Table 1. Results of variation partitioning (based on adjusted D2) using Generalized Linear Models (GLM) with binomial error distribution and Bayesian Information Criterion variable selection method. The proportion of explained variation in the distribution of each macrophyte species was divided among pure local, climate and spatial effects, along with their joint fractions (represented by ∩). Areas are represented as: FIN: Finland, SWE: Sweden, MN: Minnesota, WI: Wisconsin. NV indicates no variables were selected for species. NA indicates an area not applicable as the number of species observations was too low (prevalence < 5%) or high (prevalence > 95%) to provide a feasible model.
...read more