This paper investigates a method for improving the selection of seismic motions for designing earthquake-resistant underground structures. It is found that PGV alone is unreliable as a predictor of structural damage with increasing earthquake intensity. Therefore, based on characterizing seismic intensity by using PGV, another parameter, referred to here as the severest parameter, is introduced to distinguish potential damage capacity for different seismic motions. A numerical model of a soil-underground structure system was established using the finite element software OpenSees. A total of 120 real ground motions were selected for the model, considering the influences of eight different site groups on the underground station and the rupture distances of the input seismic motions. The results show that as seismic intensity increases, substantial variability in the response of underground structures emerges under the same amplitude of PGV, diminishing the effectiveness of the relationship between PGV and structural damage. When assessing the potential damage capacity of seismic motions with similar or close amplitudes of PGV, VSI is an appropriate severest parameter for Class III sites and ASI is suitable for Class II sites. When the correlation coefficient between the severest parameter and the structural response is greater than 0.8, it can be used to reliably assess seismic damage capacity based on the size of the severest parameter.

Combining intrusive geotechnical site investigations with non-intrusive geophysical surveys is a cost-effective approach to producing data with varying levels of accuracy, uncertainty, and different spatial scales to better characterize the site's liquefaction properties. Moreover, the demand for three-dimensional (3D) subsurface models in geotechnical engineering is increasing, but the models contain uncertainties and spatial variability associated with the use of relevant stochastic and geostatistical methods, and it remains a challenging task to obtain reliable liquefaction assessment results and the corresponding damage capacity. This study proposes a data-driven and non-parametric form of 3D multi-source fusion Bayesian compressive sampling (3D MSF-BCS) method for assessing 3D soil liquefaction-induced damage capacity. It consists of three main components: (i) 3D MSF-BCS fusing sparse geotechnical data (e.g., cone penetration test (CPT)) and geophysical data (e.g., multichannel analysis of surface waves (MASW)) for 3D site modeling, (ii) quantifying the accuracy and uncertainty of 3D MSF-BCS, and (iii) Soil liquefaction-induced damage capacity analysis in 3D space. The method was applied to numerical examples and a real case study at the Cresselly Place site, and the results showed that the proposed method performs well.