Application of a versatile aquatic macrophyte integrity index for Minnesota lakes

The objective was to develop a Minnesota aquatic macrophyte integrity index that can use plant checklist data from existing and ongoing lake plant survey programs without alteration. Using the extensive lake survey data collected by numerous state programs, we created a suite of predictive models for macrophyte richness and floristic quality and identified aquatic macrophyte community outliers to set potential impairment thresholds. The highest-ranked predictive models included total phosphorus, disturbance indices, and ecoregion variables. Models with all in-lake macrophyte taxa generally performed better than those based on just submerged aquatic macrophyte or those based on submerged and floating-leaf taxa. The best generalized linear mixed model for aquatic macrophyte richness was a model containing total phosphorus, alkalinity, lake size, maximum depth, ecoregion, survey type, and several interactions. The best linear mixed effects model for floristic quality also included these predictive variables. Richness and floristic quality thresholds were calculated using these models with associated disturbance–response breakpoints. The approach took sampling protocol into account by providing different thresholds based on sample design. These thresholds then identify potentially biologically impaired lakes. There appeared to be no disturbance–response breakpoints between aquatic macrophyte richness and floristic quality for the Northern Lakes and Forest ecoregion of northeastern Minnesota.