GNGTS 2013 - Atti del 32° Convegno Nazionale

checked for consistency using the reciprocity rules of Ackermann et al. (1986) and then inverted using a commercial software, that is part of the SeisImager ®package, distributed by Oyo Corporation Ltd. and based on the work of Hayashi and Takahashi (2001). The tomographic models, not only provide important information for interpretation of the seismic profiles, but are also used in the reflection processing sequence, integrating the semblance-based velocity models used for the normal moveout correction. First break picks were also used to apply refraction static corrections (Taner et al. , 1998). The preliminary velocity models were obtained by picking the maximum of semblance (Neidell and Taner, 1971). The preliminary velocity models and the resulting stacks were refined by two cycles of residual static corrections and velocity analysis (Clerabout and Ronen, 1985). The final stack sections were processed with a spatial filtering algorithm, and depth migrated using a Kirchhoff post-stack depth migration. Discussion and conclusions. The data acquired in Laas (BZ) are characterized by good signal-to-noise ratio. The use of a “dense wide aperture” geometry, allowed us to sample reflected and refracted phases in a wide offset range, therefore reaching a comparable depth penetration for both reflected and refracted phases. Combination of seismic reflection and first- arrival tomography provided overlapping and matching information in terms of reflectivity models and P-wave tomographic velocity models along our profiles, and allowing us to imaging from the topographic surface down to the top of the basement. All tomographic models show heterogeneous P wave velocities and strong gradients: P-wave velocities vary from a minimum of about 0.5 km/s at the surface and increase up to over 4.5 km/s at top of the metamorphic bedrock. The comparison between tomographic models and depthmigrated sections, emphasize a perfect agreement between the trend and the geometry of seismic-stratigraphic units on the migrated sections with the P-wave velocity models from the first-arrival tomography. The migrated profile “Lasa_1” shows a very shallow reflector with high amplitude at an average depth of 150-200 m (Fig. 2). Its amplitude characteristics and the high stacking velocity associated with this reflector (V RMS > 3500 m/s), suggest that this event is generated by the reflection from the top of metamorphic bedrock. This interpretation is supported by the agreement between the reflection event and the high velocity (Vp > 4 km/s) of seismic tomography. The migrated profiles of Lasa2_3 (Fig. 2) clearly image the structure of the valley and the trend of the top of the metamorphic bedrock below. The profiles also show an a sub-horizontal Fig. 2 – Depth migrated sections of profiles Lasa_1 and Lasa_2_3 with respective tomographic models. 54 GNGTS 2013 S essione 3.1