varpart with RM design

Dear all,

I would like to do variance partitioning of community dissimilarity matrix
(Y) using 4 explanatory tables:
X1=environmental characteristics
X2=species traits related to dispersal
X3=species characteristics (abundances and richnness)
X4= xy spatial coordinates

The problem is that I have repeated measures design (longitudinal data). I
treated "Time" (factor with 4 levels) as fixed factor in the main analysis
(relating community dissimilarity matrix with environmental characteristics)
and checked for sphericity etc.

Now, I would like to see what proportion of variation is due to these 4
explanatory tables, but I am not sure how to deal with spatial coordinates
(X4), without having to pool all data across dates?
Should I just repeat xy data 4 times in the table?
Should I do pcnm with X4, before using it in varpart?

I would greatly appreciate any help,
Vesna 

--
View this message in context: http://r-sig-ecology.471788.n2.nabble.com/varpart-with-RM-design-tp6672240p6672240.html
Sent from the r-sig-ecology mailing list archive at Nabble.com.