stopover model with temporal covariates

[jmcaldwell]

I am attempting to model a data set of remote detections using 3 states (not yet arrived, arrived, departed) and 2 events (captured, not captured) with a time varying model that is conditional on the first occasion from Guérin et al. 2017 (Guérin, S., Picard, D., Choquet, R., & Besnard, A. (2017). Advances in methods for estimating stopover duration for migratory species using capture–recapture data. Ecological Applications, 27(5), 1594-1604.). My data set consists of approximately 2500 individuals (2 groups) and 99 occasions. I have built the fully time varying model: f(1).g.t+f(2).g.t, which works, but I am most interested in how temporal environmental covariates affect the transition probabilities. When using a model that is conditional on the first occasion, can covariates be tested using E-SURGE? If so, would this GEMACO sentence work: f(1).g.t+f(2).g.t.+[i+t*x(1)]? The transition matrix and capture matrix I'm using are below. Thanks, in advance, for any help.

Transition Matrix
* r -
- s *
- - *

Capture Probability Matrix
* -
* p
* -

[simone77]

I'm uncertain about how to address your main question: 'Can covariates be tested using E-SURGE when utilizing a model conditional on the initial occasion?' The subsequent inquiries heavily hinge on clarifying this point. While my inclination is to say 'why not?' a more definitive answer could be obtained by reaching out to Rémi Choquet for clarification. If you do, it would be great if you shared his response here.