Dear Alexandre, If I understood your message right, removing the variation attributed to the XY coordinates matrix essentially removes most variation that can be attributed to management if your treatments are separated in space. There is simply no way to disentangle treatments effects from other potential effects (e.g. a nutrient gradient) because of that; this is a design problem, not due to the analysis.
What I suggest is that you only use a management "matrix" (binary variable 0-1) and a env. var. matrix. Don't use a XY spatial matrix. Just be cautious in your interpretations of the results. Cheers Etienne -- Etienne Laliberté ================================ Rural Ecology Research Group School of Forestry University of Canterbury Private Bag 4800 Christchurch 8140, New Zealand Phone: +64 3 366 7001 ext. 8365 Fax: +64 3 364 2124 www.elaliberte.info
