Defining the Lifecycle of the Extratropical Transition of Tropical Cyclones using the Deviation Angle Variance Technique for Remotely-Sensed Imagery

The extratropical transition (ET) of tropical cyclones (TCs) is a process in which recurving, warm-core TCs propagate poleward and interact with mid-latitude features, such as a cold-core pre-existing extratropical cyclone or trough, and transition into extratropical cyclones. During ET, TCs encounter environmental changes, such as increased Coriolis, decreased SSTs, and strong westerly flow and vertical wind shear. These changes cause drastic changes in the TC structure as it becomes an extratropical cyclone, which include, but are not limited to, an increase in the gale force wind field, increase in precipitation poleward and downstream, and a decrease in the asymmetry of the TC. Previous studies have divided the ET lifecycle into two stages – transformation and reintensification. Several of these studies objectively classify the times at which these two stages begin and end; however, these objective measures have been shown to have inconsistencies, including reliance on gridded analyses and computational difficulties.

This study attempts to define the ET lifecycle stages using an objective measure provided by the Deviation Angle Variance Technique (DAV-T) (Piñeros et. al. 2008). The DAV-T uses a signal variance calculated directly from satellite infrared brightness temperature images. A map of the gradient of brightness temperatures in an image is used to determine an accumulator matrix where the maximum value is the center of the storm. At every point around the center within a given radius, the difference between the ideal brightness gradient angle, which is pointed directly along a radial for an ideal symmetric TC, and the actual brightness gradient angle. The signal variance is then defined as the variance of the deviation brightness angles where the ideal TC has a variance of 0 and the least symmetric TC has high variance values. Results will be presented that show a distinct evolution of the signal variance for various types of ET including a characteristic dramatic increase in the signal variance values in the early stages of ET associated with the loss of symmetry of the storm and subsequent evolution depending on the particular path through ET.