- Email: [email protected]
Detailed crustal structure in the area of the southern Apennines–Calabrian Arc border from local earthquake tomography
Journal of Geodynamics 82 (2014) 87–97
Contents lists available at ScienceDirect
Journal of Geodynamics journal homepage: http://www.elsevier.com/locate/jog
Detailed crustal structure in the area of the southern Apennines–Calabrian Arc border from local earthquake tomography C. Totaro a,∗ , I. Koulakov b , B. Orecchio a , D. Presti a a b
Physics and Earth Sciences Department, Messina University, Viale F. StagnoD’Alcontres 31, 98166 Messina, Italy Institute of Petroleum Geology and Geophysics, SB RAS, Prospekt Akademika Koptyuga 3, Novosibirsk 630090, Russia
a r t i c l e
i n f o
Article history: Received 28 February 2014 Received in revised form 16 July 2014 Accepted 21 July 2014 Available online 27 July 2014 Keywords: Seismic tomography Crustal structure Calabrian Arc Southern Apennines
a b s t r a c t We present a new seismic velocity model for the southern Apennines–Calabrian Arc border region with the aim to better define the crustal structures at the northern edge of the Ionian subduction zone. This sector also includes the Pollino Mts. area, where a seismic sequence of thousands of small to moderate earthquakes has been recorded between spring 2010 and 2013. In this sector a seismic gap was previously hypothesized by paleoseismological evidences associated with the lack of major earthquakes in historical catalogs. To perform the tomographic inversion we selected ca. 3600 earthquakes that have occurred in the last thirty years and recorded by permanent and temporary networks managed by INGV and Calabria University. Using for the first time the Local Tomography Software for passive tomography inversion (LOTOS hereinafter) to crustal analysis in southern Italy, we have computed the distribution of Vp, Vs, and the Vp/Vs ratio. The obtained velocity model, jointly evaluated with results of synthetic modeling, as well as with the hypocenter distribution and geological information, gives us new constraints on the geodynamical and structural knowledge of the study area. The comparison between the shallow tomography sections and surface geology shows good correlation between velocity patterns and the main geological features of the study area. In the upper crust a lowvelocity anomaly of P- and S-waves is detectable beneath the Pollino Mts. area and seems to separate the Calabrian and southern Apennines domains, characterized by higher velocities. The distributions of high Vp/Vs ratio, representing strongly fractured rocks with likely high fluid content, clearly correlate with areas of significant seismicity. In the lower crust we detect a clear transition from high to low seismic velocities in correspondence with the Tyrrhenian coast of the study area, which may represent the transition from the thinner Tyrrhenian crust to the thicker one beneath Calabria. In this area, also characterized by a progressive detachment of a retreating lithospheric slab, the generation of a Subduction-Transform Edge Propagator (STEP) fault zone, that laterally decouples subducting lithosphere from non-subducting lithosphere in a scissor type of fashion, may have taken place. These conditions imply the existence of a kinematic decoupling which allows for differential movement between the Calabrian Arc and the southern Apennine chain. The low velocity anomaly separating the southern Apennines and the Calabrian Arc domain may be related to fluid upwelling occurring in correspondence with the northern edge of the Calabrian subducting slab. © 2014 Elsevier Ltd. All rights reserved.
1. Introduction The western Mediterranean area, located at the contact belt between the slowly convergent African and Eurasian plates (Calais et al., 2003; Nocquet and Calais, 2004; Serpelloni et al., 2007), has been the site of a continental-scale lithospheric subduction
∗ Corresponding author. Tel.: +39 090 6765102; fax: +39 090 392333. E-mail address: [email protected] (C. Totaro). http://dx.doi.org/10.1016/j.jog.2014.07.004 0264-3707/© 2014 Elsevier Ltd. All rights reserved.
process, the evolution of which in the last 30 million years is marked by the eastward migration of the retreating subduction hinge (Fig. 1a; Wortel and Spakman, 2000). Most of the subduction system has already undergone detachment of the subducting lithosphere with the exception of the central, most arcuate portion of the system, the Calabrian Arc in southern Italy (Fig. 1b; Neri et al., 2009 and references therein). Different states of the subduction process can be related to the progressive change of lithospheric structure near the retreating trench zone and to strong lithospheric heterogeneity between the Calabrian Arc and the marginal tectonic units of Sicily
88
C. Totaro et al. / Journal of Geodynamics 82 (2014) 87–97
Fig. 1. (a) Map of the Mediterranean region with the western Mediterranean plate boundary evolution in the last 30 Myrs (redrawn from Wortel and Spakman, 2000, with modifications according to Neri et al., 2009). The space–time evolution of the boundary marks the process of rollback of the subducting lithosphere and the related trench retreat until the present-day location near the Ionian shoreline of Calabria. Black arrows indicate the present motion of Africa relative to Europe (Calais et al., 2003; Nocquet and Calais, 2004; Nocquet, 2012). In (b) the solid curve with the sawtooth pattern, pointing in the direction of subduction, indicates the present-day location of the Calabrian Arc subducting system. According to the local earthquake tomography by Neri et al. (2009), black sawteeth indicate the continuous subducting slab while white sawteeth the plate boundary segments where slab detachment has already occurred. The white arrow shows the sense of the subducting slab rollback. The black arrows indicate the present motion of Africa relative to Europe (Calais et al., 2003; Nocquet and Calais, 2004). Gray dashed line is the locations of the surface projection of the northern STEP fault proposed by Rosenbaum et al. (2008) in correspondence with the Sangineto Line (SL in map). Black lines are the longitudinal limits of the continuous slab domain, according to Neri et al. (2009). West of the Aeolian Islands in the Tyrrhenian Sea, the location of the east–west trending compressive margin as proposed by several investigators (see e.g. Billi et al., 2007) is schematically reported. Circles show the locations of the earthquakes of magnitude 6.0 and larger that have occurred after 1000 A.D. according to the CPTI11 catalog (Rovida et al., 2011; http://emidius.mi.ingv.it/CPTI11); seismogenic stresses are from Montone et al. (2012). Gray box holds the Pollino Mts. area and the black dots inside represent the earthquake locations of the 2010–2013 seismic crisis.
and southern Apennines, where detachment has already occurred (Cimini and Marchetti, 2006; Faccenna et al., 2005; Lucente et al., 2006; Montuori et al., 2007; Neri et al., 2009; Spakman and Wortel, 2004). According to Govers and Wortel (2005) this scenario, characterized by a progressive detachment of a retreating lithospheric slab, may have led to the generation of a Subduction-Transform Edge Propagator (STEP) fault that laterally decouples subducting lithosphere from non-subducting lithosphere in a scissor type of fashion. The lithosphere and mantle setting of the southern Apennines– Calabrian Arc border region has been deeply investigated in the last decades by means of regional and local seismic analyses reported in several previous papers (Barberi et al., 2004; Chiarabba et al., 2008; Chironi et al., 2000; Giacomuzzi et al., 2012; Montuori et al., 2007; Neri et al., 2002, 2009; Steckler et al., 2008). Different velocity patterns have been identified, providing evidences for first order crustal and sub-crustal heterogeneities and boundaries between the main crustal domains of Southern Apennines,
Calabrian, and Tyrrhenian regions (Fig. 1b). High velocity pattern at crustal depth beneath the Tyrrhenian Sea is commonly interpreted with the thinning of the Tyrrhenian crust and incipient oceanization (Barberi et al., 2004; Chiarabba et al., 2008; Finetti, 2005a, 2005b; Orecchio et al., 2011; Pepe et al., 2000). At greater depths, tomographic analysis evidenced that the Ionian subducting slab is in-depth continuous only beneath the central part of the Arc in southern Calabria while detachment has already occurred at the northern and southwestern edges of the arc itself, e.g. northern Calabria and northeastern Sicily, respectively (Fig. 1b; Neri et al., 2009). Beneath the southern Apennines lack of subcrustal seismicity and of high-velocity anomalies down to 200 km depth, together with a southwestward dipping high-velocity body at greater depths have been detected (Cimini, 1999; Cimini and Marchetti, 2006; Wortel and Spakman, 2000). In this framework the Calabrian Arc, a curved structure characterized by very heterogeneous seismotectonic regimes along its length (Cristofolini et al., 1985; Montone et al., 2004), together with the adjacent southern Apennines
C. Totaro et al. / Journal of Geodynamics 82 (2014) 87–97
sector, has been the site of destructive, magnitude class 6 and 7 earthquakes that have occurred both in recent and historical times (Fig. 1b; Galli et al., 2008; Neri et al., 2006). In the junction area between the Southern Apennines and the Calabrian Arc domains (e.g. at the northern edge of the Calabrian subduction zone, gray box in Fig. 1b), the Pollino Mts. area shows a structural and seismotectonic setting quite intricate, representing a highly deformed, very complex sector of great geodynamic interest and still not fully understood (see e.g. Bonini et al., 2011; Frepoli et al., 2011; Neri et al., 2012; Spina et al., 2011). Furthermore, in this sector a seismic gap was previously hypothesized by paleoseismological evidences associated with the lack of major earthquakes in historical catalogs (Cinti et al., 2002; Michetti et al., 2000). Since spring 2010 to 2013, this area has been affected by a seismic sequence with more than four thousands of small to moderate earthquakes (black dots in Fig. 1b). The strongest events occurred on May 28th 2012 (Ml 4.3) and on October 25th 2012 (Ml 5.0), respectively. Seismic activity has shown a fairly regular increase in number and energy since the onset of activity (Totaro et al., 2013). Epicentral distribution of the 2010–2013 sequence and the relative focal mechanism solutions evidenced two main clusters with a faint NNW-SSE-trend, showing predominantly normal faulting mechanisms with NE extension. The recent seismic activity has been interpreted as a seismic deformation occurring inside the southern Apennines extensional domain, specifically at its southern tip (Totaro et al., 2013). In order to better characterize the crustal structures of this peculiar region of southern Italy, which includes the Pollino Mts. area, we performed a local earthquake tomography by applying the LOTOS code (Koulakov, 2009), drawing also benefit from the large amount of data coming from the 2010–2013 seismic activity. The use of the LOTOS code, that represents the first application of this algorithm in the southern Italy crustal studies, allowed us to obtain highly resolved results strongly reducing the grid spacing with respect to previous investigations carried out in the region, and also to furnish a detailed Vp/Vs model not previously available for the study area.
2. Tectonic settings The large Calabrian Arc is a Cenozoic–Quaternary curved orogen (Fig. 1a) running from the NW–SE-trending Southern Apennines to the E–W-trending Sicilian Maghrebides (Carminati et al., 2012; Catalano et al., 1996; Lavecchia et al., 2007; Malinverno and Ryan, 1986; Minelli and Faccenna, 2010; Polonia et al., 2011; Rosenbaum and Lister, 2004). The central part of the Arc is characterized by the presence of a narrow subduction zone, which appears to be the only site of residual active subduction in the framework of a larger scale subduction process that has involved the western Mediterranean in the last tens Myrs (Fig. 1a; Faccenna et al., 2004; Wortel and Spakman, 2000). Many investigators (see, among others, Faccenna et al., 1996; Malinverno and Ryan, 1986) have suggested that geologic and geophysical data in the Calabrian Arc region can be interpreted in the framework of a geodynamic model assuming the co-existence of NW-SE convergence of Nubia and Europe plates and gravity-induced SE-ward rollback of the Ionian lithospheric slab subducting to NW beneath the Tyrrhenian lithosphere (Fig. 1b). The Ionian subducting slab is apparently in-depth continuous only beneath the central part of the Arc (southern Calabria), while it has already undergone detachment at the edges of the Arc itself, beneath northern Calabria and northeastern Sicily, respectively. Two transitional zones separating the different domains of continuous vs detached slab have been also detected (Neri et al., 2009). Subduction activity is now close to end but slow retreat seems to be still present (D’Agostino et al., 2011).
89
The space–time evolution of the rollback process (characterized by reduced intensity and greater concentration in the Calabrian Arc area since the Upper Pliocene) can be referred to a clear structural differentiation between the Calabrian Arc and the marginal tectonic units of Sicily and southern Apennines (Neri et al., 2012). In this framework, the Southern Apennines and the Calabrian Arc represent two of the most seismically-active regions of the entire Mediterranean area, with their long record of destructive historical earthquakes of magnitude as large as 7 (Amoruso et al., 2006; Galli and Bosi, 2004; Galli et al., 2008; Guidoboni et al., 2007; Monaco and Tortorici, 2000; Neri et al., 2006; black circles in Fig. 1b). The Calabrian Arc, uplifting at a rate of 0.5–1.2 mm/year in the last 1–0.7 Myrs, is characterized by internal deformation mainly accommodated by normal faulting (see e.g. Catalano and De Guidi, 2003; Catalano et al., 2003; Monaco and Tortorici, 2000). The strongest earthquakes can be related to an extensional regime with opening direction varying from NW–SE in southern Calabria to NE–SW in the northern part of the region (see e.g. Neri et al., 2004, 2005). In particular, the northern sector of the Calabria Arc (i.e., the Calabria–Lucania boundary) is a particular tectonically complex area as it joins the NNE–SSW-trending, NW–SE-extending curved Calabria with the NW–SE-trending, NE–SW-extending Southern Apennines, and is located on top of the northern edge of the subducting Ionian slab (Fig. 1b; Presti et al., 2013). To the north of this peculiar sector, in the Southern Apennines, slab break-off has already occurred, active subduction is therefore substantially absent, and an extensional stress regime compatible with the post-orogenic NE–SW extension affecting the entire Apennines has been detected (Presti et al., 2013). The Tyrrhenian and axial sectors of the Southern Apennines have undergone postorogenic extensional tectonics since at least earlymiddle Pleistocene time (Amicucci et al., 2008; Barchi et al., 2007). In this area, several destructive earthquakes were sourced, in recent and historical times, by extensional faults located closely west of the Apennines, considered as major seismogenic faults and subjected to a relatively uniform extensional stress regime with a nearly NE–SW orientation (Billi et al., 2011; Galli et al., 2006, 2008; Pantosti et al., 1993; Presti et al., 2013). Nevertheless most seismically active faults of this region are still poorly known. In the junction area between the southern Apennines extensional domain and the northern edge of the Calabrian Arc subduction zone, the Pollino Mts. area shows a structural and seismotectonic setting quite intricate and different views have been proposed by several authors (see, among others, Catalano et al., 2004; DISS Working Group, 2010; Michetti et al., 2000; Spina et al., 2011). Although normal mechanisms on NW-trending structures represent the most diffuse faulting style, differently oriented structures and various mechanisms are also reported (see among others Catalano et al., 2004; DISS Working Group, 2010; Spina et al., 2011; Van Dijk et al., 2000). This region lies at the northern section of the Calabrian accretionary wedge, which, since at least the Neogene, has been drifting southeastward onto the retreating Ionian subducting slab (Fig. 1a; Billi et al., 2007; Neri et al., 2012). Previous studies have indicated this as a highly deformed zone including shallow dynamics associated with the activity of a Subduction-Transform Edge Propagator (STEP) fault (Neri et al., 2012; Presti et al., 2013). Govers and Wortel (2005) have introduced the concept of STEP fault to represent lithosphere tearing at the edges of subduction trench zones. Even if location of the northern STEP fault in this subducting system is still debated, several authors have found evidences of shallow expressions of the Ionian slab tears in correspondence with the Sangineto Line (SL in Fig. 1b), at the Calabria–Lucania boundary (Govers and Wortel, 2005; Rosenbaum and Lister, 2004; Rosenbaum et al., 2008).
90
C. Totaro et al. / Journal of Geodynamics 82 (2014) 87–97
3. Data and method In order to increase the detail of our analysis with respect to previous study carried out in the same sector (Barberi et al., 2004; Calò et al., 2013; Chiarabba et al., 2008; Orecchio et al., 2011), we performed a tomographic inversion by applying the LOTOS algorithm (Koulakov, 2009). We used the data of local seismicity occurred in the study region and collected from the available national and local networks managed by INGV (www.ingv.it) and University of Calabria, respectively. We selected seismicity that has occurred between January 1981 and October 2012 at the southern Apennines–Calabrian Arc border region, within the limits of 39◦ N–40.5◦ N and 15◦ E–17.5◦ E (Fig. 2). Time picks from 250 stations were used, including those located outside the study area. It has to be noted that the recording conditions might vary from station to station depending on the type of instruments, installing conditions and the time in which the recorded events occurred. We want to remark that seismic monitoring in Italy has been strongly improved since the 1997, after the implementation of the new National Seismic Network (Amato and Mele, 2008), and more than 70% of our starting dataset events have occurred after the 1996 (i.e. in the period of higher quality of the recording network). To select the data we applied two criteria: first, the total number of P and S picks per event should be larger or equal 11; second, after locating the sources in the 1D starting model, we rejected the data with residuals greater than 1.5 and 2 s for P- and S-data, respectively. The final dataset used for tomographic inversion consisted of 2388 events (Fig. 2) and corresponding 28,108 P- and 14,605 S-arrival times (17.8 picks per event, on average). The distribution of these events over the study area is relatively homogeneous, ensuring high resolution of tomographic inversions. By using the LOTOS code (Koulakov, 2009), the calculations start with preliminary source locations based on the grid search method with the use of travel times computed in the 1D starting velocity model. At this step, we made the estimates for the optimal 1D velocity distributions by repeating the location procedure with dozens of different 1D models parameterized with 4–6 parameters (constant Vp/Vs ratio and 3–5 velocity values in different depth levels). The best model provided the maximum number of events and picks and minimum average deviations of residuals. Based on these criteria, we found the model with Vp/Vs = 1.76 and P-velocity values: 4.9 km/s at 0 km, 5.93 km/s at 10 km, 7 km/s at 30 km and 8 km/s at 40 km. Between the indicated depth levels, the velocity was linearly interpolated. The inversion procedure includes the iteration of location and inversion steps. Source locations in the 3D velocity model use 3D ray tracer based on the bending method, which ensures fast stable calculations of travel times of seismic rays between any two points in the study volume. The velocity distribution is parameterized with a set of nodes distributed in the study area according to the distribution of rays. In map view, the nodes are distributed with a regular step (5 km in our case); in the vertical direction, the grid spacing depends on the data density, but it cannot be smaller than a predefined value (3 km in our case). To avoid any bias related to the basic orientation of the grid, we performed the inversions for several grids with different basic orientations (0◦ , 22◦ , 45◦ and 67◦ in our case). The results computed for these grids are averaged in a regular mesh which is then used as an updated 3D velocity model in the next iteration. The inversion is performed simultaneously for the P- and Svelocity distributions, source corrections (four parameters for each source) and station corrections. The latest parameter appears to be very important in our case, because in our dataset we have some stations located outside the study area. Station corrections make possible to rid-off the factors affecting the travel time between the boundary of the study area and a single station. The inversion of the
matrix is performed using the LSQR algorithm (Paige and Saunders, 1982; Nolet, 1987). We regularize the solution by minimizing the gradient between neighboring nodes. The value of smoothing coefficients is estimated based on the results of synthetic modeling. As a result of inversion, we obtain the 3D distributions of P- and Svelocity anomalies. The distribution of Vp/Vs ratio is then computed by dividing the obtained values of absolute P- and S-velocities. We performed five iterations and then we obtained a reduction of the residual average deviations of ∼25% and ∼35% for P- and S-data. The values of remnant average deviations of the residuals are 0.25 s and 0.35 s, respectively.
4. Results and discussion In order to estimate a possible effect of noise on the resolution as well as the optimal values of inversion parameters, we performed several synthetic tests. In particular, in Fig. 3 we present three different checkerboard tests. In these models we defined periodic positive and negative velocity anomalies of different sizes and empty spacing between them: 20–10 km for Model 1, 15–7 km for Model 2 and 10–5 km for Model 3. Anomalies always had amplitude equal to 5% and opposite signs for P- and S-models, in order to ensure large variations of Vp/Vs ratio. The synthetic travel times were computed for the same sourcereceiver pairs as in the case of the observed data. The locations of real sources correspond to the solution obtained after five iterations of real data inversion. The synthetic data were perturbed with random noise having the average deviations of 0.1 and 0.2 for the P- and S-data, bringing to the same values of variance reduction as in the case of real data inversion. After computing the synthetic data, we “forgot” all information about source locations and velocity model and performed the reconstructions using same procedure and inversion parameters as in the case of the real data analysis, including the absolute source location, which may bias considerably the synthetic residuals. The reconstructions of P- and S anomalies, as well as Vp/Vs ratio obtained by division of the computed absolute P- and S-velocities, are presented in Fig. 3. It can be seen that Models 1 and 2 are robustly reconstructed in most parts of the study area. Model 3, with the anomaly sizes of 10 km, appears to be resolved only in the central part where there is maximum concentration of events and stations. It is important that the used calculation method ensures robust reconstruction of Vp/Vs ratio despite significant noise in the data which might cause some instabilities in the reconstructions of P- and S models. Based on these tests, we can evaluate the minimum size of patterns which can be resolved in different parts of the area. Furthermore, these tests show that the inversion parameters, which were identical for real and synthetic data inversions, are adequate and provide optimal quality reconstructions. The new seismic velocity model obtained in the present study for the southern Apennines–Calabrian Arc border region (Figs. 4 and 5) is reported as the resulting seismic velocity anomalies with respect to the aforesaid 1D velocity model, as well as Vp/Vs ratio, in horizontal and vertical sections. Absolute velocities along the vertical sections are shown in Fig. 6. The figures present also the distributions of relocated seismic events at corresponding depths and close to the presented profiles, respectively. The horizontal sections present seismic structures beneath the southern Apennines–Calabrian Arc border region in the uppermost crust (3 km depth), middle crust (15 km) and lower crust (30 km). The model is shown only in areas with sufficient data coverage: the values are masked if distance to the nearest parameterization node is larger than 10 km. In the upper crust layer a small size low-velocity anomaly of P-waves is detectable just south of the Pollino Mts. area. It seems
C. Totaro et al. / Journal of Geodynamics 82 (2014) 87–97
91
Fig. 2. Epicentral distribution of the starting inversion dataset consisting of ∼3600 earthquakes (dots) and location of the recording seismic stations (triangles).
to separate the two adjacent domains of Calabrian Arc and southern Apennines, characterized by higher velocity values. At same depth, low S-waves velocity anomaly is also present, and high values of Vp/Vs ratio characterize the Pollino Mts. region (Fig. 4). This section appears to reproduce the variety of the sedimentary cover, intrusive and metamorphic bodies. In Fig. 7, the comparison between the shallow tomography sections for P- and S anomalies with the surface geology (Compagnoni et al., 2011) is shown. It can be seen that P-velocity patterns correlate with the main geological structures. For example, a large positive anomaly “1” lies in correspondence with the Calabride mountain complex which is composed of rigid metamorphic and igneous rocks of Hercynian age. The Pollino Mts. area lies in correspondence with the Apennine and Ligurid/Sicilid units (“2” and “3”, respectively), which is a sector characterized by high P-velocities in the shallowmost crust. In the depression between these two mountain areas we observe a small, but very contrasted feature with low P- and S-velocities (“4”) which might represent a small sedimentary basin. The Apulian Platform, which is located at the margin of the study area,
matches with high P-velocity anomaly “5”. The fair correlation of P-velocity distribution with geology can be explained by the sensitivity of bulk elastic parameters to the composition and rigidity of rocks. Significantly different structures are observed at shallow depths for the S-velocity anomalies. A strong negative anomaly is observed in the Pollino massif (“3”), where P-model presents high velocity. Coexistence of high P and low S velocities at shallow depths may testify strong fracturing and saturation of rocks with fluids. This massif is associated with high Vp/Vs ratio pattern, which is particularly strong in the area of huge seismic activity (i.e. the area affected by the 2010–2013 Pollino swarm). The Calabride block “1” is associated with high-velocity S-anomaly, having a size considerably smaller than the relative shape indicated in the geological map. For the shallow layers, the S-velocity is rather more sensitive to the fracturing and saturation with fluids than to the composition. It is interesting that Vp/Vs anomalies are mostly associated with areas of high relief where maximum intensity of tectonic processes is expected. Thus joint consideration of P, S-velocity anomalies and
92
C. Totaro et al. / Journal of Geodynamics 82 (2014) 87–97
Fig. 3. Checkerboard tests performed to assess the spatial resolution and to estimate the optimal values of inversion parameters. Periodic positive and negative velocity anomalies of different sizes and empty spacing between them have been defined: 20–10 km for Model 1 (top row), 15–7 km for Model 2 (middle row) and 10–5 km for Model 3 (bottom row). Anomalies always had amplitude equal to 5% and opposite signs for P- and S-models (left and central columns), in order to ensure large variations of Vp/Vs ratio (right column).
C. Totaro et al. / Journal of Geodynamics 82 (2014) 87–97
93
Fig. 4. New seismic velocity model for the southern Apennines–Calabrian Arc border region obtained in the present study. Tomographic results are reported in terms of percentage variation of P- and S-wave velocity with respect to the optimal 1D reference model (top and middle rows) and in terms of Vp/Vs ratio (bottom row), respectively. Plates correspond to uppermost crust (3 km depth), middle crust (15 km) and lower crust (30 km), respectively. The model is shown only in areas with sufficient data coverage. 3D earthquake locations coming from the inversion are also shown, separately plotted according to the minimum vertical grid spacing: 0–4.5 km; 13.5–16.5 km; 28.5–31.5 km, respectively. Gray dashed line in the top left panel is the location of the surface projection of the northern STEP fault by Rosenbaum et al. (2008) in correspondence with the Sangineto Line (SL in map). “PM” stands for Pollino Mts., black lines indicate the profiles relative to the vertical sections reported in Figs. 5 and 6.
Vp/Vs ratio gives independent sights to the mechanical characteristics of rocks in the uppermost layers. In the middle and lower crust (15 and 30 km depth in Fig. 4) the structure becomes considerably different with respect to the shallow section, and a clear change in the velocity pattern is evident moving from the Tyrrhenian Sea to the Calabrian Arc domains. We
observe a strong contrast between a large low-velocity anomaly beneath the continental part of Italy and high-velocity beneath the offshore areas of the Tyrrhenian Sea, which is similarly expressed in both P- and S-velocity models. These results show a general agreement with the main crustal features of the study region revealed also by seismic lines and gravimetric analyses, indicating crustal
94
C. Totaro et al. / Journal of Geodynamics 82 (2014) 87–97
Fig. 5. Vertical sections of the tomographic model along the profiles indicated in the maps of Fig. 4. Percentage variation of P- and S-wave velocity with respect to the optimal 1D reference model (top and middle rows) and in terms of Vp/Vs ratio (bottom row), respectively are reported. Distribution of relocated events lying within ±20 km of the vertical planes is also shown.
thickening moving from southern Tyrrhenian to Calabrian Arc (Neri et al., 2012 and references therein). Looking at the vertical sections, P-wave velocity patterns reported in section 1 (left panels in Fig. 5) well depict the separation between the southern Apennines and the Calabrian Arc domains by the low velocity zone present in correspondence with the Pollino Mts. area. In-depth continuity and dimension increasing of this low velocity anomaly are also evident. In cross-section 2 (right panels in Fig. 5) the sharp separation between high and low seismic velocities in correspondence with the Tyrrhenian coast is clearly evident in the whole depth range and it may represent the
transition from the thinner Tyrrhenian crust to the thicker one beneath Calabria. Looking at the absolute velocity patterns along the vertical sections in Fig. 6, we can associate the Moho interface with contour lines of ∼7.4 km/s and 4.2 km/s in P- and S-velocity models, respectively (“red-to-orange” layer). We can then estimate that the crustal thickness varies from ∼25–30 km beneath the offshore area to ∼35–40 km beneath the continent, according to the results obtained in previous studies (see e.g. Piana Agostinetti and Amato, 2009). Same absolute velocity sections can be used to identify the sedimentary cover thickness. If we assume the bottom of sediments at 5.5 km/s and 3.0 km/s for P- and S-velocities (“green”
Fig. 6. Absolute velocity patterns along the same vertical sections of Fig. 5.
C. Totaro et al. / Journal of Geodynamics 82 (2014) 87–97
95
Fig. 7. Comparison between the shallow tomography sections for P- and S anomalies and surface geology (redrawn from Compagnoni et al., 2011).
Fig. 8. Focus on the southern Apennines–Calabrian Arc transition area. Low velocity anomalies of −4% are shown for the depth levels 3 km, 15 km, and 30 km, (continuous, dashed, and dotted black lines, respectively). Gray dashed line is the locations of the shallow projection of the northern STEP fault proposed by Rosenbaum et al. (2008) in correspondence with the Sangineto Line (SL in map). Thick gray line marks the northern limit of the continuous slab domain, according to Neri et al. (2009).
The tomographic results for P- and S anomalies at the upper crust layer show a fairly good correlation between velocity patterns and the main geological features of the study area (Fig. 7). In the middle and lower crust, velocity distribution clearly evidences the separation between the southern Apennines and the Calabrian Arc domains by a low velocity zone present at all depths and showing a dimensional enlargement with depth. In this transitional area, the existence of lithospheric-scale tear faults, in particular those rooting the left lateral Sangineto Line (Fig. 8), has been previously suggested by several authors (Govers and Wortel, 2005; Panza et al., 2007; Rosenbaum et al., 2008). The lateral extension and the quite relevant distance of this sector from the slab edge do not permit to properly define this region as a direct expression of a present-day STEP activity at the northern edge of the subducting slab. It may perhaps represent the present-day activity of structural systems that eventually worked like a STEP zone in the past (Catalano et al., 2001; Govers and Wortel, 2005; Rosenbaum et al., 2008). In this context it has to be noted that the surface projection of the Sangineto Line lies in correspondence with the low velocity anomalies (−4%) identified at all depth levels in the present study (Fig. 8). This may suggests that the structural systems that eventually worked like a STEP zone in the past may have opened the route to a possible fluid upwelling, and this could be evidenced by the enlargement of the low velocity anomaly size with depth.
Acknowledgments layer), we can identify the areas with thick sediments (mostly in depressions) and zones where basement appears to be closer to the surface. High earthquake concentration in the high-Vp zones (Figs. 4 and 5) and lack of seismicity in the low-Vp and lowVs ones are evident, especially in correspondence with the Pollino Mts. area. In this small sector, low velocity anomalies are present at all depths and a constant enlargement of their size can be noted with depth increasing (Fig. 5). The distributions of high Vp/Vs ratio in the upper crust clearly correlate with areas of huge seismicity representing strongly fractured rocks with likely high fluid content. 5. Conclusions The present study represents the first application of the LOTOS algorithm to crustal analyses in southern Italy. This code allowed us to obtain a new highly resolved seismic velocity model for the southern Apennines–Calabrian Arc border region furnishing also a detailed Vp/Vs model not previously available for the study area.
This research has been supported by Istituto Nazionale di Geofisica e Vulcanologia and Dipartimento della Protezione Civile (DPC) through the INGV-DPC 2014 S1 Project and by the Project of National Interest PRIN 2010-2011 “Geodinamica attiva e recente dell’Arco Calabro e del complesso di accrezione nel Mar Ionio”, funded by Ministero Istruzione Universita` e Ricerca (MIUR).
References Amato, A., Mele, F.M., 2008. Performance of the INGV National Seismic Network from 1997 to 2007. Ann. Geophys. 51, 2–3. Amicucci, L., Barchi, M.R., Montone, P., Rubiliani, N., 2008. The Vallo di Diano and Auletta extensional basins in the southern Apennines (Italy): a simple model for a complex setting. Terra Nova 20, 475–482, http://dx.doi.org/10.1111/ j.1365-3121.2008.00841.x. Amoruso, A., Crescentini, L., Neri, G., Orecchio, B., Scarpa, R., 2006. Spatial relation between the 1908 Messina Straits earthquake slip and recent earthquake distribution. Geophys. Res. Lett. 33, http://dx.doi.org/10.1029/2006GL027227. Barberi, G., Cosentino, M.T., Gervasi, A., Guerra, I., Neri, G., Orecchio, B., 2004. Crustal seismic tomography in the Calabrian Arc region, south Italy. Phys. Earth Planet. Inter. 147, 297–314.
96
C. Totaro et al. / Journal of Geodynamics 82 (2014) 87–97
Barchi, M., Amato, A., Cippitelli, G., Merlini, S., Montone, P., 2007. Extensional tectonics and seismicity in the axial zone of the Southern Apennines. Ital. J. Geosci., Spec. Iss. 7, 47–56. Billi, A., Presti, D., Faccenna, C., Neri, G., Orecchio, B., 2007. Seismotectonics of the Nubia plate compressive margin in the south-Tyrrhenian region, Italy: clues for subduction inception. J. Geophys. Res. 112, B08302, http://dx.doi.org/10.1029/2006JB004837. Billi, A., Faccenna, C., Bellier, O., Minelli, L., Neri, G., Piromallo, C., Presti, D., Scrocca, D., Serpelloni, E., 2011. Recent tectonic reorganization of the Nubia-Eurasia convergent boundary heading for the closure of the western Mediterranean. Bull. Soc. Géol. Fr. 182, 279–303. Bonini, M., Sani, F., Moratti, G., Benvenuti, M.G., 2011. Quaternary evolution of the Lucania Apennine thrust front area (Southern Italy), and its relations with the kinematics of the Adria Plate boundaries. J. Geodyn. 51, 125–140, http://dx.doi.org/10.1016/j.jog.2010.01.010. Calais, E., DeMets, C., Nocquet, J.-M., 2003. Evidence for a post-3.16-Ma change in Nubia-Eurasia-North America plate motions? Earth Planet. Sci. Lett. 216, 81–92. Calò, M., Parisi, L., Luzio, D., 2013. Lithospheric P- and S-wave velocity models of the Sicilian area using WAM tomography: procedure and assessments. Geophys. J. Int. 195 (1), 625–649, http://dx.doi.org/10.1093/gji/ggt252. Carminati, E., Lustrino, M., Doglioni, C., 2012. Geodynamic evolution of the central and western Mediterranean: tectonics vs. igneous petrology constraints. Tectonophysics, http://dx.doi.org/10.1016/j.tecto.2012.01.026. Catalano, R., Di Stefano, P., Sulli, A., Vitale, F.P., 1996. Paleogeography and structure of the central Mediterranean: Sicily and its offshore area. Tectonophysics 260, 291–323, http://dx.doi.org/10.1016/0040-1951(95)00196-4. Catalano, R., Doglioni, C., Merlini, S., 2001. On the Mesozoic Ionian basin. Geophys. J. Int. 144, 49–64. Catalano, S., De Guidi, G., 2003. Late Quaternary uplift of northeastern Sicily: relation with the active normal faulting deformation. J. Geodyn. 36, 445–467. Catalano, S., De Guidi, G., Monaco, C., Tortorici, G., Tortrici, L., 2003. Long-term behaviour of the late Quaternary normal faults in the Straits of Messina area (Calabrian arc): structural and morphological constraints. Quat. Int. 101-102, 81–91. Catalano, S., Monaco, C., Tortorici, L., Paltrinieri, W., Steel, N., 2004. Neogene–Quaternary tectonic evolution of the southern Apennines. Tectonics 23, http://dx.doi.org/10.1029/2003TC001512. Chiarabba, C., De Gori, P., Speranza, F., 2008. The Southern Tyrrhenian Subduction zone: deep geometry, magmatism and plio-pleistocene evolution. Earth Planet. Sci. Lett. 268, 408-423, http://dx.doi.org/10.1016/j.epsl.2008.01.036. Chironi, C., De Luca, L., Guerra, I., Luzio, D., Moretti, A., Vitale, M., Sea Land Group, 2000. Crustal structures of the Southern Tyrrhenian Sea and the Sicily Channel on the basis of the M25, M26, M28, M39 WARR profiles. Boll. Soc. Geol. Ital. 119, 189–203. Cimini, G.B., 1999. P-wave deep velocity structure of the southern Tyrrhenian subduction zone from nonlinear teleseismic traveltime tomography. Geophys. Res. Lett. 26 (24), pp. 3,709–3,712. Cimini, G.B., Marchetti, A., 2006. Deep structure of peninsular Italy from seismic tomography and subcrustal seismicity. Ann. Geophys. 49 (1), 331–345. Cinti, F.R., Moro, M., Pantosti, D., Cucci, L., D’Addezio, G., 2002. New constraints on the seismic history of the Castrovillari fault in the Pollino gap (Calabria, southern Italy). J. Seismol. 6, 199–217. Compagnoni, B., Galluzzo, F., Bonomo, R., Capotorti, F., D’Ambrogi, C., Di Stefano, R., Graziano, R., Martarelli, L., Pampaloni, M.L., Pantaloni, M., Ricci, V., Tacchia, D., Masella, G., Pannuti, V., Ventura, R., Vitale, V., 2011. Carta Geologica d’Italia in Scala 1:1.000.000. http://isprambiente.gov.it/ Cristofolini, R., Ghisetti, F., Scarpa, R., Mezzani, L., 1985. Character of stress field in the Calabrian Arc and Southern Apennines (Italy) as deduced by geological, seismological and volcanolgical information. Tectonophysics 117, 39–58. ´ M.R., SeeD’Agostino, N., D’Anastasio, E., Gervasi, A., Guerra, I., Nedimovic, ber, L., Steckler, M., 2011. Forearc extension and slow rollback of the Calabria Arc from GPS measurements. Geophys. Res. Lett. 38, L17304, http://dx.doi.org/10.1029/2011GL048270. DISS Working Group, 2010. Database of Individual Seismogenic Sources (DISS), Version 3.1.1: A compilation of potential sources for earthquakes larger than M 5.5 in Italy and surrounding areas., http://dx.doi.org/10.6092/INGV.IT-DISS3.1.1 http://diss.rm.ingv.it/diss/,(INGV 2010-Istituto Nazionale%20di%20Geofisica% 20e%20Vulcanologia-All%20rights%20reserved Faccenna, C., Davy, P., Brun, J.P., Funiciello, R., Giardini, D., Mattei, M., Nalpas, T., 1996. The dynamics of back-arc extension: an experimental approach to the opening of the Tyrrhenian Sea. Geophys. J. Int. 126, 781–795. Faccenna, C., Piromallo, C., Crespo-Blanc, A., Jolivet, L., 2004. Lateral slab deformation and the origin of the western Mediterranean arcs. Tectonics 23, TC1012, http://dx.doi.org/10.1029/2002TC001488. Faccenna, C., Civetta, L., D’Antonio, M., Funiciello, F., Margheriti, L., Piromallo, C., 2005. Constraints on mantle circulation around the deforming Calabrian slab. Geophys. Res. Lett. 32 (6), L06311. Finetti, I.R., 2005a. Depth contour map of the Moho discontinuity in the central Mediterranean region from new CROP seismic data. In: Finetti, I.R. (Ed.), CROP PROJECT: Deep Seismic Exploration of the Central Mediterranean and Italy. Elsevier, Amsterdam, pp. 597–606. Finetti, I.R., 2005b. The Calabrian Arc and subducting Ionian slab from the new CROP seismic data. In: Finetti, I.R. (Ed.), CROP PROJECT: Deep Seismic Exploration of the Central Mediterranean and Italy. Elsevier, Amsterdam, pp. 393–412.
Frepoli, A., Maggi, C., Cimini, G.B., Marchetti, A., Chiappino, M., 2011. Seismotectonic of Southern Apennines from recent passive seismic experiments. J. Geodyn. 51, 110–124. Galli, P., Bosi, V., 2004. Catastrophic 1638 earthquakes in Calabria (southern Italy): new insights from paleoseismological investigation. J. Geophys. Res. 108, http://dx.doi.org/10.1029/2001JB00171. Galli, P., Bosi, V., Piscitelli, S., Giocoli, A., Scionti, V., 2006. Late Holocene earthquakes in Southern Apennines: paleoseismology of the Caggiano fault. Int. J. Earth Sci. 95, 855–870. Galli, P., Galadini, F., Pantosti, D., 2008. Twenty years of paleoseismology in Italy. Earth Sci. Rev. 88, 89–117. Giacomuzzi, G., Civalleri, M., De Gori, P., Chiarabba, C., 2012. A 3D Vs model of the upper mantle beneath Italy: insight on the geodynamics of central Mediterranean. Earth Planet. Sci. Lett. 335–336, 105–120. Govers, R., Wortel, M.J.R., 2005. Lithosphere tearing at STEP faults: response to edges of subduction zones. Earth Planet. Sci. Lett. 236, 505–523. Guidoboni, E., Ferrari, G., Mariotti, D., Comastri, A., Tarabusi, G., Valensise, G., 2007. CFTI4Med. Catalogue of Strong Earthquakes in Italy 461 B.C. 1997 and Mediterranean Area 760 B.C. 1500. An Advanced Laboratory of Historical Seismology. http://storing.ingv.it/cfti4med/ Koulakov, I., 2009. LOTOS code for local earthquake tomographic inversion: benchmarks for testing tomographic algorithms. Bull. Seismol. Soc. Am. 99 (1), 194–214, http://dx.doi.org/10.1785/0120080013. Lavecchia, G., Ferrarini, F., de Nardis, R., Visini, F., Barbano, M.S., 2007. Active thrusting as a possible seismogenic source in Sicily (southern Italy): some insights from integrated structural-kinematic and seismological data. Tectonophysics 445, 145–167, http://dx.doi.org/10.1016/j.tecto.2007.07.007. Lucente, F.P., Margheriti, L., Piromallo, C., Barruol, G., 2006. Seismic anisotropy reveals the long route of the slab through the westerncentral Mediterranean mantle. Earth Planet. Sci. Lett. 241, 517–529. Malinverno, A., Ryan, W., 1986. Extension in the Tyrrhenian Sea and shortening in the Apennines as result of arc migration driven by sinking of the lithosphere. Tectonics 5, 227–245. Michetti, A.M., Ferreli, L., Esposito, E., Porfido, S., Blumetti, A.M., Vittori, E., Serva, L., Roberts, G.P., 2000. Ground effects during the 9 September 1998, Mw = 5.6 Lauria earthquake and the seismic potential of the “aseismic” Pollino region in Southern Italy. Seismol. Res. Lett. 71, 31–46. Minelli, L., Faccenna, C., 2010. Evolution of the Calabrian accrewedge (central Mediterranean). Tectonics 29, TC4004, tionary http://dx.doi.org/10.1029/2009TC002562. Monaco, C., Tortorici, L., 2000. Active faulting in the Calabrian arc and eastern Sicily. J. Geodyn. 29, 407–424. Montone, P., Mariucci, M.T., Pondrelli, S., Amato, A., 2004. An improved stress map for Italy and surrounding regions (central Mediterranean). J. Geophys. Res. 109, B10410, http://dx.doi.org/10.1029/2003JB002703. Montone, P., Mariucci, M.T., Pierdominici, S., 2012. The Italian present-day stress map. Geophys. J. Int. 189 (2), 705–716. Montuori, C., Cimini, G.B., Favali, P., 2007. Teleseismic tomography of the southern Tyrrhenian subduction zone: new results from seafloor and land recordings. J. Geophys. Res. 112, B03311. Neri, G., Barberi, G., Orecchio, B., Aloisi, M., 2002. Seismotomography of the crust in the transition zone between the southern Tyrrhenian and Sicilian tectonic domains. Geophys. Res. Lett. 29 (23), 10, http://dx.doi.org/10.1029/2002GL015562. Neri, G., Barberi, G., Oliva, G., Orecchio, B., 2004. Tectonic stress and seismogenic faulting in the area of the 1908 Messina earthquake, south Italy. Geophys. Res. Lett. 31, 10, http://dx.doi.org/10.1029/2004GL019742. Neri, G., Barberi, G., Oliva, G., Orecchio, B., 2005. Spatial variations of seismogenic stress orientations in Sicily, south Italy. Phys. Earth Planet. Inter. 148, 175–191. Neri, G., Oliva, B., Orecchio, B., Presti, D., 2006. A possible seismic gap within a highly seismogenic belt crossing Calabria and eastern Sicily, Italy. Bull. Seismol. Soc. Am. 96, 321–331. Neri, G., Orecchio, B., Totaro, C., Falcone, G., Presti, D., 2009. Subduction beneath southern Italy close the ending: results from seismic tomography. Seismol. Res. Lett. 80, 63–70. Neri, G., Marotta, A.M., Orecchio, B., Presti, D., Totaro, C., Barzaghi, R., Borghi, A., 2012. How lithospheric subduction changes along the Calabrian Arc in southern Italy: geophysical evidences. Int. J. Earth Sci. 101, 1949–1969. Nocquet, J.-M., Calais, E., 2004. Geodetic measurements of crustal deformation in the western Mediterranean and Europe. Pure Appl. Geophys. 161, 661–681, http://dx.doi.org/10.1007/s00024-003-2468-z. Nocquet, J.-M., 2012. Present-day kinematics of the Mediterranean: a comprehensive overview of GPS results. Tectonophysics 579, 220–242. Nolet, G., 1987. Seismic wave propagation and seismic tomography. In: Nolet, G. (Ed.), Seismic Tomography. Reidel, Dordrecht, pp. 7–29. Orecchio, B., Presti, D., Totaro, C., Guerra, I., Neri, G., 2011. Imaging the velocity structure of the Calabrian Arc region (South Italy) through the integration of different seismological data. Boll. Geof. Teor. Appl. 52 (4), 625–638, http://dx.doi.org/10.4430/bgta0023. Paige, C.C., Saunders, M.A., 1982. LSQR: an algorithm for sparse linear equations and sparse least squares. ACM Trans. Math. Softw. 8, 43–71. Pantosti, D., Schwartz, D., Valensise, G., 1993. Paleoseismology along the 1980 surface rupture in Irpinia Fault: implications for earthquake
C. Totaro et al. / Journal of Geodynamics 82 (2014) 87–97 recurrence in the Southern Apennines, Italy. J. Geophys. Res. 98 (4), 6561–6577. Panza, G.F., Peccerillo, A., Aoudia, A., Farina, B., 2007. Geophysical and petrological modelling of the structure and composition of the crust and upper mantle in complex geodynamic settings: the Tyrrhenian Sea and surroundings. Earth Sci. Rev. 80, 1–46, http://dx.doi.org/10.1016/j.earscirev.2006.08.004. Pepe, F., Bertotti, G., Cella, F., Marsella, E., 2000. Rifted margin formation in the south Tyrrhenian Sea: a high-resolution seismic profile across the north Sicily passive continental margin. Tectonics 19, 241–257. Piana Agostinetti, N., Amato, A., 2009. Moho depth and Vp = Vs ratio in peninsular Italy from teleseismic receiver functions. J. Geophys. Res. Solid Earth 114, B6. Polonia, A., Torelli, L., Mussoni, P., Gasperini, L., Artoni, A., Klaeschen, D., 2011. The Calabrian Arc subduction complex in the Ionian Sea: regional architecture, active deformation, and seismic hazard. Tectonics 30, TC5018, http://dx.doi.org/10.1029/2010TC002821. Presti, D., Billi, A., Orecchio, B., Totaro, C., Faccenna, C., Neri, G., 2013. Earthquake focal mechanisms, seismogenic stress, and seismotectonics of the Calabrian Arc, Italy. Tectonophysics, http://dx.doi.org/10.1016/j.tecto.2013.01.030. Rosenbaum, G., Lister, G.S., 2004. Neogene and Quaternary rollback evolution of the Tyrrhenian Sea, the Apennines, and the Sicilian Maghrebides. Tectonics 23, TC1013, http://dx.doi.org/10.1029/2003TC001518. Rosenbaum, G., Gasparon, M., Lucente, F.P., Peccerillo, A., Miller, M.S., 2008. Kinematics of slab tear faults during subduction segmentation and implications for Italian magmatism. Tectonics 27 (2).
97
Rovida, A., Camassi, R., Gasperini, P., Stucchi, M., 2011. CPTI11, la versione 2011 del Catalogo Parametrico dei Terremoti Italiani, Milano, Bologna (in Italian) http://emidius.mi.ingv.it/CPTI11 Serpelloni, E., Vannucci, G., Pondrelli, S., Argnani, A., Casula, G., Anzidei, M., Baldi, P., Gasperini, P., 2007. Kinematics of the western Africa–Eurasia plate boundary from focal mechanism and GPS data. Geophys. J. Int., http://dx.doi.org/10.1111/j.1365-246X.2007.03367.x. Spakman, W., Wortel, R., 2004. A tomographic view on western Mediterranean geodynamics. In: Cavazza, W. (Ed.), The Transmed Atlas: The Mediterranean Region from Crust to Mantle. Springer, New York, pp. 31–52. Spina, V., Tondi, E., Mazzoli, S., 2011. Complex basin development in a wrenchdominated back-arc area: tectonic evolution of the Crati basin, Calabria, Italy. J. Geodyn. 51, 90–109. Steckler, M.S., Piana Agostinetti, N., Wilson, C.K., Roselli, P., Seeber, L., Amato, A., Lerner-Lam, A., 2008. Crustal structure in the Southern Apennines from teleseismic receiver functions. Geology 36, 155–158. Totaro, C., Presti, D., Billi, A., Gervasi, A., Orecchio, B., Guerra, I., Neri, G., 2013. The ongoing seismic sequence at the Pollino Mountains, Italy. Seismol. Res. Lett. 84 (6), 955–962, http://dx.doi.org/10.1785/0220120194. Van Dijk, J.P., Bello, M., Brancaleoni, G.P., Cantarella, G., Costa, V., Frixa, A., Golfetto, F., Merlini, S., Riva, M., Torricelli, S., Toscano, C., Zerilli, A., 2000. A regional structural model for the northern sector of the Calabrian Arc (southern Italy). Tectonophysics 324, 267–320. Wortel, R., Spakman, W., 2000. Subduction and slab detachment in the Mediterranean-Carpathian region. Science 290, 1910–1917.
Contents lists available at ScienceDirect
Journal of Geodynamics journal homepage: http://www.elsevier.com/locate/jog
Detailed crustal structure in the area of the southern Apennines–Calabrian Arc border from local earthquake tomography C. Totaro a,∗ , I. Koulakov b , B. Orecchio a , D. Presti a a b
Physics and Earth Sciences Department, Messina University, Viale F. StagnoD’Alcontres 31, 98166 Messina, Italy Institute of Petroleum Geology and Geophysics, SB RAS, Prospekt Akademika Koptyuga 3, Novosibirsk 630090, Russia
a r t i c l e
i n f o
Article history: Received 28 February 2014 Received in revised form 16 July 2014 Accepted 21 July 2014 Available online 27 July 2014 Keywords: Seismic tomography Crustal structure Calabrian Arc Southern Apennines
a b s t r a c t We present a new seismic velocity model for the southern Apennines–Calabrian Arc border region with the aim to better define the crustal structures at the northern edge of the Ionian subduction zone. This sector also includes the Pollino Mts. area, where a seismic sequence of thousands of small to moderate earthquakes has been recorded between spring 2010 and 2013. In this sector a seismic gap was previously hypothesized by paleoseismological evidences associated with the lack of major earthquakes in historical catalogs. To perform the tomographic inversion we selected ca. 3600 earthquakes that have occurred in the last thirty years and recorded by permanent and temporary networks managed by INGV and Calabria University. Using for the first time the Local Tomography Software for passive tomography inversion (LOTOS hereinafter) to crustal analysis in southern Italy, we have computed the distribution of Vp, Vs, and the Vp/Vs ratio. The obtained velocity model, jointly evaluated with results of synthetic modeling, as well as with the hypocenter distribution and geological information, gives us new constraints on the geodynamical and structural knowledge of the study area. The comparison between the shallow tomography sections and surface geology shows good correlation between velocity patterns and the main geological features of the study area. In the upper crust a lowvelocity anomaly of P- and S-waves is detectable beneath the Pollino Mts. area and seems to separate the Calabrian and southern Apennines domains, characterized by higher velocities. The distributions of high Vp/Vs ratio, representing strongly fractured rocks with likely high fluid content, clearly correlate with areas of significant seismicity. In the lower crust we detect a clear transition from high to low seismic velocities in correspondence with the Tyrrhenian coast of the study area, which may represent the transition from the thinner Tyrrhenian crust to the thicker one beneath Calabria. In this area, also characterized by a progressive detachment of a retreating lithospheric slab, the generation of a Subduction-Transform Edge Propagator (STEP) fault zone, that laterally decouples subducting lithosphere from non-subducting lithosphere in a scissor type of fashion, may have taken place. These conditions imply the existence of a kinematic decoupling which allows for differential movement between the Calabrian Arc and the southern Apennine chain. The low velocity anomaly separating the southern Apennines and the Calabrian Arc domain may be related to fluid upwelling occurring in correspondence with the northern edge of the Calabrian subducting slab. © 2014 Elsevier Ltd. All rights reserved.
1. Introduction The western Mediterranean area, located at the contact belt between the slowly convergent African and Eurasian plates (Calais et al., 2003; Nocquet and Calais, 2004; Serpelloni et al., 2007), has been the site of a continental-scale lithospheric subduction
∗ Corresponding author. Tel.: +39 090 6765102; fax: +39 090 392333. E-mail address: [email protected] (C. Totaro). http://dx.doi.org/10.1016/j.jog.2014.07.004 0264-3707/© 2014 Elsevier Ltd. All rights reserved.
process, the evolution of which in the last 30 million years is marked by the eastward migration of the retreating subduction hinge (Fig. 1a; Wortel and Spakman, 2000). Most of the subduction system has already undergone detachment of the subducting lithosphere with the exception of the central, most arcuate portion of the system, the Calabrian Arc in southern Italy (Fig. 1b; Neri et al., 2009 and references therein). Different states of the subduction process can be related to the progressive change of lithospheric structure near the retreating trench zone and to strong lithospheric heterogeneity between the Calabrian Arc and the marginal tectonic units of Sicily
88
C. Totaro et al. / Journal of Geodynamics 82 (2014) 87–97
Fig. 1. (a) Map of the Mediterranean region with the western Mediterranean plate boundary evolution in the last 30 Myrs (redrawn from Wortel and Spakman, 2000, with modifications according to Neri et al., 2009). The space–time evolution of the boundary marks the process of rollback of the subducting lithosphere and the related trench retreat until the present-day location near the Ionian shoreline of Calabria. Black arrows indicate the present motion of Africa relative to Europe (Calais et al., 2003; Nocquet and Calais, 2004; Nocquet, 2012). In (b) the solid curve with the sawtooth pattern, pointing in the direction of subduction, indicates the present-day location of the Calabrian Arc subducting system. According to the local earthquake tomography by Neri et al. (2009), black sawteeth indicate the continuous subducting slab while white sawteeth the plate boundary segments where slab detachment has already occurred. The white arrow shows the sense of the subducting slab rollback. The black arrows indicate the present motion of Africa relative to Europe (Calais et al., 2003; Nocquet and Calais, 2004). Gray dashed line is the locations of the surface projection of the northern STEP fault proposed by Rosenbaum et al. (2008) in correspondence with the Sangineto Line (SL in map). Black lines are the longitudinal limits of the continuous slab domain, according to Neri et al. (2009). West of the Aeolian Islands in the Tyrrhenian Sea, the location of the east–west trending compressive margin as proposed by several investigators (see e.g. Billi et al., 2007) is schematically reported. Circles show the locations of the earthquakes of magnitude 6.0 and larger that have occurred after 1000 A.D. according to the CPTI11 catalog (Rovida et al., 2011; http://emidius.mi.ingv.it/CPTI11); seismogenic stresses are from Montone et al. (2012). Gray box holds the Pollino Mts. area and the black dots inside represent the earthquake locations of the 2010–2013 seismic crisis.
and southern Apennines, where detachment has already occurred (Cimini and Marchetti, 2006; Faccenna et al., 2005; Lucente et al., 2006; Montuori et al., 2007; Neri et al., 2009; Spakman and Wortel, 2004). According to Govers and Wortel (2005) this scenario, characterized by a progressive detachment of a retreating lithospheric slab, may have led to the generation of a Subduction-Transform Edge Propagator (STEP) fault that laterally decouples subducting lithosphere from non-subducting lithosphere in a scissor type of fashion. The lithosphere and mantle setting of the southern Apennines– Calabrian Arc border region has been deeply investigated in the last decades by means of regional and local seismic analyses reported in several previous papers (Barberi et al., 2004; Chiarabba et al., 2008; Chironi et al., 2000; Giacomuzzi et al., 2012; Montuori et al., 2007; Neri et al., 2002, 2009; Steckler et al., 2008). Different velocity patterns have been identified, providing evidences for first order crustal and sub-crustal heterogeneities and boundaries between the main crustal domains of Southern Apennines,
Calabrian, and Tyrrhenian regions (Fig. 1b). High velocity pattern at crustal depth beneath the Tyrrhenian Sea is commonly interpreted with the thinning of the Tyrrhenian crust and incipient oceanization (Barberi et al., 2004; Chiarabba et al., 2008; Finetti, 2005a, 2005b; Orecchio et al., 2011; Pepe et al., 2000). At greater depths, tomographic analysis evidenced that the Ionian subducting slab is in-depth continuous only beneath the central part of the Arc in southern Calabria while detachment has already occurred at the northern and southwestern edges of the arc itself, e.g. northern Calabria and northeastern Sicily, respectively (Fig. 1b; Neri et al., 2009). Beneath the southern Apennines lack of subcrustal seismicity and of high-velocity anomalies down to 200 km depth, together with a southwestward dipping high-velocity body at greater depths have been detected (Cimini, 1999; Cimini and Marchetti, 2006; Wortel and Spakman, 2000). In this framework the Calabrian Arc, a curved structure characterized by very heterogeneous seismotectonic regimes along its length (Cristofolini et al., 1985; Montone et al., 2004), together with the adjacent southern Apennines
C. Totaro et al. / Journal of Geodynamics 82 (2014) 87–97
sector, has been the site of destructive, magnitude class 6 and 7 earthquakes that have occurred both in recent and historical times (Fig. 1b; Galli et al., 2008; Neri et al., 2006). In the junction area between the Southern Apennines and the Calabrian Arc domains (e.g. at the northern edge of the Calabrian subduction zone, gray box in Fig. 1b), the Pollino Mts. area shows a structural and seismotectonic setting quite intricate, representing a highly deformed, very complex sector of great geodynamic interest and still not fully understood (see e.g. Bonini et al., 2011; Frepoli et al., 2011; Neri et al., 2012; Spina et al., 2011). Furthermore, in this sector a seismic gap was previously hypothesized by paleoseismological evidences associated with the lack of major earthquakes in historical catalogs (Cinti et al., 2002; Michetti et al., 2000). Since spring 2010 to 2013, this area has been affected by a seismic sequence with more than four thousands of small to moderate earthquakes (black dots in Fig. 1b). The strongest events occurred on May 28th 2012 (Ml 4.3) and on October 25th 2012 (Ml 5.0), respectively. Seismic activity has shown a fairly regular increase in number and energy since the onset of activity (Totaro et al., 2013). Epicentral distribution of the 2010–2013 sequence and the relative focal mechanism solutions evidenced two main clusters with a faint NNW-SSE-trend, showing predominantly normal faulting mechanisms with NE extension. The recent seismic activity has been interpreted as a seismic deformation occurring inside the southern Apennines extensional domain, specifically at its southern tip (Totaro et al., 2013). In order to better characterize the crustal structures of this peculiar region of southern Italy, which includes the Pollino Mts. area, we performed a local earthquake tomography by applying the LOTOS code (Koulakov, 2009), drawing also benefit from the large amount of data coming from the 2010–2013 seismic activity. The use of the LOTOS code, that represents the first application of this algorithm in the southern Italy crustal studies, allowed us to obtain highly resolved results strongly reducing the grid spacing with respect to previous investigations carried out in the region, and also to furnish a detailed Vp/Vs model not previously available for the study area.
2. Tectonic settings The large Calabrian Arc is a Cenozoic–Quaternary curved orogen (Fig. 1a) running from the NW–SE-trending Southern Apennines to the E–W-trending Sicilian Maghrebides (Carminati et al., 2012; Catalano et al., 1996; Lavecchia et al., 2007; Malinverno and Ryan, 1986; Minelli and Faccenna, 2010; Polonia et al., 2011; Rosenbaum and Lister, 2004). The central part of the Arc is characterized by the presence of a narrow subduction zone, which appears to be the only site of residual active subduction in the framework of a larger scale subduction process that has involved the western Mediterranean in the last tens Myrs (Fig. 1a; Faccenna et al., 2004; Wortel and Spakman, 2000). Many investigators (see, among others, Faccenna et al., 1996; Malinverno and Ryan, 1986) have suggested that geologic and geophysical data in the Calabrian Arc region can be interpreted in the framework of a geodynamic model assuming the co-existence of NW-SE convergence of Nubia and Europe plates and gravity-induced SE-ward rollback of the Ionian lithospheric slab subducting to NW beneath the Tyrrhenian lithosphere (Fig. 1b). The Ionian subducting slab is apparently in-depth continuous only beneath the central part of the Arc (southern Calabria), while it has already undergone detachment at the edges of the Arc itself, beneath northern Calabria and northeastern Sicily, respectively. Two transitional zones separating the different domains of continuous vs detached slab have been also detected (Neri et al., 2009). Subduction activity is now close to end but slow retreat seems to be still present (D’Agostino et al., 2011).
89
The space–time evolution of the rollback process (characterized by reduced intensity and greater concentration in the Calabrian Arc area since the Upper Pliocene) can be referred to a clear structural differentiation between the Calabrian Arc and the marginal tectonic units of Sicily and southern Apennines (Neri et al., 2012). In this framework, the Southern Apennines and the Calabrian Arc represent two of the most seismically-active regions of the entire Mediterranean area, with their long record of destructive historical earthquakes of magnitude as large as 7 (Amoruso et al., 2006; Galli and Bosi, 2004; Galli et al., 2008; Guidoboni et al., 2007; Monaco and Tortorici, 2000; Neri et al., 2006; black circles in Fig. 1b). The Calabrian Arc, uplifting at a rate of 0.5–1.2 mm/year in the last 1–0.7 Myrs, is characterized by internal deformation mainly accommodated by normal faulting (see e.g. Catalano and De Guidi, 2003; Catalano et al., 2003; Monaco and Tortorici, 2000). The strongest earthquakes can be related to an extensional regime with opening direction varying from NW–SE in southern Calabria to NE–SW in the northern part of the region (see e.g. Neri et al., 2004, 2005). In particular, the northern sector of the Calabria Arc (i.e., the Calabria–Lucania boundary) is a particular tectonically complex area as it joins the NNE–SSW-trending, NW–SE-extending curved Calabria with the NW–SE-trending, NE–SW-extending Southern Apennines, and is located on top of the northern edge of the subducting Ionian slab (Fig. 1b; Presti et al., 2013). To the north of this peculiar sector, in the Southern Apennines, slab break-off has already occurred, active subduction is therefore substantially absent, and an extensional stress regime compatible with the post-orogenic NE–SW extension affecting the entire Apennines has been detected (Presti et al., 2013). The Tyrrhenian and axial sectors of the Southern Apennines have undergone postorogenic extensional tectonics since at least earlymiddle Pleistocene time (Amicucci et al., 2008; Barchi et al., 2007). In this area, several destructive earthquakes were sourced, in recent and historical times, by extensional faults located closely west of the Apennines, considered as major seismogenic faults and subjected to a relatively uniform extensional stress regime with a nearly NE–SW orientation (Billi et al., 2011; Galli et al., 2006, 2008; Pantosti et al., 1993; Presti et al., 2013). Nevertheless most seismically active faults of this region are still poorly known. In the junction area between the southern Apennines extensional domain and the northern edge of the Calabrian Arc subduction zone, the Pollino Mts. area shows a structural and seismotectonic setting quite intricate and different views have been proposed by several authors (see, among others, Catalano et al., 2004; DISS Working Group, 2010; Michetti et al., 2000; Spina et al., 2011). Although normal mechanisms on NW-trending structures represent the most diffuse faulting style, differently oriented structures and various mechanisms are also reported (see among others Catalano et al., 2004; DISS Working Group, 2010; Spina et al., 2011; Van Dijk et al., 2000). This region lies at the northern section of the Calabrian accretionary wedge, which, since at least the Neogene, has been drifting southeastward onto the retreating Ionian subducting slab (Fig. 1a; Billi et al., 2007; Neri et al., 2012). Previous studies have indicated this as a highly deformed zone including shallow dynamics associated with the activity of a Subduction-Transform Edge Propagator (STEP) fault (Neri et al., 2012; Presti et al., 2013). Govers and Wortel (2005) have introduced the concept of STEP fault to represent lithosphere tearing at the edges of subduction trench zones. Even if location of the northern STEP fault in this subducting system is still debated, several authors have found evidences of shallow expressions of the Ionian slab tears in correspondence with the Sangineto Line (SL in Fig. 1b), at the Calabria–Lucania boundary (Govers and Wortel, 2005; Rosenbaum and Lister, 2004; Rosenbaum et al., 2008).
90
C. Totaro et al. / Journal of Geodynamics 82 (2014) 87–97
3. Data and method In order to increase the detail of our analysis with respect to previous study carried out in the same sector (Barberi et al., 2004; Calò et al., 2013; Chiarabba et al., 2008; Orecchio et al., 2011), we performed a tomographic inversion by applying the LOTOS algorithm (Koulakov, 2009). We used the data of local seismicity occurred in the study region and collected from the available national and local networks managed by INGV (www.ingv.it) and University of Calabria, respectively. We selected seismicity that has occurred between January 1981 and October 2012 at the southern Apennines–Calabrian Arc border region, within the limits of 39◦ N–40.5◦ N and 15◦ E–17.5◦ E (Fig. 2). Time picks from 250 stations were used, including those located outside the study area. It has to be noted that the recording conditions might vary from station to station depending on the type of instruments, installing conditions and the time in which the recorded events occurred. We want to remark that seismic monitoring in Italy has been strongly improved since the 1997, after the implementation of the new National Seismic Network (Amato and Mele, 2008), and more than 70% of our starting dataset events have occurred after the 1996 (i.e. in the period of higher quality of the recording network). To select the data we applied two criteria: first, the total number of P and S picks per event should be larger or equal 11; second, after locating the sources in the 1D starting model, we rejected the data with residuals greater than 1.5 and 2 s for P- and S-data, respectively. The final dataset used for tomographic inversion consisted of 2388 events (Fig. 2) and corresponding 28,108 P- and 14,605 S-arrival times (17.8 picks per event, on average). The distribution of these events over the study area is relatively homogeneous, ensuring high resolution of tomographic inversions. By using the LOTOS code (Koulakov, 2009), the calculations start with preliminary source locations based on the grid search method with the use of travel times computed in the 1D starting velocity model. At this step, we made the estimates for the optimal 1D velocity distributions by repeating the location procedure with dozens of different 1D models parameterized with 4–6 parameters (constant Vp/Vs ratio and 3–5 velocity values in different depth levels). The best model provided the maximum number of events and picks and minimum average deviations of residuals. Based on these criteria, we found the model with Vp/Vs = 1.76 and P-velocity values: 4.9 km/s at 0 km, 5.93 km/s at 10 km, 7 km/s at 30 km and 8 km/s at 40 km. Between the indicated depth levels, the velocity was linearly interpolated. The inversion procedure includes the iteration of location and inversion steps. Source locations in the 3D velocity model use 3D ray tracer based on the bending method, which ensures fast stable calculations of travel times of seismic rays between any two points in the study volume. The velocity distribution is parameterized with a set of nodes distributed in the study area according to the distribution of rays. In map view, the nodes are distributed with a regular step (5 km in our case); in the vertical direction, the grid spacing depends on the data density, but it cannot be smaller than a predefined value (3 km in our case). To avoid any bias related to the basic orientation of the grid, we performed the inversions for several grids with different basic orientations (0◦ , 22◦ , 45◦ and 67◦ in our case). The results computed for these grids are averaged in a regular mesh which is then used as an updated 3D velocity model in the next iteration. The inversion is performed simultaneously for the P- and Svelocity distributions, source corrections (four parameters for each source) and station corrections. The latest parameter appears to be very important in our case, because in our dataset we have some stations located outside the study area. Station corrections make possible to rid-off the factors affecting the travel time between the boundary of the study area and a single station. The inversion of the
matrix is performed using the LSQR algorithm (Paige and Saunders, 1982; Nolet, 1987). We regularize the solution by minimizing the gradient between neighboring nodes. The value of smoothing coefficients is estimated based on the results of synthetic modeling. As a result of inversion, we obtain the 3D distributions of P- and Svelocity anomalies. The distribution of Vp/Vs ratio is then computed by dividing the obtained values of absolute P- and S-velocities. We performed five iterations and then we obtained a reduction of the residual average deviations of ∼25% and ∼35% for P- and S-data. The values of remnant average deviations of the residuals are 0.25 s and 0.35 s, respectively.
4. Results and discussion In order to estimate a possible effect of noise on the resolution as well as the optimal values of inversion parameters, we performed several synthetic tests. In particular, in Fig. 3 we present three different checkerboard tests. In these models we defined periodic positive and negative velocity anomalies of different sizes and empty spacing between them: 20–10 km for Model 1, 15–7 km for Model 2 and 10–5 km for Model 3. Anomalies always had amplitude equal to 5% and opposite signs for P- and S-models, in order to ensure large variations of Vp/Vs ratio. The synthetic travel times were computed for the same sourcereceiver pairs as in the case of the observed data. The locations of real sources correspond to the solution obtained after five iterations of real data inversion. The synthetic data were perturbed with random noise having the average deviations of 0.1 and 0.2 for the P- and S-data, bringing to the same values of variance reduction as in the case of real data inversion. After computing the synthetic data, we “forgot” all information about source locations and velocity model and performed the reconstructions using same procedure and inversion parameters as in the case of the real data analysis, including the absolute source location, which may bias considerably the synthetic residuals. The reconstructions of P- and S anomalies, as well as Vp/Vs ratio obtained by division of the computed absolute P- and S-velocities, are presented in Fig. 3. It can be seen that Models 1 and 2 are robustly reconstructed in most parts of the study area. Model 3, with the anomaly sizes of 10 km, appears to be resolved only in the central part where there is maximum concentration of events and stations. It is important that the used calculation method ensures robust reconstruction of Vp/Vs ratio despite significant noise in the data which might cause some instabilities in the reconstructions of P- and S models. Based on these tests, we can evaluate the minimum size of patterns which can be resolved in different parts of the area. Furthermore, these tests show that the inversion parameters, which were identical for real and synthetic data inversions, are adequate and provide optimal quality reconstructions. The new seismic velocity model obtained in the present study for the southern Apennines–Calabrian Arc border region (Figs. 4 and 5) is reported as the resulting seismic velocity anomalies with respect to the aforesaid 1D velocity model, as well as Vp/Vs ratio, in horizontal and vertical sections. Absolute velocities along the vertical sections are shown in Fig. 6. The figures present also the distributions of relocated seismic events at corresponding depths and close to the presented profiles, respectively. The horizontal sections present seismic structures beneath the southern Apennines–Calabrian Arc border region in the uppermost crust (3 km depth), middle crust (15 km) and lower crust (30 km). The model is shown only in areas with sufficient data coverage: the values are masked if distance to the nearest parameterization node is larger than 10 km. In the upper crust layer a small size low-velocity anomaly of P-waves is detectable just south of the Pollino Mts. area. It seems
C. Totaro et al. / Journal of Geodynamics 82 (2014) 87–97
91
Fig. 2. Epicentral distribution of the starting inversion dataset consisting of ∼3600 earthquakes (dots) and location of the recording seismic stations (triangles).
to separate the two adjacent domains of Calabrian Arc and southern Apennines, characterized by higher velocity values. At same depth, low S-waves velocity anomaly is also present, and high values of Vp/Vs ratio characterize the Pollino Mts. region (Fig. 4). This section appears to reproduce the variety of the sedimentary cover, intrusive and metamorphic bodies. In Fig. 7, the comparison between the shallow tomography sections for P- and S anomalies with the surface geology (Compagnoni et al., 2011) is shown. It can be seen that P-velocity patterns correlate with the main geological structures. For example, a large positive anomaly “1” lies in correspondence with the Calabride mountain complex which is composed of rigid metamorphic and igneous rocks of Hercynian age. The Pollino Mts. area lies in correspondence with the Apennine and Ligurid/Sicilid units (“2” and “3”, respectively), which is a sector characterized by high P-velocities in the shallowmost crust. In the depression between these two mountain areas we observe a small, but very contrasted feature with low P- and S-velocities (“4”) which might represent a small sedimentary basin. The Apulian Platform, which is located at the margin of the study area,
matches with high P-velocity anomaly “5”. The fair correlation of P-velocity distribution with geology can be explained by the sensitivity of bulk elastic parameters to the composition and rigidity of rocks. Significantly different structures are observed at shallow depths for the S-velocity anomalies. A strong negative anomaly is observed in the Pollino massif (“3”), where P-model presents high velocity. Coexistence of high P and low S velocities at shallow depths may testify strong fracturing and saturation of rocks with fluids. This massif is associated with high Vp/Vs ratio pattern, which is particularly strong in the area of huge seismic activity (i.e. the area affected by the 2010–2013 Pollino swarm). The Calabride block “1” is associated with high-velocity S-anomaly, having a size considerably smaller than the relative shape indicated in the geological map. For the shallow layers, the S-velocity is rather more sensitive to the fracturing and saturation with fluids than to the composition. It is interesting that Vp/Vs anomalies are mostly associated with areas of high relief where maximum intensity of tectonic processes is expected. Thus joint consideration of P, S-velocity anomalies and
92
C. Totaro et al. / Journal of Geodynamics 82 (2014) 87–97
Fig. 3. Checkerboard tests performed to assess the spatial resolution and to estimate the optimal values of inversion parameters. Periodic positive and negative velocity anomalies of different sizes and empty spacing between them have been defined: 20–10 km for Model 1 (top row), 15–7 km for Model 2 (middle row) and 10–5 km for Model 3 (bottom row). Anomalies always had amplitude equal to 5% and opposite signs for P- and S-models (left and central columns), in order to ensure large variations of Vp/Vs ratio (right column).
C. Totaro et al. / Journal of Geodynamics 82 (2014) 87–97
93
Fig. 4. New seismic velocity model for the southern Apennines–Calabrian Arc border region obtained in the present study. Tomographic results are reported in terms of percentage variation of P- and S-wave velocity with respect to the optimal 1D reference model (top and middle rows) and in terms of Vp/Vs ratio (bottom row), respectively. Plates correspond to uppermost crust (3 km depth), middle crust (15 km) and lower crust (30 km), respectively. The model is shown only in areas with sufficient data coverage. 3D earthquake locations coming from the inversion are also shown, separately plotted according to the minimum vertical grid spacing: 0–4.5 km; 13.5–16.5 km; 28.5–31.5 km, respectively. Gray dashed line in the top left panel is the location of the surface projection of the northern STEP fault by Rosenbaum et al. (2008) in correspondence with the Sangineto Line (SL in map). “PM” stands for Pollino Mts., black lines indicate the profiles relative to the vertical sections reported in Figs. 5 and 6.
Vp/Vs ratio gives independent sights to the mechanical characteristics of rocks in the uppermost layers. In the middle and lower crust (15 and 30 km depth in Fig. 4) the structure becomes considerably different with respect to the shallow section, and a clear change in the velocity pattern is evident moving from the Tyrrhenian Sea to the Calabrian Arc domains. We
observe a strong contrast between a large low-velocity anomaly beneath the continental part of Italy and high-velocity beneath the offshore areas of the Tyrrhenian Sea, which is similarly expressed in both P- and S-velocity models. These results show a general agreement with the main crustal features of the study region revealed also by seismic lines and gravimetric analyses, indicating crustal
94
C. Totaro et al. / Journal of Geodynamics 82 (2014) 87–97
Fig. 5. Vertical sections of the tomographic model along the profiles indicated in the maps of Fig. 4. Percentage variation of P- and S-wave velocity with respect to the optimal 1D reference model (top and middle rows) and in terms of Vp/Vs ratio (bottom row), respectively are reported. Distribution of relocated events lying within ±20 km of the vertical planes is also shown.
thickening moving from southern Tyrrhenian to Calabrian Arc (Neri et al., 2012 and references therein). Looking at the vertical sections, P-wave velocity patterns reported in section 1 (left panels in Fig. 5) well depict the separation between the southern Apennines and the Calabrian Arc domains by the low velocity zone present in correspondence with the Pollino Mts. area. In-depth continuity and dimension increasing of this low velocity anomaly are also evident. In cross-section 2 (right panels in Fig. 5) the sharp separation between high and low seismic velocities in correspondence with the Tyrrhenian coast is clearly evident in the whole depth range and it may represent the
transition from the thinner Tyrrhenian crust to the thicker one beneath Calabria. Looking at the absolute velocity patterns along the vertical sections in Fig. 6, we can associate the Moho interface with contour lines of ∼7.4 km/s and 4.2 km/s in P- and S-velocity models, respectively (“red-to-orange” layer). We can then estimate that the crustal thickness varies from ∼25–30 km beneath the offshore area to ∼35–40 km beneath the continent, according to the results obtained in previous studies (see e.g. Piana Agostinetti and Amato, 2009). Same absolute velocity sections can be used to identify the sedimentary cover thickness. If we assume the bottom of sediments at 5.5 km/s and 3.0 km/s for P- and S-velocities (“green”
Fig. 6. Absolute velocity patterns along the same vertical sections of Fig. 5.
C. Totaro et al. / Journal of Geodynamics 82 (2014) 87–97
95
Fig. 7. Comparison between the shallow tomography sections for P- and S anomalies and surface geology (redrawn from Compagnoni et al., 2011).
Fig. 8. Focus on the southern Apennines–Calabrian Arc transition area. Low velocity anomalies of −4% are shown for the depth levels 3 km, 15 km, and 30 km, (continuous, dashed, and dotted black lines, respectively). Gray dashed line is the locations of the shallow projection of the northern STEP fault proposed by Rosenbaum et al. (2008) in correspondence with the Sangineto Line (SL in map). Thick gray line marks the northern limit of the continuous slab domain, according to Neri et al. (2009).
The tomographic results for P- and S anomalies at the upper crust layer show a fairly good correlation between velocity patterns and the main geological features of the study area (Fig. 7). In the middle and lower crust, velocity distribution clearly evidences the separation between the southern Apennines and the Calabrian Arc domains by a low velocity zone present at all depths and showing a dimensional enlargement with depth. In this transitional area, the existence of lithospheric-scale tear faults, in particular those rooting the left lateral Sangineto Line (Fig. 8), has been previously suggested by several authors (Govers and Wortel, 2005; Panza et al., 2007; Rosenbaum et al., 2008). The lateral extension and the quite relevant distance of this sector from the slab edge do not permit to properly define this region as a direct expression of a present-day STEP activity at the northern edge of the subducting slab. It may perhaps represent the present-day activity of structural systems that eventually worked like a STEP zone in the past (Catalano et al., 2001; Govers and Wortel, 2005; Rosenbaum et al., 2008). In this context it has to be noted that the surface projection of the Sangineto Line lies in correspondence with the low velocity anomalies (−4%) identified at all depth levels in the present study (Fig. 8). This may suggests that the structural systems that eventually worked like a STEP zone in the past may have opened the route to a possible fluid upwelling, and this could be evidenced by the enlargement of the low velocity anomaly size with depth.
Acknowledgments layer), we can identify the areas with thick sediments (mostly in depressions) and zones where basement appears to be closer to the surface. High earthquake concentration in the high-Vp zones (Figs. 4 and 5) and lack of seismicity in the low-Vp and lowVs ones are evident, especially in correspondence with the Pollino Mts. area. In this small sector, low velocity anomalies are present at all depths and a constant enlargement of their size can be noted with depth increasing (Fig. 5). The distributions of high Vp/Vs ratio in the upper crust clearly correlate with areas of huge seismicity representing strongly fractured rocks with likely high fluid content. 5. Conclusions The present study represents the first application of the LOTOS algorithm to crustal analyses in southern Italy. This code allowed us to obtain a new highly resolved seismic velocity model for the southern Apennines–Calabrian Arc border region furnishing also a detailed Vp/Vs model not previously available for the study area.
This research has been supported by Istituto Nazionale di Geofisica e Vulcanologia and Dipartimento della Protezione Civile (DPC) through the INGV-DPC 2014 S1 Project and by the Project of National Interest PRIN 2010-2011 “Geodinamica attiva e recente dell’Arco Calabro e del complesso di accrezione nel Mar Ionio”, funded by Ministero Istruzione Universita` e Ricerca (MIUR).
References Amato, A., Mele, F.M., 2008. Performance of the INGV National Seismic Network from 1997 to 2007. Ann. Geophys. 51, 2–3. Amicucci, L., Barchi, M.R., Montone, P., Rubiliani, N., 2008. The Vallo di Diano and Auletta extensional basins in the southern Apennines (Italy): a simple model for a complex setting. Terra Nova 20, 475–482, http://dx.doi.org/10.1111/ j.1365-3121.2008.00841.x. Amoruso, A., Crescentini, L., Neri, G., Orecchio, B., Scarpa, R., 2006. Spatial relation between the 1908 Messina Straits earthquake slip and recent earthquake distribution. Geophys. Res. Lett. 33, http://dx.doi.org/10.1029/2006GL027227. Barberi, G., Cosentino, M.T., Gervasi, A., Guerra, I., Neri, G., Orecchio, B., 2004. Crustal seismic tomography in the Calabrian Arc region, south Italy. Phys. Earth Planet. Inter. 147, 297–314.
96
C. Totaro et al. / Journal of Geodynamics 82 (2014) 87–97
Barchi, M., Amato, A., Cippitelli, G., Merlini, S., Montone, P., 2007. Extensional tectonics and seismicity in the axial zone of the Southern Apennines. Ital. J. Geosci., Spec. Iss. 7, 47–56. Billi, A., Presti, D., Faccenna, C., Neri, G., Orecchio, B., 2007. Seismotectonics of the Nubia plate compressive margin in the south-Tyrrhenian region, Italy: clues for subduction inception. J. Geophys. Res. 112, B08302, http://dx.doi.org/10.1029/2006JB004837. Billi, A., Faccenna, C., Bellier, O., Minelli, L., Neri, G., Piromallo, C., Presti, D., Scrocca, D., Serpelloni, E., 2011. Recent tectonic reorganization of the Nubia-Eurasia convergent boundary heading for the closure of the western Mediterranean. Bull. Soc. Géol. Fr. 182, 279–303. Bonini, M., Sani, F., Moratti, G., Benvenuti, M.G., 2011. Quaternary evolution of the Lucania Apennine thrust front area (Southern Italy), and its relations with the kinematics of the Adria Plate boundaries. J. Geodyn. 51, 125–140, http://dx.doi.org/10.1016/j.jog.2010.01.010. Calais, E., DeMets, C., Nocquet, J.-M., 2003. Evidence for a post-3.16-Ma change in Nubia-Eurasia-North America plate motions? Earth Planet. Sci. Lett. 216, 81–92. Calò, M., Parisi, L., Luzio, D., 2013. Lithospheric P- and S-wave velocity models of the Sicilian area using WAM tomography: procedure and assessments. Geophys. J. Int. 195 (1), 625–649, http://dx.doi.org/10.1093/gji/ggt252. Carminati, E., Lustrino, M., Doglioni, C., 2012. Geodynamic evolution of the central and western Mediterranean: tectonics vs. igneous petrology constraints. Tectonophysics, http://dx.doi.org/10.1016/j.tecto.2012.01.026. Catalano, R., Di Stefano, P., Sulli, A., Vitale, F.P., 1996. Paleogeography and structure of the central Mediterranean: Sicily and its offshore area. Tectonophysics 260, 291–323, http://dx.doi.org/10.1016/0040-1951(95)00196-4. Catalano, R., Doglioni, C., Merlini, S., 2001. On the Mesozoic Ionian basin. Geophys. J. Int. 144, 49–64. Catalano, S., De Guidi, G., 2003. Late Quaternary uplift of northeastern Sicily: relation with the active normal faulting deformation. J. Geodyn. 36, 445–467. Catalano, S., De Guidi, G., Monaco, C., Tortorici, G., Tortrici, L., 2003. Long-term behaviour of the late Quaternary normal faults in the Straits of Messina area (Calabrian arc): structural and morphological constraints. Quat. Int. 101-102, 81–91. Catalano, S., Monaco, C., Tortorici, L., Paltrinieri, W., Steel, N., 2004. Neogene–Quaternary tectonic evolution of the southern Apennines. Tectonics 23, http://dx.doi.org/10.1029/2003TC001512. Chiarabba, C., De Gori, P., Speranza, F., 2008. The Southern Tyrrhenian Subduction zone: deep geometry, magmatism and plio-pleistocene evolution. Earth Planet. Sci. Lett. 268, 408-423, http://dx.doi.org/10.1016/j.epsl.2008.01.036. Chironi, C., De Luca, L., Guerra, I., Luzio, D., Moretti, A., Vitale, M., Sea Land Group, 2000. Crustal structures of the Southern Tyrrhenian Sea and the Sicily Channel on the basis of the M25, M26, M28, M39 WARR profiles. Boll. Soc. Geol. Ital. 119, 189–203. Cimini, G.B., 1999. P-wave deep velocity structure of the southern Tyrrhenian subduction zone from nonlinear teleseismic traveltime tomography. Geophys. Res. Lett. 26 (24), pp. 3,709–3,712. Cimini, G.B., Marchetti, A., 2006. Deep structure of peninsular Italy from seismic tomography and subcrustal seismicity. Ann. Geophys. 49 (1), 331–345. Cinti, F.R., Moro, M., Pantosti, D., Cucci, L., D’Addezio, G., 2002. New constraints on the seismic history of the Castrovillari fault in the Pollino gap (Calabria, southern Italy). J. Seismol. 6, 199–217. Compagnoni, B., Galluzzo, F., Bonomo, R., Capotorti, F., D’Ambrogi, C., Di Stefano, R., Graziano, R., Martarelli, L., Pampaloni, M.L., Pantaloni, M., Ricci, V., Tacchia, D., Masella, G., Pannuti, V., Ventura, R., Vitale, V., 2011. Carta Geologica d’Italia in Scala 1:1.000.000. http://isprambiente.gov.it/ Cristofolini, R., Ghisetti, F., Scarpa, R., Mezzani, L., 1985. Character of stress field in the Calabrian Arc and Southern Apennines (Italy) as deduced by geological, seismological and volcanolgical information. Tectonophysics 117, 39–58. ´ M.R., SeeD’Agostino, N., D’Anastasio, E., Gervasi, A., Guerra, I., Nedimovic, ber, L., Steckler, M., 2011. Forearc extension and slow rollback of the Calabria Arc from GPS measurements. Geophys. Res. Lett. 38, L17304, http://dx.doi.org/10.1029/2011GL048270. DISS Working Group, 2010. Database of Individual Seismogenic Sources (DISS), Version 3.1.1: A compilation of potential sources for earthquakes larger than M 5.5 in Italy and surrounding areas., http://dx.doi.org/10.6092/INGV.IT-DISS3.1.1 http://diss.rm.ingv.it/diss/,(INGV 2010-Istituto Nazionale%20di%20Geofisica% 20e%20Vulcanologia-All%20rights%20reserved Faccenna, C., Davy, P., Brun, J.P., Funiciello, R., Giardini, D., Mattei, M., Nalpas, T., 1996. The dynamics of back-arc extension: an experimental approach to the opening of the Tyrrhenian Sea. Geophys. J. Int. 126, 781–795. Faccenna, C., Piromallo, C., Crespo-Blanc, A., Jolivet, L., 2004. Lateral slab deformation and the origin of the western Mediterranean arcs. Tectonics 23, TC1012, http://dx.doi.org/10.1029/2002TC001488. Faccenna, C., Civetta, L., D’Antonio, M., Funiciello, F., Margheriti, L., Piromallo, C., 2005. Constraints on mantle circulation around the deforming Calabrian slab. Geophys. Res. Lett. 32 (6), L06311. Finetti, I.R., 2005a. Depth contour map of the Moho discontinuity in the central Mediterranean region from new CROP seismic data. In: Finetti, I.R. (Ed.), CROP PROJECT: Deep Seismic Exploration of the Central Mediterranean and Italy. Elsevier, Amsterdam, pp. 597–606. Finetti, I.R., 2005b. The Calabrian Arc and subducting Ionian slab from the new CROP seismic data. In: Finetti, I.R. (Ed.), CROP PROJECT: Deep Seismic Exploration of the Central Mediterranean and Italy. Elsevier, Amsterdam, pp. 393–412.
Frepoli, A., Maggi, C., Cimini, G.B., Marchetti, A., Chiappino, M., 2011. Seismotectonic of Southern Apennines from recent passive seismic experiments. J. Geodyn. 51, 110–124. Galli, P., Bosi, V., 2004. Catastrophic 1638 earthquakes in Calabria (southern Italy): new insights from paleoseismological investigation. J. Geophys. Res. 108, http://dx.doi.org/10.1029/2001JB00171. Galli, P., Bosi, V., Piscitelli, S., Giocoli, A., Scionti, V., 2006. Late Holocene earthquakes in Southern Apennines: paleoseismology of the Caggiano fault. Int. J. Earth Sci. 95, 855–870. Galli, P., Galadini, F., Pantosti, D., 2008. Twenty years of paleoseismology in Italy. Earth Sci. Rev. 88, 89–117. Giacomuzzi, G., Civalleri, M., De Gori, P., Chiarabba, C., 2012. A 3D Vs model of the upper mantle beneath Italy: insight on the geodynamics of central Mediterranean. Earth Planet. Sci. Lett. 335–336, 105–120. Govers, R., Wortel, M.J.R., 2005. Lithosphere tearing at STEP faults: response to edges of subduction zones. Earth Planet. Sci. Lett. 236, 505–523. Guidoboni, E., Ferrari, G., Mariotti, D., Comastri, A., Tarabusi, G., Valensise, G., 2007. CFTI4Med. Catalogue of Strong Earthquakes in Italy 461 B.C. 1997 and Mediterranean Area 760 B.C. 1500. An Advanced Laboratory of Historical Seismology. http://storing.ingv.it/cfti4med/ Koulakov, I., 2009. LOTOS code for local earthquake tomographic inversion: benchmarks for testing tomographic algorithms. Bull. Seismol. Soc. Am. 99 (1), 194–214, http://dx.doi.org/10.1785/0120080013. Lavecchia, G., Ferrarini, F., de Nardis, R., Visini, F., Barbano, M.S., 2007. Active thrusting as a possible seismogenic source in Sicily (southern Italy): some insights from integrated structural-kinematic and seismological data. Tectonophysics 445, 145–167, http://dx.doi.org/10.1016/j.tecto.2007.07.007. Lucente, F.P., Margheriti, L., Piromallo, C., Barruol, G., 2006. Seismic anisotropy reveals the long route of the slab through the westerncentral Mediterranean mantle. Earth Planet. Sci. Lett. 241, 517–529. Malinverno, A., Ryan, W., 1986. Extension in the Tyrrhenian Sea and shortening in the Apennines as result of arc migration driven by sinking of the lithosphere. Tectonics 5, 227–245. Michetti, A.M., Ferreli, L., Esposito, E., Porfido, S., Blumetti, A.M., Vittori, E., Serva, L., Roberts, G.P., 2000. Ground effects during the 9 September 1998, Mw = 5.6 Lauria earthquake and the seismic potential of the “aseismic” Pollino region in Southern Italy. Seismol. Res. Lett. 71, 31–46. Minelli, L., Faccenna, C., 2010. Evolution of the Calabrian accrewedge (central Mediterranean). Tectonics 29, TC4004, tionary http://dx.doi.org/10.1029/2009TC002562. Monaco, C., Tortorici, L., 2000. Active faulting in the Calabrian arc and eastern Sicily. J. Geodyn. 29, 407–424. Montone, P., Mariucci, M.T., Pondrelli, S., Amato, A., 2004. An improved stress map for Italy and surrounding regions (central Mediterranean). J. Geophys. Res. 109, B10410, http://dx.doi.org/10.1029/2003JB002703. Montone, P., Mariucci, M.T., Pierdominici, S., 2012. The Italian present-day stress map. Geophys. J. Int. 189 (2), 705–716. Montuori, C., Cimini, G.B., Favali, P., 2007. Teleseismic tomography of the southern Tyrrhenian subduction zone: new results from seafloor and land recordings. J. Geophys. Res. 112, B03311. Neri, G., Barberi, G., Orecchio, B., Aloisi, M., 2002. Seismotomography of the crust in the transition zone between the southern Tyrrhenian and Sicilian tectonic domains. Geophys. Res. Lett. 29 (23), 10, http://dx.doi.org/10.1029/2002GL015562. Neri, G., Barberi, G., Oliva, G., Orecchio, B., 2004. Tectonic stress and seismogenic faulting in the area of the 1908 Messina earthquake, south Italy. Geophys. Res. Lett. 31, 10, http://dx.doi.org/10.1029/2004GL019742. Neri, G., Barberi, G., Oliva, G., Orecchio, B., 2005. Spatial variations of seismogenic stress orientations in Sicily, south Italy. Phys. Earth Planet. Inter. 148, 175–191. Neri, G., Oliva, B., Orecchio, B., Presti, D., 2006. A possible seismic gap within a highly seismogenic belt crossing Calabria and eastern Sicily, Italy. Bull. Seismol. Soc. Am. 96, 321–331. Neri, G., Orecchio, B., Totaro, C., Falcone, G., Presti, D., 2009. Subduction beneath southern Italy close the ending: results from seismic tomography. Seismol. Res. Lett. 80, 63–70. Neri, G., Marotta, A.M., Orecchio, B., Presti, D., Totaro, C., Barzaghi, R., Borghi, A., 2012. How lithospheric subduction changes along the Calabrian Arc in southern Italy: geophysical evidences. Int. J. Earth Sci. 101, 1949–1969. Nocquet, J.-M., Calais, E., 2004. Geodetic measurements of crustal deformation in the western Mediterranean and Europe. Pure Appl. Geophys. 161, 661–681, http://dx.doi.org/10.1007/s00024-003-2468-z. Nocquet, J.-M., 2012. Present-day kinematics of the Mediterranean: a comprehensive overview of GPS results. Tectonophysics 579, 220–242. Nolet, G., 1987. Seismic wave propagation and seismic tomography. In: Nolet, G. (Ed.), Seismic Tomography. Reidel, Dordrecht, pp. 7–29. Orecchio, B., Presti, D., Totaro, C., Guerra, I., Neri, G., 2011. Imaging the velocity structure of the Calabrian Arc region (South Italy) through the integration of different seismological data. Boll. Geof. Teor. Appl. 52 (4), 625–638, http://dx.doi.org/10.4430/bgta0023. Paige, C.C., Saunders, M.A., 1982. LSQR: an algorithm for sparse linear equations and sparse least squares. ACM Trans. Math. Softw. 8, 43–71. Pantosti, D., Schwartz, D., Valensise, G., 1993. Paleoseismology along the 1980 surface rupture in Irpinia Fault: implications for earthquake
C. Totaro et al. / Journal of Geodynamics 82 (2014) 87–97 recurrence in the Southern Apennines, Italy. J. Geophys. Res. 98 (4), 6561–6577. Panza, G.F., Peccerillo, A., Aoudia, A., Farina, B., 2007. Geophysical and petrological modelling of the structure and composition of the crust and upper mantle in complex geodynamic settings: the Tyrrhenian Sea and surroundings. Earth Sci. Rev. 80, 1–46, http://dx.doi.org/10.1016/j.earscirev.2006.08.004. Pepe, F., Bertotti, G., Cella, F., Marsella, E., 2000. Rifted margin formation in the south Tyrrhenian Sea: a high-resolution seismic profile across the north Sicily passive continental margin. Tectonics 19, 241–257. Piana Agostinetti, N., Amato, A., 2009. Moho depth and Vp = Vs ratio in peninsular Italy from teleseismic receiver functions. J. Geophys. Res. Solid Earth 114, B6. Polonia, A., Torelli, L., Mussoni, P., Gasperini, L., Artoni, A., Klaeschen, D., 2011. The Calabrian Arc subduction complex in the Ionian Sea: regional architecture, active deformation, and seismic hazard. Tectonics 30, TC5018, http://dx.doi.org/10.1029/2010TC002821. Presti, D., Billi, A., Orecchio, B., Totaro, C., Faccenna, C., Neri, G., 2013. Earthquake focal mechanisms, seismogenic stress, and seismotectonics of the Calabrian Arc, Italy. Tectonophysics, http://dx.doi.org/10.1016/j.tecto.2013.01.030. Rosenbaum, G., Lister, G.S., 2004. Neogene and Quaternary rollback evolution of the Tyrrhenian Sea, the Apennines, and the Sicilian Maghrebides. Tectonics 23, TC1013, http://dx.doi.org/10.1029/2003TC001518. Rosenbaum, G., Gasparon, M., Lucente, F.P., Peccerillo, A., Miller, M.S., 2008. Kinematics of slab tear faults during subduction segmentation and implications for Italian magmatism. Tectonics 27 (2).
97
Rovida, A., Camassi, R., Gasperini, P., Stucchi, M., 2011. CPTI11, la versione 2011 del Catalogo Parametrico dei Terremoti Italiani, Milano, Bologna (in Italian) http://emidius.mi.ingv.it/CPTI11 Serpelloni, E., Vannucci, G., Pondrelli, S., Argnani, A., Casula, G., Anzidei, M., Baldi, P., Gasperini, P., 2007. Kinematics of the western Africa–Eurasia plate boundary from focal mechanism and GPS data. Geophys. J. Int., http://dx.doi.org/10.1111/j.1365-246X.2007.03367.x. Spakman, W., Wortel, R., 2004. A tomographic view on western Mediterranean geodynamics. In: Cavazza, W. (Ed.), The Transmed Atlas: The Mediterranean Region from Crust to Mantle. Springer, New York, pp. 31–52. Spina, V., Tondi, E., Mazzoli, S., 2011. Complex basin development in a wrenchdominated back-arc area: tectonic evolution of the Crati basin, Calabria, Italy. J. Geodyn. 51, 90–109. Steckler, M.S., Piana Agostinetti, N., Wilson, C.K., Roselli, P., Seeber, L., Amato, A., Lerner-Lam, A., 2008. Crustal structure in the Southern Apennines from teleseismic receiver functions. Geology 36, 155–158. Totaro, C., Presti, D., Billi, A., Gervasi, A., Orecchio, B., Guerra, I., Neri, G., 2013. The ongoing seismic sequence at the Pollino Mountains, Italy. Seismol. Res. Lett. 84 (6), 955–962, http://dx.doi.org/10.1785/0220120194. Van Dijk, J.P., Bello, M., Brancaleoni, G.P., Cantarella, G., Costa, V., Frixa, A., Golfetto, F., Merlini, S., Riva, M., Torricelli, S., Toscano, C., Zerilli, A., 2000. A regional structural model for the northern sector of the Calabrian Arc (southern Italy). Tectonophysics 324, 267–320. Wortel, R., Spakman, W., 2000. Subduction and slab detachment in the Mediterranean-Carpathian region. Science 290, 1910–1917.