Introduction
Production of aragonite shells by many marine organisms via calcification requires adequate availability of seawater carbonate ions (CO32-), which is often represented by the aragonite saturation state of seawater: ![--><$>\rOmega \, = \,\frac{{\left[ {C{{a}^{2 + }} } \right]\left[ {CO_{3}^{{\ \,2{\rm{ - }}}} } \right]}}{\lambda }<$><!--](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20160203064504941-0217:S0954102011000204_eqnU1.gif?pub-status=live){kind=small}
, where λ is the solubility coefficient of calcium carbonate in the form of aragonite. Changes in ΩARAG are explicitly linked to rises and falls in oceanic partial pressure of CO2 (pCO2) through alterations in equilibrium CO2 chemistry in seawater. The coastal Southern Ocean ecosystem in particular, is most vulnerable to future declining ΩARAG via anthropogenic CO2 uptake (Orr et al. Reference Orr, Fabry, Aumont, Bopp, Doney, Feely, Gnanadesikan, Gruber, Ishida, Joos, Key, Lindsay, Maier-Reimer, Matear, Monfray, Mouchet, Najjar, Plattner, Rodgers, Sabine, Sarmiento, Schlitzer, Slater, Totterdell, Weirig, Yamanaka and Yool2005, McNeil & Matear Reference McNeil and Matear2008) since pteropods (aragonite secreting zooplankton) make up a large proportion of its biomass (Accornero et al. Reference Accornero, Manno, Esposito and Gambi2003, Hunt et al. Reference Hunt, Pakhomov, Hosie, Siegel, Ward and Bernard2008). Spatio-temporal variability in surface ocean ΩARAG is scarce for coastal Antarctic waters. Here, we report data on seasonal variability of ΩARAG from two independent Antarctic sites (Ross Sea and Prydz Bay, see Fig. 1).
Fig. 1 Geographic location of two Antarctic coastal sites (Ross Sea and Prydz Bay), along with the surface observations over the annual cycle for aragonite saturation state (ΩARAG).
Methods
We calculated ΩARAG using CO2 dissociation constants (Dickson & Millero Reference Dickson and Millero1987) from two reported pCO2 datasets in the Ross Sea and Prydz Bay, East Antarctica. The Ross Sea CO2 measurements were made during intermittent/sea ice-free periods in 1996–97 as described elsewhere (Sweeney Reference Sweeney2003). The Prydz Bay CO2 measurements were collected one kilometre offshore from Australia's Davis Station in Prydz Bay, East Antarctica from 1993–95 as described elsewhere (Gibson & Trull Reference Gibson and Trull1999). During ice-free periods (December–March), samples were collected at the surface while during ice cover (March–November), samples were collected at 2–20 m via drill hole.
Results
Surface ocean CO2 concentrations over an annual cycle undergo extreme changes at both the Ross Sea and Prydz Bay sites, with pCO2 rapidly increasing from c. 100 μatm up to c. 450 μatm from summer to autumn. This more than fourfold increase in CO2 drives a rapid decline in ΩARAG from up to 4.5 in summer to c. 1.1 in autumn (Fig. 1). Geochemical aragonite dissolution occurs when ΩARAG < 1, which is nearly reached during October in the Ross Sea (Fig. 1). Winter data is not available, but it is probably similar to autumn conditions (Sweeney Reference Sweeney2003).
Discussion
We put the natural seasonal variations observed for ΩARAG at these two sites (Fig. 1) into context against the decadal changes likely due to anthropogenic CO2 absorption. The average ΩARAG for both sites is c. 1.9 (Fig. 1) with a mean pCO2 of 246 μatm and total alkalinity of 2300 μmol kg-1. Using a mean temperature (c. -1.1°C) and salinity (33.4), we can calculate the upper-bound equilibrium changes in ΩARAG since pre-industrial times. Under mean conditions for the year 1995, ΩARAG is estimated to be 0.55 lower since pre-industrial times (the year 1780), which is an upper bound. In reality however these Antarctic surface waters are not in equilibrium with the atmosphere due to sea ice hindrance of air-sea gas exchange and dilution from older deep waters (McNeil & Matear Reference McNeil and Matear2008). Natural seasonal variations (c. 3.2) are nearly six times the magnitude of anthropogenic acidification (c. 0.55) since pre-industrial times (Fig. 1).
The large natural variations in coastal Antarctic ΩARAG suggest that aragonite secreting species such as pteropods must be able to endure extreme changes in ΩARAG. Alternatively pteropods may undergo large variations in biomass dynamics that are partially driven by the changes in ΩARAG shown here. Studies based on a range of different methodologies suggest the lifecycle of Antarctic relevant pteropods (Limacina helicina Phipps) are longer than one year, with some juvenile species undergoing development during the winter months (Hunt et al. Reference Hunt, Pakhomov, Hosie, Siegel, Ward and Bernard2008). If this longer life cycle is true for the most abundant Antarctic pteropod (Limacina helicina), our results may suggest a high resilience to ΩARAG variations given the dominance of this pteropod in sediment traps in the Ross Sea (Accornero et al. Reference Accornero, Manno, Esposito and Gambi2003).
Understanding calcification changes for a particular species in a high CO2 ocean has required laboratory experiments where ΩARAG levels are artificially decreased from 5 to 2 (Riebesell et al. Reference Riebesell, Zondervan, Rost, Tortell, Zeebe and Morel2000). The observed threefold changes in ΩARAG shown here over a few months are of the same order of magnitude as those conditions used in artificial laboratory CO2 experiments. This makes the Antarctic coastal oceans a perfect natural mesocosm to investigate resilience of Southern Ocean marine ecosystems to ocean acidification.
