No abstract

[1] Recent research suggests that anthropogenic global warming would be associated with an increase in the intensity of tropical cyclones. A recent statistical analysis of observed tropical cyclone intensity shows that its variability with location and season is strongly tied to the variability of the thermodynamic potential intensity (PI) of tropical cyclones, as calculated using a theory described in an earlier work by the authors. Thus it is of interest to look for possible trends in global measures of PI, which are far more stable than those of actual storm intensity. We estimate global trends of PI from 1958 to 1996, averaged over the region where it exceeds 40 m s À1 , using the National Centers for Environmental Prediction/National Center for Atmospheric Research (NCEP/NCAR) Reanalysis and the NCEP Empirical Orthogonal Function (EOF) sea surface temperature (SST) analysis. We adjust the Reanalysis temperatures for a large, spurious temperature increase that occurred around 1979. We do this by subtracting from the Reanalysis the atmospheric temperature difference between pairs of years with similar tropical SST before and after 1979. The value of the global mean PI is very large for the SST of the corresponding region in the mid-1990s. Supported by a recent study on the effects of ozone decrease on tropospheric temperatures, we suggest that the ozone decrease might be one of the factors contributing to increase of PI during the 1990s.

The transformation of a mesoscale convective system into Hurricane Guillermo was captured by aircraft and Doppler wind data during the Tropical Experiment in Mexico. The early phase of the system evolves in a way very similar to previously documented mesoscale convective systems, with a midlevel mesocyclone developing in the stratiform precipitation region. More unusually, the cyclone extends to low altitudes: A weak cyclone is discernable even in the 300-m altitude wind field. After another day of evolution, a small, surface-based warmcore cyclone is observed to develop within the relatively cold air associated with the mesocyclone aloft. This mesocyclone develops into a hurricane over the subsequent day. A nonhydrostatic, axisymmetric numerical model is used to explore the evolution of the initially cold-core, midlevel vortex into a tropical cyclone. A mesoscale, midlevel ''showerhead'' is switched on in an initially quiescent, tropical atmosphere overlying a warm ocean surface. Evaporation of the falling rain cools the lower troposphere and leads to the spinup of a midlevel vortex, while divergent outflow develops near the surface. After some time, the midlevel vortex expands downward toward the boundary layer, and later a warm-core, surface-flux-driven cyclone develops within it. Experiments with the model show that both the cyclone, with its associated cold anomaly, and the relatively humid air associated with the evaporatively cooled mesoscale cyclone are important for the subsequent development of the warm-core system. The simulations also suggest that, for the warm-core development to occur, the stratiform rain must last long enough to drive the midlevel vortex down to the boundary layer. The authors present a simple argument for why this must be so and suggest that this process would be significantly impeded by the presence of background vertical wind shear.

No abstract