GNGTS 2014 - Atti del 33° Convegno Nazionale

GNGTS 2014 S essione 1.1 57 of the total Adria-Eurasia convergence rate is partitioned and absorbed to the north of the thrust belt, between the south-eastern Alps domain and the Eurasia plate. In this sense, we consider the Eastern Alps as an independent continental sliver partitioning motion between Adria and Eurasia. Modelling approach and geometry of the thrust front. We invert simultaneously the observed interseismic GPS velocities for coupling distribution on the thrust fault and for the rotation pole of a rigid eastern Alps sliver, using elastic dislocation in a half-space (Okada, 1985) and the back-slip approach of Savage (1983). To define the geometry of the model fault plane, we considered all available information on crustal structure, instrumental seismicity distribution, nodal planes of seismic moment tensors, proposed seismogenic sources, and high-resolution crustal images of the lithospheric structure of the eastern Alps. We suggest that all these observations concur to consider the surface trace of the seismogenetic structures outcropping at the front of the southern Alps, as the shallow expression of the crustal-scale thrust fault imaged by TRANSALP experiment, justifying the use of a single continuous structure outcropping at the margin of the mountain front. The adopted fault geometry (Fig. 2) is only representative of the southern Alps thrust front and is most likely incorrect west of the Schio-Vicenza line and east of the 1976 Friuli sequence epicentral regions, where the tectonic elements rapidly rotate to assume NW-SE Dinaric orientation. For this reason the calculation of seismic moment accumulation will be limited to this area. We use triangular dislocation elements (Gimbutas et al. , 2012) tessellating the fault plane to account for the arcuate shape of the mountain front and impose smoothness constraints on the slip distribution, solving the system of equations using a bounded-variable least squares algorithm (Stark and Parker, 1995) to impose slip rate bounds. Results. The results of the inversion of GPS velocities show a relatively homogeneous pattern of interseismic coupling across the south-eastern Alps (Fig. 2). On average, the fault system appears fully locked from where it emerges at the surface along the foothills to beneath the front of the range to approximately 15 km to the north (corresponding to a depth of ~10 km). Our best-fitting model yields an Eulerian pole between the Venetian-Friulian plane and the Alps located at 44.60°N, 5.69°E with a rotation rate of -1.27°/Ma, predicting any systematic residuals and supporting our hypothesis pointing out for a localized deformation at the margin of the mountain front and negligible deformation in the interiors of the Alps. A shortening rate of 1.0-1.5 mm/yr is predicted by the rigid rotation between eastern Alps and the Venetian- Friulian plain. The interseismic coupling distribution (Fig. 2) shows a concentration of ~100% of coupling in the Friuli region, in the area of the Bosco del Cansiglio event and across the Montello thrust. In the Carnia area, near Maniago and Arba Ragogna thrusts and in the western edge of the modelled thrust (across the Thiene-Bassano and Bassano-Cornuda thrusts) the coupling is moderate (~50%). Resolution and forward tests show the ability of our dataset to resolve along- strike variation of the coupling distribution on the modelled fault plane over a distance ~20 km. However, it is possible that some amount of coupling could be not well estimated where the GPS velocities still present significant uncertainties. The comparison between background seismicity and estimated interseismic coupling (Fig. 2) suggests that the seismicity correlates to the down-dip end of the locked fault zone. Thus, the relationship between geodetic strain and current seismicity suggests that small seismic events are controlled by stress build-up at the base of the locked fault where the deformation can be accommodated either by sliding on a plane or by diffusion/dislocation creep. Cut-off of seismicity, likely controlled by the transition between rate-weakening (unstable slip) and rate- strenghtening (stable creep) frictional sliding mechanisms at ~350-400°C for crustal rocks, occurs at a depth of ~15-18 km, in agreement with available information about the geothermal regime in the area (Viganó et al. , 2012). This is analogous to that observed across the Himalaya of Nepal (Cattin and Avouac, 2000) or across the Central Range of Taiwan (Dominguez et al. ,