By employing the frequency-wavenumber (FK) method to simulate the propagation of seismic wavefield in the crustal layer, using the spectral element method (SEM) to simulate the propagation of wavefield in the near-surface soil, and using the multi-degree-of-freedom (MDOF) model to simulate the seismic response of building clusters in city, this paper establishes the FK-SE-MDOF approach (a two-step method) for urban earthquake disaster analysis of fault-to-city based on the concept of domain reduction. The approach can simultaneously consider factors such as earthquake source parameters, propagation paths, local site effects, site-city interaction (SCI) effects, and the dynamic nonlinear responses of buildings (hereafter referred to as source-to-city factors) in a physics-based model. Firstly, the theories of the approach were introduced, and the correctness of the approach was verified. Furthermore, the applicability and the necessity of considering source-to-city factors were examined using a building cluster on an ideal sedimentary basin under the action of a point dislocation source. Finally, the seismic response of buildings in a region was simulated using buildings in the Nankai District of Tianjin as examples. This approach avoids the influences caused by expert experience differences in empirical and hybrid methods, establishes a connection between fault rupture and buildings dynamic response, and can more realistically reflect the distribution of seismic wavefields, building seismic responses, and damage state distribution under the earthquake scenario. It can be applied to earthquake disaster simulation for urban buildings at the scale of tens of thousands of buildings, and the simulation results can provide quantitative guidance for urban planning, earthquake-resistant design, risk assessment, post-earthquake rescue, etc.

Seismic fragility analysis is a crucial tool for assessing the seismic performance of buildings. In areas with dense clusters of tall buildings, the significant site-city interaction (SCI) effect alters wave propagation mechanisms, influencing the seismic fragility of structures. However, a significant increase in computational workload results from the need for detailed modeling of sites and building clusters for the SCI analysis. To address this challenge, this work first investigates the minimum number of earthquake waves required to characterize SCI-induced response changes. The Central Business District of Shanghai is analyzed. A table for the recommended minimum number for a given accuracy requirement and prediction reliability is provided. Moreover, a seismic fragility analysis method considering the SCI effect is proposed for low-rise buildings. The case study indicates that, buildings with similar height will exhibit various fragility changes after considering SCI. For the complete damage state, the mean intensity value of the fragility curve can be 14.4 % smaller than that without SCI. In addition, this approach provides significant computational workload reduction. For the case study, the computational workload of the proposed method is roughly 1/50 of that using traditional IDA method.