Scaling Comparison of X-Band and C-Band Radar Data for Urban Hydrology

Abstract:

Rainfall is a very complex process that can be hardly represented with the help of a system of deterministic equations. However, it does posses scaling symmetries that have been more and more used to define stochastic processes representing rainfall.

The interest of high resolution rainfall measurements for a better modeling of urban and peri-urban catchments’ responses is investigated with the help of multifractals applied to two types of rainfall data: C-band radar data provided by Météo-France at a resolution of 1 Km in space and 5 min in time and data from a newly installed X-band radar operated by Ecole des Ponts ParisTech and providing data with a resolution of 100 m in space and 2.5 min in time. Multifractal downscaled C-band radar data are first compared with X-band data at the resolution of the latter to validate this downscaling. This multifractal downscaling is then used to generate an ensemble of high resolution rainfalls that can be thereafter input into a validated hydrological model to obtain an ensemble of hydrographs.

The variability observed within the simulated ensemble corresponds to the uncertainty associated with the small scale rainfall variability that is not measured by the C-band radar. Finally the response with the X-band radar data is simulated, which enables to discuss the added value of the improved rainfall data.