EAR-PF: SPATIAL AND TEMPORAL VARIABILITY OF COSEISMIC SUBSIDENCE ALONG THE CASCADIA SUBDUCTION ZONE

Dura Cristina, Philadelphia PA

Investigators

Abstract

Dr. Cristina Dura has been granted an NSF EAR Postdoctoral Fellowship to carry out research and education plans at Humboldt State University. This investigation will expand the resolution of subsidence due to seismic events along the Cascadian margin in both space and time. This information is critical for the assessment of seismic and tsunami hazards along the Pacific coast in the USA. Dr. Dura will use computer modeling based on diatom abundance to estimate subsidence along the West Coast. The study will consist of three components: (1) expand regional modern diatom dataset from the coasts of southern Washington, Oregon, and northern California to better characterize the vertical zonation of diatoms in a range of tidal environments. This information will be used to infer ecological niches of fossil diatom assemblages in cores; (2) develop a new diatom-based computational model that will reduce errors on subsidence estimates by ~28% compared to previous models; and (3) use this information to understand the spatial and temporal distribution of earthquakes along the Cascadia subduction zone. The education plan will entail developing teaching materials for K12 schools, participating in teacher workshops, creating an online tsunami curriculum to accompany a children's book, and outreach events for the community through local community groups. Accurate and precise estimates of coseismic subsidence over multiple earthquake cycles are critical to understanding the patterns, timing, and magnitude of strain accumulation and release during past earthquakes. Dr. Dura will employ new Bayesian diatom-based transfer functions to estimate coseismic subsidence, and improve upon the accuracy of existing foraminifera-based transfer functions. New estimates of coseismic subsidence in multiple cores at four sites spanning over 600 km of the Cascadia subduction zone will be used to characterize and correlate earthquakes along strike and distinguish rupture lengths and earthquake magnitudes over the past 2000-4000 years. The high-precision subsidence estimates will be incorporated into 3-D elastic dislocation models to produce more realistic (heterogeneous vs. segmented) rupture models that will help improve future earthquake and tsunami hazard assessments. Using a modern diatom dataset and application of a new Bayesian diatom-based transfer function will advance quantitative microfossil-based techniques by significantly increasing the accuracy and precision of coseismic subsidence estimates at Cascadia.