Many declining populations of the imperiled Allegheny woodrat (Neotoma magister) function as nonequilibrium metapopulations in which rates of subpopulation extirpation exceed recolonization. Quantifying and maximizing survival rates thus becomes critical for the conservation of these spatially structured populations. We used encounter histories of individually marked woodrats from subpopulations in Indiana, monitored annually from 2005 to 2013, to 1) estimate apparent annual survival rates while accounting for imperfect detection, 2) evaluate differences in apparent survival between unaugmented subpopulations and subpopulations reestablished or restored through translocation efforts, and 3) describe the effect of genetic diversity on survival. From Cormack–Jolly–Seber models developed in a Bayesian framework, apparent survival was greater for adults than for juveniles, greater for females than males, and there was a modest negative effect of density dependence. Although heterozygosity rates at 11 microsatellites increased among reinforced subpopulations following translocations, we observed no effect of heterozygosity on apparent survival. However, after translocations, average apparent survival was approximately 14% greater among recipient subpopulations than remnant subpopulations. This suggests that viability of recipient subpopulations was limited by low connectivity and the absence of genetic benefits conveyed by immigration or the potential for patches to be recolonized following local extinction. Under conditions of reduced connectivity, translocation among subpopulations to replicate natural gene flow may be appropriate to facilitate the long-term persistence of this and perhaps other nonequilibrium metapopulations.