3D tomographic imaging of the southern Apennines (Italy): a statistical approach to estimate the model uncertainty and resolution

A new dataset of first P-wave arrival times is used to derive the 3D tomographic model of the Campania-Lucania region in the southern Apennines (Italy). We address the issue related to the non-uniqueness of the tomographic inversion solution through massive numerical experimentation based on the global exploration of the model parameter space starting from a large variety of physically plausible initial models. The average of all the realizations is adopted as the best-fit solution and the uncertainty of the model parameters is studied using a statistical approach based on a Monte Carlo-type analysis. How the uncertainty in the initial model, earthquake locations, and data influences the inversion result is studied by considering separately the individual effects. Checkerboard tests are performed to estimate the resolving power of the dataset. Re-located seismicity in a reliable new 3D tomographic model allows us to correlate the earthquake distribution with the main seismogenic structures present in the area.