Proposed Community Stress Drop Validation Experiment

Annemarie S. Baltay, William L. Ellsworth, Martin Schoenball, & Gregory C. Beroza

Submitted August 15, 2017, SCEC Contribution #7797, 2017 SCEC Annual Meeting Poster #055

We propose to undertake a rigorous community stress drop validation study, and present here a draft “experimental design” for community review and feedback.

Earthquake stress drops are notoriously variable and difficult to measure; as such, estimates of this source parameter performed by different researchers, with different methods, or using differing datasets, have historically yielded highly inconsistent stress drops. This wide scatter arises because of model dependence, measurement uncertainty, or recording limitations. Stress drop is often estimated from the cube of the corner frequency times seismic moment, so even small errors in corner frequency or magnitude can result in much larger errors in stress drop. Data selection, metadata reliability, and processing differences, as well as assumptions on material parameters and velocity models can also yield very different stress drops for the same earthquake. Various methods, ranging from spectral ratios, broadband fitting, and ground-motion inversions, have yielded different results on the same dataset or when performed by the same researcher.

We propose a stress drop validation experiment, inviting any and all interested participants to use the same data set to work towards estimating stress drops by whatever method(s) deemed appropriate. We seek to understand the nature and causes of both the variability in earthquake stress drop estimates as well as the specific discrepancies found when considering the same earthquakes. Are variations due to differing methods and model assumptions? Do data selection and processing cause disagreement? Motivated by the recent intense interest in induced earthquakes, we select the 2016 M5.0 Cushing earthquake and ~40 foreshocks and aftershocks for this experiment. The catalog has been relocated by Schoenball and Ellsworth (2017), and includes earthquakes of M2 to 5 recorded by broadband and accelerometer stations at epicentral distance from ~zero to 100 km. Participants will agree to share results of estimated stress drop in a specified format, as well as documentation on the method, description of exact data and preprocessing used, assumptions employed, and byproducts estimated (i.e., seismic moment, corner frequency, energy, ground motion parameters). We aim to set a rapid pace for the experiment, with initial results presented at a special session of the 2018 annual meeting of the SSA, and the possibility of a SCEC workshop to follow.

Baltay, A. S., Ellsworth, W. L., Schoenball, M., & Beroza, G. C. (2017, 08). Proposed Community Stress Drop Validation Experiment. Poster Presentation at 2017 SCEC Annual Meeting.

Related Projects & Working Groups