Content uploaded by Maria Adelaide Romano
Author content
All content in this area was uploaded by Maria Adelaide Romano on Nov 16, 2016
Content may be subject to copyright.
ESC2016-524, 2016
35th General Assembly of the European Seismological Commission
© Author(s) 2016. CC Attribution 3.0 License.
Probabilistic seismic hazard assessment for the Collalto area (Northern
Italy): preliminary results based on natural seismicity observations
Robin Gee (1,2), Laura Peruzza (1), Marco Pagani (2), Maria Adelaide Romano (1), and Enrico Priolo (1)
(1) Istituto Nazionale di Oceanografia e di Geofisica Sperimentale (OGS), Centro Ricerche Sismologiche (CRS), Borgo Grotta
Gigante, Italy, (2) Global Earthquake Model (GEM) Foundation, Pavia, Italy
We present preliminary results for a probabilistic seismic hazard assessment (PSHA) for an area centred around
Collalto Stoccaggio, a natural gas storage facility in Northern Italy. The Collalto site is located at the outer front
of the Venetian Alps fold and thrust belt, 50 km northwest of Venice in a region characterised by medium-high
seismic hazard. The storage exploits a depleted natural gas reservoir located within an actively growing anticline,
which is likely driven by the Montello Fault, the underlying blind thrust. The proximity of the Collalto site to this
structure and other active faults in the region requires a careful consideration of these faults in our study. At this
time, no correlation can be identified between the the gas storage activity and local seismicity, so we proceed with
a PSHA that considers only natural seismicity, where the rates of earthquakes are assumed to be time-independent.
The source model consists of faults and distributed seismicity to consider earthquakes that cannot be asso-
ciated to specific structures. For the fault sources, all active faults within 50 km of the site are considered.
They are characterised in terms of their geometry and slip rates using geological, geophysical and seismological
information. Distributed seismicity is characterised using a homogenised catalogue that incorporates the available
historical and instrumental data. Source modelling and hazard calculations are performed using the OpenQuake-
engine. We model the fault planes in 3D and investigate the impact on the hazard using ground motions prediction
equations (GMPEs) with different rupture-to-site distance metrics. We explore the sensitivity of the hazard results
to various parameters affected by epistemic uncertainty (e.g. fault geometry, maximum magnitude, different
GMPEs) and propose a logic tree to account for them.