Multi-fractal rainfall time series generator

The modeling of rainfall is a complex issue that has been subject to several research. Its complexity is related to the modeling of some important properties of rainfall (intermittency rain/ no-rain, rain extremes and rain intensity). It is generally agreed that rain can be described in a multifractal framework. Multiplicative cascades are the conceptual tool for modeling stochastic processes and fields with multifractal statistics. Among numerous possible models, the universal multifractal model (UMM) received a great attention in the geophysical literature. However, Fractionnaly Integrated UMM allows simulating time series that may not represent all actual properties of rainfall (does not generate zero value, rainfall distribution is not coherent with real distribution, multiple scaling regimes). The present work is devoted to the calibration of Fractionnaly Integrated UMM to be able to generate realistic rainfall time series reproducing different statistical properties of observed precipitation time series