Skip to Main content Skip to Navigation

Multifractal space-time downscaling of rainfall

Abstract : We present a study performed in the framework of the EDF project on the use of seasonal meteorological forecasts to improve hydro-electrical resources management. Due to the difference of space and time scales, it is indispensable to downscale the (meso-scale) GCM rainfall to the (micro-) scale of the hydrological models. To preserve the scaling properties of the rainfall field, as well as its close interrelation with the dynamics at all scales, we develop a multifractal downscaling algorithm based on the idea that the rain rate cascades from large to small scale in a multiplicative manner: a scale invariant random multiplicative increment determines the rate fraction forwarded from a parent structure to a child one. Firstly, we proceeded to a rather exhaustive space and time analysis of the Météo-France PRECIP data base (about 10 years of high resolution data for 243 rain gages distributed over France territory) in order to estimate the universal multifractal parameters α, C1 as well as the exponent Ht of the scaling anisotropy of time versus space. The latter was empirically estimated to be in full agreement with its theoretical value: Ht=1/3. Secondly, we develop a cascade model defined with these parameters from space-time pixels corresponding to 243km×243km×32days, close to those of the GCM, which are of the order of 250km×250km×30 days, i.e. 35km×35km×25days. This choice is done in agreement with the value of Ht. In conclusion, we discuss how to take into account the orographic effects.
Document type :
Complete list of metadata

Cited literature [41 references]  Display  Hide  Download
Contributor : Ecole Mines ParisTech Connect in order to contact the contributor
Submitted on : Friday, August 6, 2010 - 10:40:45 AM
Last modification on : Wednesday, November 17, 2021 - 12:30:49 PM
Long-term archiving on: : Thursday, December 1, 2016 - 6:08:15 PM


  • HAL Id : pastel-00001573, version 1


Angelbert Chabi Biaou. Multifractal space-time downscaling of rainfall. Océan, Atmosphère. École Nationale Supérieure des Mines de Paris, 2004. Français. ⟨NNT : 2004ENMP1259⟩. ⟨pastel-00001573⟩



Record views


Files downloads