Shifts in phytoplankton and zooplankton communities in three cyanobacteria-dominated lakes after treatment with hydrogen peroxide.

ABSTRACT

Cyanobacteria can reach high densities in eutrophic lakes, which may cause problems due to their potential toxin production. Several methods are in use to prevent, control or mitigate harmful cyanobacterial blooms. Treatment of blooms with low concentrations of hydrogen peroxide (H2O2) is a promising emergency method. However, effects of H2O2 on cyanobacteria, eukaryotic phytoplankton and zooplankton have mainly been studied in controlled cultures and mesocosm experiments, while much less is known about the effectiveness and potential side effects of H2O2 treatments on entire lake ecosystems. In this study, we report on three different lakes in the Netherlands that were treated with average H2O2 concentrations ranging from 2 to 5 mg L-1 to suppress cyanobacterial blooms. Effects on phytoplankton and zooplankton communities, on cyanotoxin concentrations, and on nutrient availability in the lakes were assessed. After every H2O2 treatment, cyanobacteria drastically declined, sometimes by more than 99%, although blooms of Dolichospermum sp., Aphanizomenon sp., and Planktothrix rubescens were more strongly suppressed than a Planktothrix agardhii bloom. Eukaryotic phytoplankton were not significantly affected by the H2O2 additions and had an initial advantage over cyanobacteria after the treatment, when ample nutrients and light were available. In all three lakes, a new cyanobacterial bloom developed within several weeks after the first H2O2 treatment, and in two lakes a second H2O2 treatment was therefore applied to again suppress the cyanobacterial population. Rotifers strongly declined after most H2O2 treatments except when the H2O2 concentration was ≤ 2 mg L-1, whereas cladocerans were only mildly affected and copepods were least impacted by the added H2O2. In response to the treatments, the cyanotoxins microcystins and anabaenopeptins were released from the cells into the water column, but disappeared after a few days. We conclude that lake treatments with low concentrations of H2O2 can be a successful tool to suppress harmful cyanobacterial blooms, but may negatively affect some of the zooplankton taxa in lakes. We advise pre-tests prior to the treatment of lakes to define optimal treatment concentrations that kill the majority of the cyanobacteria and to minimize potential side effects on non-target organisms. In some cases, the pre-tests may discourage treatment of the lake.