-68 ± 7 ‰). At depth, the model simulates
low levels of Δ14C in the eastern basin deep water (average value:
-64 ± 7.4 ‰), significantly lower than those simulated in
the WMed deep waters (average value: -48 ± 6.9 ‰)
(Fig. c). To conclude, the model reproduced the E–W gradient and
the mean regional values of radiocarbon age reasonably well, except for the
Aegean sub-basin where the model underestimates the regional mean value
(Table 1), but the uncertainty in the data is also high.
Distribution of post-Bomb 14C
The simulated bomb 14C ocean distribution in the whole Mediterranean Sea
in March 1977 is illustrated in Fig. . The large atmospheric
Δ14C increase is reasonably well captured by the model in the
surface layer (values up to 120 ‰) over the whole basin. The lowest
values are encountered in the known region of convection and formation of
deep and intermediate waters (i.e. Gulf of Lion and Cyprus–Rhodes area;
Fig. a). Figure b shows a high concentration of
radiocarbon at intermediate depths mainly in areas with recent water-mass
ventilation. The radiocarbon distribution is more uniform in the deep water,
except at one location where relatively high radiocarbon levels are simulated in
the deep layer as a result of mixing with the radiocarbon-enriched surface
water, particularly in the Cretan Sea (values up to ±70 ‰)
(Fig. c).
Model output for March 1977 for the post-bomb situation. Upper
panel: mean Δ14C (in ‰) in surface waters (0 to 200 m),
intermediate waters (200 to 600 m), and deep waters (600 to 3500 m). Lower
panel: Δ14C along E–W section in (d) WMed and
(e) EMed, where colour-filled dots represent in situ observations
. Both model and data are reported with the same colour
scale.
Figure d and e show the modelled Δ14C results along
vertical sections in the western and eastern basins compared with in situ
data obtained from seawater samples in the Ionian Sea during the GEOSECS
expedition in 1977 (Station 404, 35.24∘ N, 17.12∘ E; Stuiver and ostlund, 1983). Similarly to the pre-bomb situation, the
Δ14C values decrease rapidly with depth, exhibiting a significant
vertical gradient between the maximum in the surface water of around
120 ‰ and the minimum in the deep-water values of around
-50 ‰ in the western basin and around -60 ‰ in the
eastern basin. The model correctly simulates the Δ14C vertical
distribution in the first 1500 m of the water column, in agreement with
observations (Fig. e). At depth, the model tends to underestimate
the 14C penetration in the deep Ionian sub-basin, where it fails to
reproduce the high Δ14C levels associated with EMDW formation
(Fig. b).
Figure displays the modelled Δ14C evolution between 1765
and 2008 for surface waters (average depth between 0 and 10 m in dashed line
and between 0 and 100 m depths in solid line) in the Liguro-Provençal
sub-basin, plotted against the in situ values as reconstructed by
Tisnerat-Laborde et al. (2013) from a 50-year old zooxanthellate coral
C. caespitosa collected alive in 1998 along the coast of Bonassola
(44∘10′ N, 9∘36′ E; NW Mediterranean; 28 m water
depth) and from mollusc shells Tisnerat-Laborde, personal
communication, 2016.
Between 1900 and 1952, the modelled Δ14C values show a slight
decrease of ∼ 12 ‰ resulting from the Suess effect
. The model slightly overestimates the observed pre-bomb
mean value -56 ± 3 ‰, in 1949–1955
as noted previously. Between 1952 and 1980, the Δ14C proxy values
increase rapidly from -56 ‰ to almost + 85 ‰ in the
Ligurian sub-basin due to a net uptake of atmospheric bomb 14C.
Δ14C values (in ‰) in the Ligurian sub-basin from
1765 to 2008 for the surface water (0–10 m depth; blue dashed line) and
sub-surface water (0–100 m depth; blue solid line) together with available
in situ observations from coral (black dashed line) and
molluscs (cyan stars).
The model represents well the uptake of bomb 14C for the top layer
(0–10 m) and the sub-surface layer (0–100 m) until 1965. Then, a slight
difference of Δ14C is simulated between these two layers, with a
higher value in the top layer that is consistent with the observations. These
differences are the result of vertical convective mixing
, i.e. the mixing layer depth could impact the amplitude
of Δ14C peak in the surface layer. Afterwards, the Δ14C
values decreased slowly with fluctuations but reaching a value around
+50 ‰ in 2008. This gradual decline of Δ14C (values up
to +60 ‰) is well simulated in the surface water when we compare
the modelled present-day (March 2011) distribution of radiocarbon in the
surface water (Fig. ). These results demonstrate that the model
simulates the bomb 14C uptake in surface and sub-surface water with a
realistic timescale comparable to in situ data and shows a good consistency
between the observed and simulated bomb Δ14C annual average values
(Fig. ).
Model output in March 2011. Upper panel: mean Δ14C (in
‰) surface water (0 to 200 m), intermediate water (200 to 600 m),
and deep water (600 to 3500 m). Lower panel: Δ14C along E–W
section in (d) WMed and (e) EMed, where colour-filled dots
represent in situ observations from Meteor M84 .
Both model and data are reported with the same colour scale.
Figure a shows the modelled present-day (March 2011) distribution
of radiocarbon in the surface water, against Meteor M84/3 cruise
data . The Δ14C distribution pattern for the
surface water is similar to the model outputs obtained for the years 1956 and
1977, with the eastern basin generally showing higher Δ14C values
compared to the western basin, except in the areas of the formation of deep and
intermediate waters in the Mediterranean Sea (the Cyprus–Rhodes area and in
the Gulf of Lion), where the Δ14C concentration decreases rapidly
due to higher vertical convection (Fig. ).
Comparison of average vertical profiles of Δ14C in the
WMed (left) and in the EMed (right). Model results are in blue; red indicates
the in situ data.
Figure d and e, present the simulated radiocarbon content for March
2011 at intermediate and deep depth along a W–E transect together with the
available Meteor M84/3 cruise data . The two
vertical sections show a Δ14C maximum in the first 500 m (Δ14C > 40 ‰). At deeper depths, Δ14C values
exhibit a significant vertical gradient up to 1500 m (Fig. ), with
low Δ14C values simulated for the deep waters (values lower than
-40 ‰), except for the central Levantine (i.e the area south of
the Crete sub-basin) and deep water, where high values Δ14C are
simulated (around -20 ‰) due to the intense deep convection in
this area. Relatively high values are simulated in the Algerian basin
(around -30 ‰, Fig. d ).
The model correctly reproduces the Δ14C content of the surface
waters as noted previously, with values similar to observations (values about
+50 ‰, Figs. and ). For the deeper depths,
the simulated Δ14C levels tend to be underestimated by more than
20 ‰ in the WMed and by about 50 ‰ in the EMed compared to
the observations. This is the result of the too weak deep-water overflow
through the Otranto Strait from the Adriatic into the Ionian sub-basin and
the weak southern penetration of the new WMDW in the simulation compared to
the values deduced from in situ observations
. This underestimation leads to
excessively low 14C average values at depth of the eastern basin.
However, the model simulates well the Δ14C values in the surface and
deep water of the Adriatic sub-basin (Fig. a, c) compared to
Meteor M84/3 cruise data .
Δ14C evolution from 1925 to 2008 in the Gulf of Lion
(green), the Algerian sub-basin (red), the Levantine sub-basin (magenta), the
Tyrrhenian sub-basin (black), the Cretan Sea (cyan), and the Ligurian
sub-basin (blue).
The spatial and temporal variability
The temporal variability of the radiocarbon distribution was explored as a
function of sub-basin location (Fig. ). Specifically, we compared
the annual average Δ14C time series in different “boxes”
following the LIW trajectory from the Levantine sub-basin to the Gulf of
Lion (including the Tyrrhenian, Ligurian, Algerian, and Cretan sub-basins)
for the surface (Fig. a), intermediate (Fig. b), deep
(Fig. c), and whole water column (Fig. d).
Δ14C difference between EMT and NoEMT experiments along
sections in the WMed (left column) and in the EMed (right column) for 1995
(top) and 1999 (bottom).
Mean Δ14C obtained for experiment with EMT (red) and NoEMT
(green) (a) in the Levantine sub-basin deep water (2000–3500 m
depth) and (b) in the Algerian sub-basin at intermediate level
(200–600 m). The right panels illustrate the difference between EMT and
NoEMT for the corresponding left panels.
The Δ14C evolution of the surface water is very similar within the
different sub-basins until 1965. Afterwards, the Tyrrhenian, Algerian, and the
Ligurian sub-basins have similar bomb 14C peak record, while the Gulf of
Lion, the Levantine basin, and the Cretan Sea respond differently to the bomb signal
compared to the other sub-basins. The Levantine and the Cretan Sea and the Gulf of
Lion show surface values as high as 100 ‰ and as low as 60
‰, respectively (Fig. a). The differences between the
western and eastern basins are more pronounced at intermediate depths
(Fig. b), especially between the Cretan Sea and the Gulf of Lion,
which shows an almost 40 ‰ difference in Δ14C. The
Algerian and Ligurian sub-basins are characterized by a very similar
Δ14C evolution through time, showing intermediate values between
the Cretan Sea and the Gulf of Lion. The results for the Tyrrhenian
sub-basin and Cretan Sea indicate a higher transfer in intermediate water
compared to other sub-basins. Model outputs for the deep layers
(600–3500 m) reveal much higher Δ14C levels in the Cretan Sea
compared to the other locations (Fig. c) because it has a shallower
bottom depth. The Δ14C difference across the six sub-basins is more
pronounced at deeper depths than at the surface (Fig. a) and
intermediate layers (Fig. b), especially after the 14C bomb
peak. This difference decreases gradually after 1995, particularly in the
surface water where the Δ14C values are almost the same among the
different sub-basins. The integrated values for the whole water column
(Fig. 9d) show the same pattern as seen in the deep waters (Fig. 9c),
suggesting a strong role of deep layers in controlling the distribution of
Δ14C in the water column.
The impact of the EMT event on the radiocarbon distribution in the
Mediterranean was analysed by comparing the outputs of two model simulations
(shown in Figs. and ): “EMT” and “NoEMT” for the
years of 1995 and 1999, respectively (see Sect. 2.3). A substantial
penetration of radiocarbon is observed in the deep water south of Crete in
1995 as a consequence of the EMT event that increased bottom Δ14C
values by more than 60 ‰, close to 14C bomb peak values. On the
other hand, the EMT reduces the Δ14C value in the intermediate
waters in the EMed (Fig. b). The EMT-related Δ14C signal
in the deep waters decreases gradually after the event, with values around
30 ‰ in 1999 (Fig. d). For the WMed (Fig. a,
c), the contrast is particularly pronounced at intermediate levels, with
regional values shifted by almost 10 ‰ between 200 and 800 m depth
in the Algerian basin (Fig. a), as a consequence of the abrupt
change in the eastern basin during the EMT event. As shown in
Fig. , the shift begins in 1992 in the Levantine sub-basin and
reached a 60 ‰ difference in 1995 between these two simulations
(Fig. a, b).
Discussion
The radiocarbon simulations provide independent and additional constraints on
the thermohaline circulation and deep-water ventilation in the Mediterranean
Sea. The relatively simple approach of radiocarbon modelling adopted here
from and A. Mouchet, personal communication, 2016
using a high-resolution regional model, led to a realistic simulation of the
radiocarbon distribution relative to available in situ data. It also enables
the evaluation of the NEMO-MED12 model performance in the Mediterranean Sea
from the seasonal to decadal and centennial timescales. Furthermore, it
provides a unique opportunity to better constrain the variability of the
uptake of bomb 14C in the whole Mediterranean Sea and to study the
impact of important hydrological events such as the Eastern Mediterranean
Transient (EMT).
The modelled radiocarbon distribution is very sensitive to the value of the
empirical coefficient (KW) (i.e. is the constant regulating
air–sea flux). In this study we have used KW=0.25× (0.01/3600) s m-1; this value led to a better
simulation of Δ14C in the Mediterranean compared to the other
estimates available in the literature, i.e.
0.426 × (0.01/3600) s m-1 used in global-scale simulations
A. Mouchet, personal communication, 2016. The
KW value depends on the wind field and the upper ocean mixing
rate field . For the present work we
used the wind fields from the ARPERA forcing
and the atmospheric CO2 values
from . These boundary conditions enabled the model to produce
satisfactory simulations of the bomb 14C chronology. In particular, the
timing of the Δ14C peak in the surface is consistent with the
estimated 14CO2 time transfer from the atmosphere to the
ocean in the surface waters (∼10 yr; ) as shown
in Fig. .
Unlike the global ocean, where input/output of radiocarbon comes only from
the exchange with the atmosphere, in the Mediterranean Sea there is also
lateral exchange of 14C through the Strait of Gibraltar. Unfortunately,
there are no time series data of 14C concentration in that area. Hence,
simulated 14C levels in the model's AW are determined by damping to
global model estimates from A. Mouchet, personal communication, 2016 at the
western boundary of the model domain using the 3-D profile calculated between
35 and 55∘ N and from 0 to 46 ∘ W (sensitivity tests were
performed to determine this box). This large box in the North Atlantic gave
the most representative signature of radiocarbon during the bomb peak (value
up to 140 ‰ in 1980) from the global simulation of
A. Mouchet, personal communication, 2016.
The comparison between the model outputs and the sea-surface Δ14C
record (Fig. ) obtained from a 50-year-old shallow-water coral in
the western Mediterranean Sea from reveals a good model
performance in simulating the bomb/post-bomb radiocarbon distribution
(Figs. b, ). However the representation of the pre-bomb
distribution is more difficult in the simulation (Fig. a). Several
issues complicate the simulation of the natural steady-state distribution of
14C using ocean-model circulation: (i) the uncertainty associated with
the radiocarbon surface boundary conditions applied in ocean model
experiments, (ii) the climatological field to represent the wind forcing,
often based on atmospheric model outputs and/or historical data, and
(iii) the significant changes due to the human activity which affects the
radiocarbon distribution in the atmosphere and the ocean (e.g. Suess
effect). In addition, the limited spatial and temporal resolution of seawater
and carbonate organism measurements during the pre-bomb period limits our
understanding of the natural radiocarbon distribution in the Mediterranean
Sea.
On the other hand, the 14C reservoir ages for this period are
exclusively localized over the continental shelf (mainly reconstructed from
shallow-water corals and molluscs). These proxy data reveal a high regional
variability as reconstructed by between 1837 and 1951 and
, which can be attributed to both (i) the interactions
between the ocean and land by the transport of depleted freshwater and
(ii) the potential changes in the vertical mixing of the water column, with
an increase in air–sea CO2 exchanges. These processes could be
favoured by the atmospheric conditions, such as the North Atlantic
Oscillation (NAO), East Atlantic Pattern (EA), East Atlantic/West Russian
pattern (EA/WR) within stronger and frequent wind storms and stronger
precipitation over northern Europe .
After the 14C bomb peak, a large gradient of Δ14C existed
between the surface waters already enriched and saturated in bomb 14C
(values up to 120 ‰) and intermediate/deep waters with a relatively
low Δ14C level (Fig. ), associated with the long
equilibration time of the radiocarbon-depleted deep waters and with vertical
mixing. Nevertheless the model simulation shows that the bomb-produced
radiocarbon signal has reached the deep layers of the Mediterranean Sea due
to the rapid transfer of surface waters to intermediate and deep depths,
especially in the Cretan Sea, where a high Δ14C is simulated in the
deep waters (Fig. ).
The new Δ14C data obtained from the analysis of the seawater
samples collected during the Meteor M84/3 cruise represent a unique
opportunity to critically assess the dynamics of the NEMO-MED12 ocean model
and to evaluate its ability to reproduce the main features of the present-day
radiocarbon distribution in the Mediterranean. The model produces realistic
simulated Δ14C values in the surface layer that are in agreement
with in situ measurements, thus supporting our modelling approach. However,
some important aspects of the model still need to be improved, particularly
for deep water, where it underestimates Δ14C (Figs. ,
). Previous passive tracer evaluations of NEMO-MED12 have shown
that the ventilation rates of deep waters are underestimated by the model for
the whole Mediterranean e.g..
This is particularly evident in the Ionian sub-basin where the eastern
Mediterranean deep water is not properly simulated due to the too weak
formation of Adriatic Deep Water that flows at shallower depths compared to
the observations. Similarly, the southward propagation of the newly formed
WMDW in the model is slower than the observations as a consequence of a
reduced salinity content (and hence density) in the formation area. Finally,
tritium and helium and CFC
simulations have shown that the model overestimates the mixing near the
Cretan Arc.
Several factors could control the radiocarbon distribution across the
Mediterranean Sea. During the pre-industrial period, the AW inflow at
Gibraltar, together with freshwater input from rivers, could have played an
important role in the radiocarbon distribution in the Mediterranean. The
large amount of radiocarbon injected into the atmosphere during the
thermonuclear weapon testing is now the dominant control on the 14C
distribution in the surface water, completely masking the natural radiocarbon
background. This creates the opportunity to study the constraints on the
14CO2 air–sea exchange. On the other hand, the ventilation
rate is the key mechanism and the most important factor controlling the
14C distribution in the deep layer.
The model has provided, for the first time, the evolution of Δ14C
in different parts of the basin and at different depths (Fig. ).
The difference in Δ14C in surface water between the western and
eastern basins reveals enrichment of Δ14C along surface water-mass
pathway due to prolonged exposure of the surface water to the atmosphere. It
also shows the different mechanism of 14C transfer at depth, where it
depends on convection processes with higher convection occurring especially
during the bomb peak with a large amount of radiocarbon in the atmosphere; the surface water masses undergo transfer with different intensity in the
different sectors of the Mediterranean basin.
The sequence of EMT events that occurred at the beginning of the 1990s in the
eastern Mediterranean has substantially changed the deep water-mass structure
in the whole basin. Different hypotheses concerning the preconditioning of
the EMT and its timing have been proposed in the literature
. The renewal of the deep-water masses after the EMT is
satisfyingly simulated by our regional model as illustrated by
tritium-helium3 and by neodymium simulations
. These findings allow us to study the impact of
interannual variability on a very long timescale, including the exceptional
events observed in the ventilation of the deep waters. The radiocarbon
simulation documents a severe impact of the EMT on the water-mass
distribution through the transfer of a large volume of 14C-enriched
near-surface water into the deep layers, with the highest contribution being
observed in the area south of the Cretan Arc.
The EMT event generates an important accumulation of 14C-enriched water
at the bottom of the Levantine sub-basin with a more than 60 ‰ difference in 1995 compared to the pre-EMT situation. In our simulation the
LIW layer is also affected by low values in the eastern Mediterranean, where
the renewal of the bottom water masses (low concentration of radiocarbon)
during the EMT could lead to a decrease in the 14C content in the LIW
layer (200–600 m, Fig. ). On the other hand, higher values of
radiocarbon are simulated at intermediate levels in the western Mediterranean
during the EMT, with shifts of up to 10 ‰ compared to the No-EMT
values. During the EMT, part of the Levantine basin is filled by water masses
originating in the Aegean Sea, with different characteristics compared to the
Adriatic. Hence, the EMT could modify water-mass characteristics and
potentially affect the formation of deep-water masses in this basin.
Conclusions
The radiocarbon distribution of the whole Mediterranean Sea was simulated for
the first time using a high-resolution model (NEMO-MED12) at 1/12∘
horizontal resolution and compared to available in situ measurements and
proxy-based reconstructions. The present study provides a unique opportunity
to improve the interpretation and understanding of the available in situ data and could help in the design of new observational programmes for the
Mediterranean Sea. It also provides a new approach to understanding and
better constraining air–sea gas exchange and the dynamics of Mediterranean water
masses over the last decade. The air–sea exchange parameterization led to a
realistic simulation of bomb 14C in the surface water, compared to
in situ data. The model correctly simulates the main features of the radiocarbon
distribution during and after the 14C bomb perturbation, especially in
the surface and intermediate layers. On the other hand, severe mismatches between
model and observations in the deep layer are clearly associated with
shortcomings in the model parameterization.
The natural distribution of 14C in the Mediterranean Sea is mainly
affected by the inflow of Atlantic water through the Strait of Gibraltar.
Further, the large amount of radiocarbon injected into the atmosphere during
the nuclear bomb-testing period has been the dominant factor defining the
14C distribution in the surface water, largely masking the natural
radiocarbon background. More paleo-data from the pre-industrial period would
help improve the knowledge of the natural distribution of 14C in the
Mediterranean and better constrain the fluxes and exchange of radiocarbon
between the different reservoirs.
This 14C modelling provides a unique opportunity to explore the impact
of the interannual variability on the radiocarbon distribution in the whole
Mediterranean Sea and the interaction between its western and eastern basins.
The outputs of the model simulation of the EMT event reveal a significant
increase in Δ14C (by more than 60 ‰) in the Aegean deep
water and at an intermediate level (value up to 10 ‰) in the western
basin. The model results with and without EMT show that the vertical transport of
surface signals in the Mediterranean is strong, suggesting a major
contribution of the EMT in the accumulation of radiocarbon in the eastern
Mediterranean deep waters. Although the approach we adopted does not attempt
to quantify the anthropogenic carbon, the model results and observations on
the 14C distribution support the contention that a large amount of
anthropogenic carbon is being stored in the deep Mediterranean waters, in
agreement with previous findings e.g..
The model used in this work is the free surface-ocean
general circulation model NEMO (Madec and NEMO-Team, 2008) in a regional
configuration called NEMO-MED12 (Beuvier et al., 2012a)
(http://www.nemo-ocean.eu/).
The data associated with the paper are available from the
corresponding author upon request. All the data used in this study were
published by their authors as cited in the paper. Here we present the model
result against the in situ data already published in the literature.
The Supplement related to this article is available online at doi:10.5194/bg-14-1197-2017-supplement.
The authors declare that they have no conflict of
interest.
Acknowledgements
We would like to thank the editor Christoph Heinze and the three anonymous
reviewers for their careful reading of the paper and helpful remarks.
Edited by: C. Heinze Reviewed
by: three anonymous referees
References
Antonov, J. I., Locarnini, R. A., Boyer, T. P., Mishonov, A. V., and Garcia,
H. E.: World Ocean Atlas 2005, Salinity, edited by: Levitus, S., NOAA
Atlas NESDIS 62, US Government Printing Office, Washington, DC, 2, 182
pp., 2006.
Attané, I. and Courbage, Y.: Demography in the Mediterranean region:
situation and projections, Plan Bleu, Paris, Economica, 2004.Ayache, M., Dutay, J.-C., Jean-Baptiste, P., Beranger, K., Arsouze, T.,
Beuvier, J., Palmieri, J., Le-vu, B., and Roether, W.: Modelling of the
anthropogenic tritium transient and its decay product helium-3 in the
Mediterranean Sea using a high-resolution regional model, Ocean Sci., 11,
323–342, 10.5194/os-11-323-2015, 2015a.Ayache, M., Dutay, J.-C., Jean-Baptiste, P., and Fourré, E.:
Simulation
of the mantle and crustal helium isotope signature in the Mediterranean Sea
using a high-resolution regional circulation model, Ocean Sci., 11,
965–978, 10.5194/os-11-965-2015, 2015b.Ayache, M., Dutay, J.-C., Arsouze, T., Révillon, S., Beuvier, J., and
Jeandel, C.: High-resohlution neodymium characterization along the
Mediterranean margins and modelling of ϵNd distribution in the
Mediterranean basins, Biogeosciences, 13, 5259–5276,
10.5194/bg-13-5259-2016, 2016.Beuvier, J., Sevault, F., Herrmann, M., Kontoyiannis, H., Ludwig, W., Rixen,
M., Stanev, E., Béranger, K., and Somot, S.: Modeling the
Mediterranean Sea interannual variability during 1961–2000: Focus on the
Eastern Mediterranean Transient, J. Geophys. Res., 115,
C08017, 10.1029/2009JC005950, 2010.Beuvier, J., Béranger, K., Lebeaupin Brossier, C., Somot, S.,
Sevault,
F., Drillet, Y., Bourdallé-Badie, R., Ferry, N., and Lyard, F.:
Spreading of the Western Mediterranean Deep Water after winter 2005: Time
scales and deep cyclone transport, J. Geophys. Res., 117,
C07022, 10.1029/2011JC007679, 2012a.
Beuvier, J., Lebeaupin Brossier, C., Béranger, K., Arsouze, T.,
Bourdallé-Badie, R., Deltel, C., Drillet, Y., Drobinski, P., Lyard, F.,
Ferry, N., Sevault, F., and Somot, S.: MED12, Oceanic component for the
modelling of the regional Mediterranean Earth System, Mercator Ocean
Quarterly Newsletter, 46, 60–66, 2012b.Broecker, S. and Gerard, R.: Natural radiocarbon in the Mediterranean Sea,
Lamont-Doherty Geological Observatory of Columbia University, Palisades, New
York, 10964, http://onlinelibrary.wiley.com/doi/10.4319/lo.1969.14.6.0883/pdf, 1969.Broecker, W. S. and Olson, E. A.: Lamont Radiocarbon Measurements VIII,
Radiocarbon, 3, 176–204, 10.2458/AZU_JS_RC.3.1080,
1961.
Broecker, W. and Peng, T.-H.: Tracers in the Sea: Palisades, NY,
Lamont-Doherty Geol. Observ., 1982.Broecker, W. S., Peteet, D. M., and Rind, D.: Does the ocean-atmosphere
system have more than one stable mode of operation?, Nature, 315, 21–26,
10.1038/315021a0, 1985.Butzin, M., Prange, M., and Lohmann, G.: Radiocarbon simulations for the
glacial ocean: the effects of wind stress, Southern Ocean sea ice and
Heinrich events, Earth Planet. Sci. Lett., 235, 45–61,
10.1016/j.epsl.2005.03.003, 2005.
Craig, H.: Abyssal carbon and radiocarbon in the Pacific, J.
Geophys. Res., 74, 5491–5506, 10.1029/JC074i023p05491,
1969.
Delibrias, G.: Carbone-14, in: Méthodes de datation par les phénoménes nucléaires naturels, edited by: Roth, E. and Poty, B., Applications, Masson, 423–58, 1985.Diffenbaugh, N. S. and Giorgi, F.: Climate change hotspots in the CMIP5
global
climate model ensemble,, Climatic Change, 114, 813–822,
10.1007/s10584-012-0570-x, 2012.Druffel, E. M. and Suess, H. E.: On the radiocarbon record in banded corals:
Exchange parameters and net transport of 14CO2 between atmosphere and surface
ocean, J. Geophys. Res., 88, 1271,
10.1029/JC088iC02p01271,
1983.Duffy, P. B., Eliason, D. E., Bourgeois, A. J., and Covey, C. C.: Simulation
of bomb radiocarbon in two global ocean general circulation models, J.
Geophys. Res., 100, 22545, 10.1029/95JC01937, 1995.England, M. H. and Rahmstorf, S.: Sensitivity of Ventilation Rates and
Radiocarbon Uptake to Subgrid-Scale Mixing in Ocean Models, J.
Phys. Oceanogr., 29, 2802–2828,
10.1175/1520-0485(1999)029<2802:SOVRAR>2.0.CO;2, 1999.
Ferry, N., Parent, L., Garric, G., Barnier, B., and Jourdain, N. C.:
Mercator
Global Eddy Permitting Ocean Reanalysis GLORYS1V1: Description and Results,
Mercator Ocean Quarterly Newsletter, 36, 15–28, 2010.Giorgi, F.: Climate change hot-spots, Geophys. Res. Lett., 33, L08707,
10.1029/2006GL025734, 2006.Giorgi, F. and Lionello, P.: Climate change projections for the
Mediterranean
region, Glob. Planet. Change, 63, 90–104,
10.1016/j.gloplacha.2007.09.005, 2008.Guilderson, T. P., Schrag, D. P., Kashgarian, M., and Southon, J.:
Radiocarbon
variability in the western equatorial Pacific inferred from a high-resolution
coral record from Nauru Island, J. Geophys. Res.-Oceans,
103, 24641–24650, 10.1029/98JC02271,
1998.Guyennon, A., Baklouti, M., Diaz, F., Palmieri, J., Beuvier, J.,
Lebaupin-Brossier, C., Arsouze, T., Béranger, K., Dutay, J.-C., and
Moutin, T.: New insights into the organic carbon export in the Mediterranean
Sea from 3-D modeling, Biogeosciences, 12, 7025–7046,
10.5194/bg-12-7025-2015, 2015.Herrmann, M., Sevault, F., Beuvier, J., and Somot, S.: What induced the
exceptional 2005 convection event in the northwestern Mediterranean basin?
Answers from a modeling study, J. Geophys. Res., 115,
C12051, 10.1029/2010JC006162, 2010.Herrmann, M. J. and Somot, S.: Relevance of ERA40 dynamical downscaling for
modeling deep convection in the Mediterranean Sea, Geophys. Res.
Lett., 35, L04607, 10.1029/2007GL032442, 2008.IPCC: Climate Change 2013: The Physical Science Basis, Contribution of
Working Group I to the Fifth Assessment Report of the Intergovernmental Panel
on Climate Change, edited by: Stocker, T. F., Qin, D., Plattner, G.-K., Tignor, M.,
Allen, S. K., Boschung, J., Nauels, A., and Xi, Y., Cambridge University Press, Cambridge,
United Kingdom and New York, NY, USA, p. 1535,
10.1017/CBO9781107415324., 2013.Josey, S. A.: Changes in the heat and freshwater forcing of the eastern
Mediterranean and their influence on deep water formation, J.
Geophys. Res., 108, 3237, 10.1029/2003JC001778, 2003.Josey, S. A., Somot, S., and Tsimplis, M.: Impacts of atmospheric modes of
variability on Mediterranean Sea surface heat exchange, J.
Geophys. Res., 116, C02032, 10.1029/2010JC006685, 2011.Key, R. M., Kozyr, A., Sabine, C. L., Lee, K., Wanninkhof, R., Bullister,
J. L., Feely, R. A., Millero, F. J., Mordy, C., and Peng, T.-H.: A global
ocean carbon climatology: Results from Global Data Analysis Project
(GLODAP), Global Biogeochem. Cy., 18, GB4031, 10.1029/2004GB002247, 2004.
Klein, B., Roether, W., Manca, B., Bregant, D., Beitzel, V., Kovacevic, V.,
and
Luchetta, A.: The large deep water transient in the Eastern Mediterranean,
Deep-Sea Res., 46, 371–414, 1999.Lascaratos, A., Roether, W., Nittis, K., and Klein, B.: Recent changes in
deep
water formation and spreading in the eastern Mediterranean Sea: a review,
Prog. Oceanogr., 44, 5–36, 10.1016/S0079-6611(99)00019-1,
1999.Lebeaupin Brossier, C., Béranger, K., Deltel, C., and Drobinski, P.:
The Mediterranean response to different space-time resolution atmospheric
forcings using perpetual mode sensitivity simulations, Ocean Model., 36,
1–25, 10.1016/j.ocemod.2010.10.008, 2011.
Levin, I. and Hesshaimer, V.: Radiocarbon – a unique tracer of global
carbon
cycle dynamics, Radiocarbon, 42, 69–80, 2000.Levin, I., Naegler, T., Kromer, B., Diehl, M., Francey, R. J., Gomez-Pelaez,
A. J., Steele, L. P., Wagenbach, D., Weller, R., and Worthy, D. E.:
Observations and modelling of the global distribution and long-term trend of
atmospheric 14CO2, Tellus B, 62, 26–46, 10.3402/tellusb.v62i1.16511,
2010.
Locarnini, R. A., Mishonov, A. V., Antonov, J. I., Boyer, T. P., and Garcia,
H. E.: World Ocean Atlas 2005, Temperature, edited by: Levitus, S., NOAA
Atlas NESDIS 61, US Government Printing Office, Washington, DC, 1, 182 pp.,
2006.Ludwig, W., Dumont, E., Meybeck, M., and Heussner, S.: River discharges of
water and nutrients to the Mediterranean and Black Sea: Major drivers for
ecosystem changes during past and future decades?, Prog. Oceanogr.,
80, 199–217, 10.1016/j.pocean.2009.02.001, 2009.
Madec, G. and NEMO-Team.: Note du Pole de modélisation, Institut
Pierre-Simon Laplace (IPSL), France, NEMO ocean engine, 27, ISSN N1288-1619, 2008.Mahadevan, A.: An analysis of bomb radiocarbon trends in the Pacific, Mar.
Chem., 73, 273–290, 10.1016/S0304-4203(00)00113-4, 2001.Maier-Reimer, E., Mikolajewicz, U., and Hasselmann, K.: Mean Circulation of
the Hamburg LSG OGCM and Its Sensitivity to the Thermohaline Surface
Forcing, J. Phys. Oceanogr., 23, 731–757,
10.1175/1520-0485(1993)023<0731:MCOTHL>2.0.CO;2, 1993.Malanotte-Rizzoli, P., Manca, B. B., D'Alcala, M. R., Theocharis, A.,
Brenner,
S., Budillon, G., and Ozsoy, E.: The Eastern Mediterranean in the 80s and in
the 90s: the big transition in the intermediate and deep circulations,
Dynam. Atmos. Oceans, 29, 365–395,
10.1016/S0377-0265(99)00011-1, 1999.Matsumoto, K., Nishihara, S., Kamimura, M., Shiraishi, T., Otoguro, T.,
Uehara,
M., Maeda, Y., Ogura, K., Lumsden, A., and Ogura, T.: The prepattern
transcription factor Irx2, a target of the FGF8/MAP kinase cascade, is
involved in cerebellum formation, Nat. Neurosci., 7, 605–612,
10.1038/nn1249, 2004.
MEDAR-MedAtlas-group: Medar-Medatlas Protocol (Version 3) Part I: Exchange
Format and Quality Checks for Observed Profiles, P. Rap. Int.,
50,
IFREMER/TMSI/IDM/SIS002-006, 2002.MerMex-Group: Marine ecosystems' responses to climatic and anthropogenic
forcings in the Mediterranean, Prog. Oceanogr., 91, 97–166,
10.1016/j.pocean.2011.02.003, 2011.
Millot, C. and Taupier-Letage, I.: Circulation in the Mediterranean Sea. The
Mediterranean Sea in Handbook of Environmental Chemistry, Springer Berlin,
ISBN: 978-3-540-25018-0, 5, 29–66, 2005.Mouchet, A.: The Ocean Bomb Radiocarbon Inventory Revisited, Radiocarbon,
55, 1580–1594,
10.2458/azu_js_rc.55.16402, 2013.Muller, S. A., Joos, F., Edwards, N. R., and Stocker, T. F.: Water Mass
Distribution and Ventilation Time Scales in a Cost-Efficient,
Three-Dimensional Ocean Model, J. Climate, 19, 5479–5499,
10.1175/JCLI3911.1, 2006.Naegler, T.: Reconciliation of excess 14C-constrained global CO2
piston
velocity estimates, Tellus B, 61, 372–384,
10.1111/j.1600-0889.2008.00408.x,
2009.Orr, J. C., Maier-Reimer, E., Mikolajewicz, U., Monfray, P., Sarmiento,
J. L.,
Toggweiler, J. R., Taylor, N. K., Palmer, J., Gruber, N., Sabine, C. L., Le
Quéré, C., Key, R. M., and Boutin, J.: Estimates of
anthropogenic carbon uptake from four three-dimensional global ocean models,
Global Biogeochem. Cy., 15, 43–60, 10.1029/2000GB001273,
2001.Orr, J. C., Najjar, R. G., Aumount, O., Bopp, L., Bullister, J. L.,
Danabasoglu, G., Doney, S. C., Dunne, J. P., Dutay, J.-C., Graven, H.,
Griffies, S. M., John, J. G., Joos, F., Levin, I., Lindsay, K., Matear, R.
J., McKinley, G. A., Mouchet, A., Oschlies, A., Romanou, A., Schlitzer, R.,
Tagliabue, A., Tanhua, T., and Yool, A.: Biogeochemical protocols and
diagnostics for the CMIP6 Ocean Model Intercomparison Project (OMIP), Geosci.
Model Dev. Discuss., 10.5194/gmd-2016-155, in review, 2016.Palmiéri, J., Orr, J. C., Dutay, J.-C., Béranger, K., Schneider,
A., Beuvier, J., and Somot, S.: Simulated anthropogenic CO2 storage and
acidification of the Mediterranean Sea, Biogeosciences, 12, 781–802,
10.5194/bg-12-781-2015, 2015.Pinardi, N. and Masetti, E.: Variability of the large scale general
circulation of the Mediterranean Sea from observations and modelling: a
review, Palaeogeogr. Palaeocl., 158, 153–173,
10.1016/S0031-0182(00)00048-1, 2000.Reimer, P. and McCormac, F.: Marine Radiocarbon Reservoir Corrections for
the
Mediterranean and Aegean Seas, Radiocarbon, 44, 159–166,
10.2458/AZU_JS_RC.44.4088, 2002.Rixen, M., Beckers, J. M., Levitus, S., Antonov, J., Boyer, T., Maillard, C.,
Fichaut, M., Balopoulos, E., Iona, S., Dooley, H., Garcia, M. J., Manca, B.,
Giorgetti, A., Manzella, G., Mikhailov, N., Pinardi, N., and Zavatarelli, M.:
The Western Mediterranean Deep Water: A proxy for climate change,
Geophys. Res. Lett., 32, 1–4, 10.1029/2005GL022702, 2005.
Rodgers, K. B., Cane, M. A., and Schrag, D. P.: Seasonal variability of sea
surface Delta C-14 in the equatorial Pacific in an ocean circulation mode,
J. Geophys. Res., 102, 18627–18639, 1997.
Roether, W. and Weiss, W.: Die groß-skalige thermohaline Zirkulation des
Mittelmeers, Annal. Meteorol., 15, 135–137, 1980.Roether, W., Manca, B. B., Klein, B., Bregant, D., Georgopoulos, D., Beitzel,
V., Kovacevic, V., and Luchetta, A.: Recent Changes in Eastern Mediterranean
Deep Waters, Science, 271, 333–335, 10.1126/science.271.5247.333,
1996.Roether, W., Klein, B., Manca, B. B., Theocharis, A., and Kioroglou, S.:
Transient Eastern Mediterranean deep waters in response to the massive
dense-water output of the Aegean Sea in the 1990s, Prog. Oceanogr.,
74, 540–571, 10.1016/j.pocean.2007.03.001, 2007.
Sarmiento, J. L. and Gruber, N.: Ocean biogeochemical dynamics. Princeton
University Press. Princeton, New Jersey, USA, 2006.Schroeder, K., Ribotti, A., Borghini, M., Sorgente, R., Perilli, A., and
Gasparini, G. P.: An extensive western Mediterranean deep water renewal
between 2004 and 2006, Geophys. Res. Lett., 35, 1–7,
10.1029/2008GL035146, 2008.Schroeder, K., Chiggiato, J., Bryden, H. L., Borghini, M., and Ben Ismail,
S.: Abrupt climate shift in the Western Mediterranean Sea, Sci.
Rep., 6, 23009, 10.1038/srep23009, 2016.Siani, G., Paterne, M., Arnold, M., Bard, E., Mativier, B., Tisnerat, N., and
Bassinot, F.: Radiocarbon Reservoir Ages In The Mediterranean Sea And Black
Sea, Radiocarbon, 42, 271–280,
10.2458/azu_js_rc.42.3844, 2000.Soto-Navarro, J., Somot, S., Sevault, F., Beuvier, J., Béranger, K.,
Criado-Aldeanueva, F., and García-Lafuente, J.: Evaluation of regional
ocean circulation models for the Mediterranean Sea at the Strait of Gibraltar: volume transport and thermohaline properties of the outflow,
Clim.
Dynam., 44, 1277–1292, 10.1007/s00382-014-2179-4, 2014.
Stanev, E. V. and Peneva, E. L.: Regional sea level response to global
climatic change: Black Sea examples, Europe, 32, 33–47, 2002.Stuiver, M. and Polach, H.: Discussion: Reporting of 14C data,
Radiocarbon,
22, 964–966, 1977.Stuiver, M., Quay, P. D., and Ostlund, H. G.: Abyssal water carbon-14
distribution and the age of the world oceans, Science,
219, 849–51, 10.1126/science.219.4586.849, 1983.SUESS, H. E.: Radiocarbon Concentration in Modern Wood, Science, 122,
415–417, 10.1126/science.122.3166.415-a, 1955.Sweeney, M. T., Thomson, M. J., Cho, Y. G., Park, Y. J., Williamson, S. H.,
Bustamante, C. D., and McCouch, S. R.: Global Dissemination of a Single
Mutation Conferring White Pericarp in Rice, PLoS Genet., 3, E133,
10.1371/journal.pgen.0030133, 2007.
Tanhua, T., Hainbucher, D., Cardin, V., Álvarez, M., Civitarese, G.,
McNichol, A. P., and Key, R. M.: Repeat hydrography in the Mediterranean
Sea, data from the Meteor cruise 84/3 in 2011, Earth Syst. Sci. Data, 5,
289–294, 10.5194/essd-5-289-2013, 2013.
Theocharis, A., Georgopoulos, D., Karagevrekis, P., Iona, A., Perivoliotis,
L.,
and Charalambidis, N.: Aegean influence in the deep layers of the Eastern
Ionian Sea, Rapport de la Commission international pour l'Exploration
Scientifique de la Mer Méditerranée, 33, 235, 1992.
Theocharis, A., Nittis, K., Kontoyiannis, H., Papageorgiou, E., and
Balopoulos,
E.: Climatic changes in the Aegean Sea influence the Eastern Mediterranean
thermohaline circulation (1986–1997), Geophys. Res. Lett., 26, 1617–1620, 1999.Tisnérat-Laborde, N., Montagna, P., McCulloch, M., Siani, G., Silenzi,
S., and Frank, N.: A High-Resolution Coral-Based Δ14C Record of
Surface Water Processes in the Western Mediterranean Sea, Radiocarbon, 55,
1617–1630, 10.2458/azu_js_rc.55.16397, 2013.Toggweiler, J. R., Dixon, K., and Bryan, K.: Simulations of radiocarbon in a
coarse-resolution world ocean model: 1. Steady state prebomb distributions,
J. Geophys. Res., 94, 8217, 10.1029/JC094iC06p08217,
1989a.Toggweiler, J. R., Dixon, K., and Bryan, K.: Simulations of radiocarbon in a
coarse-resolution world ocean model: 2. Distributions of bomb-produced carbon
14, J. Geophys. Res., 94, 8243,
10.1029/JC094iC06p08243,
1989b.
Vörösmarty, C. J., Fekete, B. M., and Tucker, B. A.: Global
River
Discharge Database (RivDIS V1.0), International Hydrological Program, Global
Hydrological Archive and Analysis Systems, UNESCO, Paris, 1996.Wanninkhof, R.: Relationship between wind speed and gas exchange over the
ocean, J. Geophys. Res., 97, 7373, 10.1029/92JC00188,
1992.