Miller, R. S. | Search bioRxiv

bioRxiv2018

Determinants of invasive species policy: print media and agriculture determine United States invasive wild pig policy

Conflicts between wildlife, invasive species, and agricultural producers are increasing. Although direct management actions taken to mitigate these conflicts remain controversial, most stakeholders agree that better policies are needed to balance socio-economic considerations with invasive species management, wildlife conservation, and agriculture. However the interaction between societal and biological drivers that influence human-invasive species-wildlife conflict mitigation policy is poorly understood. We identify factors influencing policy leading to the establishment of a new federal program to control invasive wild pigs in the United States. We fit generalized linear models relating frequency of congressional policy activity, such as congressional hearings and reports, to frequency of print newspaper media and percent of the U.S. agricultural industry co-occurring with invasive wild pigs for 29 years preceding the establishment of the federal program in 2013. Our models explained 89% of the deviance in congressional policy activity indicating a strong linkage between congressional invasive wild pig policy activity and predictors representing the number of negative of newspaper articles, geographic distribution of print media, and percent of agricultural producers co-occurring with invasive wild pigs. These effects translated to 3.7% increase in the number of congressional policy actions for every additional five states with negative news media. Invasive wild pig congressional policy activity increased 6.7% for every additional 10 negative newspaper articles. Increases in co-occurrence of agriculture and invasive wild pigs had the largest effect, for every 1% increase in co-occurrence there was a 41% increase in congressional policy activity. Invasive wild pig congressional policy activity that explicitly addressed livestock increased at nearly twice the rate of policy activity addressing crop agriculture. These results suggest that agriculture and media coverage may act as determinants for invasive species policy. Our approach may provide early insight into emerging policy areas enabling proactive policy development by agencies or early engagement by scientists to find solutions before the policy area becomes grid locked. Our results can also support policy and program evaluation providing a means of determining if the implemented policies match the original policy determinants ensuring best alignment with public, environmental, and stakeholder interests.

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↗