TY - JOUR
T1 - Low SNR seismicity detection using features of 3D particle motions of direct P-waves
AU - Mukuhira, Yusuke
AU - Poliannikov, Oleg V.
AU - Fehler, Michael C.
AU - Moriya, Hirokazu
N1 - Funding Information:
This work was supported by ExxonMobil through its membership in the MIT Energy Initiative. We thank Brian deMartin for useful comments and data that contributed greatly to our work. We also thank KNMI for the seismic data. This study was supported by JSPS Overseas Research Fellow (20160228) and ERL, MIT.
Publisher Copyright:
© 2018 SEG
PY - 2018/8/27
Y1 - 2018/8/27
N2 - Detection and analysis of seismicity is valuable for characterizing and better understanding of reservoirs. At the P-wave arrival, particle motion has a linear polarization and the direction of arrival is oriented to the source zone along the ray path. We extract these characteristics of direct P-wave arrivals by evaluating the 3D particle motions and use them for event detection purposes. Particle motion is evaluated using spectral matrix analysis in the time-frequency domain over the frequency band having the highest power. We also use information about reference events in a catalog such as moveout and inclination of direction of arrival to detect events within a particular seismogenic region. We apply our method to field seismicity data recorded at multiple wells each having multiple receivers located in the Groningen gas field The Netherlands. Our method detects previously undetected (buried) events that are not in the catalog.
AB - Detection and analysis of seismicity is valuable for characterizing and better understanding of reservoirs. At the P-wave arrival, particle motion has a linear polarization and the direction of arrival is oriented to the source zone along the ray path. We extract these characteristics of direct P-wave arrivals by evaluating the 3D particle motions and use them for event detection purposes. Particle motion is evaluated using spectral matrix analysis in the time-frequency domain over the frequency band having the highest power. We also use information about reference events in a catalog such as moveout and inclination of direction of arrival to detect events within a particular seismogenic region. We apply our method to field seismicity data recorded at multiple wells each having multiple receivers located in the Groningen gas field The Netherlands. Our method detects previously undetected (buried) events that are not in the catalog.
UR - http://www.scopus.com/inward/record.url?scp=85098683355&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85098683355&partnerID=8YFLogxK
U2 - 10.1190/segam2018-2965365.1
DO - 10.1190/segam2018-2965365.1
M3 - Conference article
AN - SCOPUS:85098683355
SP - 2882
EP - 2886
JO - SEG Technical Program Expanded Abstracts
JF - SEG Technical Program Expanded Abstracts
SN - 1052-3812
T2 - Society of Exploration Geophysicists International Exposition and 88th Annual Meeting, SEG 2018
Y2 - 14 October 2018 through 19 October 2018
ER -