Evidence of the potential influence of planktonic community structure on the interannual variability of particulate organic carbon flux

Abstract

The functioning of the biological pump during spring blooms was assessed with biogeochemical data from JGOFS (Joint Global Ocean Flux Study) process studies in the NE Atlantic during 1989 and 1990. A comparison of the integrated primary productivity signal (estimated by the 14C technique and from changes in ambient surface-water [tCO 2] and [NO 3 −]s during the spring blooms of the two years revealed close similarities. These observations suggest that the magnitude of the biological drawdown of CO 2 via photosynthetic activity during these periods was comparable. However, despite similarities in the magnitudes of these and other surface rate processes, sediment trap particulate organic carbon (POC) fluxes at 3100 m representing the spring bloom settlement events were 1.8 times greater in 1989 than in 1990. Taking into account the spatial and temporal resolutions of pelagic and deep trap datasets, these observations suggest that the coupling between organic carbon production in surface waters and its transfer to the deep ocean was stronger in the period studied in 1989 than in 1990. That is, the biological pump was more efficient in 1989. The size of the dominant phytoplankton species was observed to be the principal difference between the two spring bloom data sets. The potential influence of the observed algal size differences on the vertical POC flux was quantified from size-fractionated productivity data in conjuction with a food web-vertical flux model. The derived POC fluxes from the surface layer were two times greater in spring 1989 than in the bloom period in 1990, and a comparison of these flux estimates with those from other methods is favourable. Extrapolation of these derived shallow POC fluxes to 3100 m with existing empirical algorithms yields deep POC fluxes that are consistent with those collected by sediment traps at this depth. Differences in algal size between the two spring blooms can thus account for the observed interannual differences in deep-water POC fluxes without the need to invoke interannual differences in unmeasured mid-water processes. This work provides a clear demonstration that although observations of oceanic productivity may yield the input to the biological pump, they cannot, on annual timescales, reliably provide information on the efficiency of the pump in transferring carbon to the deep ocean.