AI and ML used to advance Earth science and predictability.
AI has the potential to revolutionize our understanding of Earth’s systems by helping to collect, understand, and analyze large sets of data. To explore the potential of AI, the U.S. Department of Energy’s Office of Biological and Environmental Research and Advanced Scientific Computing Research program held the Artificial Intelligence for Earth System Predictability (AI4ESP) workshop from October through December 2021.
The workshop brought together over 700 participants from the private and public sectors, including representatives from Earth and environmental sciences, computing, and AI. Participants discussed 17 topics related to the integrative water cycle and extreme weather phenomena, as well as nine focal points related to Earth system predictions. Through the sessions, experts explored the potential of AI to unlock scientific discoveries using tools such as neural networks, knowledge-informed machine learning, AI architectures, and co-design.
The workshop identified several challenges that support the need for revolutionized AI technology and infrastructure. This includes managing and analyzing large sets of data to enhance the ability to observe and predict extreme events, as well as fostering the integration of human activities into theory and models. Experts also identified opportunities for AI research and development to help tackle the greatest challenges facing Earth science.
In addition to technological advancements, the workshop also identified a need for programmatic and cultural changes to support a more cohesive mission across scientific and government agencies. This includes developing AI research centers specific to environmental science, as well as frameworks that enable shared services across various communities.
The 2021 AI4ESP workshop participants continue to discuss community computational activities, including those from the American Geophysical Union and American Meteorological Society. Through further collaborations, engagements, and framework development, AI4ESP will continue to further its mission.