[R-sig-eco] NMDS trajectory plots

[Lien Sim]

Thanks Jari, ordiarrows() looks like a useful function, and I have downloaded
the latest version. Unfortunately, I am still confused about how to
incorporate factors into my NMDS analysis to begin with. I have been
searching the R forums and helpfiles trying to find a solution and was
wondering if someone could point me in the right direction.

I have a community (P/A) matrix with 78 taxa and 105 site/date combinations.
I can run an NMDS using metaMDS() on this data matrix - the problem comes
when I try to split my site/date codes into two separate factors - one with
site names (9 levels) and one with sampling years (17 levels). If I load a
data matrix with the first two columns as factors, the NMDS will no longer
run and I get this error message:

Error in if (any(autotransform, noshare > 0, wascores) && any(comm < 0)) { :
  missing value where TRUE/FALSE needed
In addition: Warning message:
In Ops.factor(left, right) : < not meaningful for factors

I don't know what this message means and am unsure why it's telling me there
are missing values, when it's the same data I successfully used for the
previous NMDS, just with site/date codes replaced with two factor vectors.
Should I be using two matrices (one with community data and one with factor
data) to deal with this problem of multiple factors? If so, how do you
relate the matrices?

I'm having the same problem with ordiarrows(). I'm not sure how to tell R to
use the sampling years (one factor) for each site (the other factor) to plot
trajectories within sites over time.

I would be very grateful for any help.

Thanks!
Lien

--
View this message in context: http://r-sig-ecology.471788.n2.nabble.com/NMDS-trajectory-plots-tp6957817p6980510.html
Sent from the r-sig-ecology mailing list archive at Nabble.com.