Hidden Megathrust Beneath Italy’s Abruzzi Coast Revealed by Seismic Tomography

Beneath the quiet hills and Adriatic coastline of Abruzzi in central-southern Italy, geologists have long suspected that a major compressive structure lurks out of sight. Now, for the first time, an international team has produced seismic images that appear to show it directly. Using a technique called local earthquake tomography, researchers led by Rita de Nardis and Donato Talone of the University G. d’Annunzio in Chieti, together with colleagues in Bologna and Trieste, have mapped a buried thrust fault that reaches depths of roughly 24 kilometers, revealing what they interpret as a doubling of the mid-crust across the Abruzzi Arc basal thrust. The findings, published in the journal Solid Earth, could reshape how scientists assess earthquake hazard in one of Italy’s most enigmatic tectonic settings.

The region in question belongs to the Outer Thrust System of Italy, a first-order arcuate fold-and-thrust belt that stretches roughly 2,500 kilometers from northern Italy to Sicily. It formed during the Late Pliocene and Quaternary as the Apennine-Maghrebian belt propagated eastward into the Adriatic foreland. Along its length, the system features two great outward-convex arcs: the Padan-Adriatic arc in the north and the Ionian-Sicilian arc in the south. The northern arc can be subdivided into smaller third-order arcs, including Monferrato, Emilia, Ferrara, Adriatic, and Abruzzi. While the northern arcs have produced damaging earthquakes in recent decades, the Abruzzi segment has remained strikingly silent, and its deep geometry has been far less understood than that of its neighbors.

That silence is precisely what makes the region scientifically troublesome. Slowly deforming zones present a fundamental dilemma for seismologists: faults that load at low rates may either be active, releasing earthquakes only at long recurrence intervals, or genuinely inactive. GPS measurements indicate that convergence across the Italian Outer Thrust System proceeds at just 1 to 3 millimeters per year, and historic and instrumental earthquakes in the contractional province rarely exceed magnitude 6.0. Yet the record shows that silent thrusts can surprise. The 2012 Emilia earthquake, the 1968 Belice earthquake in Sicily, and the 2023 magnitude 6.8 event in Morocco’s High Atlas all struck areas previously considered to have low seismicity, rupturing faults that had shown no instrumental activity.

To peer into the crust, the team inverted an enormous dataset of seismic travel times: 42,176 P-wave and 29,045 S-wave arrival times from 5,712 earthquakes with magnitudes between 0.2 and 5.5, recorded by 37 stations of the Italian National Seismic Network between January 2009 and December 2020. The earthquakes were carefully filtered and relocated using three one-dimensional velocity models proposed for the area, with the best-constrained catalogue selected as the starting point. The tomography itself was performed with the Fast-Marching Tomography algorithm FMTOMO, which solves the eikonal equation on a regular grid to compute first-arrival travel times through heterogeneous media, then reconstructs ray paths and iteratively minimizes travel-time residuals to update the velocity model.

Resolution was a central concern. The researchers optimized the grid spacing, using about 5 kilometers horizontally and 2 to 3 kilometers vertically, and determined optimal damping and smoothing parameters from classical trade-off curves. They then verified the reliability of the resulting models with synthetic checkerboard and spike tests, which showed that the smallest resolvable anomaly is roughly 15 kilometers across, with locally finer resolution of about 10 kilometers at shallow depths. The final models achieved reductions in root-mean-square misfit and covariance of roughly 73 and 93 percent for P-wave velocity, and 65 and 88 percent for S-wave velocity, providing confidence in the features interpreted down to about 20 and locally 24 kilometers depth.

The velocity models reveal a layered picture of the crust. In the west, within the Apennine extensional domain, scattered low-velocity anomalies in the upper 8 kilometers correlate well with Quaternary intra-mountain basins such as the Fucino and Sulmona basins. East of the extensional domain, at the hanging wall of the buried thrust, five low-velocity zones between 2 and 8 kilometers depth correspond to coastal and fluvial deposits and to Miocene-to-Pleistocene sandstone and clay units. Two of these anomalies coincide with a large positive magnetic anomaly whose origin is debated; because sedimentary rocks are slower than magmatic ones, the team favors a deeper magnetic source within the basement rather than volcanic rocks trapped in the sedimentary cover.

The most significant discovery, however, lies deeper. Between roughly 14 and 24 kilometers depth, in an area bounded by latitudes 41.3 to 41.8 degrees north and longitudes 14.3 to 15.0 degrees east, the tomography images a broad velocity inversion: a lower-velocity layer of 6.0 to 6.6 kilometers per second beneath a higher-velocity layer of 6.6 to 7.0 kilometers per second. This configuration is consistent with a mid-crustal overthrust in which a stack of crystalline and Mesozoic units, including dolomitic lithologies, overrides a lower-velocity footwall likely composed of Triassic evaporites and Verrucano formations. A comparable high-velocity body had been documented slightly further north by earlier work, but without the depth resolution to resolve the underlying low-velocity structure that the new study now images.

Integrating the tomographic results with geological maps, deep wells, seismic reflection profiles, and newly computed focal mechanisms, the team built a three-dimensional conceptual model of the Abruzzi Arc basal thrust. The reconstructed fault surface extends approximately 170 kilometers along strike, forming a broad eastward-convex arc with an average dip of about 22 degrees and reaching depths of up to 24 kilometers. Along strike it is segmented into three fourth-order arcs named Abruzzo Citeriore, Frentani, and Daunia. The deep tomography-derived fault patch extends about 100 kilometers and dips southwest at roughly 15 degrees. The geometry closely matches that of the Ferrara and Emilia arcs to the north, supporting the idea that the Abruzzi thrust may be the southernmost element of a larger crust-scale frontal thrust, and confirming a thick-skinned deformation style in which the crystalline basement itself is involved.

The model also clarifies how three tectonic provinces coexist in the region. To the west, east- and west-dipping normal faults of the Apennine extensional province host most of the region’s seismicity at shallow depths. To the east, right-lateral strike-slip faults of the Adriatic foreland, including the sources of the 2002 San Giuliano and 2018 Montecilfone earthquakes, remain confined to depths of 10 to 25 kilometers. Their upward propagation appears to be inhibited by the low-angle basal thrust, which acts as a mechanical barrier, a configuration recognized in several other sectors of the Italian thrust belt. New focal mechanisms computed from a temporary seismic network reveal modest compressional activity in 2009 and 2018 at depths of 8 to 18 kilometers, associated with a back-thrust splaying from the main structure.

Whether the Abruzzi Arc basal thrust is capable of generating large earthquakes remains the crucial open question. Some destructive historical events, including the 1706 and 1933 Maiella earthquakes and the catastrophic 1456 sequence, have epicenters that could align with the thrust or with adjacent domains, and interpretations differ. Morphotectonic studies document ongoing uplift and shortening along parts of the coastal sector since at least the Middle Pleistocene. The authors argue that slow deformation rates and long recurrence intervals do not rule out future activity, and that if the thrust behaves like its northern counterparts, it could release moderate to strong earthquakes. The new three-dimensional model offers a foundation for a next generation of seismic hazard models for the central-southern Apennine transition zone, where a seemingly quiet fault may be quietly loading toward its next rupture.

Subject of Research: Seismic tomographic imaging of mid-crustal doubling and the buried Abruzzi Arc basal thrust in central-southern Italy

Article Title: First tomographic imaging of mid-crustal doubling at the Abruzzi outer thrust front, central-southern Italy

Article References: First tomographic imaging of mid-crustal doubling at the Abruzzi outer thrust front, central-southern Italy. (n.d.). https://doi.org/10.5194/se-17-665-2026

Image Credits: AI Generated

DOI: 10.5194/se-17-665-2026

Keywords: seismic tomography, Abruzzi Arc basal thrust, Apennines, crustal doubling, thrust tectonics, seismic hazard, Italy, FMTOMO, focal mechanisms, thick-skinned tectonics, slowly deforming zones, Adriatic foreland