Estimation of survival rates from a mark-recapture study with tag loss

A probability model is described for a mark-recapture study where animals are released annually and where animals may be recovered and identified by a permanent band or resighted and identified by a nonpermanent collar. Survival, capture, and sighting probabilities are assumed to depend on the year. Collar-retention probabilities are assumed to depend only on the age of the collar. Closed-form maximum likelihood estimates do not exist for the model but numerical solutions are easily obtained.