Robust Results, Elegant Analyses, Data Driven Science

Session: What Can We Infer About the Earthquake Source Through Analyses of Strong Ground Motion?
Type: Oral
Date: 4/29/2020
Time: 02:30 PM
Room: 115
Description: 

It has been 52 years since Aki published the first paper to infer source parameters of an earthquake (1966 Parkfield) by synthesizing near-source strong motion records from a kinematic simulation. Although methods for analysis were being developed, there were only a few large magnitude earthquakes that had been well recorded on-scale by near-source seismometers and accelerometers in the following two decades (1970-80’s). As the instrumentation increased, so have the methods of analysis and the number of papers with inferences about the properties of the source. The most obvious change since Aki’s 1968 paper, is the spatial heterogeneity of the kinematic parameters: slip, slip rate, rupture velocity, rise time. Haskell’s (1964) seminal paper had proposed a constant slip function and constant rupture velocity as a starting point. However, he clearly knew (Haskell, 1966) that the parameters were spatially varying on the fault. With so many kinematic models being produced, the questions arose: What features are robust? What features are required by the data? How much uncertainty is there for any inferred property at any point on the fault? These questions related to the kinematic properties have consequences for inferences about the physical properties of the rupture, e.g., slip weakening distance, breakdown work, etc. and constraints on dynamic simulations of earthquakes. Many outstanding scientists have contributed to our body of knowledge about the earthquake process. I will look at some of the work of Dr. Paul Spudich and his impact. His work covers a broad area and examines the source with different methods to get the most robust result from the data. It will become clear that having an excellent match between a synthetic ground motion time history and a recorded one does not necessarily indicate a well-resolved source description. At the same time, his work shows how one can assess the quality of the model and move forward in interpreting the earthquake process.

Presenting Author: Ralph J. Archuleta

Authors