Volume 39, Issue 12 e2020TC006296
Research Article
Free Access

Disproving the Presence of Paleozoic-Triassic Metamorphic Rocks on the Island of Zannone (Central Italy): Implications for the Early Stages of the Tyrrhenian-Apennines Tectonic Evolution

Manuel Curzi

Corresponding Author

Manuel Curzi

Dipartimento di Scienze della Terra, Sapienza Università di Roma, Rome, Italy

Correspondence to:

M. Curzi,

[email protected]

Search for more papers by this author
Andrea Billi

Andrea Billi

Consiglio Nazionale delle Ricerche, IGAG, Rome, Italy

Search for more papers by this author
Eugenio Carminati

Eugenio Carminati

Dipartimento di Scienze della Terra, Sapienza Università di Roma, Rome, Italy

Consiglio Nazionale delle Ricerche, IGAG, Rome, Italy

Search for more papers by this author
Federico Rossetti

Federico Rossetti

Dipartimento di Scienze, Università Roma Tre, Rome, Italy

Search for more papers by this author
Richard Albert

Richard Albert

Institut für Geowissenschaften, Goethe University Frankfurt, Frankfurt am Main, Germany

Frankfurt Isotope and Element Research Center (FIERCE), Goethe-University Frankfurt, Frankfurt, Germany

Search for more papers by this author
Luca Aldega

Luca Aldega

Dipartimento di Scienze della Terra, Sapienza Università di Roma, Rome, Italy

Search for more papers by this author
Giovanni Luca Cardello

Giovanni Luca Cardello

Dipartimento di Scienze della Terra, Sapienza Università di Roma, Rome, Italy

Search for more papers by this author
Alessia Conti

Alessia Conti

Dipartimento di Scienze della Terra, Sapienza Università di Roma, Rome, Italy

Search for more papers by this author
Axel Gerdes

Axel Gerdes

Institut für Geowissenschaften, Goethe University Frankfurt, Frankfurt am Main, Germany

Frankfurt Isotope and Element Research Center (FIERCE), Goethe-University Frankfurt, Frankfurt, Germany

Search for more papers by this author
Luca Smeraglia

Luca Smeraglia

Consiglio Nazionale delle Ricerche, IGAG, Rome, Italy

Search for more papers by this author
Roelant Van der Lelij

Roelant Van der Lelij

Geological Survey of Norway, Trondheim, Norway

Search for more papers by this author
Gianluca Vignaroli

Gianluca Vignaroli

Dipartimento di Scienze Biologiche, Geologiche ed Ambientali–BiGeA, Università degli Studi di Bologna, Bologna, Italy

Search for more papers by this author
Giulio Viola

Giulio Viola

Dipartimento di Scienze Biologiche, Geologiche ed Ambientali–BiGeA, Università degli Studi di Bologna, Bologna, Italy

Search for more papers by this author
First published: 23 October 2020
Citations: 14

Abstract

The inner Apennines (Italy) are characterized by scattered outcrops of continent-derived orogenic metamorphic units exposed along the Tyrrhenian coast from northern to southern Apennines. At least since the 1970s, some peculiar rocks exposed on Zannone Island (central Italy) have been described as the only Paleozoic-Triassic metamorphic complex linking those exposed in the northern—with those in the southern Apennines. Assessing the protolith nature, thermobaric conditions, and structural features of what is accepted to be the metamorphic unit of Zannone is, therefore, crucial to elucidate the early paleotectonic evolution of the Apennines-Tyrrhenian orogenic system. To that end, we interpreted seismic reflection profiles offshore Zannone, we carried out mesoscale and microscale structural investigations of representative outcrops on the island, performed X-ray diffraction analysis, and K-Ar and U-Pb geochronology of representative clay gouge and syntectonic carbonate veins. Results show that the metamorphic rocks of Zannone can actually be reinterpreted as belonging to nonmetamorphic siliciclastic turbidites, likely deposited in foredeep settings, and coeval to the Oligocene-Miocene Macigno Fm. of the northern Apennines. The turbiditic sequence was overthrust by Triassic dolostone in the early Miocene (~22 Ma), weakly deformed at <200°C, and downfaulted by postorogenic extensional faults starting ~7 Ma ago. Hence, Zannone represents a unique geological location in central Italy where to study the innermost (i.e. oldest) thrust sheet of the central-northern Apennines, thereby shedding new light onto the early tectonics of the Apennines. Based on this evidence, a new evolutionary scheme for the early stage of the Apennines tectonic evolution is proposed.

Key Points

  • Discovery of a hitherto unknown early thrust in the Apennines
  • New radiometric dating of Apennine tectonic phases
  • Reinterpreting and overturning of a long-held assumption in the Apennines

1 Introduction

The early orogenic history of the Apennine fold-and-thrust belt in Italy is poorly known as the innermost (westernmost) portion of the chain was dismantled, downfaulted, and overprinted by the opening of the Tyrrhenian back-arc basin, which developed since Middle Miocene time at the rear of the eastward migrating compressional front of the orogenic wedge (e.g., Carmignani et al., 2004; Carminati et al., 2010; Faccenna et al., 1997; Malinverno & Ryan, 1986). The rare and scattered metamorphic rocks along the western inner margin of the belt have provided essential information to unravel part of the early stages of the Apennine tectonic building (e.g., Faccenna et al., 2001; Jolivet et al., 1998, 2003; Papeschi et al., 2020; Rossetti et al., 1999; Vezzoni et al., 2018; Vignaroli et al., 2009). The metamorphic complexes of the Apennine belt consist of three main lithotectonic systems: (i) European-derived continental units (e.g., Beaudoin et al., 2017; Di Rosa et al., 2020; Di Vincenzo et al., 2016; Jolivet et al., 2003; Molli & Malavieille, 2011; Rossetti et al., 2015; Vitale Brovarone et al., 2013; Figure 1); (ii) ocean-derived (Tethys) units (e.g., Carmignani et al., 1994; Rossetti, Faccenna, Goffé, et al., 2001; Rossetti et al., 2004; Vai, 2001; Vignaroli et al., 2009; Vitale Brovarone et al., 2013; Figure 1), and (iii) Adria-derived continental units (e.g., Jolivet et al., 1998; Molli et al., 2018; Rossetti et al., 1999). The Adria-derived metamorphic complexes, consisting of metamorphic Permian-Triassic siliciclastic deposits and minor Paleozoic basement rocks, are poorly exposed in northwestern Tuscany, from the Apuan Alps to the Monticiano-Roccastrada Ridge and Romani Mounts (e.g., Aldinucci et al., 2008; Brogi & Giorgetti, 2012; Jolivet et al., 1998; Theye et al., 1997; Vai, 2001; Vignaroli et al., 2009; Figure 1), and in the Calabria-Peloritani Arc of southern Italy (e.g., Rossetti, Faccenna, Goffé, et al., 2001; Vai, 2001; Vignaroli et al., 2008; Figure 1). Over a distance of about 500 km between Tuscany and Calabria, however, no metamorphic rocks are known to crop out in the Apennine belt (e.g., Bigi et al., 1990; Di Filippo & Toro, 1980; Vignaroli et al., 2009; Figure 1), with the exception of metamorphic rocks dredged from the Flavio Gioia seamount in the Tyrrhenian back-arc domain (e.g., Vai, 2001; Figure 1). A second remarkable exception, now known for over a century, are Adria-derived, low-grade metamorphic rocks of Paleozoic-Triassic age on the Island of Zannone (hereafter referred to as ZI) in tectonic contact with an overlying series of Triassic dolostone (Accordi & Carbone, 1988; Bieber, 1924; De Rita et al., 1986; Di Sabatino, 1979; Doetler, 1875; Galdieri, 1905; Pantosti & Velonà, 1986; Parotto & Praturlon, 1975; Sabatini, 1898; Segre, 1954; Figures 1 and 2c).

Details are in the caption following the image
Simplified structural-geological map of Italy showing a synthesis of the P-T paths reconstructed for the main metamorphic complexes in the Apennine belt (redrawn and modified from Vignaroli et al., 2009). Age in red: Adria-derived continental metamorphic complexes. Age in black: Europa-derived continental metamorphic complexes. Age in blue: Liguro-Piemontese (oceanic)-derived metamorphic complexes. The faults in this map are from the Structural map of Italy (Consiglio Nazionale delle Ricerche, 1992).
Details are in the caption following the image
(a) Simplified geological map of central-southern Apennines (redrawn and modified from Smeraglia et al., 2019). (b) Geological cross section (modified after Mostardini & Merlini, 1986). Cross-section trace is shown in (a). (c) Simplified geological map and stratigraphy of the Zannone Island showing the previously described Paleozoic crystalline basement (modified after De Rita et al., 1986 and Cosentino & Parotto, 1986).

As an equivalent Triassic dolostone was also found in the Mara 001 borehole offshore ZI (Figure 2a; ViDEPI database, https://www.videpi.com/videpi/videpi.asp), at depths between 2,115 and 2,735 m, the exposure of metamorphic rocks on ZI would require a large throw (~3 km) along normal faults to exhume the metamorphic rocks in their footwall. However, normal faults seen in several seismic lines on the continental shelf offshore ZI do not accommodate such displacements (e.g., Conti et al., 2017; Marani & Zitellini, 1986; Zitellini et al., 1984). Larger displacements (in the order of 2–3 km) are instead found close to the major fault system bounding the continental shelf to the east of ZI (Conti et al., 2017; Zitellini et al., 1984).

In the tectonic reconstructions of the Apennines-Tyrrhenian system, the occurrence of metamorphic rocks on ZI potentially establishes a link between the Tuscan and the Calabria-Peloritani Arc (Figure 1) metamorphic complexes (e.g., Vai, 2001). Since the most recent studies on ZI rocks date back to the 1980s, a reappraisal involving a modern multidisciplinary geological and analytical approach of such low-grade metamorphic rocks is needed to derive a better understanding of (1) their origin (i.e., their protolith), (2) the thermobarometric conditions that they experienced during deformation, and (3) the kinematics and age of the tectonic contact separating them from the overlying sedimentary cover.

This work aims at improving our understanding of the geological nature of the rocks exposed on ZI and their relationship with the overlying Triassic dolostone so as to frame their occurrence within the broader tectonic picture of the entire Apennine-Tyrrhenian geodynamic system. To this end, we performed field and laboratory studies, which include mesoscale-to-microscale structural investigations, X-ray diffraction (XRD) analysis, K-Ar dating of clay fault gouge, U-Pb dating of syntectonic carbonates, integrated with the interpretation of available seismic reflection profiles. Our main conclusion is that the rocks on ZI, which were previously ascribed to a Paleozoic-Triassic metamorphic basement, can instead be referred to foredeep turbidite deposits (likely of Oligocene-early Miocene age) that were (i) overthrust by a Triassic dolostone sequence in early Miocene time (at ca. 22 Ma) and (ii) deformed by postorogenic normal faulting during the late Miocene (at ca. 7 Ma). Correlation at the regional scale suggests that the contact between turbiditic rocks in the footwall and Triassic dolostone in the hanging wall represents the most internal known thrust sheet of the northern Apennines, attesting to the earliest stage of compressional deformation in the Apennine orogenic system. This result has major implications on the early evolutionary stages of the Apennine belt, which we explore in this paper. More generally, our findings show that the interpretation of low-grade metamorphic rocks, tectonically detached from the stratigraphic succession, as basement rocks bears significant consequences on the kinematic interpretation of fold and thrust belts (modes and timing of development and dismembering of orogenic wedges), on geological cross-section balancing, and on their retrodeformation (Dahlstrom, 1969; Woodward et al., 1989).

2 Geological Setting

The Apennine belt is the result of the Cenozoic convergence between Europe and Africa-Adria plates, which resulted in the formation of an active margin accommodating the W-directed subduction of a portion of the Liguro-Piemontese Ocean (Tethys) and of the Adria continental margin beneath the European plate (e.g. Carminati et al., 2010; Doglioni et al., 1998; Elter, 1975; Faccenna et al., 2001; Molli et al., 2010). The overall, first-order structure of the Apennine belt is the result of the superposition of an early phase of crustal shortening, followed and overprinted by Tyrrhenian back-arc extension (Figure 1), which was driven by the progressive eastward slab retreat (e.g., Faccenna et al., 1997, 2001; Rosenbaum & Lister, 2004). In the central-northern Apennines, shortening began during the late Oligocene (e.g., Boccaletti et al., 1990; Castellarin, 1992; Figure 1) and gave rise to a foreland-verging fold-and-thrust belt (e.g., Carminati et al., 2010; Faccenna et al., 2001; Malinverno & Ryan, 1986; Patacca et al., 1990; Figures 1, 2a, 2b, and 3). Postorogenic extension in the inner sectors of the Apennine belt (Tyrrhenian back-arc basin) took place during the middle-late Miocene and was associated with normal faulting, exhumation of deep-seated rocks, and magmatism (e.g., Cavinato & De Celles, 1999; Faccenna et al., 1997; Jolivet et al., 1998; Malinverno & Ryan, 1986; Sartori et al., 2004; Figure 1, 2a, and 2b).

Details are in the caption following the image
Simplified tectonostratigraphic columns showing the structural relationship between the different units of the central and northern Apennines (redrawn and modified from Jolivet et al., 1998).

The rocks involved in the Apennine building belong to different paleogeographic domains and consist of (1) Mesozoic-Cenozoic carbonate and marly carbonate deposited either in shallow or in deep marine environments along the Adria passive margin (e.g. Cosentino et al., 2010; Figures 1, 2a, 2b, and 3); (2) Adria-derived continental metamorphic complexes exposed in Tuscany, Calabria, and Sicily (Peloritani Mountains) and mainly composed of metamorphic Permo-Triassic siliciclastic sequences (Verrucano Group; Aldinucci et al., 2008; Perrone et al., 2006) and of minor Paleozoic sequences forming the metamorphic core of the Apennine belt (Brogi & Giorgetti, 2012; Carmignani et al., 1994; Jolivet et al., 1998, 2003; Molli et al., 2018; Rossetti et al., 1999; Vai, 2001; Vignaroli et al., 2008, 2009; Figures 1 and 3); (3) Europe-derived continental igneous and metamorphic complexes of Paleozoic age in Corsica, Sardinia, and the Calabria-Peloritani Arc (e.g., Jolivet et al., 2003; Rossi et al., 1994; Vai, 2001 Figures 1 and 3); and (4) ophiolite-bearing metamorphic and non-metamorphic complexes derived from a paleo-oceanic domain between the Adria and Europe continental blocks (Jurassic-Cretaceous Liguro-Piemontese ocean). Such complexes are exposed in Corsica, the Tuscan Archipelago, and Calabria (Ligurian and Subligurian Units; e.g., Carmignani et al., 1994; Jolivet et al., 1998, 2003; Molli & Malavieille, 2011; Rossetti, Faccenna, Jolivet, et al., 2001; Rossetti et al., 2004, 2015; Vai, 2001; Vignaroli et al., 2009; Vitale Brovarone et al., 2013; Vitale & Ciarcia, 2013; Figures 1 and 3). Moreover, synorogenic sequences consisting of siliciclastic sandstone and marl deposited in foredeep and wedge-top basins were progressively incorporated (and metamorphosed in the internal part of the belt) in the advancing fold-and-thrust belt (e.g., Brogi & Giorgetti, 2012; Castellarin et al., 1992; Cosentino et al., 2010; Figures 1, 2a, 2b, and 3).

The metamorphic complexes that crop out in the inner sector of the Apennine belt (Figure 1) were exhumed during synorogenic and postorogenic tectonics affecting the Apennine wedge (e.g., Aldega et al., 2011; Fellin et al., 2007; Jolivet et al., 1998, 2003; Molli, 2008; Molli et al., 2010; Rossetti et al., 1999, 2002, 2004, 2015; Vignaroli et al., 2009). The orogenic metamorphic complexes experienced different P-T conditions indicative of a progressive transition from subduction- to Barrovian-type metamorphism, evolving in space (from the internal portion of the belt toward east) and time (from the Eocene in Corsica and Calabria; Di Vincenzo et al., 2016; Maggi et al., 2012; Rossetti, Faccenna, Goffé, et al., 2001; Rossetti et al., 2004, 2015; Vitale Brovarone & Herwartz, 2013; to the Oligocene-Miocene in Tuscany and southern Apennines; e.g., Balestrieri et al., 2011; Bianco et al., 2019; Brogi & Giorgetti, 2012; Brunet et al., 2000; Molli, 2008; Vignaroli et al., 2009; Figure 1). In detail, in the northern Apennines, the metamorphic peak pressure decreases from ~2.0 GPa at 40–35 Ma in Corsica to ~0.6–0.8 GPa in southern Tuscany (ca. 26 Ma in the Tuscan Archipelago and at 24–20 Ma at Mount Argentario; e.g., Bianco et al., 2019; Brogi & Giorgetti, 2012; Brunet et al., 2000; Papeschi et al., 2020; Rossetti, Faccenna, Jolivet, et al., 2001; Figure 1).

2.1 Pontian Archipelago and Zannone Island

The Pontian Archipelago is located in the central-eastern Tyrrhenian back-arc basin on the western margin of the Latium continental shelf (Figure 1 and 2a). The present-day geology of the Pontian Archipelago mostly results from the postorogenic extensional tectonics and volcanic activity of the peri-Tyrrhenian region (Carminati et al., 2010; Malinverno & Ryan, 1986; Zitellini et al., 1984). The Pontian Archipelago consists of five islands and may be subdivided into two groups based on their structural and volcanic features (Chiocci & Orlando, 1996; De Rita et al., 2004; Pantosti & Velonà, 1986). The eastern group is composed of the Ventotene and Santo Stefano islands, which represent the subaerial portion of a large and submerged stratovolcano grown at the center of the Ventotene basin between 0.8 and 0.13 Ma (Cadoux et al., 2005; Metrich et al., 1988; Pantosti & Velonà, 1986; Peccerillo, 2005). The western group includes Ponza, Palmarola, and Zannone islands, which are located on a structural high bounded by normal faults formed during (postorogenic) early Pliocene extension (De Rita et al., 1986; Pantosti & Velonà, 1986; Zitellini et al., 1984). In the western Pontian Archipelago, volcanic rocks (Cadoux et al., 2005; Peccerillo, 2005) erupted at 4.2 Ma and 1 Ma over a complex substratum interpreted as a nappe stack consisting of three thrust sheets (De Rita et al., 1986; Pantosti & Velonà, 1986; Parotto & Praturlon, 1975; Figure 2c). The lowermost thrust sheet, exposed along the eastern coast of ZI, is described as being made up of phyllite, quartzite, and quartz-rich sandstone of Paleozoic-Triassic age (De Rita et al., 1986; Di Sabatino, 1979; Parotto & Praturlon, 1975; Segre, 1954; Figure 2c). The intermediate thrust sheet consists of Late Triassic dolostone (Segre, 1954; Figure 2c). The upper thrust sheet is composed of Upper Cretaceous-Eocene marl and marly limestone followed upward by Miocene terrigenous sedimentary sequences consisting of sandstone, marl, claystone, and gypsum-rich sandstone with a “younger-on-older” tectonic contact on the intermediate thrust sheet (Figure 2c). This contact has been interpreted as the “younger-on-older” reactivation of an early erosional surface (De Rita et al., 1986; Pantosti & Velonà, 1986; Figure 2c). The stack of imbricate thrust sheets is also seen onshore close to ZI in the Mount Circeo (Accordi, 1966; Pantosti et al., 1986), Volsci Range (Cardello et al., 2020; Rossi et al., 2002), and Mount Massico (Smeraglia et al., 2019; Vitale et al., 2019; Figure 1b) areas. Furthermore, thrusting and out-of-sequence thrusting at Mount Massico have been recently described and dated by U-Pb on syntectonic carbonate mineralizations to ~7 and ~5 Ma, respectively (Smeraglia et al., 2019).

2.2 The “Basement” Rocks of Zannone Island: An Historical Review

The geology of ZI has been traditionally described as reflecting the exposure of a classical metamorphic basement-cover transition, consisting of low-grade metamorphic rocks (known as “Scisti lucenti,” “shiny schists” in English) overlain by Triassic dolostone (Doetler, 1875; Sabatini, 1898). Due to the lack of fossils, Doetler (1875) and Sabatini (1898) suggested a Devonian-Carboniferous age for these metamorphic rocks. About 20–30 years later, Bieber (1924) and Galdieri (1905) interpreted the contact between the metamorphic rocks and the overlying Triassic dolostone as paraconcordant and therefore suggested that the metamorphic rocks could be Triassic in age. Subsequently, the contact was interpreted as a thrust fault and the metamorphic rocks were described as quarzitic sandstone intercalated with schist layers of Paleozoic age (Segre, 1954). De Rita et al. (1986) and Parotto and Praturlon (1975) reinforced the idea that the contact between the metamorphic rocks and the overlying Triassic dolostone is a thrust fault and the age of the metamorphic rocks was supposed to be Paleozoic or Triassic. De Rita et al. (1986) described the metamorphic rocks as phyllite, quartzite, and quartz-rich sandstone and, although the base of such rocks is not exposed, tentatively suggested that the metamorphic rocks at ZI may represent a large thrust block unit. Nevertheless, in his review on the metamorphic basement rocks in Italy, Vai (2001) leaned toward a reinterpretation of Paleozoic age for the metamorphic rocks at ZI. A (still loosely defined) Paleozoic-Triassic age was assigned to these rocks also on the geological map of Italy (ISPRA (APAT), 2015; Segre, 1960) with unclear relationships with the surrounding rocks.

3 Material and Methods

To address our research goals, we have combined a set of techniques and analytical methods including:

  1. Interpretation of offshore seismic reflection profiles acquired close to ZI to reconstruct the local subsurface geology and structures.
  2. Field mapping and mesostructural analysis along the eastern coast of ZI to investigate the metamorphic rocks, so far considered to be of Paleozoic-Triassic age, and the overlying Triassic dolostones. Sampling of the Paleozoic-Triassic metamorphic rocks was done for thin section analysis. Syntectonic carbonate veins, slickenfibers and clay fault gouge from the tectonic contact between the metamorphic rocks and the overlying Triassic dolostones were also collected for U-Pb and K-Ar dating, respectively.
  3. Petrotextural and microstructural observations by optical microscopy on oriented thin sections of the metamorphic rocks, dolostones, and syntectonic carbonate mineralizations.
  4. XRD analysis of the metamorphic rocks and thrust-related clay gouge to define their mineralogical assemblage.
  5. K-Ar geochronology on authigenic and synkinematic clay minerals from the fault gouge to constrain the timing of thrusting.
  6. U-Pb dating of syntectonic carbonate veins to constrain the timing of postorogenic extension.

Details on the analytical methods are provided in the supporting information.

4 Results

4.1 Interpretation of Seismic Reflection Profiles

The line drawing of seismic reflection profiles is based on the interpretation of the main structural elements, major unconformities, and reflections inside the unconformity-bounded seismostratigraphic units. The uninterpreted seismic reflection profiles are shown in the supplementary material (Figure S1). ZI is located on a horst/structural high bounded by normal faults (Figure 4) offsetting the Messinian reflectors by up to 0.5–0.7 s TWT. The structural high can be seen between shot point 2000 and 5,000 in Figure 4c and between shot point 3,000 and 13,000 in Figure 4c. Minor normal faults are also present in the surrounding basins (Figures 4b and 4c) and exhibit offsets in the order of 0.1–0.2 s TWT. The structural high is mainly composed of a seismic unit characterized by scattered discontinuous reflectors with high amplitude and variable frequency (whose top is marked by the green line in Figures 4b and 4c). Due to scarce signal penetration, the deeper part of the unit is also locally characterized by reflection-free portions (Figures 4b and 4c). The seismic unit above is characterized by discontinuous reflections with medium amplitude and medium frequency (Figures 4b and 4c). Locally, it also contains a reflection-free zone (Figure 4c between SP 3000 and SP 6000, around at 0.6 TWT). Its top is marked by a major unconformity (yellow line in Figures 4b and 4c) characterized by onlap relationships and erosional truncations. This unconformity is marked by high amplitude and strong impedance contrasts. The seismic unit above the angular unconformity displays a seismic facies generally characterized by continuous and well-layered reflectors, with medium to low amplitude and high to medium frequency. Within this unit, which can be observed in the basins adjacent to the ZI structural high, a major unconformity (blue line in Figures 4b and 4c) separates an upper and a lower portion, which differ for their internal geometries. The lower portion is characterized by thickness variations and growth geometries (Figure 4b), whereas the upper portion is characterized by subhorizontal reflectors (orange lines in Figure 4a), which locally show east-northeastward progradation geometries (Figure 4c).

Details are in the caption following the image
(a) Structural map obtained by integration of different multichannel seismic datasets (TIR-10 survey, acquired in the frame of CROP Project; http://www.ismar.cnr.it/products/reports-campagne/2010-2019, ViDEPI database), showing the trace of seismic lines surrounding the Zannone Island and the isochrones in two-way time (TWT, ms) of the top Messinian (yellow lines in b and c; Conti et al., 2017). (b and c) Interpreted seismic lines showing the top carbonates, top Messinian, and Middle Pliocene reflectors as well as normal faults bordering the structural highs.

4.2 Field Data

Field work on ZI was focused on the rocks ascribed in the literature to the Paleozoic-Triassic metamorphic basement (hereafter referred as basal unit) and on the contact with the overlying Triassic dolostone. We studied these features over an ~100 m long outcrop along the eastern coast of the island (Lat: N40°58′08.01″, Long: E13°03′46.01″, and ~2 m a.s.l.; Figures 2c and 5a).

Details are in the caption following the image
(a) Panoramic view of the outcrop studied in this work (eastern coast of the Zannone Island) with simplified geological interpretation (see location in Figure 2c). (b) Normal faults cutting the basal unit. Schmidt net (lower hemisphere projection) showing attitude of normal faults and related kinematic indicators. (c) Detail of the basal unit consisting of alternating pelite and competent sandstone beds.

The basal unit consists of subhorizontal brown and yellowish pelite and competent sandstone layers, organized in alternating 10–50 cm thick beds (Figures 5b and 5c). The maximum exposed thickness of the basal unit is of about 10 m (Figure 5a). The overlying rock unit consists of well-bedded Triassic dolostone that dips moderately (~25°) toward NE (Figures 5a and 6a). The contact between the Triassic dolostone atop the basal unit is defined by a low-angle thrust fault. It dips between 3 and 10° toward SW (~N210°) and consists of an irregular surface (Figure 6c), with dip-slip grooves, slickenlines, and carbonate slickenfibers (Figure 6a), indicating a top-to-NE (~N40°E) sense of shear (Figure 6a). In the hanging wall to the thrust, the dolostone is highly fractured and brecciated, containing a complex network of mutually crosscutting carbonate veins, arranged to form a mesh-texture (Figures 6b and 6e). Immediately below the thrust surface, the fault rock consists of foliated cataclasite and decimeter-thick lenses of fault gouge (Figures 6b and 6d). The thrust is cut by a set of NE-SW striking high-angle (60–80°) normal faults dipping both toward NW and SE (Figures 6a, 6b, and 6e). The fault surfaces are decorated by carbonate slickenfibers indicating dip-slip kinematics (pitch of about 90°; Figures 6a and 6e). The maximum estimated displacement for the normal faults is in the order of 5–10 m, whereas thrust displacement cannot be constrained.

Details are in the caption following the image
(a) Tectonic contacts between the basal unit and Triassic dolostone. Contacts consist of a low-angle thrust fault and at least two opposite-dipping (NW and SE) high-angle normal faults. Schmidt net (lower hemisphere projection) shows attitude of thrust and normal faults and related kinematic indicators. (b) Detail of the low-angle thrust cut by normal faults. The thrust juxtaposes the brecciated and fractured Triassic dolostone above the foliated basal unit. (c) Detail of the irregular thrust surface decorated by striae and carbonate slickenfibers. The thrust surface is composed of an ~2 cm thick carbonate fibers and marked with grooves. (d) Detail of the clay gouge exposed in the first 10 cm below the thrust surface with sampling site for K-Ar dating. (e) High-angle normal faults that cut the low-angle thrust. Notice the brecciated and veined Triassic dolostone in the hangingwall and the sampling site of carbonate slickenfibers (sample Z115) and carbonate vein (Z116) for U-Pb geochronology.

4.3 Petrographic and Microstructural Observations

The sandstone beds of the basal unit (Figures 7a and 7b) are mainly composed of quartz (up to 90% of the total rock volume) and subordinate grains made of feldspar, white mica, chlorite, and opaque minerals, embedded within a secondary carbonate cement (Figures 7c and 7d). Based on the mineral modes, these rocks can be classified as quartzarenite. Quartz grains are rounded-to-subrounded, commonly exhibiting evidence of undulose extinction, with size (i.e., diameter) not exceeding 200 μm (Figures 7c and 7d). Small variations in grain size define the faint planar anisotropy at the outcrop scale that is here interpreted as a primary foliation (bedding, S0; Figure 7c). The quartzarenite beds are characterized by an upward fining grain size (with the fining trend perpendicular to the S0; Figures 7a and 7e). Quartz grains are devoid of internal microfractures and evidence of pressure-solution between quartz grains is very scarce. When present, it is defined by subhorizontal pressure-solution surfaces (Figure 7f). An incipient secondary foliation (S1) subparallel to S0 is seldom observed (Figures 7b and 7f). Fine-grained white mica aggregates up to 50 μm thick and spaced tens-to-hundreds of microns apart locally outline the S1 (Figure 7f). Along S1, evidence of an incipient pressure solution is also observed, mostly localized along the white mica bands (Figure 7g).

Details are in the caption following the image
(a) High-resolution scan of a hand specimen of the competent sandstone beds of the basal unit. Note the upward decreasing size (see panel e for details) and the planar bedding (S0; see panel c for details). (b) High-resolution scan of a hand specimen of the competent sandstone beds of the basal unit showing foliation planes (S1; see panels f and g for details). (c) Microphotograph under cross-polarized light showing the mineralogical composition of the competent sandstone beds of the basal unit consisting of quartz grains of oblate shape with the long axis aligned parallel to the primary fabric (S0). (d) Microphotograph under cross-polarized light of white micas. (e) Detail of the upward decreasing size within a sandstone bed of the basal unit. (f and g) Microphotograph under cross-polarized and polarized light of white micas aligned along incipient foliation planes (S1) parallel to S0. Notice in (f) the local pressure-solution between quartz grains. Notice in (g) the pressure-solution planes localized along the alignment of white mica crystals. (h) Recrystallized Triassic dolostone and detail of a syntaxial carbonate vein.

The Triassic dolostone displays evidence of secondary recrystallisation overprinting the primary sedimentary features (Figure 7h). Carbonate-filled veins and brecciated layers are the most prominent features postdating the sedimentary fabric. Carbonate-filled veins are up to 1 cm thick (Figure 7h) with an internal texture characterized by crystal zonation oriented parallel to vein walls. Syntaxial growth of elongate to fibrous crystals from the vein walls toward the medial line is invariably recognized within these veins (Figure 7h).

4.4 XRD Analysis

4.4.1 Pelite Beds of the Basal Unit

The whole-rock composition of the less competent beds (pelite) of the basal unit consists of quartz (58–60%), calcite (1–22%), ankerite (1–24%), muscovite (4–9%), chlorite (7–14%), pyrite (1–2%), and K-feldspar (1–2%, Table 1).

Table 1. Summary of Mineralogical XRD Analyses for the Whole Rock Fraction of the Pelite Beds of the Basal Unit
Sample ID Lithology Whole-rock composition (%wt)
Qz Cal Ank Ms Chl Py Kfs
Z106 Pelite 60 1 24 4 7 2 2
Z111 Pelite 57 1 22 9 9 1 1
Z112 Pelite 58 22 1 4 14 1 tr
  • Note. Qz = quartz; Cal = calcite; Ank = ankerite; Ms = muscovite; Chl = chlorite; Py = pyrite; Kfs = K-feldspar; tr = traces.

The semiquantitative analysis of the <2 μm grain size fraction is shown in Table 2. In the pelite beds, illite is the most abundant mineral with contents between 47 and 68%, followed by chlorite (5–40%), smectite (10–41%), and subordinate kaolinite (4–12%). Occasionally, mixed layers chlorite-smectite (corrensite) were observed with contents not exceeding 2%.

Table 2. Semiquantitative X-ray Diffraction Analysis of the <2 μm Grain Size Fraction for the Pelite Beds of the Basal Unit
Sample ID Lithology <2 μm grain size fraction (%wt)
Sm I C-S K Chl Other
Z102 pelite 41 52 2 5 Qz
Z106 pelite 10 50 40 Qz
Z111 pelite 33 47 12 8 Qz
Z112 pelite 10 68 4 18 Qz
  • Note. Sm = smectite; I = illite; C-S = mixed layer chlorite-smectite (corrensite); K = kaolinite; Chl = chlorite; Qz = quartz.

4.4.2 Thrust-Related Fault Gouge

We collected one sample of the fault gouge (sample Z120) located along the thrust surface juxtaposing the Triassic dolostone in the hanging wall against the basal unit in the footwall (Figure 6d). We separated the sample into five grain size fractions from <0.1 to 10 μm (Figure 8a and Table 3). The mineralogical assemblage consists of quartz, dolomite, Na-plagioclase, hematite, illite/muscovite-2 M1, illite-1 M, smectite, and chlorite. Quartz and dolomite are present in the coarse subfractions (0.4–2, 2–6, and 6–10 μm) with contents of 11%, 27%, and 38% for quartz, and 3%, 8%, and 11% for dolomite. Na-plagioclase occurs in the 2–6 μm and 6–10 μm subfractions with contents <2%. Hematite occurs in the coarse subfractions (0.4–2, 2–6, and 6–10 μm) and does not exceed 3%. Chlorite also occurs in all subfractions with contents lower than 6%. The illite/muscovite-2 M1 content is about 1% in the <0.1 μm subfraction and ranges between 10% and 25% in the coarser subfractions. Illite-1 M does not exceed 4%. Smectite abundance progressively increases from 21% to 96% with decreasing grain size (Figure 8a).

Details are in the caption following the image
(a) X-ray semiquantitative analysis of thrust-related clay gouges (sample Z120, see Figure 6d for sampling location). Note the progressive increase of smectite with decreasing size. (b) K-Ar age versus grain size spectra. Dark gray horizontal bars broadly define geological time intervals. Notice the inclined age versus grain size curve. (c) U-Pb Tera-Wasserburg plots of syntectonic carbonate veins. Red lines are 2σ error envelopes of the regression line. Lower intercepts are age and 2σ uncertainties calculated as the lower intercept with Concordia curve; Y-intercept is 207Pb/206Pb ratio of the intercept of the isochron with Y axis (i.e., common Pb ratios of analyzed carbonate veins) and 2σ uncertainties. MSWD: mean squared weighted deviation. Isochrons for carbonate slickenfibers (sample Z115) developed along a high-angle normal fault which cuts the thrust (see Figure 6e for sampling location) and the carbonate mineralization (sample Z116) constituting the thrust surface (see Figure 6c for sampling location).
Table 3. Whole-Rock Composition of Various Subfractions From the Fault Gouge
Sample ID Grain size (μm) Whole-rock composition (%wt)
Qz Dol Pl Hem I-2 M1 I-1 M Sm Chl
<0.1 1 2 96 1
Z120 0.1–0.4 10 3 81 6
0.4–2 11 3 1 22 4 53 6
2–6 27 8 1 1 25 3 31 4
6–10 38 11 2 3 18 3 21 4
  • Note. Qz = quartz; Dol = dolomite; Pl = Na-plagioclase; Hem = hematite I-2 M1 = illite/muscovite-2 M1; I-1 M = illite/muscovite-1 M; Sm = smectite; Chl = chlorite.

4.5 K-Ar Dating of Synkinematic Clay Minerals in the Thrust-Related Fault Gouge

The five grain size fractions separated from the thrust fault gouge (Figure 6d) were analyzed by K-Ar geochronology (Table 4). K-Ar dates for synkinematic clay minerals are 133.20 ± 1.9 Ma for the 6–10 μm subfraction, 117.5 ± 1.7 Ma for the 2–6 μm subfraction, 85.3 ± 1.3 Ma for the 0.4–2 μm subfraction, 40.3 ± 0.6 Ma for the 0.1–0.4 μm subfraction, and 22.1 ± 0.6 Ma for the <0.1 μm subfraction (Figure 8b). Dates define an inclined spectrum in the sense of Pevear (1999), wherein the coarsest subfraction yields the oldest age (133.2 ± 1.9 Ma) and the finest subfraction yields the youngest age (22.1 ± 0.6 Ma). Ages of the intermediate size subfractions decrease with the grain size between these two end members (Figure 8b).

Table 4. K-Ar Data for the Fault Gouge
K 40Ar* Age data
Sample ID Grain size fraction (μm) Mass (mg) wt % σ (%) Mass (mg) mol/g σ (%) 40Ar* % Age (Ma) σ (Ma)
Z120 <0.1 51.5 0.428 1.5 4.422 1.6534E−11 2.47 2.6 22.1 0.6
Z120 0.1–0.4 52.1 1.931 1.5 3.824 1.3662E−10 0.41 22.6 40.3 0.6
Z120 0.4–2 51.8 3.450 1.5 3.688 5.2279E−10 0.26 67.0 85.3 1.3
Z120 2–6 51.6 3.567 1.5 4.320 7.5070E−10 0.24 81.8 117.5 1.7
Z120 6–10 55.4 2.986 1.5 3.826 7.1578E−10 0.36 84.9 133.2 1.9

4.6 U-Pb Geochronology of Syntectonic Carbonates

We performed U-Pb dating on one syntectonic carbonate vein and one carbonate slickenfiber associated with high-angle normal faults, which cut across and displace the studied thrust. In detail, we collected the carbonate vein (sample Z116) from a normal fault damage zone and the carbonate slickenfibers decorating the surface of the same fault (sample Z115; Figure 6e). Dates are 7.7 ± 5.8 Ma for sample Z116 and 7.2 ± 7.7 Ma for sample Z115 (Figure 8c and Table S1).

5 Discussion

5.1 No Large-Displacement Normal Faults on Zannone Island nor in Nearby Areas

The chronostratigraphic interpretation of the main unconformities and seismic units identified on the seismic profiles derives from: (1) previous seismostratigraphic works in the Tyrrhenian Basin (Conti et al., 2017; Moussat et al., 1986; Zitellini et al., 1984) and (2) the analysis of the exploration boreholes located along the Latium-Campania coast (Figures 2a and 4a).

The lower seismic unit can be assigned to the Triassic-Jurassic carbonates similar to those drilled offshore in the Michela 001 and Mara 001 boreholes (at depths of about 2,100 and 1,500 m, respectively) and cropping out in the adjacent onshore areas (e.g., Volsci Range, Mount Circeo, and Mount Massico; Figure 2a). The overlying unit can be interpreted as a succession of Miocene claystone capped by the unconformity ascribable to the top Messinian/base of Pliocene unconformity (yellow line in Figures 4b and 4c). The upper seismic unit (above the top Messinian/base of Pliocene unconformity) is represented by postorogenic Pliocene-Pleistocene siliciclastic sequences and occurs in the basin surrounding the structural high of ZI (Figures 4b and 4c). The major unconformity occurring within the Pliocene-Pleistocene unit (blue line in Figures 3b and 3c) separates prograding reflectors above from horizontal reflectors below. The unconformity is of regional extent and locally marks the end of fault activity; due to these characteristics, it is tentatively correlated with the middle Pliocene unconformity recognized by Conti et al. (2017) and Zitellini et al. (1984).

As already mentioned, the occurrence of metamorphic rocks tectonically overlain by Triassic dolostone on ZI would require a large normal throw (>3,000 m) in order to exhume metamorphic rocks and correlate them with the downfaulted blocks surrounding ZI. The seismic profiles (Figures 4a and 4b) do indeed indicate the presence of normal faults around ZI but their estimated throw is too low to be consistent with the aforementioned hypothesis. In detail, the major offsets occur along high-angle master normal faults bounding the structural high of ZI and are in the order of about 300 m (depth conversion of 0.5–0.7 s TWT by assuming an average seismic velocity of 2,000 m/s for the Pliocene-Pleistocene sequence; Figures 4b and 4c). Minor normal faults within the basins exhibit smaller displacements, in the order of 100–200 m (depth conversion of 0.1–0.2 s TWT; Figures 4b and 4c).

5.2 No Metamorphic Rocks on Zannone Island

Based on the interpretation of other authors, Vai (2001) proposed the Paleozoic-Triassic metamorphic rocks of the Monticiano-Roccastrada Ridge and Romani Mounts (Figure 1) to be the northward continuation of the basal unit of ZI, thus implying a regional correlation between the Paleozoic-Triassic metamorphic rocks on ZI with the Tuscan Metamorphic Complex (e.g., Brogi & Giorgetti, 2010; Molli, 2008; Figure 1). As documented by our field and laboratory evidence, however, the basal unit consists of alternating pelite and quartzarenite beds typical of siliciclastic turbidite successions (Mutti & Ricci Lucchi, 1978; Mutti et al., 2009; Nichols, 1999), rather than basement metamorphic rocks. Microtextural evidence reveals that the basal unit exposed on ZI experienced deformation at temperature <200°C, typical of deep diagenetic conditions (e.g., Houseknecht, 1984; Passchier & Trouw, 2005). This is attested by (i) the preserved crystal boundaries between quartz grains with limited evidence of intragranular pressure solution (Figure 7c) and (ii) the reorientation of detrital white mica, along which incipient pressure solution and stylolite formation occur (Figures 7f and 7g). In this context, the undulose extinction observed at the microscale in some quartz grains (i.e., evidence of dislocation creep mechanism associated with T > 300°C; Stipp et al., 2002) is interpreted as a crystallographic feature inherited from the eroded source rock rather than resulting from deformation of the siliciclastic turbidite sequence. It has to be pointed out that the temperature experienced by the siliciclastic turbidite rocks on ZI is not consistent with the peak temperatures reached by the metamorphic complexes of the Apennines, which range between 300°C and 550°C (e.g., Brogi & Giorgetti, 2012; Jolivet et al., 1998; Vignaroli et al., 2009; Figure 1). The compactness and deep diagenetic conditions recorded by the ZI siliciclastic turbidite rocks can be tentatively explained by thrust-related burial during orogenic thickening. By considering that the current thickness of the thrust sheet preserved above the siliciclastic turbidite rocks on ZI does not exceed 800 m (Figure 2c; De Rita et al., 1986; Parotto & Praturlon, 1975), and by assuming a geothermal gradient of 20–25°C/km during orogenic wedge accretion (as derived from exhumed Adria-derived metamorphic rocks in the inner sector of the Apennine belt; Vignaroli et al., 2009), a nowadays partly eroded overburden about 3 km thick can be estimated for the siliciclastic flysch-like rocks on ZI. Such an overburden is consistent with that estimated for the Mount Massico ridge, located onshore, ~50 km toward the east (Smeraglia et al., 2019) and with that constrained by the Mara 001 well (see location in Figure 2a). Alternatively, it is also possible that the <200°C temperature was experienced during postorogenic normal faulting and crustal thinning, when ZI was affected by magmatism and hydrothermalism associated with warm fluid circulation (De Rita et al., 1986; Vignaroli et al., 2016), which may have also led to the pervasive recrystallization of the Triassic dolostone (Figure 7h). Regardless of when and how the siliciclastic turbidite rocks experienced their maximum burial temperature, tectonic deformation produced a nonpenetrative S1 foliation (Figure 7b) marked by the reorientation of detrital white mica parallel to bedding (S0; Figure 7f). Along these mica domains, and less commonly along quartz grains, weak pressure-solution localized (Figures 7f and 7g). The rearrangement of mica flakes, which represent weak minerals among the rigid quartz grain skeleton, suggests that the development of S1, and therefore also pressure-solution textures, developed in response to the first deformation phase at the early stages of tectonic burial. In other words, it is likely that S1 developed as a multistage process during both early sedimentary (early diagenesis) and tectonic load (thrust sheet emplacement of Mesozoic-Cenozoic carbonates onto the siliciclastic turbidite rocks; Figures 9b and 9c).

Details are in the caption following the image
Simplified sketch showing the main sedimentary and tectonic events on ZI. The source area is thought to be the same as that of the Macigno Fm. (Amendola et al., 2016; Cornamusini et al., 2018). (a) During Late Oligocene-Early Miocene times, the flysch-type rocks deposited above the Mesozoic-Cenozoic carbonates in a foredeep. (b) At ~22 ma (Early Miocene) orogenic compression led to the overthrusting of Mesozoic-Cenozoic carbonates onto the siliciclastic flysch-type rocks. (c) Detail of incipient tectonic foliation beneath the thrust surface. Notice the development of clay gouge during thrusting and the incipient foliation marked by micas along which pressure-solution occurs. (d) During middle-late Miocene (?) times, the area evolved in a piggy-back basin where sandstones and marls deposited. (e) Since ~6–5 Ma (late Messinian-early Pliocene), the area of ZI was affected by postorogenic extensional tectonics. Normal faults downfaulted the Triassic dolostone and the underlying siliciclastic flysch-type rocks as well as the entire sedimentary succession exposed at ZI. (f) Present-day schematic relationship between the siliciclastic flysch-type rocks and the Triassic dolostone. Notice the development of carbonate veins and slickenfibers close by the Triassic dolostone downfaulted and juxtaposed onto the siliciclastic flysch-type rocks.

5.3 New Time Constraints on Compressional Tectonics on Zannone Island

NE verging, low-angle thrusting on ZI juxtaposed the Triassic dolostone over the turbiditic sequence of the basal unit. The thrust surface contains lenses of clay gouge mainly formed at the expense of the underlying siliciclastic rocks (Figures 6d and 9c). K-Ar dating of the finest grain size fraction of the clay gouge provides an age for thrusting of ~22 Ma (Early Miocene). The inclined age versus grain size curve defined by K-Ar ages is interpreted as the mixing of inherited components with authigenic mineral phases (Figure 8b), wherein the finest fraction reflects the authigenic/synkinematic component and corresponds to the timing of the last faulting increment (~22 Ma; Figure 8b). This is in agreement with the conceptual “Age Attractor Model” (AAM) proposed by Torgersen et al. (2015) and Viola et al. (2016), and already applied to Neogene deformation in the Northern Apennines by Viola et al. (2018). According to the AAM, the age of the last recorded increment of deformation usually acts as an attractor, toward which a mixing line converges from the oldest protolithic ages. The resulting inclined curve reflects the mixing of inherited components and authigenic mineral phases, where the slope of the curve is a function of the age difference between the two age end members. Consequently, ages of intermediate grain size fractions along the inclined curves can reflect a mixing of the two end members and are therefore generally devoid of clear geological meaning. The mineralogical composition of the dated fractions of the clay gouge shows a considerable increase of smectite content from 21% to 96% with decreasing grain size (Figure 8a and Table 3). This evidence suggests that smectite represents the primary authigenic K-bearing mineral in the clay gouge. The absence of detrital illite in the finest fraction (Figure 8b) may result from the inefficiency of mechanical comminution of large grains of such anisotropic mineral down to the <0.1 micron fraction. The age of the finest fraction (~22 Ma) should therefore realistically reflect the authigenic/synkinematic component and corresponds to the timing of the last thrusting event recorded on ZI, when the Triassic dolostone overthrust the siliciclastic turbiditic rocks toward the NE. The age of thrusting is coherent with independent geological and time constraints (biostratigraphic and stratigraphic) available for the internal portion of the central and northern Apennine belt, where compressional deformation is dated back to the late Oligocene-early Miocene time (e.g. Accordi, 1966; Carmignani et al., 1978; Castellarin et al., 1992; Malinverno & Ryan, 1986). In particular, at Mount Circeo, located ~30 km north of ZI (and representing the most internal contractional structure so far documented in the central Apennines; Figures 2a and 2b), the synorogenic stratigraphic record (Flysch di San Felice Fm.) has an Aquitanian depositional age (between about 23 and 20 Ma; Beneo, 1950; Civitelli & Corda, 1988). Therefore, a syn- to post-Aquitanian age is inferred for the compressive tectonics at Mount Circeo. This inference is consistent with our K-Ar age (~22 Ma) for thrusting on ZI and indicates a minimum deposition age of ~22 Ma for the siliciclastic turbiditic rocks of the basal unit exposed on ZI. The same inference would also imply that compressional tectonics on ZI was coeval or slightly younger than the deposition of such turbiditic rocks. In this context, our work provides the first absolute time constraint on the most internal (i.e., oldest) known thrust of the central Apennine thrust-and-fold belt. In particular, in the frame of this interpretation, ZI would represent the oldest preserved relict and evidence of the eastward migration of the Apennine compressive front, whereby Mesozoic-Cenozoic carbonates progressively overthrust younger (see section 5.5 for detailed discussion on the age) foredeep siliciclastic rocks since about Aquitanian times (Figures 9a and 9b).

5.4 New Time Constraints for the Postcompressive Extensional Tectonics on Zannone Island

The thrust studied in this work is cut by a set of NE-SW striking normal faults (Figures 6a and 6b), which offset the entire outcrop (Figure 5) as well as the surrounding basins (Figure 4). One of such normal faults downfaulted the thrust contact coupling the Triassic dolostone with the underlying turbiditic rocks, thus forming the youngest tectonic contact between these two rock units (Figure 6a and 6e). The seismic profiles interpreted in this work provide information on the timing of extensional tectonics on ZI. Extension-related growth strata from the Miocene unit and within the lower part of the Pliocene-Pleistocene unit testify to extensional tectonics in the ZI area at 6–5 Ma (Figure 4b). This observation is also consistent with the Upper Miocene-Lower Pliocene age previously proposed for the onset of crustal extension in this area (e.g., Conti et al., 2017; Malinverno & Ryan, 1986; Zitellini et al., 1984). Due to the low quality of the available seismic profiles, we do not have clear evidence of the preextensional compressional tectonics that is known in this portion of the Tyrrhenian margin. (e.g., Pantosti et al., 1986; Rossi et al., 2002; Smeraglia et al., 2019; Vitale et al., 2019). Furthermore, it should be pointed out that the occurrence of magmatic products, polyphase tectonics and of Messinian and intra-Miocene erosional unconformities, may hide compressional structures (such as folds and thrusts), which are documented elsewhere in the broad study area.

The activity of the extensional faults on ZI (Figures 5 and 6a) was likely associated with fluid overpressure, as documented by the mesh texture of syntectonic carbonate veins and carbonate cement in tectonic breccias along the normal faults (Figure 6e). Our U-Pb dating of such syntectonic carbonates indicates that normal faulting occurred at about 7 Ma (i.e., early Messinian; Figures 8c, 9e, and 9f). Even considering the large errors associated with U-Pb ages (7.2 ± 7.7 and 7.7 ± 5.8; Figure 8c), our U-Pb results are consistent with the stratigraphic evidence as derived from the analysis of the seismic reflection profiles, which suggest an age for the onset of extensional tectonics at ~6–5 Ma (late Messinian-early Pliocene; Figure 4b). Moreover, the U-Pb carbonate ages are also consistent with (1) the age of the synextensional volcanism in the area, which occurred since 4.2 Ma (e.g., Cadoux et al., 2005; Peccerillo, 2005), and (2) the age of the intrusive and volcanic rocks of the Tuscan Province (e.g., Peccerillo, 2005 Figure 1). In detail, extensional tectonics and crustal thinning in the western Pontian islands was followed and accompanied by widespread magmatism (between 4.2 and 1 Ma; Cadoux et al., 2005; Peccerillo, 2005).

According to the eastward propagation of both orogenic and postorogenic deformation in the central Apennines (Cavinato & De Celles, 1999; Patacca et al., 2008), the extensional tectonics recorded on ZI (~7 Ma) was followed, ~50 km to the east, by extension in the Mount Massico area, constrained at ~3 Ma by U-Pb dating on syntectonic carbonates (Smeraglia et al., 2019).

5.5 Regional Correlations

A correlation of the basal unit (turbiditic rocks) exposed on ZI with the siliciclastic Verrucano-like deposit of Triassic age (exposed in Tuscany, northern Apennines) is not likely since that deposit consists of (micro)conglomerates, breccias, and sandstones containing quartz clasts and metamorphic and volcanic lithic fragments (e.g., Aldinucci et al., 2008). Moreover, the inferred minimum depositional age of ~22 Ma for the siliciclastic turbiditic rocks allows us to exclude a correlation with the middle-late Miocene flysch deposits exposed in the central and southern Apennines (e.g., Centamore et al., 2007; Centamore & Rossi, 2009; Cosentino et al., 2010; Vitale & Ciarcia, 2013). Therefore, we interpret the basal unit of ZI as a flysch deposited in a foredeep basin that was older and more internal than the depositional basin of the Aquitanian flysch currently exposed to the north of ZI in the Mount Circeo area (Figure 2a). Hence, based on the Aquitanian depositional ages of the flysch successions exposed at Mount Circeo, we propose a late Oligocene-early Miocene depositional age for the turbidites (basal unit) at ZI.

A possible correlation of the flysch (basal unit) on ZI is attempted with the oldest and most internal foredeep deposits of the northern Apennines, that is, the quartz-rich turbiditic deposits of the Macigno Fm. from Tuscany (Figures 1 and 3). This formation has a late Oligocene-early Miocene depositional age, is widely exposed in the northern Apennines (Tuscan Domain, Figures 3 and 10) and consists of alternating bedded marl and sandstone (e.g., Barchi et al., 2003; Conti et al., 2020; Cornamusini et al., 2018; Molli et al., 2010). Moreover, the sandstone of the Macigno Fm. mainly consists of quartzarenite similar to that of the turbiditic sequence on ZI (e.g., Amendola et al., 2016; Cornamusini, 2002, 2004). Since the composition of the Macigno Fm. varies significantly between the proximal and distal portions of the turbidite system (i.e. from the coastal portion of Tuscany toward the east) and from its bottom to the top (e.g. Amendola et al., 2016; Cornamusini, 2004; Cornamusini et al., 2018), further investigations would be necessary to support this inference. However, based on age and lithological affinity, we deem that the Macigno Fm. contains the rocks that are the most similar and correlatable with the ZI turbiditic rocks. Our K-Ar age of 22.1 Ma represents the oldest available radiometric constraint on thrusting in the central Apennines and, if the correlation with the Macigno Fm. is right, then ZI would represent the southernmost continuation of the foredeep and related thrust front, which, in the early Miocene (~22 Ma), was advancing toward E and NE in both the northern and central Apennines (Figure 10). In this context, consistently with the source area proposed for the Macigno Fm. (Amendola et al., 2016; Cornamusini et al., 2018), we suggest that the source of the basal turbiditic unitconsists of uplifted crystalline rocks of likely European affinity (Figure 9).

Details are in the caption following the image
Time-distance diagram, including syntectonic extensional and compressional basins and periods of extensional and compressional phases in the northern Apennines, along the CROP 03 (data from Amendola et al., 2016; Barchi, 2010; Barchi et al., 2003; Beneo, 1950; Cavinato & De Celles, 1999; Cipollari & Cosentino, 1995, 1996; Cipollari et al., 1997; Curzi et al., 2020; Patacca & Scandone, 2007; Pauselli et al., 2006; Smeraglia et al., 2019; Vezzani et al., 2010). Notice that K-Ar and U-Pb ages are consistent with the proposed temporal evolution of the northern Apennine-Tyrrhenian back-arc basin system. The trace of the CROP 03 is represented by the blue line in the simplified geological map of the northern Apennines.

As mentioned above, the available seismic lines do not possess the necessary resolution to (1) resolve the internal structural configuration beneath the structural high of ZI and adjoining areas and (2) provide clear evidence for the compressive deformation close to ZI (Figure 4b and 4c). This has so far prevented to unravel the regional geometry of the thrust sheets exposed on the island. Nevertheless, the seismic unit observed below the top Messinian unconformity (Figures 4b and 4c) can be tentatively correlated with the Miocene siliciclastic deposits exposed above the Mesozoic-Cenozoic carbonates on ZI (De Rita et al., 1986; Parotto & Praturlon, 1975; Figure 2c). The siliciclastic deposits below the top Messinian unconformity can also be interpreted as tectonically equivalent to the flysch exposed onshore in the Latium-Campania area (Flysch di Frosinone and Flysch del Cilento Fms.; Centamore et al., 2007; D'Argenio et al., 1973; Parotto & Praturlon, 1975) and drilled offshore (Michela 001 and Mara 001 boreholes) and onshore (Fogliano 1 borehole; see locations in Figure 2a; Conti et al., 2017; Zitellini et al., 1984). This hypothesis would imply that the deposition of siliciclastic deposits continued through time (during the Miocene) above the thrust sheet of Mesozoic-Cenozoic carbonates at ZI, probably in a piggy-back basin setting (Figure 9d). Again, this reconstruction would imply that most of the vertical offset between Triassic dolostones on ZI and the coeval rocks in the Mara 001 borehole was acquired during shortening (thrusting of the Mesozoic-Cenozoic succession on Neogene flysch deposits), rather than during back-arc crustal thinning (and related normal faulting), reconciling the lack of large throw along normal faults in the NE offshore ZI.

5.6 Regional Implications

In late Oligocene-Early Miocene times, compressional orogenic deformation took place in the internal (western) portion of the Apennine wedge (Barchi, 2010; Castellarin et al., 1992; Malinverno & Ryan, 1986; Viola et al., 2018; Figure 10). In the northern Apennines, this early compressional deformation was associated with the overthrusting of Ligurian Units (Jurassic ophiolites and their sedimentary cover of the subducting Tethys ocean) over the Mesozoic-Cenozoic carbonates of the internal Tuscan Domain (i.e., western Adria continental margin; e.g. Caricchi et al., 2015; Carmignani et al., 1994; Molli et al., 2010; Figures 3 and 10). Early compressional deformation was accompanied by the deposition of the coeval Macigno Fm. in a foredeep basin (e.g., Barchi, 2010; Castellarin et al., 1992; Cornamusini et al., 2018; Figure 10). We propose that such early compressional deformation is radiometrically constrained on ZI, central Apennines, where Mesozoic carbonates overthrust onto the Macigno Fm.-like deposits at ~22 Ma (early Miocene; Figures 10 and 11). In the northern Apennines, the eastward advancing compressional front was accompanied by foreland-verging thrust faults and the deposition of the Cervarola Fm. in a foreland-ward advancing system of progressively younger foredeep basins (e.g. Barchi, 2010; Botti et al., 2004; Figure 10). In the central Apennines, the eastward advancing compressional front led to the thrust sheet emplacement at Mount Circeo in post-Aquitanian time (~<20 Ma; Figures 10 and 11), which is inferred by the Aquitanian depositional age of the flysch exposed at Mount Circeo (Beneo, 1950; Civitelli & Corda, 1988). In the late Miocene, during the forward breaking, postorogenic extension began in the central-eastern Tyrrhenian basin (Figure 10). Extension occurred in response to the opening of the Tyrrhenian back-arc basin and was associated with normal faulting, hydrothermalism, and magmatism (Cosentino et al., 2010; Malinverno & Ryan, 1986; Vignaroli et al., 2016). Our ~7 Ma U-Pb ages (Figure 8c) on syntectonic carbonate veins and slickenfibers are consistent with the occurrence of extensional tectonics in the coastal portion of Italy and in the area of ZI (Figures 10 and 11; e.g., Malinverno & Ryan, 1986; Vignaroli et al., 2016). From Late Miocene to present times, the compression-extension coupled system has progressively migrated eastward toward the Adriatic foreland (Figures 10 and 11; e.g., Barchi, 2010; Cardello & Doglioni, 2015; Cavinato & De Celles, 1999; Pauselli et al., 2006).

Details are in the caption following the image
Time-distance diagram, including thrusting, out-of-sequence thrusting, and postorogenic basin sedimentation in the central Apennines (modified after Cavinato & De Celles, 1999). The dashed rectangles are inferred from Beneo (1950), Cipollari and Cosentino (1995, 1996), Cipollari et al. (1997), and Cosentino et al. (2014). The U-Pb and K-Ar radiometric constraints available for the central Apennines are from Curzi et al. (2020) and Smeraglia et al. (2019). Notice that K-Ar and U-Pb ages are consistent with the proposed temporal evolution of the central Apennine-Tyrrhenian back-arc basin system. In particular, the K-Ar age is consistent with the inferred age of compressional deformation at Mount Circeo. The trace of the time-distance section across the central Apennines (with projected surrounding thrusts and postorogenic basins) is represented by the blue line in the simplified geological map of the central Apennines.

6 Conclusions

The exposure of supposedly Paleozoic-Triassic metamorphic rocks on ZI is no longer acceptable. New field observations, thin section and XRD analysis, and K-Ar dating point instead to a Miocene synorogenic turbiditic deposit that was likely deposited during the early phases of the Apennines fold-thrust development in a foredeep environment. This finding supports the identification of the innermost thrust sheet of the central-northern Apennines belt. This evidence, together with new radiometric dating on thrust-related fault gouge (ca. 22 Ma; K-Ar method) and postthrusting, synextensional carbonate mineralizations (ca.7; U-Pb method), allows us to propose a new evolutionary scheme for the central-northern Apennines-Tyrrhenian tectonic system. These results stimulate new geological and geophysical research in the Tyrrhenian offshore to identify and constrain the early evolutionary stages of the Apennine belt, which now finds a significant and novel landmark on ZI.

Acknowledgments

This work has been funded by Progetto di Ateneo Sapienza 2017 and 2019 (E. Carminati), by Progetto di Ateneo Sapienza 2018 (L. Aldega), and Bando per il finanziamento di progetti di ricerca congiunti per la mobilità all'estero di studenti di dottorato (M. Curzi and S. Franchini). FIERCE is financially supported by the Wilhelm and Else Heraeus Foundation and by the Deutsche Forschungsgemeinschaft (DFG, INST 161/921-1 FUGG and INST 161/923-1 FUGG), which is gratefully acknowledged. P. Cipollari, M. Mercuri, M. Santantonio, and D. Tentori are acknowledged for geological discussions. Ruikai Xie is thanked for assistance with K analysis. D. Mannetta is thanked for preparation of thin sections. We warmly thank the Editor, L. Jolivet, an anonymous reviewer, and reviewer Andrea Artoni for their constructive comments.

    Data Availability Statement

    Analytical data are available in the supporting information and in the Figshare external repository (https://doi.org/10.6084/m9.figshare.12252791).