Analysis of Global Atmospheric Non-Gaussian Extreme Events

Statistics of extreme events in weather and climate (e.g. rare floods or strong wind storms) are commonly based on the assumption of Gaussian statistics. Sixty-two years of National Centers for Environmental Prediction / National Center for Atmospheric Research (NCEP / NCAR) Reanalysis I data and thirty-one years of National Centers for Environmental Prediction / Department of Energy (NCEP / DOE) Reanalysis II data are analyzed to determine if this assumption is true. The mean and variance of several atmospheric variables are calculated. Furthermore, the higher statistical moments — skewness and kurtosis — are calculated for geopotential height, relative vorticity, and the meridional and zonal wind components. Zonal averages of these higher statistical moments are also analyzed. It is found that statistically significant deviations from Gaussianity are found for every variable in the atmosphere on the synoptic to global scales. This empirical analysis is linked to particular atmospheric phenomena such as tropical cyclones, sudden stratospheric warming events, and the concept of rectifica- tion. In essence, there are fundamental forcing asymmetries in the atmospheric equa- tions of motion that lead to the existence of non-Gaussian distributions. Additionally, the relationship between skewness and kurtosis and the existence of power-law tails in non-Gaussian systems is examined.