EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Deep learning to extract the meteorological by‐catch of wildlife cameras

Jamie Alison iD, Stephanie Payne iD, Jake M Alexander iD, Anne D Bjorkman iD, Vincent Ralph Clark iD, Onalenna Gwate iD, Maria Huntsaar, Evelin Iseli iD, Jonathan Lenoir iD, Hjalte Mads Rosenstand Mann iD, Sandy-Lynn Steenhuisen iD, Toke Thomas Høye iD

Global Change Biology · 2024

DOI: 10.1111/gcb.17078 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Global Change Biology, 2024. Verfasst von Jamie Alison und 11 weiteren Autor:innen. Beteiligte Einrichtung: Department of Ecoscience, Aarhus University, Aarhus, Denmark. Hauptthemen laut MEDLINE: Animals, Wild, Deep Learning. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 66 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: Department of Science and Innovation, South Africa, Innovationsfonden, Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Microclimate—proximal climatic variation at scales of metres and minutes—can exacerbate or mitigate the impacts of climate change on biodiversity. However, most microclimate studies are temperature centric, and do not consider meteorological factors such as sunshine, hail and snow. Meanwhile, remote cameras have become a primary tool to monitor wild plants and animals, even at micro‐scales, and deep learning tools rapidly convert images into ecological data. However, deep learning applications for wildlife imagery have focused exclusively on living subjects. Here, we identify an overlooked opportunity to extract latent, ecologically relevant meteorological information. We produce an annotated image dataset of micrometeorological conditions across 49 wildlife cameras in South Africa's Maloti‐Drakensberg and the Swiss Alps. We train ensemble deep learning models to classify conditions as overcast, sunshine, hail or snow. We achieve 91.7% accuracy on test cameras not seen during training. Furthermore, we show how effective accuracy is raised to 96% by disregarding 14.1% of classifications where ensemble member models did not reach a consensus. For two‐class weather classification (overcast vs. sunshine) in a novel location in Svalbard, Norway, we achieve 79.3% accuracy (93.9% consensus accuracy), outperforming a benchmark model from the computer vision literature (75.5% accuracy). Our model rapidly classifies sunshine, snow and hail in almost 2 million unlabelled images. Resulting micrometeorological data illustrated common seasonal patterns of summer hailstorms and autumn snowfalls across mountains in the northern and southern hemispheres. However, daily patterns of sunshine and shade diverged between sites, impacting daily temperature cycles. Crucially, we leverage micrometeorological data to demonstrate that (1) experimental warming using open‐top chambers shortens early snow events in autumn, and (2) image‐derived sunshine marginally outperforms sensor‐derived temperature when predicting bumblebee foraging. These methods generate novel micrometeorological variables in synchrony with biological recordings, enabling new insights from an increasingly global network of wildlife cameras.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Animals, WildDeep LearningAnimalsHumansWeatherSnowBiodiversity

Stichwörter der Autor:innen

Trifolium pratensealpine ecologyautomated monitoringbeesmicrometeorologyproximal sensingsnow melttime-lapse photography

Autor:innen und Institutionen

  • Department of Ecoscience, Aarhus University, Aarhus, Denmark.
  • Afromontane Research Unit and Department of Plant Sciences, University of the Free State, Bloemfontein, South Africa.
  • Institute of Integrative Biology, ETH Zurich, Zurich, Switzerland.
  • Department of Biological and Environmental Sciences, University of Gothenburg, Gothenburg, Sweden.
  • Gothenburg Global Biodiversity Centre, Gothenburg, Sweden.
  • Afromontane Research Unit and Department of Geography, University of the Free State, Bloemfontein, South Africa.

ORCID-Profile: Jamie Alison, Stephanie Payne, Jake M Alexander, Anne D Bjorkman, Vincent Ralph Clark, Onalenna Gwate, Evelin Iseli, Jonathan Lenoir, Hjalte Mads Rosenstand Mann, Sandy-Lynn Steenhuisen, Toke Thomas Høye

Förderung

  • Department of Science and Innovation, South Africa
  • Innovationsfonden
  • Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung

Zitieren

APA

Alison, J., Payne, S., Alexander, J. M., Bjorkman, A. D., Clark, V. R., Gwate, O., Huntsaar, M., Iseli, E., Lenoir, J., Mann, H. M. R., Steenhuisen, S. L., & Høye, T. T. (2024). Deep learning to extract the meteorological by‐catch of wildlife cameras. Global Change Biology. https://doi.org/10.1111/gcb.17078

Vancouver

Alison J, Payne S, Alexander JM, Bjorkman AD, Clark VR, Gwate O, et al. Deep learning to extract the meteorological by‐catch of wildlife cameras. Global Change Biology. 2024. doi:10.1111/gcb.17078

BibTeX
@article{Alison2024,
  author = {Alison, Jamie and Payne, Stephanie and Alexander, Jake M and Bjorkman, Anne D and Clark, Vincent Ralph and Gwate, Onalenna and Huntsaar, Maria and Iseli, Evelin and Lenoir, Jonathan and Mann, Hjalte Mads Rosenstand and Steenhuisen, Sandy-Lynn and Høye, Toke Thomas},
  title = {Deep learning to extract the meteorological by‐catch of wildlife cameras},
  journal = {Global Change Biology},
  year = {2024},
  doi = {10.1111/gcb.17078}
}
Kostenlos starten →