William A. Komaromi1 and James D. Doyle2
1. National Research Council, Monterey, CA
2. Naval Research Laboratory, Monterey, CA
Utilizing the ability of the Global Hawk (GH) and WB-57 aircraft to overfly the TC core at 60,000 ft with most GH missions exceeding 24 h, The NASA Hurricane and Severe Storm Sentinel (HS3) and Tropical Cyclone Intensity (TCI) field experiments have provided an unprecedented high-altitude dropsonde dataset of Atlantic tropical cyclones (TCs). The unique combination of range, altitude and coverage of dropsondes released into a large sample of both strengthening and weakening TCs ranging from tropical depression through major hurricane strength has made it possible to analyze and relate changes in the structure of the TC outflow to both changes in the inner-core and the environment. Here we investigate 8 missions into intensifying TCs, 4 missions into weakening TCs, and 5 missions into steady TCs from the HS3 dataset during the 2012-2014 Atlantic hurricane seasons. From the TCI dataset, we focus on a series of missions into Hurricane Joaquin (2015) from near peak intensity through its weakening stage, and Patricia (2015) during its early rapid intensification phase and record-setting peak intensity. The primary focus of this study is the strength and structure of the TC outflow and how it relates to storm intensity and intensity change. The magnitude and altitude of the warm core, as well as the presence of secondary warm cores and how the overall cyclone structure relates to the environmental shear as derived directly from the dropsondes will also be discussed. Finally, some comparisons between the dropsondes and CIMSS AMVs in terms of the level of detail in the vertical structure of the wind field are also made.
The structure and magnitude of the outflow is found to vary quite rapidly in the vertical. For most hurricanes sampled during HS3, the level of maximum radial wind u is sharp and well-defined, with values close to zero just above and below. The minimum in tangential wind v is slightly broader in the vertical and less pronounced. For both u and v, the vertical gradient above the strongest outflow is steeper than below, and appears to be capped above by greater static stability near the tropopause and below by greater inertial stability below. There are also often one or more inflow layers immediately above or below the outflow level. It is also worth noting that the level of minimum v is typically slightly, ~25-50 hPa, above the level of maximum u. Additionally, a stronger upper-level divergence signature is found to be associated with strengthening versus weakening systems, especially from 180-150 hPa with less difference at or below 200 hPa. Stronger TCs were not necessarily associated with greater upper-level divergence than weaker TCs. However, intense upper-level divergence does not guarantee a strengthening system, as two counter-examples demonstrate. For a number of cases and particularly for Nadine (2012) and Edouard (2014), the upper-level anticyclone is largest and most prominent at 200 hPa and contracts with height to increasingly smaller radius through 100 hPa. Nearly all of the hurricanes in the sample have some reflection of an upper-level anticyclone through 100 hPa, in contrast with the tropical storms which typically do not, and the ones that do tend to feature an anticyclone which is quite displaced with the center of the vortex.
An environmental shear is also computed from the dropsondes themselves by taking 4 quadrant-mean wind vectors and averaging these vectors to remove the TC-component of the shear. Most of the TCs feature a divergence-convergence couplet with the convergence upshear and the divergence downshear. This signature is quite evident in 3-dimensions, and also rotates about the center of the storm as the shear vector varies with height. In multiple cases, the location of deep convection relative to the vortex center is more consistent with the mid-level shear vector than the traditional 850-200 hPa deep-layer shear vector. There are also a few instances of the stronger shear undercutting the outflow. Lastly, significant variability in terms of the height of the warm core is observed, although it was typically located between 400-200 hPa. Several cases including Cristobal (2014) and Nadine (2012) feature double warm cores, although the higher warm care is almost always stronger in this dataset. Interestingly, the final flight into Hurricane Edouard (2014) featured a very low warm core near 700 hPa; however, the system was under strong westerly shear at the time and was classified as a strong post-tropical cyclone by the NHC shortly after the flight. Further implications of all of these findings for our understanding of TC dynamics and prediction, as well as relevance to the existing literature will also be discussed.