Study on Probabilistic Inundation Hazard along the Coast of South Pacific Islands: Case Study at Lakeba Island in Fiji

抄録

 This paper develops a system for estimation of the probabilistic inundation hazard along the coast of South Pacific Islands. The present system is based on Monte Carlo simulation and consists of three submodels: (i) a stochastic cyclone model, SCM; (ii) a offshore wave model; and (iii) a nearshore wave and surge model. The present SCM generates a number of virtual cyclones and determines the track and the time-varying central pressure of each cyclone. Offshore wave model determines the wave conditions at the offshore of the target coast for each cyclone. Obtained offshore wave conditions are then applied to the third sub-model for computations of wave deformation and time-varying surge induced by wind and waves on the nearshore area including a fringing coral reef and a lowland along the coast. The present SCM accounts for the unique features of each cyclone through the auto-regression model. The concept of minimum central pressure was introduced to improve the predictive skill of the probabilistic characteristics of extreme cyclones. The developed system is applied to Lakeba Island in Fiji. Through Monte Carlo simulation, probabilistic characteristics of wind, offshore waves and runup are investigated at four target coasts located on the north, south, east and west side of the island, respectively. While the estimated wind velocity showed similar characteristics at each coast, clearly higher offshore waves and runup height were computed on the north and west side of the island. Finally, the impact of climate change on coastal hazard of the South Pacific Islands is investigated through the numerical analysis based on GCMs.