Refining tsunami inundation limits for the 1700 CE Cascadia earthquake: detailed mapping of fine-grained deposits and diatom biostratigraphy at the Salmon River estuary, central Oregon

Abstract

Stratigraphic evidence of coseismic subsidence and sandy tsunami deposits preserved in coastal wetlands provide tangible evidence of the 1700 CE Cascadia Subduction Zone (CSZ) earthquake and tsunami. Such evidence has been used to evaluate earthquake source models by comparing simulated tsunami inundation or sediment deposition to the mapped extent of sandy tsunami deposits. However, post-tsunami surveys of modern events, such as the 2011 Tōhoku and 2010 Maule earthquakes, demonstrate that the inland limit of visible sand deposition often underestimates inundation limits especially in coastal areas with flat or gently sloping topography. Finer-grained sediments (e.g., silt) and marine microfossils can be transported farther inland beyond traces of sand and more accurately represent tsunami inundation limits. To improve estimates of tsunami inundation, we use a high-resolution, multiproxy approach to more precisely map the true inland extent of 1700 CE tsunami inundation at the Salmon River estuary, Oregon. We identify fine-grained tsunami deposits in sediment cores collected across the previously mapped tsunami sand deposition boundary using grain size analyses, diatoms, and computed tomography imagery. High-resolution grain size and diatom data reveal silty sand (>60% sand) and anomalous marine diatoms overlying the sharp 1700 CE subsidence contact up to ~1 km farther inland of the previously recognized tsunami sand extent. In inland cores with more subtle sandy silt (<35% sand) deposits above the subsidence contact, diatom analyses show that the same epipsammic marine-brackish species characteristic of tsunami-transported sand are still present, allowing us to extend the inundation line up to 1.2 km further inland of previous studies. Preliminary inverse flow modeled flow depths and velocities are generally bounded by forward modeled flow depths and velocities from high slip earthquake sources that best replicate our new inundation limit. However, this initial comparison is limited because the models are fundamentally different in their approach to calculating hydrodynamics. Our results extend the minimum mapped tsunami inundation limit and better represent the full inland reach of the 1700 CE tsunami. Expanding this approach to additional CSZ sites will improve paleotsunami reconstructions, providing more accurate constraints on earthquake and tsunami source models and enhancing future hazard assessments that are critical for improving community resilience.