Numerical simulations of the 2004 Indian Ocean tsunami deposits' thicknesses and emplacements

Abstract

Abstract. After more than a decade of recurring tsunamis, identification of tsunami deposits, a part of hazard characterization, still remains a challenging task that is not fully understood. The lack of sufficient monitoring equipment and rare tsunami frequency are among the primary obstacles that limit our fundamental understanding of sediment transport mechanisms during a tsunami. The use of numerical simulations to study tsunami-induced sediment transport was rare in Indonesia until the 2004 Indian Ocean tsunami. This study aims to couple two hydrodynamic numerical models in order to reproduce tsunami-induced sediment deposits, i.e., their locations and thicknesses. Numerical simulations were performed using the Cornell Multi-grid Coupled Tsunami (COMCOT) model and Delft3D. This study reconstructed tsunami wave propagation from its source using COMCOT, which was later combined with Delft3D to map the location of the tsunami deposits and calculate their thicknesses. Two-dimensional horizontal (2-DH) models were used as part of both simulation packages. Four sediment transport formulae were used in the simulations, namely van Rijn 1993, Engelund–Hansen 1967, Meyer-Peter–Mueller (MPM) 1948, and Soulsby 1997. Lhoong, in the Aceh Besar District, located approximately 60 km southwest of Banda Aceh, was selected as the study area. Field data collected in 2015 and 2016 validated the forward modeling techniques adopted in this study. However, agreements between numerical simulations and field observations were more robust using data collected in 2005, i.e., just months after the tsunami (Jaffe et al., 2006). We conducted pit (trench) tests at select locations to obtain tsunami deposit thickness and grain size distributions. The resulting numerical simulations are useful when estimating the locations and the thicknesses of the tsunami deposits. The agreement between the field data and the numerical simulations is reasonable despite a trend that overestimates the field observations.