Tsunami Excitation Estimation from Real-time GNSS

Tsunami early warning systems currently comprise modeling of observations from the global seismic network, deep-ocean DART buoys, and a global distribution of tide gauges. While these tools work well for tsunamis traveling teleseismic distances, saturation of seismic magnitude estimation in the near field can result in significant underestimation of tsunami excitation for local warning (Wang et al., 2012). Moreover, DART buoy and tide gauge observations cannot be used to rectify the underestimation in the available time, typically 10-20 minutes, before local runup occurs. Real-time GNSS measurements of coseismic offsets may be used to estimate finite faulting within 1-2 minutes and, in turn, tsunami excitation for local warning purposes (Blewitt et al., 2006; Melgar and Bock, 2013; Yue and Lay, 2011). Described here is a tsunami amplitude estimation algorithm, implemented for the Cascadia subduction zone, that uses continuous GNSS position streams to estimate finite faulting. The system is based on a time-domain convolution of fault slip that uses a pre-computed catalog of hydrodynamic Green’s functions generated with the GeoClaw shallow-water wave simulation software and maps seismic slip along each section of the fault to points located off the Cascadia coast in 20m of water depth and relies on the principle of the linearity in tsunami wave propagation. The system draws continuous slip estimates from a message query server (RabbitMQ), convolves the slip with appropriate Green’s functions which are then superimposed to produce wave amplitude at each coastal location. The maximum amplitude and its arrival time are then passed into a database for subsequent monitoring and display. This system was tested with data from a real earthquake for which we have continuous GNSS time series and surveyed runup heights, Tohoku, Japan 2011. This system has been implemented in the CWU Geodesy Lab for the Cascadia subduction zone and will be expanded to the circum-Pacific as real-time processing of international GNSS data streams become available.

Name thesis:degree_name

Master of Science (MS)

Discipline thesis:degree_discipline

Geological Sciences

Year dc:date.available

2018

Author dc:creator

• Jeffries, Catherine

Contributors dc:contributor

• Tim Melbourne
• Breanyn MacInnes
• Walter Szeliga

• tsunami estimation; real-time GNSS; early warning
• Earth Sciences
• Geophysics and Seismology
• Physical Sciences and Mathematics

Language dc:language

English

Repository record dc:identifier

https://digitalcommons.cwu.edu/etd/960

OAI identifier oai:identifier

oai:digitalcommons.cwu.edu:etd-1932