Probabilistic tsunami hazard analysis for Tuzla test site using Monte Carlo simulations

Abstract

In this study, time-dependent probabilistic tsunamihazard analysis (PTHA) is performed for Tuzla, Istanbul, inthe Sea of Marmara, Turkey, using various earthquake sce-narios of Prince Island Fault (PIF) within the next 50 and100 years. The Monte Carlo (MC) simulation technique isused to generate a synthetic earthquake catalogue, whichincludes earthquakes having moment magnitudes betweenMw6.5 and 7.1. This interval defines the minimum and max-imum magnitudes for the fault in the case of an entire faultrupture, which depends on the characteristic fault model.Based on this catalogue, probability of occurrence and as-sociated tsunami wave heights are calculated for each event.The study associates the probabilistic approach with tsunaminumerical modeling. The tsunami numerical code NAMIDANCE was used for tsunami simulations. According to theresults of the analysis, distribution of probability of occur-rence corresponding to tsunami hydrodynamic parameters isrepresented. Maximum positive and negative wave ampli-tudes show that tsunami wave heights up to 1 m have 65 %probability of exceedance for the next 50 years and this valueincreases by 85 % in the Tuzla region for the next 100 years.Inundation depth also exceeds 1 m in the region with proba-bilities of occurrence of 60 % and 80 % for the next 50 and100 years, respectively. Moreover, probabilistic inundationmaps are generated to investigate inundated zones and theamount of water penetrated inland. Probability of exceedanceof 0.3 m wave height ranges between 10 % and 75 % accord-ing to these probabilistic inundation maps, and the maximuminundation distance calculated in the entire earthquake cata-logue is 60 m in this test site. Furthermore, synthetic gaugepoints are selected along the western coast of Istanbul by in-cluding Tuzla coasts. Tuzla is one of the areas that showshigh probability exceedance of 0.3 m wave height, which isaround 90 %, for the next 50 years while this probabilityreaches up to more than 95 % for the next 100 years.