The accurate prediction of the level and variability of (potentially damaging) near‐source strong‐ground motions in future earthquakes is one of the key challenges for seismologists and earthquake engineers. The increasing number of near‐source recordings collected by dense strong‐motion networks exemplifies the inherent complexity of near-field ground shaking, governed by a number of (partially interacting) physical processes. Characterizing, quantifying, and modeling (either by means of empirical scaling relations or by numerical simulations) ground‐motion complexity requires the joint investigation of three dominant ingredients: (I) the physics of earthquake rupture ; (II) the details of wave‐propagation in heterogeneous media ; (III) the effects of local site conditions .

This article discusses briefly the beginnings of strong‐motion seismology and the recognition of ground‐motion complexity. Using two well recorded recent earthquakes, I introduce observational aspects of near-field ground shaking and the basic mathematical description for computing ground motion. The article proceeds by describing each of the three “ground‐motion ingredients” in some detail, but does not attempt to provide an in-depth review of all the scientific advancements in these fields. Rather, I explain the key elements for characterizing and modeling ground‐motion complexity, supplemented with a concise overview of the underlying physical processes. Current research increasingly incorporates advanced physical concepts into standard practice, therefore leading to improved strong‐motion simulation approaches to accurately predict intensity and variability of near‐source shaking.