Analyses multifractales et spatio-temporelles des précipitations du modèle méso-NH et des données radar

Abstract

In the framework of universal multi fractals, the spatiotemporal variability of rainfall over a wide range of scales is quantified with the help of three scale-invariant parameters. In this paper we estimated these multi-fractal parameters on numerical simulations performed with the help of Meso-NH, developed by Météo-France and the Laboratory of Aerology (Univ. P. Sabatier, Toulouse, France), and on composite radar images, covering the same heavy rainfall event that occurred on 5-9 September 2005 in the south of France. The comparison of the results shows that both data sets exhibit similar ranges of scaling, whose properties are in agreement with the simplest unified space-time scaling model of rainfall fields. At the same time, some parameter estimates were quite different. © 2011 IAHS Press.