Robust data-driven uncertainty quantification in water cycle extreme predictions
- Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
Climate experts apply global climate models to predict the future climate, informing decisions and adaptations. Uncertainty plays a fundamental role in this process (particularly for the water cycle simulation, the most relevant but least well-simulated). For many aspects of the water cycle there are disagreements for future climate projections, reflecting the significant uncertainty in model formulation. Also, the chaotic nature of climate dynamics, as well as indeterministic future projection emission scenarios, contribute to the uncertainty in climate predictions. New AI methods that are beginning to be adopted by the climate community have the power to improve predictions of water cycle extremes. The crucial question is how to capture all sources of uncertainty in a systematic way in order for the predictions from the AI models to be maximally leveraged by decision makers
- Research Organization:
- Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)
- Sponsoring Organization:
- USDOE Office of Science (SC), Biological and Environmental Research (BER)
- Contributing Organization:
- Artificial Intelligence for Earth System Predictability (AI4ESP)
- OSTI ID:
- 1769775
- Report Number(s):
- AI4ESP1002
- Country of Publication:
- United States
- Language:
- English
Similar Records
Water Cycle-Driven Infectious Diseases as Multiscale, Reliable, Continuously Updating Water Cycle Sensors
Autonomous reinforcement learning agents for improving predictions and observations of extreme climate events