TY - JOUR
T1 - Interaction between fault and off-fault seismic sources in hazard analysis – A case study from Israel
AU - Maiti, Soumya Kanti
AU - Kamai, Ronnie
N1 - Publisher Copyright:
© 2020 Elsevier B.V.
PY - 2020/9/5
Y1 - 2020/9/5
N2 - In this study, seismic source characterization (SSC) for purpose of seismic hazard analysis (SHA) is explored, with a focus on the moment balance and interaction between on-fault and off-fault seismicity. For the purpose of this study, a homogenous, uniform, earthquake catalog for Israel is first compiled and de-clustered. Gridded seismicity is calculated based on the de-clustered catalog, using a fixed kernel, as well as an adaptive kernel. The main faults are represented as planar features and their magnitude-frequency distribution is adjusted to match both the geological and the seismological available data. A full probabilistic seismic hazard analysis (PSHA) is then conducted for the entire state of Israel. Results are presented both in terms of the full spatial distribution, in the form of hazard maps, as well as at specific point locations - near and far from the main sources. Our results show that, based on the available data and balancing seismic moment, the characteristic portion of the main seismic sources in Israel is larger than the classical 94% (Suggested by Youngs and Coppersmith 1985) and is about 96%–98%, depending on the fault. This suggests that the exponential model, as currently used in the building code, leads to a significant underestimation of hazard. We further show that as long as the moment is balanced, hazard results are only mildly sensitive to model choices, such as size of kernel, or type of cutoff between on-fault and off-fault sources. Finally, we suggest using the adaptive-kernel with the Fault-extract interaction, which we believe is the best representation of the currently available geological, seismological and geodetic information.
AB - In this study, seismic source characterization (SSC) for purpose of seismic hazard analysis (SHA) is explored, with a focus on the moment balance and interaction between on-fault and off-fault seismicity. For the purpose of this study, a homogenous, uniform, earthquake catalog for Israel is first compiled and de-clustered. Gridded seismicity is calculated based on the de-clustered catalog, using a fixed kernel, as well as an adaptive kernel. The main faults are represented as planar features and their magnitude-frequency distribution is adjusted to match both the geological and the seismological available data. A full probabilistic seismic hazard analysis (PSHA) is then conducted for the entire state of Israel. Results are presented both in terms of the full spatial distribution, in the form of hazard maps, as well as at specific point locations - near and far from the main sources. Our results show that, based on the available data and balancing seismic moment, the characteristic portion of the main seismic sources in Israel is larger than the classical 94% (Suggested by Youngs and Coppersmith 1985) and is about 96%–98%, depending on the fault. This suggests that the exponential model, as currently used in the building code, leads to a significant underestimation of hazard. We further show that as long as the moment is balanced, hazard results are only mildly sensitive to model choices, such as size of kernel, or type of cutoff between on-fault and off-fault sources. Finally, we suggest using the adaptive-kernel with the Fault-extract interaction, which we believe is the best representation of the currently available geological, seismological and geodetic information.
UR - http://www.scopus.com/inward/record.url?scp=85086507398&partnerID=8YFLogxK
U2 - 10.1016/j.enggeo.2020.105723
DO - 10.1016/j.enggeo.2020.105723
M3 - Article
AN - SCOPUS:85086507398
SN - 0013-7952
VL - 274
JO - Engineering Geology
JF - Engineering Geology
M1 - 105723
ER -