Understanding the Climatological Connection of Tropical Cyclone Intensity and the Tropopause Variability

Substantial effort has been devoted to modeling tropical cyclone (TC) climatology in order to understand how tropical cyclone distribution and frequency will be impacted in the future climate. Various modeling studies have shown that an increase in sea surface temperature (SST) will lead to a decrease in overall TC frequency but an increase in frequency of intense TC (Category 4-5). However, little research has been done on upper level atmospheric conditions in relation to TC activity. Recent real-time TC forecasts in the north Western Pacific (WPAC) basin using the Hurricane Research and Forecasting (HWRF) model have consistently captured an intriguing TC structure in almost all Super Typhoon cases in which an upper warm core near the tropopause is seen to form on top of a pre-existing mid-level warm core. Such development of an upper warm anomaly suggests a potentially strong interaction of TCs with the lower stratosphere that has not been fully understood. This study uses the NCEP reanalysis data from 1948-2014 to analyze the connection between the conditions of the tropopause and TC intensity in the North Atlantic (NATL) and WPAC basins. Analyses of the statistical connection between the power dissipation index (PDI) and the TC intensity in the NATL basin shows a strong negative correlation between the PDI and the tropopause height (TPH) from 1948-1980 but a positive correlation during the interval 1981-2014. In contrast, the PDI-TPH correlation is consistently negative in the WPAC basin during the entire period from 1948-2014, indicating overall that the lower the tropopause height is, the stronger the TC intensity would be. In addition to the dominant negative correlation between the PDI and the TPH, we find a significant negative correlation between the stratification of the lower stratosphere and the PDI in both ocean basins. The negative correlation between the PDI and TPH is consistent with recent idealized simulations of the TC development, but appears to be unexplainable in the light of Emanuel's maximum potential intensity theory. Some of these correlations may be the result of data inhomogeneities, but potential physical mechanisms for such correlations between PDI and the tropopause characteristics will be provided.