Bats play crucial ecological roles and provide valuable ecosystem services, yet many populations face serious threats from various ecological disturbances. The North American Bat Monitoring Program (NABat) aims to use its technology infrastructure to assess status and trends of bat populations, while developing innovative and community-driven conservation solutions. Here, we present NABat ML, an automated machine-learning algorithm that improves the scalability and scientific transparency of NABat acoustic monitoring. This model combines signal processing techniques and convolutional neural networks (CNNs) to detect and classify recorded bat echolocation calls. We developed our CNN model with internet-based computing resources ('cloud environment'), and trained it on >600,000 spectrogram images. We also incorporated species range maps to improve the robustness and accuracy of the model for future 'unseen' data. We evaluated model performance using a comprehensive, independent, holdout dataset. NABat ML successfully distinguished 31 classes (30 species and a noise class) with overall weighted-average accuracy and precision rates of 92%, and >= 90% classification accuracy for 19 of the bat species. Using a single cloud-environment computing instance, the entire model training process took Synthesis and applications. Our convolutional neural network (CNN)-based model, NABat ML, classifies 30 North American bat species using their recorded echolocation calls with an overall accuracy of 92%. In addition to providing highly accurate species-level classification, NABat ML and its outputs are compatible with Bayesian and other statistical techniques for measuring uncertainty in classification. Our model is open-source and reproducible, enabling future implementations as software on end-user devices and cloud-based web applications. These qualities make NABat ML highly suitable for applications ranging from grassroots community science initiatives to big-data methods developed and implemented by researchers and professional practitioners. We believe the transparency and accessibility of NABat ML will encourage broad-scale participation in bat monitoring, and enable development of innovative solutions needed to conserve North American bat species.

Summary 1. The practice of deliberately moving animals from one site to another for conservation is increasing as a tool to re‐establish extirpated populations. Resource managers are faced with developing strategies for reintroduction attempts, but often lack experimentally derived evidence upon which to base decisions. 2. Using the northern water snake Nerodia sipedon sipedon in the USA, we compared the behaviour and performance of resident snakes with that of individuals translocated directly from the wild to a nearby nature reserve or reared in captivity prior to translocation. 3. Both translocated groups had low survivorship relative to resident snakes, but the proximal causes of their poor performance differed considerably. Captive‐reared snakes exhibited restricted surface activity and movements and abnormal habitat use, and ultimately failed to maintain appropriate body temperature and body mass, with high mortality associated with the overwintering period. Wild snakes directly translocated to an unfamiliar site maintained body temperatures and growth comparable with residents, but their more extensive movements resulted in frequent excursions off reserve and high mortality. 4. Synthesis and applications . We contend that an individual’s prior experience is an important factor in determining their behaviour and performance during the phase of early establishment at an unfamiliar site. This suggests the existence of common underlying mechanisms influencing the outcome of reintroduction attempts, and provides a potentially useful framework for improving reintroduction efforts. Resource managers would likely improve success of reintroductions by matching habitats (and associated resources and conditions) between source and release sites, by temporarily confining animals in enclosures that force new associations to be made while limiting exploratory wanderings, or by enrichment of environmental conditions in captivity.

Forest loss and fragmentation is expected to shape the genetic structure of amphibian populations and reduce genetic variation. Another factor widely understood to have impacted these same parameters in North America is the range expansion that occurred following glacial retreat at the end of the Pleistocene. The Eastern Red-Backed Salamander (Plethodon cinereus) has been subjected to both processes. In this context, we investigated the historical events that are likely to have shaped genetic variation in this species using a panel of six microsatellite markers screened on individuals sampled across ten localities in northeastern Indiana, USA. We found low genetic diversity across forest patches and minimal differentiation. We expected population structure associated with forest fragmentation to result from genetic drift in isolation but instead found that a balance between gene flow and drift was ~50 times more likely. Ratios of allele number and range (M), and coalescent modeling of population demography suggested the occurrence of marked historic decline in effective population size across the region. Taken together, the data point to a loss of genetic variation which preceded deforestation over the past 200 years. This result indicates an important role for ancient demographic processes in shaping current genetic variation that may make it difficult to detect the effect of recent habitat fragmentation.