Abstract
A set of independent prognostic variables, based on a survey of the microphysical timescales in clouds, is proposed for long-term cloud-resolving model simulations. Two of the variables are the moist entropy and the total mixing ratio of airborne water with no contributions from precipitating particles. Non-prognostic variables such as air temperature can be diagnosed from the prognostic variables easily. In this proposed modeling framework, moist thermodynamics is separated (or modularized) from cloud dynamics and microphysics. Numerical results are compared with analytic solutions to show that the proposed prognostic variables work well when a large time step (e.g., 10 s) is used for numerical integration.
