Paleoearthquakes along the Irpinia fault at Pantano di San Gregorio Magno (southern Italy)

Abstract

We excavated trenches across the causative fault of the 23 November 1980, Campania-Lucania earthquake, named Irpinia Fault. The first experiment was performed in 1989 at Piano di Pecore, along the middle section of the fault (Pantosti et al., 1989a, 1989b, 1991). Two new trenches were excavated in 1990 at Pantano di San Gregorio Magno, a closed elongated depression located near the southern end of the fault and filled with Quaternary and Holocene sediments. A careful study of the stratigraphic sequence exposed in the trenches (that is extensively described in the paper) allowed us to recognize the geologic records of four paleoearthquakes that occurred prior to 1980. The four events share a similar dislocation magnitude and geometry with the most recent earthquake. Radiometric dating of organic samples collected at different levels of the stratigraphic sequence yielded a time interval of occurrence for individual paleoevents and allowed us to calculate a fault slip rate and repeat time for 1980-type earthquakes of 0.3 mm/yr and 1,950 yr, respectively. The positive agreement between these results and those obtained at the Piano di Pecore site indicates that the 23 November 1980 earthquake is strictly characteristic for the Irpinia Fault (in the sense of Schwartz & Coppersmith, 1984). The new observations provide an unprecedented tool for understanding the main characteristics of seismicity in the southern Apennines, but also highlight a basic discrepancy with the historical evidence: based on the Italian catalogue of historical seismicity, earthquakes of magnitude ≡ 7 occur every 300-500 years, whereas the repeat time suggested by paleoseismology is four to six time larger. The paper suggests a number of possible explanations and delineates research avenues that could be followed to address this paradox.