Cyanobacterial blooms: Statistical models describing risk factors for national-scale lake assessment and lake management

Abstract

Cyanobacterial toxins constitute one of the most high risk categories of waterborne toxic biological substances. For this reason there is a clear need to know which freshwater environments are most susceptible to the development of large populations of cyanobacteria. Phytoplankton data from 134 UK lakes were used to develop a series of Generalised Additive Models and Generalised Additive Mixed Models to describe which kinds of lakes may be susceptible to cyanobacterial blooms using widely available explanatory variables. Models were developed for log cyanobacterial biovolume. Water colour and alkalinity are significant explanatory variables and retention time and TP borderline significant (R2-adj = 21.9%). Surprisingly, the models developed reveal that nutrient concentrations are not the primary explanatory variable; water colour and alkalinity were more important. However, given suitable environments (low colour, neutral-alkaline waters), cyanobacteria do increase with both increasing retention time and increasing TP concentrations, supporting the observations that cyanobacteria are one of the most visible symptoms of eutrophication, particularly in warm, dry summers. The models can contribute to the assessment of risks to public health, at a regional- to national level, helping target lake monitoring and management more cost-effectively at those lakes at the highest risk of breaching World Health Organisation guideline levels for cyanobacteria in recreational waters. The models also inform restoration options available for reducing cyanobacterial blooms, indicating that, in the highest risklakes (alkaline, low colour lakes), risks can generally be lessened through management aimed at reducing nutrient loads and increasing flushing during summer.