Physics-based Simulator of Short- and Long-Term Seismicity: application
to the Central Apennines Region
Abstract
The application of a physics-based earthquake simulation algorithm to
the central Apennines region, where the 24 August 2016 Amatrice
earthquake (Mw6.2) occurred, allowed the compilation of a synthetic
seismic catalog lasting 100,000 years, and containing more than 300,000
M ≥ 4.0 events, without the limitations that real catalogs suffer in
terms of completeness, homogeneity and time duration. The seismogenic
model upon which we applied the simulator code, was derived from the
DISS 3.2.1 database (http://diss.rm.ingv.it/diss/), selecting all the
fault systems that are recognized in the central Apennines region, with
a total of 54 fault rectangular segments. The physical model on which
the last version of our simulation algorithm is based includes, besides
tectonic stress loading and static stress transfer as in the previous
versions, also the Rate & State constitutive law. The resulting
synthetic catalog exhibits typical magnitude, space and time features,
which are comparable with those of real observations. These features
include (i) an earthquake magnitude distribution departing from the
linear Gutenberg-Richter distribution in the moderate and higher
magnitude range, (ii) long–term pseudo-periodicity of strong
earthquakes and (iii) short- and medium-term earthquake clustering. We
found in our synthetic catalog a clear trend of long-term acceleration
of seismic activity preceding M ≥ 6.0 earthquakes and quiescence
following those earthquakes. A typical aspect of the observed seismicity
in Italy and of the Central Apennines region in particular, is the
occurrence of multiple events of M ≥ 5.5 earthquakes close to each other
in space and time. A special attention was devoted to verify if the
synthetic catalog includes this feature. On this purpose, we applied a
specific code written in “Mathematica” to count the number of multiple
events contained in a seismic catalog under a quantitative definition,
finding that the synthetic catalog contains a large number of multiple
events but not as frequently as in the real catalog. Lastly, as an
example of a possible use of synthetic catalogs, a ground motion
prediction equation was applied to all the events reported in the
synthetic catalog for the production of PSHA maps (in terms of PGA).