Abstract
Magmatic activity in the western part of Elba Island at the north end of the Tyrrhenian Sea lasted approximately 1.5 Ma during the Late Miocene, building a complex of nested Christmas-tree laccoliths, a 10 km-diameter pluton (Monte Capanne) and, finally, the steeply-dipping Orano dyke swarm (ODS). This igneous activity occurred as an extensional regime and followed the wake of eastward-migrating compression of the Apennine front. The ODS consists of hybridized mantle-derived magmas, constituting about 200 dykes totalling a length of approximately 90 km. These dykes intruded the northwestern part of the pluton (NW of the Pomonte–Procchio geomorphic lineament) and its contact aureole, as well as several kilometres of sedimentary rock above. The ODS intruded near the close of pluton crystallization, above a source region marked by a positive magnetic anomaly located NW of the Pomonte–Procchio lineament. Dyke orientations are dominated by a major system trending N78E, with dykes concentrated in belts that locally produced up to 15% extension; between these belts, a minor system of Orano dykes dominates with N38W and N22E trends.
ODS emplacement patterns preserve the strain that resulted in exploitation of Riedel fractures in a NE–SW dextral shear zone; local internal zones of sinistral shear account for one set of the minor system. This shearing occurred between offset segments of the Elba Ridge in the western Elba transfer zone, where strain concentrated magma flow to build the western Elba magmatic complex. This zone developed as a result of different extension rates that produced north-trending Neogene–Quaternary sedimentary basins north and south of the zone. Such basins are connected regionally by NE-trending lineaments previously active during the formation and destruction of the Tethys Ocean. All the magmatic centres in the northern Tyrrhenian–Tuscan area are distributed along such lineaments and developed as a wave moving northeastward across the region, suggesting that magmatism was focused by transfer zone development as back-arc extension migrated in that direction and reactivated former faults.
Dykes are commonly considered evidence for pathways of magmas ascending through the crust (Clemens & Mawer 1992; Petford et al. 1993, 1994; Petford 2000). They offer the potential to reconstruct stress fields at the time of dyking because they directly record the strain associated with their emplacement (Anderson 1951; Delaney et al. 1986; Jolly & Sanderson 1995). Possible stress-related inferences include the degree of coaxiality with respect to the near-field boundary conditions, which in turn relates to whether the deforming system is compressive, tensional, transpressive or transtensional (Glazner et al. 1999; Mège & Korme 2004).
Several factors influence the geometry and the structural patterns of magma pathways, namely: (i) structure of the crust (presence and attitude of old discontinuities like bedding and thrusts faults) and its rheological heterogeneities; (ii) regional stress fields in the area at different crustal levels; and (iii) magma driving pressure (Delaney et al. 1986; Tokarski 1990; Baer et al. 1994; Hogan et al. 1998). Emplacement of multiple intrusions introduces additional transient factors to the local stress field (Vigneresse & Tikoff 1999). All these factors contribute in differing degrees to produce the final spatial distribution of dykes in plutonic–subvolcanic systems.
Studying syn- to post-plutonic dyke swarms emplaced in a granitic host can mitigate the overall complexity affecting palaeostress reconstruction. Indeed, because of the high degree of mechanical isotropy of the granitic host, these dykes can potentially provide more straightforward information about the stress fields active during dyke emplacement than can dykes in layered sedimentary or metamorphic host rocks. Dyke swarms inside granite are commonly emplaced in the waning stages of the build-up of a plutonic complex, and, in this context, both regional and local stress conditions can play a role in determining their final distribution and structural pattern. Examples of regionally controlled emplacement of dyke swarms in granitic hosts include the North Atlantic Igneous Province (Jolley & Bell 2002), the Independence dyke swarm (Glazner et al. 1999) and the Kaplice dyke swarm (Vrana et al. 2005). However, the regular radial and tangential patterns of aplite–pegmatite dykes in many granite plutons indicate local control (Marre 1986; Bergbauer et al. 1998) or interference between local and regional control (Bergbauer & Martel 1999).
The Miocene Monte Capanne pluton of Elba Island (Tuscany, Italy) includes examples of both local and regional stress control on dyke emplacement: aplitic dykes sourced in the pluton were emplaced in part under the control of cooling jointing and elsewhere by detachment above the pluton, while emplacement of mafic Orano magma rising from the mantle was controlled by a regional transfer zone. Finally, insights from the structural and tectonic history of western Elba Island allow us to speculate on the distribution patterns of magmatism throughout the northern Tyrrhenian region.
Geological background
The structure of Elba Island includes five stacked tectonic complexes (Fig. 1) assembled by about 20 Ma (Deino et al. 1992) due to the collision of the Sardinia–Corsica block and the Adria plate (Malinverno & Ryan 1986). The lower complexes (I–III) have continental affinity, namely metamorphic basement with carbonate and clastic sedimentary covers, while the upper ones (IV–V) are oceanic in character (Trevisan 1950; Keller & Pialli 1990; Pertusati et al. 1993). These units were metamorphosed and deformed in east-verging nappes as the Apennine compressional thrust–fold belt developed.
Location map and geological sketch map for Elba Island. The three main zones (western, central and eastern) of the island are separated by the Eastern Border Fault (EBF) and the Central Elba Fault (CEF), respectively. Dashed ellipses enclose the outcrop zones of the Orano dyke swarm in western and central Elba. A–A′ indicates the location of magnetic cross-section reported in Figure 3. PP line is a geomorphological lineament described in the text and in Figure 5.
While this compressive regime advanced to the east (Brunet et al. 2000), the west-dipping Adriatic plate rolled back (Serri et al. 1993), triggering delamination of the upper plate where an ensialic back-arc setting developed and propagated eastward through time. Associated igneous activity also migrated from west (14 Ma) to east (0.2 Ma), producing intrusive and extrusive products of mantle–crustal hybrid nature to build the Tuscan Magmatic Province (Barberi et al. 1971; Innocenti et al. 1992, 1997; Westerman et al. 1993; Dini et al. 2002; Savelli 2002).
Extensional processes and igneous activity affected the Elba Island region during the Late Miocene (Bouillin et al. 1993; Serri et al. 1993; Jolivet et al. 1994). Field, petrographic and geochemical data, along with intrusive relationships, have been used to define and correlate several Miocene intrusive units in western and central Elba, with the relative chronology of the units supported by isotope chronology (Dini et al. 2002, 2004).
The magmatic sequence formed a nested Christmas-tree laccolith complex of porphyritic rocks with a diameter of about 10 km, built of nine layers with a cumulate thickness of about 2400 m, all emplaced at depths of 2–3.5 km (Fig. 2) (Rocchi et al. 2002; Westerman et al. 2004) between 8.5 Ma (40Ar-39Ar muscovite plateau age after Maineri et al. 2003) and 7.4 Ma (Dini et al. 2002). This complex was intruded by the Monte Capanne pluton, a roughly circular pluton also with a diameter of 10 km, and an emplacement depth of about 6 km at approximately 6.9 Ma (Dini et al. 2002). A pluton thickness of about 3 km is suggested for the pluton from magnetic modelling (Fig. 3) (Botti 1999).
Schematic cross-section of the western Elba intrusive complex as reconstructed for the time of emplacement of the Orano dyke swarm at approximately 6.85 Ma (Westerman et al. 2004). CBa, Capo Bianco aplite, two-layer laccolith disrupted by the intrusion of later laccoliths; Pp, Portoferraio porphyry, multilayer laccolith; SMp, San Martino porphyry, multilayer laccolith; ODS, Orano dyke swarm; CEF, Central Elba Fault (see Fig. 1) along which the western Elba intrusive complex gets decapitated soon after emplacement the Orano dyke swarm.
Magnetic model (Botti 1999) based on a detailed survey of the Monte Capanne pluton and its surroundings. Magnetic susceptibilities used in the model were measured in the field and are reported in parentheses for each rock type in 10−3 SI units. The model is drawn on a NW–SE section (see Fig. 1 for location).
Two systems of dykes cut the pluton. The first is represented by widely distributed aplite–pegmatite veins and dykes, generally less than 1 m thick and extending only several metres along strike; several leucosyenogranite dykes and lenses (tens of meters thick) also occur, mainly close to the pluton's contact.
The second system is the Orano mafic dyke swarm (ODS), emplaced 6.85 Ma (Dini et al. 2002) before complete consolidation of the plutonic system (Fig. 4). In western Elba, the ODS only occurs NW of a geomorphic lineament – the Pomonte–Procchio line (PP line, Fig. 1). This is a prominent topographic feature seen in satellite imagery as an alignment of valleys and ridges extending from Pomonte through Monte Perone to the Procchio area (Fig. 5). The PP line splits the Monte Capanne pluton (including its contact aureole) into two regions characterized by different morphological textures: (i) the northwestern part displays a deeply incised drainage pattern, with valley orientations clustered in WNW and NE directions; and (ii) the southeastern part shows a smoother morphology with an overall NW-trending, poorly incised drainage system. In addition, the northwestern region is characterized by a positive magnetic anomaly (Fig. 3), modelled as a sill-like mass of moderate magnetic susceptibility lying below this part of the pluton (Botti 1999). This body is thought to represent the immediate source region for the magma that filled Orano dykes. This ENE-trending ODS includes more than 200 darkly-coloured dykes with dominant quartz monzodioritic–granodioritic compositions (Fig. 6). Orano dykes are porphyritic with plagioclase, biotite, clinopyroxene and amphibole phenocrysts, and quartz and K-feldspar xenocrysts set in a very fine-grained groundmass of plagioclase, K-feldspar and phlogopite. Some dykes, generally among the thickest, are zoned. Outer border zones, typically a few tens of centimetres thick, are distinguished from inner zones by: (i) finer grained groundmass; (ii) lower content of K-feldspar and quartz xenocrysts; and (iii) higher ferromagnesian mineral concentrations. The borders of zoned dykes are petrographically comparable to unzoned dykes, while the inner portions of zoned dykes reach monzogranitic compositions (Fig. 6). The ODS intruded all of the other intrusive units of the sequence about 6.85 Ma, representing the closing event of igneous activity in western Elba (Dini et al. 2002).
(a) Typical Orano dyke within Monte Capanne monzogranite; the hammer is for scale. Chiessi village and Colle San Bartolomeo in the background. See also Figure 9 for location. (b) Orano dyke with horn-and-bridge (side step) geometry at Colle d'Orano type locality (Fig. 7). The small sledge is for scale. The dyke also contains exotic K-feldspar megacrysts and granite blocks. (c) Example of brittle emplacement style of the Orano dyke in the Monte Capanne monzogranite, near Sant'Andrea (Fig. 7). The lens cap to the right of the circle is for scale. Note within the circle a K-feldspar megacryst of the granite cut across by the dyke. (d) Example of plastic emplacement style of the Orano dyke in the Monte Capanne monzogranite, near Pomonte (Fig. 7); the hammer is for scale. (e) Example of the Orano dyke cross-cutting an aplite dyke within the Monte Capanne monzogranite, near Pomonte (Fig. 7); the hammer is for scale. (f) Examples of sudden orientation changes of Orano dykes, along with closely occurring plastic and brittle emplacement styles. Locality near Chiessi (see also Fig. 9 for location); the hammer is for scale.
Satellite image showing the PP line as a complex alignment of valleys (dashed lines) and ridges (dotted lines) extending from Pomonte through Monte Perone to the Procchio area.
(a) Total alkali v. silica classification diagram (Le Maitre 2002) showing the composition of ODS compared to the other Miocene igneous units of Elba. Worthy of note is the compositional overlap of Orano dykes from western and central Elba. (b) The enlarged view reports tie-lines connecting inner (small symbols) and outer parts of zoned dykes (large symbols).
As construction of the domed igneous complex in western Elba neared completion, surface slopes approached 25° (Westerman et al. 2004), and detachment on the Central Elba fault décollement surface was initiated (Fig. 2). Ultimately, the top half of the laccolith complex, including the uppermost part of the ODS, was displaced at least 8 km eastwards. Concurrent or subsequent uplift of western Elba occurred with 2–3 km of vertical throw on the Eastern Border fault, coupled with westward tilting of the top half of the complex now displaced in central Elba. This produced a serendipitous sequence of exposures with the lowermost part of the ODS in its original setting (in western Elba, with a vertical extent of at least 1 km) and the uppermost part displaced in central Elba (within the laccolith–flysch system, with a vertical extent of at least 3 km) (Dini et al. 2002; Westerman et al. 2004).
Method of data collection
Detailed mapping and field measurements were carried out at a scale of 1 : 10 000, with data collected at approximately 1000 stations in the study area. Dykes were mapped by walking along contacts wherever possible. Interpretive maps for western and central Elba (Figs 7 and 8) are based on all the observations as well as inferred dyke emplacement styles.
(a) Interpretive geological map of Orano dykes (black) in western Elba. The southeastern limit of the area affected by dyke intrusion is indicated by the bold dashed line. Shades of grey within the Monte Capanne pluton represent the pluton's internal facies: white, K-feldspar megacryst-rich S. Andrea facies; dark grey, K-feldspar megacryst-poor S. Piero facies; grey, S. Francesco facies, with intermediate petrographic features. Dotted lines bound areas dominated by dykes with ENE trends (major system) and NNE or NW trends (minor system). Oblique ruling represents the pluton's contact aureole. (b) The strike frequency distribution and rose diagram of western Elba Orano dykes. (c) Rose diagrams of strike for zones dominated by the major and minor systems. The diagram on the right has a radius scaled ×4 with respect to the one on the left to emphasize the strike distribution within the minor system.
(a) Interpretive geological map of central Elba showing the distribution of Orano dykes (black, modified after Dini et al. 2004). Oblique ruling, Monte Capanne pluton country rock; grey, central Elba multilayer – laccolith complex; horizontal ruling, laccolith country rock (Complex V, Cretaceous and Paleocene flysch); oblique dashed ruling, Complex IV (Jurassic ophiolite sequence and its ocean-floor sedimentary cover). (b) Strike frequency distribution and rose diagram of central Elba Orano dykes.
To evaluate variations in dyke trends for the entire swarm, a dataset of strikes was generated from the interpretive maps (1 : 10 000 scale). As nearly all dykes are subvertical, each one was divided into segments with continuous and coherently oriented apparent strike. Individual data points were then assigned for each 10 m of a given segment (for example, a 70 m-segment striking N85E generated seven data points with that strike value). This method was modified for non-vertical dykes trending roughly orthogonal to ridges or valleys. For those dykes, length along strike was determined by connecting points of equal elevation. In this way, a dataset of approximately 9000 strike values was generated for the 90 km of dykes presented on the maps. The method also accounts for the fact that dyke orientations in the field often vary over short distances, in which case a single measurement could be significantly misleading.
The abundance of information presented in the interpretive maps (Figs 7 and 8) allows calculation of values of extension within the mapped areas. To do this, a grid was established with traverse lines spaced 500 m apart and oriented N15W, orthogonal to the average orientation of the major dyke system that accounts for most of the extension. This method was modified from Jolly & Sanderson (1995) by using projected thickness values rather than true thicknesses. This was necessary because many Orano dykes cross the traverses obliquely, in which case only a part of their true extension occurs parallel to the grid direction. Values for the percentage of extension parallel to the grid lines equal the sum of dyke extensions parallel to the traverse (t), minus half the extensions of the first and last dykes in the group, divided by the sum of the space (s) between the dykes in the set (three or more dykes make up each set):
This method provides information for each traverse as a whole, as well as for many segments along each traverse. Values are plotted at the mid-point of each evaluated segment and then contoured, permitting evaluation of how the amount of N15W–S15E extension varies throughout the study area.
Data
In western Elba, over 80 000 m total cumulate length for Orano dykes are seen NW of the PP line on the interpretive map (Fig. 7a), occurring primarily within the Monte Capanne pluton and, to a lesser extent, in the surrounding contact aureole. Orientations of Orano dykes in western Elba are shown in Figure 7b as a histogram and as a rose diagram, both constructed from the weighted dataset described above. A primary system of dykes made of a set accounting for 67 of the 80 km cumulative length shows a normal distribution with average orientation N75E, and is referred to hereafter as the ENE major system. In addition, a minor system of dykes occurs, consisting of two sets with average orientations N38W (hereafter referred to as NW) and N22E (hereafter NNE). The areal distribution of the dykes belonging to the different systems in not uniform: Figure 7a shows how dykes with different orientations are concentrated in distinct ENE-oriented ‘belts’. Figure 7c illustrates separately the strike distribution of dykes in the zones dominated by the major and minor systems, respectively, in order to better visualize the patterns of the minor systems.
Orano dykes in central Elba belong to the eastward-displaced uppermost part of the swarm, and they crop out with 9300 m of cumulate length (Fig. 8). Their distribution is restricted to the northern half of the region. Patterns of orientation and relative frequency are similar to those in western Elba (Fig. 8), even though the orientation pattern in central Elba is rotated approximately 15°–20° clockwise. This rotation probably took place during the eastward translation of the uppermost part of the ODS from western to central Elba. As a consequence, only the western Elba dykes are found in their pristine emplacement location and have their original attitudes preserved.
The entire ODS is steeply dipping with local dip variations between 90° and 40°, and rare subhorizontal small dykes. Orano dykes in western Elba are numerous (157 dykes on the interpreted map), with thicknesses ranging from a few centimetres to 50 m, and lengths ranging from metres to about 7 km. Most dykes have sharp and planar contacts that commonly change orientation abruptly (Figs 4 and 9). Thicknesses along strike are generally maintained, independent of abrupt variations in attitude, indicating that the direction of opening was perpendicular to dyke walls. This direction of opening is further supported by observations where Orano dykes cut a variety of older lithological contacts, such as between the pluton and country rock, older aplite dykes or microgranular magmatic enclaves (Fig. 4e). Given the constraints that the major system is overwhelmingly dominant, that dyke dips are steep, and that opening is perpendicular to dyke walls, the overall direction of opening was horizontal in the NNW direction.
Detailed geological map of an Orano dyke near Punta del Timone, Chiessi (west side of Monte Capanne pluton, see Fig. 7 for location). Grey, Orano dyke; white, Monte Capanne pluton; oblique ruling, contact aureole. This single, complex dyke shows: (i) variable intrusive styles from plastic to brittle; and (ii) the whole range of strikes shown by the overall ODS (see also Fig. 4a and f). Topographic contour lines are in metres above sea level.
Dykes in both the major and minor systems show the same rheological behaviour with respect to their host: although they generally have sharp and continuous contacts, some dykes locally display flow ‘readjustment’ of the dyke walls. Embayments and crenulated contacts are common and testify that the granite was not completely crystallized during dyke intrusion (Fig. 4d). In addition, detailed mapping and field observations show that individual Orano dykes commonly change orientation (from ENE to NNE to NW), and dykes have been traced continuously through such changes in orientation without petrographical variation (Figs 4 and 9). These dykes also show additional evidence of complex geometry, such as sudden side-steps (Fig. 4c, and f), horn structures and bridges (Fig. 4b), and broken bridges. Finally, Orano dykes of all orientations have petrographical and chemical characteristics that overlap; when divided by orientation and plotted in variation diagrams they show no systematic geochemical distinctions (Fig. 10). Taken together, these lines of evidence indicate that the major and minor systems constitute a single dyking episode.
Chemical variation diagrams of the two systems of Orano dykes: open squares, ENE dykes; filled circles, NW and NNE dykes. The two dyke systems display overlapping compositional fields, lending further support to a single emplacement episode.
These attitude and geometry data for the ODS have then been used to derive information about the distribution and degree of extension needed to accommodate dyke emplacement. With this aim, a map of ‘percent extension’ was created (Fig. 11) following the methods described earlier, i.e. contouring the extension percentages calculated along the N15W–S15E direction. This choice effectively documents the extension produced by the ENE major dyke system, while the extension due to the minor systems is not accounted for. Nevertheless, extension determined along directions orthogonal to the strikes of minor systems is always below 2%, thus not affecting the results described in the following paragraph.
Contour map of per cent extension (indicated by the numbers on the contour lines) as determined in the N15W S15E direction for Orano dykes in western Elba (see the section ‘Methods of data collection’ for an explanation of extension determination). Grid spacing is 500 m. Oblique ruling, contact aureole. A–A′ indicates the location of magnetic cross-section reported in Figure 1 and 3.
The pattern of extension reveals three maxima (up to 15%) located in a restricted oval area in the central part of the dyke swarm. This area is in the middle of the NW sector of the pluton, corresponding to that of the underlying susceptive body inferred from magnetic data (Fig. 3). These maxima have shapes elongated parallel to the strike of the major system, and are best illustrated by the 3% contour line. It is worth noting that extension values abruptly decrease, starting from the maxima to the peripheral zone of dyking. Thus, the amount of extension (with respect to the ENE strike of the major dyke system) decreases both transversally and longitudinally, indicating that most of the magma input was focused in a limited area.
The southeastern margin of the ODS is marked by the NE-trending PP line, a geomorphic lineament that had structural significance prior to emplacement of the ODS. Indeed, the slightly older leucocratic dyke system has pronounced differences in orientations NW and SE of the PP line (Fig. 12a). Orientations of these dykes in the northwestern part of the pluton correlate strongly with pluton radii (Fig. 12b), suggesting dykes exploited radial cooling fractures. Leucocratic dykes SE of the PP line trend NNE–SSW, perpendicular to magnetic lineations in the pluton and to mineral lineations in the surrounding country rocks (Bouillin et al. 1993). This pattern is consistent with incipient southeastward movement on a décollement surface (Central Elba Fault, Figs 1 and 2) above the eastern part of the pluton that left the western part unaffected.
(a) Orientation of aplite dykes (data from this work; Boccaletti & Papini 1989; Pezzotta 2000) and magnetic lineations in the Monte Capanne pluton (Bouillin et al. 1993); (b) frequency distribution of aplite dyke orientations in the NW zone of Monte Capanne, defined as the angle between the dyke strike and the pluton radius (radiating from the top of Monte Capanne and passing through the dyke centre); and (c) stereogram of poles of representative Orano (filled diamond) and aplite dykes occurring in NW (open square) and SE (open circle) zones, respectively.
Discussion
The collected data indicate that the ODS emplacement was not controlled either by a local, cooling-related stress field or by the overall regional Late Miocene east–west extension (Pascucci et al. 1999). Indeed, the contrast in orientations of the radial leucocratic dyke system and the slightly younger ODS in the NW portion of the pluton and its surroundings indicates that the local stress fields ruling aplite dykes emplacement shifted to a different stress field. Moreover, this latter stress field is incompatible with the overall east–west extension in the northern Tyrrhenian Sea, as the dykes are oriented nearly parallel to that direction.
ODS in the western Elba transfer zone (WETZ)
The stress field controlling ODS emplacement can be constrained by structural patterns of the dykes (Fig. 7). The major system of Orano dykes shows a pattern of linked offset fractures, best seen in the largest dykes weaving through the heart of the pluton (Fig. 7) that appear as connected side-stepping en échelon fractures. This pattern is characteristic of primary fractures (R1) of a dextral Riedel shear system trending NE–SW. A key zone for the understanding of such a geometry is the southwestern region of the swarm where the segmented, ENE-trending dyke of San Bartolomeo (belonging to the major system, Fig. 7) is almost continuous over a length of about 3 km. The two main segments are characterized by eastern terminations striking WNW, parallel to the small dykes intruding the host rock to the south of the major dyke. Additional evidence for dextral shear in that region is represented by the prominent ENE-trending Chiessi–Capanne–Perone dykes in combination with the nearby NW-oriented dyke set (Fig. 7).
Mechanisms controlling emplacement of the ODS could be related to differential rates of horizontal movement and an irregular distribution of stretching in the brittle portion of the crust on a regional scale. This requires strain accommodation along transfer faults that typically develop at relatively high differential extension rates (Acocella et al. 2005), perhaps as a result of differential slab retreat (Scrocca 2006). The term transfer fault is used here to describe a subvertical transtensive fault between adjacent zones undergoing differential movement (Gibbs 1990; Peacock et al. 2000). We propose that the ODS formed in a narrow NE-trending transfer fault zone, the western Elba transfer zone (WETZ), with the PP line marking the SE border of the zone. In this scenario, the major dyke system is aligned in correspondence with R1 shear fractures of predominantly dextral strain. Such transfer fault systems in back-arc regions are characteristically complex (Dewey et al. 1998), such that variations in the rates of translation of adjacent blocks within the transfer fault zone produce sectors with opposing senses of strain, even if the overall shear couple remains constant (Corti et al. 2003). Within the ODS, coexisting opposite senses of shear are pointed out by the occurrence of zones dominated by one or the other set of minor dykes. Where dextral shear dominated, the NW-trending set of minor dykes formed along R2 fractures (e.g. the minor system to the north of Monte Capanne; Fig. 7); in zones dominated by sinistral shear, NNE-trending R2 fractures were exploited to make the second dyke set (e.g. the minor system in the northwestern area of the ODS).
Overall, thickness and spacing of Orano dykes in the WETZ are not homogeneously distributed (Fig. 7): the extension values related to dyke emplacement depict three ENE-elongated ellipses for highest values (3–15%) enclosed in a larger elliptical area with the same orientation, bounded by 0.5% extension value (Fig. 11). Thus, the distribution of extension values indicates that significant magma volume was emplaced in a narrow, restricted zone within the overall outcrop area of the ODS. This suggests that, even though the ODS structural pattern was controlled by the WETZ, the restricted zone of rising melts appears to be linked to a local discrete feeding core. This inference is supported by the positive magnetic anomaly observed for this area and interpreted as evidence for a susceptive, mafic igneous body just below the Monte Capanne pluton.
Restricted geographical distribution is characteristic not only of the ODS, but also of the entire episode of western Elba magmatism that included intrusion of seven lithologically distinct units comprising a total estimated volume of about 200 km3 (Rocchi et al. 2002). The igneous activity evolved over 1.5 Ma from crustal products to hybrid products involving multiple crustal- and mantle-derived sources (Dini et al. 2002; Westerman et al. 2003). In all likelihood, the ODS stress field was active during the previous igneous episodes, as indicated by the 7.4 Ma San Martino laccolith for which east–west-trending feeder dykes are preserved in western Elba (Westerman et al. 2004). It is impressive that such numerous and petrogenetically distinct magma batches were emplaced at the same narrowly constrained centre, suggesting the persistence of a discrete system for magma feeding-ascent that was intermittently activated between 8.4 and 6.85 Ma. Of note here is that the northwestern zone of the Monte Capanne pluton has also been indicated as the feeding zone for that entire pluton on the basis of AMS data (Bouillin et al. 1993).
The western Elba transfer zone in the frame of northern Tyrrhenian transfer systems
The WETZ is parallel to several subvertical transverse structures oriented at high angles to the Oligocene–Aquitanian compressional (thrusts, fronts) and post-Burdigalian extensional (basins, normal faults) features of the Apennine orogenic belt (Fig. 13). These ENE- to NE-trending structures have long been recognized on the basis of geological studies (Signorini 1935; Ghelardoni 1965; Fazzini & Gelmini 1982) and geophysical data (Bartole 1995; Cassano et al. 2001), and have been interpreted as transfer fault zones (Bartole 1995). Furthermore, large-scale transfer-zone activity has been recently inferred also for the central sector of the Apennine chain on the basis of detailed seismological reconstruction of the thrust front (Scrocca 2006).
(a) Distribution of magmatic centres and regional transversal lineaments. Red stars, magmatic centres; yellow stars, intrusive bodies inferred on the basis of geophysical data; green contoured areas, positive magnetic anomalies (values in nT); dashed thick blue line, alignment of magnetic maxima; black thin lines, isobaths in metres below the sea level; diagonal hatch pattern, main post-Burdigalian sedimentary basins; diagonal ruling, Corsica Alpine tectonic units; crossed hatch, Variscan Sardinia–Corsica basement; solid black lines with ticks on the down-dropped side, normal faults; grey dashed lines, transversal geological–geophysical lineaments; yellow stripes, linear array of magmatic centres of the Tuscan Magmatic Province. Main rock association of magmatic centres: (1) Sisco, (2) Capraia Island, (3) Orciatico, (4) Montecatini Val di Cecina, (5) western Elba Island, (6) San Vincenzo, (7) Campiglia, (8) Larderello-Travale subsurface complex, (9) eastern Elba Island, (10) Gavorrano, (11) Castel di Pietra buried complex, (12) Roccastrada, (13) Montecristo Island, (14) Monte Amiata, (15) Radicofani, (16) Giglio Island, (17) Torre Alfina, (18) Tolfa, (19) Monte Cimino, (20) Manziana, (21) Cerite. For petrogenetic affinities and classification see Innocenti et al. (1992); Serri et al. (2001). Geological and geophysical data after CNR (1983), Bartole (1995), Carmignani et al. (1995), Pascucci et al. (1999), Chiappini et al. (2000) and Cassano et al. (2001). (b) Tectonic sketch of a theoretical west-dipping, back-rolling subduction system (simplified after Doglioni 1991) developing dextral transfer zones in its northern sector. (c) Close-up view of the Western Elba Transfer Zone linking offset zones of differential extension as indicated by the lengths of the white arrows.
In the Tyrrhenian–Tuscan area, transverse structures bound or offset north- to NNW-oriented geological–geophysical features such as sedimentary basins (Liotta 1991; Carmignani et al. 1994; Pascucci et al. 1999; Acocella & Rossetti 2002) and magnetic anomalies (Chiappini et al. 2000; Cassano et al. 2001). The NE-trending lineaments are thought to represent inherited Mesozoic structures that initially: (1) developed as transform faults during Mesozoic growth of the Tethys Ocean lithosphere; then (2) served as lateral ramps and transpressive structures during the closure of the Tethys oceanic realm and the consequent Oligocene–Aquitanian continental collision; and, finally, (3) focused transfer fault zone development during post-Burdigalian–recent extension (Liotta 1991; Bartole 1995; Carmignani et al. 1995). It is worth noting that the pre-Burdigalian activity along these structures was characterized by left-lateral transpression, as testified to by several occurrences described in the basement rocks of Corsica and Sardinia (Carmignani et al. 1995).
One of the most important NE-trending lineaments corresponds with Elba Island and shows an apparent left-lateral offset of the very prominent north–south Elba Ridge. That ridge coincides with a positive magnetic anomaly more than 300 km long (Fig. 13a) (Chiappini et al. 2000; Cassano et al. 2001), interpreted as a suture zone dominated by ophiolitic rocks stacked during construction of the Apenninic nappe and thrust complex (Bartole et al. 1992). The left-lateral offset of this north–south structure can be related to the pre-extensional tectonic activity. We propose that, with the onset of extension, the NE-trending structure was reactivated as a transfer zone (the WETZ) that controlled the Late Miocene emplacement of the ODS. The SE margin of this transfer zone across western Elba corresponds with the PP line that, not coincidentally, is located along the SW continuation of the most prominent NE-trending lineament on land, i.e. the Piombino–Faenza line (Fig. 13) (Bartole 1995).
This local situation of ODS emplacement controlled by dominantly dextral transtensional activity (Fig. 13) is well framed in the general scenario of the northern Tyrrhenian–Apennine system that is characterized by a strongly arcuate subduction zone, slab rollback and associated back-arc extension. In such a setting, theoretical models (Doglioni 1991) require dextral transtension along ENE-trending structures (Fig. 13b) to accommodate strain between domains affected by differential extension rates. In the area of western Elba, the interplay between the inherited pre-Burdigalian sinistral offset of the north–south Elba Ridge and the later differential extension helped to focus prolonged igneous activity in a small segment of a dextral transfer zone (Fig. 13c).
Magmatism focused by northern Tyrrhenian transfer systems
The western Elba intrusive complex is part of the Tuscan Magmatic Province, made up of some 20 exposed or cored igneous centres and several additional intrusions inferred from geophysical data (Fig. 13a) (Serri et al. 2001; Dini et al. 2003, 2005). The distribution of igneous centres in the Tyrrhenian–Tuscan area appears to be geographically linked to segments of the NE-trending lineaments (Fig. 13a) inferred by previous authors (Liotta 1991; Bartole 1995). In particular, we identify seven main alignments of igneous centres along NE-trending structures (Fig. 13a): (A) Sisco–Capraia–Orciatico–Montecatini Val di Cecina; (B) Western Elba–San Vincenzo–Botro ai Marmi–Larderello–Travale; (C) Porto Azzurro–Gavorrano–Castel di Pietra–Roccastrada; (D) Montecristo–Amiata–Radicofani; (E) Giglio–Torre Alfina; (F) Tolfa–Cimino; and (G) Cerite–Manziana. In addition, the geophysically inferred intrusions are also located along these structures (Fig. 13a).
The well-known, eastward-younging overall distribution of the ages of these igneous centres (Innocenti et al. 1992) also occurs along each of these seven alignments. This suggests an eastward sequential activation of segments of the NE-transversal structures triggering extraction, rising and emplacement of magmas.
Activity at most of these igneous centres was short-lived, resulting in the emplacement of a single magmatic mass. In contrast, formation of the western Elba intrusive complex and the Larderello–Travale buried intrusions were long-lived events (1.5–2.5 Ma). The fact that these two igneous centres lie on the ‘Western Elba–Piombino–Faenza’ transversal line suggests that it represents the most significant alignment of transfer zones in the northern Tyrrhenian–Apennine system.
Conclusions
This paper presents the first complete geological description of the ODS in western and central Elba, where approximately 90 km of dykes have been mapped over an original vertical extent exceeding 4 km. In western Elba, where most dykes were emplaced in the Monte Capanne monzogranitic pluton, two distinct systems of dykes are recognized: a major system made of one set trending ENE; and a minor system consisting of two sets trending NNE and NW. The ODS is organized into elongated zones dominated by individual dyke sets, all having developed in a complex NE–SW dextral shear zone within which differential strain created local zones of sinistral shear.
Emplacement of the ODS, marking the end of magmatism in western Elba, is the first documented evidence for the activity of a transfer fault zone in the northern Tyrrhenian region. This long-lived magmatism in western Elba suggests the transfer zone was active for up to 1.5 Ma before activity migrated northeastward along the Pomonte–Procchio–Piombino–Faenza transversal lineament. Similar distribution of all the magmatic centres of the Tuscan Magmatic Province suggests that they can be explained as being related to activity in transfer fault zone segments that developed sequentially along reactivated ancient lineaments, as back-arc extension migrated northeastward across the Tyrrhenian–Tuscan region.
- © The Geological Society of London 2008
References
Magma emplacement in a transfer zone: the Miocene mafic Orano dyke swarm of Elba Island, Tuscany, Italy
