A general deep learning model for bird detection in high-resolution airborne imagery.

Advances in artificial intelligence for computer vision hold great promise for increasing the scales at which ecological systems can be studied. The distribution and behavior of individuals is central to ecology, and computer vision using deep neural networks can learn to detect individual objects in imagery. However, developing supervised models for ecological monitoring is challenging because it requires large amounts of human-labeled training data, requires advanced technical expertise and computational infrastructure, and is prone to overfitting. This limits application across space and time. One solution is developing generalized models that can be applied across species and ecosystems. Using over 250,000 annotations from 13 projects from around the world, we develop a general bird detection model that achieves over 65% recall and 50% precision on novel aerial data without any local training despite differences in species, habitat, and imaging methodology. Fine-tuning this model with only 1000 local annotations increases these values to an average of 84% recall and 69% precision by building on the general features learned from other data sources. Retraining from the general model improves local predictions even when moderately large annotation sets are available and makes model training faster and more stable. Our results demonstrate that general models for detecting broad classes of organisms using airborne imagery are achievable. These models can reduce the effort, expertise, and computational resources necessary for automating the detection of individual organisms across large scales, helping to transform the scale of data collection in ecology and the questions that can be addressed.

Microclimate and resource quality determine resource use in a range-expanding herbivore.

The consequences of climate change for biogeographic range dynamics depend on the spatial scales at which climate influences focal species directly and indirectly via biotic interactions. An overlooked question concerns the extent to which microclimates modify specialist biotic interactions, with emergent properties for communities and range dynamics. Here, we use an in-field experiment to assess egg-laying behaviour of a range-expanding herbivore across a range of natural microclimatic conditions. We show that variation in microclimate, resource condition and individual fecundity can generate differences in egg-laying rates of almost two orders of magnitude in an exemplar species, the brown argus butterfly (Aricia agestis). This within-site variation in fecundity dwarfs variation resulting from differences in average ambient temperatures among populations. Although higher temperatures did not reduce female selection for host plants in good condition, the thermal sensitivities of egg-laying behaviours have the potential to accelerate climate-driven range expansion by increasing egg-laying encounters with novel hosts in increasingly suitable microclimates. Understanding the sensitivity of specialist biotic interactions to microclimatic variation is, therefore, critical to predict the outcomes of climate change across species' geographical ranges, and the resilience of ecological communities.