CAPTURE-RECAPTURE SURVIVAL MODELS TAKING ACCOUNT OF TRANSIENTS

摘要: The presence of transient animals, common enough in natural populations, invalidates the estimation survival by traditional capture-recapture (CR) models designed for study residents only. Also, transit is interesting itself. We thus develop here a class CR to describe transients. In order assess merits this approach we examine bias estimators transients relation power different tests detecting also compare relative efficiency an ad hoc dealing with that leaves out first observation each animal. then real example using lazuli bunting (Passerina amoena) and, conclusion, discuss design experiment aiming at transience. practice, easily detected whenever risk high. approach, which yields unbiased estimates only, satisfactory time-dependent context but poorly efficient when parameters are constant. shows intermediate situations between strict 'residence' and 'transience' may exist certain studies. Yet, most time, if takes into account expected length stay transient, it should be possible efficiently separate two categories animals.