The estimation of total lightning from various storm parameters: A cloud-resolving model study

Forecasting the electrical activity of a storm is a difficult task because of the complexity of the processes leading to cloud electrification and to lightning flash propagation. Today, only a few mesoscale models with an electrification and lightning parameterization attempt to simulate the whole life-cycle of the electric charges in a thunderstorm. Simulating the electric activity of a storm is challenging as some of the processes involved in cloud electrification and lightning propagation are poorly understood and remain topics that are still under investigation. In addition, the computational cost of these explicit electrical models prevents them from being used in forecasting models.

An attractive way to predict lightning flash rates in numerical models is to rely on correlations between the lightning flash rate and available model parameters. These correlations can then be used for other applications such as the parameterization to predict the production of nitrogen oxides from lightning. In this study the potential for some model parameters to be used as a proxy for the total lightning rate in cloud resolving model studies has been investigated using the Weather Research and Forecasting model (WRF) in two different storms: the 10 July 1996 STERAO-A severe storm that occurred in the U.S. High Plains and the 13 July 2005 airmass thunderstorm near Huntsville, AL.

It is shown that the WRF model reproduces the structure and morphology of the two storms. We test 5 storm parameters, precipitation ice mass, updraft volume, maximum updraft velocity, ice mass flux product, and ice water path, as proxies for lightning flash rate. Results show that the Price and Rind (1992) power law relation of maximum updraft velocity with flash rate has good correlations with flash rate, and ice mass flux product and precipitation ice mass can repeat the trend in flash rate but not the magnitude. The updraft volume and ice water path predict flash rate poorly.