Author(s): Paguedame Game; Philippe Audra; Philippe Gourbesville; Mingyan Wang

Linked Author(s): Philippe Audra, Philippe Gourbesville,

Keywords: Deterministic modelling; Flood management; Water resources

Abstract: Nice Côte d’Azur Metropolis faces numerous challenges in the Paillons River catchment. On one hand, the watershed provides water resources to large communities in the French Riviera. On the other hand, historical extremes rainfall events show that the area is sensitive to flooding and drought. In fact, complex hydrological processes generate runoff over the 246 km2 watershed. The steeply and mountainous area lies between 0 and 1500 m elevation. Its river drains out through the urban centre to an outlet at sea. For several years, a river monitoring system have recorded meteorological and runoff data at different locations of the river network. Previous studies assessed the hydrogeological processes or their impacts of this catchment. But there are limitations in spatial details or length of periods of studies of rainfall-runoff processes, in surface hydraulics, in the hydrodynamics of the unconfined aquifer, and in river-aquifer exchange. To fill the missing data gaps and reproduce the complex processes, deterministic distributed models are needed. Thus, this study is part of a modelling strategy that can produce detailed information and optimize the organization of massive data collected. Results show capabilities of MIKE SHE coupled with Mike 11 to model extreme rainfall events. The max discharge on November 6th was estimated at 400 m3/s at hourly average. The error of maximum (ErrMax) is 0.14 m3/s and the RSME is 6.83 m3/s , for the event of 5th to 7th November 2000. The assessments of the November 2000 events are challenging due to lack of details time series for a long period. It is more challenging to assess rainfall events over several years. Some model parameters are not known over the catchment and need to be optimized. Horizontal hydraulic conductivity and saturated hydraulic conductivity of the most important soil class drive model response to rainfall inputs. The optimized value of horizontal hydraulic conductivity falls within the range of physical parameters. Other parameters like land use Strickler coefficient and thickness of saturated zone also affect the timing of peak discharges and shape of runoff curves.

DOI: https://doi.org/10.3850/IAHR-39WC2521711920221310

Year: 2022