Temporal Coverage over Density: Parsimonious Training-Set Design for ML Climate Downscaling
Researchers have developed a new method for training machine learning models to downscale climate data, focusing on how to select training years effectively. Their study, using the CESM2 Large Ensemble, found that training models on years distributed across the entire climate trajectory, rather than contiguous historical periods, significantly improves their ability to reproduce climate variability. This approach, even with limited data, outperforms models trained solely on historical data and suggests that broad sampling of climate states is more beneficial than temporal continuity for allocating scarce high-resolution simulation resources. AI
IMPACT Optimizes training data selection for climate models, potentially improving accuracy and efficiency in climate impact assessments.