A Low-Wavenumber Analysis of the Relative Roles of the External and Internal Dynamical Processes Responsible for Rapid Intensity Changes in Landfalling Tropical Cyclones over the Bay of Bengal

Rapid Intensity changes in tropical cyclones (TCs) are a special subset of the already baffling intensification problem where forecasting failures can have dire consequences. One of the main reasons as to why forecasting these rapid intensity changes is particularly hard is that the factors responsible, span a multitude of scales. Examples of these processes impacting TC intensity changes include the microphysical process such as evaporation and precipitation at disorganized cloud scales, aggregated, organized convective processes at the vortex scale, and larger scale circulations at the atmospheric environment. Most importantly, these external and internal dynamical and thermodynamical variables seek to act simultaneously in a nonlinear fashion and these processes can either complement, amplify, inhibit or not impact the TC intensity at all. For example, the environmental vertical wind shear vector can significantly alter the location, maintenance and amplification of vortex-scale, deep convective processes, such that the TC weakens significantly, or in certain cases, result in the asymmetric rapid intensification of the TC. Likewise, the manner in which the TC vortex interacts with its upper-atmospheric circulations can lead to vastly different TC behaviors. Therefore, the key to improved prediction of intensity lies in an improved understanding of the relative roles of the external, environmental processes, and the internal storm dynamics.

The relative roles of these governing factors can vary from basin to basin, case by case and as the storm is at various stages of its lifecycle. This study specifically focuses on TCs over the Bay of Bengal region where the distinctive seasonal variations due to the variability in the monsoon cycle, and the relatively small size of the basin, set up unique environments over the atmosphere, land, and ocean. WRF-ARW simulations of all the instances at which TCs between 2012 - 2016 that underwent rapid changes in intensity are used to create a suite of cases for the wave-number analysis (WNA). The environmental and vortex-scale variables are reduced to their 0 and 1 azimuthal harmonics, as a function of distance to storm center, since they contain most of the variance (wavenumber space, cylindrical coordinate system). This reduces the less-impactful “noise” in the variables of concern. The end of the vortex and the environment is defined using an adapted version of Kurihara’s algorithm (1984).

The study aims to derive its understanding from previous anecdotal evidence of individual cases and conduct a systematic analysis on a sufficiently varied number of cases to make representative and statistically significant conclusions. Guidelines from prior research serve as the base hypotheses and the present study builds on the following conclusions:

• Within the vortex, following balance arguments, the closer the centroid of convective heating is to the radius of maximum winds, the greater is the spin-up of the vortex
• The location of deep convection in the left-shear quadrants plays a dominant role in determining if the storm is about to undergo a rapid change in intensity.
• Environmental wind shear can aid in the intrusion and transport of low entropy air into the boundary layer. If the deficit in equivalent potential temperature (mid-level and within boundary layer) outweighs the refurbishing from the surface fluxes, the ability to sustain deep convection is lost and the TC begins to rapidly weaken.

With this in mind, the WNA is used to evaluate several important characteristics in the TC environment and vortex and to quantify the contribution of the following:

1. the radial distribution of convection with respect to the radius of maximum wind and the role of the outer rain bands.
2. the azimuthal distribution of convection with respect to the direction of environmental shear, within the vortex,
3. the relative location of the peak in wavenumber 1 with respect to the shear vector
4. the environmental moisture in different quadrants with respect to the shear vector at different levels
5. the direction of the driest air with respect to the shear vector
6. the horizontal moisture flux convergence within and above the boundary layer in different quadrants with respect to the shear vector
7. the upper-level wind divergence
8. the environmental wind shear and the
9. the antecedent soil moisture and soil temperature

The modeled cases are then segmented into rapidly intensifying, rapidly weakening and those that underwent no significant change in intensity and their optimal (Fisher’s) discriminant from the variables listed above is computed. The results of the discriminant analysis that identify the variables and the combination of variables that have the most significant impact on the storm’s intensity, will be presented. The implications of these results are huge since they help in the understanding of the most essential factors influencing TC rapid intensity changes and they aid in the development of decision support tools that can aid in early warnings and guidance in disaster-prone regions.