National Center for Computational Sciences

Global climate models are helping scientists predict the response of the climate system to natural and anthropogenic aerosol forcing. A sudden aerosol forcing releases chemicals in quantities not typically circulated in the atmosphere, as would occur after a volcanic eruption. The density of aerosols affects the amount and type of solar radiation that reaches the earth’s surface, which in turn impacts climate.

This project uses the power of the petascale Cray Jaguar supercomputer to simulate global weather conditions from 1960 to the present. The global scale model will be run multiple times and assimilated to create an average climate pattern for the four decades. The capabilities of Jaguar allow the simulations to use a horizontal resolution down to 37 kilometers, which allows the model to resolve finer scale events such as hurricanes and subcontinental precipitation patterns. The team will be able to compare the simulated climate reaction to aerosol forcing with observation to test the predictive skill of the climate model at regional and decadal scales.