Impact of several site-condition proxies and ground-motion intensity measures on the spectral amplification factor using neuro-fuzzy approach: an example on the KiK-Net dataset

Mohammed Akram Ismail Zaoui (), Boumédiène Derras and Julie Régnier
Additional contact information
Mohammed Akram Ismail Zaoui: Dr Moulay Tahar University
Boumédiène Derras: Dr Moulay Tahar University
Julie Régnier: CEREMA Méditerranée

Natural Hazards: Journal of the International Society for the Prevention and Mitigation of Natural Hazards, 2025, vol. 121, issue 7, No 38, 8703-8732

Abstract: Abstract The aim of this paper is to utilize noted machine learning algorithm to examine the impact of 22 Site Condition Proxies (SCPs) and 3 Ground Motion Intensity Measurements (GMIMs) on the Response Spectral Amplification Factor (SAF). The SAF represents the ratio of Pseudo-Spectral Acceleration with 5% damping between surface and downhole measurements, with the downhole data serving as the reference site. The 22 SCPs include average shear velocities at various depths (VS10, VS20, VS30, VS50, VS100), Average shear wave velocity from the ground surface up to the depth at which shear velocity reaches 800 m/s, so-called H800 (VS800), mean shear velocity to the total depth of the profile (Vsmean), minimum and maximum shear velocities (Vsmin, Vsmax), shear wave velocity at the location of the borehole instrument (Vszhole), contrast velocities (CVs), gradient velocity to 30 m (B30), topographic slope, depth at which shear velocity reaches 800 m/s (H800), and frequency peaks of the site response (fn). The 3 GMIMs consist of peak ground acceleration, peak ground velocity (PGV), and a proxy for maximum strain (PGV/VS30). This SAF is developed using a subset of the KiK-Net database and employs the Adaptive Neuro-Fuzzy Inference System (ANFIS) machine learning algorithm. The input variables of the model are the SCPs and GMIMs, while the target variables are the SAF values at different periods, ranging from 0.01 to 4 s. A sensitivity analysis is conducted to identify the most relevant SCPs and GMIMs, using the variance reduction coefficient (RC) as a metric to evaluate model performance. Among the GMIMs, PGV/VS30 is identified as the most effective. However, no single SCP is found to be relevant across all periods; therefore, a combination of two SCPs is chosen, with the optimal pair being [f0HV: fundamental resonance frequency using horizontal-to-vertical spectral ratio, CV2 = Vsmax/VS30,]. The final combination includes [f0HV, CV2, PGV/VS30]. This combination accounts for the different characteristics of SCPs, such as rigidity and depth, as well as a ground motion intensity parameter. The ANFIS model is validated through three methods: comparison with models proposed by (Earth Planets Sp 69:1–21, 10.1186/s40623-017-0718-z, 2017; Bull Earthq Eng 17:119–139, 10.1007/s10518-018-0459-9, 2019), testing with the same data used in the training phase, and validation with new data (Bahrampouri et al. in Earthq Spectra 37:505–522, 10.1177/8755293020952447, 2021). This sensitivity study and the developed model aim to enhance the consideration of local site effects, both for seismic code revisions and site-specific studies.

Keywords: Site effect; KiK-Net; Response spectra amplification factor; Site condition proxies; Neuro fuzzy inference system machine; Learning (ANFIS) (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s11069-025-07151-0 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:nathaz:v:121:y:2025:i:7:d:10.1007_s11069-025-07151-0

Ordering information: This journal article can be ordered from
http://www.springer.com/economics/journal/11069

DOI: 10.1007/s11069-025-07151-0

Access Statistics for this article

Natural Hazards: Journal of the International Society for the Prevention and Mitigation of Natural Hazards is currently edited by Thomas Glade, Tad S. Murty and Vladimír Schenk

More articles in Natural Hazards: Journal of the International Society for the Prevention and Mitigation of Natural Hazards from Springer, International Society for the Prevention and Mitigation of Natural Hazards
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().