Covariates in "random changes" multi-season model

[Orien Richmond]

Hi all,

I would like to examine the effects of several covariates on monthly occupancy of a marsh bird, but I don't have enough data to estimate colonizations and extinctions between months (35 sites over 8 months, 2 visits/month for 2 different years). My covariates include grazing (3 levels), area (continuous), and irrigated (binary). I think the "random changes in occupancy" parameterization of the multi-season model might work here. If I understand the model correctly, psi1 = occupancy for the first season and gamma1, gamma2, etc. = occupancy for each subsequent season.

My question is this: Is there a way to estimate a single parameter for each covariate (e.g., area)? It appears that I have to estimate at least two parameters for each covariate: psi1(area),gamma1-7(area).

Thanks!
Orien

[bacollier]

Orien,
I am sure Darryl/Jim will chip in here soon, I think there is a thread roughly on this from a while back (viewtopic.php?f=11&t=1018)

Bret

[darryl]

If you right click on the design matrix, there's an option 'combine matrices' which will put all the parameter types on a single design matrix, like MARK. Then you can make effects be consistent across psi and gamma.

[Orien Richmond]

Beauty! Thanks Darryl, that's a really cool option.

-Orien