In the latter of these previous studies, it was demonstrated that the maintenance of boundary layer moisture along inflowing air parcels from the northwestern Gulf of Mexico through central Texas was a critical contributor to Erin's peak intensity. These inflowing air parcels were embedded within a southerly low-level jet feature between the remnant Erin circulation to the west and a subtropical ridge of high pressure centered over the southeastern United States. This implies that moisture transport by the low-level jet contributed favorably to Erin's overland reintensification. However, it remains unknown as to whether the low-level jet exerted a greater control, whether favorable or otherwise, upon Erin's overland reintensification, and precisely how sensitive Erin's overland reintensification was to the precise structure, intensity, and location of the low-level jet.
In this research, we further examine the potential influence of the low-level jet upon the overland reintensification of Tropical Storm Erin. We do so utilizing a thirty-member ensemble of 4 km WRF-ARW, version 3.4.1, numerical model simulations. Numerical simulations begin at 0000 UTC 18 August 2007 and extend forward for 42 h until 1800 UTC 19 August 2007. The Ensemble Kalman filter implementation within the WRF-DART data assimilation system is utilized to generate initial and lateral boundary conditions for these numerical simulations from the 1800 UTC 17 August 2007 GFS model 0-h analysis and 3-48 h forecasts, respectively. Initial conditions are updated once via the assimilation of available atmospheric observations at 0000 UTC 18 August 2007; it is these updated fields that are used to initialize the thirty-member ensemble. The Morrison double-moment microphysical, Yonsei University planetary boundary layer, NOAH land-surface, and RRTMG shortwave and longwave radiation parameterization schemes are utilized within the ensemble. Conversely, cumulus convection is treated explictly within the ensemble.
Despite the relatively short simulation duration, the thirty-member ensemble produces a diverse range of solutions for Erin's evolution. In particular, the intensity of the simulated tropical cyclone at 1200 UTC 19 August 2007, during the reintensification period, ranges linearly (r^2 = 0.95) from 997 hPa to 1005 hPa within the ensemble. Likewise, the location of the simulated tropical cyclone at 1200 UTC 19 August 2007 ranges from southwestern Oklahoma to southeastern Kansas. This lends confidence that the ensemble is meaningfully dispersive, at least in the context of Erin's reintensification. Using the ensemble simulation output, ensemble sensitivity and part correlation analyses are conducted to examine the relationship between Erin's overland reintensification and the structure, intensity, and position of the low-level jet. Composite atmospheric fields from the simulations containing the strongest and weakest simulated Erin vortices during the reintensification period are obtained and analyzed to further examine this relationship. The presentation will focus upon discussing the key research findings and their implications toward the possible role of lower-to-middle tropospheric jets, such as the African easterly jet, in modulating the potential for a tropical disturbance to undergo tropical cyclogenesis.