How do different interventions on existing levee systems influence flood maps?

Levee failure is one of the causes of many catastrophic flood damages. Consequently, different options of levee rehabilitation techniques or counter measures, e.g. covering with grass, concrete slabs, rip-rap, asphaltic revetments etc, have been used to strengthen the levee system. However, little is known on the amount of safety, against flooding, provided by the levee system due to difficulty (contains huge uncertainity) in quantifying the resisting abilities of the applied counter measures to the hydraulic load. The present research proposes a methodology to investigate the effects of two different counter measures, (1) improving grass cover quality in the levee land-ward slope, (2) heightening levee crest, on the estimation of the probability of levee failure due to overtopping and the consequent flood extent. The methodology has been applied to the 98km-braided reach of Po River, Italy, between the cross-sections of Cremona and Borgoforte. The adopted methodology was divided into four core categories. Firstly, reliability analysis of the levee system, in case of overtopping, was performed, using the geometrical levee features as deterministic variables, while the grass cover quality (non geometrical levee feature) was assumed to be a stochastic variable to account for the uncertainties associated to it. In order to estimate the fragility curves for all sections, a Monte Carlo framework was implemented. Secondly, a 1D hydrodynamic model was implemented to estimate the water level in the river in case of a synthetic 200 year return period flood event. The information of the water level was combined with the fragility curves in order to assess the probability of failure for each single cross section. As third step, a levee breach model was introduced to address the uncertainties related to the location, size and development of the breaches. Finally, a 2D hydrodynamic model CA2D_S, based on the cellular automata approach in semi-inertial formulation for flux computation, was implemented. CA2D - SCENARI (CA2D_S) is a version of the CA2D model specifically designed to simulate levee breach scenarios in low land areas. Four test scenarios were formulated with different combinations of the assumed countermeasures. These are base case scenario represents the present situation, second scenario formulated by heightening the levee, third scenario is improving the grass cover quality and the last scenario is the combination of scenario two and three, heightening levee as well as improving the rass cover quality. The previous methodological steps were repeated for each scenario and the results from CA2D, expressed in terms of flood extent considered after some time steps since the levee breach, were compared and analyzed. The analysis showed that the counter measures tested in different scenarios can reduce the probability of failure of the levee system and prevent flooding locally