Tsunami Hazard, Samoan Islands: Palaeotsunami Investigation, Numerical Modeling and Risk Implications

Tsunami investigation is a fundamental component of coastal hazard mitigation and risk reduction. Recent history reveals that such hazards can influence rapid changes in global cultural dynamics through extensive loss of life (e.g. 2004 Indian Ocean Tsunami), lifeline destruction (e.g. 2011 Tohoku Tsunami) and property damage (e.g. 2014 Chile Tsunami), affecting the mobilization of regional and global humanitarian and financial resources. The 2009 South Pacific Tsunami (2009 SPT) in the Samoan Islands, which had devastating local impacts, provided the opportunity to better understand tsunami characteristics and subsequent hazard potential in this region. Lessons were learned from the impacts of this event in the context of local and regional tsunami mitigation. Equally a number of questions emerged. What is the long-term tsunami hazard in the Samoan region? What is the future risk of near-field events of similar or greater magnitude? What evidence is there in the geohazard chronology record? If there is evidence, what does it imply with regard to risk reduction in Samoa and the broader Pacific? These questions formed the research basis for this thesis. Specific aims and objectives were devised to address the challenges and concerns identified. A range of inter-disciplinary techniques were used to yield innovative information to achieve them. Proxy characteristics (e.g. loss on ignition, grain size, elemental ratio, geochronology, resonance modeling) associated with the 2009 SPT and identified 1990 and 1991 Cyclones Ofa and Val deposits, respectively, provided unique analogues for identifying and distinguishing tsunami and cyclone signatures in the deeper Samoan geologic record. A tsunami and cyclone geochronological model spanning the last 3,000 years or so was developed. Estimation of tsunami frequency of similar or greater magnitude events than the 2009 SPT likely originating from the near-field Northern Tongan Subduction Arc (NTSA) source was also made possible. The results suggest a minimum 87 year recurrence interval of 2009 SPT-type tsunami intensities or stronger associated with a likely NTSA origin. Assessment of the contemporaneity between identified tsunamis and cyclones in the geologic record with anomalous and/or enigmatic sequences in the ethno-archaeological, oral and indigenous records provided likely indicators of the possible extent of associated hazards. Further, the discovery and association of anthropogenically-formed charcoal contemporaneous with earliest colonization in the Samoan archipelago, with the oldest tsunami identified, likely substantiates an approximate 3,000 year hazard history. Evidence of a landslide-generated tsunami which occurred during the Last Glacial Maximum (LGM), as well as non-related hypotheses concerning inland high-elevation calcareous deposits of cultural significance, were considered within the broader long-term tsunami hazard context. Knowledge gaps associated with landslide-generated tsunami processes and their hazard potential in this region were identified. The possibility of calcareous deposits found in a central highland location in these islands being of a potential tsunami or coastal marine origin is dismissed. This research demonstrates that an intrinsic tsunami hazard history covering the last 3,000 years exists in the Samoan Islands. Directions for future studies that build on the findings presented here are offered. The principal research outcomes achieved provide a basis for future refinement. Nonetheless, the thesis can be used in its present form as a guide for similar investigations, as well as in long-term coastal risk and mitigation at the local level. The techniques used and information obtained can also be developed and applied to analogous coastal environments in other countries to assist broader long-term regional and global tsunami risk reduction