Deep structure of the southern Apennines, Italy: Thin ...

TECTONICS, VOL. 24, TC3005, doi:10.1029/2004TC001634, 2005

Deep structure of the southern Apennines, Italy: Thin-skinned or thick-skinned? Davide Scrocca Istituto di Geologia Ambientale e Geoingegneria, Consiglio Nazionale delle Ricerche, Rome, Italy

Eugenio Carminati and Carlo Doglioni Dipartimento di Scienze della Terra, Universita` degli Studi di Roma La Sapienza, Rome, Italy

Received 18 February 2004; revised 12 October 2004; accepted 14 December 2004; published 24 May 2005.

[1] The deep structure of the southern Apennines (SA) accretionary wedge is still debated since industrial seismic reflection and well data provide reliable constraints only to a depth of about 10 km. As a consequence, two directly linked questions regard (1) the shortening in the accretionary prism (particularly within the buried Apulian thrust units) and (2) the degree of involvement of the lower plate basement (i.e., the Apulian crystalline basement). To address these issues, we have constructed a regional section along a recently released deep seismic reflection profile (CROP-04) which intersects the entire SA. The resulting cross section, adequately constrained to a depth of about 15 km, has been framed in a geodynamic scenario characterized by the eastward roll-back of the westward subducting ApuloAdriatic lithosphere. On the basis of this section we speculate on the deep structure, building both thin- and thick-skinned thrust models. A cross-check of these end-members models against documented tectonic, geophysical, and geochemical features shows that the thin-skinned model is generally more consistent with the available data. The development of basement slices with thicknesses of tens of kilometers is unlikely, while it remains possible that the Apulian basement could have been involved with its upper few kilometers. In the thin-skinned model, the total shortening of the allochthonous units (i.e., Apennine and Apulian carbonate platforms and Lagonegro basin) is estimated to be greater than 280–300 km. Some 90 km of shortening can be attributed to the Apulian thrust units. Citation: Scrocca, D., E. Carminati, and C. Doglioni (2005), Deep structure of the southern Apennines, Italy: Thin-skinned or thick-skinned?, Tectonics, 24, TC3005, doi:10.1029/2004TC001634.

Copyright 2005 by the American Geophysical Union. 0278-7407/05/2004TC001634$12.00

1. Introduction [2] The southern Apennines (SA) are one of the most intriguing segments of the Apennines thrust belt. Over the last decades the SA have been the object of an intensive hydrocarbon exploration program that has provided a wealth of new data. As a consequence, a better understanding of the SA main features has been developed, at least down to the depth investigated by wells and seismic data [e.g., Mostardini and Merlini, 1986; Casero et al., 1988; Roure et al., 1991; Lentini et al., 1996; Monaco et al., 1998; Mazzoli et al., 2000; Mazzotti et al., 2000; Menardi Noguera and Rea, 2000; Patacca and Scandone, 2001a, 2001b]. In spite of these extensive geological and geophysical studies, two major unsolved questions remain regarding the SA structure and geodynamic evolution: (1) What is the shortening accumulated by the accretionary prism and in particular within the deepest thrust sheets in the SA thrust belt (i.e., the Apulian carbonate platform units)? Also, as a consequence, (2) what is the degree of involvement of the lower plate basement (i.e., the Apulian crystalline basement) in the SA accretionary wedge? Two conflicting interpretations have been proposed so far (Figures 1a and 1b). [3] 1. The first (Figure 1a) suggests the existence of a basement wedge (the classic backstop proposed for most of the Alpine and Cordillera types of orogen) developed during the formation of the SA [e.g., Casero et al., 1988; Roure et al., 1991; Mazzoli et al., 2000; Menardi Noguera and Rea, 2000; Sciamanna et al., 2000; Speranza and Chiappini, 2002]. [4] 2. The second interpretation (Figure 1b) considers the SA accretionary prism to be mostly composed of sedimentary cover while the crystalline basement remains essentially undeformed [e.g., Mostardini and Merlini, 1986; Marsella et al., 1995; Doglioni et al., 1996; Mazzotti et al., 2000; Patacca and Scandone, 2001a]. [5] However, whatever is the preferred structural style, it should be noted that the available geophysical and geochemical data strongly support a geodynamic scenario (Figure 1c) characterized by the subduction of the ApuloAdriatic lithosphere below the SA [e.g., Doglioni et al., 1996; Ziegler and Roure, 1996]. [6] These models (Figures 1a and 1b) produce significantly different estimates of the shortening suffered by the Apulian carbonate platform and, consequently, of its orig-

TC3005

1 of 20

TC3005

SCROCCA ET AL.: DEEP STRUCTURE OF THE SOUTHERN APENNINES

TC3005

Figure 1. Conflicting interpretations about the deep structural setting of the southern Apennines. (a) Thick-skinned model. The Apulian crystalline basement is largely deformed and forms a large wedge within the SA accretionary prism. Shortening estimated at the top of the Apulian carbonates is small, in the order of 15– 25 km (modified after Menardi Noguera and Rea [2000] with permission from Elsevier). (b) Thin-skinned model. The southern Apennines accretionary prism is made up of completely rootless sedimentary nappes. The basement remains essentially undeformed and dips westward below the accretionary prism. The total shortening of the buried Apulian thrust sheets is estimated to be not less than 90 km (modified after Mazzotti et al. [2000]). (c) The geodynamic scenario suggested by the available geophysical and geochemical data is characterized by the subduction of the Apulo-Adriatic lithosphere below the SA.

inal length. This, in turn, has relevant implications both for any attempt to reconstruct the Mesozoic paleogeography in this area [e.g., Patacca and Scandone, 1989; Dercourt et al., 1992; Catalano et al., 2001] and also for a better understanding of the petroleum system evolution [e.g., Roure and Sassi, 1995; Sciamanna et al., 2000]. [7] To answer these unsolved questions, a geological cross section, which traverses the entire SA, from the Tyrrhenian to the Adriatic Sea, has been built along the recently released reprocessing of the CROP-04 deep seismic reflection profile [Mazzotti et al., 2000, 2005], acquired within the framework of the Italian deep crust exploration project [Scrocca et al., 2003]. The original interpretation of the CROP-04, developed by integrating available surface and subsurface data sets (industrial seismic and well data), provides further constraints to reconstruct the structure of this segment of the Apennines. [8] Building our cross section, we have considered only well-constrained structural features (with the exception of its western end), in order to develop a ‘‘conservative’’ interpretation. The portrayed setting extends to a depth of about 15 km. Moreover, this cross section has been framed in a geodynamic scenario characterized by the eastward rollback of the westward subducting Apulo-Adriatic lithosphere. The resulting interpretation may prove useful for further geophysical or geological modeling.

[9] On the basis of this well-constrained cross section, we have speculated on the possible deep (i.e., 10 – 30 km) structure, taking into consideration all the available geological and geophysical constraints. Two completely different models, which can be considered as two end-members, have been developed: (1) basement deeply involved in thrusting or (2) basement not involved (or only in the upper few kilometers). By analyzing and comparing the main tectonic implications of these models it has been possible to highlight some inconsistencies and shortcomings of the thickskinned model and to highlight some critical issues that may drive future research.

2. Geological Setting [10] The SA is a roughly NW-SE oriented segment of the Apennines thrust belt. The Apennines, with overall vergence to the NE, are characterized by low structural and topographic elevation. The accretionary prism migrated from west to east since the early Miocene being followed by coeval extensional tectonics which, progressively, crosscut the thrust sheets. The SA are bordered to the east by a foredeep, filled up by Pliocene-Pleistocene deposits [Casnedi et al., 1981; Balduzzi et al., 1982a, 1982b; Consiglio Nazionale delle Ricerche, 1992] and to the west by the Tyrrhenian back arc basin [Kastens et al.,

2 of 20

TC3005


TC3005

Figure 2. Simplified geological map of the southern Apennines (modified after Patacca et al. [1992a]). The location of the regional geological cross section shown in Figure 4 (basically coincident with the CROP04 seismic reflection profile) is shown on the map. Boxes highlight the CROP-04 segments shown in Figures 5 and 6. Letters refer to wells (A, Puglia 1; B, Gaudiano 1; C, Bellaveduta 1; D, Lavello 5; E, Lavello 1; F, S. Fele 1; G, M. Foi 1; H, Vallauria 1; I, S. Gregorio Magno 1; J, Contursi 1; K, Gargano 1). 1988, and references therein] developed at least since middle Miocene (Figure 2). [11] In the SA, the progressive propagation of the contractional deformation toward the foreland is clearly documented by the development and evolution of a series of eastward younging foredeep basins and by the occurrence of several piggyback basins, which developed on top of the advancing allochthonous units [Patacca and Scandone, 1989, 2001b]. [12] The progressive eastward migration of extensional processes along the Tyrrhenian margin, of the thrust belt fronts, and of the foreland flexure (and associated foredeep basins) are related to the retreat of the subducting ApuloAdriatic lithosphere [Malinverno and Ryan, 1986; Patacca et al., 1990; Doglioni, 1991]. [13] Because of the very complex geological setting of the SA several, often conflicting, paleogeographical models have been proposed for the passive margin of the Adriatic plate [D’Argenio et al., 1975; Mostardini and Merlini, 1986; Casero et al., 1988; Sgrosso, 1988; Patacca et al., 1992a; Marsella et al., 1995; Menardi Noguera and Rea, 2000]. These models are characterized

by a variable number of carbonate platforms and pelagic domains. It is beyond the scope of this paper to discuss in detail these differences. We adopted a paleogeographic model that, at least in the SA sector crossed by our cross section, fits the available stratigraphic and structural data. From west to east, the main Mesozoic domains are as follows: (1) the internal oceanic to transitional LigurideSicilide basinal domains (internal nappes), (2) the Apennine carbonate platform, (3) the Lagonegro-Molise basins, and (4) the Apulian carbonate platform (Figure 3). In this model, the Liguride-Sicilide nappes represent remnants of the Neotethyan oceanic domain. The Apennine and Apulian platforms and the intervening Lagonegro-Molise basin developed during the Mesozoic rifting and the subsequent passive continental margin evolution of the Apulo-Adriatic plate [Channell et al., 1979]. Remarkably, these platform carbonates and pelagic basin were all originally located on the same Apulian crystalline basement (Figure 3). [14] The forward propagation of top-to-the-east thrusts piled up these paleogeographic domains during the eastward roll-back of the subduction hinge.

3 of 20


TC3005

TC3005

Figure 3. The main paleogeographic domain recognized in the southern Apennines developed during the Mesozoic rifting and the subsequent passive continental margin evolution of the Apulo-Adriatic plate. The Apennine and Apulian platforms and the intervening Lagonegro-Molise basin were originally located on the same Apulian crystalline basement (modified after Casero et al. [1988]). [15] The complete closure of the Neotethyan domain was achieved in the SA following a stage of subduction of oceanic crust, developed in the Late Cretaceous to early Miocene [Cello and Mazzoli, 1999]. After the overthrusting of the Liguride-Sicilide units onto the Adriatic plate Mesozoic passive margin (characterized by transitional to continental lithosphere), the SA accretionary prism incorporated the sedimentary cover of the passive margin itself through a series of thrusting events. Commencing in the middle Miocene, the tectonic accretion within the thrust belt has been contemporaneous with extensional tectonics along the Tyrrhenian margin which produced thinning of the internal sectors of the belt [Casero et al., 1988; Patacca et al., 1990; Cello and Mazzoli, 1999]. 2.1. Internal Nappes [16] This group of nappes comprises sediments derived from internal domains (Figure 3), deposited on oceanic to transitional crust. It is subdivided into the following units. 2.1.1. Liguride Units [17] Early Cretaceous to early Miocene sequences with incorporated ophiolitic suites. They comprise both the metamorphic Frido Unit and the unmetamorphosed Cilento Unit [Ogniben, 1969; Knott, 1987; Bonardi et al., 1988; Monaco and Tortorici, 1995]. The Frido Melange has been interpreted as a part of an accretionary prism built up during the Cretaceous subduction of the Tethys oceanic lithosphere [Knott, 1987, 1994]. 2.1.2. Sicilide Units [18] Late Cretaceous – early Miocene succession of basinal deposits [Ogniben, 1969]. The provenance of the Sicilide units from a basinal domain located west of the Apennine Platform is inferred from their geometric position; the Sicilide units systematically overlie the Alburno-Cervati carbonates from the Cilento area to the high Agri valley. However, an external original position

(i.e., east of the Apennine carbonate platform) has also been proposed [Mostardini and Merlini, 1986; Casero et al., 1988; Pescatore et al., 1988]. A discussion on this topic can be found in the work of Menardi Noguera and Rea [2000]. 2.2. Apennine Carbonate Platform [19] The thrust units belonging to this domain, also known as Western or Campano-Lucana Platform, form the SA backbone. This domain (Figure 3) consists of a thick pile (up to 5000 m) of Late Triassic-early Miocene shallow-water carbonates [Sartoni and Crescenti, 1961; Selli, 1957, 1962] overlain by condensed hemipelagic and siliciclastic deposits related respectively to its flexural sinking and to the onset of foredeep environments [Patacca et al., 1990]. [20] In the area crossed by CROP-04 profile, this paleogeographic domain includes tidal flat and protected shelf lagoon facies (Alburno-Cervati unit), platform edge (M. Marzano) and slope facies (Monti della Maddalena). [21] All the thrust sheets derived from this domain are generally detached along an intra-Triassic detachment from their Paleozoic substratum, which has never been drilled by exploration wells. 2.3. Lagonegro-Sannio and Molise Basinal Units [22] These units derive from sedimentary successions deposited in a relatively deep basinal domain. While the Middle Triassic-Early Cretaceous portion (Lagonegro) of this basinal domain is not controversial, the nature and characteristics of the Late Cretaceous-Tertiary section is still debated. [23] The typical Lagonegro stratigraphic succession (Figure 3) is made up of the following four formations which evolve from fluvial conglomerates and shallow water carbonates of the ‘‘Monte Facito’’ (Middle Triassic) to the

4 of 20

TC3005


‘‘Calcari con Selce’’ (Late Triassic), ‘‘Scisti Silicei’’ (Jurassic), and ‘‘Galestri’’ (Early Cretaceous) deep water facies. For a detailed description of the Lagonegro formations see, among many others, Scandone [1967, 1972], Wood [1981], and Miconnet [1988]. [24] In the classical geological literature, the outcrops of the Lagonegro formations have been grouped into four different facies [Scandone, 1967] while, at a regional scale, the structural setting of the Lagonegro units has been defined in terms of two superimposed nappes [Scandone, 1972]. The structurally upper nappe is named Lagonegro type II, and displays more proximal depositional characteristics. The lower nappe is termed Lagonegro type I, and shows more distal facies. Carbone et al. [1991] and Mazzoli et al. [2001] have questioned this matching between thrust units and sedimentary facies. [25] The primary geometry of the Lagonegro basin was dramatically modified in the Miocene-Pliocene by thrusting, breaching and out-of-sequence processes (which produced complex imbricates and antiformal stacks) and by later normal faults. [26] Concerning the palinspastic reconstruction of the Lagonegro domains in the SA, there is significant agreement about the original position of the Lagonegro basin between the Apennine and Apulian Platforms (although an internal provenance of the Lagonegro units has been proposed by Marsella et al. [1995]). [27] The pelagic Late Cretaceous-early Miocene deposits ascribed to the Sannio Unit [Patacca et al., 1992a] are likely the upper portion of the Lagonegro units detached from its Triassic-Early Cretaceous part and transported farther east [Carbone et al., 1988; Carbone and Lentini, 1990]. [28] Tertiary basinal deposits, outcropping along the eastern edge of the SA thrust belt and usually named Molise units (Tufillo-Serrapalazzo and Daunia units [Patacca et al., 1992a, and references therein]), were also completely detached from their original substratum. The Early Messinian age of the Molise foredeep deposits [Patacca et al., 1992b] clearly documents the original external paleogeographic position of these units (i.e., east of the western carbonate platform and likely at the northeastern margin of the Lagonegro –Molise basin). 2.4. Apulian Carbonate Platform [29] This carbonate platform is made up of 5000 to 7000 m thick Mesozoic-Miocene shallow-water carbonates stratigraphically overlain by upper Messinian and Pliocene terrigenous deposits (Figure 3). The Apulian carbonates rest on the top of Permian volcanoclastic deposits (Puglia 1 well [Ricchetti et al., 1988; Mazzoli et al., 2000]), sometimes overlain by Carnian-Ladinian units (e.g., Gargano 1 well [Bosellini et al., 1993]). [30] These carbonates crop out in Apulia (Gargano and Salento regions) and represent the preorogenic cover of the foreland area [Ricchetti et al., 1988]. The Apulian platform can be followed westward, by means of seismic reflections and well data, below the Plio-Pleistocene foredeep deposits

TC3005

[Sella et al., 1988; Patacca and Scandone, 2001b] and also below the axial part of the SA thrust belt, where they represent the deepest tectonic units [Mostardini and Merlini, 1986; Casero et al., 1988; Mazzoli et al., 2000; Menardi Noguera and Rea, 2000].

3. Constraints to the Shallow Structure [31] Several cross sections through the SA have been published in the last 20 years. In these sections a variety of structural hypotheses have been represented [e.g., Mostardini and Merlini, 1986; Cello et al., 1989; Casero et al., 1988; Patacca and Scandone, 1989; Roure et al., 1991; Marsella et al., 1995; Prosser et al., 1996; Doglioni et al., 1996; Lentini et al., 1996; Mazzoli et al., 2000; Mazzotti et al., 2000; Menardi Noguera and Rea, 2000; Sciamanna et al., 2000; Patacca and Scandone, 2001a; Lentini et al., 2002; Catalano et al., 2004; Mazzotti et al., 2005; Scrocca et al., 2005]. The most profound differences between the models occur at depth (i.e., deeper than 10 km), while the main tectonic features at shallower levels are relatively similar (Figure 1). This observation confirms that the available seismic reflection and well data provide a well-constrained picture of the structural setting to depth shallower than of about 10 km. [32] To highlight the main available geological constraints, we have built a regional cross section that cuts across the entire SA thrust belt-foredeep-foreland system and is nearly coincident with the CROP-04 deep seismic reflection profile (Figure 2). This seismic profile, acquired between 1989 and 1990 within the framework of the Italian deep crust exploration project (CROP Project), offers new information on the structure and tectonic evolution of the SA. Nearly continuous and well-structured seismic signals can be recognized down to 8 s two-way time (TWT) or more, well below the usual limit of industrial seismic data. The uninterpreted seismic data can be found in the CROP Atlas [Scrocca et al., 2003] and in the special publication dedicated to the CROP-04 profile [Mazzotti et al., 2005]. The interpretation of the CROP-04 seismic reflection profile has been developed combining the results of field surveys, carried along the section trace, and the interpretation of industrial seismic lines and well logs, made accessible by the oil industry [see also Scrocca et al., 2005]. Thanks to the new data provided by the CROP-04 profile, our cross section can be considered a conservative representation of the SA structure to depths of about 15 km (Figure 4). [33] Unfortunately, because of the SA tectonic complexity, a reliable structural balancing cannot be properly performed along the entire length of the cross section presented in this paper. In fact, the plane strain deformation requirement, one of the fundamental requirements for the construction of a balanced cross section, is not fulfilled in some areas (due to strike-slip tectonics or motions out of the plane of section). [34] Despite these problems, when possible, balancing techniques (e.g., key-bed lengths balancing) have been applied in some segments of our cross section to estimate the predeformation extent of the sedimentary cover (e.g.,

5 of 20

TC3005


Figure 4. Regional geological cross section (location in Figure 2). This section, which is adequately constrained by surface and subsurface data sets (seismic and well data) to a depth of about 15 km, is framed in a geodynamic scenario characterized by the westward subduction of the Apulo-Adriatic lithosphere. The main features of the southern Apennines lithospheric setting are (1) the flexure of the subducting Apulian lithosphere below the SA, with the slab top at depth greater than 100 km below the Tyrrhenian coastline, and (2) the presence of a hot mantle wedge underlying a new ‘‘young’’ and hot Moho in the western side of the SA accretionary prism. Note, as highlighted by the question mark, that the structure of the SA accretionary wedge at depth greater than 15 km is not constrained by the available geological and the geophysical data. Key-beds (k) used to estimate the predeformational extent of the Apennine carbonate platform and Lagonegro units are also shown.

6 of 20

TC3005

TC3005


Apennine platform and Lagonegro basin) and to assess shortening at top Apulian carbonates horizon. [35] The cross section (Figure 4) will be described starting from its northeastern edge, located in the Apulian foreland, and moving toward the Tyrrhenian side. 3.1. Apulian Plate: Flexural Geometry and Upper Crust Stratigraphy [36] The Apulian carbonates outcropping in the foreland can be followed southwestward, below the Plio-Pleistocene clays and sands of the Bradanic Trough, the youngest SA foredeep basin [Balduzzi et al., 1982a, 1982b; Patacca and Scandone, 2001b]. The attitude of the top Apulian (TAP) carbonates horizon is easily identifiable from seismic reflection and well data (Figure 5) [Sella et al., 1988; Nicolai and Gambini, 2005] and clearly depicts the flexural geometry of the Apulian Platform [Royden et al., 1987; Royden, 1988]. The regional monocline shows increasing dip [Mariotti and Doglioni, 2000] moving from the foreland area toward the southwest, with dip values up to 10° or more below the deepest portion of the foredeep (Figure 4). The flexural bending of the Apulian lithosphere introduces a constraint that was not considered in some structural interpretations [e.g., Mostardini and Merlini, 1986; Marsella et al., 1995]. [37] Another deeper strong reflector, generally subparallel to the TAP, can be picked on seismic lines both below the foredeep [Roure et al., 1991] and in some sectors of the SA thrust belt [Mazzoli et al., 2000]. This horizon is interpreted as the bottom of the Apulian carbonates and attributed to the acoustic impedance contrast between the Upper Triassic dolomites and the underlain Permo-Triassic clastic deposits [Ricchetti et al., 1988; Mazzoli et al., 2000]. [38] At a regional scale the Apulian Platform shows an almost constant time-interval thickness, with minor variations mainly due to the variable thickness of the Cretaceous section. Below the foredeep, on the CROP-04 profile, a representative thickness of about 2.4 s TWT can be estimated corresponding to about 7000 – 7400 m. A simplified stratigraphy for the Apulian carbonate platform has been represented in our cross section with an average thickness of about 7400 m for the Late Triassic-Miocene carbonates (Figure 4). [39] Below the Apulian carbonates the Puglia 1 well drilled at least 1000 m of lower Triassic-Upper Permian clastic deposits that represent a synrift sequence, characterized by variable thickness [see Merlini et al., 2000, Figure 7]. However, considering the lack of other information, a constant thickness of 1500 m has been assumed for this Permo-Triassic unit. 3.2. Allochthonous Units [40] The easternmost (i.e., more advanced) thrust sheets cropping out in the SA thrust belt consist of Tertiary basinal deposits belonging to the Molise Unit. In the subsurface, the Molise units are usually the lowest Apenninic thrust sheets above the early Pliocene clays which disconformably overlie the Apulian carbonates (Figures 4 and 5).

TC3005

[41] Moving westward, ‘‘controversial’’ formations crop out along a NW-SE belt. These formations are made up of lithologically monotonous basinal sequences often of unknown age, because of the lack of diagnostic fossils. These allochthonous units have been assigned to diverse thrust sheets, originated from different paleogeographic domains, by different researchers. A typical example is represented by the ‘‘Argille Varicolori’’ (Varicoloured Shale) outcropping in the axial segment of the SA that have been ascribed by some researchers to the Sicilide Units [e.g., Patacca et al., 1992a], while are considered by other researchers the detached upper portion of the Lagonegro sequence [e.g., Pescatore et al., 1988]. Some of these outcrops have also been attributed to the Sannio unit, whose original position is still open to debate [see Patacca et al., 1992a]. In our interpretation, we have assumed that the large majority of these outcrops could represent the detached upper portion of the Lagonegro sequence. However, this uncertainty does not have any major impact on our structural reconstructions and on our shortening estimates. [42] Seismic reflection and well data (e.g., S. Fele 1 and M. Foi 1 wells [Patacca et al., 2005; Scrocca et al., 2005] show that the Lagonegro units are part of a complex antiformal thrust stack (Figures 4 and 5). Those structures are further complicated by younger thrusts, not older than late Pliocene. These younger thrusts show out-of-sequence features at the shallow levels (with decapitation of earlier anticlines in the Lagonegro antiformal stack) but, in reality, they turn out to be breaching thrust [Butler, 1987], responsible for the building up of the Apulian thrust units at depth [Scrocca et al., 2005]. [43] In the southwestern part of the cross section, the Lagonegro units are overthrust by the Apennine platform carbonates, which in turn are overridden by the Sicilide units (Figures 4 and 6). The east plunging homocline of the Apennine platform (M. Soprano ridge) is the forelimb of a wide anticline. This anticline developed in the hanging wall of a thrust which caused a tectonic doubling of the thickness of the entire Apennine platform. In the footwall of this thrust, lies the SW plunging M. Alburno homocline. This interpretation accounts for the anomalously elevated thickness of the Apennine platform in this area, detected in the CROP-04 seismic section (Figure 6). The ‘‘normal’’ thickness of the Apennine platform is estimated below M. Alburno to be 1.8– 2 s TWT which, assuming an average velocity for Apennine carbonates in the range of 5500– 6000 m/s, implies a total thickness of at least 5000 m for this unit (as adopted in our cross section). [44] The ‘‘transparent’’ seismic facies generated by the Apennine carbonate sequence overlies another very reflective and well-stratified seismic facies, less than 1 s TWT thick. This facies, well-known also from industrial seismic lines, can be followed across the western end of the CROP04 profile and overlies another poorly reflective seismic facies (Figure 6). These seismic data have been interpreted in different ways. Menardi Noguera and Rea [2000] suggest the occurrence of a huge slice of Paleozoic basement (originally located below the Lagonegro or below the Apennine carbonate platform domains) overlaying the bur-

7 of 20

Figure 5. Detail of the central part of the CROP-04 seismic profile (location in Figure 2). The main structural features of the allochthonous units can be recognized. Top Apulia horizon has been traced integrating well data and the interpretation of industrial seismic lines.

TC3005 SCROCCA ET AL.: DEEP STRUCTURE OF THE SOUTHERN APENNINES

8 of 20

TC3005


TC3005

Figure 6

TC3005

9 of 20

TC3005


TC3005

Figure 7. Palinspastic sketch along the regional cross section shown in Figure 4. The sedimentary covers belonging to the Apennine carbonate platform and to the Lagonegro-Molise basin are completely detached from their original basement. The amount of missing crystalline basement, originally located below these domains, has been approximately evaluated applying key-bed balancing techniques to derive the predeformational width of the sedimentary cover. On the basis of our estimates, the extent of the missing crystalline basement is estimated to be not less than 190– 210 km. ied transition between the western margin of the Apulian platform and the Lagonegro basin. The CROP-04 seismic image suggests an alternative interpretation [Mazzotti et al., 2000; Scrocca et al., 2005]. [45] Deep exploration wells (e.g., Contursi 1 well and S. G. Magno 1 wells) penetrated the well-stratified seismic facies in the NE end of Figure 6, which turned out to be Lagonegro units. Below the Lagonegro units the TAP horizon has also been encountered. On the basis of the CROP-04 seismic data (Figure 6), both the stratified seismic facies, interpretable as Lagonegro Units, and the TAP horizon can be followed deepening westward below the Apennine platform thrust sheets with good continuity. This interpretation is also supported by the strong seismic event observable in the SW end of the CROP-04 line, at about 8 s TWT, which could be interpreted as a near bottom Apulian carbonates reflector (Figure 6). According to this preferred interpretation, no basement rocks occur within the shallow part of the SA thrust stack. [46] With the exception represented by the controversial situation described below the M. Alburno area, it is generally accepted that the sedimentary cover belonging to the Apennine carbonate platform and to the Lagonegro-Molise basin is completely detached from their original basement. [47] To stress the magnitude of this phenomenon, we have approximately estimated of the amount of missing crystalline basement. To get this result we have evaluated the predeformation width of the Apennine platform and of the Lagonegro basin assuming that an equivalent amount crystalline basement was originally situated below these domains. Line lengths of key-beds, characterized by little penetrative deformation and with preserved hanging wall and footwall cutoffs, have been measured on our cross section (Figure 4) for both the Apennine carbonate platform (i.e., top Jurassic) and the Lagonegro basinal units (i.e., top

Late Triassic corresponding to the top of the‘‘ Calcari con Selce’’ Formation). [48] Although we are aware of the limitations of this simplified approach, since processes that can have altered line length during the deformation of the sedimentary cover (e.g., pressure solution phenomena, strike slip tectonics or polyphase deformation) have not been taken into account, it should be noted that for the purposes of this study an accurate calculation was not necessary. [49] On the basis of our estimates (Figure 7), the extent of the missing crystalline basement originally located below the Apennine carbonate platform and the Lagonegro basin is conservatively in the range of 190 – 210 km (respectively about 70 – 80 km and 120 – 130 km). The most likely explanation for this piece of evidence is that the sedimentary cover has been off-scraped from the subducting ApuloAdriatic lithosphere during the eastward roll-back of the subduction hinge [see also, Roure et al., 1991; Doglioni et al., 1996]. Consequently, the crystalline basement, originally underlying the sedimentary cover, should have disappeared in the subduction zone. In fact, a steady state involvement of the basement since the early stages of the accretionary prism should have generated (1) a much larger volume of rock in the hanging wall of the subduction, (2) emersion and outcrop of basement slices, and (3) a more elevated belt. The geodynamic implications related to the subduction of continental crust will be discussed in detail in a following section. [50] Geodynamic reconstructions of the Alps-Apennines interaction [e.g., Doglioni et al., 1998, 1999] suggest that the onset of the Apennine subduction could have occurred, since the Oligocene, along the southwestward prolongation of the preexisting double vergent Alps orogen, when the ‘‘eastward’’ Alpine subduction was gradually replaced by the opposite westward directed Apennine subduction. On the basis of this geodynamic interpretation it is considered

Figure 6. Detail of the western part of the CROP-04 seismic profile (location in Figure 2). On the eastern edge of the seismic data the presence of the Lagonegro units and of the top Apulian is proved by well data. In our interpretation, both Lagonegro units and the Apulian platform could extend westward below the Apennine carbonate units. 10 of 20

TC3005


very likely that crystalline basement slices were originally located on top of the previously described Apennine allochthonous units as relics of the Alpine orogen. This structural configuration is nowadays clearly recognizable in Calabria, south of M. Pollino in Figure 2 [e.g., Bonardi et al., 1994]. In our cross section these basement units are not present because of erosion and the back arc extension that affected the western side of the Apennine accretionary prism. Remnants of these units have been found scattered and disrupted, further to the west, on the floor of the Tyrrhenian Sea [e.g., Consiglio Nazionale delle Ricerche, 1992]. [51] Finally, it is noteworthy that the available subsurface data allow an estimation of the cumulative forward allochthonous nappes displacement that occurred in the late Pliocene-early Pleistocene. According to the work of Patacca and Scandone [2001b], the nappe advance, characterized by transport over a long flat, occurred in two phases (from 3.70 to 3.30 Ma and from1.83 to 1.50 Ma) and at least 30 km of displacement took place starting from 3.70 Ma. Another estimate has been provided by Sciamanna et al. [2000] who have calculated almost 40 km between 3.57 and 0.66 Ma. 3.3. Buried Apulian Antiformal Stack [52] Below the Lagonegro and Molise allochthonous units, several wells have penetrated carbonate platform sequences, Mesozoic to Tertiary in age. As documented by their facies, by the age of the stratigraphically overlying foredeep deposits and by the structures recognized in seismic sections, these carbonate deposits are considered to be imbricated units accreted from the western portion of the flexured Apulian Platform. These units form an antiformal stack (Figures 4 and 5), separated by a sole thrust from the undeformed, or weakly deformed, part of the Apulian platform, which crops out in the Apulia foreland [Mostardini and Merlini, 1986; Casero et al., 1988; Patacca and Scandone, 2001b]. The bottom of the Apulian carbonates and the underlying lower Triassic-Upper Permian clastic deposits are clearly offset by thrusts on seismic [Mazzoli et al., 2000]. [53] The main phases of the tectonic accretion of the Apulian domain occurred during the late Pliocene-early Pleistocene [Menardi Noguera and Rea, 2000; Sciamanna et al., 2000], although the onset of the deformation in the inner portion of the domain may have commenced at the end of the early Pliocene [Cello and Mazzoli, 1999]. [54] Although the overall structure of the antiformal stack is relatively well portrayed by subsurface data [Nicolai and Gambini, 2005], the westward extension at depth of these units below the outcropping Apennine carbonate platform is still discussed [e.g., Menardi Noguera and Rea, 2000; Mazzotti et al., 2000]. We have adopted a solution consistent with the CROP-04 seismic data, which is characterized by the presence of a thrust unit with Apulian affinity below the Apennine platform and Lagonegro thrust sheets [Mazzotti et al., 2000; Scrocca et al., 2005]. [55] The detailed reconstruction of the structure of the Apulian antiformal stack is made problematic by the poor resolution of seismic reflection data, as usually happens in

TC3005

complex terrains such as thrust belts, which allow only a rough description of the thrust related anticlines. Seismic reflection data provide a relatively good definition only of the hanging wall TAP horizon while both the thrust faults and their footwalls and the deeper horizons (e.g., the bottom Apulian carbonates) are generally poorly constrained (Figure 5). These uncertainties create the latitude for the dramatically different structural interpretations of the SA at depth. [56] The Apulian antiformal stack presented in our cross section has been reconstructed starting from a conservative interpretation of seismic reflection profiles and well data. This mean that first we have defined the relatively well-constrained hanging wall TAP geometry and then we have modeled the thrust faults and related footwalls, assuming a minimum displacement criterion at TAP level. [57] Noteworthy, the central and northeastern parts of the buried Apulian antiformal stack, the main ramp anticlines are substantially unaffected by strike-slip faults. Moreover, our cross section is nearly parallel to the transport direction of these ramp anticlines, as inferred from their regional trends [e.g., Casero et al., 1988; Nicolai and Gambini, 2005]. These evidences suggest that a plane strain deformation condition should be fulfilled at least for the central and northeastern parts of the buried Apulian antiformal stack. On these bases the shortening at TAP level (STAP) has been quantified by applying simple key-bed balancing techniques and by comparing restored and deformed lengths (i.e., STAP = Lr Ld, where Lr is restored length and Ld is deformed length). [58] The calculated shortening amounts to about 20 km (mainly accommodated by thrust faults affecting the TAP horizon). Since pressure solution phenomena that can possibly affect the Apulian carbonates have not been taken into account, our shortening values should be regarded as conservative estimates. [59] It should be noted that in the two westernmost ramp anticlines the position of the footwall cutoff for both the TAP and the bottom Apulian horizon is to a great extent unconstrained. As a consequence, alternative interpretations of the same seismic reflection data could drive to larger estimates of shortening. In order to highlight these uncertainties, we left a blank area below the modeled antiformal stack (Figure 4). 3.4. High-Angle Fault Systems [60] Several published sections of the SA do not show significant offset of high-angle normal faults, which are difficult to recognize in seismic lines of this area. However, such fault systems, related to both extensional and strikeslip tectonics, are widely documented in surface geology and seismicity. [61] East of the Alburni Mountains (Figure 6), a set of NW-SE normal faults affects the Monte Marzano transitional facies (considered the eastern margin of the Apennine carbonate platform). Some of these faults reflect the present day extensional tectonic field responsible for the Irpinian earthquake (1980), which has been related to the main

11 of 20

TC3005


seismogenic structure, which trends NW-SE and dips at about 60° to NE [Pingue et al., 1988]. [62] The Alburno-Cervati structure, characterized by a constant SW dipping bedding attitude (about 25°), is bounded on each edge by WNW-ESE subvertical, strikeslip, fault systems active in the late Pliocene [Ascione et al., 1992; Berardi et al., 1996]. [63] Finally, the western side of the M. Soprano ridge is cut by a fault system that shows a significant displacement (in the order of thousands of meters).

4. Constraints to the Lithospheric Setting [64] The SA deep structure is not sufficiently constrained using only field and subsurface data. More constraints can be added if our conservative section is framed in the SA geodynamic setting inferred from the available geophysical data sets. [65] Geodynamic interpretations have been proposed that do not consider the subduction as the driving process responsible for the genesis of the Apennines. However, the simple and indisputable observation that about 160– 200 km of crystalline basement (the former substratum of allochthonous units) are missing strongly suggests the westward subduction of the Apulo-Adriatic continental lithosphere under the SA (Figure 7). Several independent geophysical data sets corroborate this conclusion. [66] In the last two decades several reconstructions of the Moho depths have been proposed for the Italian area [among the others: Cassinis, 1983; Wigger, 1984; Nicolich, 1989; Consiglio Nazionale delle Ricerche, 1992; Nicolich, 2001]. An updated interpretation for the SA, based on deep refraction seismic data, has been proposed by Scarascia et al. [1994]. The Apulian crust is about 30 km thick in the Apulian foreland. The Adriatic Moho dips toward SW, at least down to a depth of about 50 km below the Tyrrhenian coast. Along the Tyrrhenian Sea, a different and shallower (at depths of 25– 30 km) Moho (named ‘‘Tyrrhenian’’) has been recognized. According to Calcagnile and Panza [1981] and Consiglio Nazionale delle Ricerche, [1992], the lithospheric thickness in foreland areas varies from 70 km in the northern Adriatic Sea to about 110 km in the Apulian foreland. [67] Positive Bouger anomalies (up to 120 mGal or more [Mongelli et al., 1975; Consiglio Nazionale delle Ricerche, 1992]) and very high heat flow values (up to 140 mW/m2 or more [Cataldi et al., 1995; Della Vedova et al., 2001]) have been measured along the Tyrrhenian margin and in the adjacent Tyrrhenian Sea. These data are consistent with the presence of asthenospheric material in the upper mantle below these regions and, noticeably, below the SA Tyrrhenian side, as suggested by the kinematics of the Apennines-Tyrrhenian Sea system [Doglioni, 1991], by the analysis of the shear waves attenuation [Mele et al., 1997], and by geochemical evidences (helium isotope ratios together with the amount of released gas [Italiano et al., 2000]). Moreover, Mele et al. [1997] infer the occurrence of a high-strength mantle LID below the Adriatic foreland.

TC3005

[68] Mantle tomography models generally show highvelocity anomalies below the Alps, the Northern Apennines and the Calabrian arc. From these anomalies it has been inferred that the Adriatic continental lithosphere and the Ionian oceanic lithosphere are subducting westward almost vertically underneath the Apennines [Spakman, 1990; Spakman et al., 1993; Amato et al., 1993; Piromallo and Morelli, 1997; Amato et al., 1998; Lucente et al., 1999]. On the basis of the weaker high-velocity anomalies detected in their tomographic models below the Southern Apennines, some authors proposed different mantle reconstructions for this area. Amato et al. [1993] and Lucente et al. [1999] speculate the presence of a ‘‘slabless window’’ below the southern Apennines (at depths shallower than 250 km), while Spakman [1990] and Spakman et al. [1993] suggest the existence of a continuous high-velocity slab at depth, detached from its upper part except below the Calabrian arc. These interpretations were also based on the absence of subcrustal subduction-related seismicity below the SA. [69] Detailed tomographic studies focused on the southern Apennines [De Gori et al., 2001] have pointed out the presence of an almost continuous subvertical high-velocity body. This feature, extending from depths of 65 km down to 285 km, has been interpreted as Adriatic lithosphere in subduction beneath the SA. According to De Gori et al. [2001], the absence in previous tomography images of clear high-velocity anomalies below the SA was due to their poor resolution at shallow depths. [70] The recognition of a continuous slab in local tomography models in areas characterized by absence of subcrustal seismicity, such as the SA, is problematic. These two apparently conflicting observations can be reconciled, as follows. Carminati et al. [2002] have suggested that the absence of a Benioff-plane below the Apennines is not the direct evidence for the absence of a subducting slab. It could be rather related to the continental composition of the subducting Adriatic lithosphere. According to these authors, subducting continental lithosphere is expected to have ductile rather than brittle behavior (and to accommodate deformation aseismically rather then seismically) at much shallower levels with respect to oceanic lithosphere. [71] In Italy, the acquisition of magnetic or aeromagnetic data brought to the release of several maps, sometimes showing significant differences because of different acquisition methodology or data processing [e.g., Arisi Rota and Fichera, 1987; Pinna, 1987; Molina et al., 1994; Servizio Geologico Nazionale, 1994; Chiappini et al., 2000]. Since magnetic data do not provide a unique solution to the deep structure of the SA they are not considered in our geodynamic model. [72] On the basis of the available geophysical data described above, the large-scale lithospheric setting has been reconstructed and the geometry of the subducting Apulian slab has been portrayed on our regional cross section (Figure 4). The main features are represented by (1) the flexure of the subducting Apulian lithosphere below the SA, with the slab top deeper than 100 km below the Tyrrhenian coastline, and (2) the presence of a hot mantle

12 of 20

TC3005


wedge underlying a new ‘‘young’’ and hot Moho in the western side of the SA accretionary prism.

5. Geodynamic Model [73] Surface geology displays a progressive northeastward migration of the SA accretionary wedge, with the consequent diachronism of the onset of the foreland ramp/ foredeep sedimentation (progressively younger toward the Adriatic foreland) followed by the extensional faults crosscutting the previously formed contractional structures. All these features are interpreted to have formed in response to the flexure-hinge retreat of the westward subduction of the Apulo-Adriatic continental lithosphere [see, e.g., Malinverno and Ryan, 1986; Patacca et al., 1990; Doglioni, 1991; Doglioni et al., 1996, 1999]. This subduction retreat model is consistent with the observation that the Tyrrhenian Moho is shallower than the Adriatic Moho. The retreating slab is replaced by asthenospheric materials (Figure 4) in a context of no, or very low, plate convergence. In this scenario, the Tyrrhenian Moho is a newly forming crustasthenosphere boundary. The lithospheric mantle originally underlying the Tyrrhenian crust has been replaced by hot asthenosphere since the beginning of the slab rollback process. This view agrees with the occurrence of high heat flow densities in the Tyrrhenian area and in the western part of the Italian peninsula. On the contrary the Adriatic Moho likely formed during the Mesozoic rifting stages. Low heat flow densities are measured above this lithosphere, which was already structured in the Mesozoic. [74] The lithospheric mantle of the Apulo-Adriatic continental plate subducts together with a large part of the crust, apart the sedimentary cover incorporated in the SA accretionary prism [Doglioni, 1991; Doglioni et al., 1999], as suggested by the geochemical signatures in the calcalkaline magmas of the volcanic arc [e.g., Peccerillo, 1985; Serri et al., 1993]. Although the subduction of continental crustal rocks is normally considered an unlikely process, because of the buoyancy of crust with respect to the adjacent mantle, the occurrence of crustal slices bearing high-pressure, lowtemperature metamorphism (e.g., in the Sesia Lanzo Zone [Dal Piaz et al., 1971; Compagnoni et al., 1977]) shows that the process can occur. Moreover, we suggest that, rather than negative buoyancy of the slab (slab pull), the main mechanism which drives the Apenninic westward subduction is the westward relative motion of lithosphere relative to the mantle [Doglioni, 1991]. Consequently, the continental crust is interpreted to subduct because of the eastward push of the asthenosphere rather than to its gravitational instability, as required by slab-pull models. The Adriatic plate subduction is confirmed by the shape of the main active detachment level which plunges steadily westward to accommodate the plate flexure. The detachment constitutes the upper boundary for the subducting plate. [75] The shortening in the accretionary wedge can be related to the shear between the downgoing and retreating lithosphere and the eastward flow of the asthenosphere compensating the subduction rollback. The shear is transferred upward and causes the peeling-off of the cover from

TC3005

the foreland lithosphere that enters the subduction zone [Doglioni et al., 1999]. The roll-back of the subduction hinge appears to have slowed during the late Pleistocene in response to the interference of the thick continental Apulian swell with the front of the SA accretionary prism [Doglioni et al., 1994]. [76] It should be noticed that in the SA almost no sediments affected by regional metamorphism crop out. This observation suggests that outcropping sediments remained at shallow structural levels. The sedimentary cover of the Adriatic/Apulian plate might have followed typical PTt paths associated with orogens because of west directed subduction zones, as suggested by Doglioni et al. [1999], where rocks are first covered with overburden while transported to relatively shallow depth, and finally uplifted or ‘‘exhumed’’. [77] However, north of M. Pollino ridge in Figure 2, Bonardi et al. [1993] described greenschist facies metasediments (Frido Unit). The upper part of this unit, comprising phyllites, quartzites and metalimestones, has been dated as late Oligocene. The metasediments of the Frido unit could be interpreted both as Alpine relics or related to the Apennine subduction, in particular to the earliest stages of the wedge formation, when middle-high pressure, lowtemperature metamorphism can develop. These crystalline rocks can produce positive magnetic anomalies although they do not represent basement rocks accreted from below the basal de´collement of the Apennine accretionary prism (which can be located near the basement/cover interface).

6. Speculations on the Structural Styles [78] Two main components can be distinguished in the SA accretionary wedge: the allochthonous units and the buried Apulian antiformal stack. The former are characterized by a thin-skinned tectonic style. Indeed, the middle Miocene-early Pliocene structuring of the SA was not accompanied by any involvement of the crystalline basement. The structure of the latter is not constrained by the available geological and the geophysical data and represents the main uncertainty about the deep structure of the SA. [79] As a consequence, two main groups of structural interpretations have been proposed. They differ because of the shortening within the buried Apulian thrust units and the degree of involvement of the crystalline basement underlying the Apulian carbonates (Figures 1a and 1b). [80] According to the first group, the Apulian basement is largely deformed and forms a huge wedge within the SA accretionary prism [Casero et al., 1988; Mazzoli et al., 2000; Menardi Noguera and Rea, 2000; Sciamanna et al., 2000]. In this scenario, the shortening at TAP level is small, in the order of 15 – 25 km [e.g., Mazzoli et al., 2000; Menardi Noguera and Rea, 2000]. [81] According to the second group, the SA are made up of completely rootless sedimentary nappes. Two different geometries have been hypothesized for the Apulian crystalline basement. In the first case, the basement remains undeformed and dips to the west under the thrust belt [e.g., Mostardini and Merlini, 1986; Marsella et al., 1995;

13 of 20

TC3005


TC3005

Figure 8. Thick-skinned tectonics interpretation of the SA deep structure (cross section location in Figure 2). The main thrusts offsetting the top Apulian carbonates are interpreted as deeply rooted. The resulting cross sections show three slices of the Apulian crystalline basement with shortening estimated at the top Apulian horizon of about 20 km. To place the basement slices and to permit their deformation, the present-day position of the subducting slab needs to be shifted more than 50 km westward. The modified structure no longer matches the crustal and lithospheric architecture constrained by the available geophysical data (compare Figures 4 and 8). The tectonic implications of this model show some inconsistencies with respect to well-documented geological features of the SA (see text). Mazzotti et al., 2000; Patacca and Scandone, 2001a]. In the second case the undeformed basement top follows the curved geometry of the subducting Apulian slab [Doglioni et al., 1996]. In a thin-skinned model, key-bed balancing of the TAP horizon yields a shortening of at least 110– 120 km [e.g., Mazzotti et al., 2000]. [82] Although, the available geological and geophysical constraints cannot resolve uncertainties about the deep structure of the SA, it is at least possible to compare the main tectonic and geodynamic implications of these alternative interpretations, (as it has been done in other sectors of the Apennine accretionary prism [e.g., Ghisetti et al.,

1993]). To achieve this, two completely different structural models have been developed, using as a base the large-scale model of Figure 4. In the thick-skinned scenario, the blank area below the Apulian antiformal stack has been filled with basement slices. In the thin-skinned model, only thrust sheets made up of sedimentary cover, derived from the deformation of the Apulian domain, have been introduced. 6.1. Model 1: Thick-Skinned Tectonics [83] One of the conceivable thick-skinned models is represented in Figure 8. The main thrusts offsetting the

14 of 20

TC3005


Figure 9. The Thick-skinned scenario implies a post-early Pliocene transition from thin- to thick-skinned tectonic style that may have occurred through a crustal down-section propagation of the main detachment during the structuring of the Apulian domain. The mismatch between total shortening at top Apulian level (20 km) and the late Pliocene-early Pleistocene advance of allochthonous nappes (30 –40 km) should be explained in the following way: (a) up to the boundary between early and late Pliocene, a very long flat (about 80 km) permitted the advance of allochthonous units above an essentially undeformed downflexing Apulian domain; (b) in late Pliocene-early Pleistocene times, the detachment above the Apulian carbonates should have been active, at least partially, during the propagation of the crustal down-section and the structuring of the Apulian thrust sheets. top Apulian carbonates are interpreted as being deeply rooted and, consequently, the resulting cross section shows three slices of the Apulian crystalline basement. In this scenario, shortening at TAP level amounts to about 20 km. Alternative models foresee the presence of a thick continental crust backstop made up by the Apulian crystalline basement delaminated from the subducting apulo-adriatic lithospheric mantle [e.g., Roure et al., 1991; Ziegler and Roure, 1996], envisage the activation of an ensialic shear

TC3005

zone responsible for the deformation of the Apulian basement following the subduction of the basement of the Lagonegro basin [e.g., Menardi Noguera and Rea, 2000], or invoke the inversion of Permo-Triassic extensional faults underlying the Apulian carbonates [e.g., Shiner et al., 2004]. [84] However, since a thin-skinned tectonic style during the middle Miocene-early Pliocene tectonic accretion of the allochthonous units is strongly supported by the available geological and geophysical data (see section 3.2), a transition from thin- to thick-skinned tectonic style is necessarily required by whatever thick-skinned model. Being initially the main detachment above the crystalline basement of the subducting lithosphere, this transition must have occurred with the activation of a deep crustal shear zone, contemporaneous with the deformation of the Apulian carbonate platform domain in the late Pliocene – early Pleistocene. In our reconstruction (Figure 9), this transition has been represented in a qualitative and simplistic way with a crustal down-section propagation of the main detachment. The down section propagation might have reached the ductile and layered lower crust where, in the case of a continental lithosphere (such as the Apulian lithosphere), major potential decoupling horizons could develop. Eventually, this crustal shear zone should have propagated upward, below the axial zone of the chain, developing three main splays. The necessity of the crustal down-section propagation of the main detachment still persists even if a slab break-off model for the SA, suggested by some authors [e.g., Spakman, 1990; Spakman et al., 1993] but not confirmed by recent tomographic model [e.g., De Gori et al., 2001], is accepted. In fact, the slab break-off below the SA should have occurred starting from the early Pleistocene (between 1.4 and 0.8 Ma [van der Meulen et al., 1998]), once the main contractional deformation of the Apulian carbonates had already taken place. [85] It should be noted that, although the crustal downsection propagation of the main detachment may seem to a certain extent unrealistic, the transition from a thin- to a thick-skinned tectonic style is implicitly or explicitly assumed, without any discussion of its tectonic implications, in several recently published papers proposing a thickskinned interpretation for the buried Apulian thrust units [e.g., Ziegler and Roure, 1996; Mazzoli et al., 2000; Menardi Noguera and Rea, 2000; Sciamanna et al., 2000; Speranza and Chiappini, 2002]. For these reasons, the tectonic implications of a thick-skinned model need to be carefully evaluated. [86] According to the geodynamic model described in section 5, the propagation of the crustal shear zone should be accompanied by a roll-back of the subducting plate (Figures 8 and 9). Therefore, to emplace the basement slices and to permit their deformation, the present day position of the subducting slab needs to be shifted more than 50 km westward (being further west at the onset of thrusting in late Pliocene). The modified location of the slab would, however, no longer match the crustal and lithospheric architecture provided by the available geophysical constraints (compare Figures 4 and 8). As an example, along the

15 of 20

TC3005


Tyrrhenian side of the SA there is no more space for the asthenospheric wedge and it becomes almost impossible to reconcile the position of the subducting slab with tomographic data [e.g., De Gori et al., 2001]. [87] Other relevant implications concern the uplift distribution in the SA that could be predicted in case of late Pliocene-Pleistocene development of basement slices tens of kilometers thick. Although an exact calculation is beyond the aims of this study, we have a tried to evaluate the ‘‘uplift of rock’’ (in the sense of England and Molnar [1990]) predicted by the thick-skinned model of Figure 8. [88] On the basis of our cross section, a cumulative vertical throw of at least 12 km in the last 3.5 Myr can be estimated for the thrust faults responsible for the generation of the basement slices. From the deflection profile built by Royden [1988] using the base of the Pliocene sediments, about 8 – 9 km of flexural subsidence can be estimated for the top of Apulian carbonates, now located on the crest of the antiformal stack. Since the flexure of the Apulian plate trough time is not constrained in the SA, we have assumed a self-similar flexural behavior. With these assumptions about 6 km of flexural subsidence can be estimated from the beginning of late Pliocene to recent. From these values, approximate ‘‘uplift of rock’’ rates in excess of 1.7 mm/yr could be expected. [89] Using both geomorphological observations and stratigraphical and structural data, Schiattarella et al. [2003] have calculated regional ‘‘uplift of rock’’ rates of about 0.6 mm/yr for the last 2 Myr in the axial zone of chain (roughly corresponding in our cross section to the segment comprised between S. G. Magno 1 e S. Fele 1 wells). This value is in good agreement with data from other areas of the SA [e.g., Amato and Cinque, 1999; Amato, 2000]. Remarkably, the ‘‘uplift of rock’’ rates predicted by the thickskinned model significantly exceed the measured regional uplift rates. [90] Another main implication of this structural interpretation is the small total shortening at TAP level: no more than about 20 km in our cross section but even less in other reconstructions (e.g., about 15 km in the work of Mazzoli et al. [2000]). This cannot be reconciled with the late Pliocene-early Pleistocene advance of allochthonous nappes (not less than 30 –40 km [Sciamanna et al., 2000; Patacca and Scandone, 2001b]). [91] This significant discrepancy requires a discussion on the processes responsible for the advancement of nappes in this thick-skinned thrust tectonic scenario (unless it is assumed that the allochthonous nappes should have advanced by processes other than thrusting, e.g., gravity sliding, that do not seem to be supported by the available data). [92] First of all, it should be postulated that, up to the boundary between early and late Pliocene, a very long (not less than 70– 80 km) flat permitted the displacement of allochthonous units above an essentially undeformed downflexing Apulian domain (Figure 9a). In late Pliocene-early Pleistocene times, to reconcile the gap between the estimated shortening at TAP level (about 20 km) and allochthonous advance (30 –40 km), the detachment above the Apulian

TC3005

carbonates should have been active, at least partially, also during the propagation of the crustal down-section and the deformation of the Apulian thrust sheets (Figure 9b). The reliability of similar kinematics needs to be better analyzed and verified from a mechanical and rheological perspective. [93] Moreover, it should be noted that shortening in an accretionary prism should equal the amount of subduction (unless it is an erosional margin, but there are no evidences for tectonic erosion in the SA). The thick-skinned model would predict a total shortening much smaller than the imaged subduction (e.g., De Gori et al., 2001]. [94] The discovery in the SA of relevant oil fields provides other stimulating opportunities to cross-check the main implication of structural models against known geochemical, thermal and maturity data. Sciamanna et al. [2000] developed a 2D thermal and geochemical modeling along a SW-NE striking geological profile that crosses the Val d’Agri area, where major oil discoveries are located (about 50 km southeastward with respect to regional cross section analyzed in our study). This modeling, calibrated using present-day temperature and vitrinite reflectance derived from well data, was performed adopting a thickskinned structural interpretation. According to their modeling results, the small displacement assumed in their cross section between the innermost Apulian thrust sheets (few kilometers) cannot explain the observed difference in maturity. They conclude that the thrust sheets presently juxtaposed should have been originally more distant (i.e., the shortening greater, in the order of tens of kilometers) than predicted by thick-skinned models. 6.2. Model 2: Thin-Skinned Tectonics [95] The cross section of Figure 10 shows our thinskinned tectonics interpretation of the SA deep structure. It is known that only the TAP horizon is well defined by seismic and well data while the bottom of the Apulian carbonates is usually poorly imaged although clearly involved in thrusting. The conservative regional cross section of Figure 4 has been completed honoring the TAP attitude and assuming the basement not involved at all (i.e., it has been fully subducted). The main thrusts recognized at the TAP level have been extrapolated westward, flattening at depth. The resulting setting is characterized by the partial overlap of three main thrust sheets, made up of both the Apulian carbonates and the underlying Permo-Triassic clastics. The lowest thrust body becomes subhorizontal at depth, above the Tyrrhenian Moho. Applying key-bed balancing techniques at the TAP level, the total shortening of the buried Apulian thrust sheets is estimated to be not less than 90 km. Of course, the proposed structure is just one of the possible solutions and, likely, a simplistic one. Whatever the adopted structural solution, the total shortening of the Apulian carbonates units in the frame of a thin-skinned tectonic scenario should be about five times larger than in a thick-skinned one. [96] In some of the proposed thin-skinned interpretations, whose deeper part was constrained by magnetic modeling, the undeformed west dipping basement is extended west-

16 of 20

TC3005


TC3005

Figure 10. Thin-skinned tectonics interpretation of the SA deep structure (cross section location in Figure 2). The main thrusts recognized at TAP level have been extrapolated westward, flattening at depth. The resulting setting is characterized by the partial overlap of three main thrust sheets, made up by both the Apulian carbonates and the underlying Permo-Triassic clastics. The lowest thrust body becomes subhorizontal at depth, above the Tyrrhenian Moho. The total shortening of the Apulian thrust units should be at least than 90 km, while the total shortening of the allochthonous units (including the Apennine carbonate platform, the Lagonegro-Molise basin, and the Apulian carbonate platform) is estimated to be not less than 280 –300 km. ward below the SA to the Tyrrhenian coastline [e.g., Mostardini and Merlini, 1986], in conflict with available geophysical data. In our model, the large-scale geodynamic framework, provided by the geophysical data, has been respected (Figure 10). [97] Remarkably, the ca. 35– 40 km late Pliocene-early Pleistocene front advance of the allochthonous nappes can be easily justified in a thin-skinned model, characterized

by more than 90 km of shortening at the TAP level, assuming that the allochthonous nappes advance was contemporaneous with the deformation of the underlying Apulian units. In this scenario, at the early/late Pliocene boundary, the allochthonous nappes were already partially emplaced over the flexured Apulian domain. The allochthonous nappes front could have been positioned no more than few tens of kilometers eastward with

17 of 20

TC3005


respect to the active thrusts affecting the underlying Apulian domain. As a consequence, during the piling up of the Apulian antiformal stack, at least part of the displacement responsible for the emplacement of each Apulian thrust sheet could have been transferred upward to the main basal detachment of the allochthonous nappes, causing their eastward transport. [98] For a thin-skinned scenario, during the late Plioceneearly Pleistocene, the frontal accretion occurred with the incorporation into the wedge of relatively thin thrust sheets (about 8500 – 9000 m thick) made up of sedimentary cover detached from the subducting and rolling-back Apulian slab. As already stated this process took place with no or very low plate convergence at the transition between an expanding back arc basin and a subducting and rolling-back Apulo-Adriatic slab; in this respect, the SA significantly differ from typical collisional orogens [Doglioni, 1991; Doglioni et al., 1999]. As a consequence, the SA accretionary wedge and the induced topography, rather than growing vertically, developed horizontally moving as a fast wave toward the east, at rates of 10– 30 mm/yr or more [Patacca et al., 1990; Gueguen et al., 1998]. The average expected uplift rates are generally lower than 1 mm/yr [Doglioni et al., 1999], with peak values reached only for a short time span in which the tectonic wave crosses an area. These conclusions are in good agreement with the data available in the SA (already described in the previous section). [99] Displacements among the internal Apulian thrust sheets in the order of 30 km, predicted by our thin-skinned model, could explain the observed maturity trends in the Apulian carbonates inferred from vitrinite reflectance data, as suggested by Sciamanna et al. [2000]. [100] Moreover, the thin-skinned interpretation of the SA is also supported by global tectonics models. Comparing orogens worldwide [Royden and Burchfiel, 1989; Doglioni, 1992; Royden, 1993], two different classes can be recognized, which display pronounced and systematic characters. The orogens developing on ‘‘east’’ directed subductions (e.g., Alps, Himalayas) show a conjugate retrobelt, high structural and topographic elevation, two shallow foredeeps (with low subsidence rate), and outcropping rocks characterized by medium- to high-grade metamorphism, testifying high exhumation rates. The orogens developing on ‘‘west’’ directed subductions (e.g., Carpathians, Barbados) are instead characterized by the occurrence of a back arc basin, low structural and topographic elevation, one deep foredeep (with high subsidence rates), low to none metamorphism. Generally, in the first group the entire crust and the upper mantle are deeply involved in thrusting, whereas in the second group (which includes the Apennines) the accretionary prism is mainly formed by the tectonic stacking of the sedimentary cover [Doglioni et al., 1999].

7. Conclusion [101] A review of the existing geological and geophysical data set highlights that the available data do not

TC3005

fully constraint the deep structure of the SA. The main uncertainties are as follows: (1) the shortening of the SA accretionary prism and, in particular, within the buried Apulian thrust units and (2) the degree of involvement of the crystalline basement, underlying the Apulian carbonates, in the accretionary wedge. Given these uncertainties, we have modeled the deep structure of the SA adopting both a thick-skinned (Apulian basement deeply involved) and a thin-skinned style (Apulian basement not involved at all), as both these models are admissible from a geometric point of view. The robustness of these models, which must be regarded as two end-members, has been cross-checked against well-documented tectonic, geophysical, and geochemical features of the SA. [102] The thick-skinned interpretation necessarily requires a transition from thin- to thick-skinned tectonic style, since the available data document a thin-skinned deformation during the middle Miocene-early Pliocene tectonic accretion of the allochthonous units (e.g., Apennine carbonate platform and Lagonegro-Molise basin). This post-early Pliocene transition may have occurred through a crustal down-section propagation of the main detachment during the contractional deformation of the Apulian domain. The analysis of the main implications of this thick-skinned model revealed some inconsistencies with both the crustal and lithospheric setting, deduced from geophysical data, and the known Plio-Pleistocene vertical motions. Moreover, shortening at TAP level of about 20 km, implied by this model, cannot explain both the observed maturity trends in the Apulian carbonates, inferred from vitrinite reflectance data, and the late Pliocene-early Pleistocene advance of allochthonous nappes. On the contrary, the thin-skinned interpretation is generally more coherent with the available constraints. [103] On the basis of these results, it emerges that thinskinned interpretations of the SA deep structure should be preferred and that the involvement of slices of Apulian crystalline basement, tens of kilometers thick, is unlikely. However, since our models represent end-members, it remains possible that minor involvement of the Apulian basement could have occurred just at its top (i.e., the upper few kilometers). [104] In a thin-skinned model the total shortening of the Apulian thrust units, derived applying key-bed balancing techniques at TAP level, should be at least 90 km. The total shortening of the allochthonous units in the SA (including thrust sheets derived from the deformation of the Apennine carbonate platform, the Lagonegro-Molise basin, and the Apulian carbonate platform), based on the estimate of the original widths of these paleogeographic domains, should be not less than 280– 300 km, a value relatively consistent and more coherent with the present length of the slab beneath the SA.

[105] Acknowledgments. Discussions with Simone Sciamanna, Paolo Scandone, and Carlo Nicolai were very helpful. Andy Nicol, Sharon Mosher, Adrian Pfiffner, and an anonymous reviewer are thanked for constructive criticism. Research supported by ASI, and COFIN 2001.

18 of 20

TC3005


TC3005

References Amato, A. (2000), Estimating Pleistocene tectonic uplift rates in the southeastern Apennines (Italy) from erosional land surfaces and marine terraces, in Geomorphology, Human Activity and Global Environmental Change, Publ. Int. Assoc. Geomorphol., vol. 9, edited by O. Slaymaker, pp. 67 – 87, John Wiley, Hoboken, N. J. Amato, A., and A. Cinque (1999), Erosional landsurfaces of the Campano-Lucano Apennines (S. Italy); Genesis, evolution, and tectonic implications, Tectonophysics, 315, 251 – 267. Amato, A., B. Alessandrini, G. B. Cimini, A. Frepoli, and G. Selvaggi (1993), Active and remnant subducted slabs beneath Italy: Evidence from seismic tomography and seismicity, Ann. Geofis., 36, 201 – 214. Amato, A., et al. (1998), Passive seismology and deep structure in central Italy, in Geodynamics of Lithosphere and Earth’s Mantle: Seismic Anisotropy as a Record of the Past and Present Dynamic Processes, edited by J. Plomerova, R. C. Liebermann, and V. Babuska, Pure Appl. Geophys., 151, 479 – 493. Arisi Rota, F., and R. Fichera (1987), Magnetic interpretation related to geo-magnetic provinces: The Italian case history, Tectonophysics, 138, 179 – 196. Ascione, A., A. Cinque, and M. Tozzi (1992), La Valle del Tanagro (Campania): Una depressione strutturale ad evoluzione complessa, Stud. Geol. Camerti, Spec. Vol. 1992, 209 – 219. Balduzzi, A., R. Casnedi, U. Crescenti, and M. Tonna (1982a), Il Plio-Pleistocene del sottosuolo del Bacino pugliese (Avanfossa appenninica), Geol. Romana, 21, 1 – 28. Balduzzi, A., R. Casnedi, U. Crescenti, F. Mostardini, and M. Tonna (1982b), Il Plio-Pleistocene del sottosuolo del Bacino lucano (Avanfossa appenninica), Geol. Romana, 21, 89 – 111. Berardi, F., G. De Rosa, and M. Tozzi (1996), Vincoli Strutturali di Superficie per una Ricostruzione Geometrica e Cinematica del Massiccio dei Monti Alburni (Appennino Meridionale), Mem. Soc. Geol. Ital., 51, 13 – 21. Bonardi, G., F. O. Amore, G. Ciampo, P. De Capoa, P. Miconnet, and V. Perrone (1988), Il Complesso Liguride auct.: Stato delle conoscenze e problemi aperti sulla sua evoluzione pre-appenninica ed i suoi rapporti con l’Arco Calabro, Mem. Soc. Geol. Ital., 41, 17 – 35. Bonardi, G., P. De Capoa, B. Fioretti, and V. Perone (1993), L’Age des metacalcaires de l’unite du Frido (region Calabro-Lucanienne, Italie) et ses implications geodynamiques, C. R. Acad. Sci., 317(7), 955 – 962. Bonardi, G., P. De Capoa, B. Fioretti, and V. Perone (1994), Some remarks on the Calabria-Peloritani arc and its relationship with the southern Apennines, Boll. Geof. Teor. Appl., 36(141 – 144), 483 – 492. Bosellini, A., C. Neri, and V. Luciani (1993), Guida ai carbonati cretaceo-eocenici di scarpata e bacino del Gargano (Italia Meridionale), Ann. Univ. Ferrara Nuova Ser., 4, 1 – 77. Butler, R. H. W. (1987), Thrust sequences, J. Geol. Soc. London, 144, 619 – 634. Calcagnile, G., and G. F. Panza (1981), The main characteristics of the lithosphere-asthenosphere system in Italy and surrounding regions, Pure Appl. Geophys., 119, 865 – 879. Carbone, S., and F. Lentini (1990), Migrazione neogenica del sistema catena-avampaese nell’Appennino meridionale: Problematiche paleogeografiche e strutturali, Riv. Ital. Paleontol. Stratigr., 96, 271 – 296. Carbone, S., S. Catalano, F. Lentini, and C. Monaco (1988), Le unita’ stratigrafico-strutturali dell’Alta Val d’Agri (Appennino Lucano) nel quadro dell’evoluzione del sistema catena-avanfossa, Mem. Soc. Geol. Ital., 41, 331 – 341.

Carbone, S., S. Catalano, S. Lazzari, F. Lentini, and C. Monaco (1991), Presentazione della carta geologica del Bacino del Fiume Agri (Basilicata), Mem. Soc. Geol. Ital., 47, 129 – 143. Carminati, E., F. Giardina, and C. Doglioni (2002), Rheological control on subcrustal seismicity in the Apennines subduction (Italy), Geophys. Res. Lett., 29(18), 1882, doi:10.1029/2001GL014084. Casero, P., F. Roure, L. Endignoux, I. Moretti, C. Muller, L. Sage, and R. Vially (1988), Neogene geodynamic evolution of the southern Apennines, Mem. Soc. Geol. Ital., 41, 109 – 120. Casnedi, R., U. Crescenti, C. D’amato, F. Mostardini, and U. Rossi (1981), Il Plio-Pleistocene del sottosuolo molisano, Geol. Romana, 20, 1 – 42. Cassinis, R. (1983), Seismicity and crustal structure in the Italian region: A preliminary zoning, Boll. Geofis. Teor. Appl., 25(97), 3 – 26. Catalano, R., C. Doglioni, and S. Merlini (2001), On the Mesozoic Ionian basin, Geophys. J. Int., 144, 49 – 64. Catalano, S., C. Monaco, L. Tortorici, W. Paltrinieri, and N. Steel (2004), Neogene-Quaternary tectonic evolution of the southern Apennines, Tectonics, 23, TC2003, doi:10.1029/2003TC001512. Cataldi, R., F. Mongelli, P. Squarci, L. Taffi, G. Zito, and C. Calore (1995), Geothermal ranking of the italian territory, Geothermics, 24, 115 – 129. Cello, G., and S. Mazzoli (1999), Apennine tectonics in southern Italy: A review, J. Geodyn., 27, 191 – 211. Cello, G., L. Tortorici, N. Martini, and W. Paltrinieri (1989), Structural styles in the frontal zones of the southern Apennines, Italy: An example from the Molise district, Tectonics, 8, 753 – 768. Channell, J. E. T., B. D’Argenio, and F. Horvath (1979), Adria, the African promontory, in Mesozoic Mediterranean paleogeography, Earth Sci. Rev., 15, 213 – 292. Chiappini, M., A. Meloni, E. Boschi, O. Faggioni, N. Beverini, C. Carmisciano, and I. Marson (2000), Shaded relief magnetic anomaly map of Italy and surrounding marine areas, Ann. Geofis., 43, 983 – 989. Compagnoni, R., G. V. Dal Piaz, J. C. Hunziker, G. Gosso, B. Lombardo, and P. F. Williams (1977), High pressure-low temperature metamorphism of the oceanic and continental crust in the Western Alps, Rend. Soc. Ital. Miner. Petr., 33, 281 – 334. Consiglio Nazionale delle Ricerche (1992), Structural model of Italy and gravity map, scale 1:500,000, Progetto Finalizzato Geodin, vol. 3, Quad. Ric. Sci., 114. Dal Piaz, G. V., J. Hunziker, and G. Martinotti (1971), La zona Sesia Lanzo e l’evoluzione tettonico-metamorfica delle Alpi nordoccidentali interne, Mem. Soc. Geol. Ital., 11, 433 – 466. D’Argenio, B., T. Pescatore, and P. Scandone (1975), Structural pattern of the Campania-Lucania Apennines, in Structural Model of Italy, Quad. Ric. Sci., 90, 313 – 327. De Gori, P., G. P. Cimini, C. Chiarabba, G. De Natale, C. Troise, and A. Deschamps (2001), Teleseismic tomography of the Campanian volcanic area and surrounding Apenninic belt, J. Volcanol. Geotherm. Res., 109, 55 – 75. Della Vedova, B., S. Bellani, G. Pellis, and P. Squarci (2001), Deep temperatures and surface heat flow distribution, in Anatomy of an Orogen: The Apennines and Adjacent Mediterranean Basins, edited by G. B. Vai and I. P. Martini, pp. 65 – 76, Springer, New York. Dercourt, J., L. E. Ricou, and B. Vrielynck (Eds.) (1992), Atlas Tethys Palaeoenvironmental Maps, 307 pp., Elsevier, New York. Doglioni, C. (1991), A proposal of kinematic modelling for W-dipping subductions—Possible applications to the Tyrrhenian-Apennines system, Terra Nova, 3, 423 – 434.

19 of 20

Doglioni, C. (1992), Main differences between thrust belts, Terra Nova, 4, 152 – 164. Doglioni, C., F. Mongelli, and P. Pieri (1994), The Puglia uplift (SE Italy): An anomaly in the foreland of the Apenninic subduction due to buckling of a thick continental lithosphere, Tectonics, 13, 1309 – 1321. Doglioni, C., P. Harabaglia, G. Martinelli, F. Mongelli, and G. Zito (1996), A geodynamic model of the southern Apennines accretionary prism, Terra Nova, 8, 540 – 547. Doglioni, C., F. Mongelli, and G. P. Pialli (1998), Boudinage of the Alpine belt in the Apenninic back-arc, Mem. Soc. Geol. Ital., 52, 457 – 468. Doglioni, C., P. Harabaglia, S. Merlini, F. Mongelli, A. Peccerillo, and C. Piromallo (1999), Orogens and slabs vs. their direction of subduction, Earth Sci. Rev., 45, 167 – 208. England, P., and P. Molnar (1990), Surface uplift, uplift of rocks, and exhumation of rocks, Geology, 18, 1173 – 1177. Ghisetti, F., M. Barchi, A. W. Bally, I. Moretti, and L. Vezzani (1993), Conflicting balanced structural sections across the central Apennines (Italy): Problems and implications, in Generation, Accumulation and Production of Europe’s Hydrocarbons III, edited by A. M. Spencer, Spec. Publ. Eur. Assoc. Pet. Geosci., vol. 3, pp. 219 – 231, Springer, New York. Gueguen, E., C. Doglioni, and M. Fernandez (1998), On the post-25 Ma geodynamic evolution of the western Mediterranean, Tectonophysics, 298, 259 – 269. Italiano, F., M. Martelli, G. Martinelli, and P. M. Nuccio (2000), Geochemical evidence of melt intrusions along lithospheric faults of the southern Apennines, Italy: Geodynamic and seismogenic implications, J. Geophys. Res., 105(B6), 13,569 – 13,578. Kastens, K., et al. (1988), ODP Leg 107 in the Tyrrhenian Sea: Insights into passive margin and back-arc basin evolution, Geol. Soc. Am. Bull., 100, 1140 – 1156. Knott, S. D. (1987), The Liguride complex of southern Italy¯—A Cretaceous to Paleogene accretionary wedge, Tectonophysics, 142, 217 – 226. Knott, S. D. (1994), Structure, kinematics and metamorphism of the Liguride complex, southern Apennines, Italy, J. Struct. Geol., 16, 1107 – 1120. Lentini, F., S. Catalano, and S. Carbone (1996), The external thrust system in southern Italy: A target for petroleum exploration, Pet. Geosci., 2, 333 – 342. Lentini, F., S. Carbone, A. Di Stefano, and P. Guarnieri (2002), Stratigraphical and structural constraints in the Lucanian Apennines (southern Italy): Tools for reconstructing the geological evolution, J. Geodyn., 34, 141 – 158. Lucente, F. P., C. Chiarabba, G. B. Cimini, and D. Giardini (1999), Tomographic constraints on the geodynamic evolution of the italian region, J. Geophys. Res., 104(B9), 20,307 – 20,327. Malinverno, A., and W. B. F. Ryan (1986), Extension in the Tyrrhenian sea and shortening in the Apennines as a result of arc migration driven by sinking of lithosphere, Tectonics, 5, 227 – 245. Mariotti, G., and C. Doglioni (2000), The dip of the foreland monocline in the Alps and Apennines, Earth Planet. Sci. Lett., 181, 191 – 202. Marsella, E., A. W. Bally, G. Cippitelli, B. D’argenio, and G. Pappone (1995), Tectonic history of the Lagonegro domain and southern Apennine thrust belt evolution, Tectonophysics, 252, 307 – 330. Mazzoli, S., S. Corrado, M. De Donatis, D. Scrocca, R. Butler, D. Di Bucci, G. Naso, C. Nicolai, and V. Zucconi (2000), Time and space variability of ‘‘thin-skinned’’ and ‘‘thick-skinned’’ thrust tectonics in the Apennines (Italy), Rend. Fis. Accad. Lincei, 11(1), 5 – 39. Mazzoli, S., S. Barkham, G. Cello, R. Gambini, L. Mattioni, P. Shiner, and E. Tondi (2001), Recon-

TC3005


struction of continental margin architecture deformed by the contraction of the Lagonegro Basin, southern Apennines, Italy, J. Geol. Soc., 158, 309 – 319. Mazzotti, A., E. Stucchi, G. Fradelizio, L. Zanzi, and P. Scandone (2000), Seismic exploration in complex terrains: A processing experience in the southern Apennines, Geophysics, 65, 1402 – 1417. Mazzotti, A., E. Patacca, and P. Scandone (Eds.) (2005), The CROP 04 transect – southern Apennines, Boll. Soc. Geol. Ital., in press. Mele, G., A. Rovelli, D. Seber, and M. Baranzagi (1997), Shear wave attenuation in the lithosphere beneath Italy and surrounding regions: Tectonic implications, J. Geophys. Res., 102(B6), 11,863 – 11,875. Menardi Noguera, A., and G. Rea (2000), Deep structure of the Campanian-Lucanian Arc (southern Apennine, Italy), Tectonophysics, 324(4), 239 – 265. Merlini, S., G. Cantarella, and C. Doglioni (2000), On the seismic profile Crop M5 in the Ionian Sea, Boll. Soc. Geol. Ital., 119, 227 – 236. Miconnet, P. (1988), Evolution mesozoique du secteur de Lagonegro, Mem. Soc. Geol. Ital., 41, 321 – 330. Molina, F., et al. (1994), Geomagnetic survey of Italy: Repeat station network and magnetic maps, Internal Publ. 554, Ist. Naz. di Geofis. e Vulcanol., Rome. Monaco, C., and L. Tortorici (1995), Tectonic role of ophiolite-bearing terranes in the devolopment of the southern Apennines orogenic belt, Terra Nova, 7, 153 – 160. Monaco, C., L. Tortorici, and W. Paltrinieri (1998), Structural evolution of the Lucanian Apennines, southern Italy, J. Struct. Geol., 20, 617 – 638. Mongelli, F., M. Loddo, and G. Calcagnile (1975), Some observations on the Apennines gravity field, Earth Planet. Sci. Lett., 24, 385 – 393. Mostardini, F., and S. Merlini (1986), Appennino centro meridionale, sezioni geologiche e proposta di modello strutturale, Mem. Soc. Geol. Ital., 35, 177 – 202. Nicolai, C., and R. Gambini (2005), Structural architecture of the Adria platform-and-basin system, Boll. Soc. Geol. Ital., in press. Nicolich, R. (1989), Crustal structure from seismic studies in the frame of the European Geotraverse (southern segment) and CROP Projects, in The Lithosphere in Italy: Advances in Earth Science Research, Atti Conv. Lincei Ser., vol. 80, edited by A. Boriani et al., pp. 41 – 61, Accad. Naz. dei Lincei, Rome. Nicolich, R. (2001), Deep seismic transects, in Anatomy of an Orogen: The Apennines and Adjacent Mediterranean basin, edited by G. B. Vai and I. P. Martini, pp. 47 – 52, Springer, New York. Ogniben, L. (1969), Schema introduttivo alla geologia del confine calabro-lucano, Mem. Soc. Geol. Ital., 8, 453 – 763. Patacca, E., and P. Scandone (1989), Post-Tortonian mountain building in the Apennines: The role of the passive sinking of a relic lithospheric slab, in The Lithosphere in Italy: Advances in Earth Science Research, Atti Conv. Lincei Ser., vol. 80, edited by A. Boriani et al., pp. 157 – 176, Accad. Naz. dei Lincei, Rome. Patacca, E., and P. Scandone (2001a), Line CROP 04 – southern Apennines, geological interpretation and section balance, paper presented at the Third Forum Italiano di Scienze della Terra, Geoitalia 2001, Fed. Ital. di Sci. della Terra, Chieti, Italy, 5 – 8 Sept. Patacca, E., and P. Scandone (2001b), Late thrust propagation and sedimentary response in the thrust belt-fore-deep system of the southern Apennines, in Anatomy of an Orogen: The Apennines and Adjacent Mediterranean Basin, edited by G. B. Vai and I. P. Martini, pp. 401 – 440, Springer, New York. Patacca, E., R. Sartori, and P. Scandone (1990), Tyrrhenian basin and Appeninic Arcs: Kinematic relations since late Tortonian times, Mem. Soc. Geol. Ital., 45, 425 – 451.

Patacca, E., P. Scandone, M. Bellatalla, N. Perilli, and U. Santini (1992a), The Numidian-sand event in the southern Apennines, Mem. Sci. Geol., 18, 297 – 337. Patacca, E., P. Scandone, M. Bellatalla, N. Perilli, and U. Santini (1992b), La zona di giunzione tra l’arco appenninico settentrionale e l’arco appenninico meridionale nell’Abruzzo e nel Molise, in Studi Preliminari All’acquisizione Dati del Profilo CROP 11 Civitavecchia-Vasto, edited by M. Tozzi et al., Stud. Geol. Camerti, Spec. Vol., 1991/2, 417 – 441. Patacca, E., A. Cascella, and D. Merola (2005), Stratigraphy of selected wells along the CROP-04 transect, Boll. Soc. Geol. Ital., in press. Peccerillo, A. (1985), Roman comagmatic province central Italy: Evidence for subduction-related magma genesis, Geology, 13, 103 – 106. Pescatore, T., P. Renda, and M. Tramutoli (1988), Rapporti tra le unita’ lagonegresi e le unita’ sicilidi nella media Valle del Basento, Lucania (Appennino meridionale), Mem. Soc. Geol. Ital., 41, 353 – 361. Pingue, F., G. De Natale, V. Bonasia, V. Cagnetti, E. Del Pezzo, M. Martini, R. Scarpa, and A. Zollo (1988), Il terremoto Irpino-Lucano del 23 Novembre 1980, Mem. Soc. Geol. Ital., 41, 1163 – 1169. Pinna, E. (1987), Italy, Total Magnetic Field Anomaly, Selca, Florence. Piromallo, C., and A. Morelli (1997), Imaging the Mediterranean upper mantle by P-wave travel time tomography, Ann. Geofis., 40, 963 – 979. Prosser, G., M . Schiattarella, M. Tramutoli, C. Doglioni, P. Harabaglia, and A. Bigozzi (1996), Una sezione rappresentativa dell’Appennino meridionale, in Proceedings of the Conferenza Sulla Ricerca Scientifica in Basilicata, pp. 150 – 154, Univ. della Basilicata-Regione Basilicata, Potenza, Italy. Ricchetti, G., N. Ciaranfi, E. Luperto Sinni, F. Monelli, and P. Pieri (1988), Geodinamica ed evoluzione sedimentaria e tettonica dell’avampaese apulo, Mem. Soc. Geol. Ital., 41, 57 – 82. Roure, F., and W. Sassi (1995), Kinematics of deformation and petroleum system appraisal in Neogene foreland fold-and-thrust belts, Pet. Geosci., 1, 253 – 269. Roure, F., P. Casero, and R. Vially (1991), Growth processes and melange formation in the southern Apennines accretionary wedge, Earth Planet. Sci. Lett., 102, 395 – 412. Royden, L. E. (1988), Flexural behaviour of the continental lithosphere in Italy: Constraints imposed by gravity and deflection data, J. Geophys. Res., 93(B7), 7747 – 7766. Royden, L. E. (1993), The tectonic expression of slab pull at continental convergent boundaries, Tectonics, 12, 303 – 325. Royden, L. E., and B. C. Burchfiel (1989), Are systematic variations in thrust belt style related to plate boundary processes? (the Western Alps versus the Carpathians), Tectonics, 8, 51 – 62. Royden, L. E., E. Patacca, and P. Scandone (1987), Segmentation and configuration of subducted lithosphere in Italy: An important control on thrust-belt and foredeep-basin evolution, Geology, 15, 714 – 717. Sartoni, S., and U. Crescenti (1961), Ricerche biostratigrafiche nel Mesozoico dell’Appennino meridionale, Giorn. Geol., 29, 161 – 302. Scandone, P. (1967), Studi di geologia lucana: La serie calcareo-silico-marnosa e i suoi rapporti con l’Appennino Calcareo, Boll. Soc. Nat. Napoli, 76, 301 – 469. Scandone, P. (1972), Studi di geologia lucana: La carta dei terreni della serie calcareo-silico-marnosa e note illustrative, Boll. Soc. Nat. Napoli, 81, 225 – 300. Scarascia, S., A. Lozej, and R. Cassinis (1994), Crustal structures of the Ligurian, Tyrrhenian and Ionian seas and adjacent onshore areas interpreted from wide-angle seismic profiles, Boll. Geofis. Teor. Appl., 36, 5 – 19. Schiattarella, M., P. Di Leo, P. Benedice, and S. I. Giano (2003), Quaternary uplift vs tectonic loading:

20 of 20

TC3005

A case study from the Lucanian Apennine,southern Italy, Quat. Int., 101 – 102, 239 – 251. Sciamanna, S., R. Gambini, F. Mosca, C. Nicolai, J. L. Rudkiewicz, and W. Sassi (2000), Southern Apennine Thrust Belt: Forward structural and geochemical modelling since Messinian times, paper presented at Geology and Petroleum Geology of Mediterranean Basins, Eur. Assoc. of Geosci. and Eng., Malta, 1 – 4 Oct. Scrocca, D., , C. Doglioni, F. Innocenti, P. Manetti, A. Mazzotti, L. Bertelli, L. Burbi, and S. D’Offizi. (Eds.) (2003), CROP Atlas: Seismic reflection profiles of the Italian crust, Mem. Descr. Carta Geol. Ital., 62, 194 pp. Scrocca, D., S. Sciamanna, E. Di Luzio, M. Tozzi, C. Nicolai, and R. Gambini (2005), Structural setting along the CROP-04 deep seismic profile (southern Apennines-Italy), Boll. Soc. Geol. Ital., in press. Sella, M., C. Turci, and A. Riva (1988), Sintesi geopetrolifera della Fossa Bradanica, Mem. Soc. Geol. Ital., 41, 87 – 108. Selli, R. (1957), Sulla trasgressione del Miocene nell’Italia meridionale, Giorn. Geol., 26, 1 – 54. Selli, R. (1962), Il Paleogene nel quadro della geologia dell’Italia meridionale, Mem. Soc. Geol. Ital., 3, 733 – 789. Serri, G., F. Innocenti, and P. Manetti (1993), Geochemical and petrological evidence of the subduction of delaminated Adriatic continental lithosphere in the genesis of the Neogene-Quaternary magmatism of central Italy, Tectonophysics, 223, 117 – 147. Servizio Geologico Nazionale (1994), Carta Aeromagnetica d’Italia, scale 1:100,000, Serv. Geol. d’Ital., Ist. Poligr. e Zecca dello Stato, Rome. Sgrosso, I. (1988), Nuovi elementi per un piu` articolato modello paleogeografico nell’Appennino centromeridionale, Mem. Soc. Geol. Ital., 41, 225 – 242. Shiner, P. R., S. Mazzoli, and G. Cello (2004), Interactions between thin and thick-skinned thrusting in the Val d’Agri area of the S. Apennines, paper presented at the 32nd International Geological Congress, Int. Union of Geol. Sci., Florence, Italy, 20 – 28 Aug. Spakman, W. (1990), Tomographic images of the upper mantle below central Europe and the Mediterranean, Terra Nova, 2, 542 – 553. Spakman, W., S. van der Lee, and R. van der Hilst (1993), Travel-time tomography of the EuropeanMediterranean mantle down to 1400 km, Phys. Earth Planet. Inter., 79, 3 – 74. Speranza, F., and M. Chiappini (2002), Thick-skinned tectonics in the external Apennines, Italy: New evidence from magnetic anomaly analysis, J. Geophys. Res., 107(B11), 2290, doi:10.1029/2000JB000027. van der Meulen, M. J., J. E. Meulenkamp, and M. J. R. Wortel (1998), Lateral shift of Apenninic foredeep depocentres reflecting detachment of subducted lithosphere, Earth Planet. Sci. Lett., 154, 203 – 219. Wigger, P. J. (1984), Die Krustenstruktur der Nordapennins und angrenzender Gebiete mit besondere Ber} ucksichtigung der geotermischen Anomalie der Toskana, Berlin. Geowiss. Abh. Reihess Heft, 9, 1 – 87. Wood, A. W. (1981), Extensional tectonics and birth of the Lagonegro Basin (southern Italian Apennines), Neues Jahrb. Geol. Palaeontol. Abh., 161, 93 – 131. Ziegler, P. A., and F. Roure (1996), Architecture and petroleum systems of the Alpine orogen and associated basin, in Pery-Tethys Memoir 2: Structure and Prospects of Alpine Basins and Forelands, edited by P. A. Ziegler and F. Horva`th, Mem. Mus. Natl. Hist. Nat., 170, 15 – 45.

E. Carminati and C. Doglioni, Dipartimento di Scienze della Terra, Universita` degli Studi di Roma La Sapienza, P.le A. Moro 5, I-00185 Rome, Italy. D. Scrocca, Istituto di Geologia Ambientale e Geoingegneria, Consiglio Nazionale delle Ricerche, c/o Dipartimento di Scienze della Terra, Universita` degli Studi di Roma La Sapienza, P.le A. Moro 5, Box 11, I-00185 Rome, Italy. ([email protected])