Artificial Intelligence for Climate Change Biology: From Data Collection to Predictions.

This article reviews the emerging role of artificial intelligence (AI) and machine learning (ML) in advancing thermal ecology research, particularly in understanding animal responses to climate change through improved microclimate modeling and behavioral tracking. It highlights how AI techniques, including deep neural networks and other ML models, enhance bias correction and downscaling of climate data to fine spatial resolutions, enabling more accurate predictions of the microclimates animals experience. The integration of AI with sensor technologies such as accelerometers, acoustic loggers, and camera traps facilitates detailed classification of thermoregulatory behaviors and microhabitat usage, providing critical data for refining biophysical niche models. These advancements support more precise ecological predictions and inform conservation strategies aimed at identifying climate refugia and designing climate-resilient habitats. The article also discusses challenges in AI adoption, including data limitations and the need for combining mechanistic models with ML approaches to improve robustness under novel climate scenarios.