Abstract
Machine learning regression can predict macroscopic fault properties such as shear stress, friction, and time to failure using continuous records of fault zone acoustic emissions. Here we show that a similar approach is successful using event catalogs derived from the continuous data. Our methods are applicable to catalogs of arbitrary scale and magnitude of completeness. We investigate how machine learning regression from an event catalog of laboratory earthquakes performs as a function of the catalog magnitude of completeness. We find that strong model performance requires a sufficiently low magnitude of completeness, and below this magnitude of completeness, model performance saturates.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 13,269-13,276 |
| Journal | Geophysical Research Letters |
| Volume | 45 |
| Issue number | 24 |
| DOIs | |
| State | Published - Dec 28 2018 |
Fingerprint
All Science Journal Classification (ASJC) codes
- Geophysics
- Earth and Planetary Sciences(all)
Cite this
}
Earthquake Catalog-Based Machine Learning Identification of Laboratory Fault States and the Effects of Magnitude of Completeness. / Lubbers, Nicholas; Bolton, David C.; Mohd-Yusof, Jamaludin; Marone, Chris J.; Barros, Kipton; Johnson, Paul A.
In: Geophysical Research Letters, Vol. 45, No. 24, 28.12.2018, p. 13,269-13,276.Research output: Contribution to journal › Article
TY - JOUR
T1 - Earthquake Catalog-Based Machine Learning Identification of Laboratory Fault States and the Effects of Magnitude of Completeness
AU - Lubbers, Nicholas
AU - Bolton, David C.
AU - Mohd-Yusof, Jamaludin
AU - Marone, Chris J.
AU - Barros, Kipton
AU - Johnson, Paul A.
PY - 2018/12/28
Y1 - 2018/12/28
N2 - Machine learning regression can predict macroscopic fault properties such as shear stress, friction, and time to failure using continuous records of fault zone acoustic emissions. Here we show that a similar approach is successful using event catalogs derived from the continuous data. Our methods are applicable to catalogs of arbitrary scale and magnitude of completeness. We investigate how machine learning regression from an event catalog of laboratory earthquakes performs as a function of the catalog magnitude of completeness. We find that strong model performance requires a sufficiently low magnitude of completeness, and below this magnitude of completeness, model performance saturates.
AB - Machine learning regression can predict macroscopic fault properties such as shear stress, friction, and time to failure using continuous records of fault zone acoustic emissions. Here we show that a similar approach is successful using event catalogs derived from the continuous data. Our methods are applicable to catalogs of arbitrary scale and magnitude of completeness. We investigate how machine learning regression from an event catalog of laboratory earthquakes performs as a function of the catalog magnitude of completeness. We find that strong model performance requires a sufficiently low magnitude of completeness, and below this magnitude of completeness, model performance saturates.
UR - http://www.scopus.com/inward/record.url?scp=85058989103&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85058989103&partnerID=8YFLogxK
U2 - 10.1029/2018GL079712
DO - 10.1029/2018GL079712
M3 - Article
VL - 45
SP - 13,269-13,276
JO - Geophysical Research Letters
JF - Geophysical Research Letters
SN - 0094-8276
IS - 24
ER -