ecomix is a package to implement model based species level (Species Archetype Models) or site level (Regions of Common Profile) grouping of community data.
Use phylogenetic branch length to inform this process. When modelling species in a sites by species matrix you add in ‘new’ species which are species complexes. So for a simple example you model sp1, sp2 and sp12. This will blur the biogeographic delineation, as you are essential modelling the distributions of sp1 and sp2 as a single species. How deep you go in the tree will determine how high in the hierarchy you go. Presumably the deeper you go the broader the regions will get – all the way to the terminal branch you are then at a global scale.
Methodologically this should be quite easy – we’ll just need a function in the ecomix function that can create new ‘species’ based on their respective phylogenetic relationship, I’m sure there will be existing R packages that do this ‘ape’ ect.
The species matrix will then contain sites X sp + sp_phylo- combinations.
You could also weight the species and species complexes by their respective branch length. Thus each species would get a weighting - I'll have to talk to Scott about how to do this.
Use phylogenetic branch length to inform this process. When modelling species in a sites by species matrix you add in ‘new’ species which are species complexes. So for a simple example you model sp1, sp2 and sp12. This will blur the biogeographic delineation, as you are essential modelling the distributions of sp1 and sp2 as a single species. How deep you go in the tree will determine how high in the hierarchy you go. Presumably the deeper you go the broader the regions will get – all the way to the terminal branch you are then at a global scale.
Methodologically this should be quite easy – we’ll just need a function in the ecomix function that can create new ‘species’ based on their respective phylogenetic relationship, I’m sure there will be existing R packages that do this ‘ape’ ect.
The species matrix will then contain sites X sp + sp_phylo- combinations.
You could also weight the species and species complexes by their respective branch length. Thus each species would get a weighting - I'll have to talk to Scott about how to do this.