NABat ML: Utilizing deep learning to enable crowdsourced development of automated, scalable solutions for documenting North American bat populations

Journal Article

Title: NABat ML: Utilizing deep learning to enable crowdsourced development of automated, scalable solutions for documenting North American bat populations

Publication Date:

August 23, 2022

Journal:

Journal of Applied Ecology

Volume:

Issue:

Pages:

2849-2862

Publisher:

British Ecological Society

Affiliation

Colorado State University, US Geological Survey (USGS), US Fish and Wildlife Service (USFWS), Wyoming Game and Fish Department, US Army Engineer Research and Development Center, National Park Service, Florida Fish and Wildlife Conservation Commission, University of Nebraska, University of Maine, Bat Conservation International, University of Edinburgh, University of Prince Edward Island, University of Illinois, US Bureau of Land Management (BLM), Canadian Wildlife Health Cooperative, Colorado State University, Montana State University

Receptor:

Bats

Document Access

Website:

External Link

Attachment:

Access File

Citation

Khalighifar, A.; Gotthold, B.; Adams, E.; Barnett, J.; Beard, L.; Britzke, E.; Burger, P.; Chase, K.; Cordes, Z.; Cryan, P.; Ferrall, E.; Fill, C.; Gibson, S.; Haulton, G.; Irvine, K.; Katz, L.; Kendall, W.; Long, C.; Aodha, O.; McBurney, T.; McCarthy, S.; McKown, M.; O'Keefe, J.; Patterson, L.; Pitcher, K.; Rustand, M.; Segers, J.; Sappanen, K.; Siemers, J.; Stratton, C.; Straw, B.; Weller, T.; Reichert, B. (2022). NABat ML: Utilizing deep learning to enable crowdsourced development of automated, scalable solutions for documenting North American bat populations. Journal of Applied Ecology, 59(11), 2849-2862. https://doi.org/10.1111/1365-2664.14280

Download Citation File

Abstract

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 <16 hours.
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.