14 April 2024 in Archaeological Prospection
We performed magnetic and GPR measurements to image the buried ruins of the Middle Age abbey San Pietro in Crapolla, on the Sorrento‐Amalfi Coast (Massa Lubrense, Southern Italy). The site represents an important religious location, which is nowadays partially buried along the cliff. An integrated study was necessary to map the buried structures and address the archaeological excavation. For this reason, we carried out the surveys on two main grids in order to reconstruct the structures of the abbey and of its related church. The magnetic data were filtered through the discrete wavelet transform (DWT) and then transformed to total gradient maps. The obtained maps were interpreted with depth from extreme points (DEXP) imaging method to assess the horizontal and depth positions of the top. The GPR data were processed and time‐depth converted. Results from the integrated interpretation of these data suggest the possible presence of different vaulted rooms and an elongated structure at 0.3‐m depth from ground surface. This latter is interpretable in terms of perimetral and internal walls of the abbey and its church. These outcomes were crucial to successfully address archaeological excavations, which targeted one of the modelled areas and unearthed a wall at the predicted depths.
6 December 2023 in IEEE Transactions on Geoscience and Remote Sensing
We have established a workflow for a multiorder sequential joint inversion (MOSJI) of gravity and gravity gradients, that aims at modeling vertically stacked sources in various geological scenarios. We consider the joint inversion of the gravity data and one of the $h$ th-order derivatives of the gravity data. The first step involves separate inversions, which are fundamental to fully exploit the different wavelength-content of the two quantities to invert. The joint inversion is warranted by using the scheme of a sequential joint inversion with a cross-gradient constraint. The algorithm is able to exploit different types of a priori information, such as compactness and inhomogeneous model-weighting function. First, we test this approach on a realistic synthetic model from the SEg Advanced Modeling (SEAM) Phase I model, involving salt and mother salt structures. Then, we consider a synthetic model containing either shallower or deeper karst cavities. These tests produced a better modeling of both shallower and deeper sources, when compared to the separate unconstrained inversions. Thanks to these good results, we apply our method to a real case for cavity detection in Southern Spain. The method shows an accurate modeling of the expected sources. In all the aforementioned tests, we obtain a strong decrease of the cross-gradient values and a meaningful linearization in the scatter plots of physical parameters, both indicating the good performance of the joint inversion.
23 February 2024 in Acta Geophysica
We analyzed magnetic data of a deep-water area in the Eastern Mediterranean, offshore Egypt where there are numerous salt domes. Regional-residual separation of the reduced to pole (RTP) magnetic anomalies was performed with the discrete wavelet transform, and the residual magnetic anomalies were interpreted as mainly due to the susceptibility contrast between sediments and salt bodies. A multiscale boundary analysis of the anomalies yielded meaningful lineaments at both large and short scales. Salt bodies lineaments are very well defined in the short-scale map. Then, we inverted the data by a nonlinear and non-iterative 3D inversion technique, to delineate the salt bodies and estimate their magnetization contrast. We used the depths obtained from the Euler deconvolution to form a set of maxima and minima constraints in the inversion. Our procedure yielded an interesting map of the top of salt in the deep-water region, entirely based on the magnetic data, which agrees well with the seismic top of salt. This is a somewhat surprising result, in some respects not obvious, demonstrating that magnetic prospecting can be an important and low-cost tool for the exploration of salt diapirs, especially when seismic data are partially or completely inaccessible or of poor quality.
19 August 2025 in IEEE Transactions on Geoscience and Remote Sensing
We describe a mixed Machine Learning (ML)-Inversion algorithm for sedimentary basin modelling of gravity data. We consider two typical targets of basin modelling, i.e., the basement depth and the diapirs, both being of fundamental importance for energy exploration or tectonic studies. First, Machine Learning is implemented to interpret gravity anomalies along a set of profiles crossing the whole anomaly. Then, the 2D interpreted models are merged into a reference model to constrain the final 3D inversion of the whole set of data. The learning of our Convolutional Neural Network is based on the simplest gravity sources, i.e., faults. Such choice transfers a strong a priori information to the interpretative problem, since gravimetric anomalies due to diapirs, or to the morphology of the basement, can be seen as composed by the constructive interference of anomalies generated by the edges of the sources. We considered up to 221.100 different fault-models with variable parameters (depth, density, thickness and depth-to-top), corresponding to low/ high dip-angle faults and mixed faults. Our workflow was shown suitable for reconstructing different structures in either synthetic or real-data cases: a diapir-shaped source and the basement of a huge sedimentary basin, the Caltanissetta basin (Italy), respectively. In both cases it is shown that our mixed approach produces significant results thanks to the self-constraints extracted during the Machine Learning step.