Abstract

Statistical Model for Meteorolocal Elements Based on Local Radiosonde Measurements

Y.Shtemler (Department of Mechanical Engineering, Ben-Gurion University of the Negev, Beer-Sheva, Israel); A.Virtser, I.Kupershmidt (Institute for Industrial Mathematics, Beer-Sheva, Israel)

e-mail: shtemler@hotmail.com

A comprehensive statistical model is developed for vertical profiles of meteorological elements (horizontal wind and temperature). The model can be useful for several applications: safety in high-risk areas such as chemical and nuclear plants, space-vehicle starts etc. The model is based on the 3-year local radiosonde measurements for the most important for practice strong winds for which the altitude level of the maximum wind is highly expressed. Vertical profiles of meteorological elements measured under quite different atmospheric conditions exhibit qualitative similarity, but differ by the reference scales of wind, temperature and altitudes. A proper choice of the reference scales for a scaling procedure make the measurements meteorologically homogeneous. The main hypothesis of the present approach is universality of the homogenized data, which are considered as realizations of a random process with statistical characteristics independent from atmospheric conditions. The Principal Component Analysis (PCA) is applied to the homogenized data. The statistical model of the local meteorology is developed based on the hypothesis of universality for means, PCA basic functions and statistical parameters of standard distributions for the coefficients of the PCA transform. The dimensional reference scales are also simulated as standard distributions, but with non-universal statistical characteristics.

Section : 6