Procedures are described for establishing mean chlorophyllawater quality standards based on the probability and frequency that any specified “maximum” chlorophyllaconcentration will be exceeded by a given mean chlorophyllaconcentration. The models developed employ frequency distribution relationships and provide a means through which water quality standards can be established for mean chlorophyllaconcentrations while accounting for the impacts of peak algal growths. This is felt to be a more appropriate approach than establishing arbitrary maximum chlorophyll a standards. The relationships developed can be used in conjunction with other models to predict the effect that remedial programs could have on the probabilities that chlorophyll a concentrations would exceed desired levels. The procedures permit the establishment of rational sampling programs for compliance/ enforcement monitoring. These models are calibrated on data for North Carolina lakes. The complications arising from the dependency of “maximum” chlorophyllaon sample size is accounted for by these models. A linear regression model is developed for predicting the largest expected chlorophyllaconcentration from the mean chlorophyllaconcentration, and is compared with the model based upon frequency distribution relationships.