Evaluation of sampling and screening techniques for tiered monitoring of toxic cyanobacteria in lakes

Exposure to cyanotoxins can pose serious human health consequences both through drinking water and recreational use of a contaminated waterway. We adapted a tiered framework proposed by the World Health Organization (WHO) for monitoring cyanobacteria populations and assessing public health risks and implemented the program in Lake Champlain. This study focused on evaluating the sampling protocols employed in this adapted monitoring program over two field seasons (2003 and 2004). Using a paired sampling design, we evaluated whether 63-mu m Wisconsin net samples adequately represented whole-water conditions and whether chlorophyll a concentration could serve as a useful predictor of cyanobacteria density and microcystin concentration. We also evaluated the spatial and temporal dynamics of blooms and their implications for monitoring. Our results suggest that using threshold values of either potentially toxic cyanobacteria density counted using a rapid screening protocol or chlorophyll a concentration serve as initial indicators of potentially high levels of cyanotoxin; however, the utility of chlorophyll a is system-dependent. Whole-water samples provide more accurate estimates of population density and higher microcystin concentrations than net samples, offering a more cautionary approach to assessing risk to recreational lake users. Shoreline samples generally showed higher cyanotoxin concentrations than offshore, but restricting sampling to the shoreline may miss early warnings of bloom development. (C) 2007 Elsevier B.V. All rights reserved.