R-sig-ecology Digest, Vol 36, Issue 2

Fri, Mar 4, 2011 6:25 AM
Hi Bier,

I've also thought about such analyses. I don't have the definitive
answer, but one option may be to look at the analysis on Hawaiian bird
abundance time series in Zuur et al Mixed Effects Models and
Extensions in Ecology with R. They use the gamm function in "mgcv" to
model the time-series where each species has its own smoothing term
for time (substitute plot for birds)

BM1<-gamm(Birds ? Rain + ID + s(Time, by = as.numeric(ID ==
"Stilt.Oahu")) +s(Time, by = as.numeric(ID == "Stilt.Maui")) +s(Time,
by = as.numeric(ID == "Coot.Oahu")) +s(Time, by = as.numeric(ID ==
"Coot.Maui")), weights = varIdent(form =? 1 | ID))
... where ID is the species name.

To make it analogous to your situation, perhaps the next step would be
to make another model (BM0) where time is the same for all species (in
your case plot), then test whether the reduced model fits as well as
the complex model. e.g.,
BM0<-gamm(Birds ? Rain + ID + s(Time))
anova(BM0, BM1)

Robert Rankin
Bird Studies Canada
Port Rowan, ON N0E 1M0