Linking lightning to tropical cyclone intensity change: a composite study of frequency and distribution

The United States was directly impacted by no less than twenty tropical storms or hurricanes between 2004 and 2008. Of these, four achieved category five status at some period during their lifetimes. Although tropical cyclone (TC) track forecasts have improved considerably during the last few years, predicting intensity continues to be a challenge for both meteorologists and numerical model simulations. Although a storm's path is influenced greatly by large-scale atmospheric circulations, its strength appears to be dominated by small-scale mechanisms within the storm itself.

Most previous research on TC intensity change has employed either numerical modeling or diagnostic approaches using traditional meteorological parameters. Only recently have studies started to examine electrification as a means for assessing the potential for intensification. Several papers have considered lightning as a proxy for storm intensification mostly using data from Vaisala's National Lightning Detection Network (NLDN) and Long-Range Lightning Detection Network (LLDN). Some of these studies have looked at TCs of varying intensity. Others have focused only on the convective properties of the most intense hurricanes. Most of this prior research has considered individual storms and has produced conflicting results. Some results showed high correlations between lightning frequency and subsequent intensification, while others showed no solid relationship between the two. We are seeking to better understand these relations by investigating a large number of storms and a variety of parameters that might influence both lightning and intensity.

We have prepared a dataset describing Atlantic Basin storms between the years 2004 and 2008. TCs were considered west of 70º W and between 15º N and 40º N. Our dataset divides each storm into fifty 10 km radial bins and twenty 18 degree sectors. Parameters such as wind shear, storm motion and speed, and average flash rate were compiled for each storm and subset by category. This allows us to examine TCs with high spatial and temporal resolution as a function of intensity, and to form relations between storm intensity and convective frequency and distribution.

The proposed paper will describe our dataset and methods and present a summary of results. We will focus on the role of lightning (if any) in predicting future storm intensity. We will also examine categorical relationships between wind shear and storm speed/motion on convective distributions and average flash rates. These results will facilitate a new understanding of the role of lightning in TCs and intensity change.