Biology subjects

Wolfson, D. W.

Publications and source records attributed to Wolfson, D. W..

4 recordsLinked to original sources

bioRxiv2025

Landscape suitability and range expansion estimates for the North American Interior Population of trumpeter swans

The Interior Population of trumpeter swans (Cygnus buccinator) has grown and expanded substantially since reintroduction efforts began in the 1950s. To support management of this population, we developed a landscape suitability model and range expansion estimates (2023 2033). We assessed landscape suitability for breeding trumpeter swans with a use-available modeling design based on GPS collar data (20191iJ2023) and derived home range information from swans in the western Great Lakes Region of the U.S. and Canada. We related occurrence and pseudo-absences to landscape-scale summaries of upland and wetland conditions in a logistic mixed-effects model. We spatially applied the model and incorporated these estimates into a range-expansion model that identified the likelihood of a 50-km grid cell being occupied by 2033. We summarized time-series citizen-science data within cells and related newly occupied range cells and unoccupied cells (2004 2023) to year, survey effort, the amount of suitable breeding landscapes, and distance to previously occupied cells in a logistic regression model. Landscape suitability was positively related to greater amounts of wetland perimeter and foraging areas. New range cells were positively associated with suitable breeding landscapes and survey intensity and negatively associated with year and distance to previously occupied cells. We estimated a 4.4% (95% CI: 2.0-6.9%) annual range expansion rate from 2023 to 2033, with expansion occurring in the Prairie Pothole Region of the Dakotas and the Boreal Shield and James Bay Lowlands of Canada. North Dakota and South Dakota allow tundra swan (Cygnus columbianus) hunting but not for trumpeter swans. Our models can be useful for targeting hunter education at a local scale to mitigate unintended take of the similar looking species, allowing for pioneering trumpeter swans to establish persistent populations. Furthermore, our range expansion model can help jurisdictions with no or low breeding trumpeter swan populations anticipate future swan distribution and abundance.

ecology↗

bioRxiv2021

Transient population dynamics drive the spread of invasive wild pigs in North America

Invasion of nonindigenous species is considered one of the most urgent problems affecting native ecosystems and agricultural systems. Mechanistic models that account for short-term population dynamics can improve prediction because they incorporate differing demographic processes that link the environmental conditions of a spatial location explicitly with the invasion process. Yet short-term population dynamics are rarely accounted for in spatial models of invasive species spread. Accounting for transient population dynamics, we predict the population growth rate and establishment probability of wild pigs following introduction into any location in North America. We compared predicted population growth rate with observed geographic rates of spread and found significant relationships between the annual rate of spread and population growth rates. We used geospatial data on the distribution of mast producing tree species (a principle forage resource of wild pigs) and agricultural crops that can replace mast in their diets to predict population dynamics using transient population simulations. We simulated populations under different initial population sizes (i.e. number of introduced individuals, often termed propagule size) and for different amounts of time following introduction. By varying the initial population size and simulation time, we were able to identify areas in North America with high probability for establishment of wild pigs if introduced. Our findings can be used to inform surveillance and removal efforts to reduce the potential for establishment and spread of wild pigs.

ecology↗

bioRxiv2020

Improving the accessibility and transferability of machine learning algorithms for identification of animals in camera trap images: MLWIC2

O_LIMotion-activated wildlife cameras (or "camera traps") are frequently used to remotely and non-invasively observe animals. The vast number of images collected from camera trap projects have prompted some biologists to employ machine learning algorithms to automatically recognize species in these images, or at least filter-out images that do not contain animals. These approaches are often limited by model transferability, as a model trained to recognize species from one location might not work as well for the same species in different locations. Furthermore, these methods often require advanced computational skills, making them inaccessible to many biologists. C_LIO_LIWe used 3 million camera trap images from 18 studies in 10 states across the United States of America to train two deep neural networks, one that recognizes 58 species, the "species model," and one that determines if an image is empty or if it contains an animal, the "empty-animal model." C_LIO_LIOur species model and empty-animal model had accuracies of 96.8% and 97.3%, respectively. Furthermore, the models performed well on some out-of-sample datasets, as the species model had 91% accuracy on species from Canada (accuracy range 36-91% across all out-of-sample datasets) and the empty-animal model achieved an accuracy of 91-94% on out-of-sample datasets from different continents. C_LIO_LIOur software addresses some of the limitations of using machine learning to classify images from camera traps. By including many species from several locations, our species model is potentially applicable to many camera trap studies in North America. We also found that our empty-animal model can facilitate removal of images without animals globally. We provide the trained models in an R package (MLWIC2: Machine Learning for Wildlife Image Classification in R), which contains Shiny Applications that allow scientists with minimal programming experience to use trained models and train new models in six neural network architectures with varying depths. C_LI

ecology↗

bioRxiv2018

Machine learning to classify animal species in camera trap images: applications in ecology

1. Motion-activated cameras (\"camera traps\") are increasingly used in ecological and management studies for remotely observing wildlife and have been regarded as among the most powerful tools for wildlife research. However, studies involving camera traps result in millions of images that need to be analyzed, typically by visually observing each image, in order to extract data that can be used in ecological analyses.\n\n2. 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.\n\n3. 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; MLWIC) that allows the users to A) implement the trained model presented here and B) train their own model using classified images of wildlife from their studies.\n\n4. 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 analyzing images. We present an R package making these methods accessible to ecologists. We discuss the implications of this technology for ecology and considerations that should be addressed in future implementations of these methods.

ecology↗