- Email: [email protected]
Crust-mantle mechanical structure in the Central Mediterranean region
Tectonophysics 603 (2013) 89–103
Contents lists available at ScienceDirect
Tectonophysics journal homepage: www.elsevier.com/locate/tecto
Crust-mantle mechanical structure in the Central Mediterranean region Raffaele Splendore, Anna Maria Marotta ⁎ Università degli Studi di Milano, Dipartimento di Scienze della Terra “Ardito Desio”, L. Cicognara 7, 20129 Milano, Italy
a r t i c l e
i n f o
Article history: Received 1 October 2012 Received in revised form 9 May 2013 Accepted 21 May 2013 Available online 31 May 2013 Keywords: Lithosphere thermal field Rheological analysis Lithosphere strength heterogeneities Central Mediterranean
a b s t r a c t A thermo-rheological analysis is performed to elucidate the mechanical behaviour of the lithosphere in the Mediterranean domain. The thermal field of the lithosphere is calculated by means of a finite element 3D thermal model. The decrease in radiogenic heat production with depth is taken into account, together with the compositional layering of the lithosphere. The predicted thermal field is analysed in terms of the temperatures and depth of the thermal lithosphere base. Based on the predicted thermal field, a rheological analysis is conducted, accounting for both the brittle and ductile behaviour of each lithospheric layer. The effects of the choice of wet and dry rheology are also investigated. Our rheological analysis reveals a strongly heterogeneous lithosphere strength pattern in the Central Mediterranean, characterised by strong lateral strength gradients and the occurrence of non-competent crustal layers of significant thickness in the eastern portion of the study area. © 2013 Elsevier B.V. All rights reserved.
1. Introduction An accurate knowledge of the lithosphere strength and its variation in space is crucial when an appropriate 3D analysis of the deformation is pursued, especially in very complex tectonic areas, such as the Mediterranean, where localised deformation features are expected and an average rheological statement of the lithosphere results not adequate (e.g. Marotta et al., 2004). Furthermore, when a sufficiently refined thermo-rheological model of the lithosphere is available, it is possible to reveal intra-lithosphere soft zones that might be responsible for mechanical decoupling at different crust or mantle levels. One crucial aspect of lithosphere rheology analysis is an adequate knowledge of the thermal state of the lithosphere, particularly when small-scale rheological discontinuities may occur. While other rheological analyses use indirect information about the thermal field of the lithosphere, based on, for example, mantle tomographic studies (e.g., Tesauro et al., 2009b), we prefer to use a 3D thermal model based on the measures of surface heat flow and appropriate assumptions for the thermal parameters of the crust and mantle rocks. Thus, we obtain a 3D lithosphere thermal field that is completely consistent with the subsequent rheological analysis. Over the past 10 years, the knowledge of the thermal structure of the lithosphere has extended from the global scale (e.g., Artemieva, 2006 and reference therein) to the regional scale of the Mediterranean region (e.g., Jimenez-Munt et al., 2003; Splendore et al., 2010). One of the main objectives of these studies was to identify the depth of the thermal base of the lithosphere by the resolution of the steady-state heat conduction equation for the lithosphere, which is constrained by ⁎ Corresponding author. Tel.: +39 02 50318470; fax: +39 02 50318489. E-mail address: [email protected] (A.M. Marotta). 0040-1951/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.tecto.2013.05.017
a fixed crust thickness and surface heat flow. In particular, Artemieva (2006) developed a global 1°x1° thermal model for the continental lithosphere using borehole measurements of heat flow and xenolith data and a 1°x1° crustal thickness model (CRUST5.1, Mooney et al., 1998). Jimenez-Munt et al. (2003) performed a more regional analysis that was based on an isostatically rather than seismically defined crust model. Splendore et al. (2010) performed a 3D thermal analysis of the surroundings of the Italian Peninsula that did not account for an accurate crustal stratification, thus producing smooth thermal gradients and a rather thin lithosphere. The present study aims to improve the thermal model developed by Splendore et al. (2010) in two major respects: (1) by considering a spatially refined modelled volume, both vertically and horizontally and (2) by introducing a more sophisticated compositional stratification of the crust. Predictions from the thermal model are corroborated by a qualitative comparison with the available seismic tomography of the area. Seismic tomography, in fact, maps the seismic velocity heterogeneities of Earth's interior. Moreover, while wave velocities within the crust are mainly controlled by compositional heterogeneities, in the upper mantle, changes in wave velocities are mainly attributed to temperature variation (Piromallo and Morelli, 2003), with a small contribution, i.e., less than 1%, from mantle composition (Goes et al., 2000; Sobolev et al., 1997). Positive seismic velocity anomalies are generally related to cold material and thus to depressed thermal fields, while negative seismic velocity anomalies are typical of hot material and high temperatures. Both P- and S-wave velocity distributions are used to construct tomographic models, but S-wave models are less consistent than those based on P-waves, providing a lower horizontal resolution (Koulakov et al., 2009). Finally, the predicted thermal field is used to determine the strength of the lithosphere and its heterogeneities at both crust and mantle
90
R. Splendore, A.M. Marotta / Tectonophysics 603 (2013) 89–103
levels. In the present work, we analyse the main features of lithosphere strength through the central Mediterranean region, focusing on whether and where the interplay between regional thermal fields, crust and mantle thicknesses, lithosphere composition stratification and strain rates weakens intra-plate levels, that may result in decoupling features. 2. Thermal analysis The lithosphere thermal field is calculated by the numerical integration of the steady state 3D energy equation ∂ ∂T ∂ ∂T ∂ ∂T þ þ þ ρH ðzÞ ¼ 0 K⋅ K⋅ K⋅ ∂x ∂x ∂y ∂y ∂z ∂z
ð1Þ
over the domain shown in Fig. 1 (panel a), where K is the thermal conductivity, T is the temperature, ρ is the density and H is the rate of radiogenic heat production per mass unit. No hydrothermal contribution is accounted for through the crust. Boundary conditions correspond to fixed temperature at the upper (300 K) and lower (1600 K) boundaries and to zero heat flow across the vertical sidewalls. For what concerns the geometry of the 3D domain where the numerical integration is performed, both lateral and upper boundaries of the study area are fixed and coincide with the major tectonic lineaments characterising the central Mediterranean region (the Alpine chain to the north, the Carpathian belt and the Anatolian faults system to the east and the Africa and Eurasia boundary to the south) and with the regional topography (Fig. 1, panel b), respectively.
Fig. 1. (a) Scheme of the 3D grid used for the thermo-rheological analysis. (b) Topography used to define the upper boundary of the 3D model. Data are taken from EuCRUST07 model (Tesauro et al., 2008).
R. Splendore, A.M. Marotta / Tectonophysics 603 (2013) 89–103
The lower boundary of the model, on the contrary, has not an a-priory fixed geometry but it is a model unknown itself. In order to solve for this unknown we use an iterative procedure in which, at each iteration, the geometry of the lower boundary of the model is modified accordingly to the predicted thermal field. In particular, let ZL(ite) be the depth of the thermal lithosphere base used at iteration ite to calculate the 3D thermal field, T(ite), and qs(ite) the predicted surface heat flow, ZL is modified assuming h i obs ΔZ L ðiteÞ ¼ α qs ðiteÞ−qs
ð2Þ
where qobs is the observed surface heat flow (Fig. 2). s In order to minimize the occurrence of abrupt lateral gradients, the proportional constant α is chosen which is equal to 0.05 and a maximum variation in ΔZL(ite) of ± 1 km is allowed. We use a classical χ2 analysis to calculate the quantity h 2
χ ðiteÞ ¼
qs ðiteÞ−qobs s
i2
ndof
ð3Þ
with ndof number of nodes at the surface. The iterative procedure stops when 2
2
χ ðiteÞ−χ ðite−1Þ≤0:01:
ð4Þ
At the first iteration the base of the thermal lithosphere (depth of the 1600 K) is derived from the 1D simple analytical solution T ðzÞ ¼ T s þ
qobs s z K
ð5Þ
obtained for a homogeneous layer with prescribed surface temperature (Ts) and heat flow (qobs s ), equal to 300 K and to the observed heat flow, respectively. The lithosphere is compositionally differentiated into an upper crust, lower crust and mantle. The 3D domain is numerically discretised into 26,805 triangular-based linear prismatic elements with an average dimension of 0.5°× 0.5° in the horizontal direction and a length of 2 km (within the crust) to 20 km (in the lithosphere mantle) in the vertical direction. The geometries of the upper–lower crust transition and Moho (Fig. 3, panels a and c) are based on the EuCRUST07 model (Tesauro et al., 2008). Table 1 lists the physical and thermal rock
Fig. 2. Distribution of the surface heat flow in the Adriatic region, obtained by linear interpolation of the global database of Pollack et al. (1993), augmented by Fernandez et al. (1998) and Artemieva (2006). The extremely high values of surface heat flow in the Larderello and Stromboli areas, which are likely related to non-conductive processes, such as local hydrothermal convection or volcanic processes, are excluded.
91
properties assumed for the upper and lower crusts and mantle. An exponential decrease in radiogenic heat production with depth is assumed within the crust according to −hz
HðzÞ ¼ H s e
r
ð6Þ
where Hs is the radiogenic heat production at the surface, z is the depth, and hr is a length scale for the decrease in H with depth. We use hr = 10 km, in agreement with the average value for continental crust (Turcotte and Schubert, 2002). Constant radiogenic heat production with depth is assumed within the mantle. Lithosphere temperature will be discussed in terms of the geometry of the base of the thermal lithosphere, defined as the depth of the 1600 K isotherm, the temperature at the upper crust–lower crust interface and the temperature at the Moho. To strengthen our thermal analysis, we also calculate the predicted thermal anomalies, computed as the percentage of temperature deviance with respect to the average temperature characterising the study area at the same depth, both horizontally and vertically, and qualitatively compare them to the regional seismic tomography. We are aware that the assumption of steady state thermal field is rather strong when a relatively recent tectono-thermal region, such as the Mediterranean, is analysed. However, since the poor knowledge of temporal constraints does not allow developing a rigorous time dependent 3D thermal model in the area, we adopt the stationary simple formulation used by several authors for the same area (e.g. Cloetingh et al., 2010; Jimenez-Munt et al., 2003). 2.1. Predicted 3D thermal field Fig. 4 shows the predicted base of the lithosphere in terms of the depth of the 1500 K (panel a) and 1600 K (panel b) isotherms. The Tyrrhenian and Pannonian Basins, where high values of observed surface heat flow are accompanied by a thin seismic crust, are characterised by a thin lithosphere, whose base depth ranges from 40 to 70 km. The lithosphere base deepens to approximately 100 km in the Algero-Provençal Basin and to more than 200 km between the Apennines and Dinarides belts, while it reaches its uppermost values in the Ionian region, where very low surface heat flow is measured and subduction has depressed the thermal field. The style of lateral variation in the lithosphere thickness predicted by our model is in agreement with that predicted by previous thermal models, such as those of Jimenez-Munt et al. (2003) and Artemieva (2006), in which the thermal lithosphere base follows the major tectonic and geodynamic features of the Mediterranean area. In particular, the thickest lithosphere is located below the Calabrian zone and the Apennines and Dinarides belts, while the Tyrrhenian and Pannonian Basins are paved by the thinnest lithosphere. However, some differences can be observed. With respect to the models of Artemieva (2006) and Jimenez-Munt et al. (2003), our model suggests a deeper lithosphere base under the Apennines and Dinarides belts and below the Calabrian area. It is worth mentioning that the thick lithosphere that our model predicts below the AlgeroProvençal Basin stands in contrast with the thin lithosphere predicted by Jimenez-Munt et al. (2003). One possible reason for these discrepancies may be the fact that a different dataset was used to define the crust: the global analysis by Artemieva (2006) is based on the CRUST5.1 crustal model (Mooney et al., 1998); Jimenez-Munt et al. (2003) accounted for isostasy within a 1D thermal model when deriving crustal thickness. Our 3D thermal analysis is based on the crustal model EuCRUST07 (Tesauro et al., 2008), derived by seismic reflection, refraction and receiver function studies, which provides a fine database for topography, upper–lower crust transition and Moho depth over a uniform grid of 15′ × 15′, making it more suitable to resolve small crust and mantle heterogeneities.
92
R. Splendore, A.M. Marotta / Tectonophysics 603 (2013) 89–103
Fig. 3. Depth of the upper–lower crust transition (panel a), depth of the Moho (panel c) and thickness of the upper crust (panel b) and the lower crust (panel d). Data are taken from EuCRUST07 model (Tesauro et al., 2008).
Fig. 5 shows the temperature predicted at the upper/lower crust interface and at the Moho in panels a and b, respectively. A cold crust characterises the basinal areas, which feature a very thin crust and a shallow Moho. The highest temperatures are predicted below the Dinarides and eastern Alps, where the roots of mountain belts reach great depths (50–60 km, Fig. 3, panel c), with a temperature of 800 K predicted at the interface between the upper and lower crusts and a temperature above 1100 K predicted at the Moho depth. The Apennines chain is characterised by a lower temperature at the Moho depth of approximately 900 K; in the northern part of the chain, however, a temperature of 1100 K is reached at a depth of approximately 45 km (Fig. 3, panel c). The peak temperature of 900 to 1000 K at the upper–lower crust transition depth is achieved below the eastern domain of the eastern Alps and along the Dinaries (Fig. 5, panel a), where a very thick upper crust is observed (Fig. 3, panel b), which is associated with a thin lower crust (Fig. 3, panel d).
Table 1 Thermal and physical properties used in the present study. Layer
Composition
Densitya (kg m−3)
Thermal conductivityb,c (Wm−1 K−1)
Radiogenic heat productiona (μW m−3)
Upper crust Lower crust Mantle
Granite Granulite Depleted peridotite
2700 2900 3200
3.1 2.0 4.15
2.43 0.4 0.002
a b c
Vilà et al. (2010). Rybach (1988). Artemieva and Mooney (2001).
2.2. Temperature anomalies In order to better enlighten deep hot and cold domains, suitable for comparison with the available tomography of the Mediterranean area (e.g., Boschi et al., 2004; Brandmayr et al., 2011; Fry et al., 2008; Koulakov et al., 2009; Marone et al., 2004; Piromallo and Morelli, 2003), at each point within the 3D study domain, we calculate the percentage of deviation of the predicted temperature from the average horizontal temperature at the same depth. Our model predicts a negative temperature anomaly, between −5% and more than −40%, below the Adria microplate and Calabrian arc, continuous in depth within the entire lithosphere (Fig. 6), with two peak zones: the first located in the northern Adriatic Sea (panels a and e) and the second below the Ionian Sea (panel c), where a negative peak value of more than −40% is reached within a depth of 50 km. The depressed thermal field associated with the thermal negative anomalies is consistent with that produced by subduction. The gap in the thermal negative anomaly between the northern portion of the Adria microplate and the Ionian Sea may be the result of hot material rising at shallow depths through the mechanical discontinuity separating the Adriatic and Ionian slabs, as supported by tomographic studies (Koulakov et al., 2009) and explained by Chiarabba et al. (2008) as the result of updoming astenospheric material that mechanically separates the Ionian and Adriatic slabs. The Tyrrhenian Basin is characterised by a positive thermal anomaly, with the peak value located in the southwestern zone (Fig. 6, panels c, d and f). Moving to the east, another significant positive thermal anomaly localises below the Pannonian Basin (Fig. 6, panel b), compatible with the upwelling of hot mantle during the opening of the basin (Huismans et al., 2001). In both cases, the positive thermal anomalies affect the entire lithosphere thickness.
R. Splendore, A.M. Marotta / Tectonophysics 603 (2013) 89–103
93
For ductile behaviour, we assume the power law (Weertman and Weertman, 1975) σD ¼
ε_ ε_ 0
1 =
n
Ea eðnRTÞ
ð8Þ
where ε_ is the strain rate (ranging from 10−16 to 10−14 s−1), R is the universal constant of gas, T is the temperature, and ε_ 0 , n, and Ea are the rheological parameters of the rock (Table 2). Other creep mechanisms, such as low-T plasticity (e.g. Mei et al. 2010) are not included in our model. The depth-dependent local strength σY is then defined as σ Y ¼ minðσ B ; σ D Þ:
ð9Þ
A minimum value of the yield stress equal to 10 MPa is assumed to define the base of the mechanical lithosphere (McNutt et al., 1988; Ranalli, 1994), while the base of the competent crustal layers is defined as the depth at which σY becomes less than 1–5% of the lithostatic pressure (Cloetingh and Burov, 1996). Two different crust–mantle stratifications have been considered for the rheological analysis, representative for a dry and a wet rheology. The values of the chosen parameters are listed in Table 2. 3.1. Lithosphere mechanical base
Fig. 4. Depth of the 1500 K isotherm (a) and 1600 K isotherm (b), assumed as the thermal lithosphere base, predicted by thermal model.
Fig. 7 shows the depth of the mechanical lithosphere base for the wet (panels a1 and a2) and dry (panels b1 and b2) rheology and for strain rates of 10−16 s−1 (panels a1 and b1) and 10−14 s−1 (panels a2 and b2). As expected, the features resemble the spatial variation of the depth of the thermal lithosphere rather well (isotherm 1600 K, Fig. 4), since the main control is due to the ductile behaviour
Finally, our model predicts a negative thermal anomaly below the Algero-Provençal Basin that disappears within a depth of 100 km (Fig. 6, panel a). These results are in general agreement with the seismic velocity anomaly pattern derived in tomographic studies, in particular with the positive velocity anomaly predicted below the suture zones of the Apennines and Dinarides and below the Ionian Sea and the AlgeroProvençal Basin, which is related to cold and more rigid material, and with the negative velocity anomalies predicted below the Tyrrhenian and Pannonian Basins (Koulakov et al., 2009; Piromallo and Morelli, 2003).
3. Rheological analysis The 3D lithosphere rheological model accounts for brittle and ductile behaviour within each lithosphere layer, the upper crust, the lower crust and the mantle. For brittle behaviour, we assume the linear failure criterion σ B ¼ β ρ g zð1−λÞ
ð7Þ
which is suitable for the behaviour of the fractured rocks (e.g. Sibson, 1974), where z is the depth, ρ is the density, g is the acceleration of gravity, and β is a parameter that depends on the type of faulting and varies from 3 (for thrust faulting) to 1.2 (for strike-slip) to 0.75 for normal faulting (Ranalli and Murphy, 1987). Parameter λ is the pore pressure factor, varying, in our analysis, from 0 (for a dry rheology) to the hydrostatic value 0.4 (representative of a wet crust (Ranalli, 1995)).
Fig. 5. Predicted temperature at the upper–lower crust transition depth (panel a) and at the Moho depth (panel b).
94
R. Splendore, A.M. Marotta / Tectonophysics 603 (2013) 89–103
Fig. 6. Predicted thermal anomaly along six vertical sections crossing the study area. Black lines represent the Moho (dashed) and the thermal lithosphere base (dotted). Anomalies are expressed as percentage of deviation from the average temperature at each depth.
Table 2 Rheological parameters used in the present study. (a) Ranalli and Murphy (1987), (b) Wilks and Carter (1990), (c) Afonso and Ranalli (2004).
Upper crust Lower crust Mantle
Lithology
ε_ 0 (Pa−n·s−1)
Wet granite (a) Dry granite (a) Undried felsic granulite (b) Dry mafic granulite (b) Wet peridotite (c) Dry peridotite (c)
7.94328 79.245 20.095 8.8334 2 4
· · · · · ·
10−16 10−30 10−22 10−22 10−21 10−12
n
Ea (J·mol−1)
1.9 3.2 3.1 4.2 4.0 3.0
137 123 243 445 471 540
· · · · · ·
103 103 103 103 103 103
of the deep lithosphere mantle. Thus, a relatively thin mechanical lithosphere paves the Pannonian Basin, from a depth of 35 km for the wet rheology at a strain rate of 10−16 s−1 (panel a1) to a depth of 45 km for the dry rheology at a strain rate of 10−14 s−1 (panel b2). A relatively thin mechanical lithosphere also characterises the central portion of the Tyrrhenian Sea, with a depth ranging from 35 km (panel a1) to 55 km (panel b2). Excluding the southernmost portion of the study domain, where the very high lithosphere thickness is a consequence of the depressed thermal field typical of a subduction setting, two relative picks localise below the Algero-Provençal (up to 70 km) and the northern sector of the Adria microplate (more than 130 km).
R. Splendore, A.M. Marotta / Tectonophysics 603 (2013) 89–103
95
Fig. 7. Predicted depth of the mechanical lithosphere for wet (panels a1 and a2) and dry (panels b1 and b2) rheology and for a strain rate of 10−16 s−1 (panels a1 and b1) and of 10−14 s−1 (panels a2 and b2). A minimum value of the yield stress equal to 10 MPa is assumed.
To better understand how the lithosphere would react under stress and, in particular, to discern its mechanical behaviour at different depths, it is useful to perform an accurate analysis of the vertical strength of the lithosphere throughout the entire study area. Particular attention should be paid to the thickness of the competent lithosphere layers, as they would affect the vertical transmission of the stress between adjacent layers of different rigidity and, thus, could be responsible for an unexpected redistribution of the deformation within the lithosphere. 3.2. Lithosphere strength Fig. 8 shows the lithosphere strength predicted by assuming strain rates equal to 10−16 s−1 (panels a1, a2, b1 and b2) and 10−14 s−1 (panels c1, c2, d1 and d2) for the wet (panels a1, a2, c1 and c2) and the dry (panels b1, b2, d1 and d2) rheology and for normal (panels a1, b1, c1 and d1) and thrust faults (panels a2, b2, c2 and d2). Significant strength gradients occur throughout the study domain, ranging from three to one and a half order of magnitude for strain rates of 10−16 s−1 and 10−14 s−1, respectively. In particular, our analysis indicates a relatively softer lithosphere below the Pannonian Basin with respect to the rest of the study domain, with a lithosphere strength ranging from 0.3 TN/m (for wet rheology, under extension and for a strain rate of 10− 16 s− 1 — Fig. 8, panel a1) to 3 TN/m (for dry rheology, under compression and for a strain rate of 10− 14 s− 1 — Fig. 8, panel d2). A strong lithosphere paves the region extending between the Tyrrhenian and the Algero-Provençal Basins, where a lithosphere strength ranging from 3 TN/m (for wet rheology, under extension
and for a strain rate of 10− 16 s− 1 — Fig. 8, panel a1) to more than 50 TN/m (for dry rheology, under compression and for a strain rate of 10−14 s−1 — Fig. 8, panel b2) is predicted. The sole exception resides below the Corsica–Sardinia complex, which shows a strength comparable to that predicted below the Pannonian area. An alternating strength pattern characterises the Adria microplate, with a very strong lithosphere below the northern and southern sectors separating a central relative softer sector, with a maximum variation of about two orders of magnitude (Fig. 8, panel b2). A progressive decrease in the strength of the lithosphere also occurs moving through the Alps from west to east. 3.3. Crust and mantle strength In order to better enlighten the contribution of crust and mantle layers in the magnitude of the lithosphere strength, here we discuss the strength of the singular layers separately. Fig. 9 shows the crust strength predicted for wet (panels a1, a2, c1 and c2) and dry (panels b1, b2, d1 and d2) rheology, assuming a strain rate equal to 10−16 s−1 (panels a1, a2, b1 and b2) or a strain rate equal to 10−14 s−1 (panels c1, c2, d1 and d2), for normal faults (panels a1, b1, c1 and d1) and thrust faults (panels a2, b2, c2 and d2). When the entire crust is considered, our analysis shows that the crustal strength predicted by wet rheology is globally softer than that predicted by the dry rheology. In particular, our model predicts a relatively strong crust through the Italian Peninsula, with the sole exception of the Tuscany area and of the Adria microplate, where the northern sector is stronger than the southern sector.
96
R. Splendore, A.M. Marotta / Tectonophysics 603 (2013) 89–103
Fig. 8. Lithosphere strength predicted assuming a wet (panels a1, a2, c1 and c2) and a dry (panels b1, b2, d1 and d2) rheology, a strain rate equal to 10−16 s−1 (panels a1, a2, b1 and b2) and 10−14 s−1 (panels c1, c2, d1 and d2), for normal faults (panels a1, b1, c1 and d1) and thrust faults (panels a2, b2, c2 and d2).
This strong crust separates the two soft crust domains paving the Pannonian Basin and the region extending between the Tyrrhenian and Algero-Provençal Basins. Furthermore, while the Pannonian Basin
shows rather uniform crustal strength, in particular for the wet rheology, significant lateral variations occur west of the Apennines, with the softest crust occurring below the Marsili–Vavilov and Algero-Provençal Basins.
R. Splendore, A.M. Marotta / Tectonophysics 603 (2013) 89–103
97
Fig. 9. Strength predicted for the whole crust assuming a wet (panels a1, a2, c1 and c2) and a dry (panels b1, b2, d1 and d2) rheology, a strain rate equal to 10−16 s−1 (panels a1, a2, b1 and b2) or a strain rate equal to 10−14 s−1 (panels c1, c2, d1 and d2), for normal faults (panels a1, b1, c1 and d1) and thrust faults (panels a2, b2, c2 and d2).
The crust belonging to the Alpine domain shows variations in strength up to one order of magnitude along the axis of the chain, from 0.5 TN/m in the westernmost sector, for wet rheology, to 40 TN/m in the easternmost sector, for dry rheology.
Although a general agreement can be observed between the crust strength pattern predicted by the present study and that predicted by previous studies (e.g., Tesauro et al., 2009b, 2011), some local differences can be worth highlighted, such as the predicted strong crust
98
R. Splendore, A.M. Marotta / Tectonophysics 603 (2013) 89–103
characterising the Italian Peninsula in contrast with the average soft crust along the same area predicted by Tesauro et al. (2009b). When the strength of the upper and lower crust are analysed separately (Fig. 10) more information can be derived. For the wet rheology, in the Pannonian Basin, the strength of the crust is controlled by the upper crust, since the strength of the lower crust is at least one order of magnitude lower than that of the upper crust. The lower crust controls the rheological differentiation below the Italian Peninsula, along the Apennines, for both wet and dry rheology: in the central Italian Peninsula, the lower crust is stronger than the upper crust, while the strength of the lower crust becomes at least one order of magnitude lower than the strength of the upper crust in the northern and southern portions of the Apennines. Dry rheology predicts a lower crust below the Adria and the Alpine region that is stronger than the upper crust, while wet rheology predicts the reverse. Finally, both the upper and lower crusts contribute to the crust strength below the region extending between the Tyrrhenian and Algero-Provençal Basins and to its lateral heterogeneities, where the strength of the two crust layers is of the same order of magnitude and exhibits similar spatial features. Fig. 11 shows the lithosphere mantle strength predicted for wet (panels a1, a2, c1 and c2) and dry (panels b1, b2, d1 and d2) rheology, a strain rate equal to 10−16 s−1 (panels a1, a2, b1 and b2) or a strain rate equal to 10−14 s−1 (panels c1, c2, d1 and d2), for normal faults (panels a1, b1, c1 and d1) and thrust faults (panels a2, b2, c2 and d2). It is worth noting the rheological contrast between the soft mantle below the Pannonian Basin with respect to the strong mantle below the Algero-Provençale and the Marsili–Vavilov Basins and below the
northern portion of the Adria microplate, where the strength magnitude reaches values greater than 100 TN/m when the dry rheology is considered, within a compressive regime. Furthermore, within the Algero Provençale and the Marsili–Vavilov basins' strong domain a remarkably rapid decrease in strength below the Sardinia–Corsica complex occurs, within 2° of longitude (up to two orders of magnitude for wet rheology and up to more than three orders of magnitude for dry rheology). Along the Italian Peninsula the mantle shows alternating strength pattern, with two soft localised areas at the Tuscany region and north of the Marsili–Vavilov basins. Moreover, unlike Southern Italy, where significant horizontal east–west strength gradients do not occur, in the northern sector a significant variation in strength occurs across the Apennines axis, separating the soft mantle in the west from the strong mantle in the east sectors. The Adria microplate can be subdivided into three sectors: two strong sectors, one in close proximity to the Gulf of Venice and the other south at about 40° latitude, separated by a third relatively softer domain, with a difference in the strength of up to more than two orders of magnitude. Moving along the axis of the Italian Alps, a progressive softening of the mantle occurs, with a rather strong domain with a strength of up to 30 TN/m (Fig. 11, panel d2) located below the westernmost Alps; the strength abruptly decreases by more than two orders of magnitude moving to the east. In order to further elucidate the relative contribution of crust and mantle to the lithosphere strength, the ratio between the strengths of the crust and mantle is shown in Fig. 12. Our results show that the relative role of crust and mantle is almost invariant with respect to rheology (wet or dry), strain rate (10−16 or 10−14 s−1) and tectonic regime (thrust or normal).
Fig. 10. Strength predicted for the upper crust (panels a1 and a2) and for the lower crust (panels b1 and b2), assuming a strain rate equal to 10−16 s−1 and for thrust faults.
R. Splendore, A.M. Marotta / Tectonophysics 603 (2013) 89–103
99
Fig. 11. Strength predicted for the lithosphere mantle, assuming a wet (panels a1, a2, c1 and c2) and a dry (panels b1, b2, d1 and d2) rheology, a strain rate equal to 10−16 s−1 (panels a1, a2, b1 and b2) or a strain rate equal to 10−14 s−1 (panels c1, c2, d1 and d2), for normal faults (panels a1, b1, c1 and d1) and thrust faults (panels a2, b2, c2 and d2).
Furthermore, in the Tyrrhenian and Algero-Provençal Basinal areas and in the northern sector of the Adria microplate the style and magnitude of the lithosphere strength is directly correlated to
the strength of the mantle and its gradients. In addition, in the Tyrrhenian and Algero-Provençal Basins, the magnitude of mantle strength is very high with respect to the crust strength, in agreement
100
R. Splendore, A.M. Marotta / Tectonophysics 603 (2013) 89–103
Fig. 12. Ratio between the crustal strength σic and the mantle strength σim, for a wet (panels a1, a2, c1 and c2) and a dry (panels b1, b2, d1 and d2) rheology, for a strain rate equal to 10−16 s−1 (panels a1, a2, b1 and b2) and 10−14 s−1 (panels c1, c2, d1 and d2), for normal faults (panels a1, b1, c1 and d1) and thrust faults (panels a2, b2, c2 and d2).
R. Splendore, A.M. Marotta / Tectonophysics 603 (2013) 89–103
with the results of Tesauro et al. (2009a). In the areas characterised by thickened crust (Alps, Dinarides, Carpathians and the Sardinia– Corsica complex) the lithosphere strength is always controlled by
101
the crust. In the central Pannonian Basin crustal and mantle contribution to the lithosphere strength are comparable. Along the Italian Peninsula, crust and mantle play the same role in controlling the
Fig. 13. Thickness and percentage of the non-competent portion of upper (panels a1, a2, b1 and b2) and lower (panels c1, c2, d1 and d2) crust, obtained for a wet (panels a1, a2, c1 and c2) and a dry (panels b1, b2, d1 and d2) rheology, assuming a strain rate equal to 10−16 s−1 and a minimum value of the yield stress equal to 1% of the lithostatic pressure.
102
R. Splendore, A.M. Marotta / Tectonophysics 603 (2013) 89–103
lithosphere strength, with the exceptions of Tuscany and the area north of the Marsili–Vavilov basin, where the strength of the mantle is negligible and the lithosphere strength is controlled by the crust. A final discussion concerns the thickness of the mechanically non-competent portions of upper and lower crust, defining the base a competent layer as the depth where the yield stress becomes less than 1–5% of the lithostatic pressure (Cloetingh and Burov, 1996). Fig. 13 shows the thickness and the percentage of the mechanically non-competent portion of the upper (panels a1, a2, b1 and b2) and lower (panels c1, c2, d1 and d2) crust, obtained for wet (panels a1, a2, c1 and c2) and dry (panels b1, b2, d1 and d2) rheology, assuming a strain rate equal to 10−16 s−1 and a minimum value of the yield stress equal to 1% of the lithostatic pressure. The primary feature that is worth highlighting is the striking difference between the western and eastern parts of the study domain. The crustal layers paving the region extending from the Tyrrhenian to the Algero-Provençal Basins result totally and unconditionally (with respect to the rheology) competent, with the sole exception of the Sardinia–Corsica complex. Here, only the 50% of the upper crust is competent for wet rheology, while for dry rheology, almost the whole upper crust is competent. For wet rheology the lower crust is totally non-competent (Fig. 13, panels c1 and c2), while for dry rheology at least the 50% of the lower crust remains competent (Fig. 13, panels d1 and d2). In the eastern part of the study domain wet rheology predicts significantly thick upper and lower crust non-competent layers, up to 70% of the total thickness for the upper crust and to 100% of the total thickness for the lower crust. Both the lateral extension and the magnitude of the thickness of the non-competent crustal layers significantly reduce when dry rheology is considered (Fig. 13, panels b1, b2, d1 and d2). The thickness of the crustal competent layers below the Italian Peninsula and the Adria microplate strongly vary from dry to wet rheology. In the first case, both the upper and the lower crusts are totally competent, while, for wet rheology, only 50% of upper crust remain competent, while up to 100% of the lower crust becomes non-competent. The central sector of the Italian peninsula shows a totally competent crustal layer. 4. Conclusions A 3D thermo-rheological model has been developed to analyse the mechanical behaviour of the crust–mantle system in the Central Mediterranean region and to investigate the occurrence of local intra-layers features. From a purely thermal point of view, our results show that the thermal field follows the major geodynamic features in the study area, with a depressed field below the suture zones, while hot temperatures characterise active extensional areas. In particular, the thickest thermal lithosphere localises below the Calabrian arc and the northern portion of the Adria microplate. A relatively thick thermal lithosphere also resides below the AlgeroProvençal Basin, compatible with thermal subsidence affecting the area in the post-rift phase, which began in the Middle Miocene (Cavazza et al., 2004). The thinnest thermal lithosphere paves the young Tyrrhenian, still actively extending, back-arc Basin (Cavazza et al., 2004; Marotta and Sabadini, 2008) and the Pannonian Basins, where extension tectonics and active volcanism are also observed. From a rheological point of view, our analysis reveals a strongly heterogeneous lithosphere strength pattern in the Central Mediterranean, characterised by steep lateral strength gradients and the occurrence of non-competent layers of significant thickness at crustal levels, located in the eastern portion of the study area. In particular, the study domain can be divided into two main parts: the western part characterised by a strong lithosphere in
which the mantle contribution is dominant and the eastern part in which a softer lithosphere is predicted and the crustal contribution to the total strength becomes more important. The sole exceptions reside in the soft Sardinia-complex and in the centre of the Pannonian Basin where a thin crust and a relatively stronger lithosphere, with respect to its boundaries, occur. The Italian Peninsula and the Adria microplate constitute a transition zone between the two main domains, exhibiting a heterogeneous rheological pattern, in which the softer Tuscany area and the region north of the Marsily–Vavilov basins and the central sector of the Adria microplate are surrounded by the harder lithosphere paving the northern sector of the Adria microplate and the Tyrrhenian Basin. The upper crust controls the total crust strength throughout the study domain when wet rheology is accounted, with the exception of small sectors of the Ionian area, Sicily and central Italy, where the contribution of the lower crust prevails. In dry rheology the upper crust controls the total crust strength only in the western portion of the study domain, from the Dinarides to the Pannonian Basin; in the rest of the study domain, with the exception of local features, the lower crust plays the main role. Non-competent crustal layers locate below the whole Pannonian area and the Sardinia–Corsica complex, that can affect up to 60% of the upper crust thickness and to 100% of the lower crust thickness. Shifting from wet to dry rheology slightly affects the general style of the rheological pattern and the mutual ratio between the contributions of the mantle and crust to the lithosphere strength. However, it does affect the extension of the non-competent domains. Acknowledgments The authors thank the editor Hans Thybo and the two anonymous reviewers for their constructive criticisms. All figures were created using GMT plotting software (Wessel and Smith 2001). References Afonso, J.C., Ranalli, G., 2004. Crustal and mantle strength in continental lithosphere: is the jelly sandwich model obsolete? Tectonophysics 394, 221–232. Artemieva, I., 2006. Global 1° × 1° thermal model TC1 for the continental lithosphere: implications for lithosphere secular evolution. Tectonophysics 416, 245–277. Artemieva, I., Mooney, W.D., 2001. Thermal thickness and evolution of Precambrian lithosphere: a global study. Journal of Geophysical Research 106 (B8), 16387–16414. Boschi, L., Ekström, G., Kustowski, B., 2004. Multiple resolution surface wave tomography: the Mediterranean basin. Geophysical Journal International 157, 293–304. Brandmayr, E., Marson, I., Romanelli, F., Panza, G.F., 2011. Lithosphere density model in Italy: no hint for slab pull. Terra Nova 23, 292–299. Cavazza, W., Roure, F., Ziegler, P.A., 2004. The Mediterranean area and the surrounding regions: active processes, remnants of former Tethyan oceans and related thrust belts. In: Cavazza, W., Roure, F., Spakman, W., Stampfli, G.M., Ziegler, P.A. (Eds.), The Transmed Atlas — The Mediterranean Region From Crust to Mantle. Springer, pp. 1–29. Chiarabba, C., De Gregori, P., Speranza, F., 2008. The southern Tyrrhenian subduction zone: deep geometry, magmatism and Plio-Pleistocene evolution. Earth and Planetary Science Letters 268, 408–423. Cloetingh, S.A.P.L., Burov, E.B., 1996. Thermomechanical structure of European continental lithosphere: constraints from rheological profiles and EET estimates. Geophysical Journal International 124, 695–723. Cloetingh, S.A.P.L., Van Wees, J.D., Ziegler, P.A., Lenkey, L., Beekman, F., Tesauro, M., Forster, A., Norden, B., Kaban, N., Hardebol, N., Bontè, D., Genter, A., GuillouFrottier, L., Ter Voorde, M., Sokoutis, D., Willingshofer, E., Cornu, T., Worum, G., 2010. Lithosphere tectonics and thermo-mechanical properties: an integrated modelling approach for enhanced geothermal system exploration in Europe. Earth Science Review 102, 159–206. Fernández, M., Marzan, I., Correia, A., Ramalho, E., 1998. Heat flow, heat production, and lithospheric thermal regime in the Iberian Peninsula. Tectonophysics 291, 29–53. Fry, L., Boschi, L., Ekström, G., Giardini, D., 2008. Europe–Mediterranean tomography: high correlation between new seismic data and independent geophysical observables. Geophysical Research Letters 35, L04301. Goes, S., Govers, R., Vacher, P., 2000. Shallow mantle temperatures under Europe from P and S wave tomography. Journal of Geophysical Research 105, 11153–11169. Huismans, R., Podladchikov, Y.Y., Cloetingh, S.A.P.L., 2001. Dynamic modeling of the transition from passive to active rifting, application to the Pannonian basin. Tectonics 20 (6), 1021–1039.
R. Splendore, A.M. Marotta / Tectonophysics 603 (2013) 89–103 Jiménez-Munt, I., Sabadini, R., Gardi, A., Bianco, G., 2003. Active deformation in the Mediterranean from Gibraltar to Anatolia inferred from numerical modeling, geodetic and seismological data. Journal of Geophysical Research 108 (B1). Koulakov, I., Kaban, M.K., Tesauro, M., Cloetingh, S.A.P.L., 2009. P- and S-velocity anomalies in the upper mantle beneath Europe from tomographic inversion of ISC data. Geophysical Journal International 179, 345–366. Marone, F., Van der Lee, S., Giardini, D., 2004. Three-dimensional upper–mantle Svelocity model for the Eurasia–Africa plate boundary region. Geophysical Journal International 158, 109–130. Marotta, A.M., Mitrovica, J.M., Sabadini, R., Milne, M., 2004. Combined effects of Tectonics and Glacial Isostatic Adjustment on intraplate deformation in Central and Northern Europe: applications to Geodetic Baseline Analysis. Journal of Geophysical Research 109, B01413-1-B01413-22. http://dx.doi.org/10.1029/2002JB002337. Marotta, A.M., Sabadini, R., 2008. Africa-Eurasia kinematics control of long-wavelength tectonic deformation in the Central Mediterranean. Geophysical Journal International 175, 742–754. http://dx.doi.org/10.1111/j.1365-246X.2008.03906.x. McNutt, M.K., Diament, M., Kogan, M.G., 1988. Variations of elastic plate thickness at continental thrust belts. Journal of Geophysical Research 93, 8825–8838. Mei, S., Suzuki, A.M., Kohlstedt, D.L., Dickson, N.A., Duram, W.B., 2010. Experimental constraints on the strength of the lithospheric mantle. Journal of Geophysical Research 115, B08204. http://dx.doi.org/10.1029/2009JB006873. Mooney, W.D., Laske, G., Masters, T.G., 1998. CRUST 5.1: a global crustal model at 5° × 5°. Journal of Geophysical Research 103, 727–747. Piromallo, C., Morelli, A., 2003. P wave tomography of the mantle under the Alpine– Mediterranean area. Journal of Geophysical Research 108 (B2). Pollack, H.N., Hurter, S., Johnson, J.R., 1993. Heat flow from the Earth's interior: analysis of the global data set. Reviews of Geophysics 31 (3), 267–280. Ranalli, G., 1994. Nonlinear flexure and equivalent mechanical thickness of the lithosphere. Tectonophysics 240, 107–114. Ranalli, G., 1995. Rheology of the Earth, second edition. Chapman & Hall (413 pp.). Ranalli, G., Murphy, D.C., 1987. Rheological stratification of the lithosphere. Tectonophysics 132, 281–295. Rybach, L., 1988. Determination of heat production rate. In: Haenel, R., Raybach, L., Stegena, L. (Eds.), Handbook of Terrestrial Heat Flow Density Determination. Kluwer Acad, Dordrecht, Netherlands.
103
Sibson, R.H., 1974. Frictional constraints on thrust, wrench and normal faults. Nature 249, 542–544. Sobolev, S.V., Zeyen, H., Granet, M., Achauer, U., Bauer, C., Werling, F., Altherr, R., Fuchs, K., 1997. Upper mantle temperatures and lithosphere–asthenosphere system beneath the French Massif Central constrained by seismic, gravity, petrologic and thermal observations. Tectonophysics 275, 143–164. Splendore, R., Marotta, A.M., Barzaghi, R., Borghi, A., Cannizzaro, L., 2010. Block model versus thermomechanical model: new insights on the present-day regional deformation in the surroundings of the Calabrian Arc. In: Spalla, M.I., Marotta, A.M., Gosso, G. (Eds.), Advances in Interpretation of Geological Processes: Refinement of Multiscale Data and Integration in Numerical Modelling: Geological Society, London, Special Publications, 332, pp. 129–147. Tesauro, M., Kaban, M.K., Cloetingh, S.A.P.L., 2008. EuCRUST-07: a new reference model for the European crust. Geophysical Research Letters 35, L05313. Tesauro, M., Kaban, M.K., Cloeting, S.A.P.L., 2009a. How rigid is Europe's lithosphere? Geophysical Research Letters 36, L16303. Tesauro, M., Kaban, M.K., Cloeting, S.A.P.L., 2009b. A new thermal and rheological model of the European lithosphere. Tectonophysics 476, 478–495. Tesauro, M., Burov, E.B., Kaban, M.K., Cloeting, S.A.P.L., 2011. Ductile crustal flow in Europe's lithosphere. Earth and Planetary Science Letters 312, 254–265. Turcotte, D.L., Schubert, G., 2002. Geodynamics. Ed. Cambridge Univ. Press, New York. Vilà, M., Fernández, M., Jiménez-Munt, I., 2010. Radiogenic heat production variability of some common lithological groups and its significance to lithospheric thermal modeling. Tectonophysics 490, 152–264. Weertman, J., Weertman, J.R., 1975. High temperature creep of rock and mantle viscosity. Annual Review of Earth and Planetary Sciences 3, 293–315. Wessel, P., Smith, W.M.F., 2001. New improved version of generic mapping tools released. EOS, Trans. Am. Geophys. Un. 79, 579. Wilks, K.R., Carter, N.L., 1990. Rheology of some continental lower crustal rocks. Tectonophysics 182, 57–77.
Contents lists available at ScienceDirect
Tectonophysics journal homepage: www.elsevier.com/locate/tecto
Crust-mantle mechanical structure in the Central Mediterranean region Raffaele Splendore, Anna Maria Marotta ⁎ Università degli Studi di Milano, Dipartimento di Scienze della Terra “Ardito Desio”, L. Cicognara 7, 20129 Milano, Italy
a r t i c l e
i n f o
Article history: Received 1 October 2012 Received in revised form 9 May 2013 Accepted 21 May 2013 Available online 31 May 2013 Keywords: Lithosphere thermal field Rheological analysis Lithosphere strength heterogeneities Central Mediterranean
a b s t r a c t A thermo-rheological analysis is performed to elucidate the mechanical behaviour of the lithosphere in the Mediterranean domain. The thermal field of the lithosphere is calculated by means of a finite element 3D thermal model. The decrease in radiogenic heat production with depth is taken into account, together with the compositional layering of the lithosphere. The predicted thermal field is analysed in terms of the temperatures and depth of the thermal lithosphere base. Based on the predicted thermal field, a rheological analysis is conducted, accounting for both the brittle and ductile behaviour of each lithospheric layer. The effects of the choice of wet and dry rheology are also investigated. Our rheological analysis reveals a strongly heterogeneous lithosphere strength pattern in the Central Mediterranean, characterised by strong lateral strength gradients and the occurrence of non-competent crustal layers of significant thickness in the eastern portion of the study area. © 2013 Elsevier B.V. All rights reserved.
1. Introduction An accurate knowledge of the lithosphere strength and its variation in space is crucial when an appropriate 3D analysis of the deformation is pursued, especially in very complex tectonic areas, such as the Mediterranean, where localised deformation features are expected and an average rheological statement of the lithosphere results not adequate (e.g. Marotta et al., 2004). Furthermore, when a sufficiently refined thermo-rheological model of the lithosphere is available, it is possible to reveal intra-lithosphere soft zones that might be responsible for mechanical decoupling at different crust or mantle levels. One crucial aspect of lithosphere rheology analysis is an adequate knowledge of the thermal state of the lithosphere, particularly when small-scale rheological discontinuities may occur. While other rheological analyses use indirect information about the thermal field of the lithosphere, based on, for example, mantle tomographic studies (e.g., Tesauro et al., 2009b), we prefer to use a 3D thermal model based on the measures of surface heat flow and appropriate assumptions for the thermal parameters of the crust and mantle rocks. Thus, we obtain a 3D lithosphere thermal field that is completely consistent with the subsequent rheological analysis. Over the past 10 years, the knowledge of the thermal structure of the lithosphere has extended from the global scale (e.g., Artemieva, 2006 and reference therein) to the regional scale of the Mediterranean region (e.g., Jimenez-Munt et al., 2003; Splendore et al., 2010). One of the main objectives of these studies was to identify the depth of the thermal base of the lithosphere by the resolution of the steady-state heat conduction equation for the lithosphere, which is constrained by ⁎ Corresponding author. Tel.: +39 02 50318470; fax: +39 02 50318489. E-mail address: [email protected] (A.M. Marotta). 0040-1951/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.tecto.2013.05.017
a fixed crust thickness and surface heat flow. In particular, Artemieva (2006) developed a global 1°x1° thermal model for the continental lithosphere using borehole measurements of heat flow and xenolith data and a 1°x1° crustal thickness model (CRUST5.1, Mooney et al., 1998). Jimenez-Munt et al. (2003) performed a more regional analysis that was based on an isostatically rather than seismically defined crust model. Splendore et al. (2010) performed a 3D thermal analysis of the surroundings of the Italian Peninsula that did not account for an accurate crustal stratification, thus producing smooth thermal gradients and a rather thin lithosphere. The present study aims to improve the thermal model developed by Splendore et al. (2010) in two major respects: (1) by considering a spatially refined modelled volume, both vertically and horizontally and (2) by introducing a more sophisticated compositional stratification of the crust. Predictions from the thermal model are corroborated by a qualitative comparison with the available seismic tomography of the area. Seismic tomography, in fact, maps the seismic velocity heterogeneities of Earth's interior. Moreover, while wave velocities within the crust are mainly controlled by compositional heterogeneities, in the upper mantle, changes in wave velocities are mainly attributed to temperature variation (Piromallo and Morelli, 2003), with a small contribution, i.e., less than 1%, from mantle composition (Goes et al., 2000; Sobolev et al., 1997). Positive seismic velocity anomalies are generally related to cold material and thus to depressed thermal fields, while negative seismic velocity anomalies are typical of hot material and high temperatures. Both P- and S-wave velocity distributions are used to construct tomographic models, but S-wave models are less consistent than those based on P-waves, providing a lower horizontal resolution (Koulakov et al., 2009). Finally, the predicted thermal field is used to determine the strength of the lithosphere and its heterogeneities at both crust and mantle
90
R. Splendore, A.M. Marotta / Tectonophysics 603 (2013) 89–103
levels. In the present work, we analyse the main features of lithosphere strength through the central Mediterranean region, focusing on whether and where the interplay between regional thermal fields, crust and mantle thicknesses, lithosphere composition stratification and strain rates weakens intra-plate levels, that may result in decoupling features. 2. Thermal analysis The lithosphere thermal field is calculated by the numerical integration of the steady state 3D energy equation ∂ ∂T ∂ ∂T ∂ ∂T þ þ þ ρH ðzÞ ¼ 0 K⋅ K⋅ K⋅ ∂x ∂x ∂y ∂y ∂z ∂z
ð1Þ
over the domain shown in Fig. 1 (panel a), where K is the thermal conductivity, T is the temperature, ρ is the density and H is the rate of radiogenic heat production per mass unit. No hydrothermal contribution is accounted for through the crust. Boundary conditions correspond to fixed temperature at the upper (300 K) and lower (1600 K) boundaries and to zero heat flow across the vertical sidewalls. For what concerns the geometry of the 3D domain where the numerical integration is performed, both lateral and upper boundaries of the study area are fixed and coincide with the major tectonic lineaments characterising the central Mediterranean region (the Alpine chain to the north, the Carpathian belt and the Anatolian faults system to the east and the Africa and Eurasia boundary to the south) and with the regional topography (Fig. 1, panel b), respectively.
Fig. 1. (a) Scheme of the 3D grid used for the thermo-rheological analysis. (b) Topography used to define the upper boundary of the 3D model. Data are taken from EuCRUST07 model (Tesauro et al., 2008).
R. Splendore, A.M. Marotta / Tectonophysics 603 (2013) 89–103
The lower boundary of the model, on the contrary, has not an a-priory fixed geometry but it is a model unknown itself. In order to solve for this unknown we use an iterative procedure in which, at each iteration, the geometry of the lower boundary of the model is modified accordingly to the predicted thermal field. In particular, let ZL(ite) be the depth of the thermal lithosphere base used at iteration ite to calculate the 3D thermal field, T(ite), and qs(ite) the predicted surface heat flow, ZL is modified assuming h i obs ΔZ L ðiteÞ ¼ α qs ðiteÞ−qs
ð2Þ
where qobs is the observed surface heat flow (Fig. 2). s In order to minimize the occurrence of abrupt lateral gradients, the proportional constant α is chosen which is equal to 0.05 and a maximum variation in ΔZL(ite) of ± 1 km is allowed. We use a classical χ2 analysis to calculate the quantity h 2
χ ðiteÞ ¼
qs ðiteÞ−qobs s
i2
ndof
ð3Þ
with ndof number of nodes at the surface. The iterative procedure stops when 2
2
χ ðiteÞ−χ ðite−1Þ≤0:01:
ð4Þ
At the first iteration the base of the thermal lithosphere (depth of the 1600 K) is derived from the 1D simple analytical solution T ðzÞ ¼ T s þ
qobs s z K
ð5Þ
obtained for a homogeneous layer with prescribed surface temperature (Ts) and heat flow (qobs s ), equal to 300 K and to the observed heat flow, respectively. The lithosphere is compositionally differentiated into an upper crust, lower crust and mantle. The 3D domain is numerically discretised into 26,805 triangular-based linear prismatic elements with an average dimension of 0.5°× 0.5° in the horizontal direction and a length of 2 km (within the crust) to 20 km (in the lithosphere mantle) in the vertical direction. The geometries of the upper–lower crust transition and Moho (Fig. 3, panels a and c) are based on the EuCRUST07 model (Tesauro et al., 2008). Table 1 lists the physical and thermal rock
Fig. 2. Distribution of the surface heat flow in the Adriatic region, obtained by linear interpolation of the global database of Pollack et al. (1993), augmented by Fernandez et al. (1998) and Artemieva (2006). The extremely high values of surface heat flow in the Larderello and Stromboli areas, which are likely related to non-conductive processes, such as local hydrothermal convection or volcanic processes, are excluded.
91
properties assumed for the upper and lower crusts and mantle. An exponential decrease in radiogenic heat production with depth is assumed within the crust according to −hz
HðzÞ ¼ H s e
r
ð6Þ
where Hs is the radiogenic heat production at the surface, z is the depth, and hr is a length scale for the decrease in H with depth. We use hr = 10 km, in agreement with the average value for continental crust (Turcotte and Schubert, 2002). Constant radiogenic heat production with depth is assumed within the mantle. Lithosphere temperature will be discussed in terms of the geometry of the base of the thermal lithosphere, defined as the depth of the 1600 K isotherm, the temperature at the upper crust–lower crust interface and the temperature at the Moho. To strengthen our thermal analysis, we also calculate the predicted thermal anomalies, computed as the percentage of temperature deviance with respect to the average temperature characterising the study area at the same depth, both horizontally and vertically, and qualitatively compare them to the regional seismic tomography. We are aware that the assumption of steady state thermal field is rather strong when a relatively recent tectono-thermal region, such as the Mediterranean, is analysed. However, since the poor knowledge of temporal constraints does not allow developing a rigorous time dependent 3D thermal model in the area, we adopt the stationary simple formulation used by several authors for the same area (e.g. Cloetingh et al., 2010; Jimenez-Munt et al., 2003). 2.1. Predicted 3D thermal field Fig. 4 shows the predicted base of the lithosphere in terms of the depth of the 1500 K (panel a) and 1600 K (panel b) isotherms. The Tyrrhenian and Pannonian Basins, where high values of observed surface heat flow are accompanied by a thin seismic crust, are characterised by a thin lithosphere, whose base depth ranges from 40 to 70 km. The lithosphere base deepens to approximately 100 km in the Algero-Provençal Basin and to more than 200 km between the Apennines and Dinarides belts, while it reaches its uppermost values in the Ionian region, where very low surface heat flow is measured and subduction has depressed the thermal field. The style of lateral variation in the lithosphere thickness predicted by our model is in agreement with that predicted by previous thermal models, such as those of Jimenez-Munt et al. (2003) and Artemieva (2006), in which the thermal lithosphere base follows the major tectonic and geodynamic features of the Mediterranean area. In particular, the thickest lithosphere is located below the Calabrian zone and the Apennines and Dinarides belts, while the Tyrrhenian and Pannonian Basins are paved by the thinnest lithosphere. However, some differences can be observed. With respect to the models of Artemieva (2006) and Jimenez-Munt et al. (2003), our model suggests a deeper lithosphere base under the Apennines and Dinarides belts and below the Calabrian area. It is worth mentioning that the thick lithosphere that our model predicts below the AlgeroProvençal Basin stands in contrast with the thin lithosphere predicted by Jimenez-Munt et al. (2003). One possible reason for these discrepancies may be the fact that a different dataset was used to define the crust: the global analysis by Artemieva (2006) is based on the CRUST5.1 crustal model (Mooney et al., 1998); Jimenez-Munt et al. (2003) accounted for isostasy within a 1D thermal model when deriving crustal thickness. Our 3D thermal analysis is based on the crustal model EuCRUST07 (Tesauro et al., 2008), derived by seismic reflection, refraction and receiver function studies, which provides a fine database for topography, upper–lower crust transition and Moho depth over a uniform grid of 15′ × 15′, making it more suitable to resolve small crust and mantle heterogeneities.
92
R. Splendore, A.M. Marotta / Tectonophysics 603 (2013) 89–103
Fig. 3. Depth of the upper–lower crust transition (panel a), depth of the Moho (panel c) and thickness of the upper crust (panel b) and the lower crust (panel d). Data are taken from EuCRUST07 model (Tesauro et al., 2008).
Fig. 5 shows the temperature predicted at the upper/lower crust interface and at the Moho in panels a and b, respectively. A cold crust characterises the basinal areas, which feature a very thin crust and a shallow Moho. The highest temperatures are predicted below the Dinarides and eastern Alps, where the roots of mountain belts reach great depths (50–60 km, Fig. 3, panel c), with a temperature of 800 K predicted at the interface between the upper and lower crusts and a temperature above 1100 K predicted at the Moho depth. The Apennines chain is characterised by a lower temperature at the Moho depth of approximately 900 K; in the northern part of the chain, however, a temperature of 1100 K is reached at a depth of approximately 45 km (Fig. 3, panel c). The peak temperature of 900 to 1000 K at the upper–lower crust transition depth is achieved below the eastern domain of the eastern Alps and along the Dinaries (Fig. 5, panel a), where a very thick upper crust is observed (Fig. 3, panel b), which is associated with a thin lower crust (Fig. 3, panel d).
Table 1 Thermal and physical properties used in the present study. Layer
Composition
Densitya (kg m−3)
Thermal conductivityb,c (Wm−1 K−1)
Radiogenic heat productiona (μW m−3)
Upper crust Lower crust Mantle
Granite Granulite Depleted peridotite
2700 2900 3200
3.1 2.0 4.15
2.43 0.4 0.002
a b c
Vilà et al. (2010). Rybach (1988). Artemieva and Mooney (2001).
2.2. Temperature anomalies In order to better enlighten deep hot and cold domains, suitable for comparison with the available tomography of the Mediterranean area (e.g., Boschi et al., 2004; Brandmayr et al., 2011; Fry et al., 2008; Koulakov et al., 2009; Marone et al., 2004; Piromallo and Morelli, 2003), at each point within the 3D study domain, we calculate the percentage of deviation of the predicted temperature from the average horizontal temperature at the same depth. Our model predicts a negative temperature anomaly, between −5% and more than −40%, below the Adria microplate and Calabrian arc, continuous in depth within the entire lithosphere (Fig. 6), with two peak zones: the first located in the northern Adriatic Sea (panels a and e) and the second below the Ionian Sea (panel c), where a negative peak value of more than −40% is reached within a depth of 50 km. The depressed thermal field associated with the thermal negative anomalies is consistent with that produced by subduction. The gap in the thermal negative anomaly between the northern portion of the Adria microplate and the Ionian Sea may be the result of hot material rising at shallow depths through the mechanical discontinuity separating the Adriatic and Ionian slabs, as supported by tomographic studies (Koulakov et al., 2009) and explained by Chiarabba et al. (2008) as the result of updoming astenospheric material that mechanically separates the Ionian and Adriatic slabs. The Tyrrhenian Basin is characterised by a positive thermal anomaly, with the peak value located in the southwestern zone (Fig. 6, panels c, d and f). Moving to the east, another significant positive thermal anomaly localises below the Pannonian Basin (Fig. 6, panel b), compatible with the upwelling of hot mantle during the opening of the basin (Huismans et al., 2001). In both cases, the positive thermal anomalies affect the entire lithosphere thickness.
R. Splendore, A.M. Marotta / Tectonophysics 603 (2013) 89–103
93
For ductile behaviour, we assume the power law (Weertman and Weertman, 1975) σD ¼
ε_ ε_ 0
1 =
n
Ea eðnRTÞ
ð8Þ
where ε_ is the strain rate (ranging from 10−16 to 10−14 s−1), R is the universal constant of gas, T is the temperature, and ε_ 0 , n, and Ea are the rheological parameters of the rock (Table 2). Other creep mechanisms, such as low-T plasticity (e.g. Mei et al. 2010) are not included in our model. The depth-dependent local strength σY is then defined as σ Y ¼ minðσ B ; σ D Þ:
ð9Þ
A minimum value of the yield stress equal to 10 MPa is assumed to define the base of the mechanical lithosphere (McNutt et al., 1988; Ranalli, 1994), while the base of the competent crustal layers is defined as the depth at which σY becomes less than 1–5% of the lithostatic pressure (Cloetingh and Burov, 1996). Two different crust–mantle stratifications have been considered for the rheological analysis, representative for a dry and a wet rheology. The values of the chosen parameters are listed in Table 2. 3.1. Lithosphere mechanical base
Fig. 4. Depth of the 1500 K isotherm (a) and 1600 K isotherm (b), assumed as the thermal lithosphere base, predicted by thermal model.
Fig. 7 shows the depth of the mechanical lithosphere base for the wet (panels a1 and a2) and dry (panels b1 and b2) rheology and for strain rates of 10−16 s−1 (panels a1 and b1) and 10−14 s−1 (panels a2 and b2). As expected, the features resemble the spatial variation of the depth of the thermal lithosphere rather well (isotherm 1600 K, Fig. 4), since the main control is due to the ductile behaviour
Finally, our model predicts a negative thermal anomaly below the Algero-Provençal Basin that disappears within a depth of 100 km (Fig. 6, panel a). These results are in general agreement with the seismic velocity anomaly pattern derived in tomographic studies, in particular with the positive velocity anomaly predicted below the suture zones of the Apennines and Dinarides and below the Ionian Sea and the AlgeroProvençal Basin, which is related to cold and more rigid material, and with the negative velocity anomalies predicted below the Tyrrhenian and Pannonian Basins (Koulakov et al., 2009; Piromallo and Morelli, 2003).
3. Rheological analysis The 3D lithosphere rheological model accounts for brittle and ductile behaviour within each lithosphere layer, the upper crust, the lower crust and the mantle. For brittle behaviour, we assume the linear failure criterion σ B ¼ β ρ g zð1−λÞ
ð7Þ
which is suitable for the behaviour of the fractured rocks (e.g. Sibson, 1974), where z is the depth, ρ is the density, g is the acceleration of gravity, and β is a parameter that depends on the type of faulting and varies from 3 (for thrust faulting) to 1.2 (for strike-slip) to 0.75 for normal faulting (Ranalli and Murphy, 1987). Parameter λ is the pore pressure factor, varying, in our analysis, from 0 (for a dry rheology) to the hydrostatic value 0.4 (representative of a wet crust (Ranalli, 1995)).
Fig. 5. Predicted temperature at the upper–lower crust transition depth (panel a) and at the Moho depth (panel b).
94
R. Splendore, A.M. Marotta / Tectonophysics 603 (2013) 89–103
Fig. 6. Predicted thermal anomaly along six vertical sections crossing the study area. Black lines represent the Moho (dashed) and the thermal lithosphere base (dotted). Anomalies are expressed as percentage of deviation from the average temperature at each depth.
Table 2 Rheological parameters used in the present study. (a) Ranalli and Murphy (1987), (b) Wilks and Carter (1990), (c) Afonso and Ranalli (2004).
Upper crust Lower crust Mantle
Lithology
ε_ 0 (Pa−n·s−1)
Wet granite (a) Dry granite (a) Undried felsic granulite (b) Dry mafic granulite (b) Wet peridotite (c) Dry peridotite (c)
7.94328 79.245 20.095 8.8334 2 4
· · · · · ·
10−16 10−30 10−22 10−22 10−21 10−12
n
Ea (J·mol−1)
1.9 3.2 3.1 4.2 4.0 3.0
137 123 243 445 471 540
· · · · · ·
103 103 103 103 103 103
of the deep lithosphere mantle. Thus, a relatively thin mechanical lithosphere paves the Pannonian Basin, from a depth of 35 km for the wet rheology at a strain rate of 10−16 s−1 (panel a1) to a depth of 45 km for the dry rheology at a strain rate of 10−14 s−1 (panel b2). A relatively thin mechanical lithosphere also characterises the central portion of the Tyrrhenian Sea, with a depth ranging from 35 km (panel a1) to 55 km (panel b2). Excluding the southernmost portion of the study domain, where the very high lithosphere thickness is a consequence of the depressed thermal field typical of a subduction setting, two relative picks localise below the Algero-Provençal (up to 70 km) and the northern sector of the Adria microplate (more than 130 km).
R. Splendore, A.M. Marotta / Tectonophysics 603 (2013) 89–103
95
Fig. 7. Predicted depth of the mechanical lithosphere for wet (panels a1 and a2) and dry (panels b1 and b2) rheology and for a strain rate of 10−16 s−1 (panels a1 and b1) and of 10−14 s−1 (panels a2 and b2). A minimum value of the yield stress equal to 10 MPa is assumed.
To better understand how the lithosphere would react under stress and, in particular, to discern its mechanical behaviour at different depths, it is useful to perform an accurate analysis of the vertical strength of the lithosphere throughout the entire study area. Particular attention should be paid to the thickness of the competent lithosphere layers, as they would affect the vertical transmission of the stress between adjacent layers of different rigidity and, thus, could be responsible for an unexpected redistribution of the deformation within the lithosphere. 3.2. Lithosphere strength Fig. 8 shows the lithosphere strength predicted by assuming strain rates equal to 10−16 s−1 (panels a1, a2, b1 and b2) and 10−14 s−1 (panels c1, c2, d1 and d2) for the wet (panels a1, a2, c1 and c2) and the dry (panels b1, b2, d1 and d2) rheology and for normal (panels a1, b1, c1 and d1) and thrust faults (panels a2, b2, c2 and d2). Significant strength gradients occur throughout the study domain, ranging from three to one and a half order of magnitude for strain rates of 10−16 s−1 and 10−14 s−1, respectively. In particular, our analysis indicates a relatively softer lithosphere below the Pannonian Basin with respect to the rest of the study domain, with a lithosphere strength ranging from 0.3 TN/m (for wet rheology, under extension and for a strain rate of 10− 16 s− 1 — Fig. 8, panel a1) to 3 TN/m (for dry rheology, under compression and for a strain rate of 10− 14 s− 1 — Fig. 8, panel d2). A strong lithosphere paves the region extending between the Tyrrhenian and the Algero-Provençal Basins, where a lithosphere strength ranging from 3 TN/m (for wet rheology, under extension
and for a strain rate of 10− 16 s− 1 — Fig. 8, panel a1) to more than 50 TN/m (for dry rheology, under compression and for a strain rate of 10−14 s−1 — Fig. 8, panel b2) is predicted. The sole exception resides below the Corsica–Sardinia complex, which shows a strength comparable to that predicted below the Pannonian area. An alternating strength pattern characterises the Adria microplate, with a very strong lithosphere below the northern and southern sectors separating a central relative softer sector, with a maximum variation of about two orders of magnitude (Fig. 8, panel b2). A progressive decrease in the strength of the lithosphere also occurs moving through the Alps from west to east. 3.3. Crust and mantle strength In order to better enlighten the contribution of crust and mantle layers in the magnitude of the lithosphere strength, here we discuss the strength of the singular layers separately. Fig. 9 shows the crust strength predicted for wet (panels a1, a2, c1 and c2) and dry (panels b1, b2, d1 and d2) rheology, assuming a strain rate equal to 10−16 s−1 (panels a1, a2, b1 and b2) or a strain rate equal to 10−14 s−1 (panels c1, c2, d1 and d2), for normal faults (panels a1, b1, c1 and d1) and thrust faults (panels a2, b2, c2 and d2). When the entire crust is considered, our analysis shows that the crustal strength predicted by wet rheology is globally softer than that predicted by the dry rheology. In particular, our model predicts a relatively strong crust through the Italian Peninsula, with the sole exception of the Tuscany area and of the Adria microplate, where the northern sector is stronger than the southern sector.
96
R. Splendore, A.M. Marotta / Tectonophysics 603 (2013) 89–103
Fig. 8. Lithosphere strength predicted assuming a wet (panels a1, a2, c1 and c2) and a dry (panels b1, b2, d1 and d2) rheology, a strain rate equal to 10−16 s−1 (panels a1, a2, b1 and b2) and 10−14 s−1 (panels c1, c2, d1 and d2), for normal faults (panels a1, b1, c1 and d1) and thrust faults (panels a2, b2, c2 and d2).
This strong crust separates the two soft crust domains paving the Pannonian Basin and the region extending between the Tyrrhenian and Algero-Provençal Basins. Furthermore, while the Pannonian Basin
shows rather uniform crustal strength, in particular for the wet rheology, significant lateral variations occur west of the Apennines, with the softest crust occurring below the Marsili–Vavilov and Algero-Provençal Basins.
R. Splendore, A.M. Marotta / Tectonophysics 603 (2013) 89–103
97
Fig. 9. Strength predicted for the whole crust assuming a wet (panels a1, a2, c1 and c2) and a dry (panels b1, b2, d1 and d2) rheology, a strain rate equal to 10−16 s−1 (panels a1, a2, b1 and b2) or a strain rate equal to 10−14 s−1 (panels c1, c2, d1 and d2), for normal faults (panels a1, b1, c1 and d1) and thrust faults (panels a2, b2, c2 and d2).
The crust belonging to the Alpine domain shows variations in strength up to one order of magnitude along the axis of the chain, from 0.5 TN/m in the westernmost sector, for wet rheology, to 40 TN/m in the easternmost sector, for dry rheology.
Although a general agreement can be observed between the crust strength pattern predicted by the present study and that predicted by previous studies (e.g., Tesauro et al., 2009b, 2011), some local differences can be worth highlighted, such as the predicted strong crust
98
R. Splendore, A.M. Marotta / Tectonophysics 603 (2013) 89–103
characterising the Italian Peninsula in contrast with the average soft crust along the same area predicted by Tesauro et al. (2009b). When the strength of the upper and lower crust are analysed separately (Fig. 10) more information can be derived. For the wet rheology, in the Pannonian Basin, the strength of the crust is controlled by the upper crust, since the strength of the lower crust is at least one order of magnitude lower than that of the upper crust. The lower crust controls the rheological differentiation below the Italian Peninsula, along the Apennines, for both wet and dry rheology: in the central Italian Peninsula, the lower crust is stronger than the upper crust, while the strength of the lower crust becomes at least one order of magnitude lower than the strength of the upper crust in the northern and southern portions of the Apennines. Dry rheology predicts a lower crust below the Adria and the Alpine region that is stronger than the upper crust, while wet rheology predicts the reverse. Finally, both the upper and lower crusts contribute to the crust strength below the region extending between the Tyrrhenian and Algero-Provençal Basins and to its lateral heterogeneities, where the strength of the two crust layers is of the same order of magnitude and exhibits similar spatial features. Fig. 11 shows the lithosphere mantle strength predicted for wet (panels a1, a2, c1 and c2) and dry (panels b1, b2, d1 and d2) rheology, a strain rate equal to 10−16 s−1 (panels a1, a2, b1 and b2) or a strain rate equal to 10−14 s−1 (panels c1, c2, d1 and d2), for normal faults (panels a1, b1, c1 and d1) and thrust faults (panels a2, b2, c2 and d2). It is worth noting the rheological contrast between the soft mantle below the Pannonian Basin with respect to the strong mantle below the Algero-Provençale and the Marsili–Vavilov Basins and below the
northern portion of the Adria microplate, where the strength magnitude reaches values greater than 100 TN/m when the dry rheology is considered, within a compressive regime. Furthermore, within the Algero Provençale and the Marsili–Vavilov basins' strong domain a remarkably rapid decrease in strength below the Sardinia–Corsica complex occurs, within 2° of longitude (up to two orders of magnitude for wet rheology and up to more than three orders of magnitude for dry rheology). Along the Italian Peninsula the mantle shows alternating strength pattern, with two soft localised areas at the Tuscany region and north of the Marsili–Vavilov basins. Moreover, unlike Southern Italy, where significant horizontal east–west strength gradients do not occur, in the northern sector a significant variation in strength occurs across the Apennines axis, separating the soft mantle in the west from the strong mantle in the east sectors. The Adria microplate can be subdivided into three sectors: two strong sectors, one in close proximity to the Gulf of Venice and the other south at about 40° latitude, separated by a third relatively softer domain, with a difference in the strength of up to more than two orders of magnitude. Moving along the axis of the Italian Alps, a progressive softening of the mantle occurs, with a rather strong domain with a strength of up to 30 TN/m (Fig. 11, panel d2) located below the westernmost Alps; the strength abruptly decreases by more than two orders of magnitude moving to the east. In order to further elucidate the relative contribution of crust and mantle to the lithosphere strength, the ratio between the strengths of the crust and mantle is shown in Fig. 12. Our results show that the relative role of crust and mantle is almost invariant with respect to rheology (wet or dry), strain rate (10−16 or 10−14 s−1) and tectonic regime (thrust or normal).
Fig. 10. Strength predicted for the upper crust (panels a1 and a2) and for the lower crust (panels b1 and b2), assuming a strain rate equal to 10−16 s−1 and for thrust faults.
R. Splendore, A.M. Marotta / Tectonophysics 603 (2013) 89–103
99
Fig. 11. Strength predicted for the lithosphere mantle, assuming a wet (panels a1, a2, c1 and c2) and a dry (panels b1, b2, d1 and d2) rheology, a strain rate equal to 10−16 s−1 (panels a1, a2, b1 and b2) or a strain rate equal to 10−14 s−1 (panels c1, c2, d1 and d2), for normal faults (panels a1, b1, c1 and d1) and thrust faults (panels a2, b2, c2 and d2).
Furthermore, in the Tyrrhenian and Algero-Provençal Basinal areas and in the northern sector of the Adria microplate the style and magnitude of the lithosphere strength is directly correlated to
the strength of the mantle and its gradients. In addition, in the Tyrrhenian and Algero-Provençal Basins, the magnitude of mantle strength is very high with respect to the crust strength, in agreement
100
R. Splendore, A.M. Marotta / Tectonophysics 603 (2013) 89–103
Fig. 12. Ratio between the crustal strength σic and the mantle strength σim, for a wet (panels a1, a2, c1 and c2) and a dry (panels b1, b2, d1 and d2) rheology, for a strain rate equal to 10−16 s−1 (panels a1, a2, b1 and b2) and 10−14 s−1 (panels c1, c2, d1 and d2), for normal faults (panels a1, b1, c1 and d1) and thrust faults (panels a2, b2, c2 and d2).
R. Splendore, A.M. Marotta / Tectonophysics 603 (2013) 89–103
with the results of Tesauro et al. (2009a). In the areas characterised by thickened crust (Alps, Dinarides, Carpathians and the Sardinia– Corsica complex) the lithosphere strength is always controlled by
101
the crust. In the central Pannonian Basin crustal and mantle contribution to the lithosphere strength are comparable. Along the Italian Peninsula, crust and mantle play the same role in controlling the
Fig. 13. Thickness and percentage of the non-competent portion of upper (panels a1, a2, b1 and b2) and lower (panels c1, c2, d1 and d2) crust, obtained for a wet (panels a1, a2, c1 and c2) and a dry (panels b1, b2, d1 and d2) rheology, assuming a strain rate equal to 10−16 s−1 and a minimum value of the yield stress equal to 1% of the lithostatic pressure.
102
R. Splendore, A.M. Marotta / Tectonophysics 603 (2013) 89–103
lithosphere strength, with the exceptions of Tuscany and the area north of the Marsili–Vavilov basin, where the strength of the mantle is negligible and the lithosphere strength is controlled by the crust. A final discussion concerns the thickness of the mechanically non-competent portions of upper and lower crust, defining the base a competent layer as the depth where the yield stress becomes less than 1–5% of the lithostatic pressure (Cloetingh and Burov, 1996). Fig. 13 shows the thickness and the percentage of the mechanically non-competent portion of the upper (panels a1, a2, b1 and b2) and lower (panels c1, c2, d1 and d2) crust, obtained for wet (panels a1, a2, c1 and c2) and dry (panels b1, b2, d1 and d2) rheology, assuming a strain rate equal to 10−16 s−1 and a minimum value of the yield stress equal to 1% of the lithostatic pressure. The primary feature that is worth highlighting is the striking difference between the western and eastern parts of the study domain. The crustal layers paving the region extending from the Tyrrhenian to the Algero-Provençal Basins result totally and unconditionally (with respect to the rheology) competent, with the sole exception of the Sardinia–Corsica complex. Here, only the 50% of the upper crust is competent for wet rheology, while for dry rheology, almost the whole upper crust is competent. For wet rheology the lower crust is totally non-competent (Fig. 13, panels c1 and c2), while for dry rheology at least the 50% of the lower crust remains competent (Fig. 13, panels d1 and d2). In the eastern part of the study domain wet rheology predicts significantly thick upper and lower crust non-competent layers, up to 70% of the total thickness for the upper crust and to 100% of the total thickness for the lower crust. Both the lateral extension and the magnitude of the thickness of the non-competent crustal layers significantly reduce when dry rheology is considered (Fig. 13, panels b1, b2, d1 and d2). The thickness of the crustal competent layers below the Italian Peninsula and the Adria microplate strongly vary from dry to wet rheology. In the first case, both the upper and the lower crusts are totally competent, while, for wet rheology, only 50% of upper crust remain competent, while up to 100% of the lower crust becomes non-competent. The central sector of the Italian peninsula shows a totally competent crustal layer. 4. Conclusions A 3D thermo-rheological model has been developed to analyse the mechanical behaviour of the crust–mantle system in the Central Mediterranean region and to investigate the occurrence of local intra-layers features. From a purely thermal point of view, our results show that the thermal field follows the major geodynamic features in the study area, with a depressed field below the suture zones, while hot temperatures characterise active extensional areas. In particular, the thickest thermal lithosphere localises below the Calabrian arc and the northern portion of the Adria microplate. A relatively thick thermal lithosphere also resides below the AlgeroProvençal Basin, compatible with thermal subsidence affecting the area in the post-rift phase, which began in the Middle Miocene (Cavazza et al., 2004). The thinnest thermal lithosphere paves the young Tyrrhenian, still actively extending, back-arc Basin (Cavazza et al., 2004; Marotta and Sabadini, 2008) and the Pannonian Basins, where extension tectonics and active volcanism are also observed. From a rheological point of view, our analysis reveals a strongly heterogeneous lithosphere strength pattern in the Central Mediterranean, characterised by steep lateral strength gradients and the occurrence of non-competent layers of significant thickness at crustal levels, located in the eastern portion of the study area. In particular, the study domain can be divided into two main parts: the western part characterised by a strong lithosphere in
which the mantle contribution is dominant and the eastern part in which a softer lithosphere is predicted and the crustal contribution to the total strength becomes more important. The sole exceptions reside in the soft Sardinia-complex and in the centre of the Pannonian Basin where a thin crust and a relatively stronger lithosphere, with respect to its boundaries, occur. The Italian Peninsula and the Adria microplate constitute a transition zone between the two main domains, exhibiting a heterogeneous rheological pattern, in which the softer Tuscany area and the region north of the Marsily–Vavilov basins and the central sector of the Adria microplate are surrounded by the harder lithosphere paving the northern sector of the Adria microplate and the Tyrrhenian Basin. The upper crust controls the total crust strength throughout the study domain when wet rheology is accounted, with the exception of small sectors of the Ionian area, Sicily and central Italy, where the contribution of the lower crust prevails. In dry rheology the upper crust controls the total crust strength only in the western portion of the study domain, from the Dinarides to the Pannonian Basin; in the rest of the study domain, with the exception of local features, the lower crust plays the main role. Non-competent crustal layers locate below the whole Pannonian area and the Sardinia–Corsica complex, that can affect up to 60% of the upper crust thickness and to 100% of the lower crust thickness. Shifting from wet to dry rheology slightly affects the general style of the rheological pattern and the mutual ratio between the contributions of the mantle and crust to the lithosphere strength. However, it does affect the extension of the non-competent domains. Acknowledgments The authors thank the editor Hans Thybo and the two anonymous reviewers for their constructive criticisms. All figures were created using GMT plotting software (Wessel and Smith 2001). References Afonso, J.C., Ranalli, G., 2004. Crustal and mantle strength in continental lithosphere: is the jelly sandwich model obsolete? Tectonophysics 394, 221–232. Artemieva, I., 2006. Global 1° × 1° thermal model TC1 for the continental lithosphere: implications for lithosphere secular evolution. Tectonophysics 416, 245–277. Artemieva, I., Mooney, W.D., 2001. Thermal thickness and evolution of Precambrian lithosphere: a global study. Journal of Geophysical Research 106 (B8), 16387–16414. Boschi, L., Ekström, G., Kustowski, B., 2004. Multiple resolution surface wave tomography: the Mediterranean basin. Geophysical Journal International 157, 293–304. Brandmayr, E., Marson, I., Romanelli, F., Panza, G.F., 2011. Lithosphere density model in Italy: no hint for slab pull. Terra Nova 23, 292–299. Cavazza, W., Roure, F., Ziegler, P.A., 2004. The Mediterranean area and the surrounding regions: active processes, remnants of former Tethyan oceans and related thrust belts. In: Cavazza, W., Roure, F., Spakman, W., Stampfli, G.M., Ziegler, P.A. (Eds.), The Transmed Atlas — The Mediterranean Region From Crust to Mantle. Springer, pp. 1–29. Chiarabba, C., De Gregori, P., Speranza, F., 2008. The southern Tyrrhenian subduction zone: deep geometry, magmatism and Plio-Pleistocene evolution. Earth and Planetary Science Letters 268, 408–423. Cloetingh, S.A.P.L., Burov, E.B., 1996. Thermomechanical structure of European continental lithosphere: constraints from rheological profiles and EET estimates. Geophysical Journal International 124, 695–723. Cloetingh, S.A.P.L., Van Wees, J.D., Ziegler, P.A., Lenkey, L., Beekman, F., Tesauro, M., Forster, A., Norden, B., Kaban, N., Hardebol, N., Bontè, D., Genter, A., GuillouFrottier, L., Ter Voorde, M., Sokoutis, D., Willingshofer, E., Cornu, T., Worum, G., 2010. Lithosphere tectonics and thermo-mechanical properties: an integrated modelling approach for enhanced geothermal system exploration in Europe. Earth Science Review 102, 159–206. Fernández, M., Marzan, I., Correia, A., Ramalho, E., 1998. Heat flow, heat production, and lithospheric thermal regime in the Iberian Peninsula. Tectonophysics 291, 29–53. Fry, L., Boschi, L., Ekström, G., Giardini, D., 2008. Europe–Mediterranean tomography: high correlation between new seismic data and independent geophysical observables. Geophysical Research Letters 35, L04301. Goes, S., Govers, R., Vacher, P., 2000. Shallow mantle temperatures under Europe from P and S wave tomography. Journal of Geophysical Research 105, 11153–11169. Huismans, R., Podladchikov, Y.Y., Cloetingh, S.A.P.L., 2001. Dynamic modeling of the transition from passive to active rifting, application to the Pannonian basin. Tectonics 20 (6), 1021–1039.
R. Splendore, A.M. Marotta / Tectonophysics 603 (2013) 89–103 Jiménez-Munt, I., Sabadini, R., Gardi, A., Bianco, G., 2003. Active deformation in the Mediterranean from Gibraltar to Anatolia inferred from numerical modeling, geodetic and seismological data. Journal of Geophysical Research 108 (B1). Koulakov, I., Kaban, M.K., Tesauro, M., Cloetingh, S.A.P.L., 2009. P- and S-velocity anomalies in the upper mantle beneath Europe from tomographic inversion of ISC data. Geophysical Journal International 179, 345–366. Marone, F., Van der Lee, S., Giardini, D., 2004. Three-dimensional upper–mantle Svelocity model for the Eurasia–Africa plate boundary region. Geophysical Journal International 158, 109–130. Marotta, A.M., Mitrovica, J.M., Sabadini, R., Milne, M., 2004. Combined effects of Tectonics and Glacial Isostatic Adjustment on intraplate deformation in Central and Northern Europe: applications to Geodetic Baseline Analysis. Journal of Geophysical Research 109, B01413-1-B01413-22. http://dx.doi.org/10.1029/2002JB002337. Marotta, A.M., Sabadini, R., 2008. Africa-Eurasia kinematics control of long-wavelength tectonic deformation in the Central Mediterranean. Geophysical Journal International 175, 742–754. http://dx.doi.org/10.1111/j.1365-246X.2008.03906.x. McNutt, M.K., Diament, M., Kogan, M.G., 1988. Variations of elastic plate thickness at continental thrust belts. Journal of Geophysical Research 93, 8825–8838. Mei, S., Suzuki, A.M., Kohlstedt, D.L., Dickson, N.A., Duram, W.B., 2010. Experimental constraints on the strength of the lithospheric mantle. Journal of Geophysical Research 115, B08204. http://dx.doi.org/10.1029/2009JB006873. Mooney, W.D., Laske, G., Masters, T.G., 1998. CRUST 5.1: a global crustal model at 5° × 5°. Journal of Geophysical Research 103, 727–747. Piromallo, C., Morelli, A., 2003. P wave tomography of the mantle under the Alpine– Mediterranean area. Journal of Geophysical Research 108 (B2). Pollack, H.N., Hurter, S., Johnson, J.R., 1993. Heat flow from the Earth's interior: analysis of the global data set. Reviews of Geophysics 31 (3), 267–280. Ranalli, G., 1994. Nonlinear flexure and equivalent mechanical thickness of the lithosphere. Tectonophysics 240, 107–114. Ranalli, G., 1995. Rheology of the Earth, second edition. Chapman & Hall (413 pp.). Ranalli, G., Murphy, D.C., 1987. Rheological stratification of the lithosphere. Tectonophysics 132, 281–295. Rybach, L., 1988. Determination of heat production rate. In: Haenel, R., Raybach, L., Stegena, L. (Eds.), Handbook of Terrestrial Heat Flow Density Determination. Kluwer Acad, Dordrecht, Netherlands.
103
Sibson, R.H., 1974. Frictional constraints on thrust, wrench and normal faults. Nature 249, 542–544. Sobolev, S.V., Zeyen, H., Granet, M., Achauer, U., Bauer, C., Werling, F., Altherr, R., Fuchs, K., 1997. Upper mantle temperatures and lithosphere–asthenosphere system beneath the French Massif Central constrained by seismic, gravity, petrologic and thermal observations. Tectonophysics 275, 143–164. Splendore, R., Marotta, A.M., Barzaghi, R., Borghi, A., Cannizzaro, L., 2010. Block model versus thermomechanical model: new insights on the present-day regional deformation in the surroundings of the Calabrian Arc. In: Spalla, M.I., Marotta, A.M., Gosso, G. (Eds.), Advances in Interpretation of Geological Processes: Refinement of Multiscale Data and Integration in Numerical Modelling: Geological Society, London, Special Publications, 332, pp. 129–147. Tesauro, M., Kaban, M.K., Cloetingh, S.A.P.L., 2008. EuCRUST-07: a new reference model for the European crust. Geophysical Research Letters 35, L05313. Tesauro, M., Kaban, M.K., Cloeting, S.A.P.L., 2009a. How rigid is Europe's lithosphere? Geophysical Research Letters 36, L16303. Tesauro, M., Kaban, M.K., Cloeting, S.A.P.L., 2009b. A new thermal and rheological model of the European lithosphere. Tectonophysics 476, 478–495. Tesauro, M., Burov, E.B., Kaban, M.K., Cloeting, S.A.P.L., 2011. Ductile crustal flow in Europe's lithosphere. Earth and Planetary Science Letters 312, 254–265. Turcotte, D.L., Schubert, G., 2002. Geodynamics. Ed. Cambridge Univ. Press, New York. Vilà, M., Fernández, M., Jiménez-Munt, I., 2010. Radiogenic heat production variability of some common lithological groups and its significance to lithospheric thermal modeling. Tectonophysics 490, 152–264. Weertman, J., Weertman, J.R., 1975. High temperature creep of rock and mantle viscosity. Annual Review of Earth and Planetary Sciences 3, 293–315. Wessel, P., Smith, W.M.F., 2001. New improved version of generic mapping tools released. EOS, Trans. Am. Geophys. Un. 79, 579. Wilks, K.R., Carter, N.L., 1990. Rheology of some continental lower crustal rocks. Tectonophysics 182, 57–77.