The choice of the Lower Rhine Embayment as study area for strong ground motion modeling may be puzzling at first glance. This region in the northwest of the European continent is characterized by active tectonics on a complex system of fault-zones with relatively low deformation rates. Consequently, the area has shown low to moderate seismicity in the time frame covered by observational seismology. However, historical and geological evidence proves that the fault systems of the Lower Rhine Embayment have the potential of producing large earthquakes with magnitudes 6 and above accompanied by surface rupture. The presence of large sediment deposits in this region leads to local amplification of ground motion with large lateral variations. Dense population and an agglomeration of industry results in an elevated seismic risk.

Assessment of seismic hazard in regions characterized by low recent seismicity is afflicted with large uncertainties. This is mainly due to the dearth of observational data on strong ground motions associated with large earthquakes. Numerical simulations of earthquake scenarios can account for estimates on peak ground motion and waveforms and therefore help closing this gap.

Naturally the first step consists in accurate reproduction of the few observed events. An additional crucial quantity is the range of variations of the simulation results within the uncertainty margins associated with input parameters. Knowledge about this behavior enlarges the significance of numerical simulation results.

Four historical and recent earthquake scenarios are modeled using a finite difference approach. Results are analyzed with special emphasis given to their intrinsic variability with model complexity and simulation settings. The choice of investigated parameters is adopted to the differing scope of observational data available for the individual events. In general encouraging similarity between synthetic and observed ground motions is found, even when a simplified model is used. However, detailed investigation carried out for the most recent earthquake scenario - the magnitude 4.9 July 22 2002 Alsdorf event - strongly suggests the significance of an appropriate source description and the modeling of anelastic behavior on simulation results. Finally a web-based application for storage and visualization of synthetic ground motion data is presented.