Hydrologic and Hydraulic modelling as a support for an early warning system for flash floods

Urban floods are amongst the most dangerous and frequent natural hazards. Despite the great uncertainty and errors of the computation and input information, the estimation of flow discharges and flood levels are paramount for a successful flood management. In small watersheds with high slopes and small concentration time, reliable early warning systems should be based on real-time monitoring of the forcing variables and hydrology and hydraulic modelling to evaluate the flooded areas and the inundation levels in the vulnerable area.

In this work, hydrologic analysis and hydrodynamic modelling as a support for an early warning system are presented. Together with the modelling component, the monitoring setup including meteorological and hydrometric stations is showed. The case study focuses on the Vinhas Creek basin, where steep slopes from the mountainous headwaters converge to a final flat reach located in the highly urbanized area of the city of Cascais, close to the Tagus estuary mouth. This area is rather prone to torrential floods that inundate the lower part of the city.

The hydrologic and hydraulic modelling were conducted with the HEC models. After the definition and characterization of the watershed, the first modelling step was made by applying the HEC-HMS software coupled with geomorphological analysis and landcover to estimate the curve number and all related variables to generate flood hydrographs. Calibration was achieved using data from the historical flood of February of 1983.

As for the hydraulic modelling of the flood wave it was performed using HEC-RAS 2D software. A Digital Terrain Model was constructed taking advantage of 2x2m² topographic information. The model was validated against data from the flood of February 2021 for the validation of the results.

Finally, the integration of the validated HEC models for hydrology and hydrodynamics in the Water Information Forecast Framework is briefly presented. Based on meteorological data, this system aims at providing real time flood levels in the vulnerable areas.