Geotechnical data based seismic microzonation in Seoul using region-specific and code-based site amplification models

Abstract

We develop seismic microzonation maps in Seoul employing a GIS-aided approach. We use extensive borehole database of Seoul, which include stratigraphical information and standard penetration test (SPT) results. Two outlier detection methods are used to remove unreliable data. VS profiles are developed empirically both from SPT blow counts where available and averaged values representative of geologic soil layers where only stratigraphical information are provided. The sites are classified using the Korean design system (KDS) and a revised site classification system of Aaqib et al. (Earthq Eng 26:8257-8279, 2022, DOI:10.1080/13632469.2021.1990164), (AEA22). It is revealed that the KDS system, which uses the bedrock depth and average VS of soil, is not effective in categorizing the sites, resulting in significant overlapping of site parameters for different classes. The site period dependent AEA22 system better captures the surface geologic variations in Seoul, producing cleaner yet more representative classification map. We further develop the surface spectral acceleration maps of Seoul subjected to a 2400-year return period earthquake. The seismic amplifications are calculated using three procedures, which are (1) simulation-based site amplification model of Aaqib et al. (Earthq Spec 37:1900-1930, 2021, DOI:10.1177/8755293020981984) developed for shallow bedrock sites of Korea, (2) KDS and (3) AEA22 amplification factors. The site amplification model is considered as the reference method. The KDS system produces unrealistic predictions, underestimating the short period accelerations while overestimating the mid-to-long period responses. The AEA22 system produces an enhanced predictions of the surface accelerations compared with the KDS system. We propose to utilize the site amplification model or the AEA22 system for estimation of the seismic hazard in Seoul.