Deep learning to extract the meteorological by-catch of wildlife cameras
Permanent lenke
https://hdl.handle.net/10037/34992Dato
2023-12-09Type
Journal articleTidsskriftartikkel
Peer reviewed
Forfatter
Alison, Jamie; Payne, Stephanie; Alexander, Jake M.; Bjorkman, Anne D.; Clark, Vincent Ralph; Gwate, Onalenna; Huntsaar, Maria; Iseli, Evelin; Lenoir, Jonathan; Mann, Hjalte Mads Rosenstand; Steenhuisen, Sandy-Lynn; Høye, Toke ThomasSammendrag
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 snow falls 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) imagederived 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.
Forlag
WileySitering
Alison, Payne, Alexander, Bjorkman, Clark, Gwate, Huntsaar, Iseli, Lenoir, Mann, Steenhuisen, Høye. Deep learning to extract the meteorological by-catch of wildlife cameras. Global Change Biology. 2024;30(1)Metadata
Vis full innførselSamlinger
Copyright 2023 The Author(s)