GNGTS 2014 - Atti del 33° Convegno Nazionale

GNGTS 2014 S essione 3.1 23 phase of calderization (Barberi et al. , 1978). In the eastern sector, the sedimentary processes related to the Sarno-Sebeto coastal plain controlled the deposition of marine and coastal seismic units during Late Pleistocene and Holocene, often interlayered with volcanic deposits related to Somma-Vesuvius (Milia et al. , 1998; Aiello et al. , 2001). The Capri island represents a key sector for the comprehension of the paleogeographic setting of the southern Apennines during the Mesozoic and the Cenozoic (Barattolo and Pugliese, 1987). In fact, it represents one of the few areas where the inner margin of the Apenninic platform is preserved (Mostardini and Merlini, 1986). The recognized and distinguished units are numerous and include a wide stratigraphic interval ranging between the Early Jurassic and the Miocene. Due to the complex tectonic setting of the island, determined from the Oligo-Miocene Apenninic deformation e then from the Plio-Quaternary neotectonics, the vertical and horizontal relationships among the different units are often very complex, not preserved or unaltered. In particular, this happens in the eastern sector of the island, where the geologic situation is complicated by an intense urbanization and by the wide occurrence of Quaternary deposits. The main formations are represented by the Cala Ventroso dolomites (Lias), the Grotta Delle Felci bioclastic limestones (Middle-Upper Lias),the Torina oolitic limestones (Aalenian-Bathonian), the Migliara cherty limestones (Middle Lias-Kimmeridgian), the Ellipsactinia limestones (Callovian-Valanginian), the Limmo white limestones (Aptian), the Tiberio rudites (Aptian-Turonian p.p.), the Scaglia di Punta Carena (Upper Turonian-Coniacian), the Faro conglomerates (Maastrichtian). Deep seismic sections on the Southern Tyrrhenian continental margin and in the Tyrrhenian bathyal plain have been collected. Three regional seismic profiles have been processed and interpreted, for an overall length of 160 km. The acquisition parameters are represented by the type of seismic source (N.2 Airguns, G/I gun SI/Sodera), by the record length (5 s, TWT), by the shot interval (25 m) and by the hydrophones interval (12.5 m). The software used for the seismic processing are the “Promax2D” (Landmark Ltd.) and the “Seismic Unix” (Colorado School of Mines). Some advanced processes have been applied to a basic flux of elaboration to improve the useful signal occurring in the seismic data. The seismic data have been promoted to produce stacked sections, which have been interpreted. The Fourier analysis has been carried out on the seismic traces to identify the frequency content of the signal in several seismograms after the application of a band-pass filter. This enabled to recognize the frequency interval in which the useful signal was concentrated. The gain application consisted of the compensation of the delay of the signal due to absorption, scattering and decay of the amplitude. A large amount of the lost signal has been restored in order to obtain levels of amplitude similar throughout all the seismic data. The corresponding process is the Automatic Gain Control (AGC). The first stacked seismic sections have been produced through the velocity analysis on sorted seismic traces in CMP-gathers. The Common MID Points Gathers (CDP) represent the in-phase sum of the seismic traces coming from the same CDP. The same process has been repeated after applying different processes of elaboration to the seismic data, trying to understand if the same processes have produced or not significant improvements of the seismic signal. The reflection hyperbola alignment obtained using different velocity values have been calculated. The NMO correction has been obtained using previously estimated correct velocity values. The signal/noise ratio was increased through the stacking, reducing the casual noise included in the data. During the stacking, the coherent signal has increased its width by constructive interference of a factor equivalent to the coverage of the data; on the other side, the casual signal has been added to another noise by slightly increasing its width. The applied filter resulted to be enough conservative (0-20-50-70 Hz) allowing to eliminate the occurring high-frequency noise. A frequency spectrum before and after the band-pass filter has also been constructed. Theprocedureofmultipleattenuationconsistedof bothstackingandpredictivedeconvolution. The move-out between the primary reflections and the multiple ones was discriminated through