Software

FocMecDR: A Cross-Correlation-Based Double-Ratio Earthquake Focal Mechanism Inversion Algorithm (coming soon)

FocMecDR is a new earthquake focal mechanism inversion algorithm, which is based on cross-correlation and double-ratio techniques. FocMecDR is designed to automatically invert relative focal mechanisms for earthquakes, especially for those small ones. Similar to the concept of double-difference in earthquake location, FocMecDR uses an earthquake with a known focal mechanism as a reference and inverts focal mechanisms for nearby target events by minimizing the difference between the ratio of P/S wave amplitude ratio and theoretical ratios (not waveforms, just radiation pattern coefficients), as well as matching their relative polarities via cross-correlation.  

Reference: Zhang M. FocMecDR: A Cross-Correlation-Based Double-Ratio Earthquake Focal Mechanism Inversion Algorithm, in prep. (presented at the SSA  annual meeting, 2023)

OBSTransformer:  A Generalized Machine Learning Phase Picker for OBS data

OBSTransformer is a transfer-learned seismic phase picker for Ocean Bottom Seismometer (OBS) data adopted from the EqTransformer model. OBSTransformer has been trained on an auto-labeled tectonically inclusive OBS dataset comprising ~36k earthquake and 25k noise samples. The code was released by our PhD student Alireza Niksejel.

The OBSTransformer is released and maintained at https://github.com/alirezaniki/OBSTransformer

Reference:

Niksejel A. and Zhang M. OBSTransformer: A Deep-Learning Seismic Phase Picker for OBS Data Using Automated Labelling and Transfer Learning. Geophysical Journal International, 2024, in press

DiTing-FOCALFLOW: An End-to-End Machine Learning-Based Earthquake Focal Mechanism Inversion Workflow

DiTing-FOCALFLOW utilizes a well-trained machine learning first motion classifier DiTingMotion and the HASH software to automatically invert focal mechanisms of earthquakes. The code was released by our visiting scientist Dr. Ming Zhao.

The DiTing-FOCALFLOW is released and maintained at https://github.com/mingzhaochina/DiTing-FOCALFLOW.

Reference:

Zhao, M., Xiao, Z. W., Zhang, M., Yang, Y., Tang, L., Chen, S. DiTingMotion: a deep-learning first-motion-polarity classifier and its application to focal mechanism inversion (Frontiers in Earth Science, 2023).

LOC-FLOW: An End-to-End Machine Learning-Based High-Precision Earthquake Location Workflow

LOC-FLOW is a “hands-free” machine learning-based earthquake location workflow to process continuous seismic records: from raw waveforms to well located earthquakes with magnitude calculations. The package assembles several popular routines for sequential earthquake location refinements, suitable for catalog building ranging from local to regional scales (a Chinese tutorial). 

The LOC-FLOW is released and maintained at https://github.com/Dal-mzhang/LOC-FLOW.

Reference:

Miao Zhang, Min Liu, Tian Feng, Ruijia Wang and Weiqiang Zhu. LOC-FLOW: An End-to-End Machine-Learning-Based High-Precision Earthquake Location Workflow, Seismological Research Letters, 2022, https://doi.org/10.1785/0220220019


Single Station Earthquake Location Through Full Waveform Matching 

https://github.com/Dal-mzhang/single-station-earthquake-location

A demo showing you how to locate earthquakes using a single station and waveform matching (a talk)


Reference:

Zhang M., Liu M., Plourde A., Bao F., Wang R. and Gosse J. Source characterization for two small earthquakes in Dartmouth, Nova Scotia, Canada: pushing the limit of single station. Seismological Research Letters, 2021, https://doi.org/10.1785/0220200297


Rapid Earthquake Association and Location (REAL)

https://github.com/Dal-mzhang/REAL

Method description:

 REAL (Rapid Earthquake Association and Location) associates arrivals of different seismic phases and locates seismic events primarily through counting the number of P and S picks and secondarily from traveltime residuals. A group of picks are associated with a particular earthquake if there are enough picks within the theoretical traveltime windows. The location is determined to be at the grid point with most picks. If multiple locations have the same maximum number of picks, the grid point with smallest traveltime residual is selected. We refine seismic locations using a least-squares location method (VELEST) and a high-precision relative location method (hypoDD). [ raw data -> high-resolution earthquake catalog]

References:

Match&Locate: A Package for Small Event Detection and Location

1. M&L1.0 package automatic download link      [An example could be requested by E-mail]

2. An updated version MatchLocate2.0 is available on Github

      (multiple templates, multiple phases, location resolution, MAD threshold, bandpass filtering et al.)

3. GPU-Match&Locate1.0 is available on Github

4. A recent application of high-precision earthquake detection and location using M&L (Liu et al., 2022)

Method description:

Compared to the current methods of small event detection (template matching/matched filter), the M&L method places event detection to a lower magnitude level and extends the capability of detecting small events that have large distance separations from the template. The method has little dependence on the accuracy of the velocity models used, and, at the same time, provides high-precision location information of the detected small-magnitude events.

References:

1. Zhang M. and Wen L. An effective method for small event detection: match and locate (M&L). Geophysical Journal International, 200 (3), 1523-1537, 2015. [download]

2. Zhang M. and Wen L. Seismological Evidence for a Low‐Yield Nuclear Test on 12 May 2010 in North Korea. Seismological Research Letters, 86 (1), 138-145, 2015. [download]

3. Zhang M. and Wen L. Earthquake characteristics before eruptions of Japan's Ontake volcano in 2007 and 2014. Geophysical Research Letters, 42 (17), 6982–6988, 2015. [download]