Ocean acidification is a pervasive threat to coral reef ecosystems, and our understanding of the ecological processes driving patterns in tropical benthic community development in conditions of acidification is limited. We deployed limestone recruitment tiles in low aragonite saturation (Ωarag) waters during an in-situ field experiment at Puerto Morelos, Mexico, and compared them to tiles placed in control zones over a 14-month investigation. The early stages of succession showed relatively little difference in coverage of calcifying organisms between the low Ωarag and control zones. However, after 14 months of development, tiles from the low Ωarag zones had up to 70% less cover of calcifying organisms coincident with 42% more fleshy algae than the controls. The percent cover of biofilm and turf algae was also significantly greater in the low Ωarag zones, while the number of key grazing taxa remained constant. We hypothesize that fleshy algae have a competitive edge over the primary calcified space holders, coralline algae, and that acidification leads to altered competitive dynamics between various taxa. We suggest that as acidification impacts reefs in the future, there will be a shift in community assemblages away from upright and crustose coralline algae toward more fleshy algae and turf, established in the early stages of succession.
Citation: Crook ED, Kroeker KJ, Potts DC, Rebolledo-Vieyra M, Hernandez-Terrones LM, Paytan A (2016) Recruitment and Succession in a Tropical Benthic Community in Response to In-Situ Ocean Acidification. PLoS ONE 11(1): e0146707. https://doi.org/10.1371/journal.pone.0146707
Editor: Mónica Medina, Pennsylvania State University, UNITED STATES
Received: April 24, 2015; Accepted: December 21, 2015; Published: January 19, 2016
Copyright: © 2016 Crook 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.
Data Availability: All data files are available from the BCO-DMO database at the following URL: (http://www.bco-dmo.org/dataset/564766).
Funding: This work was supported by the National Science Foundation-1040952 (AP) (nsf.gov) and the Environmental Protection Agency-Science to Achieve Results (EDC) (epa.gov). 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.
Declining surface ocean pH (ocean acidification) is a global environmental issue likely to be deleterious for a wide range of marine organisms [1–3]. Coral reef systems are expected to be particularly susceptible to ocean acidification and may see significant declines in calcification over the 21st century due to declining aragonite saturation state (Ωarag) [4–6]. Laboratory studies have described responses of many individual species to acidification [7–9], but ecosystem responses to acidification are complex [10,11]. Field studies are essential for understanding how complex assemblages of species may respond to declining pH [12–19].
Ocean acidification is predicted to directly impact calcifying organisms by reducing calcification rates. However, a key question for ecosystems is how acidification may impact communities by altering competitive interactions between organisms, resulting in phase shifts [16, 20–22]. Recently, Kroeker et al. at (2012) found that calcareous species were rapidly outcompeted by fleshy algae in acidified conditions; that is, competitive interactions between fleshy algae and calcifying species alter community structure in reduced pH environments in the temperate Mediterranean Sea . Growing evidence suggests that non-calcareous algae appear to benefit in low pH conditions [16, 23–24], while calcifying organisms are either directly impacted (i.e. reduced calcification rates) or outcompeted by fleshy algal species [14– 16]. However, most studies have been observation-based, and our understanding of the processes driving these patterns in community development is limited. Here, we conducted a field experiment to investigate recruitment and early succession near a tropical coral reef to study how competitive interactions drive ecosystem responses to acidification in a tropical community. We investigate how acidification affects dominance of space among various taxa between sites that vary naturally in carbonate chemistry on the Mesoamerican Barrier Reef. Our study design is similar to that of Kroeker et al., (2012), allowing for comparison between the responses in temperate and tropical systems. We focused on interactions among organisms under lower than ambient Ωarag conditions to determine whether, and how, reduced Ωarag may affect this community.
The Mesoamerican Barrier Reef lies off the east coast of the Yucatan Peninsula. Rainwater rapidly infiltrates the porous karstic limestone of Quintana Roo, and then flows towards the ocean through interconnected caves and fractures. Along the flow path, the groundwater mixes extensively with seawater in underground aquifers before discharging into the lagoon between the shore and the offshore reefs at localized submarine springs (known locally as “ojos”) . Although these submarine groundwater springs have near-oceanic salinities and temperatures  the water has high dissolved inorganic carbon (CT), high total alkalinity (AT), low pH and low Ωarag . The ojos are typically at 4–7 m depth, and the chemistry of the water affects the diversity, abundance, and calcification rates of corals that settle and grow at the springs [18,27].
Understanding the ecological processes leading to the observed differences in diversity and abundance of organisms along these naturally varying pH-saturation conditions is a critical step for predicting future impacts of acidification on reef environments. We deployed limestone recruitment tiles in low pH-Ωarag waters at the ojo centers and compared them to those concurrently placed in control zones of ambient pH-saturation within a few meters of the springs. A subset of tiles was collected on three occasions (3 months, 6 months and 14 months) for analysis of recruitment and community succession. Although the average saturation state at the ojo centers (Ωarag = 1.5) is much lower than most predictions for the late 21st century, this study assesses potential impacts of ocean acidification on developing reef communities that may be particularly relevant if atmospheric CO2 follows more extreme IPCC scenarios  or if local conditions (e.g. river or groundwater inputs, upwelling) exacerbate global acidification.
Materials and Methods
The experiment took place at two ojos (Sites A and B), approximately 500 m offshore near Puerto Morelos Reef Natural Park, Quintana Roo, Mexico (20.853° N, 86.898° W). The ojos were chosen based on previous monitoring, which suggested their water had low saturation (Ojo A Ωarag = 1.4 ± 0.4, Ojo B Ωarag = 1.6 ± 0.4) for much of the time each year and relatively high salinities (> 30) in the immediate vicinity of the discharge (Table 1) [26, 27]. We used a 2 x 2 factorial design (2 Ωarag levels x 2 ojo sites), which allowed us to compare the response as a function of chemical changes (e,g. different chemistry regimes with low pH-Ωarag and ambient pH-Ωarag) and location (site A and site B). If aragonite saturation state or pH is an important controlling factor we expect little difference between sites at similar pH-Ωarag and larger differences regardless of site for different pH-Ωarag. To mimic the natural karst substrate, we deployed 40 limestone tiles (15 x 15 cm), acquired from a quarry near Puerto Morelos. Twenty tiles were deployed at each site; 10 in a low saturation zone (Ωarag ~ 1.5, hereafter referred to as “ojo”) in the direct vicinity of the spring discharge, and 10 in an ambient zone (Ωarag ~ 3.8, hereafter referred to as “control”) about 5 to 10 m from the area of influence of the spring discharge. The tiles were bolted to concrete masonry blocks with stainless steel screws through a hole drilled in the center of each tile (Fig 1). We deployed the tiles on 28 August 2010, immediately preceding a coral mass-spawning event. We removed subsets of three randomly selected tiles from each treatment after 3 months (25 November 2010) and 6 months (14 March 2011), and removed the four remaining tiles after 14 months (19 October 2011). Upon removal, the tiles were photographed, fixed in a 4% formalin solution for 48 hours, and then stored in 70% ethanol until analyzed (Fig 2).
Ojo A (A) and Ojo B (B) during recruitment substrate deployment (time zero). The low pH-Ωarag zones (x) are within 10m of ambient zones (y) at each site.
Examples of tiles collected at 3 months (A, B), 6 months (C,D), and 14 months (E,F), at ojo centers (A,C,E) and controls (B,D,F).
We collected discrete water samples for dissolved inorganic carbon (CT), total alkalinity (AT), salinity, and nutrients at initial deployment, during each recovery, and at monthly intervals throughout the 14-month deployment. Additionally, in situ monitoring was conducted intermittently over a 4 year period before, during, and after deployment in an attempt to capture the variability in chemistry at the ojo sites over shorter (e.g., diurnal) and longer (e.g., interannual) time frames. These include a 24 hour monitoring at Ojo A. Values reported here (Table 1) include discrete samples taken over a several year period to give a greater indicator of conditions at the site. CT was measured using a CM5011 Carbon Coulometer (UIC, Inc.) and AT was measured with an automated, open-cell potentiometric titration procedure. Certified Reference Materials (batch 118) from the laboratory of Dr. Andrew Dickson at Scripps Institution of Oceanography were used to calibrate each instrument. CT and AT were used to calculate aragonite saturation state (Ωarag) and pH via CO2sys software , using CO2 dissociation constants from Merhbach et al. (1973) refitted by Dickson and Millero (1987) [30,31]. pH is reported in total scale (pHT). Salinity was measured with a salinometer (Guildline 8410 PortaSal), and nutrients were analyzed on a flow injection autoanalyzer (FIA, Lachat Instruments Model QuickChem 8000). In addition to the discrete samples, pH, temperature, and salinity at the ojos were monitored using a SeapHOx sensor, which suggest that the discharge at the spring was continuous throughout the experiment (S1 Fig).
The main focus of this study was to determine how acidification may impact community level changes, and specifically, to determine functional differences between the communities inside and outside of the springs. We therefore focus on functional groups rather than conducting species level analyses. This approach is consistent with previous investigations  and thus allows comparison between the Caribbean and Mediterranean sites. Organisms on the tiles were assigned to eleven functional groups (Fig 3). The tiles were divided into 1.5 x 3 cm subplots on the edges and 3 x 3 cm subplots on the face of the tiles for visual estimates of percent cover . Subplot estimates were then summed for total percent cover. The cover of erect fleshy algae forming a canopy over the tile was analyzed first and then removed to estimate the percent cover of encrusting groups. Encrusting foraminifera, molluscs, and polychaetes were counted and measured using a Celestron digital microscope (0.1 mm accuracy).
Percent cover can be greater than 100% due to the multiple layers of organisms present on the tiles. Foraminifera, molluscs, and polychaetes are not represented due to their minor contribution to overall % cover.
Community composition, defined as the presence or absence of functional groups, was compared between chemistry regime (ojo or control zones) and sites (Site A or B) for the 14-month tiles on a Bray-Curtis (BC) similarity matrix of presence/absence of functional groups. Community structure, defined as the relative abundance of functional groups, was analyzed on a zero-adjusted BC similarity matrix of square-root transformed total percent cover of functional groups. Permutational Multivariate Analysis of Variance (PERMANOVAs) were used to test variation in community composition and structure, with site and chemistry regime as categorical, fixed factors using 9,999 unrestricted permutations of the transformed data and Type III SS. Non-metric multi-dimensional scaling (nMDS) plots were made to visualize the variability in community structure. NMDS plots are ordinations of the multivariate data, in this case community data, where each point on the ordination represents the community on a single tile. In an nMDS plot, the multivariate data are placed into two-dimensional space so that the rank differences among the data are preserved, based on the BC dissimilarity matrix. Thus communities that are more dissimilar to one another are farther apart on the plot.
Variation in succession was tested on the community structure measurements among site x saturation x time using PERMANOVA with site (Site A or B), chemistry regime (ojo or control zone), and time (3, 6 or 14 months) as fixed factors. In addition, we tested for differences in the percent cover of select fauna. Due to numerous zero values that violated the assumptions of parametric statistics, we used permutation-based analysis of variance to test univariate variables (α = 0.05).
All work was conducted at the Puerto Morelos Reef Natural Park. Samples were collected under Secretaría de Agricultura, Ganadería, Desarrollo Rural, Pesca y Alimentación (SAGARPA) permit DGOPA.00153.170111.-0051 and were exported with a Convention on International Trade in Endangered Species (CITES) Permit MX52912.
Eleven functional groups were common on the tiles (Figs 2 and 3). Five groups were comprised of calcareous organisms: including erect and crustose coralline algae (CCA), vermetid molluscs, tubicolous polychaetes, and encrusting foraminifera. Six groups were comprised of non-calcareous forms: including red, green and brown erect fleshy algae, turf algae, encrusting fleshy algae, and bacterial biofilm. Because only a total of 4 individual corals settled (Siderastrea radians) on tiles, only at the control sites, they were excluded from the analyses. The tiles from the ojos were generally dominated by fleshy algae, turf, and biofilm, while those in control conditions were dominated by CCA (Fig 3). Erect coralline algae were often entirely absent from ojo tiles.
After 14 months of development, the community structure, defined as the relative abundance of functional groups, was significantly different between ojo and control conditions (PERMANOVA chemistry, F1,12 = 14.89, p = 0.0001; Fig 4). The differences in community structure between ojos and controls were driven primarily by higher abundances of CCA and erect calcified algae in the ambient zone, while the ojos had higher abundances of biofilm, erect fleshy algae, and turf algae (Fig 5). After 14 months, there were relatively minor differences between sites A and B (F1,12 = 2.96, p = 0.05). There were only marginal differences in the relative abundance between sites, with erect red algae being slightly more abundant at Site A and erect green algae and encrusting calcified algae being slightly more abundant at Site B. However, the site variable contributed only minimally to variation, and the impact of the sites was negligible compared to the impact of Ωarag (Fig 4).
In an nMDS, distance on the plot is a measure of dissimilarity. Each dot is representative of a single tile, labeled by site (open circles (site A) or closed circles (site B)), time (3, 6, or 14 months) and saturation state (red for ojo centers, black for controls). In an nMDS plot, multivariate data are placed into two-dimensional space so that the rank differences among the data are preserved, based on the BC dissimilarity matrix. Thus communities that are more dissimilar to one another are farther apart on the plot.
Open black symbols represent controls and closed red symbols represent the ojo centers. Diamonds depict Site A and circles depict Site B. Error bars are ± S.D. Note that the vertical (y-axis) scales on each figure are different.
In addition to the differences in the relative abundance of functional groups, the composition of the assemblages, defined as the presence or absence of functional groups, also varied between chemistry regimes (PERMANOVA, F1,12 = 7.44, p = 0.002). These differences in community composition were mainly due to the absence of erect and crustose coralline algae (CCA) on some, but not all, of the ojo tiles. In contrast, the community composition did not differ between sites (PERMANOVA site F1,12 = 0.24, p = 0.76) after 14 months of development.
Community structure changed through time (PERMANOVA time F2,28 = 17.64, p = 0.0001) in both chemistry regimes. Significant differences between chemistry regimes (F1,28 = 18.14, p = 0.0001) and sites (F1,28 = 3.54, p = 0.02) were maintained through time. Within each chemistry regime and site, however, the community structure only differed between 3 months and the following time points, but not between 6 and 14 months.
Foraminifera abundance (Fig 6B) was affected by both chemistry regime and time (ANOVA pH x time, F2,28 = 6.44, p = 0.005). On control tiles, the number of foraminifera increased from 3 to 6 months (paired t-test, p = 0.01), and then declined by 14 months (p = 0.0007) (Fig 6a). The number of foraminifera was greater on control than ojo tiles at 3 months (p = 0.04), but did not differ at 6 months (p = 0.1). At 14 months, there were more foraminifera at the ojos than at the controls (p = 0.03), stemming from a significant decline in their abundance at the controls. The number of foraminifera at the ojo centers increased marginally from 3 to 14 months (p = 0.06). Conversely, no trends were found in relation to chemistry regime for either polychaete or vermetid molluscs (Fig 6c) abundance, although vermetids increased in number over time (PERMANOVA time, F2,28 = 5.42, p = 0.009; Fig 6b and 6c).
(A,D,E) Average encrusting foraminifera, polychaete, and vermetid mollusc abundance (# of individuals) for each set of tiles by month. Open black symbols represent controls and closed red symbols represent the ojo centers. Diamonds depict Site A and circles depict Site B. Error bars are ± S.D. Note that the vertical (y-axis) scales on each figure are different. Visual examples of encrusting foraminifera (B) and vermetid molluscs (C) found on the tiles. Images taken with a 0.1mm accuracy digital microscope.
To address the small relative percent cover of fauna present on the tiles, the percent cover estimates of all calcified taxa (calcareous algae and all animals) were grouped for a univariate comparison. The total percent cover of all calcified taxa was greater in the controls than in the ojos at all time periods (Fig 7a). Importantly, the percent cover of all calcified taxa increased significantly among all time periods on the control tiles, but did not increase substantially between 6 and 14 months on the ojo tiles. Rather, the percent cover of all calcified taxa stagnated on the ojo center tiles at 6 months.
Aggregate calcified taxa (A) and combined fleshy algal indices (B) by month. Each point represents the mean across both sites for Sites A and B. Black triangles are controls and red squares are ojo centers. Error bars are ± S.D.
An additional univariate comparison was made for aggregate fleshy algal indices (summing erect green, red, and brown algae and turf algae). Total fleshy algal coverage was significantly greater at the ojo centers than at the controls at each time point, and by 14 months, was 42% greater at the ojos (Fig 7b).
After 14 months of recruitment and development, there were significant differences in species composition and relative abundance of species present on the ojo tiles compared to the controls. Differences in community structure between ojos and controls were primarily due to greater percent cover of erect and crustose coralline algae on the control tiles. In our study, both upright and crustose coralline algae (CCA) were found at the ojos in low saturation conditions, although coverage was significantly reduced compared to control sites. CCA appear to recruit early regardless of saturation, suggesting that these species may be physiologically tolerant to low saturation waters during early settlement and growth. However, the development of the CCA ceased after 3 months at the ojos, and by 14 months, the ojo tiles had 82% less CCA than controls. This finding is similar to laboratory results of Kuffner et al. (2007), who estimated a drop in percent cover of CCA of more than 90% for a similar decrease in saturation (Ωarag = 1.5) . On average, erect calcified algae had 89% less cover near the ojos and were conspicuously absent from numerous ojo tiles altogether.
This cessation of development in the percent cover of the coralline algae on the ojo tiles coincided with an increase in percent cover of fleshy algae, a trend that was consistent through time and which resulted in 42% greater coverage by fleshy algae at ojo centers after 14 months of development. While there was variability among the functional groups present on the tiles, the aggregated fleshy algal coverage was greater on ojo tiles compared to controls at each time point. Similar trends were observed by Kroeker et al. (2012) at the CO2 vent site at Ischia, Italy, in which calcareous species did not increase in percent cover after 6 months, which was attributed to overgrowth or space occupation by fleshy and turf algae. Fleshy and turf algae can limit CCA by overgrowth and subsequent light limitation [33–35] or by occupying the space needed for recruitment. Numerous studies have highlighted the importance of disturbance (e.g., grazing , scour , or epithelial sloughing  in mediating algal overgrowth of CCA, and without these mechanisms, CCA are generally subordinate to fleshy algae. Importantly, we did not find grazers or have high sources of disturbance on our tiles. Our study is consistent with the idea that despite some species of CCA being physiologically capable of recruitment and growth in conditions of low pH-saturation, competitive interactions between fleshy algae and coralline algae at high pCO2 could reduce the percent cover of coralline algae . This competitive advantage of fleshy alga over other taxa was also noted by Connell and Russell (2010), who observed that space occupation by fleshy algae increase at high pCO2, and Kuffner et al. (2007) who concluded that turf algae can limit CCA abundance in high CO2 [19,32]. This is in contrast, however to a recent study by Fabricius et al. (2015) at naturally acidified reefs in Papua New Guinea, which found CCA cover decreased with pH despite low cover of fleshy algal species . Differences among these studies could in part be attributed to higher grazing pressure at the PNG site, which could have limited fleshy algal cover at this site, or species-specific differences in response to high CO2. Our study implies that other anthropogenic stressors that allow fleshy algae to flourish on coral reefs, such as nutrient loading [40,41] or overfishing [42,43], could exacerbate the effects of ocean acidification on CCA development and cover. That is, fleshy algae appear to have a competitive edge over corallines in high CO2 (but see Short et al. 2014) , and additional stressors to coralline communities, or any factor that would give more advantage to the fleshy algae (i.e. higher nutrient levels or decreased herbivory) [40,45,46], could compound the indirect effects of acidification to coralline communities.
It is likely that changes in pCO2 may impact a range of physiological processes in both fleshy and coralline algae, with consequences to the community assemblages that can preferentially recruit and grow . There is a critical need to understand how pCO2 will impact photosynthesis, enzyme systems, and reproduction rates in various marine algae, and thus predict which species of algae will flourish in future reef environments. Our study lends support to the idea that advantages in carbon transport in fleshy macroalgae are equally as important to understand as carbon transport and calcifiction in coralline algae.
Coralline algae are important ecological components of a coral reef, as they cement the reef framework and provide chemical settlement cues and settlement substrate for coral larvae [48,49]: understanding the response of coralline algae to ocean acidification is therefore of critical importance. A dramatic decline in coverage under conditions of low pH-saturation indicates that both the basic framework of reefs and the recruitment of corals could decrease with ocean acidification. Doropoulos et al. (2012) investigated coral recruitment in response to reduced coralline algae abundance at high pCO2 and found that coral settlement was mediated by settlement cues from CCA’s, which were most heavily impacted by saturation state . Their study noted a greater than 45% decline in coral recruitment due to declining CCA and loss of settlement cues. In our study, the experimental substrates were deployed in August (a likely time for coral mass-spawning) and retrieved 14 months later in the hopes of capturing at least one mass-spawning event. Despite this, the number of corals that recruited and settled on the tiles was not sufficient to address how acidification may impact coral recruitment and growth. As the recruitment substrates were placed in a lagoon with low overall coral coverage, this was not unexpected. However, the 14-month control tiles had a total of 4 colonies of Siderastrea radians, while no corals were present on the low saturation tiles, a trend that we feel is worth noting. The small juvenile colonies present suggested the corals were recent recruits. As previously noted, corals rely on important settlement cues from coralline algae, and as the low saturation tiles had approximately 80% less CCA than the control tiles, they may have been less hospitable to coral larvae. Our study suggests that this potential decline in coral recruitment can in part be attributed to the competitive advantage of fleshy algae over CCA in more acidic conditions. This connection between the impact of fleshy algal coverage, CCA, and coral recruitment warrants further study.
When considering all calcareous species (flora and fauna combined), the difference in percent cover between control and ojo tiles was similar at all time points, with approximately 70% less cover on the ojo tiles. The percent cover at the controls increased over time; however, in the ojo zones, there was no increase in calcifying cover after 6 months. Unlike the Ischia site, where no calcareous species were found at Ωarag = 1.2 , low saturation at Puerto Morelos (Ωarag≤1.5) still had up to 30% cover of calcifying organisms. This could be because the calcareous species present at Puerto Morelos are naturally more resilient to acidification, or because calcification continues to drop as the saturation level decreases reaching negligible levels at Ωarag = 1.2. Regardless, this finding has significant implications for future reef development, as the reduction in calcified taxa was immediate and persistent throughout the duration of the deployment.
The observed patterns are also consistent with competition between calcifying taxa. The trends observed for encrusting foraminifera (Fig 6) over the 14-month study suggest that CCA have a decreased ability to compete under acidification conditions. While the abundance of encrusting foraminifera increased from 3–6 months by 40% on the control tiles, at 14 months when CCA became established, the trend reversed and they decreased by 70%. Visual analysis of the tiles revealed that many of the foraminifera were overgrown by CCA between 6 months and 14 months, suggesting CCA out-competed the foraminifera for remaining space. This trend was not seen on the ojo tiles. Instead, the number of foraminifera continued to increase on the ojo center tiles where CCA were not able to establish dominance and were 80% less abundant. As with the calcifying algae, our results suggest that the calcifying foraminifera were able to grow in the ojo conditions, and that community changes were primarily driven by competition among species, where those that did not compete with CCA occupied more space on the hard substrate.
Differences in community structure between chemistry regimes were also reflected by a higher abundance of biofilm on the low saturation tiles. By 14 months, biofilm cover was significantly higher on the ojo than at the control tiles, a trend that is also consistent with the Kroeker et al. (2012) study. Additionally, in our study, there was a significant increase in the amount of biofilm in both zones at 6 months. Biofilms are essential components of marine ecosystems, as they are food for a number of grazers [50,51], and more importantly to this study, likely mediate the settlement and metamorphosis of benthic organisms [52–55]. Biofilms begin to settle on substrates within hours of submersion and are associated with early stages of succession [51,52]. The significantly higher percent cover of biofilm on the ojo tiles after 14 months of development could suggest they remained in an earlier stage of development for a longer time compared to the control tiles (as seen at Ischia). The increase in biofilm on both tiles at 6 months of growth can potentially be explained by seasonal variability (the tiles were removed in the winter months with lower incident radiation), as biofilm abundance has been shown to be inversely related to solar stress [50,51].
There were no obvious trends in vermetid mollusk or polychaete abundance by site or zone. In fact, the vermetid mollusk abundance increased significantly by 14 months on the control tiles. This observation of resilience by certain taxa to acidification is mirrored by several studies to date [8,16,17,56,57]. However, as these organisms were likely only exposed to low saturation conditions after settlement on the tiles, it is not clear from our study if acidification may impact earlier life history stages. It is possible that these taxa are impacted more in the early larval stages before settlement, and that “carry-over” effects which are expressed only in adults that were exposed to acidification during early stages of growth  are responsible for the discrepancy seen between the populations in this study and those of laboratory experiments.
While in-situ field studies are valuable for investigating how complex assemblages may respond to acidification, multiple environmental parameters may co-vary, making it difficult to resolve the influence of Ωarag on the community assemblages from that of other factors, or to assess the extent to which the influence of Ωarag may be modulated by other, co-varying factors. While temperature and light are comparable between the ojo and control zones in our study, salinity and nutrients often co-vary with changes in saturation state in the ojos as they are all dependent on the flux of submarine groundwater discharge. It is therefore important to compare our results to those derived from additional field studies where Ωarag conditions are not coupled with groundwater discharge. At the Ischia volcanic vent site, there were no salinity or nutrient changes associated with pH zones, and the similarity of our results with those of Kroeker et al. (2012) (after which this investigation was designed) lends weight to the idea that the carbonate chemistry is the primary driver of change in the communities at the ojos. Specifically, the similarities in community structure and composition with respect to CCA and fleshy algae suggests these changes may be due to low pH-saturation exposure.
Our study illustrates that while acidification will have significant direct impacts on calcification, the altered competitive interactions between organisms will also impact community assemblages in the future. That is, we expect to see a shift in communities from coralline algal coverage to fleshy algae over time as pCO2 increases over the 21st century. It is important to note that the tropical benthic calcifying organisms were able to recruit and grow in low Ωarag conditions. However, we hypothesize that competition for space as the community developed was the leading driver in the community shifts observed. Our study illustrates the importance of observing the response of entire communities to OA, as interactions between organisms will compound the direct effects of acidification and likely increase reef degradation beyond the estimates derived from species-specific observational studies. This study thus illustrates the need for conservation and policy decisions that will consider community-wide responses to acidification, particularly with regard to the increased competition between calcifying and fleshy algal species at decreased saturation. For instance, if fleshy algal species are more successful at high nutrient levels, then mitigation strategies that reduce eutrophication of surface waters may help fleshy algae from having an even greater competitive edge over calcifying species. As the oceans become more acidic over the 21st century, it will become essential to alleviate human impacts that have the potential to compound competitive interactions between organisms.
S1 Fig. Salinity and pH over time as measured by an autonomous sensor.
Salinity and pH were measured at 15 minute time intervals for a period of 3 months (August-October 2010) for a total of over 5500 data points at a single spring. Salinity is plotted against pH (a), and grouped according to the number of data points occurring in a given salinity range (b). As depicted, 93% of data points fall above a salinity of 30, and salinity never drops below 27 at the center of discharge. The lower salinity conditions are during low tide in the rainy season and the conditions do not prevail for more than a one hour. Agreement (1 s.d.) between the pH of the sensor values and the discrete measurements (calculated pH) is approximately ±0.07.
We thank A. Martinez-Fernandez (UCSC) for her help with tile deployment and recovery. We also thank Y. Takeshita (UCSD), and T. Martz (UCSD) for their help with instrument deployment and data analysis. L. Fox (UCSC) helped with statistical analyses and R. Franks (UCSC) helped with chemical analyses. The water quality team at CICY was instrumental in obtaining monthly water samples for chemical analyses.
Conceived and designed the experiments: EDC KJK DCP AP. Performed the experiments: EDC DCP AP MRV LH. Analyzed the data: EDC KJK DCP AP. Contributed reagents/materials/analysis tools: EDC KJK DCP AP LH MRV. Wrote the paper: EDC KJK DCP AP LH MRV.
- 1. Orr JC, Fabry VJ, Aumont O, Bopp L, Doney SC, Feely RA, et al. Anthropogenic ocean acidification over the twenty-first century and its impact on calcifying organisms. Nature. 2005; 437: 681–686. pmid:16193043
- 2. Fabry VJ, Seibel BA, Feely RA, Orr JC. Impacts of ocean acidification on marine fauna and ecosystem processes. ICES J Mar Sci. 2008; 65(3): 414–432.
- 3. Doney SC, Fabry VJ, Feely RA, Kleypas JA. Ocean acidification: the other CO2 problem. Ann Rev Mar Sci. 2009; 1:169–192. pmid:21141034
- 4. Hoegh-Guldberg O, Mumby PJ, Hooten AJ, Steneck RS, Greenfield P, Gomez E, et al. Coral reefs under rapid climate change and ocean acidification. Science. 2007; 318: 1737–1742. pmid:18079392
- 5. De’ath G, Lough JM, Fabricius KE. Declining coral calcification on the Great Barrier Reef. Science. 2009; 323: 116–119. pmid:19119230
- 6. Pandolfi JM, Connolly SR, Marshall DJ, Cohen AL. Response. Science. 2011; 334: 1495–1496.
- 7. Anthony KRN, Kline DI, Diaz-Pulido D, Dove S, Hoegh-Guldberg O. Ocean acidification causes bleaching and productivity loss in coral reef builders. Proc Natl Acad Sci USA. 2008; 105: 17442–17446. pmid:18988740
- 8. Jokiel PL, Rodgers KS, Kuffner IB, Anderson AJ, Cox EF, Mackenzie FT. Ocean acidification and calcifying reef organisms: a mesocosm investigation. Coral Reefs. 2008; 473–483.
- 9. de Putron SJ, McCorkle DC, Cohen AL, Dillon AB. The impact of seawater saturation state and bicarbonate ion concentration on calcification by new recruits of two Atlantic corals. Coral Reefs. 2011; 30: 321–328.
- 10. Ries JB, Cohen AL, McCorkle DC. Marine calcifiers exhibit mixed responses to CO2 induced ocean acidification. Geology. 2009; 37: 1131–1134.
- 11. Kroeker KJ, Kordas RL, Crim RN, Singh GS. Meta-analysis reveals negative yet variable effects of ocean acidification on marine organisms. Ecol Lett. 2010; 13: 1419–1434. , 2010.
- 12. Hall-Spencer JM, Rodolfo-Metalpa R, Martin S, Ransome E, Fine M, Turner SM, et al. Volcanic carbon dioxide vents show ecosystem effects of ocean acidification. Nature. 2008; 454: 96–99. pmid:18536730
- 13. Manzello DP. Coral growth with thermal stress and ocean acidification: lessons from the Eastern Tropical Pacific. Coral Reefs. 2010; 29: 749–758.
- 14. Fabricius KE, Langdon C, Uthicke S, Humphrey C, Noonan S, De’ath G, et al. Losers and winners in coral reefs acclimatized to elevated carbon dioxide concentrations. Nat Clim Chang. 2011; 1: 165–169.
- 15. Price NN, Martz TR, Brainard RE, Smith JE. Diel variability in seawater pH relates to calcification and benthic community structure on coral reefs. PLoS ONE. 2012; 7(8): e43843. pmid:22952785
- 16. Kroeker KJ, Micheli F, Gambi MC. Ocean acidification causes ecosystem shifts via altered competitive interactions. Nature. 2012.
- 17. Kroker KJ, Micheli F, Gambi MC, Martz TR. Divergent ecosystem responses within a benthic marine community to ocean acidification. Proc Natl Acad Sci USA. 2011; 108(35):14515–14520. pmid:21844331
- 18. Crook ED, Potts D, Rebolledo-Vieyra M, Hernandez L, Paytan A. Calcifying coral abundance near low pH springs: implications for future ocean acidification. Coral Reefs. 2011; 31(1): 239–245.
- 19. Andersson AJ, Kline DI, Edmunds PJ, Archer SD, Bednarsek N, Carpenter RC, et al. Understanding ocean acidification impacts on organismal to ecological scales. Oceanography. 2015; 28(2): 16–27.
- 20. Russell SD, Connell BD. The direct effects of increasing CO2 and temperature on non-calcifying organisms: increasing the potential for phase shifts in kelp forests. Proc R Soc B. 2010.
- 21. Connell SD, Kroeker KJ, Fabricius KE, Kline DI, Russell BD. The other ocean acidification problem: CO2 as a resource among competitors for ecosystem dominance. Proc R Soc B. 2013; 368: 20120442.
- 22. Falkenberg LJ, Russell BD, Connell SD. Future herbivory: the indirect effects of enriched CO2 may rival its direct effects. Mar Ecol Prog Ser. 2013; 492: 85–95.
- 23. Connell SD, Russell BD, Irving AD. Can strong consumer and producer effects be reconciled to better forecast ‘catastrophic’ phase-shifts in marine ecosystems? J Exp Mar Bio Ecol. 2011; 400 (1–2): 296–301.
- 24. Porzio L, Buia MC, Hall-Spencer JM. Effects of ocean acidification on macroalgal communities. J Exp Mar Bio Ecol. 2011; 400: 278–287.
- 25. Beddows PA, Smart PL, Whitaker FF, Smith SL. Decoupled fresh-saline groundwater circulation of a coastal carbonate aquifer: spatial patterns of temperature and specific electrical conductivity. J Hydrol. 2007; 246: 18–32.
- 26. Paytan A, Crook ED, Cohen AL, Martz T, Takeshita Y, Rebolledo-Vieyra M, et al. Reply to Iglesias-Prieto et al.: Combined field and laboratory approaches for the study of coral calcification. Proc Natl Acad Sci USA. 2014; 111 (3): E302–E303. pmid:24596954
- 27. Crook ED, Cohen AL, Rebolledo-Veiyra M, Hernandez L, Paytan A. Reduced calcification and lack of acclimatization by coral colonies growing in areas of persistent natural acidification. Proc Natl Acad Sci USA. 2013; 110 (27): 1044–1049.
- 28. Solomon S, Qin D, Manning M, Chen Z, Marquis M, Averyt KB, et al. Climate Change 2007: The Physical Science Basis, Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge: Cambridge University Press; 2007.
- 29. Pierrot D, Lewis E, Wallace DWR. MS excel program developed for CO2 system calculations. ORNL/CDIAC-105a Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, U S Department of Energy; 2006.
- 30. Mehrbach C, Culberson CH, Hawley JE, Pytkowicz RM. Measurement of the apparent dissociation constants of carbonic acid in seawater at atmospheric pressure. Limnol Oceanogr. 1973; 18: 897–907.
- 31. Dickson AG, Millero FJ. A comparison of the equilibrium constants for the dissociation of carbonic acid in seawater media. Deep Sea Res. 1987; 34: 1733–1743.
- 32. Kuffner IB, Andersson AJ, Jokiel PL, Rodgers KS, Mackenzie FT. Decreased abundance of crustose coralline algae due to ocean acidification. Nature. 2007; 1: 114–117.
- 33. Williams SL, Carpenter RC. Photosynthesis/photon flux density relationships among components of coral reef algal turfs. J Phycol. 1990; 26: 36–40.
- 34. Sebens KP. Spatial relationships among encrusting marine organisms in the New England subtidal zone. Ecol Monogr. 1986; 56(1): 73–96.
- 35. Kendrick GA. Recruitment of coralline crusts and filamentous turf algae in the Galapagos archipelago: effect of simulated scour, erosion, and accretion. J Exp Mar Biol Ecol. 1991; 147(1): 47–63.
- 36. Adey WH. Temperature control of reproductivity and productivity in a subarctic coralline alga. Phycologia. 1973; 2: 111–118.
- 37. Littler MM, Littler DS. Relationships between macroalgal functional form groups and substrata stability in a tropical rocky-intertidal system. J Exp Mar Biol Ecol. 1984; 74: 13–34.
- 38. Littler MM. The crustose Corallinaceae. Ann Rev Oceanog Mar Biol. 1972; 10: 31–47.
- 39. Fabricius KE, Kluibenschedl A, Harrington L, Noonan S, De’ath G. In-situ changes of tropical crustose coralline algae along carbon dioxide gradients. Scientific Reports. 2015; 5: 9537. pmid:25835382
- 40. Lapointe BE. Nutrient thresholds for bottom-up control of macroalgal blooms on coral reefs in Jamaica and southeast Florida. Limnol Oceanogr. 1997; 42(5): 1119–1131.
- 41. Fabricius KE. Effects of terrestrial runoff on the ecology of corals and coral reefs: review and synthesis. Mar Pollut Bull. 2005; 50: 125–146. pmid:15737355
- 42. Bellwood DR, Hughes TP, Folke C, Nystrom M. Confronting the coral reef crisis. Nature. 2004; 429: 827–833. pmid:15215854
- 43. Mumby PJ, Hastings A, Edwards HJ. Thresholds and the resilience of Caribbean reef corals. Nature. 2007; 450: 98–101. pmid:17972885
- 44. Short J, Kendrick GA, Falter J, McCulloch MT. Interactions between filamentous turf algae and coralline algae are modified under ocean acidification. J Exp Mar Biol Ecol. 2014; 456: 70–77.
- 45. Thacker RW, Ginsburg DW, Paul VJ. Effects of herbivore exclusion and nutrient enrichment on coral reef macroalgae and cyanobacteria. Coral Reefs. 2001; 19: 318–329.
- 46. Koop K, Booth D, Broadbent A, Brodie J, Bucher D, Capone D, et al. ENCORE: The effect of nutrient enrichment on coral reefs. Synthesis of results and conclusions. Mar Pollut Bull. 2001; 42(2): 91–120. pmid:11381890
- 47. Porzio L, Buia MC, Hall-Spencer JM. Effects of ocean acidification on macroalgal communities. J Exp Mar Biol Ecol. 2011; 400: 278–287.
- 48. Morse DE, Hooker N, Morse ANC, Jensen RA. Control of larval metamorphosis and recruitment in sympatric agariciid corals. J Exp Mar Bio Ecol. 1988; 116(3): 193–217.
- 49. Heyward AJ, Negri AP. Natural inducers for coral larval metamorphosis. Coral Reefs. 1999; 18: 273–279.
- 50. Doropoulos C, Ward S, Diaz-Pulido G, Hoegh-Guldberg O, Mumbay PJ. Ocean acidification reduces coral recruitment by disrupting intimate larval-algal settlement interactions. Ecol Lett. 2012; 15: 338–346. pmid:22321314
- 51. Hill AS, Hawkins SJ. Seasonal and spatial variation of epilithic micro algal distribution and abundance and its ingestion by Patella vulgata on a moderately exposed rocky shore. J Mar Biol Assoc. 1991; UK 71 (2): 403–423.
- 52. Thompson RC, Norton TA, Hawkins SJ. Physical stress and biological control regulate the producer-consumer balance of intertidal biofilms. Ecology. 2004; 85: 1372–1382.
- 53. Meadows PS, Williams GB. Settlement of Spirorbis borealis daudin larvæ on surfaces bearing films of micro-organisms. Nature. 1963; 198: 610–611.
- 54. Thompson RC, Norton TA, Hawkins SJ. The influence of epilithic microbial films on the settlement of Semibalanus balanoides cyprids—a comparison between laboratory and field experiments. Hydrobiologia. 1998; 375: 203–216.
- 55. Tebben J, Tapiolas DM, Motti CA, Abrego D, Negri AP, Blackall LL, et al. Induction of larval metamorphosis of the coral Acropora millepora by tetrabromopyrrole isolated from a Pseudoalteromonas bacterium. PLoS One. 2011; 6(4), e19082. pmid:21559509
- 56. Johnson VR, Brownlee C, Rickaby REM, Graziano M, Milazzo M, Hall-Spencer JM. Responses of marine benthic microalgae to elevated pCO2. Mar Biol. 2013; 160: 1813–1824.
- 57. Fabricius KE, De'ath G, Noonan S, Uthicke S. Ecological effects of ocean acidification and habitat complexity on reef-associated macroinvertebrate communities. Proc R Soc B. 2014; 281: 20132479. pmid:24307670
- 58. Hettinger A, Sanford E, Hill TM, Russell AD, Sato KNS, Hoey J, et al. Persistent carry-over effects of planktonic exposure to ocean acidification in the Olympia oyster. Ecology. 2012; 93: 2758–2768. pmid:23431605