The aim of this paper is to quantify the sensitivity of an operational semi-distributed urban hydrological/hydraulic model to rainfall variability. A 3400 ha urban area located in Seine-Saint-Denis (North-East of Paris, France), is used as a case study. First, spatio-temporal multifractal downscaling techniques are implemented on the C-band radar data (whose resolution is 1 km in space and 5 min in time) of the 9 February 2009 rainfall event to quantify the uncertainty associated with small scale unmeasured rainfall variability. It appears that the variability should not be neglected. This finding highlights the need to implement X-band radars (whose resolution is hectometric) in urban areas. Then, multifractal tools are used on both rainfall and simulated discharges that also exhibit a scaling behaviour. It appears that the rainfall drainage system basically transmits the rainfall variability without damping it, at least in terms of multifractal statistics.