Anthropogenic aerosol pollution continues to be one of the main sources of uncertainty in present day human influence on the climate. Most of the present-day anthropogenic aerosol forcing (which is negative) will likely disappear in the coming 20-30 years, unmasking an uncertain amount of greenhouse warming. Unfortunately, observations of the quantities necessary to constrain anthropogenic aerosol forcing are difficult, making Earth System Models (ESMs) an important tool for constraining aerosol forcing. A significant portion of the spread in present day anthropogenic aerosol forcing among ESMs is likely due to fundamental differences in aerosol scheme formulation, rather than differences in physical parameters. I will show results from a newly available 2-moment aerosol microphysics scheme (the piecewise lognormal approximation aerosol model, PAM) in the Canadian ESM (CanESM) to isolate uncertainty associated with aerosol model formulation in this model. Specifically, I will analyze a pair of CanESM configurations with identical model parameters that differ only in their aerosol schemes: one configuration uses PAM while the other uses the default bulk aerosol scheme, which was the workhorse configuration used in CMIP5 and CMIP6. Both the present-day aerosol climatology and anthropogenic forcing differ substantially between PAM and bulk, due to differences in cloud droplet activation, in-cloud sulfate chemistry, and gravitational settling, among others.