[R-sig-eco] Intepreting a plot from a "constrained" NMDS

tyler tyler.smith at mail.mcgill.ca
Mon Dec 8 02:46:26 CET 2008


Manuel Spínola <mspinola10 at gmail.com>
writes:

> Is there any reference or document on how to interpret a "constrained"
> non-metrical multidimensional scaling using ecological data? 

The vegan vignette includes a worked example.

> By "constrained" I mean after fitting environmental covariables,
> using, for example, the "envfit" function in the vegan package. 

'envfit' can be used on any ordination result, including NMDS. It does
not produce a constrained ordination, though. The ordination itself is
unaltered, but the envfit result can be used to plot the environmental
variable in the ordination space. You can then interpret this as you
would a PCA biplot. This assumes that the environmental variable has a
simple linear relationship to the ordination axes. If it doesn't,
ordisurf will give you a clearer picture.

> Is it possible to interpret the resulting plot in the same way that a
> constrained ordination, for example CCA?

No, you're still doing an unconstrained ordination, also referred to as
an indirect gradient analysis. 

Cheers,

Tyler


-- 
Friends don't let friends send Word documents

http://www.nothingisreal.com/dfki/no-word



More information about the R-sig-ecology mailing list