New research from Columbia Engineering Professor Pierre Gentine demonstrates that machine-learning techniques can be used to accurately represent clouds and their atmospheric heating and moistening, and better represent clouds in coarse resolution climate models, with the potential to narrow the range of climate prediction. This could be a major advance in accurate predictions of global warming in response to increased greenhouse gas concentrations that are essential for policy-makers (e.g. the Paris climate agreement).
- Geophysical Research Letters
- Department of Energy SciDac and Early Career Programs, National Science Foundation