Keith Johnson

We initiated and mapped a diatom bloom in the northeast subarctic Pacific by concurrently adding dissolved iron and the tracer sulfur hexafluoride to a mesoscale patch of high-nitrate, low-chlorophyll waters. The bloom was dominated by pennate diatoms and was monitored for 25 d, which was sufficiently long to observe the evolution and termination of the bloom and most of the decline phase. Fast repetition–rate fluorometry indicated that the diatoms were iron-replete until day 12, followed by a 4–5-d transition to iron limitation. This transition period was characterized by relatively high rates of algal growth and nutrient uptake, which pointed to diatoms using intracellularly stored iron. By days 16–17, the bloom was probably limited simultaneously by both iron and silicic acid supply, because low silicic acid concentrations were evident. Modeling simulations, using data from our study, provided an estimate of the critical threshold for algal aggregation. Observed diatom abundances during the bloom exceeded this threshold between days 13 and 17. Mass sedimentation of diatoms and diatom aggregates was recorded in surface-tethered free-drifting sediment traps at 50 m in depth on day 21. Although the termination of the bloom was probably controlled by the availability of both iron and silicic acid, we cannot rule out the role of algal aggregation. The bloom decline was likely triggered by the onset of mass sedimentation. During our study, evidence of both diatom species succession and species-specific aggregation point to important links between algal nutrient stress and the initiation of algal aggregation.

The roles of heterotrophic organisms (microzooplankton, mesozooplankton, bacteria and heterotrophic nanoflagellates) were examined during a nutrient enrichment experiment using a mesocosm in Saanich Inlet, British Columbia, Canada. Grazing rates of microzooplankton, copepods, and Noctiluca scintillans were respectively estimated by the dilution method, from the egg production, and the apparent growth rate. The primary production increased by about 11 times during the initial 3 days, and the grazing rate by zooplankton also increased by 7.4 times. The primary production exceeded the grazing rate during the initial 5 days, after that, almost balanced rates were observed. Biomass peaks of bacteria and HNFs (heterotrophic nanoflagellates) were observed after the decline of the phytoplankton bloom. Bacterial production and HNF bacterivory gradually increased from the beginning to the end of the experiment. Microzooplankton consistently removed about half of the primary production. The contribution of microzooplankton to grazing was largest during the initial 7 days. Heterotrophic dinoflagellates were the most dominant component of the microzooplankton, but oligotrich ciliates showed the fastest growth response to phytoplankton production. Noctiluca scintillans became an important grazer after the bloom. Overall, the contribution of microzooplankton grazing was the largest of the processes through which phytoplankton were lost. Cell sinking was a minor component contributing to loss of phytoplankton. Thus, oligotrich ciliates and heterotrophic dinoflagellates were the most plausible organisms contributing to the steady state of phytoplankton concentrations.