Motion‐activated cameras (“camera traps”) are increasingly used in ecological and management studies for remotely observing wildlife and are amongst the most powerful tools for wildlife research. However, studies involving camera traps result in millions of images that need to be analysed, typically by visually observing each image, in order to extract data that can be used in ecological analyses. We trained machine learning models using convolutional neural networks with the ResNet‐18 architecture and 3,367,383 images to automatically classify wildlife species from camera trap images obtained from five states across the United States. We tested our model on an independent subset of images not seen during training from the United States and on an out‐of‐sample (or “out‐of‐distribution” in the machine learning literature) dataset of ungulate images from Canada. We also tested the ability of our model to distinguish empty images from those with animals in another out‐of‐sample dataset from Tanzania, containing a faunal community that was novel to the model. The trained model classified approximately 2,000 images per minute on a laptop computer with 16 gigabytes of RAM. The trained model achieved 98% accuracy at identifying species in the United States, the highest accuracy of such a model to date. Out‐of‐sample validation from Canada achieved 82% accuracy and correctly identified 94% of images containing an animal in the dataset from Tanzania. We provide an r package (Machine Learning for Wildlife Image Classification) that allows the users to (a) use the trained model presented here and (b) train their own model using classified images of wildlife from their studies. The use of machine learning to rapidly and accurately classify wildlife in camera trap images can facilitate non‐invasive sampling designs in ecological studies by reducing the burden of manually analysing images. Our r package makes these methods accessible to ecologists.

Telomeres are highly conserved regions of DNA that protect the ends of linear chromosomes. The loss of telomeres can signal an irreversible change to a cell's state, including cellular senescence. Senescent cells no longer divide and can damage nearby healthy cells, thus potentially placing them at the crossroads of cancer and ageing. While the epidemiology, cellular and molecular biology of telomeres are well studied, a newer field exploring telomere biology in the context of ecology and evolution is just emerging. With work to date focusing on how telomere shortening relates to individual mortality, less is known about how telomeres relate to ageing rates across species. Here, we investigated telomere length in cross-sectional samples from 19 bird species to determine how rates of telomere loss relate to interspecific variation in maximum lifespan. We found that bird species with longer lifespans lose fewer telomeric repeats each year compared with species with shorter lifespans. In addition, phylogenetic analysis revealed that the rate of telomere loss is evolutionarily conserved within bird families. This suggests that the physiological causes of telomere shortening, or the ability to maintain telomeres, are features that may be responsible for, or co-evolved with, different lifespans observed across species.This article is part of the theme issue ‘Understanding diversity in telomere dynamics'.

BackgroundThe movement behavior of an animal is determined by extrinsic and intrinsic factors that operate at multiple spatio-temporal scales, yet much of our knowledge of animal movement comes from studies that examine only one or two scales concurrently. Understanding the drivers of animal movement across multiple scales is crucial for understanding the fundamentals of movement ecology, predicting changes in distribution, describing disease dynamics, and identifying efficient methods of wildlife conservation and management.MethodsWe obtained over 400,000 GPS locations of wild pigs from 13 different studies spanning six states in southern U.S.A., and quantified movement rates and home range size within a single analytical framework. We used a generalized additive mixed model framework to quantify the effects of five broad predictor categories on movement: individual-level attributes, geographic factors, landscape attributes, meteorological conditions, and temporal variables. We examined effects of predictors across three temporal scales: daily, monthly, and using all data during the study period. We considered both local environmental factors such as daily weather data and distance to various resources on the landscape, as well as factors acting at a broader spatial scale such as ecoregion and season.ResultsWe found meteorological variables (temperature and pressure), landscape features (distance to water sources), a broad-scale geographic factor (ecoregion), and individual-level characteristics (sex-age class), drove wild pig movement across all scales, but both the magnitude and shape of covariate relationships to movement differed across temporal scales.ConclusionsThe analytical framework we present can be used to assess movement patterns arising from multiple data sources for a range of species while accounting for spatio-temporal correlations. Our analyses show the magnitude by which reaction norms can change based on the temporal scale of response data, illustrating the importance of appropriately defining temporal scales of both the movement response and covariates depending on the intended implications of research (e.g., predicting effects of movement due to climate change versus planning local-scale management). We argue that consideration of multiple spatial scales within the same framework (rather than comparing across separate studies post-hoc) gives a more accurate quantification of cross-scale spatial effects by appropriately accounting for error correlation.Electronic supplementary materialThe online version of this article (doi:10.1186/s40462-017-0105-1) contains supplementary material, which is available to authorized users.

All rights reserved. No reuse allowed without permission.(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. from Canada. We also tested the ability of our model to distinguish empty images from those 56 with animals in another out-of-sample dataset from Tanzania, containing a faunal community 57 that was novel to the model. (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint . http://dx.doi.org/10.1101/346809 doi: bioRxiv preprint first posted online Jun. 13, 2018; 3 4. The use of machine learning to rapidly and accurately classify wildlife in camera trap images 66 can facilitate non-invasive sampling designs in ecological studies by reducing the burden of 67 manually analyzing images. We present an R package making these methods accessible to 68 ecologists. We discuss the implications of this technology for ecology and considerations that 69 should be addressed in future implementations of these methods. 70