Publication Date:
2024-01-30
Description:
In this study, we investigate the dependencies between ground-motion intensity measures (GMIM) and earthquake magnitudes (M), in order to evaluate the dynamic stress parameter (Δσ) magnitude scaling. To achieve this, two types of datasets are used: a large subset ofthe NGA-West 2 (next generation attenuation) dataset including 1700 records from 426 sites and 271 earthquakes. The other datasets are generated through the stochastic method (Boore 2003)assuming various magnitude dependencies (constant and variable) of the stress parameter with magnitude. Adaptive neuro-fuzzy inference systems (ANFIS) are used to derive datadriven ground-motion prediction models (Ameur et al. 2018). Stiff soil (Vs30 〉 500 m/s) data are selected and the ground-motion models are depending on two input parameters: the moment magnitude (Mw)and thehypocentral distance (Rhyp). Following Molkenthin et al. (2014), we assume that Δσ is the dominating controlling factor of GMIM for stiff site conditions at Rhyp = 30 km, at the frequency (f) = 3.33 Hz for moderate earthquakes in the magnitude range Mw = [4.5–6.5]. This study confirms that the relations between magnitude and stress parameter control the scaling of ground motions. We show that the magnitude-dependent stress drops better fit the latest generation of NGA-West 2 datasets and empirical ground-motion equations. We finally calibrate a relation between dynamic stress parameter and earthquake magnitude in the magnitude range Mw =[4.5–6.5].
Language:
English
Type:
info:eu-repo/semantics/article
Format:
application/pdf
Permalink
