Appendix A: Statement of Task

The National Academies of Sciences, Engineering, and Medicine will organize a workshop to bring together experts to explore the opportunities and challenges of machine learning and artificial intelligence (ML/AI) to advance Earth system science. The workshop will explore how these approaches can contribute to improving understanding, analysis, modeling, prediction, and decision making. Specific topics to be addressed could include:

• Review current applications of ML/AI to Earth system science.
• Survey emerging ML/AI technologies and approaches that could be useful for Earth system science, such as:
  • Integrating physics, expert knowledge, and ML/AI techniques to enrich models and forecasts;
  • Understanding and interpreting results produced by ML models; and
  • Bridging research and operations to support decisionmaking.
• Consider challenges and risks of using ML/AI for Earth system science, and discuss ways to mitigate these risks, such as:
  • Justice, equity, diversity, and inclusion (JEDI) and ethical implications;
  • The interface of ML/AI with existing and emerging hardware, software, tools, and approaches;
  • Workforce capacity and skill sets; and
  • Data requirements and limitations.
• Identify future opportunities to accelerate progress, such as:
  • Novel funding mechanisms and partnerships; and
  • Potential disruptions that could lead to rapid scientific, societal, and/or institutional progress.

This page intentionally left blank.