Unveiling the Faults in the Southern Raton Basin Using Nodal Array

Session: Mechanisms of Induced Seismicity: Pressure Diffusion, Elastic Stressing and Aseismic Slip
Type: Oral
Date: 4/29/2020
Time: 09:30 AM
Room: 215 + 220
Description: 

Waste-water injection induced earthquakes have been occurring in the Raton Basin for the past two decades (Rubinstein et al., 2014; Nakai et al., 2017). Several M>4 earthquakes have been reported with normal-faulting focal regimes in the Colorado section of the basin (Bamhart et al.,2014). While regional monitoring has been enhanced by the installation of seven UNM stations since 2016 (network code YX), the spatial resolution is still insufficient to completely detect and locate low-magnitude (M<1) seismicity and identify small-scale faults. In the summer of 2018, 96 high-frequency three-component nodal instruments were installed in the southern (NM) section of the basin. During the month of deployment, two M>2.5 earthquakes were recorded and their source mechanisms are dominated by strike-slip – dip-slip motion, different from the normal-faults in the CO section. We first applied STA/LTA detection and REAL (Zhang et al., 2019) methods to automatically detect and associate the arrival phases. Then we adopted a machine-learning based phase picker (PhaseNet) to detect the P and S phases. The outcome is compared to our hand-picked arrivals over 21 days (15648 picks; 60% of the deployment period). Finally, the hypocenter locations of the seismicity are refined using hypoDD (Waldhauser & Ellsworth, 2000). The updated earthquake catalogue contains thousands of earthquakes with source depths (4-8 km) deeper than injection depth (1-2 km); their spatial pattern is consistent with that from broadband stations. Moment magnitudes of these identified events will be calibrated and estimated based on the two M>2.5 events. Taking advantage of the dense array, our source analysis provides a unique opportunity to unveil the fault structures in the southern Raton Basin. Furthermore, successful application of the above routine sheds light on the potential and power of “hands-free” processing of large-volume three-component nodal data. Potential advantages and challenges will be discussed based on our experience from the Raton array.

Presenting Author: Ruijia Wang

Authors