The BOUSSOLE mooring (2013–2015) time series
Temperature and fCO2 were measured from February 2013 to February 2016.
All seasons were well represented, with missing data only in May–July 2013.
For some periods, simultaneous measurements were made at the 3 and 10 m
depths
(Fig. 2a–c).
The range of temperature (Fig. 2a) extends from 13 ∘C in winter up
to 27 ∘C in summer, followed by progressive cooling in fall. The
coldest temperature, 13 ∘C, results from the winter vertical
mixing with the deeper Levantine Intermediate Water (LIW) marked by extremes
in temperature and salinity (Copin-Montégut and Bégovic, 2002). Temperature provides the main
control of the seasonality of fCO2, from 350 µatm to more than 550 µatm in summer 2013 (Fig. 2b).
The fugacity of CO2 in seawater is
a function of temperature, DIC, alkalinity, salinity and dissolved
nutrients. In the oligotrophic surface waters of the Mediterranean Sea, the
effect of nutrients may be neglected. Temperature and DIC have the strongest
influences. By normalising fCO2 to a constant temperature, the
temperature effect can be removed and changes in fCO2 resulting from
changes in DIC can be more easily identified. Figure 2c shows the
variability of fCO2 normalised to the constant temperature of
13 ∘C, (fCO2@13) using the equation of
Takahashi et al. (1993). The underlying processes that govern the seasonal
variability of fCO2@13 are successive winter mixing, biological
activity (organic matter formation and remineralisation) and the deepening of
mixed layer in fall (Bégovic and Copin-Montegut, 2002; Hood and Merlivat, 2001). Biological processes account for
the decline in fCO2@13 observed from March–April to late summer; the
ensuing increase in surface fCO2@13 is associated with the deepening of
the mixed layer in the fall or convection in winter, as the vertical
distribution of fCO2@13 at DYFAMED shows a maximum in the 50–150 m layer. Where a large remineralisation of organic matter occurs, the
productive layer is mostly between 0 and 40 m (Copin-Montégut and Bégovic, 2002). The
contribution of air–sea exchange is not significant (Bégovic and Copin-Montegut, 2002). Over the
period 2013–2015, the air–sea CO2 flux from the atmosphere to the ocean
surface is equal to -0.45 mol m-2 yr-1.
During summer 2014, large differences between measurements at 3 and 10 m were
observed (Fig. 2a–c between dashed lines). A detailed analysis of
the temporal variability during that period underscores the role of inertial
waves at the frequency of 17.4 h that create the observed differences
between the 2 depths of observations, with the deeper waters being colder and
enriched in fCO2@13. Temperature and fCO2@13 variability is
dominated by inertial waves. In particular, from 15 to 26 August 2014,
the difference in T between the two depths is as large as 7.6 (5.1 ∘C on average). fCO2 decreases on average by
32.7 µatm,
corresponding to an increase in fCO2@13 equal to 42.8 µatm.
The 2013–2015 seasonal and inter-annual variability of T, fCO2 and
fCO2@13 is illustrated in Fig. 2–f. The larger inter-annual
changes in temperature (Fig. 2d) are observed during summer, both at 3 and 10 m depth, while over February and March, a constant value of
13 ∘C is observed as the result of vertical mixing with the LIW. A
very large inter-annual variability of fCO2@13 is observed for
T < 14 ∘C (Fig. 2f). This is associated with the winter
mixing at the mooring site, which is highly variable from year to year.
Winter mixed-layer depth (MLD) varies between 50 and 160 m, at the top of
the LIW over the 2013–2015 period (Coppola et al., 2016). The variable depth of the
winter vertical mixing causes the difference in fCO2@13, as fCO2
increases with depth (Copin-Montégut and Bégovic, 2002). The deepening of the MLD is driven by
episodic and intense mixing processes characterised by a succession of
events lasting several days, which are related to atmospheric forcing
(Antoine et al., 2008) and lead to increase in fCO2@13. Figure 2e
illustrates the solubility control of the variability of fCO2, as
fCO2
increases when T increases. Another cause of the inter-annual variability of
fCO2 for T∼14 ∘C is the timing of the spring
increase in biological activity, which differs by a month between years. For
instance, the increase happened at the beginning of April in 2013, when T∼15–16 ∘C and by mid March in 2014, when T∼14 ∘C. Another cause is the deepening of the mixed layer due to the fall
cooling,
which varies by a month between years.
Decadal changes in hydrography
Sea surface temperature changes
Monthly mean values of temperature have been computed for the two 3-year
periods, 1995–1997 and 2013–2015. In 1995–1997, fCO2 and
T at 2 m were measured with CARIOCA sensors installed on a buoy at DYFAMED
(Hood and Merlivat, 2001). The mean annual temperature of hourly CARIOCA data
is equal to 18.21 ∘C. For 2013–2015, temperature measurements made
on the BOUSSOLE mooring at 3 and 10 m have been used. For the April to
September time interval, there are only data at the 3 m depth. In addition,
temperature data measured half hourly at 0.7 m at a nearby meteorological
buoy (43∘23′ N, 7∘50′ E)
(https://doi.org/10.6096/hymex.azurbuoy.thermosalinograph.20100308, last access: 19 September 2018; Rolland and Bouin, 2010) have been used
(Fig. 3d). Mean annual temperatures are equal to 18.29 and
17.97 ∘C, respectively, based on the meteorological buoy and the BOUSSOLE
mooring data. The two sets of data differ essentially during July and August,
with the temperatures at 3 m colder than those at 0.7 m, which indicates a
thermal gradient between the two depths during summer. Therefore, for
2013–2015, we select the mean annual value computed with the meteorological
buoy, 18.29 ∘C, as a better representation of the sea surface. This value
is close to 18.21 ∘C computed for 1995–1997. Thus, no significant
change in SST is found between the 2 decades, with a mean value equal to
18.25 ∘C.
Sea surface salinity changes
The mean value of salinity and the standard error of the mean computed from
56 samples taken at BOUSSOLE in 2013–2015 are respectively 38.19 and 0.02.
In 1998–1999, ship measurements of surface salinity were made during monthly
cruises at the DYFAMED site (Copin-Montégut et al., 2004). The mean salinity and the
standard error of the mean of this set of 19 data are respectively 38.21 and
0.03. Thus, there is no significant salinity change between the two decades.
(a) fCO2@13 as a function of temperature for hourly data in
2013, 2014 and 2015.The yellow dots indicate mean fCO2@13. (b) is similar
to (a) but represents all hourly data in 1995–1997 (black) and in 2013–2015 (red).
(c) is similar to (b) but represents average values per 1 ∘C interval (standard
deviation as dotted line). The difference between the two periods is also
displayed by the dashed blue curve (scale on the right axis; the mean difference
over all SST is represented by the horizontal blue line). (d) Mean monthly
sea surface temperature for 1993–1995 (black curve for CARIOCA sensors),
2013–2015 (green for CARIOCA sensors) and 2013–2015 (red for meteorological buoy).
Corresponding mean annual values are indicated by dotted lines.
Decadal changes in fCO2@13
Time series of fCO2@13 in 1995–1997 and 2013–2015
The two time series of high-frequency data were analysed in order to
quantify the change in fCO2@13 2 decades apart at the sea surface. To
account for the inter-annual seasonal variability as well as irregular
sampling, we performed an analysis of the change in fCO2@13 as a
function of SST (Fig. 3a and b). For the 2013–2015 data set, we excluded
summer data measured at 10 m depth as they were not representative of the
surface mixed layer due to a strong stratification. Much larger fCO2@13
values are observed at low temperatures than at high temperatures, the
decrease being similar and strongly nonlinear for the two studied periods.
As described in Sect. 3.1, large values at low temperatures result from
mixing with enriched deep waters during winter, and low values for
26–28 ∘C temperatures occur at the end of summer
after the biological drawdown of carbon. An increase in fCO2@13 between the
two periods is evident across the range of temperatures.
Trend analysis and statistics
To quantify the change in fCO2@13 between the two data sets, we proceed
as follows: data are binned by 1 ∘C temperature intervals, thereby
removing any potential seasonal weighting, especially towards the
13–14 ∘C winter months temperature. The measurements made in this
temperature interval represent about 25 % of the total number of data for
both periods. For each of the fourteen 1 ∘C steps, the mean and
standard deviation of hourly fCO2@13 measurements are reported
in Table 1 and in Fig. 3c. The mean temperature within each of the 1∘ steps differ
for the two periods, as the distribution of individual measurements are not
identical.
For both data sets, a monotonic relationship between fCO2@13 and T is
observed with correlation coefficients respectively equal to -0.861 and
-0.857. The difference in fCO2@13 between the two periods, dfCO2@13, is
derived in each temperature step, seen in the difference between column 2 and 6
of Table 1. The variability of this difference is estimated as the quadratic
mean of the standard deviation in each time series. Both values are reported
in Table 1, column 9 and 10 and in Fig. 3c. The distribution of each
dfCO2@13 values around the mean over all SST of dfCO2@13 seems
random and indicates no trend of dependency with SST (Fig. 3c). This suggests
that the processes that control the seasonal variation of fCO2@13 at
the sea surface have not changed over the last two decades.
Distribution of temperature, fCO2@13, and increase dfCO2@13 data binned by
1 ∘C temperature interval for the two periods 1995–1997 and
2013–2015.
Time interval 1995–1997
Time interval 2013–2015
Temporal change
T1
fCO2@13
N
Standard
T1
fCO2@13
N
Standard
dfCO2@13
Standard
∘C
µatm
deviation
∘C
µatm
deviation
µatm
deviation
µatm
µatm
µatm
13.45
331.58
1212
28.09
13.55
363.14
6869
18.07
31.56
33.40
14.45
305.28
495
26.02
14.43
337.16
3270
16.65
31.87
30.89
15.37
281.54
447
9.62
15.57
321.10
3112
11.09
39.56
14.68
16.44
274.43
182
8.53
16.42
313.79
1818
11.09
39.36
13.99
17.58
275.54
190
7.04
17.56
306.83
1528
14.65
31.29
16.25
18.47
277.34
300
9.04
18.45
296.57
2621
10.95
19.23
14.20
19.62
265.43
342
15.58
19.41
291.84
1406
13.45
26.40
20.59
20.50
258.08
529
14.15
20.50
293.16
1135
18.21
35.08
23.06
21.56
271.15
239
12.98
21.54
297.96
1200
20.41
26.82
24.19
22.49
250.75
742
13.66
22.49
290.27
2385
18.57
39.52
23.05
23.57
252.22
320
13.00
23.47
296.92
747
21.77
44.70
25.36
24.41
245.85
506
7.08
24.40
280.44
959
14.82
34.59
16.43
25.50
250.06
215
10.77
25.53
284.05
456
14.81
33.99
18.31
26.42
256.29
279
6.24
26.29
286.71
249
11.23
30.42
12.85
We have estimated the uncertainties in the estimates of the difference
dfCO2@13 with two methods. Firstly, the arithmetic mean of dfCO2@13
is equal to 33.17 µatm, with a standard deviation (SD) and standard
error (SE), respectively equal to 6.29 µatm and 1.68 µatm. A
95 % confidence interval is thereby achieved within 1.96 SE, i.e. 3.29 µatm.
A second approach consists of computing a weighted average of
the mean of dfCO2@13. In this case, mean weighted value of
dfCO2@13 over the whole range of temperatures is estimated, the weights
being equal to the variance of dfCO2@13 in each temperature step.
The value of dfCO2@13 is equal to 32.70 µatm. The weighted SD, and the associated SE, of the
14 data points are respectively equal to 4.85 and 1.30 µatm. A 95 % confidence interval is achieved within 2.54 µatm. The
difference between the two mean dfCO2@13 estimates is 0.47 µatm,
well below SE. In the following, we have chosen the former method.
Changes of seawater carbonate chemistry in surface waters
We estimated the DIC and pH changes related to the increase in fCO2@13
measured at the sea surface 18 years apart, assuming a mean salinity equal
to 38.2, a mean alkalinity equal to 2562.3 µmol kg-1 following
Eq. (1) and a mean in situ temperature (T) equal to 18.25 ∘C. The dissociation constants of Mehrbach refitted by Dickson and Millero
(Dickson and Millero, 1987; Mehrbach et al., 1973) were used. pH is calculated on the seawater scale.
The uncertainty of dfCO2@13, 3.3 µatm, has been propagated to
compute the combined uncertainty in dDIC and dpHSWS. The uncertainties
in the equilibrium constants are neglected in this propagation of
uncertainties. Likewise, an implicit assumption is that there is no
systematic error on DIC and pHSWS derived from fCO2@13 between the two
time periods; in particular, mean temperature and salinity remain the same
(Sect. 3.2). This is further discussed in Sect. 4.1. We compute an
increase in DIC and dDIC equal to 25.2±2.7 µmol kg-1
(1.40±0.15 µmol kg-1 yr-1) and a decrease in
pHSWS and dpHSWS equal to -0.0397±0.0042 pHSWS
(-0.0022±0.0002 pHSWS yr-1) (Table 2).
Seasonally detrended long-term and annual trends
of seawater carbonate chemistry and atmospheric composition.
dfCO2a
dfCO2a
dDICa
dpHSWSc
dfCO2
dDIC annual
dpHSWSc
@13 µatm
@T µatm
µmol kg-1
pH unit
@T annual
µmol kg1 yr1
annual pH
µatm yr-1
unit yr1
Sea surface
33.2±3.3
41.4±4.1
25.2±2.7
-0.0397±0.0042
2.30±0.23
1.40±0.15
-0.0022±0.0002
Atmosphere
34.3±2.3
20.8±1.3b
1.91±0.13
1.15±0.07
Lampedusa data
dfCO2@Tair/
0.83±0.10
0.83±0.09
dfCO2@Tsea
T, mean annual temperature equal to 18.25 ∘C;
a change from 1995–1997 to 2013–2015;
b dDICant;
c dpHSWS computed at T.
Changes in atmospheric and seawater fCO2
The increase in atmospheric fCO2 from 1995–1997 to 2013–2015 was
computed from the monthly atmospheric xCO2 concentrations measured at the
Lampedusa Island station (Italy) (35∘31′ N, 12∘37′ E)
(http://ds.data.jma.go.jp/gmd/wdcgg/, last access: 14 September 2018) (see Eq. 3 in
Hood and Merlivat, 2001). Considering a mean annual in situ temperature equal to
18.25 ∘C and an atmospheric pressure of 1 atm, we derived a mean
atmospheric fCO2 equal to 355.3±0.8 µatm for 1995–1997 and
389.6±0.9 µatm for 2013–2015, which is an increase of 34.3±2.3 µatm (95 % confidence interval) (Table 2). At this temperature, the
change in fCO2 at the sea surface is 41.4±4.1 µatm. Thus the
contribution of the increase in atmospheric CO2 is responsible for
84%±5 % of the increase in fCO2 measured in the surface waters.
With the same salinity and alkalinity as previously, the measurement of the corresponding
change in surface DIC, assuming air–sea equilibrium, would be 20.8±1.3 µmol kg-1 (Table 2).