Problems with paleoproductivity proxies

Difficulties associated with reconstructing past changes in export production were highlighted recently by Averyt and Paytan (2004), who reported substantial disagreement among records developed using different paleoproductivity proxies extracted from two equatorial Pacific piston cores. Proxies included the accumulation rates of barite, excess Ba, and excess Al, as well as elemental ratios of Al/Ti and Ba/Ti. Here we build upon their work by presenting evidence for two factors that contributed to these discrepancies. First, elemental (Ba/Ti and Al/Ti) ratios are influenced by variability in space and time of the flux of Ti to equatorial Pacific sediments, so these proxies cannot be expected to hold a constant relationship to export production. Second, the late Holocene increase in CaCO3 dissolution has caused concentrations of barite, excess Ba, and excess Al to be enriched in surface sediments relative to the depth interval over which sediment accumulation rates were evaluated in developing the algorithms used by Averyt and Paytan (2004). This produces an error in the accumulation rates of these proxies that varies from core to core, ranging from a few tens of percent to as much as a factor of 3. These errors would have been propagated into the export production algorithms on the basis of fluxes of barite, excess Ba, and excess Al. Furthermore, this bias created by the late Holocene increase in CaCO3 dissolution will affect the development of any algorithm based on fluxes of sedimentary constituents. These factors must be taken into account in future paleoceanographic reconstructions.