NCAR scientists used machine learning to emulate the results of the "bin microphysics" parameterization, a package of equations used to simulate the formation and evolution of clouds inside a climate model. While bin microphysics gives a more realistic representation of clouds than the simpler "bulk microphysics" scheme, scientists often cannot afford the computing resources needed to run bin microphysics for long periods of time. The use of machine learning may allow scientists to approximate the results of bin microphysics in a computationally efficient way. To answer critical questions about the climate and how it's changing, scientists are pressing sophisticated Earth system models to solve increasingly complex equations. The result is more detailed simulations -- and also more demand for the scarce supercomputing resources needed to run them.