Impacts of ocean acidification on marine fauna and ecosystem processes Victoria J. Fabry, Brad A. Seibel, Richard A. Feely, and James C. Orr Fabry, V. J., Seibel, B. A., Feely, R. A., and Orr, J. C. 2008. Impacts of ocean acidification on marine fauna and ecosystem processes. ��� ICES Journal of Marine Science, 65: 414���432. Oceanic uptake of anthropogenic carbon dioxide (CO2) is altering the seawater chemistry of the world���s oceans with consequences for marine biota. Elevated partial pressure of CO2 (pCO2) is causing the calcium carbonate saturation horizon to shoal in many regions, particularly in high latitudes and regions that intersect with pronounced hypoxic zones. The ability of marine animals, most impor- tantly pteropod molluscs, foraminifera, and some benthic invertebrates, to produce calcareous skeletal structures is directly affected by seawater CO2 chemistry. CO2 influences the physiology of marine organisms as well through acid-base imbalance and reduced oxygen transport capacity. The few studies at relevant pCO2 levels impede our ability to predict future impacts on foodweb dynamics and other ecosystem processes. Here we present new observations, review available data, and identify priorities for future research, based on regions, ecosystems, taxa, and physiological processes believed to be most vulnerable to ocean acidification. We conclude that ocean acidification and the synergistic impacts of other anthropogenic stressors provide great potential for widespread changes to marine ecosystems. Keywords: anthropogenic CO2, calcification, ecosystem impacts, hypercapnia, ocean acidification, physiological effects, zooplankton. Received 11 July 2007 accepted 14 February 2008 V. J. Fabry: Department of Biological Sciences, California State University San Marcos, San Marcos, CA 92096���0001, USA. B. A. Seibel: Department of Biological Sciences, University of Rhode Island, Kingston RI 02881, USA. R. A. Feely: Pacific Marine Environmental Laboratory, NOAA, Seattle, WA 98115���6349, USA. J. C. Orr: Marine Environmental Laboratories, International Atomic Energy Agency, Monaco MC-98000, Monaco. Correspondence to V. J. Fabry: tel: ��1 760 7504113 fax: ��1 760 7503440 e-mail: fabry@csusm.edu. Introduction Rising atmospheric carbon dioxide (CO2) concentration is causing global warming and ocean acidification (Caldeira and Wickett, 2003, 2005 Feely et al., 2004 Orr et al., 2005), which increasingly are recognized as important drivers of change in biological systems (Lovejoy and Hannah, 2005). For at least 650 000 years prior to the industrial revolution, atmospheric CO2 concentrations varied between 180 and 300 ppmv (Siegenthaler et al., 2005). As a result of human activity, today���s atmospheric CO2 concentration is 380 ppmv and currently is rising at a rate of 0.5% year21 (Forster et al., 2007), which is 100 times faster than any change during the past 650 000 years (Royal Society, 2005 Siegenthaler et al., 2005). Approximately one-third of the anthropogenic CO2 produced in the past 200 years has been taken up by the oceans (Sabine et al., 2004). The global ocean inven- tory of anthropogenic carbon was 118+19 Pg C in 2004 (Sabine et al., 2004), which can be adjusted upwards to 140 Pg C in 2005 based on Denman et al. (2007, Table 7.1). Without this ocean sink, the anthropogenic change in atmospheric CO2 concentration would be 55% higher than the observed change from 280 to 380 ppmv (Sabine et al., 2004). Although oceanic uptake of anthropogenic CO2 will lessen the extent of global warming, the direct effect of CO2 on ocean chemistry may affect marine biota profoundly. Elevated partial pressure of CO2 (pCO2) in seawater (also known as hypercapnia) can impact marine organisms both via decreased calcium carbonate (CaCO3) saturation, which affects calcification rates, and via disturbance to acid���base (metabolic) physiology. Recent work indicates that the oceanic uptake of anthropogenic CO2 and the concomitant changes in seawater chemistry have adverse consequences for many calcifying organ- isms, and may result in changes to biodiversity, trophic inter- actions, and other ecosystem processes (Royal Society, 2005 Kleypas et al., 2006). Most research has focused on tropical coral reefs and planktonic coccolithophores. Little information is avail- able for other important taxa, for processes other than calcifica- tion, or for potential ecosystem-level consequences emerging from the oceanic pCO2 levels that are predicted to occur over the next 100 years. Here we discuss the present and projected changes in ocean carbonate chemistry, and assess their impacts on pelagic and benthic marine fauna and ecosystem processes. We exclude corals from this discussion, but note that excellent recent reviews on this topic exist (Langdon and Atkinson, 2005 Guinotte et al., 2006 Kleypas and Langdon, 2006). We highlight many of the gaps in our knowledge and identify critical questions for future research. The ocean���s inorganic carbon system: present and future changes The CO2 ���carbonate system in seawater The inorganic carbon system is one of the most important chemi- cal equilibria in the ocean and is largely responsible for controlling the pH of seawater. Dissolved inorganic carbon (DIC) exists in # 2008 International Council for the Exploration of the Sea. Published by Oxford Journals. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org 414

seawater in three major forms: bicarbonate ion (HCO3 2), carbon- ate ion (CO32), 2 and aqueous carbon dioxide (CO2(aq)), which here also includes carbonic acid (H2CO3). At a pH of 8.2, 88% of the carbon is in the form of HCO3 2, 11% in the form of CO32, 2 and only 0.5% of the carbon is in the form of dissolved CO2. When CO2 dissolves in seawater, H2CO3 is formed (Figure 1). Most of the H2CO3 quickly dissociates into a hydrogen ion (H+) and HCO3 2. A hydrogen ion can then react with a CO32 2 to form bicarbonate. Therefore, the net effect of adding CO2 to sea- water is to increase the concentrations of H2CO3, HCO3 2, and H+, and decrease the concentration of CO32 2 and lower pH (pH = 2log[H+]). These reactions are fully reversible, and the basic thermodynamics of these reactions in seawater are well known (Millero et al., 2002). The atmospheric CO2 value today is 100 ppmv greater than the pre-industrial value (280 ppmv), and the average surface ocean pH has dropped by 0.1 unit, which is about a 30% increase in [H+]. Under the IPCC emission scenarios (Houghton et al., 2001), average surface ocean pH could decrease by 0.3���0.4 pH units from the pre-industrial values by the end of this century (Caldeira and Wickett, 2005 Figure 2). Present and future changes in carbonate saturation The reaction of CO2 with seawater reduces the availability of car- bonate ions that are necessary for marine calcifying organisms, such as corals, molluscs, echinoderms, and crustaceans, to produce their CaCO3 shells and skeletons. The extent to which such organisms are affected depends largely upon the CaCO3 sat- uration state (V), which is the product of the concentrations of Ca2+ and CO32, 2 divided by the apparent stoichiometric solubility product (Ksp*) for either aragonite or calcite, two types of CaCO3 commonly secreted by marine organisms: V �� ��Ca��2����CO3 2 ��=Ksp ��1�� where the calcium concentration is estimated from the salinity, and [CO32] 2 is calculated from DIC and total alkalinity (TA) measurements (Feely et al., 2004). Increasing CO2 concentrations in the atmosphere, and thus in the surface ocean, will continue to decrease the [CO32] 2 in the upper ocean, thereby lowering CaCO3 saturation levels by means of the reaction: CO2 �� CO3 2 �� H2O �� 2HCO3 : ��2�� In regions where Varag or Vcal is .1.0, the formation of shells and skeletons is favoured. For values ,1.0, seawater is corrosive to CaCO3 and, in the absence of protective mechanisms (e.g. Corliss and Honjo, 1981 Isaji, 1995), dissolution will begin. Saturation states are generally highest in the tropics and lowest in the high latitudes, because the solubility of CaCO3 increases with decreas- ing temperature and increasing pressure. Consequently, there is significant shoaling of the aragonite saturation horizons in the Pacific, north of 408N, at the equator, and 108N, especially towards the east, because of the higher DIC concentrations relative to TA at shallower depths. These patterns result from enhanced upwelling that brings deeper waters rich in nutrients and DIC to the upper ocean (Figure 3) and supports high animal biomass. As one moves north, the aragonite saturation depth shoals from 1000 m near 308S to 300 m at the equator. Moving farther north, it deepens to 550 m near 308N, then shoals to 100 m north of 508N (Figure 3). In the North Pacific, the upward migration of the aragonite saturation horizon from anthropogenic CO2 uptake is currently 1���2 m year21 (Feely et al., 2006). Orr et al. (2005) developed model scenarios of future changes in surface ocean carbonate chemistry as a function of changes in atmospheric CO2, using the IPCC IS92a ���business-as-usual��� CO2 emission scenario, with the median projection of DIC changes from 13 ocean models that participated in the OCMIP-2 project. Based on their model outputs and global gridded data (Key et al., 2004), we plotted the projected aragonite Figure 1. Concentrations of carbon species (in units of mmol kg21), pH values, and aragonite and calcite saturation states of average surface seawater for pCO2 concentrations (ppmv) during glacial, preindustrial, present day, two times pre-industrial CO2, and three times pre-industrial CO2. Changes in the inorganic carbon system were computed by assuming equilibrium with atmospheric CO2, assuming PO4 = 0.5 mmol l21 and Si = 4.8 mmol l21, and using the carbonic acid dissolution constants of Mehrbach et al. (1973) as refitted by Dickson and Millero (1987). pH is based on the seawater scale. The last column shows the changes from the pre-industrial levels to three times atmospheric CO2 (modified from Feely et al. (2004) and Kleypas et al. (2006)). Figure 2. Atmospheric CO2 concentration projected under the IS92a ���business-as-usual��� IS92a CO2 emissions scenario, bounded by the most and least conservative SRES scenarios, B1 and A1F1, respectively, and projected global average surface seawater pH (modified from Meehl et al. (2007)). Impacts of ocean acidification on marine fauna and ecosystem processes 415

saturation state of the surface oceans for the years 1765, 1994, 2050, and 2100 (Figure 4). The model results indicate that, by the time atmospheric CO2 reaches 780 ppmv near the end of this century under the IPCC IS92a ���business-as-usual��� CO2 emis- sion scenario, portions of the Subarctic North Pacific and all of the Southern Ocean south of 608S will become undersaturated with respect to aragonite (Orr et al., 2005). At that point, the global average surface water CO32 2 concentration and aragonite and calcite saturation state will be nearly half of what they are today. The aragonite saturation horizons would also shoal from its present average depth of 730 m to the surface in the Southern Ocean, from 2600 to 115 m in the North Atlantic, and from 140 m to the surface in parts of the North Pacific (Orr et al., 2005). In the cold, high-latitude surface waters typical of polar and subpolar regions of the Southern Ocean, aragonite and calcite undersaturation will occur when seawater pCO2 values reach 560 and 900 ppmv, respectively. In the slightly warmer surface waters of the subpolar North Pacific, aragonite and calcite undersaturation will occur later, when pCO2 reaches 740 and 1040 ppmv, respectively. The cold waters of the Arctic Ocean are also naturally low in CO32 2 concentration. Continuing research is evaluating how the Arctic Ocean���s changes in carbonate chemistry during the 21st century will differ from those in the Southern Ocean (Orr et al., 2006). The warm surface waters of the tropics and subtropics will not become undersaturated with respect to aragonite or calcite over the range of these projected conditions although, in some regions associated with upwelling, shoaling aragonite saturation horizons now impinge on the depth ranges of many pelagic animals (Feely et al., 2004). Priority areas for ocean acidification research are therefore high-latitude regions, which are projected to experience the greatest changes in carbonate chemistry over decadal to century Figure 3. Distribution of (a) aragonite saturation (b) partial pressure of CO2 seawater (pCO2) and (c) dissolved oxygen along the March 2006 P16 N transect along 1528W in the North Pacific. 416 V. J. Fabry et al.