Résumé
Survival is a key life-history trait in animals. However, most methods of survival estimation require substantial human and economic investment in the long term, particularly in species occurring in low densities, the case of most endangered species. An alternative to traditional recapture (CR) methods is estimation of adult survival based indirectly on either age ratios (AGR) or turnover rates (TOR) in territorial species. These 2 methods are applicable to bird species in which recruited individuals enter into the breeding population whilst still exhibiting the external traits that distinguish those animals from experienced adults. The main advantages of these methods are that survival can be easily estimated for all monitored individuals after just 1 or 2 breeding seasons and that disturbance to the species is minimized. The main constraints of indirect methods are that the assumptions are more restrictive than in CR methods, and survival estimates, although comparable between sites and years, may be biased. We used data from a long-term monitoring survey of 2 populations of the endangered Bonelli's eagle (Aquila fasciata), one in Catalonia (NE Spain) and the other in Provence and Languedoc-Roussillon (SE France). We evaluated survival estimates using the AGR and TOR methods and compared them with CR methods and provide suitable corrections for refining survival estimates based on indirect methods. In Catalonia (2002-2008), survival was estimated at 0.84 by CR methods (SE = 0.047; n = 25 radio tagged eagles), at 0.86 by the corrected AGR method (SE = 0.011; n = 558 bird * year), and at 0.86 by the corrected TOR method (SE = 0.022; n = 547 bird * year). In France (1999-2008), survival was estimated at 0.88 by CR methods (SE = 0.040; n = 45 darvic banded eagles), at 0.87 by the corrected AGR method (SE = 0.015; n = 443 bird * year), and at 0.87 by the corrected TOR method (SE 0.015; n = 438 bird * year). All analyses suggest that females survive better than males and that individuals from the French population survive better than individuals from the Catalan population. We conclude that indirect methods, which should not be regarded as a substitute of CR methods, will allow wildlife managers and researchers to estimate accurately adult survival in a territorial species over a short period of time and to monitor survival across populations over large geographic ranges and over time. (C) 2011 The Wildlife Society.