Search for a command to run...
ABSTRACT Reintroductions are frequently conducted to restore carnivore populations, though many fail despite following best management practices. Although demographic approaches are increasingly used to assess reintroduced populations, immigration is often overlooked in reintroduction planning and assessment. American martens ( Martes americana ) were reintroduced to 2 sites in Wisconsin, the Nicolet and Chequamegon national forests, which have anecdotally experienced differential population recovery. Via non‐invasive sampling and a mark‐recapture framework, we explored the possible mechanisms behind the divergent trajectories of the Nicolet National Forest (Nicolet) population, which appears to be recovering, and the Chequamegon National Forest (Chequamegon) population, which appears to be declining toward extinction. We tested for site‐level differences in survival, estimated population size, and calculated immigration rates based on simulations of the original reintroductions to the Nicolet. We then projected each population forward, both with and without our estimated immigration rate. We calculated a current population size of 66 female martens in the Nicolet, and estimated that the reintroduced population would have reached this size if supplemented by 1 female immigrant per year. We simulated future populations with the estimate of 1 female immigrant per year; this sustained level of immigration to the Nicolet reduced the probability of extinction to <1% over the next 100 years. When we applied the same rate of immigration to the starting population of 22 female martens in the Chequamegon, extinction probabilities were reduced by approximately 30%. Despite similarities between the 2 reintroduction sites, we propose that the differential population recovery was primarily due to modest rates of immigration to the Nicolet and the relative isolation of the Chequamegon population. Martens in the Chequamegon have been projected to go extinct in the foreseeable future, yet increasing connectivity with the Nicolet or Upper Peninsula of Michigan could improve persistence. Our results illustrate the effect of modest immigration on reintroduction success and the importance of planning for connectivity across reintroduction sites. © 2019 The Wildlife Society.