[R-sig-eco] clustering communities by an environmental variable

[Torsten Hauffe]

Hi Irene,

I'm not sure whether I understand your aim completely. It sounds similar to
stratigraphically constrained clustering (CONISS) - an approach often used
in paleoecology and implemented in the packages rioja and adjclust. Instead
of stratigraphy/age you could use elevation (I guess all values of the
gradient need to be different).
Alternatively, Borcard & Legendre (Numerical ecology with R) suggest to use
the package mvpart for this aim.

Cheers!

On Fri, Apr 24, 2020 at 5:48 PM Irene Adamo <i.adamo90 using gmail.com> wrote:

> Dear all,
> I am trying to do a dendrogram  by trying to see if my communities  by
> altitude but I am not sure on how to do it.
> lets say I have this example
>
> # calculate Bray-Curtis distance among samples comm.bc.dist <-
> vegdist(comm,
> method = "bray") # cluster communities using average-linkage algorithm
> comm.bc.clust <- hclust(comm.bc.dist, method = "average") # plot cluster
> diagram plot(comm.bc.clust, ylab = "Bray-Curtis dissimilarity")
>
> Now, I would like to cluster my communities by Altitude always using the
> distance matrix and the dendrogram.
> Is it possible?
>
> Thanks a lot for any help!
>
>         [[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-ecology mailing list
> R-sig-ecology using r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-ecology
>

	[[alternative HTML version deleted]]