Uncertainty in broad-scale coastal flood damage cost assessments under sea-level rise: critical needs for more accurate protection data

Abstract

Broad-scale coastal flood risk assessments (CFR) face several methodological limitations and require modeling simplifications, leading to a cascade of uncertainties. Yet the relative influence of these uncertainties remains poorly quantified. To this end, we have developed a global sensitivity analysis that considers the multiple uncertainties associated with data and scenario choices throughout the CFR modeling chain of a recently launched pan-European coastal climate service. The application in five countries, i.e. Italy, France, Spain, the United Kingdom, and Finland, reveals that uncertainty in coastal protection levels dominates damage costs for residential buildings exposed to a 100 y return period event in 2100. When uniform low estimates of coastal protection are assumed, the choice of future climate and societal scenarios exerts strong influence on uncertainty, particularly for high sea-level outcomes. Fixing these scenarios shows how significant the modeling assumptions used to calculate asset-level damage can be, more specifically the uncertainty in: (1) estimates of damaged asset area, and (2) selection of damage curve databases. Our study underlines that methodological choices are not secondary considerations: they are primary drivers of uncertainty in broad-scale CFR assessments. This highlights the need for transparent communication regarding protection levels and damage assessments’ assumptions to improve the reliability of CFR assessments.