Most treatments to control freshwater algal blooms are not effective: Meta-analysis of field experiments

Harmful algal blooms negatively impact freshwater, estuarine, and marine systems worldwide, including those used for drinking water, recreation, and aquaculture, through the production of toxic and nontoxic secondary metabolites as well as hypoxic events that occur when algal blooms degrade. Consequently, water resource managers often utilize chemical, bacterial, physical, and/or plant-based treatments to control algal blooms and improve water quality. However, awareness of available treatments may be limited, and there is ambiguity among the effects of algal bloom treatments across studies. Such variation within the literature and lack of knowledge of other tested treatments leave uncertainty for water resource managers when deciding what treatments are best to control algal blooms and improve water quality. Our primary objective was to synthesize data from 39 published and unpublished studies that used one of 28 chemical, bacterial, physical, and/or plant-based treatments in field experiments on various water quality measurements, including phytoplankton pigments and cell density, cyanobacterial toxins (microcystin), and common off-flavors (i.e., taste and odor compounds; geosmin and 2-methylisoborneol). We hypothesized that treatments would improve water quality. Across all studies and treatment types (227 effect sizes), water quality improvements were observed when measured at the time of greatest decline following treatment or at the end of the experiment. However, these findings were primarily mediated by only four chemicals, namely copper sulfate, hydrogen peroxide, peracetic acid, and simazine. None of the bacterial, physical, or plant-based treatments were shown to significantly improve water quality by themselves. Results from this synthesis quantitatively showed that most treatments fail to improve water quality in the field and highlight the need for more research on existing and alternative treatments.