Multi-season covariate query

[geoffwah]

Hi All

I'm currently undertaking a single-species, multi-season analysis in which my primary objective is to estimate relationships between site covariates and epsilon and gamma.

My query is in regard to the treatment of the site covariates. I'm interested in relationship between conditions at time t and epsilon or gamma from t-t+1. However, my dataset actually spans 6 breeding seasons, and some of the site covariates changed a little in this time (e.g. surrounding urban density). My undertsanding is that 'season-specific' measures of the site covariates cannot be included, in which case I must either: (i) assume that the conditions at the start of the study are a reasonable approximation of the conditions in each subsequent season, or; (ii) quantify the change over time, and use that as my covariate. Am I correct in this regard??

Greatly appreciate your thoughts!

Cheers

Geoff

[jhines]

Geoff,

You can enter your seasonal covariates as sample covariates, even though you'll only use some of the columns. Just enter zeros in the ramaining columns to fill out the sample covariate table.

PRESENCE will use the first column for the first season (for epsilon or gamma), the 2nd column for the 2nd season...etc.

Cheers,

Jim

[geoffwah]

Wow, so it is possible!! Thanks Jim - much appreciate the advice.

Re the set-up of the sample covariate table, can I just check that I have this right? Say I have two seasons, each with 4 sampling occasions and a site covariate that changed between seasons. Sounds like I need to set up the table like this:
20,35,0,0,0,0,0,0

rather than:
20,0,0,0,35,0,0,0

Cheers

Geoff

[jhines]

Geoff,

The first one is correct (20,35,0,0,0,0,0,0 )

Cheers,

Jim

[geoffwah]

Beauty, thanks again Jim - I'll give it a go....

Cheers

Geoff