Seismic microzonation studies supported by dynamic numerical simulation. Application to Dushanbe (Tajikistan), Aachen (Germany) and Bukit Timah (Singapore)

RWTH Publications (RWTH Aachen)

Abstract

Urban areas located in seismically active zones require reliable and detailed seismic hazard assessments to ensure public safety and support sustainable urban development. This thesis presents a comprehensive multi-component approach to seismic microzonation based on the integration of passive seismic methods, active geophysical surveys, and two-dimensional (2D) dynamic numerical modelling. The study encompasses three geologically distinct and seismotectonically contrasting regions: Dushanbe (Tajikistan), Aachen (Germany), and the Bukit Timah granite area in Singapore. In Dushanbe (Chapter 4), a high seismic hazard zone exposed to both shallow crustal earthquakes on the Hissar–Kokshaal and Ilyak–Vaksh fault systems and to intermediate- to deep-focus earthquakes from the Hindu Kush–Pamir zone, where outdated hazard maps and building codes fail to account for soil heterogeneity and anthropogenic influence, extensive microtremor measurements, seismic refraction tomography, and both horizontal-to-vertical spectral ratio (HVSR) and standard spectral ratio (SSR) analyses were conducted using permanent and temporary seismic stations. The collected data were used to construct a three-dimensional (3D) geological model (12 × 12 km²), from which 2D geological cross-sections were extracted for numerical modelling of seismic wave amplification and estimation of local peak ground acceleration (PGA). The results revealed significant seismic amplification driven by local geological and topographic conditions, enabling refined PGA estimations and a more reliable basis for microzonation. A further refinement and quantification of site effects in Dushanbe was carried out through the development of a detailed 3D geological model and 2D dynamic numerical simulations (Chapter 5). The simulations provided new insights into the spatial variation of PGA, resonance frequencies, and the influence of subsurface and topographic conditions, improving seismic risk evaluation for the city. The modelled PGA distribution was validated against accelerations recorded during the 1980 Dushanbe (Mw 5.0) and 1989 Hissar (Mw 5.8) earthquakes, confirming that thick loess deposits in the northeastern and northwestern parts of the city are particularly susceptible to strong ground motion amplification. In Aachen (Chapter 6), located in a region of moderate intraplate seismicity within the Lower Rhine Embayment extensional system, a detailed interdisciplinary geophysical survey was conducted in previously underexplored districts using 475 HVSR measurements, Rayleigh-wave ellipticity inversion, microtremor array measurements (MAM), electrical resistivity tomography (ERT), and geotechnical borehole data. Resonance frequency maps derived from HVSR and ellipticity inversion allowed for improved estimation of the average shear-wave velocity in the upper 30 metres (Vs30). Particular attention was given to active normal faults such as the Laurensberg Fault, investigated through 2D dynamic numerical modelling along cross-section A–A′. The comparison between two profiles, one including and the other excluding the fault, highlighted the importance of incorporating fault structures and stratigraphic features into site-response models. In addition, the integrated analysis of ERT, HVSR, and MAM data at the Ferberpark sector provided, for the first time, a precise geophysical localisation of a previously unnamed fault. The computed PGA values along A–A′ (0.23–0.35 g) demonstrated that, even within a single earthquake zone defined by DIN 4149 (2005), ground motion parameters vary considerably, underscoring the need for site-specific microzonation. In Singapore (Chapter 7), which combines the two former sub-studies on the Bukit Timah granite massif, the first component demonstrated the effectiveness of Rayleigh-wave ellipticity inversion of HVSR data, constrained by borehole information, in constructing shear-wave velocity (Vs) profiles. The resulting Vs30 values showed good agreement with those obtained from multichannel analysis of surface waves (MASW) and MAM, confirming the cost-efficiency and reliability of HVSR inversion for site classification in accordance with international standards (NEHRP site class D). The second component assessed the impact of granite weathering on local seismic response through 2D dynamic numerical modelling. The modelling results showed a high level of agreement with HVSR observations, emphasising the role of lithological weathering in shaping the local seismic response, which is particularly relevant for improving design practice in tectonically stable regions subjected to far-field seismicity from the Sumatran subduction zone. The synthesis of the thesis (Chapter 8) consolidates the data and findings from the three case studies and provides a coherent evaluation of the applied methodology across the full spectrum of seismotectonic contexts, from near-field shallow and deep-focus seismicity (Dushanbe), through moderate intraplate seismicity on active normal faults (Aachen), to far-field subduction-related seismicity (Bukit Timah). The results demonstrate the effectiveness of a unified methodology that integrates passive seismic techniques, active geophysical surveys, and 2D dynamic numerical modelling, offering a flexible and scalable framework for seismic microzonation applicable to diverse urban environments worldwide. Future research directions (Chapter 9) include high-resolution 3D dynamic simulations explicitly accounting for the different types of seismic sources, targeted geophysical campaigns to further delineate active fault structures, probabilistic hazard assessments, and the integration of machine learning and large language models for site classification and seismic risk communication. The results provide a scientific basis for the refinement of seismic building codes and the development of site-specific mitigation strategies in both high- and moderate-seismicity regions.

Authors 1

1. Farkhod Hakimov corresponding Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen