COMMUNITY STRUCTURE, PICOPLANKTON GRAZING AND ZOOPLANKTON CONTROL OF HETEROTROPHIC NANOFLAGELLATES IN A EUTROPHIC RESERVOIR DURING THE SUMMER PHYTOPLANKTON MAXIMUM

An intensive 5 wk study was conducted to investigate the role of proti sts, especially heterotrophic nanoflagellates (HNF), in microbial food webs during the summer phytoplankton bloom in the epilimnion and meta limnion of the eutrophic Rimov reservoir (South Bohemia, Czech Republi c). On average, protists consumed similar to 90% of bacterial producti on in both layers. The community composition of HNF and the relative i mportance of different HNF groups as picoplankton consumers were deter mined. Small HNF (<8 mu m), as chrysomonads, bodonids and choanoflagel lates, usually accounted for <30% of total HNF biomass but numerically dominated the community in both layers. They consumed most of (simila r to 70 to 85 %) the bacterioplankton as well as autotrophic picoplank ton (APP, exclusively cyanobacteria) production in the reservoir, with the rest consumed by ciliates. Both ciliates and HNF had higher clear ance rates on APP than on bacteria and their grazing was likely respon sible for a sharp decrease in APP abundance (from 3-4 x 10(5) to <2 x 10(3) ml(-1)) and a very constant size structure of bacterioplankton i n which short rods in the size class of 0.4 to 0.8 mu m constituted 55 to 80 % of the total bacterial biomass in both layers. The proportion of HNF to total picoplankton biomass in the epilimnion indicated that the picoplankton biomass was sufficiently high to support HNF growth for most of the study. Uptake of picoplankton by less numerous, but la rger, HNF (kathablepharids, Goniomonas sp., and Streptomonas sp.) was negligible, while their biomass, especially in the metalimnion, exceed ed that of small HNF and the total biomass of picoplankton. This sugge sted that food items other than picoplankton were consumed to meet the ir carbon requirements. Analyzing potential bottom-up and top-down fac tors controlling HNF numbers and biomass, we did not find a tight rela tionship between HNF and the concentration of bacteria and chlorophyll . Variability of HNF abundance and biomass in the epilimnion could lar gely be explained by cladocerans or by pooled abundances of all potent ial crustacean consumers of HNF. In the metalimnion, the mean cell vol ume of HNF was positively linked to chlorophyll but negatively to the abundance of Cyclopidae and to the pooled abundances of Ceriodaphnia q uadrangula and Diaphanosoma brachyurum.