Comparing the invertebrate communities among invaded (Phragmites), uninvaded (cattail, meadow) and herbicide-treated (treated) sites in Long Point. There were 9 sites in each vegetation type and they spanned a water depth gradient. Were collected in two years, 2017 and 2018, over the field season (4 collections in 2017, 6 in 2018). Will combine them since differences between years are not of interest.
-
Univariate analyses
- one-way ANOVAs with habitat type for abundance and richness
- did a GLMM of these too (for no good reason)
- made violin plots for abundance, richness, Simpsons, Shannon-Weiner, and Pielou's
-
Mulivariate analyses
- one-way perMANOVA with pairwise comparison
- NMDS to visualize the groups
- Univariate
- two-way ANOVAs with habitat and year for abundance and richness
- used a GLMM and treated year and number of collections as a random effect; compared to a null
- NMDS to visualize
- Did hierarchical clustering of 2 - 10 groups then used Indicator Species Analysis to choose 3 groups as optimal (in PC-ORD)
- Made an NMDS figure to visualize the groups