GNGTS 2018 - 37° Convegno Nazionale

564 GNGTS 2018 S essione 3.1 The iterative working flow for the horizon-based tomography approach can be summarized into these main steps: 1. Build initial velocity-depth model. Therefore, we interpreted the main horizons on the MS29 pre-stack time migrated section. We picked the water-bottom and 5 horizons (Water layer, Quaternary base, 2 strong reflectors of the Plio-Pleistocene age, Messinian unconformity, Carbonate platform), subdividing the section into 7 formations. It is convenient to pick significant horizons, that are those with major changes in velocity (principal reflectors). 2. Perform the demigration of the picked horizon using the RMS velocity field. As a matter of fact, coherency inversion (Landa et al. , 1988) requires time horizons picked. This kind of inversion estimates the layer velocity using the ray tracing for a set of trial velocities. We used the velocity model obtained from the grid tomography as set of trial velocities. 3. Pick the highest semblance value velocity on the CMP locations along the picked horizon. 4. Smooth this velocity to avoid unrealistic oscillations caused by noisy data. The coherency inversion method honors the ray bending at layer boundaries, considering a non-hyperbolic CMP traveltime so, it is more realistic (Yilmaz, 2001) (Fig.2). Finally, we have performed a first PSDM with the initial velocity-depth model. Thereafter, to check the consistency of our velocity-depth model, we analyzed the CIG by a RMO analysis, which represent the input (the measure of non-flatness of the CIG) to the tomography. After tomography we newly migrated our data. We performed this process as iterative way for three times to minimize the errors and to get best as possible velocity-depth model. Conclusions. The MS29 seismic line investigates the Otranto Channel along its NW-SE axis. It intersects the Apulia carbonate platform and the two sedimentary basins: Ionian/South Adriatic Basin and South Apulia basin, respectively located at NW and at SE. To have a real geometry of the geological structures we proceeded to a PSDM, also using the well data present in the study area, achieving a good result allowing to better identify the following horizons: Quaternary base, Middle Pliocene reflector, top of the Messinian layer, top of the Apulia carbonate platform and base of pelagic sediment basins. A preliminary result obtained by grid tomography showed a velocity-depth model with high frequencies lateral variations that affect the migrated section, producing unrealistic oscillations of the horizons. This effect is probably due to the noise present on the data, which affect the compute of the structural attributes and the successive tomographic inversion. Nevertheless, grid tomography flattens the CIG’s. On the other hand, the layer-by-layer coherence inversion lead to a starting velocity-depth model that produce a realistic depth migrated section. Finally, the model-base tomography improves the velocity field with a low RMO error on CIG’s. The Fig. 3 - Part of the pre-stack depth migrated MS29 seismic line. In detail it is shown the carbonate platform with reef, the adjacent sedimentary basin and the topping sedimentary sequence. (WB=Water Bottom; QUAT=Quaternary Base; MID PLIO=Middle Pliocene; BS=Bright Spot MES=Messinian Layer; CARB=Carbonate Platform).