The sources of dissolved inorganic carbon (DIC) used to produce scleractinian coral skeletons are not understood. Yet this knowledge is essential for understanding coral biomineralization and assessing the potential impacts of ocean acidification on coral reefs. Here we use skeletal boron geochemistry to reconstruct the DIC chemistry of the fluid used for coral calcification. We show that corals concentrate DIC at the calcification site substantially above seawater values and that bicarbonate contributes a significant amount of the DIC pool used to build the skeleton. Corals actively increase the pH of the calcification fluid, decreasing the proportion of DIC present as CO 2 and creating a diffusion gradient favouring the transport of molecular CO 2 from the overlying coral tissue into the calcification site. Coupling the increases in calcification fluid pH and [DIC] yields high calcification fluid [CO 3 2 À ] and induces high aragonite saturation states, favourable to the precipitation of the skeleton.

No abstract

[1] We used ion microprobe analysis to determine the Sr/Ca composition of fasciculi (deposited during the day) and centers of calcification (COCs, deposited at night) across transects of two Porites lobata corals, of different linear extension rates, from Oahu, Hawaii. The COCs of both corals contained significantly higher Sr/Ca than the fasciculi (at the 95% confidence level). We observed no significant differences between the Sr/Ca ratios of the fasciculi (or of the COCs) of the fast and slow growing corals, and we conclude that variations in the extension rate of each colony have not affected Sr incorporation in these corals. The fasciculi and COCs in both corals exhibit large Sr/Ca heterogeneity, which is not temperature dependent. Our data do not support the hypothesis that COC analyses provide a SST signature which is unaffected by biological or kinetic effects. The heterogeneity of both features may reflect short-term (daily to weekly) variations in calcification rate which are known to occur during the day and night and which influence the relative transport rates of Sr and Ca through the coral tissue. We plotted running means through the fasciculi and COC chronologies and compared maximum and minimum Sr/Ca values in each annual cycle with the corresponding minimum and maximum mean sea surface temperature (SST) values (calculated over equivalent time periods). We found that the constants C and M of the linear equation Sr/Ca = C + (M Â SST) became smaller as the time interval used to calculate the running means increased from 1 day to $77 days. This decrease in C and M reflects the gradual removal of the short-term Sr heterogeneity (dependent on biological and/or kinetic processes) from the data set as progressively larger numbers of data points are used to calculate the Sr/Ca running mean. We hypothesize that variations in M and C between different published Sr/Ca-SST calibrations may reflect the relative importance of biological or kinetic processes in corals at different locations.

No abstract

No abstract