EcoCast: A Spatio-Temporal Model for Continual Biodiversity and Climate Risk Forecasting

Increasing climate change and habitat loss are driving unprecedented shifts in species distributions. Conservation professionals urgently need timely, high-resolution predictions of biodiversity risks, especially in ecologically diverse regions like Africa. We propose EcoCast, a novel spatio-temporal model designed for continual biodiversity and climate risk forecasting. Utilizing multisource satellite imagery, climate data, and citizen science occurrence records, EcoCast predicts near-term shifts in species distributions. We describe a three-phase approach: (1) constructing a large-scale, pre-trained spatio-temporal architecture, (2) continually fine-tuning with new data streams, and (3) developing interactive dashboards for conservation practitioners. Our pilot study in Africa shows promising improvements in forecasting distributions of selected bird species compared to a Random Forest baseline, highlighting EcoCast's potential to inform targeted conservation policies. By providing an end-to-end system, from data ingestion and model refinement to user-focused impact tools, EcoCast aims to bridge the gap between cutting-edge machine learning and biodiversity management, ultimately guiding data-driven strategies for climate resilience and ecosystem conservation throughout Africa.