Using morphometric analysis of tropical cyclone structures in a new passive microwave imagery dataset to diagnose and predict intensity

The structure of a tropical cyclone (TC) is a spatial representation of its organizational pattern and distribution of energy acquisition and release. The shape of a TC depicts a specific phase in the TC's meteorological lifecycle, reflecting its past and potentially constraining its future development. For a number of reasons, a thorough objective definition of TC structures and an intercomparison of their varieties have been neglected. This lack of knowledge may be a key reason why TC intensity forecasts, despite numerical model improvements and theoretical advances, have been stagnant in recent years relative to track forecasts.

Passive microwave satellite imagers provide multiple benefits in discerning TC structure, but until recently, compiling a research quality data set has been problematic due to several inherent technical and logistical issues. Inter-comparison between the multiple available sensors is limited by the different channels on each sensor, spatial resolutions, and calibration metrics between satellites, all of which provide inconsistencies in resolving TC structural features. To remedy these difficulties, a global archive of TCs was compiled and standardized. Using global historical best track data, TC microwave data is retrieved from the Defense Meteorological Satellite Program (DMSP) series (including all SSM/I and SSMIS), TMI, AMSR-E, and WindSat sensors. Standardization between sensors for each TC overpass are performed, including: 1) Recalibration of data from the “ice scattering” channels to a common frequency (89GHz); 2) Resampling the DMSP series to a higher resolution using the Backus-Gilbert technique; and 3) Re-centering the TC center more precisely using the ARCHER technique (Wimmers and Velden 2010).

Using this passive microwave archive, a global climatology of TC structures is objectively developed using morphometric analysis techniques (i.e. size and shape analysis of TC features). Structural metrics such as threshold analysis of the outlines of the TC shape as well as methods to diagnose the inner-core size, asymmetry, and magnitude will be introduced. The observed structural existence of TCs will be discussed and compared with intensity and structural evolution. The use of these structural measures may then be used to facilitate operational TC analysis, comparisons to numerical model output, and potentially short-term forecasts of intensity and structural change. The process of developing the dataset and various objective definitions of TC structures using passive microwave imagery will be described, with preliminary results suggesting new methods to identify TC structures that may interrogate and expand upon physical and dynamical theories.