[R-sig-eco] Designing null models to account for natural variation

[V. Coudrain]

Dear list members,

I am performing a study in which I analyse species abundance and composition under four different treatments + 1 control, each replicated four times. In order to have an idea on baseline natural variation in species abundance and composition, I think about using null models based on the observations from the control. My idea was to use the mean and standard deviation to generate random replicates with which the observations from the different treatments would be compared and effect sizes measured. I like to perform similar analysis for the multivariate approach based on species composition, but I do not really figure out how to do this. 
Do you think it is a reasonable approach and would you have some indications how to design these null models (possibly in R)?

Thank you very much
Valerie

___________________________________________________________
Mode, hifi, maison,… J'achète malin. Je compare les prix avec 
	[[alternative HTML version deleted]]