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.
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 No reuse allowed without permission.(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.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 (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint .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
Heterogeneous landscapes and fluctuating environmental conditions can affect species dispersal, population genetics, and genetic structure, yet understanding how biotic and abiotic factors affect population dynamics in a fluctuating environment is critical for species management. We evaluated how spatio-temporal habitat connectivity influences dispersal and genetic structure in a population of boreal chorus frogs (Pseudacris maculata) using a landscape genetics approach. We developed gravity models to assess the contribution of various factors to the observed genetic distance as a measure of functional connectivity. We selected (a) wetland (within-site) and (b) landscape matrix (between-site) characteristics; and (c) wetland connectivity metrics using a unique methodology. Specifically, we developed three networks that quantify wetland connectivity based on: (i) P. maculata dispersal ability, (ii) temporal variation in wetland quality, and (iii) contribution of wetland stepping-stones to frog dispersal. We examined 18 wetlands in Colorado, and quantified 12 microsatellite loci from 322 individual frogs. We found that genetic connectivity was related to topographic complexity, within- and between-wetland differences in moisture, and wetland functional connectivity as contributed by stepping-stone wetlands. Our results highlight the role that dynamic environmental factors have on dispersal-limited species and illustrate how complex asynchronous interactions contribute to the structure of spatially-explicit metapopulations.
BioOne Complete (complete.BioOne.org) is a full-text database of 200 subscribed and open-access titles in the biological, ecological, and environmental sciences published by nonprofit societies, associations, museums, institutions, and presses.
A functionalized dialkylperylene and a modified terrylenetetracarboxdiimide (TTCDI) were joined together by a hexanediyl spacer. The resulting bichromophoric molecule 4 is a suitable model system for donor ± acceptor energy transfer studies at the single-molecule level. With its absorption and fluorescence maximum at shorter wavelengths (l max 450 nm, l fl 458 nm) the dialkylperylene acts as the donor, while the TTCDI with l max 665 nm and l fl 705 nm is the acceptor molecule. The synthetic route to the new bichromophoric molecule and its optical properties (UV/Vis, fluorescence spectroscopy) are presented herein. Energy transfer from the perlyene to the terrylene moiety was confirmed by conventional ensemble fluorescence excitation and emission spectroscopy. Single bichromophores 4 embedded in polyvinylbutyral were imaged with a scanning confocal optical microscope at room temperature by selectively exciting the perylene chromophore and detecting the terrylene emission.
Contact heterogeneity among hosts determines invasion and spreading dynamics of infectious disease, thus its characterization is essential for identifying effective disease control strategies. Yet, little is known about the factors shaping contact networks in many wildlife species and how wildlife management actions might affect contact networks. Wild pigs in North America are an invasive, socially structured species that pose a health concern for domestic swine given their ability to transmit numerous devastating diseases such as African swine fever (ASF). Using proximity loggers and GPS data from 48 wild pigs in Florida and South Carolina, USA, we employed a probabilistic framework to estimate weighted contact networks. We determined the effects of sex, social group and spatial distribution (monthly home‐range overlap and distance) on wild pig contact. We also estimated the impacts of management‐induced perturbations on contact and inferred their effects on ASF establishment in wild pigs with simulation. Social group membership was the primary factor influencing contacts. Between‐group contacts depended primarily on space use characteristics, with fewer contacts among groups separated by >2 km and no contacts among groups >4 km apart within a month. Modelling ASF dynamics on the contact network demonstrated that indirect contacts resulting from baiting (a typical method of attracting wild pigs or game species to a site to enhance recreational hunting) increased the risk of disease establishment by ~33% relative to direct contact. Low‐intensity population reduction (<5.9% of the population) had no detectable impact on contact structure but reduced predicted ASF establishment risk relative to no population reduction. We demonstrate an approach for understanding the relative role of spatial, social and individual‐level characteristics in shaping contact networks and predicting their effects on disease establishment risk, thus providing insight for optimizing disease control in spatially and socially structured wildlife species.
Home range is expected to vary with ecological conditions to minimize size while still meeting the biological needs of the individual animal. Understanding the determinants of variation in home range size can be important when trying to manage or control an invasive species. Wild pigs (Sus scrofa) have been introduced throughout the globe and cause notable damage to native ecosystems. We quantified relationships of wild pig home range with environmental conditions across varying spatial and temporal scales to better understand population space use in areas invaded by wild pigs during March 2011 to August 2012 in Kent County, Texas, USA. We used mixed‐effects linear‐regression models to assess how allometric effects of body size and environmental variables (temp, elevation, latitude, and rainfall) could be used to predict home range size at a local scale. We then used general linear models with published data from 31 studies from the species’ global distribution to investigate the efficacy of environmental parameters as home range predictors. On account of either temporal or incomplete variables, home range was not well‐predicted at a local scale. Across their global distribution, the top ranked model included all 4 variables with home range positively associated with temperature, elevation, and latitude, but negatively associated with rainfall. Use of the global model represents a cost‐efficient way to estimate home ranges to control or eradicate wild pig populations. This information can be valuable for management of both established and newly introduced populations for population estimation and trapping density. © 2016 The Wildlife Society.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.