The organic-poor character of surface sediments in the APFZ, and throughout the Southern Ocean, indicates that those processes that deliver diatom opal to the sea floor do not also carry significant quantities of organic carbon to the sea floor. However, areas of significant opal export from the surface layer must also be areas of significant carbon export unless the decoupling of the cycles of silica and carbon is virtually complete within the surface layer. That is clearly not the case in the Ross Sea, where sediment-trap data indicate that the C/Si ratio of particles sinking through the 250 m depth horizon is 4 - 5 times as great as the C/Si ratio in the sediments, even though those sediments are at much shallower depths and have considerably greater organic carbon content than those in the APFZ (DeMaster et al., 1992). Thus, those processes delivering opal to the seabed may greatly intensify the transport of organic carbon to the deep ocean in the APFZ, even though the decomposition of that organic material within the deep water column is virtually complete.
The APFZ appears to be an area of significant net uptake of CO2 by the ocean (e.g., Takahashi et al., 1986). To the extent that biological pumping processes contribute to that uptake, the biogenic particle export indicated by the opal sediments of the APFZ appears to be reflected in the pCO2 signal. So even though opal sediments may be a poor indicator of total primary productivity they may be a much better indicator of organic matter export and biologically mediated CO2 uptake by the ocean.
Implications for field measurements: It would be useful to know whether the pelagic decoupling between the cycles of silica and carbon observed in the Ross Sea pertains also to the APFZ, where the largest and most quantitatively important area of modern opal sediment accumulation in the ocean is found. Of more direct importance to the study of carbon cycling would be an assessment of the upper-ocean processes that result in enhanced transport of diatom silica to the seabed and the degree to which organic carbon is carried to the deep ocean by those same processes. On the basis of presently available data from the Ross and Weddell Seas (e.g., DeMaster et al., 1992; Quéguiner et al., 1995), bloom areas--even though spatially restricted and of minor importance in the overall annual primary productivity of the Southern Ocean--may be of great quantitative importance in the vertical export of both siliceous and organic biogenic material.
Implications for modeling: The seabed signal in the APFZ is derived from the production and vertical export of biogenic opal while the biological component of the atmosphere-to-ocean CO2 flux is driven by the production and export of organic carbon. This difference implies that biogeochemical models of the Southern Ocean should consider the individual dynamics of carbon, nitrogen and silica. To resolve the sediment signal and its relationship to upper-ocean processes the models should also include at least two functionally different categories of phytoplankton--diatoms and smaller forms that neither take nor require Si. Moreover, such models should be formulated to address the possibility that there are strongly nonlinear relationships between production and export--at least for opal and perhaps for organic matter as well. Models with that degree of flexibility would be of great help in identifying relationships that result in enhanced export of opal and (perhaps) organic matter from surface waters during diatom blooms.