Understanding the behavior of tropical cyclones in order to better predict their intensities and path is the ultimate goal of this research. One way to gain insight is to use numerical models to examine various aspects of the circulation. Two topics will be pursued under this award. The first is a study of the role of latent heat release from cumulus convection in cyclone development and how the effects can best be represented in numerical models. An important part of the investigation will be to compare simulations using two different cumulus parameterization schemes. The second area involves a study of the role of the beta-effect, or latitudinal variation of the Coriolis acceleration, on the motion of tropical storms. The study will be done with a spectral model which was developed under a previous NSF grant. The research to be conducted under this award will be for the completion of a graduate student thesis.