Diversity predictors
Two macrophyte measures were previously shown to influence susceptibility to invasion (Capers et al. 2007): diversity andplant cover . We therefore developed three complementary indicators from the macrophyte PVI data to estimate macrophyte diversity: native species richness, the Shannon–Wiener diversity index (henceforth referred to as native Shannon diversity) and native macrophyte beta diversity. These indicators address key elements of native macrophyte community structure: the number of species in a lake (richness), their relative abundances within a lake (Shannon diversity), and their compositional variation across space and time (beta diversity). The native macrophyte cover data (%) collected during the PVI assessments at each sampling point were assigned to native plant cover (total sum of native macrophyte species cover) and to two functional growth forms (submerged plant cover or floating plant cover). Beta diversity was calculated as the compositional variation in the native macrophyte species’ PVI values (excluding E. canadensis ) across all sampling points as estimated via principal curves analysis (PC; De’ath 1999). PC analyses were run using the prcurvefunction in the “analogue” package in R (Simpson et al. 2020), using Canonical Analysis as the starting point on square-root transformed PVI data. The plant macrofossil data were similarly grouped into five native diversity predictors: taxon richness, Shannon diversity, native plant cover, submerged plant cover and floating plant cover.