Calcification in Three Common Calcified Algae from Phuket, Thailand: Potential Relevance on Seawater Carbonate Chemistry and Link to Photosynthetic Process

Calcifying macroalgae contribute significantly to the structure and function of tropical marine ecosystems. Their calcification and photosynthetic processes are not well understood despite their critical role in marine carbon cycles and high vulnerability to environmental changes. This study aims to provide a better understanding of the macroalgal calcification process, focusing on its relevance concerning seawater carbonate chemistry and its relationship to photosynthesis in three dominant calcified macroalgae in Thailand, Padina boryana, Halimeda macroloba and Halimeda opuntia. Morphological and microstructural attributes of the three macroalgae were analyzed and subsequently linked to their calcification rates and responses to inhibition of photosynthesis. In the first experiment, seawater pH, total alkalinity and total dissolved inorganic carbon were measured after incubation of the macroalgae in the light and after equilibration of the seawater with air. Estimations of carbon uptake into photosynthesis and calcification and carbon release into air were obtained thereafter. Our results provide evidence that calcification of the three calcified macroalgae is a potential source of CO2, where calcification by H. opuntia and H. macroloba leads to a greater release of CO2 per biomass weight than P. boryana. Nevertheless, this capacity is expected to vary on a diurnal basis, as the second experiment indicates that calcification is highly coupled to photosynthetic activity. Lower pH as a result of inhibited photosynthesis under darkness imposes more negative effects on H. opuntia and H. macroloba than on P. boryana, implying that they are more sensitive to acidification. These effects were worsened when photosynthesis was inhibited by 3-(3,4-dichlorophenyl)-1,1-dimethylurea, highlighting the significance of photosynthetic electron transport-dependent processes. Our findings suggest that estimations of the amount of carbon stored in the vegetated marine ecosystems should account for macroalgal calcification as a potential carbon source while considering diurnal variations in photosynthesis and seawater pH in a natural setting.


Introduction
Since the mid-18th century, atmospheric CO 2 has risen from 280 to 400 ppm, and is predicted to increase to 730-1200 ppm by 2100 as a result of human activities, such as the burning of fossil fuel, deforestation, agriculture and industrialization [1,2]. This has led to an increase in sea-surface temperature (ocean warming) and a decrease in ocean pH (ocean acidification), which have affected a range of marine floras and faunas [3][4][5]. Macroalgae play important roles as primary producers in the marine food chain and serve as habitat for various animals, which directly contribute to the livelihood and food security of the coastal community [6,7]. Recent studies have suggested that macroalgae may play a role in carbon sequestration, thus mitigating the changing climate, and offset anthropogenic carbon emissions to the atmosphere [8][9][10][11].
The biological processes of the three macroalgae incubated in the light presented changes in seawater carbonate chemistry ( Figure 2, Table 2). No significant alteration in controls (seawater without macroalgae) confirmed that the observed change in other treatments was associated with biological processes. Repeated-measures ANOVA showed that macroalgal species (p < 0.001), time of measurements (initial, after 3 h in the light and after equilibration with air, p < 0.001) and an interaction of the two factors (p < 0.001) significantly affected seawater pH, TA and total DIC ( Figure 2). After exposure to the light for 3 h, a significant increase in seawater pH was detected in incubated P. boryana (reaching 8.79 ± 0.15, Fisher's LSD test, p < 0.05), a reduction in seawater pH was detected in incubated H. opuntia (Fisher's LSD test, p < 0.05) and no significant change in seawater pH compared to the initial values was observed in incubated H. macroloba (Figure 2A). Seawater pH was restored to the initial level after P. boryana was removed and the seawater was allowed to be equilibrated with air. In contrast, seawater pH remained low in treatment with H. opuntia (Fisher's LSD test, p < 0.05 compared to the initial value) and reduced in treatment with H. macroloba (Fisher's LSD test, p < 0.05). A significant decrease in TA was detected in the seawater incubated with all species tested after light exposure (Fisher's LSD test, p < 0.05, Figure 2B). After equilibration, a recovery of TA was observed in treatment with P. boryana, whereas TA in the other two treatments remained low. Total DIC followed the same trend as observed in TA ( Figure 2C). These shifts in pH and TA resulted in a change in CO 2 , HCO 3 − , CO 3 2− , saturation of calcite (ΩCa) and saturation of aragonite (ΩAr) ( Table 2). Incubation of P. boryana in the light resulted in a decrease in CO 2 and HCO 3 − and an increase in CO 3 2− , ΩCa and ΩAr in seawater (Fisher's LSD test, p < 0.05). After equilibration, CO 2 and HCO 3 − were restored to the initial levels, whereas CO 3 2− ΩCa and ΩAr in the seawater became slightly lower than the initial values ( except for CO2, which showed an increase in H. opuntia (Fisher's LSD test, p < 0.05) and a slight decrease from the initial value, although not statistically significant in H. macroloba. Equilibration reinstated CO2 concentration but did not restore HCO3 − CO3 2− ΩCa and ΩAr (Table 2, Fisher's LSD test, p < 0.05). The salinity and temperature of the seawater remained steady throughout the experiments (30 ppt; 30 ℃-31 ℃).  Table 2. A summary of seawater carbonate chemistry before an incubation with Padina boryana, Halimeda opuntia and Halimeda macroloba (initial), after 3 hours of incubation and after equilibration with air. The values presented were calculated from the average total alkalinity and pH (see Figure 2) using CO 2 sys.xls program (version 2.3). Data are means ± SE (n = 9). Different uppercase letters indicate statistical difference (Fisher's LSD test, p < 0.05). Fates of the DIC associated with biological processes differed among species ( Figure 3). Photosynthetic carbon uptake was highest in P. boryana (Fisher's LSD test, p < 0.05), whereas H. opuntia and H. macroloba displayed comparable rates ( Figure 3A), corroborating the pH drift and DIC results previously described in Figure 2 and Table 2. Opposite trends were observed in carbon removal by calcification and carbon loss due to biological processes ( Figure 3B,C). The amount of carbon associated with both processes was lowest in P. boryana (Fisher's LSD test, p < 0.05), whereas no difference was detected between H. opuntia and H. macroloba. With lightly calcified algae, P. boryana was more engaged in photosynthetic carbon uptake than in calcification, resulting in no net carbon loss. In contrast, two heavily calcified macroalgae removed a larger amount of carbon for calcification than for photosynthesis. The consequential carbon loss was almost as high as their carbon removal for calcification.  [20]. Different letters above the bar indicate statistical difference (Fisher's LSD test, p < 0.05). Data are means ± SE (n = 9).

Initial
There was a significant positive relationship between carbon removal by calcification and carbon loss to air in P. boryana and H. opuntia ( Figure 4A, p = 0.005 in P. boryana and p = 0.0124 in H. opuntia), although both were not significantly related in H. macroloba (p = 0.0658). A trend line with an intercept = 0 was fitted using all the data points, and the ratio of carbon loss to air to carbon removal by calcification was 0.78. In addition, a significant positive relationship between photosynthetic carbon uptake and removal by calcification  [20]. Different letters above the bar indicate statistical difference (Fisher's LSD test, p < 0.05). Data are means ± SE (n = 9).
There was a significant positive relationship between carbon removal by calcification and carbon loss to air in P. boryana and H. opuntia ( Figure 4A, p = 0.005 in P. boryana and p = 0.0124 in H. opuntia), although both were not significantly related in H. macroloba (p = 0.0658). A trend line with an intercept = 0 was fitted using all the data points, and the ratio of carbon loss to air to carbon removal by calcification was 0.78. In addition, a  Inhibition of light reactions of photosynthesis affected the calcification rates of the three macroalgae and the pH drift of the seawater ( Figure 5). Significant effects of incubation conditions (light and darkness) on calcification rates of the three macroalgae were detected (repeated-measures ANOVA, p < 0.001, Figure 5A). Similar to the results reported in Figure 3, calcification rates in P. boryana were lower than those in H. opuntia and H. macroloba (Fisher's LSD test, p < 0.05). Calcification rates were lowered when macroalgae were incubated in darkness (Fisher's LSD test, p < 0.05). P. boryana and H. opuntia maintained positive net calcification rates in darkness; however, net dissolution was detected in H. macroloba. Treatment with DCMU had a strong impact on the calcification process (repeatedmeasures ANOVA, p < 0.001) and induced dissolution in all species tested ( Figure 5B). Seawater pH ( Figure 5C,D) was affected by macroalgal species (repeated-measures ANOVA, p < 0.001), incubation conditions (repeated-measures ANOVA, p < 0.001) and interaction of the two factors (repeated-measures ANOVA, p < 0.001). Incubation conditions did not result in significant change in seawater pH of controls, indicating that changes observed in other treatments were associated with biological processes. An increase in seawater pH was detected after P. boryana was exposed to light for 1 h (Fisher's LSD test, p < 0.05), whereas seawater from the incubation with H. opuntia and H. macroloba remained comparable to controls. Incubation in darkness resulted in a lower seawater pH as compared to incubation in the light in all treatments containing macroalgae (Fisher's LSD test, p < 0.05, Figure 5C). Similar results were observed in the presence of DCMU ( Figure 5D). all species tested ( Figure 5B). Seawater pH ( Figure 5C,D) was affected by macroalgal species (repeated-measures ANOVA, p < 0.001), incubation conditions (repeatedmeasures ANOVA, p < 0.001) and interaction of the two factors (repeated-measures ANOVA, p < 0.001). Incubation conditions did not result in significant change in seawater pH of controls, indicating that changes observed in other treatments were associated with biological processes. An increase in seawater pH was detected after P. boryana was exposed to light for 1 h (Fisher's LSD test, p < 0.05), whereas seawater from the incubation with H. opuntia and H. macroloba remained comparable to controls. Incubation in darkness resulted in a lower seawater pH as compared to incubation in the light in all treatments containing macroalgae (Fisher's LSD test, p < 0.05, Figure 5C). Similar results were observed in the presence of DCMU ( Figure 5D).

Discussion
Our study showed that the biological processes of P. boryana, H. opuntia and H. macroloba significantly alter seawater chemistry but to a different extent, which is probably due to differences in photosynthetic carbon uptake and calcification. While both photosynthetic carbon fixation and calcification contributed to total DIC uptake in P. boryana, photosynthesis appears to be the main driver for a drastic decrease in DIC and a rise in seawater pH. In contrast, DIC uptake of H. opuntia and H. macroloba were largely associated with calcification and to a lesser extent, with photosynthesis, resulting in a change in pH with smaller magnitude. Higher calcification rates in H. opuntia and H. macroloba compared to P. boryana may also explain their higher accumulation of CaCO3 (1.6-1.7 times higher than in P. boryana).

Discussion
Our study showed that the biological processes of P. boryana, H. opuntia and H. macroloba significantly alter seawater chemistry but to a different extent, which is probably due to differences in photosynthetic carbon uptake and calcification. While both photosynthetic carbon fixation and calcification contributed to total DIC uptake in P. boryana, photosynthesis appears to be the main driver for a drastic decrease in DIC and a rise in seawater pH. In contrast, DIC uptake of H. opuntia and H. macroloba were largely associated with calcification and to a lesser extent, with photosynthesis, resulting in a change in pH with smaller magnitude. Higher calcification rates in H. opuntia and H. macroloba compared to P. boryana may also explain their higher accumulation of CaCO 3 (1.6-1.7 times higher than in P. boryana).
Both Padina and Halimeda were identified as active HCO 3 − and CO 2 users by means of CCMs [48]. Both can readily utilize HCO 3 − for photosynthesis, although their actual mechanisms may differ depending on their modes of calcification [48,49]. Morphological characteristics and site of calcification influence calcification and photosynthetic pro-cess [39,50,51]. The sheet-like thallus of P. boryana has higher specific surface area than H. opuntia and H. macroloba thus they are expected to exhibit higher productivity and more efficient nutrient uptake for the same weight [52]. The morphology of primary utricle in Halimeda (e.g., size, orientation) and their lateral adhesion between adjacent primary utricles likely affect the diffusion pathway and ions uptake for CaCO 3 precipitation (e.g., Ca 2− , CO 3 2− ) [39,50,51]. Diffusion pathways between primary utricles also control H + fluxes from the bulk seawater into the primary interutricle spaces or removal of H + , a product of calcification, from the primary interutricle spaces to prevent H + accumulation [39]. In addition, shorter diffusion pathway may lead to lower diffusive resistance for the influx of Ca 2+ and potentially HCO 3 − or CO 3 2− from seawater to primary interutricle space, which potentially increases CaCO 3 precipitation [39,50]. This suggests lower energetic cost to H. opuntia than H. macroloba in calcification [53][54][55], which might explain the higher CaCO 3 content with larger crystal length and width and smaller crystal density in H. opuntia with a smaller surface diameter of the primary utricle and lower diffusion pathway length than H. macroloba. Our findings were also consistent with Peach et al. [39], which found strong inverse relationship between average CaCO 3 content and primary utricle surface diameter.
Variability in calcification rates in relation to photosynthesis among the three macroalgae corresponded well with reduced total alkalinity which did not recover after equilibration. In H. opuntia and H. macroloba, TA and DIC were decreased after incubation and remained low after equilibration, implying that CO 2 loss from the seawater took place, which was consistent with a recent study in Corallina officinalis [20]. In contrast, with lightly calcified algae, P. boryana was more engaged in photosynthetic carbon uptake than in calcification, thus resulting in no net carbon loss. In this study, a positive correlation between carbon loss to air and calcification was presented, and the ratio of carbon loss to air to carbon removal by calcification was 0.78. This estimate is comparable to the values reported in C. officinalis (0.78; [20]) and slightly higher than those reported in the coral reefs (0.6; [56]). Notably, when each species was analyzed separately, different functional constants were obtained. These constants suggest that in order to obtain net carbon loss, a certain rate of calcification must be reached (negative y-intercept). While the results generally imply that macroalgal calcification may act as a source of CO 2 , the carbon economy of the coastal community is controlled by multiple processes. Kalokora et al. [20] suggested that the released CO 2 may be refixed by a photosynthetic process, particularly of marine primary producers, such as seagrasses and other macrophytes. In Thai waters and other tropical habitats, calcified macroalgae, especially species from genus Halimeda, were often found coexisting with fleshy macroalgae and seagrasses [57][58][59]. Therefore, community photosynthesis may enhance calcification by increasing the pH and saturation state of CaCO 3 in the surrounding seawater and simultaneously capture CO 2 released from calcification during daytime [12,57].
There was a significant positive relationship between photosynthetic carbon uptake and carbon removal by calcification in all species tested, similar to previous studies in other calcifying algae [20,28,[60][61][62]. Our findings provide supporting evidence that calcification and photosynthesis in P. boryana, H. opuntia and H. macroloba are closely coupled [24,28,32,63,64] but contradict the results of a study on the congeneric species, Padina pavonica and Padina japonica, which reported a lack of correlation between photosynthesis and carbonate deposition [40]. These discrepant results may be due to differences in calcification capacity, which was significantly higher in P. boryana in the present study [40]. This coupling between photosynthesis and calcification is mediated by a localized modification of seawater chemistry, generating a suitable microenvironment for photosynthesis and calcification. Calcium carbonate deposition of an individual promotes photosynthesis via an increase in proton extrusion, which subsequently increases CO 2 availability readily for photosynthetic carbon uptake of the same individual [20,25,32,65]. Meanwhile, photosynthesis elevates the pH and saturation state of CaCO 3 at the macroalgal thalli surface diffusive boundary layer and at the calcification site, thus promoting calcification [25,31,42,66,67]. Beer and Larkum [64] suggested that calcification in Halimeda discoidea is linked to pH in intercellular spaces controlled by the ratio of photosynthesis to respiration. Recent studies have revealed that higher pH at the diffusive boundary layer in the light in many tropical calcified macroalgae, including Halimeda, is maintained by photosynthetic CCM involving a HCO 3 − -H + symport, OH − antiport and HCO 3 − dehydration catalyzed by external carbonic anhydrase [14,28]. Our study showed that approximately 4.5-4.7 mols of carbon were taken in photosynthesis for every mol carbon fixed into calcification in P. boryana. While regressions in H. opuntia and H. macroloba did not show the x and y intercepts of zero, approximately 0.5-1.0 and 0.5-1.2 mols of carbon were taken in photosynthesis for every mol carbon fixed into calcification in H. opuntia and H. macroloba, respectively. Different ratio numbers were reported in previous studies: 4-8 for Halimeda [68], 1.6 for four deep-water Halimeda [63] and 1 and 1.5 for C. officinalis, respectively [20,60], indicating that this is a variable depending on local adaptation and not only species-specific.
The dependency of calcification on photosynthesis was further illustrated in the subsequent experiment in which both light-dependent and photosystem II (PSII) electron transport-dependent calcification was observed. Incubation in darkness significantly reduced calcification rates in all tested species and simultaneously reduced seawater pH. As H. opuntia and H. macroloba deposit aragonite crystals in the intercellular space between utricles, which is semi-separated from external seawater, these algae may be more resistant to the pH change in surrounding water than P. boryana, which precipitate aragonite crystals externally at the thallus surface [40,68,69]. However, our results indicate that P. boryana was the least pH-sensitive, sustaining the highest percentage of calcification rate relative to the rates in the light (approx. 33%) at the lowest final pH (7.67 ± 0.02). This is evident as Padina is among the persistent macroalgal genus in acidified environments [38,41,[70][71][72][73][74]. Meanwhile, H. macroloba was the most sensitive species, displaying net dissolution after 1 h of darkness at the highest pH (7.74 ± 0.04) among all species. Minimal calcification and dissolution were also observed in species from genus Halimeda in the field condition at nighttime [13,14,75]. Differences in sensitivity to pH between H. opuntia and H. macroloba may be due to different morphological characteristics, such as branching and crystal microstructure [76]. While multitudinous branches forming tight complex structures in H. opuntia may generate greater diffusive boundary layer thickness and consequently allowing better regulation of their microenvironment as previously observed in coralline algae [77], shorter and thinner crystals found in H. macroloba may be more susceptible to dissolution and may require more crystal nucleation, which is a major rate-limiting step for precipitation of CaCO 3 [78]. In addition, inhibiting PSII electron transport using DCMU imposed stronger negative effects on calcification of all tested species, with P. boryana as the most affected. This emphasizes that certain mechanisms associated with calcification are species specific, and the processes and factors associated with functioning photosynthetic electron transport are more important facilitators of calcification than other light-dependent but PSII-independent processes, such as light-triggered H + transport pumps [28,31,64]. Nevertheless, more adverse effects of DCMU could be attributed to a lower pH in the seawater, promoting dissolution in all species [13,38]. Comparisons among different species of calcified macroalgae in our study suggest that CaCO 3 precipitation is linked to photosynthesis and pH, and may vary depending on the morphological features. However, these interrelations remain unclear and require further investigations.
Our results provide experimental evidence that calcification of the three common calcified macroalgae in Phuket, Thailand is a potential source of CO 2 , supporting a recent study on C. officinalis [20]. Calcification by H. opuntia and H. macroloba lead to greater loss in CO 2 per biomass weight than P. boryana. Nevertheless, this capacity varies on a diurnal basis, as calcification is highly coupled to photosynthetic activity. Our results also imply that future acidification will impose more negative effects on H. opuntia and H. macroloba than on Padina boryana, and the effects on calcified macroalgae may be worsened in darkness. Therefore, further investigations on the carbon budget of the coastal ecosystem should account for calcification as a potential carbon source while considering diurnal variations and the buffering capacity of the primary production of marine vegetation in the natural setting [12,26,57].

Sampling and Acclimatization
P. boryana, H. opuntia and H. macroloba were carefully collected with their holdfasts intact from Tung Khen Bay, Phuket (7.810057; 98.404100), where all species were found in high densities and growing on calcareous sediment with depth between 0.5 and 1.0 m. As our investigations were solely focused on the populations growing in the upper sublittoral areas, differences due to varying depth and irradiance were not considered. Cleaned samples were transported to the aquaria facility of Coastal Oceanography and Climate Change Research Center, Prince of Songkla University, in cool boxes containing seawater collected from the sampling site. Additional seawater was collected in sealed dark polypropylene bottles and fixed when necessary. Seawater pH, salinity and total alkalinity (TA) were immediately determined upon arrival at the laboratory (pH = 8.1, salinity = 30 and TA = 2.04 ± 0.05 mM).
At our aquaria facility, the samples were rinsed to remove the remaining sediment and epiphytes. Collected samples were maintained in aquaria filled with sterile seawater (pH = 8.1, salinity = 30 PSU and TA = 1.90 ± 0.08 mM) with constant aeration under ambient temperature (27 • C-32 • C). The irradiance of approximately 150 µmol photons m −2 s −1 was provided with aquarium LED light (A601, Chihiros, Cixi, Zhejiang, China) with a 12:12 h light-dark cycle. An irradiance of 150 µmol photons m −2 s −1 (measured using LI-250A light meter; LICOR, Germany) was chosen based on the minimal saturating irradiance (E k ) derived from the rapid light curves (RLC; Ralph and Gademann [79]) conducted using the collected samples (data not shown). Throughout the acclimation period, pH, salinity and TA were monitored regularly to avoid changes due to evaporation, and distilled water was added as necessary. Prior to the experiment, the health of the macroalgae was evaluated by a pulse amplitude modulated fluorometer (Diving-PAM, Walz, Effeltrich, Germany) using the maximum quantum yield (F v /F m ) as a biomarker. Subsequent experiments were carried out with algal thalli with an F v /F m ≥ 0.7. Experiments were conducted within 7 days of the collection date after which the samples were discarded and fresh materials were collected. Each experimental run consists of all species and all treatments. The run was repeated with new biological materials.

Experimental Design
Each macroalga was weighed to reach the wet weight of approximately 5 g and placed in 250 mL flasks containing a magnetic bar (0.4 × 1.0 cm in size) and filled with 250 mL of seawater. The flasks were tightly closed with rubber stoppers and securely wrapped with parafilm. Flasks filled with seawater without macroalgae were used as controls and were included in the experiment. The flasks were arranged on top of magnetic stirrers where mixing occurs and the effect of the diffusive boundary layer was minimized. The assembly of the experimental set-up was done in dimmed light. Subsequently, an irradiance of approximately 150 µmol photons m −2 s −1 was provided by aquarium LED light (A601, Chihiros, China) for 3 h. The light source was positioned on the side so that the macroalgae were not shaded by the stoppers. When incubation in the light was finished, the macroalgae were removed from the flasks. The remaining seawater without macroalgae was then allowed to be equilibrated with air using a compressed air pump for an additional 4 h (the minimum duration for the seawater to reach stable pH, predetermined from the preliminary results). The macroalgae from each flask were rinsed and dried using a paper towel and kept at −20 • C for further assessment of the surface area, CaCO 3 accumulation and microstructural analyses (See below).
Parameters related to seawater chemistry, namely seawater pH, TA and salinity, were measured at three steps: (1) the initial step before light exposure, (2) after 3 h of light exposure and (3) after equilibration with air. Seawater temperature was also recorded. These parameters were used to calculate total dissolved inorganic carbon (DIC), calcification rates, photosynthetic carbon uptake and carbon loss to air, further detailed in Section 4.2.3. Measurements for each species and each treatment were based on nine biological replicates (n = 9).

Determination of Surface Area, CaCO 3 Accumulation and Microstructural Analyses
Thalli of P. boryana, H. macroloba and H. opuntia were cut into parts (individual lobe for P. boryana and individual segment for H. macroloba and H. opuntia) and spread on a surface lined with rulers, and the image of each sample was taken. The surface area was determined using image processing software (ImageJ V. 1.53e, Rasband W.S., U.S. National Institute of Health, Bethesda, Maryland, USA) from images of samples [80,81].
Samples were rinsed with distilled water and dried until constant weight at 60 • C. The CaCO 3 accumulation of each sample was investigated by dissolving thalli in 5% HCl for 1 h and measuring the dry weights before and after CaCO 3 dissolution using an electronic balance [27].
Images of aragonite CaCO 3 crystals were examined using a scanning electron microscope (SEM Apreo, FEI, Czech Republic) at the Office of Scientific Instrument and Testing, Prince of Songkla University. The instrument was operated at 20 kV w in high vacuum mode and imaged using the secondary in-lens detector. Samples were mounted on aluminum stubs and then placed in a gold coating unit (SPI Supplies, USA) operated at a 40 mm working distance. Length, width and density of aragonite crystals (n = 4-6) were determined using image processing software. Aragonite crystals with complete needle structure at the top layer in the images were chosen. Ten crystals per replicate were used to calculate average size following the method described by Sinutok et al. [35].
Thalli of P. boryana and apical segments of Halimeda were fixed in 2% glutaraldehyde in 0.05 M phosphate buffered saline and stored at 4 • C. Fixed samples were posted-fixed in 1% OsO4 buffer, and dehydrated through a series of ethanol concentrations (70%, 80%, 90% and 100%). Samples embedded in epoxy resin and were cut into thin sections (80 nm) using a microtome (MT XL microtome, RMC, USA) and stained with 2% uranyl acetate (3 min). Stained sections of P. boryana were examined with a transmission electron microscope at 160 kV (JOEL, JEM-2010, Japan), while stained sections of H. opuntia and H. macroloba were examined using a compound microscope (DM500, Leica Microsystems, Germany). Surface diameter of primary utricle (pU) and diffusion pathway length were determined using image processing software as previously described.

Determination of Carbonate Chemistry-Related Parameters and Fates of Carbon Related to Biological Processes
To avoid creating large headspace in incubation flasks, parameters related to seawater chemistry at the initial step were determined from the samplings drawn from the prepared seawater for the experiment and not directly from the flasks. After light exposure and after equilibration, seawater was collected from each flask using a 50 mL syringe with an extended rubber tube. Total alkalinity was determined using a total alkalinity mini-titrator for water analysis (HI-84531-02, Hanna Instruments, Woonsocket, RI, USA) according to the manufacturer's instruction. Salinity was measured using an optical refractometer, and seawater pH was measured using a bench-top pH meter (Ohaus RL150, Russell, USA) according to the manufacturer's instructions.
The measured seawater chemistry parameters were later used to calculate DIC, its distribution among different forms (CO 2 , HCO 3 − and CO 3 2− ) and saturation state of aragonite (ΩAr) and calcite (ΩCa). Calculation was performed using the software CO 2 sys.xls [82].
Calcification rates were calculated following alkalinity anomaly protocol (as described in [20]), whereby a decrease in TA by two equivalents when one mole of CaCO 3 is precipitated [83,84]. Calcification rates were normalized against macroalgal wet weight.
In addition to calcification rates, the difference in DIC and TA in seawater collected at the initial time point after incubation in the light and after equilibration can estimate photosynthetic carbon uptake and carbon loss from the seawater surrounding the macroalga. Estimation of photosynthetic carbon uptake and carbon loss was done as described by Kalokora et al. [20] using the following equations: Photosynthetic carbon uptake = (DIC initial − DIC after light incubation ) − DIC calcification (1) Carbon loss = (DIC initial − DIC after equilibration ) − DIC calcification (2)

Interrelation between Photosynthesis and Calcification
Downscaled and shortened setups were adopted for this experiment. Each macroalga with known and similar wet weight (5 g) was placed in 125 mL flasks filled with 140 mL of sterilized seawater. Measurements for each species and each treatment were based on five biological replicates (n = 5).
This experiment was designed to compare calcification rates in the light to those obtained when photosynthetic light reactions were inhibited. First, we compared calcification rates in light and darkness. During the light exposure, an irradiance of approximately 150 µmol photons m −2 s −1 was provided for 1 h (Aquarium LED light A601, Chihiros, China). Calcification rates in the light were calculated from the TA measured before and after light exposure as previously described. The seawater was subsequently replaced, and macroalgae were maintained in fresh seawater in darkness for an additional 1 h. Calcification rates in the dark (Calcification dark ) were calculated from the TA measured before and after incubation in darkness. Flasks filled with seawater without macroalgae were used as negative controls. Second, we compared calcification rates in the light and in the presence of 0.1 mM 3(3,4-dichlorophenyl)-1,1-dimethylurea (DCMU, the concentration was selected based on the preliminary test, data not shown). Calcification rates in the light were calculated from the TA measured before and after 1 h of light exposure as previously described. The seawater was subsequently replaced with seawater containing 0.5 mM DCMU, and irradiance was provided for an additional 1 h. DCMU inhibits photosynthetic electron transport from PSII to plastoquinone and is widely used as an inhibitor of photosynthesis [85]. The use of DCMU can determine whether the calcification and photosynthetic electron transport are interrelated [64,86]. Seawater prepared for the measurements of calcification rates in the light contained the same volume of ethanol used as the solvent in DCMU treatment (70 µL, corresponding to 0.05% v/v). Flasks filled with seawater without macroalgae were used as negative controls. Calcification rates in the presence of DCMU (Calcification dcmu ) were calculated from the TA measured before and after incubation in the light. Seawater pH at the end of each step was also recorded.

Statistical Analyses
All the analyses were performed using Statistica academic platform version 13.0 (Stat-Soft, Tulsa, OK, USA). Cochran's test was used to verify the assumption of homogeneity of variance prior to conducting the analysis of variance (ANOVA). The statistical significance level was set at p < 0.05 in all analyses. Differences in %CaCO 3 and the CaCO 3 crystal morphology and densities among species were analyzed using one-way ANOVA. The difference in surface diameter of primary utricles (pU) diffusion pathway length was analyzed using a t-test.
To characterize the effects of calcification and its effect on seawater chemistry, water chemistry and calcification rates were analyzed by repeated-measures ANOVA using the time of measurements as the within-group factor and macroalgal species as the categorical factor. Differences in each fate of carbon (carbon uptake by photosynthesis, carbon removal by calcification and carbon lost to the air) among species were analyzed using one-way ANOVA. Fisher's least significant difference (LSD) test was used to compare these parameters across treatments. The relationships between calcification and carbon loss and between photosynthesis and calcification were analyzed by linear regression.
Effects of inhibition of photosynthesis on seawater pH and calcification rates were tested by repeated-measures ANOVA using time of measurements as the within-group factor and macroalgal species as the categorical factor. Fisher's LSD test was used to compare these parameters across treatments.
Author Contributions: Conceptualization, all authors; methodology, all authors; software, all authors; validation, all authors; formal analysis, all authors; investigation, all authors; resources, all authors; data curation, all authors; writing-original draft preparation, all authors; writing-review and editing, all authors., visualization, all authors; supervision, all authors; project administration, all authors; funding acquisition, all authors. All authors contributed equally to this work. All authors have read and agreed to the published version of the manuscript.