On quantification of future tropical cyclone flood hazard under sea-level rise.

---

Jennifer Irish, Scott Hagen, Matthew Bilskie

Thursday 2 july 2015

16:20 - 16:40h at Antarctica (level 0)

Themes: IAHR/COPRI Symposium on Long Waves and Relevant Extremes

Parallel session: 13D. COPRI Symposium: Long waves and relevant extremes

---

It is a near certainty that global sea levels will continue to rise this century (e.g., Intergov. Panel Clim. Change 2013), thereby accelerating the tropical-cyclone flooding risk (e.g., Woodruff et al. 2013 Nature). In many areas, the complex interactions between storm surge generation and coastal terrain result in nonlinear coupling between surge generation and sea-level rise (SLR) (e.g., Smith et al. 2010 Ocean Eng., Bilskie et al. 2014 Geophys. Res. Lett.), making summation of existing statistical flood elevation with SLR projections inaccurate. Here, we investigate and compare two approaches for more robust quantification of the future 100-yr return period flood elevation under sea-level rise at Panama City, FL using the joint probability method with optimal sampling (JPM-OS) (e.g., Resio et al. 2009 Nat. Hazards). In both approaches, we use a small set of seven storms for which high-resolution hydrodynamic simulations of flooding under specific SLR conditions are carried out; these storms are selected because they generate flood elevations near the 100-yr level under present-day conditions. In the first approach, we use the simulations to directly adjust the 100-yr flood elevation, e.g., the average of these simulations is assumed to represent the new 100-yr flood elevation. In the second approach, we use the same simulations to quantify the relationship between SLR, present-day surge (z_o), and future flood elevation (z_n), and then apply this relationship to all storm possibilities to develop new continuous surge probability density functions within the JPM-OS framework (Irish & Resio 2013 ASCE Waterw.). Taylor et al. (in review Nat. Hazards) shows that a linear relationship in the form of z_n = a(z_o + SLR) +b, where a and b are location-specific constants, well represents the simulated results at Panama City (%RMS error<7%). Here, we compare the resulting future 100-yr flood elevation estimates and discuss the benefits and shortcomings of both approaches.