STUDY ON NEAR-FIELD TSUNAMI INUNDATION FORECAST FOR TSUNAMI HAZARD ASSESSMENT
IN WESTERN JAVA, INDONESIA
Supervisor: Kenji SATAKE
Near-field Tsunami Inundation Forecasting (NearTIF) is a procedure developed by Gusman et al. (2014) to permit site-specific forecasting on high-resolution topography with a short computation time. The NearTIF algorithm is designed to work with a database of pre-computed tsunami waveforms and inundation, a tsunami numerical model, and a tsunami database search engine.
In this study, the NearTIF algorithm was used to generate tsunami inundation forecasting in Pelabuhan Ratu, West Java, Indonesia. 340 fault model scenarios were used to build a database of simulated tsunami waveforms at four offshore points and the resulting tsunami inundation. The high-resolution topography data for tsunami inundation modeling in Pelabuhan Ratu used in this study comes from previous research by Tanioka et al. (2012). Tsunami wave propagation and inundation was simulated with JAGURS tsunami simulation code (Baba et al., 2015a; 2015b). A nested grid system for tsunami simulation was used, where the coarsest grid size was 1 arc minute (~1853 m) and the finest was 20/81 arc seconds (~7.62 m). Three hypothetical earthquakes of different magnitude, type, and location were used. The tsunami inundation results from the NearTIF algorithm were then compared with those from direct numerical forward modeling (NFM) and analyzed for the speed values needed to forecast tsunami inundation in Pelabuhan Ratu.
The forecasts produced by NearTIF and direct NFM were similar in terms of tsunami height and inundation distance. For the computation time, NearTIF only needed approximately 45 seconds (43 seconds for tsunami waveform simulation and 2 seconds for database searching). On the other hand, direct NFM, even using a high performance computer, needed approximately 39 minutes to calculate a tsunami inundation forecast.