Natural Hazards and Earth System Sciences (Sep 2020)

Uncertainties in coastal flood risk assessments in small island developing states

  • M. U. Parodi,
  • A. Giardino,
  • A. van Dongeren,
  • S. G. Pearson,
  • S. G. Pearson,
  • J. D. Bricker,
  • A. J. H. M. Reniers

DOI
https://doi.org/10.5194/nhess-20-2397-2020
Journal volume & issue
Vol. 20
pp. 2397 – 2414

Abstract

Read online

Considering the likely increase in coastal flooding in small island developing states (SIDSs) due to climate change, coastal managers at the local and global levels have been developing initiatives aimed at implementing disaster risk reduction (DRR) and adaptation measures. Developing science-based adaptation policies requires accurate coastal flood risk (CFR) assessments, which in the case of insular states are often subject to input uncertainty. We analysed the impact of a number of uncertain inputs on coastal flood damage estimates: (i) significant wave height, (ii) storm surge level and (iii) sea level rise (SLR) contributions to extreme sea levels, as well as the error-driven uncertainty in (iv) bathymetric and (v) topographic datasets, (vi) damage models, and (vii) socioeconomic changes. The methodology was tested through a sensitivity analysis using an ensemble of hydrodynamic models (XBeach and SFINCS) coupled with a direct impact model (Delft-FIAT) for a case study of a number of villages on the islands of São Tomé and Príncipe. Model results indicate that for the current time horizon, depth damage functions (DDFs) and digital elevation models (DEMs) dominate the overall damage estimation uncertainty. When introducing climate and socioeconomic uncertainties to the analysis, SLR projections become the most relevant input for the year 2100 (followed by DEM and DDF). In general, the scarcity of reliable input data leads to considerable predictive uncertainty in CFR assessments in SIDSs. The findings of this research can help to prioritize the allocation of limited resources towards the acquisitions of the most relevant input data for reliable impact estimation.