Well, here I am again…
I’m working with Southern Right Whales in Península Valdés, Argentina. Since 12 years ago there’s a photoidentification research in a little area in the New Gulf, the most important reproduction area.
I’ve been trying to analyze this data with a MS model with unobservable states, but after did GOF tests (U-CARE) I found that there’s no significance in TEST2.CT, TEST2.CL and TEST3.SM, only TEST3.SR is significant (a transient effect). So there’s no non-temporary emigration.
And this is an expected result when we only sample a little area within the reproduction area in my case; there is a random emigration. In that case my best option is a robust design (not possible for me) or CJS (if S is the same between states) (Kendall and Nichols, 2002), but no MS models.
So, now I return to a CJS model with a survival TSM effect. Is it ok? I’m not sure if I’m seeing all the aspects around the problem, so I’m very afraid to do wrong analysis, and bring wrong results.
Again, I’m a beginner in capture-recapture field so sorry if the question is too basic…
Thanks in advance!
María Soledad Lindner