The Genesis of Hurricane Humberto (2001)

During 3 consecutive days in September 2001, over 200 Global Positioning System dropwindsondes (GPS sondes) were deployed from NOAA and NASA aircraft as tropical storm Humberto intensified to a category 2 hurricane and then weakened to a category 1. The 2 NOAA WP-3ds deployed GPS sondes from altitudes of 1.5 km and 4 km and were concentrated within 200 km of the circulation center. The NASA DC-8 and ER-2 sampled both the inner core and the environment, launching GPS sondes from above 11.5 km and 16 km, respectively. This combined effort has made Humberto the most densely sampled hurricane with GPS sondes to date.

Early on September 22nd, Humberto was a tropical depression with a minimum central pressure of 1010 hPa. When the first GPS sondes were being jettisoned, late on the 22nd, Humberto had reached tropical storm strength and would intensify further to a category 2 storm late on the 23rd. We have developed multiple horizontal fields based on the GPS sonde data using cubic splines, objective analysis techniques and subjective analysis and have combined these fields with the lower fuselage radar data from the NOAA WP-3ds. These analyses provide clues as to how a warm core and the accompanying circulation might form.

Reflectivity fields reveal that an arc of deep convection, oriented east-west, is located 50 km north of the nascent circulation center. GPS sondes jettisoned near the area of lowest pressure show an overlying dry adiabatic layer with dew point depressions that exceed 10 C centered around 800 hPa. In the upper troposphere thermodynamic conditions are similar to the surrounding environment with no evidence of the expected positive temperature perturbation that has been observed in mature hurricanes (e.g., Hawkins and Imbembo 1976). The warm, dry layer in the lower troposphere has a thermodynamic structure which closely resembles the region under a stratiform precipitation region in a tropical squall line (e.g., Zipser 1977). The hydrostatic effect of this warm layer has been estimated based on a comparison with the structure of other soundings on the periphery of Humberto. This layer can account for a reduction of pressure between 7-9 hPa, which accounts for more than 70% of the observed pressure deficit. This structure is unlike what one expects if concepts summarized by the WISHE theory (Emmanuel 1986, 1991) were responsible for genesis.

In WISHE the intensification of a hurricane depends on the fluxes at the air-sea interface to increase the equivalent potential temperature of the boundary layer air. As a parcel with warmer equivalent potential temperature reaches the eyewall, it will ascend along a warmer moist adiabat than the previous parcel. Moist adiabats diverge with height resulting in the greatest temperature differences between the eyewall and the environment in the upper troposphere. Observations of mature tropical cyclones (Hawkins and Imbembo, 1976) show that the maximum temperature perturbation exists in the upper troposphere, consistent with the WISHE theory.

The analyses for the 22nd September demonstrate that the warm core is located in the lower troposphere, not the upper levels, and is due to a subsiding warm dry layer, not higher equivalent potential temperature air moving up the eyewall column. Only later does the ascent of increasingly higher and higher values of equivalent potential temperature build the temperature perturbation in the upper troposphere. The juxtaposition of the stable layer in the lower troposphere inhibits premature ascent and allows the boundary layer air to achieve higher energy content via the fluxes at the surface. We speculate that the warm layer in the lower troposphere is the initial cause of the pressure reduction at the surface and is thus a potential mechanism for TC genesis.