Though the recent rise in induced seismicity in Oklahoma has been attributed to wastewater injection, exact triggering mechanisms remain unclear. In particular, the Mw 5.7 earthquake occurring near Prague in November, 2011 remains the largest injection induced event in the area. For this project, I will be applying the "Match & Locate" method to the Prague sequence in order to improve catalogue completeness at lower magnitudes. This method correlates template waveforms across continuous data, as in a matched-filter approach, while stacking data from different stations to detect smaller events. I will then be looking at the resulting seismicity distribution in order to better understand event migration and underlying faults and structures.
I arrived in Pasadena a little over a week ago and have begun my placement at the USGS.
Over the past week, I ran the example Match & Locate code in order to figure out how it worked. I also applied corrections to the data as it had been converted from Antelope to SAC, and created a preliminary GMT map of the events and stations. I then ran the Match & Locate code on an hour of continuous data, and templates from the OGS catalogue. I am working on looking at how changing the different parameters (search area, search interval, SNR/cross-correlation thresholds, time windows, phases, etc.) affects the results and running time, before I run the code on the entire dataset.
Here are my goals over the course of the internship:
1st third: Finish choosing parameters, use Prague velocity model and TauP to help constrain phase arrival picks, and run event detection code over the entire dataset.
2nd third: Analyze results from new catalogue for patterns, eventually relocate and characterize events (magnitudes, stress drops).
3rd third: Continue analysis, create figures and AGU poster.
Skills I would like to develop both during this internship and beyond are oral communication/public speaking and the ability to make decisions with regards to the direction of a project.