Predicting In Situ Responses of Taste- and Odor-Producing Algae in a Southeastern US Reservoir to a Sodium Carbonate Peroxyhydrate Algaecide Using a Laboratory Exposure-Response Model

Abstract

Efficacy of an in situ algaecide treatment can be predicted prior to an application by physically modeling exposures and responses with laboratory experiments. A sodium carbonate peroxyhydrate (SCP) algaecide was used in a drinking water reservoir (Hartwell Lake, Anderson, SC) to control a benthic algal assemblage putatively producing 2-methylisoboreol (MIB) and geosmin, compounds with adverse taste and odor attributes. These SCP applications provided an opportunity to test hypotheses regarding potential convergence of laboratory and in situ exposures and responses. Objectives of this study were to (1) measure responses of a benthic algal assemblage from Hartwell Lake to 7-day laboratory exposures of SCP [measured as hydrogen peroxide (H2O2) concentrations], (2) to measure the exposure of SCP (as H2O2) applied in a cove of Hartwell Lake and consequent responses of the algal assemblage, and (3) compare exposures and responses measured in the laboratory and in situ. Results demonstrated that in laboratory exposures, H2O2 released by SCP dissipated within 48 h. Significant responses of the algal assemblage in terms of phycocyanin concentrations and cell densities were measured 4 days after treatment (4-DAT) and 7-DAT following exposures of 453, 615, and 812 mg H2O2 m−2. The H2O2 exposure measured in situ was comparable to effective laboratory exposures in terms of initial exposure (619 ± 428 mg H2O2 m−2) and exposure duration (dissipation within 30 h), but the in situ exposure had a large deviation initially (i.e., ±428 mg H2O2 m−2) and was an order of magnitude less than the targeted initial exposure. Therefore, comparison of measured responses was critical to infer comparable exposures and confirm accuracy of the laboratory model. Significant in situ responses were measured 4-DAT and 7-DAT in terms of phycocyanin concentrations and cell densities, and were comparable to responses obtained from effective laboratory exposures (i.e., 453, 615, and 812 mg H2O2 m−2). Decreases in measured concentrations of MIB and geosmin at the intake of the drinking water treatment facility provided additional evidence that algae were sufficiently exposed to H2O2 from SCP. Results of this experiment provide evidence for the design and use of physical laboratory models to predict responses of algae in the field.