# Seagrass can mitigate negative ocean acidification effects on calcifying algae

## Abstract

The ultimate effect that ocean acidification (OA) and warming will have on the physiology of calcifying algae is still largely uncertain. Responses depend on the complex interactions between seawater chemistry, global/local stressors and species-specific physiologies. There is a significant gap regarding the effect that metabolic interactions between coexisting species may have on local seawater chemistry and the concurrent effect of OA. Here, we manipulated CO2 and temperature to evaluate the physiological responses of two common photoautotrophs from shallow tropical marine coastal ecosystems in Brazil: the calcifying alga Halimeda cuneata, and the seagrass Halodule wrightii. We tested whether or not seagrass presence can influence the calcification rate of a widespread and abundant species of Halimeda under OA and warming. Our results demonstrate that under elevated CO2, the high photosynthetic rates of H. wrightii contribute to raise H. cuneata calcification more than two-fold and thus we suggest that H. cuneata populations coexisting with H. wrightii may have a higher resilience to OA conditions. This conclusion supports the more general hypothesis that, in coastal and shallow reef environments, the metabolic interactions between calcifying and non-calcifying organisms are instrumental in providing refuge against OA effects and increasing the resilience of the more OA-susceptible species.

## Introduction

Seagrass meadows and calcifying algae beds are benthic communities that play unique roles in the removal, storage and release of carbon from seawater, via photosynthesis and/or calcification1. Coastal communities are metabolically responsible for 85% of the organic carbon and 45% of the inorganic carbon (Ci) buried in coastal sediments2,3,4. CO2 is essential to photosynthesis, yet its increase in seawater reduces pH and carbonate ions, threatening the calcification process5. However, these ecosystems naturally experience large vertical and horizontal variations in abiotic parameters, namely pCO2 and temperature6, that can vary from 400 to 10,000 µatm2 and 15 to 30 °C7, respectively. Research has suggested that exposure to natural fluctuations alongside possession of phenotypic plasticity may help organisms and populations to resist or acclimate to novel anthropogenic conditions6,8.

Little is known about the existing interactions between calcifying and non-calcifying primary producers under OA and temperature rise. Whether it is via alteration of seawater chemistry, allelopathy or other molecular signaling, neighboring marine plants interact by influencing each other’s metabolisms9. Changes in benthic macrophyte communities are projected for the future10 where altered competition dynamics between fleshy and calcifying algae already have been shown to drive ecosystem shifts under elevated CO2 conditions11. The current incomplete understanding of these interactions and the consequent mechanisms that drive ecosystem changes limit our ability to make realistic predictions for the effects of OA and warming on future community structure.

Seagrasses can act as buffers to OA by absorbing large quantities of CO2 and increasing the pH of seawater12,13,14. Diel pH fluctuations of 0.7–1 pH due to the photosynthesis and respiration of seagrass beds, have been reported in different locations13,15. Increased ambient pH levels during the day can become locally significant to the point where they have a positive effect on the calcification of co-occurring calcifying algae12,13. However, since oceanic conditions are rapidly changing, information is needed about how the presence of seagrasses will affect calcifying algae responses under OA and temperature rise.

Most of the studies regarding the impact of global stressors evaluate the isolated responses of primary producers, using unifactorial models or eventually considering the combined role of OA and temperature rise in the fitness of a specific and isolated biological indicator16. Thus far, the expected trend for seagrasses is neutral to positive physiological responses to OA1, yet the magnitude of change and affinity for DIC species varies17,18. The isolated effects of temperature and CO2 on the seagrass genus Halodule Endlicher17,19,20 and their isolated and combined effects on the calcifying green algae genus Halimeda J.V. Lamouroux have been widely addressed21,22,23,24,25,26,27,28. The general consensus of OA studies on Halimeda indicates negative to neutral calcification responses and neutral to positive photosynthetic responses to CO2-enriched seawater, due to species specificity22,24,26,27,28,29,30,31,32,33.

To date, two studies have considered the effects of seagrass-calcifying algae interactions under ambient conditions12,13, but none have addressed how OA and temperature rise influence these ecophysiological interactions. The species-specific nature of the isolated responses emphasizes the necessity to conduct studies that address OA and temperature rise together in order to better understand the mechanisms behind the presence/absence of interactions between these drivers. Short-term mesocosm experiments that simulate rapid heat waves and acidification, as observed in different regions, are fundamental tools to predict complex ecosystem interactions34,35. It is also necessary to introduce realism in these simulations by representing the high-frequency semidiurnal or diurnal variability that dominates coastal or shallow environments36. Recent studies reveal that under OA, net photosynthesis of the kelp Ecklonia radiata was almost 50% lower when pH fluctuated than when it was static37. This natural variability imposes particularities that can limit or stimulate primary production and must be reproduced in order to properly simulate the predictable future scenarios.

Here we investigate the effects of OA on the photosynthesis and calcification of the seagrass Halodule wrightii and the green alga Halimeda cuneata via a full factorial mesocosm design. We simulate OA and warming by exposing the calcifying alga and the seagrass to the following four combinations of ambient and elevated pCO2 and temperature: 28 °C & 320 µatm, 28 °C & 822 µatm, 30 °C & 320 µatm and 30 °C & 822 µatm. Most importantly, we examine the degree to which the photosynthetic carbon uptake of H. wrightii influences seawater chemistry under OA through short-term incubations. We aim to determine whether this can act as a metabolic feedback on the photosynthesis and calcification of H. cuneata, considering that these two species of macrophytes coexist in the shallow tropical waters off the Brazilian coast38. We hypothesize that H. wrightii is capable of using the excess DIC resulting from OA to increase its photosynthetic activity. In mitigating the effects of OA on seawater chemistry, we hypothesize that the negative effects of OA on the calcification rate of H. cuneata may consequently be ameliorated.

## Results

After exposure to treatments for 10 d, short-term incubations (illustrated in Supplementary Fig. S1) of H. cuneata and H. wrightii separately, as well as together, were conducted in order to understand their physiological responses to OA. We also sought to obtain the magnitude of the effect that the primary production of these macrophytes, particularly the seagrass, has on surrounding seawater chemistry under the stress of OA. The ultimate objective was to determine whether biologically altered seawater might be sufficient enough to mitigate the effects of OA on H. cuneata calcification.

H. cuneata and H. wrightii fared differently under OA conditions. While the calcifying alga experienced negative physiological consequences (Fig. 1), the seagrass showed a neutral response (Fig. 2, Table 1). We report a significant effect of CO2 enrichment on H. cuneata calcification (p = 0.017, Table 1), causing it to suffer a 72% decrease under elevated pCO2. Simultaneously, we observed a shift in carbonate chemistry when H. cuneata was incubated in the elevated pCO2 treatment, where HCO3 decreased by 56% and total DIC, by 60%, but carbonate and CO2 remained the same (Fig. 3, Table 2).

We did not detect a significant effect of CO2 on the gross primary production (GPP) of either H. cuneata (p = 0.370) or H. wrightii (p = 0.321) when incubated separately (Table 1), yet for H. cuneata there was an increasing trend (Fig. 2, Table 2). When H. cuneata and H. wrightii were incubated together, the resulting GPP also did not differ from ambient to elevated pCO2 (Table 2), but the overall values were clearly driven by H. wrightii production. The GPP of H. wrightii was about 53 times higher than that of H. cuneata at ambient conditions and 23 times higher under elevated CO2 (Table 2).

We observed significant changes in seawater chemistry (Table 3) resulting from the incubation of H. cuneata and H. wrightii separately and together. It’s worthy to note, however, that observed shifts that we attribute to an organism’s metabolism also reflect the natural chemical equilibrium change that occurs following biological inorganic carbon uptake. The challenge of teasing apart the biological effect and the equilibrium change warrants relative interpretation of DIC uptake. Surprisingly, H. wrightii did not increase DIC uptake under OA, showing no significant differences in total DIC consumption between ambient and elevated CO2 (p = 0.356, Table 3). However, due to its comparably higher GPP, H. wrightii still took up 40 times more total DIC than H. cuneata under elevated CO2 (Table 2). Consequently, when incubated alone under elevated CO2, H. wrightii was able to metabolically increase seawater pH by 0.27 ± 0.05 units and aragonite saturation state (ΩAr) by 7.57 ± 1.11 units (Table 2). In contrast, under the same conditions, H. cuneata only increased seawater pH by 0.08 ± 0.04 and ΩAr by 0.08 ± 0.03 (Fig. 4, Table 2). When incubated with H. cuneata, seagrass presence was a significant factor in determining the evolution of HCO3, CO32− and total DIC as well as ΔΩAr and ΔpH in seawater (Table 3). We were not able to quantitatively separate the photosynthetic rates of the alga and the seagrass when they were incubated together. So in order to estimate the effect of the seagrass, we relied on the magnitude of metabolic change for which the seagrass was solely responsible when incubated alone, as mentioned previously.

The seagrass was able to mitigate OA and significantly impact the alga’s physiology. We report significant effects of seagrass presence on H. cuneata calcification (p = 0.038), but no interactive effect was found between CO2 treatment and the latter (p = 0.281, Table 1). When H. wrightii and H. cuneata were incubated together at ambient pCO2, there was interestingly no observed change in calcification. However, at elevated pCO2, the metabolic interaction between the two mitigated the negative impact of OA and the calcification rate of H. cuneata was reduced by only 34% (as opposed to 72% when alone; Fig. 1, Table 2).

During the experimental period, the ambient CO2 treatment had a mean pCO2 of 380 µatm ± 7 SEM and a mean pH of 8.197 ± 0.006 SEM and for the elevated CO2 treatment, a mean pCO2 of 822 µatm ± 16 SEM and a mean pH of 7.923 ± 0.007 SEM (Table 4). pH fluctuated throughout the experiment due to natural variation from the adjacent reef, as shown in Supplementary Fig. S2. Mean AT was 2278 ± 2 (ambient) and 2280 ± 7 (elevated; Table 4). There was no effect of CO2 treatments on mean seawater TA (p = 0.723). The ranges of these parameters as well as the remaining physicochemical characterisation of seawater are found in Table 4. Temperature showed no significant effects or interactions on all descriptors (Table 1). Supplementary Fig. S3 shows the average PAR values observed throughout the day during the experimental period.

## Discussion

The presence of the seagrass H. wrightii mitigated the negative effect of OA on the calcification of the alga H. cuneata. This is the first study to confirm that under elevated CO2 concentrations, seagrass is still capable of maintaining comparatively high photosynthetic rates, and in turn, creating seawater conditions that are conducive to the calcification of sympatric, and otherwise ill-fated, calcifying algae. On their own, H. cuneata and H. wrightii responded differently to OA. The alga suffered decreased calcification and both the alga and the seagrass showed no significant photosynthetic response. It has widely been shown that calcifying organisms respond negatively to OA, whereas fleshy plants respond neutrally or positively1,39. The difference in the magnitude of the metabolic effect that each organism had on the surrounding seawater was substantial, where the dominant effect of H. wrightii played to the calcifying alga’s advantage under OA. It is worthy to note that the range of temperature initially tested (28–30 °C) to simulate ocean warming need not be considered a frontrunner threat to H. cuneata and H. wrightii photosynthesis and calcification due to the absence of an observable effect of temperature on these processes.

One of the more interesting findings of our study is the apparent asymmetry between photosynthetic and calcification responses to OA in H. cuneata. The calcified macroalgae showed signs of physiological stress, seeing as we observed a substantial decrease (72%) in its calcification. On the other hand, the photosynthetic response was not significant due to variation within treatment, however the fact that there was a 54% increase is worthy of consideration. These two processes occur side by side at the cellular level. Calcification occurs in intercellular spaces (inter-utricle spaces, or IUSs), which are separated from bulk seawater by a layer of utricles where photosynthesis is concentrated40. DIC uptake during photosynthesis is found to increase the pH of IUSs and calcium carbonate precipitation is favoured. Conversely, calcification produces H+ and CO2, which balance the change in pH and CO2 concentration produced by photosynthesis40. Due to the close proximity of these processes, under ambient conditions, calcification is reported to be closely coupled to photosynthesis in the genus Halimeda41,42. Thus, when photosynthesis increases, calcification is expected to increase, and vice versa. However, when we elevated CO2 in this study, calcification was compromised despite an increasing trend in GPP (Table 2), which suggests that there may be a certain degree of independence between these processes.

The understanding of this uncoupling lies in the carbonate chemistry dynamics of the location where photosynthesis and calcification intersect, the IUSs. Ultrastructure data suggests that the structure and size of utricles and IUSs in Halimeda may help to explain the carbonate chemistry of the IUSs and thus, OA responses24,41. Peach et al. 2017 found an inverse relationship between diffusive pathway type and mineral content, where species with longer utricles and thinner pathways contained more aragonite than those with shorter utricles and wider pathways. Morphological parameters were not one of our response variables, but based on our results and previously established calcifying mechanisms for Halimeda40,43 and aquatic plants44, we suggest that the diffusive pathway of H. cuneata permits corrosive bulk seawater to replenish the IUSs at a much faster rate than it can be biologically-regulated, thus partially inhibiting calcification45. We also suggest that dissolution may be contributing to the low pH and DIC-rich environments in the IUSs. The degree to which the alga may be experiencing dissolution is also not evident, due to the difficulty of disentangling the effects of dissolution and decreased calcification in OA studies42,46. The alga may be using DIC directly from dissolution as substrate for photosynthesis, thus explaining why DIC from bulk seawater was in lesser demand, shown by the 60% decrease in DIC consumption. However, the GPP of the alga was not capable of ameliorating the imbalance in carbonate chemistry of IUSs enough to stimulate calcification, thus our data supports the hypothesis that photosynthesis and calcification become uncoupled under OA1. Nonetheless, we cannot be certain of the source of DIC for this increasing trend in photosynthesis, nor the reason that the alga was unable to further increase GPP. Further research on this species using microsensors would be essential to ascertain these unknown thresholds that explain the apparent disparity between photosynthetic and calcification responses to OA46.

We initially expected an increase in the GPP of H. wrightii, since additional CO2 substrate is expected to stimulate primary production in fleshy marine plants. However, our results show that there was no significant change in photosynthesis at elevated CO2. Recent meta-analyses report neutral to positive photosynthetic responses of seagrasses to elevated CO21,47, which maintains our results within the range of expected responses. In addition, a study that exposed tropical H. wrightii to reduced pH observed an increase of only 20% in its photosynthetic rate, followed by a prominent plateau that was attributed to a preference for HCO3 use17. In the same study, the absence of a change in photosynthesis in H. wrightii after the exposure to acetazolamide (AZ), an inhibitor of carbonate anhydrase (CA), a common enzyme that aids in conversion of HCO3 to CO2, indicated that this species has an alternate and more efficient mechanism for HCO3 use when compared to other seagrasses17. Our observations likely indicate that H. wrightii’s neutral response to OA is due to the efficiency of its mechanism of HCO3 use. Due to its much higher photosynthetic rate (464 ± 60 (SEM) µmol O2 gDW−1 h−1), when compared to H. cuneata (19.5 ± 2.7 (SEM) µmol O2 gDW−1 h−1) at elevated pCO2, H. wrightii removes 40 times more DIC from seawater, thus increasing the pH48, CO32− availability and aragonite/calcite saturation states49. Although the seagrass did not increase its GPP, its capacity to biologically alter its surrounding seawater chemistry was enough to influence the metabolism of coexisting H. cuneata. We did not quantify the density of the studied Halodule bed, which is a factor that is shown to affect the magnitude of Halimeda-seagrass interactions12,50. Our results show that Halodule populations are likely to withstand intermediate OA scenarios, yet local irradiance, temperature and nutrient conditions may very well play a determinant role in the magnitude of the metabolic interactions between seagrasses and sympatric calcifying macroalgae. Interspecific variations in seagrass photosynthesis due to diverse DIC assimilation mechanisms will also put some species at an advantage over others17,51. This was observed at volcanic CO2 vent sites, where seagrass community composition shifted according to seawater pH52. The extent to which populations are acclimated to elevated conditions may determine their long-term resilience.

Acidified seawater is ultimately unfavourable for H. cuneata calcification, however we demonstrate that high-performing primary producers such as H. wrightii are capable of providing significant refuge for these calcifying algae via biologically altering seawater chemistry. Previously, a 1.6-fold increase in the calcification rate of Halimeda renchii was observed in seagrass beds at ambient CO2 levels13. In our study, it was unexpected that seagrass presence did not also increase H. cuneata calcification at ambient CO2. Based on the high GPP of H. wrightii observed at ambient CO2, one would expect the consequent IUS carbonate chemistry to be exceptionally favourable and to stimulate calcification. Regardless, the issue is that future oceans will possess a much higher pCO2 than that of today’s oceans. Our results show that under OA, the presence of seagrass will likely foster calcification rates during the day that are comparable to those at current pCO2. Recent findings anticipate, however, that other factors of different functional scales will cause variation in this buffering capacity. There are often other marine macrophytes coexisting with Halimeda and seagrass, namely macroalgae. Benthic community composition is known to alter seawater chemistry at different magnitudes53,54, greatly due to species-specific irradiance optima and CCM mechanisms, therefore influencing the community’s OA buffering capacity. Modeled projections incorporating effects of OA and net community metabolism (NCM) on carbonate chemistry in seagrass meadows predict long-term offsets of CO2, but also NCM-driven extremes in carbonate chemistry under OA55. In particular, future pH levels at night are expected to be extremely low due to the intensified effect of OA on respiration55. This has implications for the net calcification of Halimeda that we weren’t capable of addressing and would need to be analysed in future studies. Additionally, Cyronak et al. (2018) reveal that the spatial and short temporal variation of carbonate chemistry in seagrass beds can be even greater than diel variability, thus potentially impacting the buffering capability of seagrasses across even smaller scales56. The fate of calcifying algae under OA may very well lie in the composition of the accompanying photoautotroph community as well as their associated NCM dynamics54.

Solid predictions of whether and which calcifying algae will adapt to OA & temperature rise are generally still insufficient57, partially due to the lack of incorporation of species interactions effects and natural seawater variability. Most of the available information produced until now has been based on stable values of pH and temperature37 and few global studies exist that address marine plant interactions alongside OA and temperature rise58. Despite the academic value of these efforts, their utilization in depicting future scenarios should be considered with caution, since natural variability of physical/chemical conditions is a selective pressure and a major driver of marine ecosystem functioning. Likewise, although there are known limitations to not manipulating CO2 directly into each experimental tank59, our design was chosen as the most feasible, which gave priority to the incorporation of diel pH and CO2 variability. Each tank was an isolated experimental unit with a certain degree of intrinsic variability and our results do not suggest that our design has biased the experimental outcome.

The degree of physiological tolerance or increased performance to changes in CO2 and temperature in the marine environment can be due to trans-generational plasticity, phenotypic buffering, or plasticity within generations (or ‘classical’ plasticity) from which ‘true’ evolutionary adaptation may arise6. Data supports that genetic variation in traits important for OA and temperature rise is prevalent in near-shore plants6. Based on our results, we suggest that the large natural variability of temperature and CO2 in shallow coastal environments has selected for phenotypic plasticity and co-evolutionary tools involving the metabolic interaction between H. cuneata and H. wrightii, thus potentially providing resilience and adaptability to OA. Plasticity in response to OA and temperature rise will help maintain population resilience under changing environments60. If Halimeda species have adequate genetic variability to generate phenotypes with different CO2 tolerances and optima, then it is likely that inter or intraspecific variability in fitness will be observed, where OA winners are likely to be those coexisting with seagrasses. The responses we observed are a contribution to the understanding of possible shifts in composition of relevant communities61, but they also highlight the relevance of coastal plant metabolic interactions as a dynamic biological factor that should be considered in the management of natural habitats, namely marine protected areas, in view of future climate scenarios.

## Methods

### Study area and experimental design

The experiment was performed in a large-scale, flow-through mesocosm designed by Projeto Coral Vivo, located at its research station on Araçaípe beach, Bahia, Brazil. The mesocosm was designed to test the effects of ocean acidification and warming (among other factors) on reef organisms62, while closely mimicking the adjacent reef conditions. Aracaípe beach’s (16° 29′ 28.6′′ S 39° 3′ 58.4′′ W) fringing reef develops 100 m off the coast of the Marine Mesocosm, where the seagrass H. wrightii and the upright calcifying green alga H. cuneata coexist. The open flow and proximity of the mesocosm to the fringing reef make this system highly realistic since it is able to maintain experimental conditions (seawater composition, temperature, diel pH and CO2 variability, turbidity, salinity, plankton density, photoperiod, rainfall, irradiance…etc) that are very similar to those in the adjacent reef.

With the interest of simulating moderate predictions of ocean acidification and warming, levels of pCO2 and temperature were chosen based on the IPCC RCP 6.0 scenario63 and modeling of future atmospheric emissions64,65. We initially established a full factorial design of ambient temperature and pCO2 and elevated temperature and pCO2, totaling four treatments. The ambient levels were the unaltered present temperature (28 °C) and pCO2 (380 µatm) of seawater from the adjacent reef, and the elevated CO2 and temperature treatments were achieved by manipulating seawater to target +2 °C (30 °C) and +0.25 pH, or +442 µatm (822 µatm). The pCO2 was calculated for each treatment using the TA and mean pH values via CO2SYS66.

Seawater from 500 m offshore was continuously pumped into four 5,000-L underground sumps where the four CO2 and temperature treatments were applied. pCO2 was manipulated in two sumps using a custom reactor system that introduced fine bubbles of CO2 into constantly mixed seawater. Similarly, a 1.9 m 15,000 W heater was placed in each of the two underground sumps where seawater temperature was to be elevated. Seawater pCO2 and temperature were not fixed. We used a custom-made Reef Angel Open-Source Controller, which elevated and regulated pCO2 and temperature levels with respect to ambient fluctuations. Treatments were applied to header sumps and not directly to experimental tanks based on feasibility and limitation of resources. Mixed treatment water was fed to four 310-L reservoir tanks, from which flow was regulated to 16, fully randomized 130-L raceway experimental tanks (n = 4 per treatment). Experimental tanks were continuously supplied with seawater at a flow rate of ~10 L•min−1, achieving a renewal rate of 5 x per hour. The experimental tank area was covered in shade cloth to uniformly reduce the intensity of natural sunlight by 70%, simulating the amount of incident solar radiation measured at approximately 2 m where organisms were collected on the reef. Duarte et al. (2015) reported a complete description of the mesocosm design and functioning.

### Sampling

Approximately 160 specimens of H. cuneata and 1,500 shoots of H. wrightii were collected at a depth of 2 m using SCUBA by carefully removing the entire holdfast and rhizome, respectively, and were brought to the holding aquariums of the Marine Mesocosm for sorting and removal of epibionts. Sediment from the first 10 cm of the sampling area was also collected and used as substrate for subsequent planting in the mesocosm. Ten H. cuneata thalli were placed upright in a plastic tray (40 × 17 × 4 cm), with the holdfasts anchored in 3 cm of sand. One tray was placed in each of the 16 experimental tanks. Approximately 50 seagrass shoots were replanted in each of 2 plastic trays with 3 cm of sediment in each experimental tank. Each of the 16 experimental tanks thus possessed 3 trays, 1 with H. cuneata and 2 with H. wrightii. Organisms were acclimated at ambient temperature and pCO2 for 15 days. Treatments commenced upon completion of the acclimation period, reaching target levels within 24 hours and were applied for a total of 10 days.

### Abiotic parameters

Salinity (Refractometer: Instrutherm RTS-101ATC), dissolved oxygen & temperature (Portable dissolved oxygen meter: Instrutherm MO-900), incident irradiance (Quantometer: apogee MQ-200) and pH (pHmeter: Gehaka ISO 9001) were measured daily in each experimental tank. Handheld pH meter and pH sensors were calibrated to NBS buffers daily and sensor drift was checked weekly using a bench top Gehaka pH meter. The remaining abiotic parameter meters were calibrated as per recommended in their factory manuals, using appropriate calibration solutions. The daily average photosynthetically active radiation (PAR) was monitored with light loggers (HOBO), which were positioned underwater at the level of the organisms in the experimental tanks. Nutrient concentrations were monitored every 3 days in each tank67. For the monitoring of seawater carbonate chemistry, water samples were retrieved from each experimental raceway tank (n = 3) and were immediately refrigerated. Total alkalinity measurements, were performed using a custom USB4000 spectrophotometer (Ocean Optics, Dunedin, USA) and compared to certified reference material (Scripps Institute of Oceanography, USA).

### Primary production and calcification

In order to isolate and assess the potential metabolic interactions between H. cuneata and H. wrightii, short-term (2.5 h) incubations were administered at the beginning and end of the experiment. Incubations were conducted on each species separately, as well as with the two species together (n = 3 per species/combination). Oxygen evolution and change in total alkalinity were measured; the former was used to calculate gross primary production (GPP) and the latter, for calcification rates and change in dissolved inorganic carbon (ΔDIC). GPP and ΔDIC were calculated for each species, whereas calcification was only calculated for the alga. An irradiance of 750 µmol quanta m−2 s−1 was used during the incubations since this was the average midday irradiance during the time period when the incubations were conducted.

Incubations did not take place in the experimental raceway tanks. An incubation setup was constructed outside of the experimental tanks using 28-L boxes, impermeable plastic bags as chambers and an illuminator. The illuminator structure was equipped with four metallic vapor lamps (220 V, REV426A4, Serwal) and positioned over four independent dark 28L boxes (22.5 × 35 × 50 cm, Marfinite). Each box was connected to an individual raceway experimental tank by a 12 mm (diam.) hose so that seawater was constantly renewed in the box. Thus, each box corresponded to a treatment (Supplementary Material Fig. S1). In each box, four transparent 29 × 39 cm incubation bags impermeable to dissolved oxygen served as incubation chambers and were filled with approximately 500 ml of seawater. Three of the bags received the following biological material: (1) only 1–2 H. cuneata thalli, (2) 1–2 H. cuneata thalli and 10 shoots of H. wrightii or (3) only 10 shoots of H. wrightii (illustrated in Supplementary Material Fig. S1). The fourth bag contained only seawater in order to monitor any background changes in oxygen concentration and total alkalinity (AT) due to microorganisms. Water samples were taken directly from treatment boxes for initial measurements of dissolved oxygen concentration (DO) and AT immediately prior to commencement of incubations. All visible air bubbles were removed from the bags before their sealing. At the end of the incubation period, water samples were taken from each bag for the final measurements of DO and AT. Water volume was measured and H. cuneata and H. wrightii were removed, dried at 60 °C and weighed. Basal segments were removed from each sample, decalcified with nitric acid (0.6M HNO3) and weighed to determine the dry decalcified weight.

### Oxygen evolution

Five initial water samples (12 ml) were collected from each treatment box (n = 3) at the beginning of the incubation. At the end of the incubations, five water samples were taken from each bag (n = 3) with a 60 ml syringe fitted with a small tube. Samples were immediately treated with manganese chloride and alkaline-iodide reagents upon removal and refrigerated for 72 hours. They were then treated with a sulfuric acid reagent and analysed spectrophotometrically according to the Winkler method adapted by Labasque68 in order to calculate the dissolved oxygen concentration, or O2 production.

GPP values were calculated by normalizing O2 production to incubation time, volume of water and the decalcified dry weight of the incubated tissue (µmol O2 gDW−1 h−1), after removing background O2 fluctuations due to microbial activity.

### Calcification

One initial water sample (180 ml) was collected from each treatment box pre-incubation and one final water sample was taken from each bag post-incubation (n = 3). Water samples were immediately refrigerated until analysis. Alkalinity anomaly measurements were performed using the aforementioned Ocean Optics equipment. The CO2SYS program66 was used to calculate all DIC species and components of seawater carbonate chemistry. Changes in each DIC species (ΔHCO3, ΔCO2 and ΔCO32−) and total DIC (ΔDIC) were calculated by subtracting the pre-incubation value from the post-incubation value.

Calcification rates were calculated for H. cuneata using the following equation69:

$$g=-0.5\,\frac{{\rm{\Delta }}{A}_{T}V}{DW{\rm{\Delta }}t}$$
(1)

where g = µmol CaCO3 g−1 h−1, ΔAT = change in total alkalinity, V = volume of incubated seawater, DW = dry, decalcified weight of H. cuneata and Δt = incubation time (h).

### Statistical analysis

A three-way analysis of variance (ANOVA), was performed on the H. cuneata calcification and GPP data (log(x + 1) transformed), with the factors CO2 (two levels), temperature (two levels), and seagrass presence (two levels). A two-way ANOVA was used to test H. wrightii GPP data (log(x + 1) transformed), with the factors CO2 and temperature (two levels each). Significance level was set at p = 0.05. Due to the absence of any temperature effect and a strong trend in the data with respect to CO2, we proceeded to pool the samples from the same CO2 treatment for a more robust analysis and in order to preserve important ecological implications. All data passed assumptions of normality of residuals and homogeneity of variances. Subsequently, two-way ANOVAs were applied to H. cuneata calcification, GPP data (log(x + 1) transformed), ΔHCO3, ΔCO2, ΔCO32−, ΔΩAr, ΔpH and Δtotal DIC data (log(x + 1) transformed), with the factors CO2 (two levels) and seagrass presence (two levels). One-way ANOVAs were applied to H. wrightii GPP data (log(x + 1) transformed), ΔHCO3, ΔCO2, ΔCO32− and Δtotal DIC data (log(x + 1) transformed), with CO2 (two levels) as a factor. Fisher’s LSD post hoc tests were used for pairwise comparisons of significant effects. All statistical analyses were performed using IBM SPSS Statistics 24.

## Data Availability

The datasets generated during and/or analysed during the current study are available from the corresponding author upon reasonable request.

## References

1. 1.

Koch, M., Bowes, G., Ross, C. & Zhang, X.-H. Climate change and ocean acidification effects on seagrasses and marine macroalgae. Global Change Biol 19, 103–132, https://doi.org/10.1111/j.1365-2486.2012.02791.x (2013).

2. 2.

Noriega, C. E. D., Araujo, M. & Lefèvre, N. Spatial and temporal variability of the CO2 fluxes in a tropical, highly urbanized estuary. Estuaries Coast. 36, 1054–1072, https://doi.org/10.1007/s12237-013-9608-1 (2013).

3. 3.

Gattuso, J.-P., Frankignoulle, M. & Wollast, R. Carbon and carbonate metabolism in coastal aquatic ecosystems. Annu. Rev. Ecol. Evol. Syst. 29, 405–434, https://doi.org/10.1146/annurev.ecolsys.29.1.405 (1998).

4. 4.

Wollast, R. Evaluation and comparison of the global carbon cycle in the coastal zone and in the open ocean. The sea 10, 213–252 (1998).

5. 5.

Harley, C. D. G. et al. The impacts of climate change in coastal marine systems. Ecol Lett 9, 228–241, https://doi.org/10.1111/j.1461-0248.2005.00871.x (2006).

6. 6.

Reusch, T. B. H. Climate change in the oceans: evolutionary versus phenotypically plastic responses of marine animals and plants. Evol Appl. 7, 104–122, https://doi.org/10.1111/eva.12109 (2014).

7. 7.

Thompson, D. W. J., Barnes, E. A., Deser, C., Foust, W. E. & Phillips, A. S. Quantifying the role of internal climate variability in future climate trends. J. Clim 28, 6443–6456, https://doi.org/10.1175/jcli-d-14-00830.1 (2015).

8. 8.

Valladares, F. et al. The effects of phenotypic plasticity and local adaptation on forecasts of species range shifts under climate change. Ecol Lett 17, 1351–1364, https://doi.org/10.1111/ele.12348 (2014).

9. 9.

Hay, M. E. Marine chemical ecology: chemical signals and cues structure marine populations, communities, and ecosystems. Ann Rev Mar Sci. 1, 193–212, https://doi.org/10.1146/annurev.marine.010908.163708 (2009).

10. 10.

Buosi, A. & Sfriso, A. Macrophyte assemblage composition as a simple tool to assess global change in coastal areas. Freshwater impacts and climatic changes. Sci Total Environ 605–606, 559–568, https://doi.org/10.1016/j.scitotenv.2017.06.196 (2017).

11. 11.

Kroeker, K. J., Micheli, F. & Gambi, M. C. Ocean acidification causes ecosystem shifts via altered competitive interactions. Nat. Clim. Change 3, 156, https://doi.org/10.1038/nclimate1680 (2012).

12. 12.

Barry, S. C., Frazer, T. K. & Jacoby, C. A. Production and carbonate dynamics of Halimeda incrassata (Ellis) Lamouroux altered by Thalassia testudinum Banks and Soland ex König. J Exp Mar Bio Ecol. 444, 73–80, https://doi.org/10.1016/j.jembe.2013.03.012. (2013).

13. 13.

Semesi, I. S., Beer, S. & Björk, M. Seagrass photosynthesis controls rates of calcification and photosynthesis of calcareous macroalgae in a tropical seagrass meadow. Mar Ecol Prog Ser. 382, 41–47 (2009).

14. 14.

Richard, K. F. U., Catherine, J. C., Gideon, M. H. & Len, J. M. Tropical seagrass meadows modify seawater carbon chemistry: implications for coral reefs impacted by ocean acidification. Environ Res Lett. 7, 024026 (2012).

15. 15.

Hendriks, I. E. et al. Photosynthetic activity buffers ocean acidification in seagrass meadows. Biogeosciences 11, 333, https://doi.org/10.5194/bg-11-333-2014 (2014).

16. 16.

Yang, Y., Hansson, L. & Gattuso, J.-P. Data compilation on the biological response to ocean acidification: an update. Earth Syst. Sci. Data Discuss. 8, 889–912 (2015).

17. 17.

Campbell, J. E. & Fourqurean, J. W. Mechanisms of bicarbonate use influence the photosynthetic carbon dioxide sensitivity of tropical seagrasses. Limnol Oceanogr 58, 839–848, https://doi.org/10.4319/lo.2013.58.3.0839 (2013).

18. 18.

Palacios, S. L. & Zimmerman, R. C. Response of eelgrass Zostera marina to CO2 enrichment: possible impacts of climate change and potential for remediation of coastal habitats. Mar Ecol Prog Ser. 344, 1–13 (2007).

19. 19.

Ow, Y. X., Collier, C. J. & Uthicke, S. Responses of three tropical seagrass species to CO2 enrichment. Mar Biol 162, 1005–1017, https://doi.org/10.1007/s00227-015-2644-6 (2015).

20. 20.

Schneider, G. et al. Structural and physiological responses of Halodule wrightii to ocean acidification. Protoplasma 255, 629–641, https://doi.org/10.1007/s00709-017-1176-y (2018).

21. 21.

Campbell, J. E., Fisch, J., Langdon, C. & Paul, V. J. Increased temperature mitigates the effects of ocean acidification in calcified green algae (Halimeda spp.). Coral Reefs 35, 357–368, https://doi.org/10.1007/s00338-015-1377-9 (2016).

22. 22.

Campbell, J. E., Craft, J. D., Muehllehner, N., Langdon, C. & Paul, V. J. Responses of calcifying algae (Halimeda spp.) to ocean acidification: implications for herbivores. Mar Ecol Prog Ser. 514, 43–56 (2014).

23. 23.

Meyer, F. W., Vogel, N., Teichberg, M., Uthicke, S. & Wild, C. The physiological response of two green calcifying algae from the Great Barrier Reef towards high dissolved inorganic and organic carbon (DIC and DOC) availability. PLoS One 10, e0133596 (2015).

24. 24.

Price, N. N., Hamilton, S. L., Tootell, J. S. & Smith, J. E. Species-specific consequences of ocean acidification for the calcareous tropical green algae Halimeda. Mar Ecol Prog Ser. 440, 67–78 (2011).

25. 25.

Robbins, L., Knorr, P. & Hallock, P. Response of Halimeda to ocean acidification: field and laboratory evidence. Biogeosci. Discuss. 6, 4895–4918 (2009).

26. 26.

Vogel, N. et al. Calcareous green alga Halimeda tolerates ocean acidification conditions at tropical carbon dioxide seeps. Limnol Oceanogr 60, 263–275, https://doi.org/10.1002/lno.10021 (2015).

27. 27.

Vogel, N., Meyer, F. W., Wild, C. & Uthicke, S. Decreased light availability can amplify negative impacts of ocean acidification on calcifying coral reef organisms. Mar Ecol Prog Ser. 521, 49–61 (2015).

28. 28.

Wizemann, A., Meyer, F. W., Hofmann, L. C., Wild, C. & Westphal, H. Ocean acidification alters the calcareous microstructure of the green macro-alga Halimeda opuntia. Coral Reefs 34, 941–954, https://doi.org/10.1007/s00338-015-1288-9 (2015).

29. 29.

Comeau, S., Edmunds, P. J., Spindel, N. B. & Carpenter, R. C. The responses of eight coral reef calcifiers to increasing partial pressure of CO2 do not exhibit a tipping point. Limnol Oceanogr 58, 388–398, https://doi.org/10.4319/lo.2013.58.1.0388 (2013).

30. 30.

Johnson, M. D., Price, N. N. & Smith, J. E. Contrasting effects of ocean acidification on tropical fleshy and calcareous algae. PeerJ 2, e411 (2014).

31. 31.

Sinutok, S., Hill, R., Doblin, M. A., Wuhrer, R. & Ralph, P. J. Warmer more acidic conditions cause decreased productivity and calcification in subtropical coral reef sediment-dwelling calcifiers. Limnol Oceanogr 56, 1200–1212, https://doi.org/10.4319/lo.2011.56.4.1200 (2011).

32. 32.

Sinutok, S., Hill, R., Doblin, M. A., Kühl, M. & Ralph, P. J. Microenvironmental changes support evidence of photosynthesis and calcification inhibition in Halimeda under ocean acidification and warming. Coral Reefs 31, 1201–1213, https://doi.org/10.1007/s00338-012-0952-6 (2012).

33. 33.

Meyer, F. W. et al. Effect of inorganic and organic carbon enrichments (DIC and DOC) on the photosynthesis and calcification rates of two calcifying green algae from a Caribbean reef lagoon. PLoS One 11, e0160268, https://doi.org/10.1371/journal.pone.0160268 (2016).

34. 34.

Gruber, N. et al. Rapid progression of ocean acidification in the California current system. Science 337, 220 (2012).

35. 35.

Gouvêa, L. P. et al. Interactive effects of marine heatwaves and eutrophication on the ecophysiology of a widespread and ecologically important macroalga. Limnol Oceanogr 62, 2056–2075, https://doi.org/10.1002/lno.10551 (2017).

36. 36.

Hofmann, G. E. et al. High-frequency dynamics of ocean pH: a multi-ecosystem comparison. PLoS One 6, e28983, https://doi.org/10.1371/journal.pone.0028983 (2011).

37. 37.

Britton, D., Cornwall, C. E., Revill, A. T., Hurd, C. L. & Johnson, C. R. Ocean acidification reverses the positive effects of seawater pH fluctuations on growth and photosynthesis of the habitat-forming kelp, Ecklonia radiata. Sci Rep. 6, 26036, https://doi.org/10.1038/srep26036 (2016).

38. 38.

Bandeira-Pedrosa, M. E., Pereira, S. M. B. & Oliveira, E. C. Taxonomy and distribution of the green algal genus Halimeda (Bryopsidales, Chlorophyta) in Brazil. Braz J Bot. 27, 363–377 (2004).

39. 39.

Kroeker, K. J. et al. Impacts of ocean acidification on marine organisms: quantifying sensitivities and interaction with warming. Global Change Biol 19, 1884–1896, https://doi.org/10.1111/gcb.12179 (2013).

40. 40.

Borowitzka, M. A. & Larkum, A. W. D. Calcification in the green alga Halimeda III. The sources of inorganic carbon for photosynthesis and calcification and a model of the mechanism of calcification. J Exp Bot 27, 879–893, https://doi.org/10.1093/jxb/27.5.879 (1976).

41. 41.

Peach, K. E., Koch, M. S., Blackwelder, P. L. & Manfrino, C. Calcification and photophysiology responses to elevated pCO2 in six Halimeda species from contrasting irradiance environments on Little Cayman Island reefs. J Exp Mar Bio Ecol. 486, 114–126, https://doi.org/10.1016/j.jembe.2016.09.008 (2017).

42. 42.

Cyronak, T., Schulz, K. G. & Jokiel, P. L. The Omega myth: what really drives lower calcification rates in an acidifying ocean. ICES J Mar Sci. 73, 558–562, https://doi.org/10.1093/icesjms/fsv075 (2016).

43. 43.

Borowitzka, M. A. & Larkum, A. W. D. Calcification in algae: mechanisms and the role of metabolism. CRC Crit Rev Plant Sci. 6, 1–45, https://doi.org/10.1080/07352688709382246 (1987).

44. 44.

Borowitzka, M. A. Calcification in aquatic plants. Plant Cell Environ 7, 457–466 (1984).

45. 45.

Peach, K. E., Koch, M. S., Blackwelder, P. L., Guerrero-Given, D. & Kamasawa, N. Primary utricle structure of six Halimeda species and potential relevance for ocean acidification tolerance. Bot Mar 60, 1, https://doi.org/10.1515/bot-2016-0055 (2017).

46. 46.

Ries, J. B. A physicochemical framework for interpreting the biological calcification response to CO2-induced ocean acidification. Geochim Cosmochim Ac. 75, 4053–4064, https://doi.org/10.1016/j.gca.2011.04.025 (2011).

47. 47.

Kroeker, K. J., Kordas, R. L., Crim, R. N. & Singh, G. G. Meta-analysis reveals negative yet variable effects of ocean acidification on marine organisms. Ecol Lett 13, 1419–1434, https://doi.org/10.1111/j.1461-0248.2010.01518.x (2010).

48. 48.

Invers, O., Zimmerman, R. C., Alberte, R. S., Pérez, M. & Romero, J. Inorganic carbon sources for seagrass photosynthesis: an experimental evaluation of bicarbonate use in species inhabiting temperate waters. J Exp Mar Bio Ecol. 265, 203–217, https://doi.org/10.1016/S0022-0981(01)00332-X (2001).

49. 49.

Bach, L. T. Reconsidering the role of carbonate ion concentration in calcification by marine organisms. Biogeosciences 12, 4939 (2015).

50. 50.

Davis, B. C. & Fourqurean, J. W. Competition between the tropical alga, Halimeda incrassata, and the seagrass, Thalassia testudinum. Aquat. Bot. 71, 217–232, https://doi.org/10.1016/S0304-3770(01)00179-6 (2001).

51. 51.

Beer, S., Axelson, L. & Björk, M. Modes of photosynthetic bicarbonate utilisation in seagrasses, and their possible roles in adaptations to specific habitats. Biol. Mar. Medit. 13, 3–7 (2006).

52. 52.

Takahashi, M., Noonan, S. H. C., Fabricius, K. E. & Collier, C. J. The effects of long-term in situ CO2 enrichment on tropical seagrass communities at volcanic vents. ICES J Mar Sci. 73, 876–886, https://doi.org/10.1093/icesjms/fsv157 (2016).

53. 53.

Page, H. N. et al. Differential modification of seawater carbonate chemistry by major coral reef benthic communities. Coral Reefs 35, 1311–1325, https://doi.org/10.1007/s00338-016-1490-4 (2016).

54. 54.

Takeshita, Y., Cyronak, T., Martz, T. R., Kindeberg, T. & Andersson, A. J. Coral Reef Carbonate Chemistry Variability at Different Functional Scales. Front Mar Sci 5, https://doi.org/10.3389/fmars.2018.00175 (2018).

55. 55.

Pacella, S. R., Brown, C. A., Waldbusser, G. G., Labiosa, R. G. & Hales, B. Seagrass habitat metabolism increases short-term extremes and long-term offset of CO2 under future ocean acidification. PNAS, https://doi.org/10.1073/pnas.1703445115 (2018).

56. 56.

Cyronak, T. et al. Short-Term Spatial and Temporal Carbonate Chemistry Variability in Two Contrasting Seagrass Meadows: Implications for pH Buffering Capacities. Estuaries Coast. 41, 1282–1296, https://doi.org/10.1007/s12237-017-0356-5 (2018).

57. 57.

Hofmann, L. C. & Bischof, K. Ocean acidification effects on calcifying macroalgae. Aquatic Biol. 22, 261–279 (2014).

58. 58.

Tait, L. W. Impacts of natural and manipulated variations in temperature, pH and light on photosynthetic parameters of coralline–kelp assemblages. J Exp Mar Bio Ecol. 454, 1–8, https://doi.org/10.1016/j.jembe.2014.01.016 (2014).

59. 59.

Cornwall, C. E. & Hurd, C. L. Experimental design in ocean acidification research: problems and solutions. ICES J Mar Sci. 73, 572–581 (2015).

60. 60.

Chevin, L.-M., Lande, R. & Mace, G. M. Adaptation, plasticity, and extinction in a changing environment: towards a predictive theory. PLoS Biol. 8, e1000357, https://doi.org/10.1371/journal.pbio.1000357 (2010).

61. 61.

Kangwe, J., Semesi, I. S., Beer, S., Mtolera, M. & Björk, M. Carbonate production by calcareous algae in a seagrass-dominated system: the example of Chwaka Bay. Chapter 8 in People, Nature and Research in Chwaka Bay, Zanzibar, Tanzania. (WIOMSA, 2012).

62. 62.

Duarte, G. et al. A novel marine mesocosm facility to study global warming, water quality, and ocean acidification. Ecol. Evol. 5, 4555–4566, https://doi.org/10.1002/ece3.1670 (2015).

63. 63.

Pachauri, R. K. et al. Climate change 2014: synthesis report. Contribution of Working Groups I, II and III to the fifth assessment report of the Intergovernmental Panel on Climate Change. (IPCC, 2014).

64. 64.

Caldeira, K. & Wickett, M. E. Oceanography: anthropogenic carbon and ocean pH. Nature 425, 365, https://doi.org/10.1038/425365a (2003).

65. 65.

Caldeira, K. & Wickett, M. E. Ocean model predictions of chemistry changes from carbon dioxide emissions to the atmosphere and ocean. J. Geophys. Res. Oceans 110, https://doi.org/10.1029/2004JC002671 (2005).

66. 66.

Pierrot, D., Lewis, E. & Wallace, D. MS Excel program developed for CO2 system calculations. Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, US Department of Energy. ORNL/CDIAC-IOS (2006).

67. 67.

Grasshoff, K., Kremling, K. & Ehrhardt, M. Methods of seawater analysis. (John Wiley & Sons, 1983).

68. 68.

Labasque, T., Chaumery, C., Aminot, A. & Kergoat, G. Spectrophotometric Winkler determination of dissolved oxygen: re-examination of critical factors and reliability. Mar Chem. 88, 53–60, https://doi.org/10.1016/j.marchem.2004.03.004 (2004).

69. 69.

Chisholm, J. R. M. & Gattuso, J.-P. Validation of the alkalinity anomaly technique for investigating calcification of photosynthesis in coral reef communities. Limnology and Oceanography 36(6), 1232–1239 (1991).

## Acknowledgements

E.B. would like to thank the Coordenação de Aperfeiçoamento de Pessoas de Nível Superior (CAPES) for Masters funding. Funding for this project came from the Synergism grant (CNPq 407365/2013-3). We extend our thanks to the Brazil-based Projeto Coral Vivo and its sponsor PetroBras Ambiental for providing the Marine Mesocosm structure and experimental assistance.

## Author information

E.B. contributed to all steps of the experimental design, experiment, data analysis and manuscript preparation. J.S. contributed to the experimental design, data analysis and manuscript preparation. C.M. contributed to the experimental design and experiment. P.H. contributed to experimental design, experiment, data analysis and manuscript preparation.

Correspondence to Ellie Bergstrom or Cíntia Martins.

## Ethics declarations

### Competing Interests

The authors declare no competing interests.

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

## Rights and permissions

Reprints and Permissions

Bergstrom, E., Silva, J., Martins, C. et al. Seagrass can mitigate negative ocean acidification effects on calcifying algae. Sci Rep 9, 1932 (2019) doi:10.1038/s41598-018-35670-3

• ### The biogeomorphology of Shark Bay's microbialite coasts

• Thérèse E. Morris
• , Pieter T. Visscher
• , Micheal J. O'Leary
• , Peter R.C.S. Fearns
•  & Lindsay B. Collins

Earth-Science Reviews (2019)

• ### Ecotoxicological responses of a reef calcifier exposed to copper, acidification and warming: A multiple biomarker approach

• Joseane A. Marques
• , Douglas P. Abrantes
• , Laura FB. Marangoni