Accurate assessment of 3D crustal velocity and density parameters: Application to Vesuvius data sets

Abstract The three-dimensional P-wave velocity and density structure of Mt. Vesuvius volcano has been revealed by tomographic inversion of seismic and gravity data. We apply the tomographic inversion method of sequential integrated inversion, now extended to 3D structures, to the first arrival travel time data collected during 1994 and 1996 TOMOVES experiment and to a new gravity map of the Campanian area. The approach, now significantly improved, allows us to reconstruct both velocity and density structures, while reducing traveltimes and gravity residuals. Inversion of traveltimes, modelled through a ray-tracing technique, is performed using an iterative algorithm which uses the second order Sobolev norm as the regularization term of the ill conditioned seismic inverse problem. This speeds up the optimization problem and allows us to obtain stable solutions with a weak dependence on the starting model and on the chosen discretization. In order to improve the match between seismic and gravity modelling and to avoid artificial density contrasts across the interfaces, the density model is parametrized with polyhedral bodies whose density is linearly dependent on the three coordinates. This study brings insight into the main velocity and density structural units of Mt. Vesuvius, which include a C shaped negative anomaly surrounding the flanks of the crater up to depths of 6 km. This anomaly is bordered in the south by a strip of high density gradients which run parallel to the Sorrento peninsula, and in the east by an intruding Mesozoic carbonate basement structure. We observe evidence for the thinning of the hard dolomites in the north-western and south-eastern part of the model, related to previously observed NW–SE and SW–NE collapse faults within the Plio-Pleistocene graben of the Campanian plain. Beneath the volcano edifice, we find a positive and thick intrusion anomaly which we compare to the results of previous studies. The reliability of the reconstructed models is quantified through a restoring test and the estimation of traveltimes and gravity residuals.