The lack of a well-recognized relation between large earthquakes and their surface effects in Italy has motivated the detailed reconstruction of the surface expression of the 1980 Campania-Lucania normal faulting event (MS=6.9). Instrumental investigations showed that it was a complex event, composed of four or more subevents occurring within 40 s of the origin time. This earthquake was the first in Italian history, and among the few in the Mediterranean basin, to produce sizable and coherent surface faulting. On the basis of field work we reconstructed three main strands which together form a 38-km-long, northwest trending fault scarp. Leveling observations have been used together with field observations to describe the faulting model; geological and geodetic estimates of the focal parameters inferred for each subevent have been compared with similar estimates on the basis of instrumental observations. A good consistency is found between the different models of the 1980 earthquake, with respect to both the magnitude and the geometry of the subevents. The total moment inferred from surface observations is comparable with that computed using long-period waveform modeling and strong motion accelerograms and is 30% lower than the moment tensor inversion estimate. Much effort has been put into understanding the coseismic deformation-surface topography interaction.

Observations of back-dipping scarps which tend to reverse the present topography indicate the young age of the presently active tectonic regime. The variability of the surface faulting and the diversion of parts of the scarp from the main trend have been interpreted in terms of intrinsic fragmentation of the characteristic earthquake-generating 1980 segment; gaps in the scarp height and changes in the fault trend and relation to topography have been identified with persisting ''relaxation geometric barriers,'' which have been found coincident with the actual rupture initiation sites. The peculiarities of the 1980 earthquake and the youthfulness of its tectonic framework illustrate the first stages of landscape development in a tectonically active region. A correct understanding of this process in well-monitored regions is essential to recognize long-term surface features related to large earthquakes of the past and therefore to provide robust estimates for seismic harzard analyses. ¿ American Geophysical Union 1990