Ocean acidification caused by anthropogenic uptake of CO2 is perceived to be a major threat to calcifying organisms. Cold-water corals were thought to be strongly affected by a decrease in ocean pH due to their abundance in deep and cold waters which, in contrast to tropical coral reef waters, will soon become corrosive to calcium carbonate. Calcification rates of two Mediterranean cold-water coral species, Lophelia pertusa and Madrepora oculata, were measured under variable partial pressure of CO2 (pCO2) that ranged between 380 µatm for present-day conditions and 930 µatm for the end of the century. The present study addressed both short- and long-term responses by repeatedly determining calcification rates on the same specimens over a period of 9 months. Besides studying the direct, short-term response to elevated pCO2 levels, the study aimed to elucidate the potential for acclimation of calcification of cold-water corals to ocean acidification. Net calcification of both species was unaffected by the levels of pCO2 investigated and revealed no short-term shock and, therefore, no long-term acclimation in calcification to changes in the carbonate chemistry. There was an effect of time during repeated experiments with increasing net calcification rates for both species, however, as this pattern was found in all treatments, there is no indication that acclimation of calcification to ocean acidification occurred. The use of controls (initial and ambient net calcification rates) indicated that this increase was not caused by acclimation in calcification response to higher pCO2. An extrapolation of these data suggests that calcification of these two cold-water corals will not be affected by the pCO2 level projected at the end of the century.
Citation: Maier C, Schubert A, Berzunza Sànchez MM, Weinbauer MG, Watremez P, Gattuso J-P (2013) End of the Century pCO2 Levels Do Not Impact Calcification in Mediterranean Cold-Water Corals. PLoS ONE 8(4): e62655. https://doi.org/10.1371/journal.pone.0062655
Editor: John Murray Roberts, Heriot-Watt University, United Kingdom
Received: October 3, 2012; Accepted: March 24, 2013; Published: April 30, 2013
Copyright: © 2013 Maier et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: Financial support was provided by the European Commission through a Marie-Curie Fellowship to CM (MECCA, project no 220299) and the project COMP via the Prince Albert II of Monaco Foundation, the German Academic Exchange Service (DAAD). This work is a contribution to the ‘European Project on Ocean Acidification’ (EPOCA) which received funding from the European Community’s Seventh Framework Programme FP7/2007–2013) under grant agreement no. 211384. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Ocean acidification is one of the major threats to the marine environment and has become a central research focus in marine science during the last decade. The ocean and atmosphere exchange carbon dioxide (CO2) , and the net uptake of CO2 by the ocean causes the pH to decline. Due to anthropogenic activity, the oocean pH has already declined by 0.1 units since pre-industrial times and will further decline by about 0.4 units until the end of the century . The concentration of CO2 in seawater is steadily increasing at an unprecedented rate of change and it is anticipated that atmospheric CO2 concentrations increase four-fold between 1750 and 2100, reaching values above 1000 ppm . Many calcifying organisms will be affected by ocean acidification with a decrease of the growth of their shells or skeletons –. Also, reproduction and larval growth are thought to be impeded –. For tropical coral reefs, the predictions as to the detrimental effects of ocean acidification are almost unison and foresee a decline in reef growth – and a shift in species composition with a decrease in diversity , . On the other hand, it has been shown, that the response to ocean acidification can be highly variable for different taxonomic groups . Investigations on cold-water corals or deep-sea corals are scarce. So far only three experimental studies on the effects of ocean acidification have been published. The rate of net calcification of Lophelia pertusa from Norwegian waters was shown to decline under future pCO2 conditions , , while no or even a positive long-term (6 months) response of net calcification was observed . A third study investigated the short-term response of net calcification of the Mediterranean coral Madrepora oculata under pCO2 values reflecting past, pre-industrial and future conditions. No effect on net calcification was found between ambient and elevated pCO2, while calcification rates were twice as high under pre-industrial pCO2 than under ambient or elevated pCO2. This suggests that present-day calcification rates of M. oculata have already drastically declined since pre-industrial times . Cold-water corals were initially thought to become affected by ocean acidification before their tropical relatives because they inhabit deeper and colder waters where the aragonite saturation state (Ωa) is lower than in shallower, warmer regions . More than 70% of cold-water coral communities are found in regions that will be undersaturated with respect to aragonite by the end of the century . However, experimental as well as observational evidence has shown that some cold-water corals are able to cope and maintain positive skeletal growth even in waters undersaturated in Ωa , , . Furthermore, the study of Form & Riebesell  found a decline of net calcification at elevated pCO2 in a short-term experiment and a stimulating effect in a long-term experiment from which they concluded that the cold-water coral L. pertusa is able to acclimate to higher pCO2. In the present study, the question of acclimation is addressed using a thorough experimental design in which net calcification of individual coral samples is repeatedly measured as a function of exposure time to respective pCO2 levels. The two branching cold-water coral species present in the Mediterranean Sea, L. pertusa and M. oculata are investigated to discern whether one of the two species is more resistant than the other or has a higher potential for acclimation. In the Mediterranean Sea, M. oculata is more widespread than L. pertusa . This could indicate that the prevailing conditions in the Mediterranean Sea are more favourable for M. oculata than for L. pertusa and this might result in a divergent species response to changing ocean pH. It is well known, that the distribution of cold-water corals is controlled by temperature within a range of 4 to 12°C . In the Mediterranean Sea, the temperature at depths where cold-water corals occur ranges between 12.5 to almost 14°C ,  which is, therefore, above the common temperature range. Additionally, total alkalinity is higher in the Mediterranean Sea than open oceans and consequently absorbs more atmospheric CO2. It has been shown, that anthropogenic CO2 has already affected the Mediterranean Sea and that the pH has decreased by 0.05 to 0.14 pH units, depending on the depth considered, since pre-industrial times .
Materials and Methods
The present study aimed at elucidating (1) the effect of an increase in CO2 (ocean acidification) on rates of net calcification on the two Mediterranean cold-water coral species, M. oculata and L. pertusa, (2) to find out whether there is a species effect and (3) to evaluate short- (shock) and longer-term (acclimation) responses in calcification. This has been addressed by measuring net calcification before, immediately after and 1, 2, 3 and 9 months after pCO2 has been adjusted.
Sampling and Experimental Set-up
During the MedSeaCan cruise in June 2009, the cold-water corals L. pertusa and M. oculata were sampled in the canyon of Lacaze Duthiers using a remotely operated vehicle (ROV) and the vessel MINIBEX (COMEX, France). Corals were sampled at water depths of 500 (42°32.98'N, 03°25.21'E), 267 (42°34.98'N, 03°24.15'E) and 260 m (42°35.07'N, 03°24.14'E). On board, corals were maintained in a plastic container (1040×640×515 mm) at a controlled temperature of 12.5±0.5°C using a chilling unit attached to a water pump (1000 l h−1). Corals were transported back to the laboratory and maintained in a climate room at 13°C until the experiments started in August 2009. Coral branches were sub-divided into smaller fragments (Table 1) and placed into vials of 4.5 or 8 cm inner diameter and a volume of ca 300 and 1000 ml. The vials were placed in 4 aquaria that served as water baths (13°C) as well as overflow basins for seawater from the vials. Each vial received running seawater (Mediterranean surface water with a salinity of 38) and air using silicone tubings of 0.5/2.5 and 1.0/3.0 mm inner/outer diameter, respectively. The seawater was filtered through 2 layers of micron bags (5 and 1 µm) into two 100 l storage tanks in a climate room which was set to 11°C. Temperature in the water baths containing the vials was maintained at ±0.1°C using electronic temperature controllers (Corema) and heaters (Tetratec® HT75). Temperature homogeneity was obtained by circulation pumps (JBL Pro Flow 500, 500 l h−1). Seawater was distributed by gravity from the storage tanks to the vials at a flow rate of 32±14 ml h−1. Air was supplied through a small tube (ca. 8 cm height×0.7/1.0 cm inner/outer diameter) inserted vertically in the vials, preventing the bubbles to be in direct contact with the coral fragments and providing an airlift and mixing. The air was pre-mixed using mass flow controllers (MFCs, ANALYT MC-GFC17, 0–10 l for air and 0–10 ml for pure CO2) and an air compressor (Jun-Air OF302-25B) at a flow rate of 4×1 l min−1 distributed to 4×21 vials. Corals were fed 3 times a week with freshly hatched Artemia larvae and 1 time a week with frozen krill. The water bath containing the vials and overflowing seawater was also adjusted to the target pCO2 by bubbling with an air stone (HOBBY ceramic air diffuser, 150 mm). Prior to each feeding the seawater of the water bath with respective pCO2 was filtered (Tetratec EX 1200, 1200 l h−1) and the strong water flow generated by the filtration unit was used to clean vials and remove old prey and detritus.
Determination of Rates of Net Calcification
Before changing the pCO2 levels in the experimental set-up, net calcification for each coral fragments was determined using the alkalinity anomaly technique  in order to provide an initial control (T0) at ambient pCO2. Subsequently, the CO2 concentration of the air used to bubble the vials was adjusted to 4 pCO2 treatment levels (A–D) with treatment A at 280 (low), B at 400 (ambient), C at 700 (elevated) and D at 1000 ppm (elevated) using the MFCs. To adjust to lower pCO2 than ambient (treatment A, 280 ppm), soda lime was used to generate low-CO2-air (5–10 ppm) which was mixed with pure CO2. For treatment B ambient air was used and for treatment C (700 ppm) and D (1000 ppm) ambient air was mixed with pure CO2. The exact mixing of air or CO2-free air with pure CO2 was adjusted by a LI-COR CO2-analyser (LI-6252). The pH in the different treatments during coral maintenance was monitored on a weekly base using a commercial pH module and electrode (IKS aquastar) which was calibrated to the NBS standard (buffers 4 and 7). This was done to control that treatment levels remained relatively constant, however, these measurements were not used to assess the carbonate chemistry, which was determined by analysis of AT and CT (see below for details) 1) during 2-day incubation to determine calcification rates via the alkalinity anomaly technique and 2) in maintenance vials containing corals and in blanks (vials without corals) after 9 months when net calcification rates were established using the buoyant weight technique.
To discriminate between short- and longer-term effects, net calcification was determined immediately after changing the pCO2 by transferring coral fragments directly from ambient into incubation vials adjusted to target air-CO2 mix (T1) and at about monthly intervals during the first 3 months (T2, T3 and T4, with 29, 57 and 89 days of exposure, respectively) using the alkalinity anomaly technique. Additionally, net calcification was again determined after approximately 9 months (267±14 days) with the buoyant weight technique  using a balance (Mettler Toledo) with a precision of 1 mg.
For determination of net calcification rates using the alkalinity anomaly technique, corals and blanks (seawater without corals) were placed for 2 days in the same type of vials than those in which they were maintained but with a constant volume of 200 or 700 ml. To maintain pCO2 levels constant vials were aerated with the same air-CO2 mix that was also used during maintenance periods for treatments A–D. At the end of the incubation, seawater was sub-sampled to determine inorganic nutrients, dissolved inorganic carbon (CT) and total alkalinity (AT) as described in Maier et al. . Other parameters of the carbonate chemistry (pCO2, pH on the total scale, pHT, and Ωa) were determined from CT, AT, temperature (13°C), salinity (38) and hydrostatic pressure (0) using the software package seacarb  (Table 2 and S1). Rates of net calcification were calculated from differences in AT from blanks (seawater without corals incubated in parallel to coral samples) and coral incubations and were corrected for changes in the concentration of inorganic nutrients. For the incubation at T3, no samples for inorganic nutrient concentrations were taken. Therefore, the regression functions between calcification rates corrected (or uncorrected) for inorganic nutrient release (Figure S1) were used to make the nutrient correction for T3. Data were normalized to the initial skeletal dry weight of coral fragments and reported in % d−1 ,  using the exponential growth function G [% d−1] = ((Wn/W0)1/n −1) * 100; with G = net calcification rate, Wn = Weight after n days, W0 = initial weight and d time interval (days) for growth increment (alkalinity anomaly method = 2 days, buoyant weight method = 267 days).
Statistical analyses were conducted using the software package Statistica 7.0. A repeated measures analysis of variance (ANOVA) was used. Only coral fragments for which net calcification rates were established at all time steps (T0–T4 and buoyant weight after 9 months) were considered for analysis as this is a pre-requisite for repeated measures testing. Therefore N is the same for all repeated measurements (Table S1). For independent samples (non-repeated measures analyses) a one-way ANOVA comparison of carbonate chemistry between pCO2 treatments or initial size and initial net calcification rates (T0) was used. If applicable, a post-hoc test was performed using the Honest Significance Difference (HSD) test for either equal or unequal N. The respective statistical tests used are also given in the text, tables and figure legends. Values are given as mean ± S.D. unless stated otherwise.
Results and Discussion
Initial Carbonate Chemistry, Concentration of Inorganic Nutrients and Calcification
For the initial incubation at ambient pCO2 (T0), the average AT and CT of blank controls (incubation without corals) was 2616±27 and 2374±21 µmol kg−1 (± standard deviation, throughout this paper; N = 30). pHT, pCO2 and Ωa were 8.05±0.03, 447.3±31.7 µatm and 2.68±0.16, respectively. The concentrations of phosphate and ammonium were 0.03±0.07 and 1.03±1.48 µmol kg−1 (N = 16). AT and CT decreased during the 2-day incubations of both coral species; AT was on average 2546±47 and 2571±32 µmol kg−1 and CT was 2333±39 and 2356±41 µmol kg−1 (N = 23 and 18), respectively. As a consequence, pHT and Ωa were also lower in coral vials than in blanks while pCO2 increased (Table 2). For both species, pHT was lower by 0.05 units and Ωa was lower by 0.03, whereas pCO2 increased by 52 for M. oculata and 60 µatm for L. pertusa. Cold-water corals can release significant amounts of inorganic nutrients , and phosphate concentrations increased by 0.413 and 0.483 µmol kg−1 to 0.44±0.49 and 0.51±0.34 µmol kg−1, while ammonium increased by 6.80 and 3.81 µmol kg−1 to 7.83±4.97 and 4.84±5.17 µmol kg−1 for M. oculata and L. pertusa (N = 18 and 22), respectively.
Before changing the pCO2 to respective treatment levels A–D (low-high pCO2), mean calcification rates were 0.006±0.006% d−1 for L. pertusa (N = 18) and 0.023±0.022% d−1 for M. oculata (N = 23) (Table 1). For both species, the rates of net calcification were significantly correlated with skeletal weight. Smaller coral fragments exhibit higher net calcification rates following a negative logarithmic trend (Figure S2). This negative dependence of size and age for cold-water coral calcification is in accordance with earlier studies , , . However, despite the fact that we worked with a relatively large size range, neither size (weight or polyp number), nor initial calcification rates measured under ambient pCO2 (Table 1) differed significantly between the coral fragments used in the four pCO2 treatments (1-way ANOVA, p≥0.15) and net calcification rates were well within the range of earlier findings –, , .
Carbonate Chemistry of Blanks after Adjusting pCO2 Levels
After the CO2 was adjusted to the 4 pCO2 levels (A–D), the carbonate chemistry changed accordingly (Table 2). For the blanks, AT was relatively uniform around 2600 µmol kg−1 for treatments A to D. The CT of blanks increased from 2313±3 to 2492±4 µmol kg−1, and pHT decreased from 8.14±0.01 to 7.76±0.01, Ωa from 3.2±0.04 to 1.5±0.03 and pCO2 levels ranged from 349±6 to 929±25 µatm for treatments A to D, respectively. For time step T1–T4, AT was not significantly different between treatments A–D (One-way ANOVA, p>0.2), while it was significant between maintenance vials buoyant weight, treatment B and D, and buoyant weight, treatment D and T4 treatment D. Other parameters of the carbonate chemistry (CT, pCO2, pH and Ωa) were in general significantly different between single pCO2 treatments (One-way ANOVA, Tukey Honest-Significant Difference (HSD) post-hoc test, p<0.05) with exceptions for adjacent treatment levels where p-values between treatments were >0.05. The actual pCO2 values of seawater differed from those applied by the MFCs and revealed a reduced range between 349 to 929 instead of 280 to 1000 µatm, respectively. This is probably due to the fact that all 4 pCO2 treatments were maintained and incubated in the same climate room and mixing of seawater with the overlying air most likely took place due to the vertical water circulation that was generated by the aeration system.
Carbonate Chemistry and Calcification Rates in Coral Incubation after Adjusting pCO2 Levels
For coral incubation of T1–T4, the AT of seawater containing L. pertusa was on average 2496±45, while AT for M. oculata incubation was on average 2520±21 (Table 2 and Table S1). The CT of L. pertusa increased from treatment A–D from 2236±23 to 2492±4 µmol kg−1 while CT of M. oculata increased from 2256±11 to 2408±7 µmol kg−1 for treatment A to D, respectively. Similar as for the T0, the DIC, pHT and Ωa were slightly lower in coral vials than in corresponding blanks while pCO2 was higher (Table 2). This means, that also other parameters of the carbonate system than the AT changed during the 2-day incubation as a consequence of coral calcification and metabolism. It is evident, that these shifts cannot be avoided as the determination of calcification rates is based on the fact, that the precipitation of 1 mole CaCO3 decreases the AT by 2 mole and CT by 1 mole . Also, the CO2 released by coral respiration and calcification into the surrounding seawater was apparently not completely equilibrated by aeration with respective air-CO2 gas mix and the pCO2 levels of coral vials increased by an average 20 to 40 µatm during the 2-day incubation (Table 2). The AT decreased by about 100 µmole relative to blanks, and there was no significant difference of AT between pCO2 treatments or repeated incubation T1 to T4 at the end of incubation (Table S2). The excretion of inorganic nutrients by cold-water corals increases the AT, i.e. it counteracts the decrease caused by calcification. Thus, calcification rates determined by the alkalinity anomaly technique are an underestimate of actual calcification rates if not corrected for. In a previous study using freshly collected cold-water corals it has been demonstrated that calcification rates of both species would be underestimated by 10% if inorganic nutrients were not taken into account . In the present study, inorganic nutrient concentrations were used to correct calcification rates using the following linear equations: GL.persusa = 1.025*Guncorrected +000019 (R = 0.997, N = 81, p<<0.001) and GM.oculata = 1.027*Guncorrected +0.00093 (R = 0.993, N = 102, p<<0.001) (Figure S1). This means that the ratio of inorganic nutrient release to net calcification was 4-times higher for freshly collected corals than for corals maintained in aquaria.
Calcification Rates and pCO2 Treatment Effects
After the pCO2 had been adjusted to the intended treatment levels, calcification rates were measured immediately (T1) and 1, 2 and 3 months (T2–T4) after adjusting the pCO2 levels using the alkalinity anomaly technique and additionally after 9 months using the buoyant weighing technique. Pooled data from repeated measurements (average T1–T4) of alkalinity anomaly corresponding to approximately 3-months of coral growth and the buoyant weight after 9-months provided similar results and average calcification rates of L. pertusa slightly varied between treatments A and D from 0.011±0.008 to 0.017±0.012% d−1 for the alkalinity anomaly method and between 0.010±0.008 and 0.021±0.037% d−1 for the buoyant weight technique (Figure 1, Table S3). For M. oculata average values ranged between 0.023±0.012 and 0.035±0.025% d−1 and between 0.017±0.014 and 0.038±0.057% d−1 for the alkalinity anomaly and buoyant weight technique, respectively. For both coral species, there was neither a significant effect between methods used to measure calcification rates (time span of maintenance at respective pCO2 levels) nor a significant pCO2 treatment effect or a combined effect of methods (exposure time) and pCO2 levels (Repeated measures ANOVA, p>0.05, Table S4A). As we used two different methods to establish net calcification rates over mid-term and longer-term calcification, we provide a methods comparison for a similar time interval (mid-term) between alkalinity anomaly and buoyant weight showing that the 2 methods provide similar results (SI 1). These data were not included in the manuscript, as they do not comprise the same N, which is mandatory for the repeated measures design used here.
Short-term (Shock) and Longer-term (Acclimation) Response to Variable pCO2
The experimental approach used in the present study was designed to discriminate between short-term “shock” effects due to fast changes in pCO2 levels and longer-term acclimation response of net calcification. The results from repeated determination of calcification rates over time revealed no short-term, or long-term response in calcification to higher pCO2 for either of the two species investigated (Figure 2). There was a continuous increase in calcification rates for the coral L. pertusa in treatment D which had highest pCO2 levels (Figure 1), however this increase was statistically not significant (repeated measures ANOVA, Table S4B). Thus, our data do not support an earlier suggestion for a potential short-term shock response with long-term acclimation of net calcification rates as proposed by Form & Riebesell . Yet, there was an effect of time on net calcification rates for both coral species. In general, average calcification rates (pooled for pCO2 treatments) increased until T3 and then decreased at T4 again (repeated measures ANOVA, unequal N HSD, Table S4C). This pattern was independent of pCO2 treatment and similar for L. pertusa and M. oculata. This indicates, that factors other than pCO2 must have been responsible for the changes in calcification rates with time, that were either driven by intrinsic controls (e.g. reproductive cycles) or a general acclimation to the maintenance conditions in the vials or aquarium (independent of pCO2).
oculata and b) L. pertusa at T0–T4 at the 4 pCO2 treatment levels (pCO2 values at different time steps and for the two species are given in Table S1). The pCO2 at T0 was at ambient pCO2 and served as initial control, while T1–T4 was after pCO2 has been changed to respective treatment levels. Values are mean ± SE and SD.
The lack of pCO2 treatment effects seems to contradict other studies reporting on a short-term calcification response of L. pertusa to elevated pCO2 levels , . In the following, an attempt is made to reconcile these contradictory findings. Form & Riebesell found a negative correlation between calcification rates and CO2 concentration in their short-term experiment. However, this negative correlation might have been due to one very high value for calcification rates at the lowest pCO2 which forced the slope to a negative trend while all other values were within a certain range independent of pCO2 (; Figure 2) similar to the findings of the present study. In the short-term study by Maier et al.  there was also a clear negative response to increasing pCO2 on net calcification rates. In that study, the experimental set up was different as very small incubation vials (50 ml) and a closed system approach was used. Thus, the initial values for carbonate chemistry were comparable with the present study and that of Form & Riebesell , however, the carbonate chemistry changed drastically during incubation and mean pCO2 values as high as 2160 µatm were reached (; Table S4). The conclusion with respect to an expected 50% decrease in calcification rates by the end of the century might thus have been a misinterpretation with respect to relating net calcification rates to the initial pCO2 range and not to the much higher pCO2 values that were actually reached during incubation due to the closed system approach and small incubation volume.
For M. oculata, so far only one study exists with respect to ocean acidification and it addresses the short-term pCO2 effect on net calcification rates . In that study, the pCO2 was changed in two ways: 1) to that of pre-industrial concentrations (285 ppm) and 2) to values projected for the end of the century (865 ppm). Similar to the present study, there was no change in net calcification rates between ambient and future pCO2 levels, while calcification rates doubled when pCO2 was set to values of pre-industrial times indicating that the increase in pCO2 that took place since pre-industrial times had already a negative impact on M. oculata calcification.
The results of the present study are in contrast to the acclimation hypothesis postulated by Form & Riebesell . Their conclusion with respect to acclimation was based on the fact, that there was a negative short-term response, while in the long-term study higher pCO2 had even higher calcification rates than the two lower pCO2 levels. However, there is indication that their data can be interpreted differently. First, the long-term experiment of Form & Riebesell  lacked the ambient pCO2 control and did thus not cover the same pCO2 range than their short-term study. However, an ambient pCO2 treatment in the long-term study would have been pivotal, specifically because the negative response in the short-term experiment was caused by a higher calcification rate at ambient pCO2. Second, the study by Form & Riebesell  also lacked initial controls. It is therefore possible, that the coral fragments used in the high pCO2 treatment had already higher initial calcification rates under ambient pCO2 conditions. In contrast, the experimental design of the present study comprised both initial and ambient controls and a time-series for calcification rates which allowed a better evaluation of shock or short-term responses versus long-term acclimation in calcification rates. Due to a lack in short-term response of calcification for the range in pCO2 levels studied, our data do not provide evidence that acclimation had played a role in the long-term calcification response to increasing pCO2. This does not mean that no acclimation took place, it rather means that the mechanism(s) enabling cold-water corals to maintain calcification rates constant over a large pCO2 range (independent of the duration of the exposure) remain to be identified.
Summary View of pCO2 Effects on Cold-water Coral Calcification
For a pCO2 range between 350 and 1000 µatm, no effect on net calcification rates, neither for short- nor long-term exposure could be distinguished, and there is evidence that the recently postulated cold-water coral acclimation hypothesis  does not hold as such. The present study revealed significant changes observed as function of time for all pCO2 treatments, but this must be attributed to other causes than pCO2 either related to aquarium conditions or coral biology. Including the previous 3 studies on cold-water coral response to ocean acidification and the results of the present study a certain pattern emerges: the response of cold-water corals L. pertusa and M. oculata to increasing pCO2 is non-linear and net calcification rates remain constant for a pCO2 range between ambient and somewhere above 1000 µatm where Ωa is already close to or even below 1. The negative short-term response in the study by Maier et al. (2009) indicated, that once a threshold at high pCO2 has been reached, a significant decline in net calcification rates can be expected with increasing pCO2 it can further be assumed that this threshold lies somewhere below 2000 µatm. For the Mediterranean coral M. oculata it appears that a 1st threshold had already been surpassed since pre-industrial as indicated by the increase in calcification rates of M. oculata at reduced pCO2 . In this respect, the exceptionally high calcification value of L. pertusa at lowest pCO2 in the short-term experiment of Form & Riebesell  might be indicative of such a threshold between present-day and pre-industrial pCO2, however, this needs further investigation.
A non-linear response is contrary to findings for tropical, zooxanthellate corals, that generally reveal a linear, negative calcification response to increasing pCO2 –. However, a non-linear response had already been revealed for temperate zooxanthellate corals which, similar to the Mediterranean cold-water corals, remained unaffected within a large range of present-day to end of the century projections of pCO2 , ,  and a drastic reduction in calcification at a pCO2 of 2850 µatm . The responses of the organisms to increasing ocean acidification are variable and complex  and even enhanced calcification at higher pCO2 had been proposed for some taxa –.
Up to now it is unclear how the corals are able to resist increasing pCO2 levels and how they maintain calcification rates constant over such a large pCO2 gradient. It has been proposed that cold-water corals are able to resist increasing ocean acidification by their ability to maintain a high pH within their calicoblastic, calcifying fluid . The way the calcifying fluid is sheltered and replenished with cations from ambient seawater is crucial in how a coral responds to increasing ocean acidification –. Also, an explanation why temperate corals can resist to higher pCO2 levels are their lower growth rates  as less carbonate ions will be required in the same time for calcification. This could also explain why cold-water corals grow in waters with an Ωa around 1, while fast growing tropical corals are found in waters with an Ωa above 3.5. Overall, cold-water corals seem well adapted to low Ωa which may explain their resistance to increasing pCO2 levels up to a certain threshold. Tropical corals can experience drastic short-term changes of pCO2 in the close environment due to diurnal changes in CO2 uptake and release driven by light-dependent changes in the metabolic functioning of reef organisms . Nothing is known on naturally occurring short-term changes in the seawater carbonate chemistry close to cold-water corals but due to the lack of photosynthetic activity in these depths they are likely less pronounced than in tropical reef systems. Nevertheless the question remains if any naturally occurring short-term changes render cold-water corals resistant to fast changes in pCO2 and a large range of pCO2 with values reaching more than 1000 µatm. Finally, it is still questionable if acclimation of calcification to increasing pCO2 is a likely scenario in the natural environment. The resistance to increasing pCO2 levels and the maintenance of constant calcification requires energy and might thus be sustainable during short-term exposure while energy requirements might not be sustained over longer time of exposure to higher pCO2. This might especially be the case in the natural cold-water coral environment, where regular food supply as usually provided during aquarium maintenance is not always guaranteed and where other stressors such as predation, disease and temperature abnormalities may further impede coral growth.
Calcification rates (G) of M. oculata and L. pertusa corrected and uncorrected for inorganic nutrient reslease during incubation. Correlation analysis were significant with R = 0.993, N = 102, p<<0.001 and R = 0.997, N = 81, p<<0.001 for M. oculata and L. pertusa, respectively.
Calcification rates (G) versus skeleletal weight (SW) of M. oculata and L. pertusa under ambient pCO2 conditions (T0). Logarithmic regressions are significant with p = 0.01 for both species; N = 23 and 18 for M. oculata and L. pertusa, respectively.
Parameters of the carbonate chemistry (CC) and the inorganic nutrients (IN) phosphate (PO4) and ammonium (NH4) for the incubation times T0–T4 and pCO2 treatments A–D after 2-day incubation using the alkalinity anomaly technique. Additionally, CC was established after 9 months (267 days) when net calcification rates were measured using the buoyant weight (BW) technique. T0 was established prior to adjusting pCO2 at ambient for all treatments, while T1–T4 and BW are values derived from 2-day incubations immediately and 1, 2 and 3 months after adjusting pCO2 and after 9 months, respectively. LP (Lophelia pertusa), MO (Madrepora oculata) and blank (incubated in parallel without coral) at respective pCO2 treatment levels. Values are given as mean ± S.D.
Post-hoc results (Tukey honest-significance difference test for unequal N) of breakdown ANOVA for parameters of the carbonate chemistry (AT, CT, pCO2, pHT, Ωa) for the coral incubations with T1–T4 (directly and 1, 2, 3 months after pCO2 was changed to respective treatment levels) established during 2-day incubation using alkalinity anomaly and after 9 months (267 days) using buoyant weight (BW) technique to determine net calcification rates. Matrix of p-values for L. pertusa at upper right and for M. oculata at lower left part of the table with p<0.05 (italic). Bold values are corresponding treatments at different incubation times.
Average calcification rates (G) of the cold-water corals Lophelia pertusa (LP) and Madrepora oculata (MO) at T0 to T4 determined by the total anomaly technique in monthly time intervals and 2 days of incubation, and calcification rates determined by buoyant weight (G(BW)) after maintenance of corals for 9–10 months under respective pCO2 treatment levels.
Statistical results for repeated measures ANOVA of calcification rates (G). A Comparison between total alkalinity (TA) method (average G, pooled T1–T4) and Buoyant Weighting (BW) for pCO2 treatments A–D; and B for comparison of repeated measurements T0–T4 for the 4 pCO2 treatments. Table C gives the matrix for p-values of the Tukey-Honest-Significance post-hoc comparison for unequal N of the variable R1 (T0–T4) for M. oculata (lower left) and L. pertusa (upper right). Significant p are marked in bold, italic
We thank captain, crew and scientific shipboard staff of R/V Minibex (COMEX). Cruises on RV Minibex were part of a project on canyons of the Mediterranean Sea (MedSeaCan) lead by the French Marine Protected Areas Agency (AAMP), France. Thanks to F. Bils for help with coral maintenance and experiments and S. Comeau for assistance in analysis of AT.
Conceived and designed the experiments: CM MW JPG PW. Performed the experiments: CM AS MMBS. Analyzed the data: CM AS MMBS. Contributed reagents/materials/analysis tools: CM JPG MW PW. Wrote the paper: CM.
- 1. IPPC (2007) Climate change 2007: the physical science basis. Cambridge University Press, Cambridge, UK.
- 2. Orr JC (2011) Recent and future changes in ocean carbonate chemistry. In Gattuso, J-P & Hansson, L, editors. Ocean Acidification. Oxford University Press, Oxford. 41–66.
- 3. Plattner G-K, Joos F, Stocker TF, Marchal O (2001) Feedback mechanisms and sensitivities of ocean carbon uptake under global warming. Tellus 53B: 564–592.
- 4. Riebesell U, Zondervan I, Rost B, Tortell PD, Zeebe RE, et al. (2000) Reduced calcification of marine plankton in response to increased atmospheric CO2. Nature 407: 364–367.
- 5. Fabry VJ (2008) Marine calcifiers in a high-CO2 ocean. Science 320: 1020 .
- 6. Hall-Spencer JM, Rodolfo-Metalpa R, Martin S, Ransome E, Fine M, et al. (2008) Volcanic carbon dioxide vents show ecosystem effects of ocean acidification. Nature 454: 96–99.
- 7. Martin S, Gattuso J-P (2009) Response of Mediterranean coralline algae to ocean acidification and elevated temperature. Global Change Biol 15: 2089–2100.
- 8. Albright R (2011) Reviewing the effects of ocean acidification on sexual reproduction and early life history stages of reef-building corals. J Mar Biol , 14pp.
- 9. Gazeau F, Gattuso J-P, Dawber C, Pronker AE, Peene F, et al. (2010) Effect of ocean acidification on the early life stages of the blue mussel Mytilus edulis. Biogeosciences 7: 2051–2060.
- 10. Doropoulos C, Ward S, Diaz-Pulido G, Hoegh-Guldberg O, Mumby PJ (2012) Ocean acidification reduces coral recruitment by disrupting intimate larval-algal settlement interactions. Ecol Lett 15: 338–346 .
- 11. Kleypas JA, Buddemeier RW, Gattuso J-P (2001) The future of coral reefs in an age of global change. Int J Earth Sciences 90: 426–437.
- 12. Kleypas JA, Yates KK (2009) Coral reefs and ocean acidification. Oceanography 22: 108–117.
- 13. Langdon C, Atkinson MJ (2005) Effect of elevated pCO2 on photosynthesis and calcification of corals and interactions with seasonal change in temperature/irradiance and nutrient enrichment. J Geophys Res 110: C09S07 .
- 14. De’ath G, Lough JM, Fabricius KE (2009) Declining coral calcification on the Great Barrier Reef. Science 323: 116–119.
- 15. Gattuso J-P, Allemand D, Frankignoulle M (1999) Photosynthesis and calcification at cellular, organismal and community levels in coral reefs: A review on interactions and control by carbonate chemistry. Amer Zool 39: 160–183.
- 16. Hoegh-Guldberg O, Mumby PJ, Hooten AJ, Steneck RS, Greenfield P, et al. (2007) Coral reefs under rapid climate change and ocean acidification. Science 318: 1737 .
- 17. Fabricius KE, Langdon C, Uthicke S, Humphrey C, Noonan S, et al. (2011) Losers and winners in coral reefs acclimatized to elevated carbon dioxide concentrations. Nature Climate Change 1: 165–169 .
- 18. Ries JB, Cohen AL, McCorkle DC (2009) A nonlinear calcification response to CO2-induced ocean acidification by the coral Oculina arbuscula. Geology 37: 1057–1152.
- 19. Maier C, Hegeman J, Weinbauer MG, Gattuso J-P (2009) Calcification of the cold-water coral Lophelia pertusa under ambient and reduced pH. Biogeosciences 6: 1671–1680.
- 20. Form AU, Riebesell U (2012) Acclimation to ocean acidification during long-term CO2 exposure in the cold-water coral Lophelia pertusa. Global Change Biol 18: 843–853 .
- 21. Maier C, Watremez P, Taviani M, Weinbauer MG, Gattuso J-P (2012) Calcification rates and the effect of ocean acidification on Mediterranean cold-water corals. Proc. R. Soc. Lond., B 279: 1713–1723 .
- 22. Orr JC, Fabry VJ, Aumont O, Bopp L, Doney SC, et al. (2005) Anthropogenic ocean acidification over the twenty-first century and its impact on calcifying organisms. Nature 437: 681–686.
- 23. Guinotte JM, Orr J, Cairns S, Freiwald A, Morgan L, et al. (2006) Will human-induced changes in seawater chemistry alter the distribution of deep-sea scleractinian corals? Front Ecol Environ 4: 141–146.
- 24. Thresher RE, Tilbrook B, Fallon S, Wilson NC, Adkins J (2011) Effects of chronic low carbonate saturation levels on the distribution, growth and skeletal chemistry of deep-sea corals and other seamount megabenthos. Mar Ecol Prog Ser 442: 87–99.
- 25. Freiwald A, Beuck L, Rüggeberg A, Taviani M, Hebbeln D (2009) The white coral community in the central Mediterranean Sea revealed by ROV surveys. Oceanography 22: 36–52.
- 26. Roberts JM, Wheeler AJ, Freiwald A (2006) Reefs of the deep: The biology and geology of cold-water coral ecosystems. Science 312: 543–547.
- 27. Taviani M, Freiwald A, Zibrowius H (2005) Deep coral growth in the Mediterranean Sea: an overview. In: Freiwald A & Roberts JM editors. Cold-water corals and ecosystems. Springer-Verlag, Berlin Heidelberg. 137–156.
- 28. Touratier F, Goyet C (2011) Impact of the Eastern Mediterranean Transient on the distribution of anthropogenic CO2 and first estimate of acidification for the Mediterranean Sea. Deep Sea Res I 58: 1–15.
- 29. Chisholm JRM, Gattuso J-P (1991) Validation of the alkalinity anomaly technique for investigating calcification and photosynthesis in coral reef communities. Limnol Oceanogr 36: 1232–1239.
- 30. Davies PS (1989) Short-term growth measurements of corals using an accurate buoyant weighing technique. Mar Biol 101: 389–395.
- 31. Reynaud S, Ferrier-Pagès C, Meibom A, Mostefaoui S, Mortlock R, et al. (2007) Light and temperature effects on Sr/Ca and Mg/Ca ratios in the scleractinian coral Acropora sp. Geochim. Cosmochim Acta 71 (2): 354–362.
- 32. Lavigne H, Gattuso J-P (2011) seacarb: seawater carbonate chemistry with R. R package version 2.4. Available: http://CRAN.R-project.org/package=seacarb. Accessed 9 September 2012.
- 33. Maier C, Kluijver AD, Agis M, Brussaard CPD, van Duyl FC, et al. (2011) Dynamics of nutrients, total organic carbon, prokaryotes and viruses in onboard incubations of cold-water corals. Biogeosciences 8: 2609–2620.
- 34. Orejas C, Ferrier-Pagès C, Reynaud S, Gori A, Beraud E, et al. (2011) Long-term growth rates of four Mediterranean cold-water coral species maintained in aquaria. Mar Ecol Prog Ser 429: 57–65.
- 35. Orejas C, Gori A, Gili JM (2008) Growth rates of live Lophelia pertusa and Madrepora oculata from the Mediterranean Sea maintained in aquaria. Coral Reefs 27: 255.
- 36. Mortensen PB (2001) Aquarium observations on the deep-water coral Lophelia pertusa (L., 1758) (Scleractinia) and selected associated invertebrates. Ophelia 54: 83–104.
- 37. Wolf-Gladrow DA, Zeebe RE, Klaas C, Körtzinger A, Dickson AG (2007) Total alkalinity: The explicit conservative expression and its application to biogeochemical processes. Deep Sea Res I 106: 287–300.
- 38. Rodolfo-Metalpa R, Martin S, Ferrier-Pagès C, Gattuso J-P (2010) Response of the temperate coral Cladocora caespitosa to mid-and long-term exposure to pCO2 and temperature levels projected for the year 2100 AD. Biogeosciences 7: 289–300.
- 39. Rodolfo-Metalpa R, Houlbrèque F, Tambutté É, Boisson F, Baggini C, et al. (2011) Coral and mollusc resistance to ocean acidification adversely affected by warming. Nature Climate Change 1: 308–311.
- 40. Ries JB, Cohen AL, McCorkle DC (2010) A nonlinear calcification response to CO2-induced ocean acidification by the coral Oculina arbuscula. Coral Reefs 29: 661–674.
- 41. Fiorini S, Middelburg JJ, Gattuso J-P (2011) Testing the effects of elevated pCO2 on coccolithophores (Prymnesiophyceae): Comparison between haploid and diploid life stages. J Phycol: .
- 42. Iglesias-Rodriguez MD, Halloran PR, Rickaby REM, Hall IR, Colmenero-Hidalgo E, et al. (2008) Phytoplankton calcification in a high-CO2 world. Science 320: 336–340.
- 43. Wood HL, Spicer JI, Widdicombe S (2008) Ocean acidification may increase calcification rates, but at a cost. Proc R Soc Lond B 275: 1767–1773.
- 44. McCulloch M, Trotter J, Montagna P, Falter J, Dunbar R, et al. (2012) Resilience of cold-water scleractinian corals to ocean acidification: Boron isotopic systematics of pH and saturation state up-regulation. Geochim Cosmochim Acta 87: 21–34 .
- 45. Tambutté S, Holcomb M, Ferrier-Pagès C, Reynaud S, Tambutté E, et al. (2011) Coral biomineralization: From the gene to the environment. J exp mar Biol Ecol 408: 58–78.
- 46. Gagnon AC, Adkins JF, Erez J (2012) Seawater transport during coral biomineralization. Earth Planet Sci 329–330: 150–161.
- 47. Gattuso J-P, Frankignoulle M, Smith SV (1999) Measurement of community metabolism and significance in the coral reef CO2 source-sink debate. Proc Natl Acad Sci 96: 13017–13022.