Because Equation 2 is not important in the present-day ocean, the link between Mg flux into the ocean and CO2 drawdown at present is indirect. The formation of Mg-silicate secondary minerals has been observed to be coupled to Ca liberation originating from the dissolution of the Ca-rich minerals and glasses present in the mid-oceanic–ridge basalts (i.e., Mottle, 1983). This coupling between Mg precipitation and Ca silicate dissolution is likely via both 1) the production of protons, and 2) by the consumption of Al and Si during the Mg-silicate precipitation. The link between “Mg-clay” mineral precipitation and Ca liberation in the mid-ocean rifts is so efficient that this process has been referred to as a “huge rock-fluid ion-exchange systems for Ca2+ and Mg2+” (e.g., Stanley and Hardie, 1999, p. 3). This exchange is probably analogous to the processes that occur when Ca-Mg–rich basaltic river suspended material interacts with the ocean. Recent experimental work reported by Stefánsdóttir and Gislason (2005) and referred to by Schwartzman in his Comment, suggests that, like the alteration of mid-ocean ridges, Ca liberation from basalt and basaltic glass is concurrent with Mg consumption. The degree to which Mg consumption is required for the liberation of Ca from either the basaltic suspended material or the mid-ocean ridges is unclear. Ca release from basalt in either the mid-ocean ridges or suspended material in deltas is due to the dissolution of Ca-bearing minerals and glasses. These Ca-bearing minerals and glasses are undersaturated, and their dissolution reactions have been demonstrated to proceed in natural seawater, meteoric waters, and experimental solutions regardless of the aqueous Mg concentration (e.g., Mottle, 1983; Oelkers and Gislason, 2001; Stefánsdóttir and Gislason, 2005). These observations demonstrate that the presence of aqueous Mg is not necessary for the liberation of silicate-bound Ca. Moreover, although Mg precipitation may promote Ca-silicate dissolution via proton production and/or Si and Al consumption, this enhancement may be balanced by the slowing of these dissolution reactions by either blocking reactive surfaces (e.g., Cubillas et al., 2005) or by clogging fluid flow paths by Mg-rich secondary phases. It is, therefore, unclear if dissolved Mg either enhances or inhibits the liberation of Ca from mid-oceanic–ridge and river suspended basalts.
We showed that (1) the flux of Ca to the ocean originating from silicate minerals transported via suspended material is of same order of magnitude as that transported by aqueous transport in river water, and (2) the suspended Ca flux is far more climate dependent than the aqueous river Ca flux and therefore has a stronger negative feedback for Earth's climate through the greenhouse effect. As evidenced by the arguments presented above, Ca liberated from suspended material undoubtedly has a greater effect on carbon fixation in the present-day ocean than that of Mg consumed by the digenesis of these sediments. The degree to which this Mg may have effected carbon fixation in the ancient (>100 m.y. ago) oceans is related to the relative efficiency of Mg carbonate versus Mg silicate precipitation in these waters.
The critical Comment of Schwartzman does, however, open up an interesting question: Could the Mg-for-Ca exchange in deltas play a role in the historic seawater Mg/Ca ratio evolution? Until now only two processes have been suggested: 1) changing ocean floor spreading rates (Hardie, 1996) and 2) dolomitisation rates (Holland, 2005). Schwartzman's Comment suggests that the weathering of volcanogenic material in deltas, such as those of volcanic islands and North and South America, could play a role in seawater Mg/Ca ratio evolution and perhaps, most interestingly, provide a link between climate and this ratio.