Interannual variability in phytoplankton biomass and species composition in a subtropical reservoir

• 1 The interannual variability of the dominant phytoplankton populations is described in a subtropical reservoir in Queensland using weekly data for a 16-year period between 1978 and 1994. North Pine Dam, Brisbane, is in an area characterized by strong interannual variability in rainfall. This variability is linked to El Nino Southern Oscillation (ENSO) events. Between 1978 and 1994 periods of drought (during strong ENSO events) were interspersed by periods of flooding rains. Rainfall on the catchment and temperature and oxygen within the dam showed strong 40-day periodicities which also varied in strength interannually in response to ENSO events. Similar patterns of fluctuations in the 40-day periodicity have been found elsewhere in SE Australia. Seasonal cycles of stratification in the dam were a function of both hydrographic and hydrological events. Intermittent rain storms caused partial turnovers and large outflows. As much as 90% of the dam volume was exchanged in a single flood event.
• 2 The dominant phytoplankton species were similar to those frequently found in tropical and subtropical lakes and reservoirs. The phytoplankton community switched between cyanobacterial blooms (Cylindrospermopsis, Microcystis) during drought and falling water levels and diatom blooms (Aulacoseira) in response to inflows and seasonal turnovers. There appeared to be a subtle interaction between inflows, water column stability, the periodic overturns and the occurrence of the dominant species. All the dominant species showed long periods (2–4 years) of exponential increase or decrease superimposed on top of the seasonal fluctuations in abundance. These patterns of abundance led to marked interannual variability in the phytoplankton biomass. Climate variability had a major impact on the seasonal and interannual changes of the dominant phytoplankton species.
• 3 Phytoplankton biomass tended to be depressed for about 3 months after individual storm events but the data also displayed long-term lag effects (2–4 years) which destroyed any significant correlation between water residence time and biomass. Summer maxima of biomass dominated by cyanobacteria disappeared between 1985 and 1990 and were replaced by smaller winter peaks. The data presented here are not capable of unequivocally identifying the precise reason for these longer-term effects. Because of the implications for water quality management in subtropical and tropical reservoirs they warrant further study.