Emerging machine learning approaches, combined with growing observational records and open data, are transforming ecosystem sciences by facilitating hypothesis testing and unlocking data-driven discoveries. Innovations in AI as well as creative integration of data-driven and process-based approaches grounded in biophysical laws and causal reasoning are deepening our insights into ecosystem functioning and driving applied knowledge to address land and resource management challenges posed by climate variability and global change.

This session aims to spark dialogue on opportunities and challenges of applying AI and machine learning to enhance ecosystem process understanding and management. We invite contributions using a range of advances, including foundation models, generative AI, large language models, knowledge-guided or physics-informed machine learning, differentiable modeling, digital twins, causal inference, trustworthy AI, transfer learning, information theory, symbolic regression, explainable AI, self-supervised learning, and uncertainty-aware modeling. We welcome applications across ecosystem ecology, ecohydrology, biogeochemistry, agroecology, biodiversity, nature-based climate solutions, conservation biology etc.

• GC - Global Environmental Change