Multi-proxy evidence for small historical tsunamis leaving little or no sedimentary record

Abstract

There has been considerable progress in tsunami research in recent years, yet most work has been focussed on identifying and understanding the evidence of large events. This study aimed to identify the evidence of small historical tsunamis in Lyttelton Harbour, New Zealand. The study area has been inundated by multiple relatively small historical tsunamis that left atypical sedimentary evidence (no sand) or no sediment. Shallow trenches revealed discontinuous layers of small grey mud clasts at various depths across the study area, most likely transported from the nearby tidal flats. The origin of these mud clast layers was investigated using a multi-proxy approach comprising sedimentological, geochemical and diatom analyses complemented by dating from 137Cs activity profiles and historical data. Subtle variations consistent with inclusions of marine mud such as a decrease in organic content and magnetic susceptibility and increases in geochemical markers (e.g. Ca/kcps, K/kcps, K/Rb, Si/Rb and Sr/Rb) were found in the sediment profile. Variations in diatom assemblages suggesting a marine influence were also recorded at similar depths, aligning with layers of mud clasts. Using 137Cs activity profiles and historical data, these deposits were attributed to the 1960 Chile Valdivia and 1964 Alaska tsunamis, as well as a possible earlier event. Sedimentary evidence for the 2010 Chile Maule tsunami was not found at the study site, but geochemical analysis of surface samples revealed marked changes in Ca, Cl, Sr and Ti concentrations, indicative of a change from a terrestrial to marine influence. This was used to identify the landward extent of inundation by the 2010 event. Ultimately, this study shows that a broad multi-proxy analysis can distinguish even the subtle signatures of an inconspicuous deposit laid down by a small tsunami.