Why corals care about Ocean acidification uncovering the Mechanism

Author Posting. © Oceanography Society, 2009. This article is posted here by permission of Oceanography Society for personal use, not for redistribution. The definitive version was published in Oceanography 22 no. 4 (2009): 118-127.

Ocean acidification reduces the pH and thus the abundance of carbonate ions in seawater.Corals living in acidified seawater continue to produce CaCO 3 and expend as much energy as their counterparts in normal seawater to raise the pH of the calcifying fluid.However, in acidified seawater, corals are unable to elevate the concentration of carbonate ions to the level required for normal skeletal growth.In several experiments, we found that boosting the energetic status of corals by enhanced heterotrophic feeding or moderate increases in inorganic nutrients helped to offset the negative impact of ocean acidification.However, this built-in defense is unlikely to benefit corals as levels of CO 2 in the atmosphere continue to rise.Most climate models predict that the availability of inorganic nutrients and plankton in the surface waters where corals live will decrease as a consequence of global warming.Thus, corals and coral reefs may be significantly more vulnerable to ocean acidification than previously thought.
Because the calcium ion concentra- that prevent nucleation and/or crystal growth.These kinetic barriers include the high hydration energy of the calcium ions (e.g., Lippmann, 1973), the low concentration and activity of the carbonate ions (e.g., Garrels and Thompson, 1962;Lippmann, 1973), and the presence of high concentrations of sulfate and magnesium (e.g., Usdowski, 1968;Kastner, 1984).
Most marine calcifiers, therefore, indicate that seawater is likely the starting fluid for calcification (Cohen et al., 2001(Cohen et al., , 2009;;Braun and Erez, 2004;Gaetani and Cohen, 2006;Holcomb et al., 2009), and it is transported to the site of calcification located between the base of the calicoblastic epithelium and the existing skeletal surface (Braun and Erez, 2004).The route by which seawater enters this space remains unclear, although calcein dye tracer studies rule out cross-membrane transport (Braun and Erez, 2004)   and Figure 1D) and that radiate from discrete aggregations of tiny granular crystals, each several nanometers in diameter (g in Figure 1B, and Figure 1C).
In many corals, there is a distinct diurnal cycle to the formation of these crystals: the granular crystals form at night, whereas the needle-shaped crystals form during the day (Holcomb, 2009).
Each aggregate of granular crystals and its associated bundle of needles is called a sclerodermite (Wells, 1956).
Sclerodermites are the basic building blocks or "bricks" of the coral skeleton.In this treatment, crystal growth is so slow that crystal morphology assumes an orthorhombic shape characteristic of aragonite grown slowly at near equilibrium conditions (Figure 5D).From Cohen et al., 2009 In Figure 5E, we plot the aspect ratio of crystals accreted by corals reared in seawater with Ω ar ranging from ambient  Conversely, if the energy budget for calcification were fixed and the coral could expend only enough energy to remove 4500 nmol protons from the calcifying fluid independent of the saturation state of the "intake" water, then our calculations show that a coral reared at Ω ar ~ 2.4 would be able to elevate the internal saturation state from 2.4 to about 15 (Figure 6).And a coral reared in seawater with Ω ar of 1.5 would elevate the internal saturation state to approximately 10 by removing the same number of protons.These numbers are consistent with our estimate of actual calcifying fluid saturation state based on crystal aspect ratios (Figure 5E), and suggest that corals reared under high-CO 2 conditions may have worked as hard, and pumped as many protons from their calcifying fluid, as corals in the ambient treatment.
However, in our experimental aquaria, there appears to be a limit to the amount of work the corals will do even at the expense of building the healthy skeleton that is so critical to recruitment success.This result suggests that the coral's energy budget for calcification might be limited by the availability of nutrients or food.

the pOteNtial rOle OF NutrieNtS aNd FOOd iN MOdulatiNg cOr al reSpONSe tO OceaN acidiFicatiON
If it is true that the coral's energy budget for calcification is limited by the availability of nutrients, then elevating the nutritional status of a coral calcifying in low-saturation-state seawater might provide the additional fuel it needs to maintain healthy calcification rates.This could be done by enhanced heterotrophic feeding or by adding moderate amounts of inorganic nutrients, such as nitrates and phosphates, to stimulate zooxanthellate photosynthesis, providing the coral with carbohydraterich photosynthate to fuel calcification.
Normally, nutrient addition stimulates zooxanthellate photosynthesis, which, under ambient CO 2 conditions, can lead to CO 2 limitation and a decline in calcification (e.g., Marubini and Davies, 1996).However, under elevated  .calcification achieved by removing protons from the internal calcifying fluid is energetically expensive.The energy expended in calcification is represented here by the number of protons pumped from each milliliter of calcifying fluid.corals reared in seawater with a range of aragonite saturation states (depicted by different colors) remove the same number of protons from that seawater (4500 nmol ml -1 ) but achieve different calcifying saturation states (shown by grey bar).This suggests that corals reared in acidified seawater did not divert more energy to proton removal even at the expense of building a normal skeleton.in this cO 2 Sys calculation, the initial calcifying fluid is seawater with an initial alkalinity of 2470 µmol kg -1 in equilibrium with the specified pcO 2 (400, 740, or 1340 µatm).The fluid is isolated from the surrounding seawater; proton pumping elevates the alkalinity and saturation state within the calcifying space while cO 2 diffuses through the calicoblastic epithelium to maintain equilibrium pcO 2 .calculations were made with a Matlab implementation of cO 2 Sys using constants of Mehrbach et al. (1973) refit by dickson andMillero (1987) for carbonate, and dickson (1990) for sulfate; input conditions are as follows: S = 30, t = 25, atmospheric pressure = 1atm.concentrations of silicate, phosphate, ammonia and hS were set at 0. ca and B concentrations were calculated from salinity.
of hectares of tropical coastline are dominated by coral reef ecosystems that exist only because coral animals, small anemone-like creatures called "polyps, " can produce calcium carbonate (CaCO 3 ) crystals faster than their skeletons are eroded by the sea.Within an extracellular space beneath the coral polyp, micron-sized CaCO 3 crystals in the form of aragonite are nucleated and grown around the clock, 365 days of the year.They are packed into bundles and stacked meticulously one on top of the other to form an intricately designed skeleton that both supports and protects the animal that built it.Most reef-building corals exist as colonies in which the individual skeletons of hundreds, sometimes thousands, of polyps can form enormous domed or branching structures sometimes reaching B Y a N N e l .c O h e N a N d M i c h a e l h O l c O M B S p e c i a l i S S u e F e at u r e B Y a N N e l .c O h e N a N d M i c h a e l h O l c O M B aBStr act.Stony corals build hard skeletons of calcium carbonate (CaCO 3 ) by combining calcium with carbonate ions derived, ultimately, from seawater.The concentration of carbonate ions relative to other carbonate species in seawater is rather low, so corals expend energy to raise the pH of seawater sequestered in an isolated, extracellular compartment where crystal growth occurs.This action converts plentiful bicarbonate ions to the carbonate ions required for calcification, allowing corals to produce CaCO 3 about 100 times faster than it could otherwise form.It is this rapid and efficient production of CaCO 3 crystals that enables corals to build coral reefs.
required for successful skeleton-building increases as the acidity of the ocean increases.We present evidence that corals growing under nutrient-replete conditions can redirect the extra energy provided by slightly elevated levels of inorganic nutrients or food toward calcification, thus dampening the negative impact of ocean acidification.Finally, we consider the implications of a changing climate, including predicted reductions in surface ocean nutrient concentrations and primary productivity, for the future of corals and coral reefs.BiOMiNer alizatiON BaSicS Understanding the fundamentals of coral calcification (i.e., the processes involved in the nucleation and growth of aragonite crystals) is a crucial first step in predicting the response of corals and coral reef ecosystems to future climate change, including ocean acidification.A useful place to start thinking about calcification strategies in the marine environment is to recognize that all marine calcifiers, including the reef-building or "stony" corals, have to overcome the kinetic barriers to CaCO 3 precipitation that exist naturally in seawater.
tion [Ca 2+ ] in seawater is very high and relatively constant, variations in Ω ar are determined mainly by the carbonate ion concentration [CO 3 2-].Today, the [CO 3 2-] of seawater in the low-latitude surface ocean is about 250 µmol kg -1 .Aragonite saturation is reached at a [CO 3 2-] of about 60 µmol kg -1 .Thus, the tropical surface ocean is, in general, but with notable exceptions such as the eastern tropical Pacific, about four times supersaturated with respect to aragonite.When Ω ar > 1 (i.e., [CO 3 2-] > 60 µmol kg -1 ), aragonite should, theoretically, precipitate from seawater.Conversely, when Ω ar < 1, aragonite should dissolve in seawater.The warm tropical ocean where most coral reefs are found is highly supersaturated with respect to aragonite, in other words, Ω ar can be significantly greater than 1.Nevertheless, neither calcite nor aragonite will form spontaneously because there are kinetic barriers must nucleate and grow CaCO 3 crystals within compartments that are isolated or semi-isolated from the external seawater, and within which they can modify, regulate, and control conditions, including several meters in height.Countless skeletons accreted over many millennia provide the reef framework, the concrete jungles of the tropical ocean that provide myriad marine and terrestrial species with habitat, nesting grounds, and food.For human populations along tropical coastlines, healthy reefs provide natural buffers to beach erosion, barriers to tsunamis and hurricanes, and significant income through tourism and fisheries.By one estimate, the net global economic value of coral reefs worldwide is a staggering $29.8 billion each year(Cesar et al., 2003).Over the next century, rising levels of atmospheric CO 2 will increase the concentration of total dissolved inorganic carbon (DIC) in seawater, but simultaneously reduce seawater pH and the abundance of carbonate ions [CO 3 2-] that corals and other marine calcifiers use to build their skeletons.For this reason, there is growing concern that the so-called "acidification" of the surface ocean may slow rates of CaCO 3 production or "calcification" by reef-building corals to a point where rates of reef erosion exceed rates of skeletal accretion, leading to the gradual, worldwide loss of coral reef ecosystems as we know them.Data from a wide range of laboratory experiments have fueled this concern, demonstrating that coral calcification (skeleton-building) can be highly sensitive to changes in seawater carbonate ion concentration.In this paper, we show how this process might work using a model based on the behavior of nonbiological (abiogenic) aragonites when grown in increasingly "acidified" seawater.We show that coral calcification under any circumstances is energetically costly to the coral animal and that the energy in some cases the carbonate chemistry of the calcifying fluid, to enable CaCO 3 precipitation to occur.Coccolithophores and gorgonians, for example, produce calcite and high-magnesium calcite intracellularly, the marine alga Halimeda accretes aggregations of acicular aragonite needles extracellularly within spaces created between cell membranes, and corals and mollusks accrete their CaCO 3 crystals within compartments created between the tissue and the existing skeleton or shell.In addition, there is substantial evidence for the involvement of organic molecules in the biomineralization processes of many marine calcifiers.Organic molecules may play a role in reducing the surface free energy required for nucleation (e.g., Teng et al., 1998), in guiding site-specific nucleation that results in species-specific skeletal architectures, or in buffering the pH of the calcifying fluid against large fluctuations in internal CO 2 concentrations caused by cycles in photosynthesis and respiration (Holcomb et al., 2009).In stony corals, skeletal formation is entirely external to the organism (extracellular) but the skeleton is not exposed to seawater.Corals either sit on top of or wrap themselves around their skeletons (Figure 1A) and the skeletal surface is completely enveloped by tissue, separated from the external seawater environment by four layers of cells.Data from a combination of geochemical, mineralogical, and tracer dye studies

Figure 1 .
Figure1.The coral animal (polyp) manipulates the chemistry of seawater sequestered in an isolated calcifying compartment to produce many millions of tiny calcium carbonate (aragonite) crystals that are assembled into a skeleton.in (a), each polyp (p) in the colony builds its own skeleton, the corallite (cl).The polyp sits atop of, and completely covers, the skeletal surface.The arrow points to the interface between the basal epithelial cells and the skeletal surface, where calcification occurs.in (B), the skeleton is composed of radiating arrays of needle-shaped crystals (f) that grow on aggregates of fine granular crystals (g). in (c), the granular crystals are accreted at night and in (d), the needle-shaped crystals are accreted mainly in the day.Scale bars: (a) 500 µm, (B) 10 µm, (c) and (d) 1 µm.(A) is adapted fromVeron (1993)

Figure
Figure 1B-D shows the morphology and microstructural arrangement of crystals in a coral skeleton.All the different components of the skeleton seen in Figure 1A are built of bundles of fine, needle-shaped crystals that are several microns in length (f in Figure 1B Aragonite crystals precipitated in the laboratory from a highly supersaturated seawater solution in the absence of a coral (i.e., abiogenic aragonites) bear remarkable resemblance to the coral sclerodermites (Figure2).Nucleation, which occurs at very high aragonite saturation states (Ω ar > 20), produces aggregates of submicron-sized granular crystals.At lower aragonite saturation states (Ω ar ~ 6-19), crystal growth is favored over nucleation.The growth phase produces bundles of fine, needleshaped crystals.Each aggregate of granular crystals and its associated needle-shaped crystals is called a "spherulite" (Figure2B;Holcomb et al., 2009).Aragonitic spherulites precipitated from a supersaturated seawater solution are compositionally (i.e., chemically) and morphologically similar to coral "sclerodermites, " as evident in Figure2.Based on this similarity, it is feasible that one mechanism corals may employ to achieve aragonite nucleation and growth is by elevating the saturation state of seawater trapped in an isolated or semiisolated calcifying compartment(Cohen and McConnaughey, 2003).Abiogenic aragonites grown experimentally from seawater over a range of aragonite saturation states show systemmatic and progressive changes in crystal morphology caused by changes in crystal growth rate (Figure3).Crystals grown experimentally at relatively low aragonite saturation states (Ω ar ~ 6) are short, wide, and highly faceted, consistent with low crystal growth rates (e.g.,Lofgren, 1974Lofgren,    , 1981; Figure3C).When the saturation state of the experimental seawater is increased during growth, the crystals produced are longer and thinner, more bladelike in appearance, and less faceted, consistent with a progressive increase in crystal growth rate (Figure3A,B).This change in crystal morphology with increasing seawater saturation state can be quantified using the crystal aspect ratio or the ratio of its length to its width.In Figure3D, the aspect ratio of aragonite crystals grown under experimental conditions is plotted against the saturation state of the seawater in which they grew.The crystal aspect ratio is linearly related to aragonite saturation state.The lower the seawater saturation state, the lower the crystal growth rate, and thus the lower the crystal aspect ratio.The relationship between seawater saturation state and crystal aspect ratio in these experiments can be described by the following equation: Ω ar = 0.93 (± 0.06) x crystal aspect ratio (µm) + 0.20(± 0.89); r 2 = 0.94.This relationship between crystal aspect ratio and aragonite saturation state can be applied to coral skeletons to

Figure 2 .
Figure 2. aragonite crystals produced by corals (B) and aragonite crystals produced experimentally from a highly supersaturated seawater solution (a) share many similarities that allow us to conclude that corals expend energy to elevate the saturation state of seawater in order to calcify. in both (a) and (B), bundles of aragonite fibers radiate out from a central region occupied by aggregates of submicronsized granular crystals.Scale bars are 1 µm in both.Images fromCohen and McConnaughey (2003) andHolcomb et al. (2009)

Figure 3 .
Figure3.aragonites precipitated experimentally from seawater over a range of aragonite saturation states display systematic changes in crystal morphology (a-c) that can be quantified using the crystal aspect ratio (the ratio of crystal length to width). in (d), crystal aspect ratio is plotted against aragonite saturation state of the seawater in which they grew.This relationship is used in Figure5to estimate the saturation state of the coral's calcifying fluid.Scale bars are 1 µm.

Figure
Figure 4. Ocean acidification negatively impacts the skeletal growth of eight-dayold corals reared from planula larvae in laboratory aquaria. in these images, the polyp has been removed to reveal the first skeleton or "primary corallite." in a, Ω ar of the aquarium seawater is ~ 3.7; in B, Ω ar ~ 2.4; in c, Ω ar ~ 1, and in d, Ω ar ~ 0.2.a-d scale bars are ~ 100 µm.FromCohen et al., 2009 (3.7) to strongly undersaturated (0.2).Note that the aspect ratio of the crystals decreases as the Ω ar of the aquarium seawater decreases.Applying the relationship between crystal aspect ratio and fluid saturation state established for experimental abiogenic aragonites, we can show that the saturation state of the coral's internal calcifying fluid is maintained significantly above that of the external seawater in all experimental conditions (Figure 5E).However, the saturation state of the coral's internal calcifying fluid decreases as that of the external seawater decreases.When the Ω ar of the external seawater is ~ 3.7 (i.e., at ambient levels), the coral pumps enough protons from the initial calcifying "intake" water to maintain an internal calcifying fluid saturation state significantly higher than it is outside (Ω ar ~ 19).At this level of aragonite supersaturation, crystals growth is very fast, allowing the coral to quickly build a massive and dense skeleton.When the Ω ar of the external seawater is ~ 2.4, the coral works to maintain the calcifying fluid Ω ar at around 15.At this level of aragonite supersaturation, crystal growth slows enough to cause the rate of skeletal formation to slow down as well.When the Ω ar of the external seawater is ~ 1, the coral works to maintain the calcifying fluid Ω ar at ~ 7.At this level of aragonite supersaturation, the rate of crystal growth slows dramatically and building a normal skeleton becomes almost impossible.When reared in strongly undersaturated seawater, the coral is able to maintain the fluid saturation state just above that at which its minute skeleton will dissolve (Ω ar ~ 2).Thus, coral calcification declines with declining seawater saturation state because the saturation state of the internal calcifying fluid declines in concert with it.And, as the saturation state of the internal calcifying fluid declines, the rate of CaCO 3 crystal growth becomes too slow to sustain growth of a normal skeleton.WhY dON't cOr alS SiMplY puMp MOre prOtONS?If corals grow aragonite from seawater by pumping protons to raise the saturation state within a calcifying compartment, why don't corals reared under elevated CO 2 (ocean acidification) conditions simply pump more protons in an effort to build a normal skeleton?After all, under elevated CO 2 , there is even more carbon around, in the form of bicarbonate.And, removing protons from the calcifying fluid would turn that excess bicarbonate into even more carbonate

Figure 5 .
Figure 5. in (a) aragonite crystals accreted by corals (shown in Figure 4) reared in seawater with a range of aragonite saturation states, display systematic and progressive changes in crystal morphology (a-d).in(e), these changes can be quantified using the crystal aspect ratio (the ratio of crystal length to width) and used to estimate changes in the saturation state of the coral's internal calcifying fluid in response to ocean acidification.Scale bars are 1 µm.FromCohen et al., 2009 photosynthesis without reducing the total amount of carbon available for calcification.Increased photosynthesis means increased photosynthate and more energy for calcification.Therefore, combining elevated nutrients with elevated CO 2 could help to offset the negative impact on calcification of elevated CO 2 alone.In at least four separate experiments conducted to date, five different species of coral were reared under significantly elevated CO 2 conditions (780-1200 ppm, Ω ~ 1.5-2) with the simultaneous addition of food or inorganic nutrients.Calcification rates under these

Figure 6
Figure6.calcification achieved by removing protons from the internal calcifying fluid is energetically expensive.The energy expended in calcification is represented here by the number of protons pumped from each milliliter of calcifying fluid.corals reared in seawater with a range of aragonite saturation states (depicted by different colors) remove the same number of protons from that seawater (4500 nmol ml -1 ) but achieve different calcifying saturation states (shown by grey bar).This suggests that corals reared in acidified seawater did not divert more energy to proton removal even at the expense of building a normal skeleton.in this cO 2 Sys calculation, the initial calcifying fluid is seawater with an initial alkalinity of 2470 µmol kg -1 in equilibrium with the specified pcO 2 (400, 740, or 1340 µatm).The fluid is isolated from the surrounding seawater; proton pumping elevates the alkalinity and saturation state within the calcifying space while cO 2 diffuses through the calicoblastic epithelium to maintain equilibrium pcO 2 .calculations were made with a Matlab implementation of cO 2 Sys using constants ofMehrbach et al. (1973) refit by dickson  and Millero (1987)  for carbonate, and dickson (1990) for sulfate; input conditions are as follows: S = 30, t = 25, atmospheric pressure = 1atm.concentrations of silicate, phosphate, ammonia and hS were set at 0. ca and B concentrations were calculated from salinity.