A cloudless future? The mystery at the heart of climate forecasts

Analyses of global climate models consistently show that clouds constitute the biggest source of uncertainty and instability in predictions. New research on the Frontera supercomputer seeks to better incorporate clouds into global models by breaking models into two parts: a coarse-grained, lower-resolution (100km) planetary model and many small patches with 100 to 200 meter resolution. These simulations can capture the physical processes and turbulent eddies involved in cloud formation and do not produce unwanted side-effects.