Restrictions to the application of ‘diagnostic’ criteria for recognizing ancient seismites

AbstractSoft-sediment deformation structures induced by seismic liquefaction and/or fluidization receive much attention in sedimentological, structural and palaeoseismic studies. The direct record of larger earthquakes is restricted to instrumental and historical data; the recognition of prehistoric earthquakes requires criteria to recognize seismites in the geological record. The areal distribution of seismites can sometimes be related to active faults since distances to the epicenter (for a given magnitude) tend to be related to the liquefaction effects of seismic shocks.The use of soft-sediment deformation structures for palaeoseismic studies has limitations, however. Hardly anything is known, for instance, about the effects that modern seismic events have on the sediments in most environments. Moreover, criteria for the recognition of seismites are still under discussion. The following characteristics seem, particularly in combination, the most reliable: (1) Soft-sediment deformation structures should occur in laterally continuous, preferably recurring horizons, separated by undeformed beds; (2) These deformation structures should be comparable with structures known to have been triggered by modern seismic activity; (3) The sedimentary basin should have experienced tectonic activity at the time when the deformations were formed; and (4) The intensity or abundance of the soft-sediment deformation structures in a presumed seismite should change laterally, depending on the distance to the epicenter. It turns out that all of these four criteria have important exceptions. (1) Soft-sediment deformation structures occurring over large lateral distances in a specific layer can be triggered also by other processes. Moreover, in environments with a low sedimentation rate, the time between successive earthquakes is often too short to allow accumulation of beds that remain undisturbed. Furthermore, total liquefaction of a sandy bed may result in the absence of deformation features. (2) No truly diagnostic soft-sediment deformation structures exist to prove seismic activity. Moreover, the final configuration of a soft-sediment deformation structure is independent of the type of trigger. (3) Seismites occur frequently in areas where seismic activity is low today. (4) The lateral changes in the intensity of soft-sediment deformation structures in seismites as a factor presumed to depend on the distances to the epicenter, pose a complicated problem. The 2012 Emilia earthquakes, for instance, affected sandy fluvial channels but not the fine-grained floodplains.It must thus be deduced that specific soft-sediment deformation structures cannot be used without additional evidence to identify seismites. In particular, the magnitude of seismic shocks and the recurrence time of main events (the most important features that allow recognition of seismites) seem to be sedimentological in nature: facies changes in space and time seem the parameters that most strongly control the occurrence, morphology, lateral extent and the vertical repetition of seismites