ABSTRACT. The colloidal fraction of &ssolved organic carbon in seawater is one of the largest reservoirs of organic carbon on the planet, outweighing phytoplankton or the bacteria by a considerable margin. Even though this colloid-sized material is a carbon reservoir of global significance, it is not easily accessible to the bacteria and may escape extensive biological degradation by virtue of its particle size characteristics. However, when larger colloids (between 0.2 and 2.0 pm in diameter) are incorporated into microaggregates (that are tens of pm to mm across), colloidal organic carbon (COC) 1s broken down as the aggregates become bioreactors for organic material. For example, the aggregation of colloids and bacteria by surface coagulation triggers a brief (2 to 4 h) episode of bacterial respiration. The bioreactive nature of aggregates is confirmed in their development as sites of intense bacterial exoenzyme activity, even though the organic material released by this activity remains largely uncoupled from bacterial growth. The degradation of COC in aggregates is a process that is missing from current models of carbon transport and ocean productivity. In addition, while the COC caught up into aggregates may be more bioreactive than previously suspected, respiration of the aggregated material persists for only a few hours. Realistic measurements of respiration should take this short-lived, but intense, response to aggregation lnto account.

During the summer of 1994, bacterioplankton abundance and metabolism were examined in seawater from Bedford Basin, Nova Scotia. Canada. Two new methods were applied to independently assess metabolic activity: (1) flow cytometnc analysis of bimodal nucleic acid distributions in bacterioplankton stained with a novel fluorescent dye, and (2) utilization of single carbon substrates in Biolog GN Microplatcsl'' Both sets of results were compared to bacterial production estimates obtained using a standard technlque for 'H-thymidine and 3H-leucine incorporation. Flow cytometry quantified the relative abundance of the bacterioplankton cells with a high apparent nucleic acid content, which was expressed as an active cell index (ACI) of dividing and/or actively metabolizing cells. The ACI was positively correlated with sole carbon source utilization at 1, 5 and 10 m (p < 0 05). Variations in ACI and sole carbon source utilization followed trends similar to, but could not be directly correlated with, bacterial production throughout the summer period. The new methods provided information that could not be obtained using the standard techniques for bacterial production. Instead, they yielded new and complementary information on the metabolic state of dividing and/or metabolizing cells and insight into the regulation of bacterial production.

Dissolved organic carbon (DOC) in seawater collected during a spring bloom in Bedford Basin, Nova Scotia. Canada, was separated into low molecular weight and colloidal size fractions by cross-flow ultrafiltration. Total DOC and the organic carbon in the 2 size fractions were analyzed by high temperatwe catalytic oxidation DOC accumulated in the mixed layer at the beginning of the bloom and contalned a substantial amount of surface active material associated with an increase in river discharge There was llttle varlatlon in total DOC at the helght of the bloom. Instead, the low molecular weight fractlon increased and reached a maxlmum 3 to 6 d after the chlorophyll maximum. The colloidal fraction of DOC also increased but, at its maximum, accounted for only 16 % of the total. This maximum in colloidal DOC was achieved 8 d after the low molecular weight maximum and almost 2 wk after the chlorophyll maximum. Colloidal material may have coagulated with phytoplankton aggregates at the height of the bloom and limited the accumulation of DOC in the mixed layer. During the decline of the bloom, the downward transport of this material associated with aggregates may have contributed to the eventual accumulat~on of DOC below the mixed ldyer As a result, the downward transport of DOC should be determined from estimates of the low molccular weight and colloidal DOC produced by phytoplankton and include the colloidal carbon associated with aggregates.

No abstract