The origin of the Papua New Guinea tsunami that killed over 2100 people on 17 July 1998 has remained controversial, as dislocation sources based on the parent earthquake fail to model its extreme run-up amplitude. The generation of tsunamis by submarine mass failure had been considered a rare phenomenon which had aroused virtually no attention in terms of tsunami hazard mitigation. We report on recently acquired high-resolution seismic reflection data which yield new images of a large underwater slump, coincident with photographic and bathymetric evidence of the same feature, suspected of having generated the tsunami. T -phase records from an unblocked hydrophone at Wake Island provide new evidence for the timing of the slump. By merging geological data with hydrodynamic modelling, we reproduce the observed tsunami amplitude and timing in a manner consistent with eyewitness accounts. Submarine mass failure is predicted based on fundamental geological and geotechnical information.

The source of the local tsunami of 17 th July 1998 that struck the north shore of Papua New Guinea remains controversial, and has been postulated as due either to seabed dislocation (fault) or sediment slump. Alternative source mechanisms of the tsunami were addressed by offshore investigation using. multibeam bathymetry, sub-bottom profiling, sediment sampling and observation from the JAMSTEC Dolphin 3K Remotely Operated Vehicle and Shinkai 2000 Manned Submersible. The area offshore of Sissano is a complex active convergent margin with subduction taking place along the New Guinea Trench. Dominant transpressional convergence results in diachronous collision of the highstanding North Bismarck Sea Plate in a westerly direction. The result is a morphological variation along the Inner Trench Slope, with the boundary between eastern and western segments located offshore Sissano in an area of on-and offshore subsidence. This subsidence, together with nearshore bathymetric focusing, is considered to increase the tsunamigenic potential of the Sissano area. The offshore data allow discrimination between tsunami generating mechanisms with the most probable source mechanism of the local tsunami as a sediment slump located offshore of Sissano Lagoon. The approximately 5-10 km 3 slump is located in an arcuate, amphitheatreshaped structure in cohesive sediments that failed through rotational faulting. In the area of the amphitheatre there is evidence of recent seabed movement in the form of fissures, brecciated angular sediment blocks, vertical slopes, talus deposits and active fluid expulsion that maintains a chemosynthetic vent fauna. Dating of the slump event may be approximated 2 by the age of the chemosynthetic faunas as well as by a seismic signal from the failing sediment mass. Faults in the area offshore Sissano are mainly dip-slip to the north with recent movement only along planes of limited lateral extent. A possible thrust fault is of limited extent and with minimal (cm) reverse movement. Further numerical modeling of the tsunami also supports the slump as source over fault displacements.

Abstract. Case studies of landslide tsunamis require integration of marine geology data and interpretations into numerical simulations of tsunami attack. Many landslide tsunami generation and propagation models have been proposed in recent time, further motivated by the 1998 Papua New Guinea event. However, few of these models have proven capable of integrating the best available marine geology data and interpretations into successful case studies that reproduce all available tsunami observations and records. We show that nonlinear and dispersive tsunami propagation models may be necessary for many landslide tsunami case studies. GEOWAVE is a comprehensive tsunami simulation model formed in part by combining the Tsunami Open and Progressive Initial Conditions System (TOPICS) with the fully nonlinear Boussinesq water wave model FUNWAVE. TOPICS uses curve fits of numerical results from a fully nonlinear potential flow model to provide approximate landslide tsunami sources for tsunami propagation models, based on marine geology data and interpretations. In this work, we validate GE-OWAVE with successful case studies of the 1946 Unimak, Alaska, the 1994 Skagway, Alaska, and the 1998 Papua New Guinea events. GEOWAVE simulates accurate runup and inundation at the same time, with no additional user interference or effort, using a slot technique. Wave breaking, if it occurs during shoaling or runup, is also accounted for with a dissipative breaking model acting on the wave front. The success of our case studies depends on the combination of accurate tsunami sources and an advanced tsunami propagation and inundation model.

No abstract

Based on numerical simulations presented in Part I, we derive predictive empirical equations describing tsunami generation by submarine mass failure ͑SMF͒ that are only valid in the vicinity of the tsunami sources. We give equations for slides and slumps, along with some cautions about their appropriate use. We further discuss results obtained here and in Part I and their practical application to case studies. We show that initial acceleration is the primary parameter describing SMF center of mass motion during tsunami generation. We explain an apparent paradox, raised in Part I, in slump center of mass motion, whereby the distance traveled is proportional to shear strength along the failure plane. We stress that the usefulness of predictive equations depends on the quality of the parameters they rely on. Parameter ranges are discussed in the paper, and we propose a method to estimate slump motion and shear strength and discuss SMF thickness to length values, for case studies. We derive the analytical tools needed to characterize SMF tsunami sources in propagation models. Specifically, we quantify three-dimensional ͑3D͒ effects on tsunami characteristic amplitude, and we propose an analytical method to specify initial 3D tsunami elevations, shortly after tsunami generation, in long wave tsunami propagation models. This corresponds to treating SMF tsunami sources like coseismic displacement tsunami sources. We conduct four case studies of SMF tsunamis and show that our predictive equations can provide rapid rough estimates of overall tsunami observations that might be useful in crisis situations, when time is too short to run propagation models. Thus, for each case, we show that the characteristic tsunami amplitude is a reasonable predictor of maximum runup in actual 3D geometry. We refer to the latter observation as the correspondence principle, which we propose to apply for rapid tsunami hazard assessment, in combination with the predictive tsunami amplitude equations.