One of the most challenging topic in earthquake geology is to characterize the seismogenic sources, i.e. the
potential causative faults of earthquakes. The main seismogenic layer is located in the upper brittle crust.
Nevertheless it does not mean that a fault take up the whole schizosphere: i.e. from the brittle-plastic transition to
the surface. Indeed, latest damaging earthquakes were generated by blind or “hidden” faults: 23 Oct. 2011, Van
earthquake (Mw 7.1, Turkey); 3 Sep 2010, Darfield earthquake (Mw 7.1, New Zealand); 12 January 2010 Haiti
earthquake (Mw 7.0); 6 April 2009 L’Aquila earthquake (Mw 6.3, Italy). Therefore understand how a fault grows
and develops is a key question to evaluate the seismogenic potential of an area. Analogue model was used to
understand kinematics and geometry of the geological structures since the beginning of the modern geology. On
the other hand, numerical model develops much more during the last thirty years. Nowadays we can use these two
methods working together providing mutual interactions. In the two-three most recent years we tried to use both
numerical and analogue models to investigate the long-term and short-term evolution of a blind normal fault.
To do this we improved the Analogue Model Laboratory of the University of Pavia with a laser scanner, a stepper
motor and other high resolution tools in order to detect the distribution of the deformation mainly induced by
blind faults. The goal of this kind of approach is to mimic the effects of the faults movements in a scaled model.
We selected two seismogenic source cases: the causative fault of the 1908 Messina earthquake (Mw 7.1) and that
of the 2009 L’Aquila earthquake (Mw 6.3). In the first case we investigate the long term evolution of this structure
using a set of analogue models and afterwards a numerical model of our sandbox allow us to investigate stress
and strain partitioning. In the second case we performed only an analogue model of short-term evolution of the
L’Aquila seismogenic source comparing our result with pre-existing numerical models. In both cases we obtain
mutual advantages using together experimental results. We believe that the analogue modelling approach coupled
with numerical modelling applied to the study of active faults can provide useful insights to investigate the seismic
potential of a structure with important appliances also for the seismic risk assessment.
