Using population viability analysis to inform and adapt ex situ conservation activities benefitting a critically endangered butterfly

Abstract Population viability analysis (PVA) can be a useful decision‐support tool to leverage uncertainty to avoid conservation hesitancy and paralysis. We developed a novel PVA model to help evaluate the effectiveness of a range of ex situ management options for the Critically Endangered Poweshiek skipperling butterfly. The PVA indicated that the last three known United States populations face high likelihoods of extinction in the absence of supportive ex situ management, particularly when population sizes and in situ survival rates are low. Modelling indicated that a headstarting program initiated in 2017 to reinforce these populations was only likely delaying extirpation. Persistence probabilities fell once the program was concluded, regardless of population size or survival rate. An alternative ex situ breeding program provided significant boosts to the persistence of all populations by producing significantly more individuals available for releases than a headstarting program reliant on in situ population sizes could provide. Withholding individuals to establish a large ex situ breeding population did not impose a notable penalty on wild populations and improved persistence. Persistence was boosted further when an ex situ breeding program was combined with headstarting to reinforce populations. Based on these model outcomes, we elected to pivot management strategies from headstarting to reinforcement via an ex situ breeding programme. Effective PVAs are contingent on the incorporation of as much information as possible, acknowledgement of uncertainty, and flexibility. PVAs provide research opportunities and have been under‐utilised in insect conservation. Our model is likely the first to link ex situ management to in situ dynamics for an invertebrate.