Use of a Detailed Planetary Boundary Layer Model for Parameterization Purposes

Abstract

Use is made of a three-dimensional model of the planetary boundary layer to investigate features of non-precipitating convection and its parameterization. The fine-grid mesh (50 m) model was first developed by J.W. Deardorff (1973) and modified by G. Sommeria (1976) to include possible cloud formation; the model and data set are briefly described in Section 2. In Section 3 the model data are utilized in order to test the validity of certain hypotheses concerning individual cloud features as well as those of a cloud population. These hypotheses, for the most part unverified in the real atmosphere, are frequently used in cumulus parameterization schemes. Sections 4 and 5 present the results of three recent cumulus parameterization schemes by Betts (1975, 1976) and Fraedrich (1976), and tests are made of these schemes with the present model. In order to illustrate the versatility of such a model when compared to the possibilities of intensive studies using atmospheric data, an attempt is made in Section 6 to link turbulent quantities with directly observable cloud features. The turbulent fluxes of heat and moisture, for example, are not easily observable but are important to cloud development and evolution. The empirical relations presented in this section present a reasonable alternative to atmospheric observations for the estimation of various important turbulent quantities. The study presented here emphasizes that despite many drawbacks, a numerical model is a useful tool to increase our understanding of a particular atmospheric phenomenon and can be used as a complement to observations for the improvement of parameterization methods in larger scale models.