A Cumulus Parameterization Based on a Cloud Model of Intermittently Rising Thermals

Abstract

The author presents a cumulus parameterization that uses a cloud model that describes atmospheric convection as consisting of a sequence of intermittently rising thermals. The total mass of thermals in a convection event is determined by the amount of convective available potential energy in local soundings. In the cloud model, it is assumed that a thermal entrains environmental air only at a thin layer around the top frontier of its rising body. The entrained air mass mixes with the thermal?s air and produces ?mixtures? that then detach themselves from the thermal. This limited mixing prevents deep erosion to the thermal?s buoyancy by entrainment and mixing processes. The remainder of the thermal continues rising to higher levels and forming more mixtures on its way to its own level of neutral buoyancy. The mixtures also rise or sink from the levels where they form to their level of neutral buoyancy. Evaluation of this scheme using Global Atmospheric Research Program Atlantic Tropical Experiment data shows that the parameterized convective heating and drying rates are consistent with observations. The calculated convective precipitation also shows a distribution similar to the observed total precipitation, except at the trough of the easterly waves where calculated precipitation is smaller than observed. The capability of this scheme in describing cumulus convection is further tested in a fully prognostic one-dimensional climate model. Results from this evaluation show reasonable climatological temperature and relative humidity profiles in the troposphere.