1074 Conservation Biology, Pages 1074–1085 Volume 16, No. 4, August 2002 Dynamics and Viability of a New Zealand Robin Population Reintroduced to Regenerating Fragmented Habitat DOUG P. ARMSTRONG* AND JOHN G. EWEN† Wildlife Ecology Group, Institute of Natural Resources, Massey University, Private Bag 11222, Palmerston North, New Zealand Abstract: New Zealand Robins ( Petroica australis) were reintroduced to Tiritiri Matangi, a highly modified island undergoing an extensive revegetation program. The island had only 13 ha of fragmented forest likely to be suitable for robins. We wished to determine whether the population was likely to persist until additional habitat became available through maturation of the revegetation and, therefore, whether it was sensible to reintroduce this species early in the restoration program. We used an information-theoretic approach to model survival, fecundity, sex ratio, and juvenile dispersal for this population, based on data collected in the first 6 years after reintroduction. This involved nominating a limited set of plausible models for each process and then selecting the models that best explained the data based on Akaike’s information criterion. We used mark-recapture analysis on survey data to model survival and general linear modeling on fecundity data. The best models were as follows: (1) juvenile survival declined as population size increased, whereas adult survival was constant among years; (2) fecundity was substantially lower for females in their first year after translocation and varied among forest patches, but did not vary among age classes or years; (3) the expected sex ratio of recruits was 50:50; (4) the distribution of recruits depended on the number of existing residents in patches and patch sizes, but not on patch isolation. We used VORTEX to simulate the dynamics of the pop- ulation based on these models and associated parameter estimates. The population was predicted to have negligible chance of extinction over several decades under current conditions, and this prediction was not sensitive to uncertainty in parameter estimates or model structure. Maturation of the planted matrix should allow the population to expand to several hundred birds over the next few decades, preventing extreme loss of genetic variation. We therefore conclude that the early reintroduction of this species was reasonable, de- spite the limited habitat available at the time. Our results show that it is critical to obtain data for the first few years after reintroduction if habitat is likely to be limited, because it is otherwise impossible to detect den- sity-dependent processes critical to population viability. Results also show that the first year after reintroduc- tion may be misleading if considered alone. Dinámica y Viabilidad de una Población de Petirrojo de Nueva Zelanda Reintroducida en un Fragmento de Hábitat en Regeneración Resumen: Los petirrojos de Nueva Zelanda ( Petroica australis) fueron reintroducidos a Tiritiri Matangi, una isla muy transformada en la cuál se está llevando a cabo un programa extensivo de reintroducción de vege- tación. La isla tenía solo 13 Ha de bosque fragmentado adecuado para los petirrojos. Quisimos determinar si la población persistiría hasta que hubiera hábitat adicional disponible por la maduración de la vegetación, y, por lo tanto, si era sensato reintroducir esta especie al inicio del programa de restauración. Utilizamos un enfoque información-teórico para modelar la supervivencia, fecundidad, proporción de sexos y dispersión de juveniles para esta población con base en la colecta de datos en los primeros seis años después de la reintro- ducción. Esto significó nombrar un conjunto limitado de modelos plausibles para cada proceso y luego selec- *email d.p.armstrong@massey.ac.nz † Current address: Laboratoire d’Ecologie, University Pierre et Marie Curie, 7 quai St. Bernard, Paris Cedex 05, France. Paper submitted May 22, 2000; revised manuscript accepted August 29, 2001.