Reply on RC2

Jonatan Glehman

doi:10.5194/nhess-2021-280-ac2

Abstract

Abstract. In Israel, due to low seismicity rates and a sparse seismic network, the temporal and spatial coverage of ground motion data is insufficient to estimate the variability in moderateâstrong (M>6) ground motions required to construct a local ground motion model (GMM). To fill this data gap and to study the ground motion variability in M>6 events, we performed a series of 3-D numerical simulations of Mâ6 and Mâ7 earthquakes. Based on the results of the simulations, we developed a parametric attenuation model (AM) and studied the residuals between simulated and AM peak ground velocities (PGVs) and the single station variability. We also compared the simulated ground motions with a global GMM in terms of PGV and significant duration (DsÂ 595). Our results suggested that the AM was unable to fully capture the simulated ground motion variability mainly due to the incorporation of super-shear rupture and effects of local sedimentary structures. We also showed that an imported GMM considerably deviates from simulated ground motions. This work sets the basis for future development of a comprehensive GMM for Israel, accounting for local source, path, and site effects.

Highlights

Disaster Risk Reduction (UNDRR) – Human Cost of Disasters, 2000 - 2019 – clearly shows that earthquakes are the deadliest natural disasters
The hypocenter for the Dead Sea Transform (DST) events was placed in the middle of the seismogenic depth; 11 and 13 Km, for the M 6 and M 7 respectively, for the M 6 Carmel Fault Zone (CFZ), the value was set to 12 Km
Our analysis shows that the attenuation model (AM) was unable to fully capture the variability of the simulated ground motions

Summary

Introduction

Disaster Risk Reduction (UNDRR) – Human Cost of Disasters, 2000 - 2019 – clearly shows that earthquakes are the deadliest natural disasters. Showing that accounting for path effects leads to a smaller value of the aleatory variability and results in different median predictions, depending on source and site location To achieve this improvement, Kuehn et al, (2019) divided California into a grid with a cell size of 30 km by 30 km and used 12,039 records from 274 events recorded at 1504 stations. The shift from ergodic models to nonergodic models, which account for local source-site and path effects such as numerical models, leads to large epistemic uncertainty in the median ground motion, resulting in increased epistemic uncertainty of the hazard (Walling & Abrahamson, 2012).

Seismicity and seismic hazard in Israel

Numerical model

Source maximum and average slip

Jordan Valley

IM a b c d e

Findings

Data and resources