Ocean Acidification Research Paper

Abstract

Ocean acidification has been identified as a risk to marine ecosystems, and substantial scientific effort has been expended on investigating its effects, mostly in laboratory manipulation experiments. However, performing these manipulations correctly can be logistically difficult, and correctly designing experiments is complex, in part because of the rigorous requirements for manipulating and monitoring seawater carbonate chemistry. To assess the use of appropriate experimental design in ocean acidification research, 465 studies published between 1993 and 2014 were surveyed, focusing on the methods used to replicate experimental units. The proportion of studies that had interdependent or non-randomly interspersed treatment replicates, or did not report sufficient methodological details was 95%. Furthermore, 21% of studies did not provide any details of experimental design, 17% of studies otherwise segregated all the replicates for one treatment in one space, 15% of studies replicated CO2 treatments in a way that made replicates more interdependent within treatments than between treatments, and 13% of studies did not report if replicates of all treatments were randomly interspersed. As a consequence, the number of experimental units used per treatment in studies was low (mean = 2.0). In a comparable analysis, there was a significant decrease in the number of published studies that employed inappropriate chemical methods of manipulating seawater (i.e. acid–base only additions) from 21 to 3%, following the release of the “Guide to best practices for ocean acidification research and data reporting” in 2010; however, no such increase in the use of appropriate replication and experimental design was observed after 2010. We provide guidelines on how to design ocean acidification laboratory experiments that incorporate the rigorous requirements for monitoring and measuring carbonate chemistry with a level of replication that increases the chances of accurate detection of biological responses to ocean acidification.

Introduction

Ocean acidification is a potential threat to marine ecosystems through its effect on the physiology and ecology of many marine species (e.g. Kroeker et al., 2013b). As the research effort being invested into examining the effects of ocean acidification on marine systems increases, publications on the topic are increasing exponentially (Gattuso et al., 2012; Riebesell and Gattuso, 2015). Experimental manipulations of CO2 concentrations in the field are difficult, and the number of field studies are limited to a few locales where high CO2 naturally occurs, or where large scale, costly, and labour intensive experiments have been employed (Barry et al., 2010; Gattuso et al., 2014). Consequently, the most studies are conducted in the laboratory (see reviews by Wernberg et al., 2012; Kroeker et al., 2013b), where CO2 concentrations can potentially be controlled and reported accurately, and their effects isolated from those of other environmental variables. A multidisciplinary field of research such as ocean acidification brings together various research expertise including engineering sophisticated systems to manipulate and monitor seawater carbonate chemistry, building systems that can house organisms for long periods, and expertise in measuring the appropriate physiological/biogeochemical/ecological responses to ocean acidification. Crucially, if the field of ocean acidification is to progress, these manipulation experiments need to be performed in such a way that clear conclusions can be drawn from the results.

One of the major challenges for ocean acidification research is that seawater carbonate chemistry must be manipulated correctly in order for experimental treatments to approximately simulate future high CO2 oceans. Outwardly, this seems rather simple: seawater pH is manipulated using CO2 gas or the chemically equivalent method of HCl and dissolved inorganic carbon (DIC; usually in the form of NaHCO3 or Na2CO3), in a way that mimics changes in the future seawater, increasing in CO2 concentrations and declines in pH. That is, total alkalinity (AT) remains constant and DIC increases (Rost et al., 2008; Hurd et al., 2009; Gattuso et al., 2010).

Other logistical constraints, however, make ocean acidification experiments more difficult than those in other related fields of biological research. Experimental tanks often need to be sealed because CO2 can degas when exposed to the atmosphere, and the tanks require sufficient rates of seawater flow-through because the metabolic activity of organisms can modify seawater DIC and AT (Rost et al., 2008). The constant addition of chemicals (i.e. CO2 or HCl/DIC) into experimental tanks over long periods of time, without directly exposing organisms to un-mixed chemicals and seawater, adds another aspect of logistic difficulty to ocean acidification research that is not present in many other manipulation experiments (Hurd et al., 2009; Riebesell et al., 2010b). Appropriate methods must also be employed to determine whether the seawater that organisms are exposed to represents the desired treatments required for ocean acidification research; at least two components of the seawater carbonate system must also be measured (pH, AT, DIC, or pCO2). If pH is used to parametrize the seawater carbonate system, then it must be measured using spectrophotometers or electrodes calibrated using TRIS buffers, not with electrodes calibrated using NBS buffers (Dickson et al., 2007; Dickson, 2010). Also, to eliminate or reduce the chance of experimental artefacts or pre-existing gradients in other factors influencing treatments differently, treatments within manipulation experiments must all contain adequate numbers of randomly interspersed and independent treatment replicates (Hurlbert, 1984; Hurlbert and White, 1993; Hurd et al., 2009; Hurlbert, 2009; Wernberg et al., 2012). To complicate matters, ocean acidification will not be occurring in isolation, as other anthropogenic effects such as global warming and localized altered salinity, nutrients (nitrogen, phosphorous), light regimes, storm occurrence, and land-born pollutants will be occurring in synergy (Feely et al., 2004; Boyd, 2011; Ciais et al., 2013). Factorial designs where CO2 is manipulated in combination with other factors therefore add further complexity and logistical challenges to experimental designs (Havenhand et al., 2010; Wernberg et al., 2012).

The inappropriate assignment of experimental units can be a problem in any form of research, and Hurlbert (1984) defines this inappropriate assignment of experimental units for a given treatment during statistical analyses as “pseudoreplication”. Hurlbert defines the appropriate procedures for eliminating the risk of pseudoreplication by randomly interspersing replicates of different treatments with each other, and by removing any interdependence within replicates from the same treatment, i.e. experimental units are randomly interspersed replicates of a treatment. These procedures detailed by Hurlbert (1984) will not be repeated here in detail, but solutions to common problems in ocean acidification manipulation experiments will be mentioned in Discussion. Hurlbert (2013a) also defines experimental units as: “the smallest … unit of experimental material to which a single treatment (or treatment combination) is assigned by the experimenter and is dealt with independently … ”, and defines independent experimental units as being units assigned to the same treatment that will not be subject to conditions that are more similar than conditions that are imposed on units from another treatment, other than the treatment under investigation (Cox, 1958; Kozlov and Hurlbert, 2006; Hurlbert, 2013a). The experimental unit can be constrained by two principles: (i) experimental units within one treatment must not influence each other more than they influence experimental units within another treatment and (ii) factors other than the treatment in question (e.g. seawater source, light, etc.) must, on average, be equal across all treatments (Hurlbert, 2013a). If these principles are adhered to, it will greatly reduce the risk of non-treatment effects differentially influencing one treatment and not others; these should be the basic tenets of experimental design. Regardless of the degree of precision that the treatment is applied and its effects measured, if treatment effects are confused with the effects of other factors not under investigation, then an accurate assessment of the effects of the treatment cannot be made. Hurlbert and White (1993) further define three types of pseudoreplication: (i) simple pseudoreplication, where there is one experimental unit per treatment and multiple individuals in one experimental unit whose responses to the treatment are measured (defined by Hurlbert, 2009 as the “evaluation unit”) and treated as though they are independent experimental units; (ii) temporal pseudoreplication, where multiple measurements are made though time on the same experimental unit and treated as independent experimental units of a treatment; (iii) sacrificial pseudoreplication, where there are multiple experimental units per treatment and multiple individuals within each experimental unit, but the individuals are treated as the experimental units during statistical analysis. These three definitions demonstrate how a misinterpretation as to what constitutes an experimental unit vs. an “evaluation unit” could lead to inappropriate design and analysis.

The field of ocean acidification research is rapidly expanding, and so far the scientific community has revealed information crucial to understanding its future impacts at an impressive rate unprecedented in many other fields of research (Riebesell and Gattuso, 2015). However, if the field is to progress, then we need to maximize the information provided by each future study. The purpose of this study is not to exhaustively detail how to correctly replicate experiments of all types; these have already been fully explained previously (Cox, 1958; Hurlbert, 1984; Mead, 1988). Nor is it our goal to undermine previously conducted research, which has significantly advanced understanding of the potential effects of ocean acidification on biological systems (Riebesell and Gattuso, 2015). The objectives of this study is to highlight how well the basic principles for the design and analysis of experiments are followed (e.g. Cox, 1958; Hurlbert, 1984; Mead, 1988; Hurd et al., 2009; Havenhand et al., 2010; Wernberg et al., 2012), and to highlight how the mistreatment of experimental units (not just during statistical analysis) hinders the ability to accurately predict the effects of ocean acidification in certain circumstances. Importantly, it was our goal to provide solutions to many commonly encountered problems in experimental design. The appropriate procedures to replicate experimental units in ocean acidification manipulation experiments are identified, and the prevalence of these appropriate methodological approaches is quantified. As well as presenting potential pitfalls in experimental design, a range of solutions to logistical limitations that are imposed by different CO2-manipulation system designs are provided, to increase the inference that can be drawn from future manipulation experiments.

Methods

To measure the prevalence of different designs in ocean acidification manipulation experiments, we searched the database Scopus http://www.scopus.com/ using the term “ocean acidification”. We cross checked this search with the database used by Kroeker et al. (2013b) and the ocean acidification blog http://news-oceanacidification-icc.org/. Four hundred and sixty-five studies published between 1993 and February 2014 were examined, along with any others that came across our desk thereafter (Supplementary Table SI). Sixty-two of these studies were also analysed by Wernberg et al. (2012). Studies were only analysed if they met specific criteria. That is, the studies had to report: (i) the results of research containing the assessment of biological responses to experimental manipulations of seawater CO2; (ii) state that their results were directly applicable to predicting the impacts of ocean acidification, increases in seawater CO2 concentrations predicted for the future, or another analogous term. Studies that were investigating the impacts of carbon sequestration in the seabed, and made no mention of future high CO2 seawater, ocean acidification, etc., were excluded. (iii) They needed to be conducted for longer than 24 h, in other words, we excluded short-term “response-assays”. (iv) They had to be published in peer reviewed journals to be included. We excluded field-based correlative surveys due to the logistical limitations imposed on them by lack of availability of multiple “treatment” sites in some locations, and because treatments were not “manipulated” by the researchers usually. This is not to say that we viewed these studies as less worthy than laboratory studies.

After papers were identified that met the above criteria, their experimental design in respect to replication was examined. Specific designs were grouped as per the scheme outlined in Figure 1 in Hurlbert (1984) which we have re-constructed in the context of ocean acidification research (Figure 1), showing potential aquaria or “experimental tank” arrays. Hulbert considers designs whose letters start with an “A” are an appropriate design choice, while those with “B” are an inappropriate choice. They are A-1 completely randomized, A-2 randomized block, A-3 systematic, B-1 simple segregation, B-2 clumped segregation, B-3 isolative segregation, B-4 randomized but with interdependent treatment replicates, and B-5 no replication. In each study, the number of actual experimental units per treatment was also recorded, rather than those stated by the authors. In the context of ocean acidification research, we provide the following examples for each of the B type designs (Figure 1): (B-1) three different pCO2 treatments are used and experimental tanks are lined up on a bench in order from lowest to highest pCO2; (B-2) three different shaker tables are used next to each other, and each table contains the replicates of only one treatment; (B-3) three different experimental tanks are used, and each experimental tank contains all the individuals of one treatment only; (B-4) three different header tanks are used to supply seawater for an entire experiment, and one header tank provides seawater for all the replicates of one treatment only; (B-5) one individual per treatment is placed in one tank, and only one tank is used for each treatment of an experiment. In all these examples, pre-existing gradients in other factors or chance events could obscure the effects of the factor under investigation.

Figure 1.

Different designs of tank arrays, modified and re-drawn from Hurlbert (1984) in the context of ocean acidification research. Different coloured tanks correspond to different CO2 treatments. Design types preceded with an A are acceptable, while those preceded by a B are unacceptable ways to replicate experimental units of a treatment according to Hulbert. (A-1) Completely randomized. (A-2) Randomized block design. (A-3) Systematic. (B-1) Simple segregation. (B-2) Clumped segregation. (B-3) Isolative segregation. (B-4) Randomized but all replicates of one treatment interdependent with themselves more than other treatments (e.g. one header tank of seawater per treatment). (B-5) No replication. An example of replicates within treatments that are interdependent are treatment replicates that all share a common header tank that is not shared with replicates of other treatments large white boxes denote tanks of seawater that do not contain the organisms that are housed. Smaller boxes denote tanks that the organisms are housed in, with different colours representing different pCO2 levels. n = number of studies using this design type in ocean acidification research. See results and methods for more details.

Figure 1.

Different designs of tank arrays, modified and re-drawn from Hurlbert (1984) in the context of ocean acidification research. Different coloured tanks correspond to different CO2 treatments. Design types preceded with an A are acceptable, while those preceded by a B are unacceptable ways to replicate experimental units of a treatment according to Hulbert. (A-1) Completely randomized. (A-2) Randomized block design. (A-3) Systematic. (B-1) Simple segregation. (B-2) Clumped segregation. (B-3) Isolative segregation. (B-4) Randomized but all replicates of one treatment interdependent with themselves more than other treatments (e.g. one header tank of seawater per treatment). (B-5) No replication. An example of replicates within treatments that are interdependent are treatment replicates that all share a common header tank that is not shared with replicates of other treatments large white boxes denote tanks of seawater that do not contain the organisms that are housed. Smaller boxes denote tanks that the organisms are housed in, with different colours representing different pCO2 levels. n = number of studies using this design type in ocean acidification research. See results and methods for more details.

In our survey, we assumed that cylinders/containers of CO2 gas, HCl, or a form of DIC is not likely to be a source of experimental artefact. In other words, we considered that using one source of chemicals to modify carbonate chemistry, and applying it to seawater in all elevated CO2 treatments would not result in a lack of independence, as we considered that the likelihood of a contaminant such as a type of micro-organism (fungi, bacteria, diatoms, etc.) contaminating that container and changing the treatment was extremely low compared with the likelihood of micro-organisms living in tanks of seawater. This could be an issue, but was outside of the bounds of our analysis, as it is difficult to expect studies to report the number of containers of chemicals used.

The “Guide to Best Practices for Ocean Acidification Research and Data Reporting” (Dickson, 2010; Gattuso et al., 2010; Riebesell et al., 2010a) provided rationale and methods detailing how to manipulate seawater carbonate chemistry in a way that most accurately mimics the changes predicted for a future ocean, and guidelines on the appropriate methods to characterize seawater carbonate chemistry. To determine the impact of this publication on the frequency of studies that used appropriate methods to manipulate and monitor seawater carbonate chemistry, we examined the same 465 papers to determine if they altered seawater chemistry in a way that simulates future ocean acidification (using methods that should increase DIC and keep total alkalinity (AT) constant) or not (methods resulting in constant DIC and decreased AT). In other words, we examined whether HCl was used by itself to reduce seawater pH without the addition of DIC. Two components of the seawater carbonate system must be measured (pH, AT, DIC, or pCO2) along with temperature and salinity to appropriately measure seawater carbonate chemistry, and if pH is measured it should be measured on the total scale: involving using TRIS buffers to calibrate electrodes, or using spectrophotometric measurements (see Dickson et al., 2007; Dickson, 2010 for complete guidelines). We also recorded the number of papers that measured two or more of seawater pHT, AT, DIC, or pCO2.

Statistical analysis

We analysed whether or not there was a difference pre- and post-2010 (the year of the publication of The Guide to Best Practices in Ocean Acidification and Data Reporting) between the proportion of studies that possibly used inappropriate methods: (i) to manipulate seawater carbonate chemistry; (ii) to measure two or more of pHT, AT, DIC, or pCO2; and (iii) to design/analyse experimental units. This was done using a z-test to examine differences between two proportions.

Definitions of tank types

In ocean acidification experiments, different containers (referred to here as tanks) have different purposes, and for clarity we assign names to each tank type. All manipulation experiments must have at least one of these tanks types, but not all laboratory manipulation experiments will use all types, and designs could incorporate tanks that have dual purposes. They are as follows: (i) storage tank: location where seawater is stored before being altered to create CO2 treatments. (ii) Mixing tank: location where chemicals and seawater are mixed together to create CO2

Abstract

Ocean acidification is likely to have direct negative physiological consequences for many marine organisms, and cause indirect effects on marine ecosystems. Ocean acidification could also affect the oceans' current role as a net carbon sink by altering the oceanic calcium carbonate budget. Although ocean acidification and climate change are both caused by greenhouse gas emissions, ocean acidification is not climate change per se, and is often referred to as “the other carbon dioxide (CO2) problem.” As the United States considers actions in response to climate change, it is critical to take into account not only the impact of CO2 emissions on the climate but also their ramifications for ocean chemistry. The metrics that currently guide the climate change debate are dominated by strategies to reduce thermal impacts on the terrestrial environment. In this article, I examine the effects of ocean acidification and why they should help guide decisionmakers in setting CO2 emissions goals.

The oceans have absorbed approximately 30% of the atmospheric carbon dioxide (CO2) emitted by humankind since the beginning of the Industrial Revolution, and they serve as an important net carbon sink (Sabine et al. 2004). Although oceans, as carbon sinks, reduce the rate of CO2 increase in the atmosphere, the absorption process has a direct and measurable impact on ocean chemistry. Since the late 18th century, the mean surface ocean pH has dropped 0.1 units from 8.2 to 8.1. Because pH is measured on a logarithmic scale (a unit change of 1.0 is equal to a tenfold change in concentration), this change is roughly equivalent to a 30% increase in the concentration of hydrogen ions (Raven et al. 2005). Unlike changes in temperature due to global warming, which are difficult to predict, the mean magnitude and rate of ocean acidification can be projected with high confidence under different CO2 emissions scenarios, derived from a series of highly predictable chemical reactions (Caldeira and Wickett 2003, Raven et al. 2005). Under the Intergovernmental Panel on Climate Change (IPCC) business-as-usual (BAU) IS92a scenario, in which atmospheric concentrations of CO2 are expected to reach 800 parts per million by volume (PPMV) by 2100 (current atmospheric carbon is around 384 PPMV), the mean surface ocean pH is projected to drop another 0.3 to 0.4 units (Orr et al. 2005)—a 150% increase in the concentration of hydrogen ions.

When CO2 dissolves in seawater, it generates carbonic acid (H2CO3), which breaks down into bicarbonate (HCO3), carbonate (CO32−), and hydrogen ions (H+; figure 1). These reactions are reversible and near equilibrium (Millero et al. 2002). Their directionality is determined by ion concentrations (including noncarbon ions), temperature, salinity, and pressure. Under the oceans' current physical and biological conditions, carbonate and bicarbonate ions act as a buffer by absorbing and storing excess carbonic acid. This process buffers seawater against larger changes in pH. Under high-CO2 conditions, the reactions forming hydrogen ions and bicarbonate are favored, and the result is lowered pH and fewer carbonate ions (figure 2).

Figure 1.

Seawater carbonate chemistry equations. Carbon dioxide dissolves into seawater from the atmosphere and generates carbonic acid (H2CO3), which breaks down into bicarbonate (HCO3), carbonate (CO32−), and hydrogen ions (H+). When protons combine with carbonate ions to form bicarbonate, the concentration of carbonate decreases, making it unavailable to marine calcifiers to form calcium carbonate (CaCO3). All of these reactions are reversible, and directionality depends on concentration, temperature, salinity, and pressure. Source: Adapted from Hoegh-Guldberg et al. 2007.

Figure 1.

Seawater carbonate chemistry equations. Carbon dioxide dissolves into seawater from the atmosphere and generates carbonic acid (H2CO3), which breaks down into bicarbonate (HCO3), carbonate (CO32−), and hydrogen ions (H+). When protons combine with carbonate ions to form bicarbonate, the concentration of carbonate decreases, making it unavailable to marine calcifiers to form calcium carbonate (CaCO3). All of these reactions are reversible, and directionality depends on concentration, temperature, salinity, and pressure. Source: Adapted from Hoegh-Guldberg et al. 2007.

Figure 2.

Bjerrum plot. The relative proportions of carbonate species concentrations (in moles per kilogram) vary as a function of pH. Depicted here is a hypothetical shift in mean surface water pH from 8.2 (preindustrial) to 7.75, the value at which calcium carbonate becomes undersaturated with respect to aragonite. Under high-CO2 conditions, the reactions forming hydrogen (H+) and bicarbonate (HCO3) are favored, and the result is fewer carbonate ions (CO32+).

Figure 2.

Bjerrum plot. The relative proportions of carbonate species concentrations (in moles per kilogram) vary as a function of pH. Depicted here is a hypothetical shift in mean surface water pH from 8.2 (preindustrial) to 7.75, the value at which calcium carbonate becomes undersaturated with respect to aragonite. Under high-CO2 conditions, the reactions forming hydrogen (H+) and bicarbonate (HCO3) are favored, and the result is fewer carbonate ions (CO32+).

A paleoclimatic event relevant to the current ocean situation occurred 55 million years ago when the atmosphere had higher concentrations of greenhouse gases and a higher mean global temperature. It has been hypothesized that there was an input of CO2 to the deep ocean, presumably the result of a massive volcanic methane release of about 2000 gigatons of carbon over approximately 10,000 years (Zachos et al. 2005, 2008). In this scenario, methane would have been rapidly oxidized into CO2, lowering the pH and carbonate ion concentration in the deep sea. (For reference, since the beginning of the Industrial Revolution, humans will have released 5000 gigatons by 2400 under the IS92a IPCC BAU scenario [Caldeira and Wickett 2003]). The fossil record indicates that this massive methane-release event was followed by the extinction of several bottom-dwelling foraminifera species, single-celled organisms with calcareous shells (Zachos et al. 2005). Although knowledge of this ancient event may help us to predict what might happen to ocean pH and biota under contemporary conditions (Zachos et al. 2005), today's world is much different. Current rates of ocean acidification are faster, preindustrial levels of CO2 and temperature were lower, and a vastly different marine biota occupies the oceans today (Doney et al. 2009).

Impacts of ocean acidification on marine biota

Changes in ocean chemistry will probably affect marine life in three different ways: (1) decreased carbonate ion concentration could affect the calcification process for calcifying organisms (e.g., corals); (2) lowered pH could affect acid-base regulation, as well as a variety of other physiological processes; and (3) increased dissolved CO2 could alter the ability of primary producers to photosynthesize. Most of the research in the field has focused on calcification effects. In this article, I focus on how the concentration of carbonate ions affects calcification and dissolution, and then summarize what is known about other physiological effects, including those relating to photosynthesis. Following these species-specific responses, I provide an overview of theoretical community and ecosystem effects that might be expected into the future.

The importance of the carbonate ion for calcification. Many organisms use calcium and carbonate ions from seawater to produce calcium carbonate (figure 1), a compound used for skeletal support (e.g., corals) and protection (e.g., snail shells). The three mineral forms of calcium carbonate commonly produced are aragonite, calcite, and high-magnesium calcite (Raven et al. 2005). Aragonite calcifiers (e.g., stony corals and shelled pteropods) and high-magnesium calcite calcifiers (e.g., coralline algae and sea urchins) are likely to be affected by ocean acidification more strongly than calcite calcifiers (e.g., foraminifera and coccolithophores). This distinction is because of differences in the solubility of mineral forms; for example, aragonite is approximately 50% more soluble than calcite (Doney et al. 2009).

The saturation state, denoted by the Greek letter Ω, refers to the degree to which seawater is saturated with a carbonate mineral (i.e., aragonite, calcite, or high-magnesium calcite), and is inversely related to the mineral's solubility. The saturation state is determined by the concentrations of calcium and carbonate ions in relation to the solubility coefficient (K'sp) for the particular calcium carbonate mineral (Ω = [Ca2+][CO32−]/K'sp). The solubility coefficient varies with temperature, salinity, and pressure. Calcium carbonate solubility rises with decreasing temperature and increasing pressure and therefore increases with ocean depth. Since the saturation state and the solubility coefficient are inversely related for a given ion concentration, the saturation state is highest in shallow, warm tropical waters and lowest in deep and cold high-latitude waters (Feely et al. 2004).

When the saturation state is equal to 1 (Ω = 1), there is an equal chance of dissolution or formation of calcium carbonate. This chemical threshold defines a two-dimensional surface in the ocean interior called the saturation horizon. When saturation is greater than 1 (Ω > 1), formation of calcium carbonate is favored; when saturation is less than 1 (Ω < 1), dissolution is favored. The saturation horizon normally occurs at an ocean depth separating shallow, warm surface waters from deeper, cooler water. However, ocean acidification is decreasing the concentration of carbonate ions and therefore decreasing the saturation state of calcium carbonate, thus bringing the saturation horizon closer to the surface (Raven et al. 2005). Because of the higher solubility of aragonite and high-magnesium calcite, the saturation horizons for these mineral forms are even shallower than for calcite. Saturation horizons also vary in depth between ocean basins; aragonite and calcite saturation horizons are shallower in the Indian and Pacific oceans than in the Atlantic Ocean because of their longer deep-water circulation pathways, resulting in a greater accumulation of biologically respired CO2 (Broecker 2003).

For many organisms, a decrease in the calcium carbonate saturation state has been correlated with a reduction in calcification rates, even when waters remain supersaturated (Ω > 1) with respect to aragonite and calcite (Kleypas et al. 2005, Fabry et al. 2008). Under laboratory conditions, several calcifying organisms, including abundant planktonic species (e.g., coccolithophores, pteropods, foraminifera) and benthic invertebrates (e.g., coral, calcifying algae, molluscs, echinoderms), have shown a reduction in calcification rates as a result of reduced carbonate ion concentration (Raven et al. 2005, Fabry et al. 2008, Guinotte and Fabry 2008). However, the impact of lower carbonate concentration on calcification rates varies among species, and there is evidence that acidification may even enhance calcification in some taxa (Iglesias-Rodriguez et al. 2008, Ries et al. 2008, Wood et al. 2008).

A decrease in calcium carbonate saturation can also increase the dissolution rate of existing exposed and unprotected (i.e., by organic coatings) calcium carbonate structures (Jokiel et al. 2008, Andersson et al. 2009). As saturation horizons become shallower, more organisms will inhabit undersaturated waters (Ω < 1) where unprotected calcium carbonate structures will dissolve. The rate and extent of this process varies across the oceans because the solubility coefficient depends in part on temperature, pressure, and salinity (Raven et al. 2005).

Other physiological effects. Other potential negative physiological effects of ocean acidification include impairment of acid-base regulation, reproduction (sea urchins), respiration (squid, crabs), metabolism (mussels, worms), and behavior, as well as increased mortality (fish, sea urchins, krill; reviewed in Fabry et al. 2008, Pörtner 2008). The investigation of effects of ocean acidification on marine organisms is fairly recent, and only a limited number of studies address them (reviewed in Raven et al. 2005, Fabry et al. 2008, Doney et al. 2009). Even closely related species may respond differently to acidification: For example, sister species of sea urchins differ in their early life-history-stage responses to ocean acidification (reviewed in Dupont et al. 2010). This example highlights the infancy of the research in this field and our inability to generalize physiological effects. The diversity of species-specific positive and negative physiological effects in response to ocean acidification has been documented in recent reviews (table 1).

One thought on “Ocean Acidification Research Paper

Leave a Reply

Your email address will not be published. Required fields are marked *