The sensitivity of a modelled tropical cyclone to SST and CAPE

In this thesis the sensitivity of a modelled tropical cyclone to sea surface temperature (SST) and Convective Available Potential Energy (CAPE) is investigated. In Chapter 1 previous studies on this topic are discussed and an overview is given of our current knowledge of the effects of latent and sensible heat fluxes from the ocean on tropical cylcones. In Chapter 2 the experiments that make up this study are descibed in some detail. They have been carried out to compare the importance of SST to that of CAPE on tropical cyclone intensity. In order to achieve this, three experiments were designed, a control experiment, an experiment with increased SST and finally an experiment with increased SST, but CAPE equal to that of the control run. The simulations were carried out with the CSIRO Division of Atmospheric Research Limited area model using the Betts Miller cumulus parameterisation scheme. Both the model and this scheme are described in Chapter 3. The tropical cyclone used in this study is Tropical Cyclone Connie, which occurred in the Timor Sea from 15 to 23 January. The track and central pressure of the control simulation are compared with the observations and other simulations of the same storm. The simulation in this study compares very well to the other simulations and given its coarse resolution of 76 x 76 km, produces an acceptable simulation of cyclone Connie, given its coarse resolution of 76 x 76 km. When comparing the three experiments to each other it is found that the storm with increased SST and increased CAPE produces the strongest tropical cyclone. Finally, the sensitivity of the Betts Miller scheme to a model parameter called the relaxation time is discussed briefly. The relaxation time simulates the lag between the large scale forcing and the convective response. It was found that a decrease in relaxation time leads to faster convective adjustment and deeper tropical cyclones. It is concluded in Chapter 5 that SST is not as obviously tied to tropical cyclone intensity as was previously thought. Evaporation from the surface was found to play a very important role, but the importance of CAPE is still unclear. For that reason it will be necessary to perform more simulations with other convective schemes that respond directly to CAPE and are independent of external parameters that have not been verified against observations.