We have been investigating the phenomena of fault displacements and long-period ground motions near the surface rupture of past earthquakes for critical structures such as super high-rise buildings, long-span bridges, large oil tanks and nuclear installations. We are aware that for the evaluation and further understanding of such phenomena near the source, physics-based rupture models, as a complement of empirical approaches, may be appropriate, so that the predictions can be supported by physics, rather than purely empirical approaches. The main reason to use physics-based rupture models is that such phenomena are source-dominated phenomena, a feature that current empirical models are not capable to capture, mainly because of the sparseness of observed data near the source that posed challenges to provide meaningful and reliable ground motion and fault displacement predictions. This issue of lack of observed data is not only in magnitude and distance range, but also in frequency content range of ground motion as well as in site conditions such as on hard rock. The motivation to carry out this study is that the coseismic fault displacement and long period ground motion near the source associated to earthquakes may seriously compromise the safety of critical infrastructures and buildings located near faults. Therefore, this issue becomes critical for site-specific Seismic Hazard Analysis (SHA) and Fault Displacement Hazard Analysis (FDHA) of such critical infrastructures. Our final goal of this research is to bring the physics-based rupture models into practice of SHA and FDHA. For this purpose, simple and practical dynamic asperity fault models are used to evaluate past and future earthquake. The 2016 Mw 7.0 Kumamoto (Japan) earthquake is used as a case study to evaluate fault displacement and long period near-source ground motion. Surface-rupturing was observed along 34km of the main fault reaching values of fault displacement up to around 2.0m, as well as very near source ground motion has been recorded at stations with distance less than 1.0km from the surface rupture. The general procedure of dynamic rupture simulation follows the method proposed by Dalguer et al. (2019), in which the dynamic asperity models, as a first step, are constrained with kinematic asperity models already consistent with the observed near-source strong ground motion, and then the shallow layer (SL) zone of the fault is calibrated to predict fault displacement and long period ground motion at the very near-source. The main conclusions are as follow: 1) Two SL zones with enhanced energy absorption mechanism, respectively, in the NE and SW side of the fault are required to predict fault displacement consistent with observation. 2) In general, buried rupture penetrating the SL zone and surface rupture models can equally fit the observed ground motion and permanent displacement at the near-source stations. Implying that the surface rupturing contribution to the near-source ground motion is low, but the contribution of the SL zone is significant. 3) The surface rupture extension and fault displacement amplitudes are controlled by the shallow asperities, but the SL zone characteristics define the final fault displacement. 4) the ratio between the fault displacement and final slip at the SL zones is around 0.5. We argue that this ratio can serve as a metric to quantify the contribution of surface rupturing to the near source ground motion and permanent displacement.