Natural Hazards and Earth System Sciences (Jan 2018)

High-resolution marine flood modelling coupling overflow and overtopping processes: framing the hazard based on historical and statistical approaches

  • A. Nicolae Lerma,
  • T. Bulteau,
  • S. Elineau,
  • S. Elineau,
  • F. Paris,
  • P. Durand,
  • B. Anselme,
  • R. Pedreros

DOI
https://doi.org/10.5194/nhess-18-207-2018
Journal volume & issue
Vol. 18
pp. 207 – 229

Abstract

Read online

A modelling chain was implemented in order to propose a realistic appraisal of the risk in coastal areas affected by overflowing as well as overtopping processes. Simulations are performed through a nested downscaling strategy from regional to local scale at high spatial resolution with explicit buildings, urban structures such as sea front walls and hydraulic structures liable to affect the propagation of water in urban areas. Validation of the model performance is based on hard and soft available data analysis and conversion of qualitative to quantitative information to reconstruct the area affected by flooding and the succession of events during two recent storms. Two joint probability approaches (joint exceedance contour and environmental contour) are used to define 100-year offshore conditions scenarios and to investigate the flood response to each scenario in terms of (1) maximum spatial extent of flooded areas, (2) volumes of water propagation inland and (3) water level in flooded areas. Scenarios of sea level rise are also considered in order to evaluate the potential hazard evolution. Our simulations show that for a maximising 100-year hazard scenario, for the municipality as a whole, 38 % of the affected zones are prone to overflow flooding and 62 % to flooding by propagation of overtopping water volume along the seafront. Results also reveal that for the two kinds of statistic scenarios a difference of about 5 % in the forcing conditions (water level, wave height and period) can produce significant differences in terms of flooding like +13.5 % of water volumes propagating inland or +11.3 % of affected surfaces. In some areas, flood response appears to be very sensitive to the chosen scenario with differences of 0.3 to 0.5 m in water level. The developed approach enables one to frame the 100-year hazard and to characterize spatially the robustness or the uncertainty over the results. Considering a 100-year scenario with mean sea level rise (0.6 m), hazard characteristics are dramatically changed with an evolution of the overtopping ∕ overflowing process ratio and an increase of a factor 4.84 in volumes of water propagating inland and 3.47 in flooded surfaces.