Figure 4: Relationship between physicochemical variables and sample taxonomic composition.

dbRDA ordination of the distLM model describing the relationship between physicochemical variables and the taxonomic dissimilarity between samples. Vectors represent the effect of each predictor variable on the two visualized axes. Vector length corresponds to the relative size of the effect, whereas direction represents the correlations to the two displayed axes. The first axis (dbRDA1) captures 51% of fitted and 26% of total variation between the samples’ taxonomic profiles; the second (dbRDA2) captures 16% of fitted and 8% of total variation. AAIW, light blue stars; SAMW, orange crosses; AABW, dark blue squares; AZ, green circles; PFZ, yellow triangles; CDW, red diamonds.