International Journal of Earth Sciences

, Volume 104, Issue 5, pp 1167–1183

The multistage crystallization of zircon in calc-alkaline granitoids: U–Pb age constraints on the timing of Variscan tectonic activity in SW Iberia

  • M. F. Pereira
  • M. Chichorro
  • P. Moita
  • J. F. Santos
  • A. M. R. Solá
  • I. S. Williams
  • J. B. Silva
  • R. A. Armstrong
Original Paper

DOI: 10.1007/s00531-015-1149-3

Cite this article as:
Pereira, M.F., Chichorro, M., Moita, P. et al. Int J Earth Sci (Geol Rundsch) (2015) 104: 1167. doi:10.1007/s00531-015-1149-3

Abstract

CL imaging and U–Th–Pb data for a population of zircons from two of the Évora Massif granitoids (Ossa-Morena Zone, SW Iberia) show that both calc-alkaline granitoids have zircon populations dominated by grains with cores and rims either showing or not showing differences in Th/U ratio, and having ages in the range ca. 350–335 Ma (Early Carboniferous). Multistage crystallization of zircon is revealed in two main growth stages (ca. 344–342 Ma and ca. 336–335 Ma), well represented by morphologically complex zircons with cores and rims with different ages and different Th/U ratios that can be explained by: (1) crystallization from melts with different compositions (felsic peraluminous to felsic-intermediate metaluminous; 0.001 < Th/U ratio < 0.5) and (2) transient temperature fluctuations in a system where anatectic felsic melts periodically underwent injection of more mafic magmas at higher temperatures. The two studied calc-alkaline granitoids do not include inherited zircons (pre-Carboniferous), probably because they were formed at the highest grade of metamorphism (T > 837 °C; granulite facies) and/or because they were derived from inheritance-poor felsic and mafic rocks from a previous cycle, as suggested by the internal structures of zircon cores. These Variscan magmatic rocks with crystallization ages estimated at ca. 336–335 Ma are spatially and temporally related to high-temperature metamorphism, anatexis, processes of interaction between crustal- and mantle-derived magmas and intra-orogenic extension that acted in SW Iberia during the Early Carboniferous.

Keywords

Igneous zircon U–Th–Pb SHRIMP data Crustal- and mantle-derived magmas Intra-orogenic extension Early Carboniferous Ossa-Morena Zone 

Introduction

The suture of the Rheic Ocean in SW Iberia, resulting from the collision of two continental blocks during the amalgamation of Pangea (Laurussia and Gondwana), is located at the boundary between the Ossa-Morena and South Portuguese zones (Eden 1991; Quesada et al. 1994; Díaz Azpiroz et al. 2006; Ribeiro et al. 2010; Braid et al. 2011; Pereira et al. 2012a, b). One of the most important discussions surrounding the Variscan orogen of SW Iberia concerns the tectonic setting of Early Carboniferous calc-alkaline magmatism occurring in both the Ossa-Morena Zone (Santos et al. 1990; Sánchez-Carretero et al. 1990; Casquet and Galindo 2004; Romeo et al. 2006; Pin et al. 2008; Moita et al. 2009) and the South Portuguese Zone (Dunning et al. 2002; de la Rosa et al. 2002; de la Rosa and Castro 2004; Rosa et al. 2008; Gladney et al. 2014). At the beginning of the Early Carboniferous, the Rheic Ocean had completely closed as a result of collision between the Ossa-Morena (Gondwana) and South Portuguese (Meguma) zones (Pereira et al. 2012a, and references therein).

In collisional settings where conditions favored the thickening of the crust and the building of mountain belts, several physical factors acted in conjunction with deform the crust through extensional tectonics. During intra-orogenic extensional tectonics, the lithosphere undergoes significant thinning coevally with heat input from an upwelled asthenosphere. Deformation of the ductile basement under high-temperature metamorphic conditions promotes partial melting and the formation of syn-kinematic crustal-derived magmas that interact with magmas of mantle origin. There is reason to argue that as a consequence of the Variscan oblique convergence, SW Iberia developed into a wrenching regime with the significant spread and formation of extensive calc-alkaline magmatism on both sides of the Rheic suture, inferred to have been emplaced after the Rheic Ocean had already closed, whose origin is not well understood (de la Rosa and Castro 2004; Romeo et al. 2006; Jesus et al. 2007; Pereira et al. 2007, 2009, 2012a; Moita et al. 2009, 2013; Rosa et al. 2008; Braid et al. 2010; Cambeses et al. 2014; Gladney et al. 2014).

This is a topic of great interest, which requires further investigation using zircon geochronology.

Recent studies have shown that the internal structures of the zircon of granitoid rocks studied using cathodoluminescence (CL) imaging, when combined with geochronology, geological and geochemical data, can provide useful evidence for determining the origin and evolution of granitic rocks (Koksal and Goncuoglu 2008; Jeon et al. 2012; Pereira et al. 2014). In the present study, a new set of U–Th–Pb SHRIMP zircon data is presented for two calc-alkaline granitoids of the Évora Massif (Ossa-Morena Zone) which, combined with previous geological, geochemical and geochronological findings for the Ossa-Morena Zone, are a key factors in: (1) determining the timing of textural (internal structures) and compositional (Th/U ratio) changes in zircon crystallization in calc-alkaline magmas (granitoids); (2) discussing whether or not these granitoids resulted from the interaction of different magmas; and (3) providing improved constraints for the timing of Variscan tectonic activity in SW Iberia.

Geological setting

The complex history of closer of the Rheic Ocean and terminal collision between Gondwana and Laurussia is recognizable in the western part of the Variscan European chain (Franke 2000; Matte 2001; Simancas et al. 2005; Ribeiro et al. 2010). The Rheic suture has been defined consensually, in western and central Europe, separating tectonostratigraphic zones with Gondwana or Laurussia affinity (Franke 2000; Martinez-Catalan et al. 2007; Murphy et al. 2009). In SW Iberia, the Rheic suture separates two distinct tectonostratigraphic zones (Díaz Azpiroz et al. 2006; Simancas et al. 2009; Braid et al. 2010; Ribeiro et al. 2010; Pereira et al. 2012a): (1) the Ossa-Morena-Zone and (2) the South Portuguese Zone.

In the Ossa-Morena Zone, the formation and erosion of the Cadomian magmatic arc in the Neoproterozoic (Eguiluz et al. 2000; Pereira et al. 2006, 2008, 2011, 2012c) were followed by Cambrian ensialic rifting (Chichorro et al. 2008; Sánchez-García et al. 2003, 2008, 2010). As a result, Gondwana continental margin experienced extensional tectonics that led to the opening of the Rheic Ocean through the Ordovician–Silurian and Lower Devonian (Robardet and Gutierrez-Marco 1990, 2004; Linnemann et al. 2008). Carboniferous syn-orogenic sedimentary and bimodal volcanic rocks are well preserved in the Ossa-Morena Zone, and there are some outcrops of mid-upper Devonian sedimentary rocks that may represent or not olistoliths within the Carboniferous turbidites (Pereira et al. 2006, 2012a; Machado et al. 2009).

The South Portuguese Zone includes: (1) the Pulo do Lobo Zone, which has been interpreted to be an accretionary prism (Eden 1991; Silva et al. 1990; Quesada et al. 1994; Onézime et al. 2003; Braid et al. 2010, 2011); (2) the Iberian Pyrite Belt with upper Devonian to lower Carboniferous sedimentary and volcanic rocks (Dunning et al. 2002; Rosa et al. 2008); and (3) the Carboniferous Culm series turbidites (Oliveira 1990; Oliveira and Quesada 1998; Pereira et al. 2013a).

In SW Iberia, Carboniferous deformation, metamorphism, emplacement of voluminous magmatism and development of marine basins were consequence of collision between Gondwana and Laurussia that followed the closure of the Rheic Ocean (Onézime et al. 2003; Braid et al. 2010; Pereira et al. 2012a; Gladney et al. 2014).

The Évora Massif is located in the Ossa-Morena Zone as part of Évora-Aracena-Lora del Rio metamorphic belt (Pereira et al. 2009, 2012a) (Fig. 1). The stratigraphy of the Évora Massif is dominated by Ediacaran to Ordovician sedimentary and igneous rocks (Pereira et al. 2007) (Fig. 1). The oldest rocks of the Serie Negra Group are Ediacaran pelites and graywackes deposited in a Cadomian magmatic arc setting in North Gondwana (Pereira et al. 2008). The Cambro–Ordovician sequence is related to the formation of ensialic rift basins with voluminous felsic and mafic magmatism (Chichorro et al. 2008). Cambro–Ordovician rocks are unconformably overlain by Early Carboniferous turbiditic strata and felsic volcanics (Cabrela basin), which are the youngest Paleozoic rocks of the Évora Massif (Pereira et al. 2012a, b).
Fig. 1

Inset showing major divisions of Iberia: CZ—Cantabrian Zone; WALZ—West-Asturian-Leonese Zone, GTZ—Galicia–Trás-os-Montes Zone, CIZ—Central Iberian Zone; OMZ—Ossa-Morena Zone, which includes the Coimbra–Cordoba shear zone and the Evora Massif; SPZ—South Portuguese Zone (SPZ); Schematic representation of the geology in the Évora Massif. Sample locations used for U–Pb analyses; Cross-section A1A2, showing the Late Carboniferous (Variscan) structure of the Évora Massif; (adapted from Pereira et al. 2012a, 2013a, b)

The overall structure of the Évora Massif has been interpreted as being the result of Variscan wrenching with significant extension (Pereira et al. 2009, 2012b) (Fig. 1). The Évora Massif includes a metamorphic core with a footwall comprising high-grade gneisses, migmatites and granitoids, and a few gabbros and diorites (Évora high-grade metamorphic terrains; Pereira et al. 2007, 2012b) (Fig. 1). The footwall is separated by mylonitic shear zones with sinistral sense of shear from two hanging-wall blocks of relatively low metamorphic grade, mainly composed of medium-grade gneisses, schists and amphibolites (Évora medium-grade metamorphic terrains and Montemor-o-Novo shear zone) (Fig. 1). Variscan extensional tectonics is associated with the growth of high- to medium-temperature and medium- to low-pressure mineral assemblages related to isobarically cooled P–T paths along major shear zones (Chichorro 2006; Pereira et al. 2009). The available geochronological data show that mylonitization in the Variscan shear zones took place in the period ca. 356–322 Ma (Pereira et al. 2012b). Early Carboniferous voluminous magmatism is dominated by granitoids with U–Pb zircon ages within the interval ca. 353–323 Ma (Pereira et al. 2009; Lima et al. 2011; Moita et al. 2013), but it is also represented by gabbros and diorites that occur extensively at the SW border of the Ossa-Morena Zone in the Beja massif, with U–Pb zircon ages ranging from ca. 353 Ma to ca. 318 Ma (Pin et al. 2008; Jesus et al. 2007). Mylonitic foliation and stretching lineation are locally folded by later contractional deformation with the development of chevron, open and upright folds, and slaty, fracture or crenulation cleavages (Pereira et al. 2013b). The later deformation affects not only the Tournaisian–Visean turbidites of the Cabrela basin in the Évora Massif (Chichorro 2006; Pereira et al. 2007) but also the Moscovian–Kasimovian sedimentary rocks of the Santa Susana basin (Quesada et al. 1990) that are exposed a few kilometers south of the Évora Massif.

Calc-alkaline granitoids: field relations, petrography and geochemistry

The field relations and geochemistry of different types of granitoids of the Évora Massif suggest the complex physical and chemical interaction of distinct melts with crustal, mantle and hybridized sources (Moita et al. 2009). The two types of granitoids used for geochronology in this study were sampled from the Évora high-grade metamorphic terrains (footwall) at Almansor–Montemor-o-Novo (foliated granitoid—Fg; 38°38′45″N, 8°13′47″W) and Alto de São Bento-Évora (porphyritic granitoid—Pg; 38°34′50″N, 7°56′15″W) (for location, see Fig. 1).

Fg is a monzogranite–granodiorite composed of plagioclase, quartz, K-feldspar and biotite. In a previous work (Moita et al. 2009), this rock unit was named as ‘weakly foliated granitoid’ to be distinguished from spatially associated strongly foliated diatexites. The monzogranite–granodiorite has a dominantly hypidiomorphic granular texture but some monomineralic aggregates (2–3 mm) with parallel oriented biotite grains, suggesting the preservation of metamorphic domains confers some anisotropy to this lithology. Field relationships at Almansor–Montemor-o-Novo, where Fg was sampled, show interpenetrations of different melts indicating mingling and mixing (Fig. 2). Fg occurs spatially associated with diatexites, tonalites and trondhjemites as sub-vertical meter-scale thick layers (Pereira and Lucio 2007; Moita et al. 2009). Diatexites are characterized by a S-L fabric defined by oriented biotite aggregates and schlieren and include centimeter- to meter-scale amphibolite, quartzite and metatexites enclaves (Fig. 2). Fg, with a weak planar fabric defined by oriented biotite and containing inclusions of diatexites, is often difficult to distinguish from diatexite. Fg, together with the diatexites, is locally folded and intruded by trondhjemite, tonalite and andesite dykes. The shear criteria related to the Variscan ductile deformation that folds and stretches the diatexites and granitoid veins indicate a sinistral sense of movement (Pereira et al. 2013a).
Fig. 2

a Dyke of foliated granitoid containing inclusions of diatexite (right side) intruding the compositional layering of a diatexite (left side) (Almansor–Montemor-o-Novo); b Porphyritic granitoid containing a porphyritic mafic enclave (Alto de São Bento-Évora)

Pg contains centimeter-length megacrysts of K-feldspar surrounded by a groundmass with a grain size of 2–3 mm in which euhedral to subhedral feldspar (plagioclase and K-feldspar) are dominant. At Alto de São Bento-Évora, mixing and mingling of compositionally distinct melts are less common, but locally, there are mafic microgranular enclaves within Pg (Moita et al. 2009) (Fig. 2). Pg has a well-defined magmatic foliation marked by the alignment of feldspar megacrysts. Pg occurs in the form of sub-horizontal dykes with meter-scale thickness emplaced in coarse- to medium-grained granite (Moita 2007). The host rocks of Pg are two-mica granites and tonalites. These granitoids are cut by a network of granitic pegmatite and aplite meter- to centimeter-scale wide dykes.

Major element geochemistry of Fg (monzogranite–granodiorite) indicates a weak peraluminous nature (I-type: A/CNK = 1.07–1.09), which differs to some extent from the weak peraluminous to metaluminous character of trondhjemites (I-type: A/CNK = 0.99–1.03) and considerably from the diatexites (S-type: A/CNK = 1.27–1.52), also outcropping at Almansor–Montemor-o-Novo (Moita et al. 2009). At Alto de São Bento-Évora, Pg (monzogranite–granodiorite) also has a weak peraluminous nature (I-type: A/CNK = 0.9), close to the weakly peraluminous to metaluminous nature of tonalites (I-type: A/CNK = 1.03–0.99) and distinct from the strongly peraluminous nature of two-mica leucogranites (S-type: A/CNK = 1.19–1.2), which are interpreted to be spatially and temporally associated (Moita 2007; Moita et al. 2009). Trace element patterns of both Fg and Pg show enrichment of the most incompatible elements and negative Nb and Ti anomalies consistent with a calc-alkaline signature (Moita 2007; Moita et al. 2009).

SHRIMP U–Th–Pb results

Sample preparation and analytical methods are described in a supplementary section. Representative CL imaging of zircons is shown in Fig. 3. The results of zircon U–Th–Pb analyses are listed in Table 1 (supplementary data repository) and plotted on Tera–Wasserburg diagrams in Fig. 4.
Fig. 3

CL imaging of representative zircons with analytical sites and their resulting ages indicated, of sample Fg—foliated granitoid (Almansor–Montemor-o-Novo) and Pg—porphyritic granitoide (Alto de São Bento-Évora); analysis spots and ages are listed in Table 1

Table 1

SHRIMP U–Pb zircon data of the (a) foliated granitoid (Fg—Almansor–Montemor-o-Novo) and (b) Porphyritic granitoid (Pg—Alto de São Bento-Évora)

Grain. Spot

Structure

% 206Pbc

ppm U

ppm Th

232Th/238U

ppm 206Pb*

206Pba/238U age

207Pba/206Pb age

(a)

8.1

bCL UZ-C

0.90

127

54

0.44

5.99

341.2

±4.9

273

±190

9.1

SZ-C

1.14

114

29

0.26

5.51

347.7

±5.3

327

±220

10.1

BZ-C

0.83

107

32

0.31

5.08

343.2

±4.9

245

±130

11.1

dCL-UZ-C

0.88

153

80

0.54

7.34

348.1

±4.7

247

±150

12.1

dCL-CZ-R

0.21

1,966

603

0.32

93.4

346

±12

318

±35

13.1

dCL-CZ-R

0.50

1,292

37

0.03

60.9

342.8

±3.6

323

±39

14.1

dCL-CZ-R

0.00

1,523

85

0.06

70.8

339.8

±5.9

371

±36

14.2

bCL UZ-C

1.13

80

22

0.28

3.85

347.5

±7.2

357

±160

15.1

bCL-SZ-C

1.52

87

31

0.37

4.03

333.8

±5.5

277

±230

16.1

bCL-CZ to UZ-C

0.34

144

65

0.47

6.58

331.7

±4.5

353

±120

17.1

SZ

0.97

111

41

0.38

5.18

337.1

±5.6

340

±310

18.1

dCL-UZ-R

0.51

1,582

289

0.19

72.1

331.4

±3.7

313

±52

19.1

UZ

1.15

133

35

0.28

6.32

343.2

±4.8

306

±150

20.1

UZ

0.67

172

68

0.41

8.15

343.7

±4.5

331

±100

21.1

dCL-UZ-R

0.12

1,251

208

0.17

59.4

346.2

±3.6

310

±22

21.2

BZ-C

0.22

92

29

0.33

4.23

336.5

±5

351

±110

21.3

dCL-UZ-R

0.02

4,808

31

0.01

226

342.6

±3.4

347.6

±7.5

22.1

dCL-UZ-R

0.03

1,834

26

0.01

87.5

348.3

±3.5

345

±14

23.1

dCL-UZ-R

1.28

1,294

174

0.14

61

340.1

±3.7

355

±60

24.1

SZ to CZ-C

0.64

164

40

0.25

7.67

340.1

±4.6

276

±130

25.1

SZ to CZ

0.82

114

28

0.26

5.32

339.3

±7.8

337

±330

26.1

BZ-C

0.24

90

25

0.29

4.15

337.4

±4.2

313

±67

27.1

dCL-UZ-R

0.07

1,494

168

0.12

69.1

337.8

±3.4

358

±16

28.1

dCL-UZ-R

0.05

1,966

299

0.16

93

345.4

±3.4

337

±13

29.1

CZ

0.21

112

28

0.26

5.29

343.4

±4.3

328

±51

(b)

1.1

dCL-CZtoUZ-R

0.04

1,539

589

0.40

69.3

328.9

3.6

312

16

1.2

dCL-CZtoUZ-R

0.19

1,176

347

0.30

54.1

335.6

3.5

329

35

2.1

dCL-CZtoUZ-R

0.06

1,778

567

0.33

83.1

341.2

3.7

322

31

3.1

dCL-CZtoUZ-R

1.58

2,839

831

0.30

135.9

344.0

4.1

335

55

3.2

dCL-CZtoUZ-R

0.30

3,393

1,074

0.33

169.7

363.8

3.6

298

38

4.1

dCL-CZtoUZ-R

0.10

1,471

547

0.38

69.8

345.8

3.8

302

26

5.1

dCL-CZ-R

−0.05

2,338

929

0.41

116.5

363.1

3.9

310

17

6.1

dCL-CZ-R

0.87

1,445

440

0.31

68.0

340.2

3.7

279

124

7.1

dCL-CZ-R

0.05

1,758

822

0.48

95.0

392.1

4.3

318

31

8.1

CZ Int.R

2.03

1,541

468

0.31

69.4

323.4

3.6

393

72

9.1

dCL-UZ Int.R

0.06

1,959

575

0.30

92.3

343.9

3.7

322

23

10.1

dCL-UZ Int.R

0.20

898

244

0.28

42.1

341.7

3.7

288

21

10.2

BZ to UZ-C

0.48

791

185

0.24

37.1

341.2

3.8

222

73

11.1

UZ-R

0.54

1,722

28

0.02

79.5

335.7

3.4

263

36

11.2

UZ-R

0.18

1,890

8

0.00

97.2

374.3

3.8

282

21

11.3

bC-CZ to UZ-C

1.74

223

168

0.78

10.5

337.1

4.5

198

174

12.1

dCL-CZtoUZ-R

0.39

1,517

475

0.32

70.2

337.0

3.4

355

36

13.1

dCL-CZtoUZ-R

0.07

2,576

563

0.23

121.0

343.0

3.4

305

17

14.1

dCL-CZtoUZ-R

0.91

1,529

485

0.33

72.8

344.5

3.5

313

47

15.1

dCL-CZ Int.R

0.23

971

121

0.13

45.6

342.0

3.6

354

64

16.1

dCL-UZ-R

0.44

2,321

1,002

0.45

112.0

350.9

3.5

307

38

16.2

BZ-C

0.12

861

156

0.19

40.7

344.9

3.7

315

90

17.1

dCL-CZtoUZ-R

1.99

2,545

675

0.27

114.1

321.5

3.3

344

85

18.1

dCL-CZ

0.17

2,620

477

0.19

127.4

354.4

3.6

338

31

19.1

dCL-CZtoUZ-R

−0.06

2,079

2,088

1.04

96.8

340.4

3.4

386

15

20.1

dCL-CZ Int.R

0.61

2,019

749

0.38

94.1

338.5

3.4

333

36

21.1

dCL-UZ-R

0.39

4,706

1,160

0.25

240.6

371.3

3.7

291

21

22.1

dCL-UZ-R

0.65

2,342

958

0.42

110.8

343.6

3.5

292

39

Grain. Spot

Structure

% Dis-cor-dant

207Pb*a/206Pb*

±%

207Pb*a/235U

±%

206Pb*a/238U

±%

Err corr

(a)

8.1

bCL UZ-C

−25

0.0517

8.1

0.388

8.2

0.05436

1.5

0.180

9.1

SZ-C

−6

0.053

9.8

0.405

9.9

0.05541

1.6

0.157

10.1

BZ-C

−40

0.0511

5.8

0.385

6

0.05468

1.5

0.246

11.1

dCL-UZ-C

−41

0.0511

6.4

0.391

6.6

0.05549

1.4

0.210

12.1

dCL-CZ-R

−9

0.05275

1.6

0.401

3.8

0.0552

3.5

0.913

13.1

dCL-CZ-R

−6

0.05287

1.7

0.3982

2

0.05462

1.1

0.531

14.1

dCL-CZ-R

8

0.05399

1.6

0.4029

2.4

0.05413

1.8

0.749

14.2

bCL UZ-C

3

0.0537

7

0.41

7.3

0.0554

2.1

0.294

15.1

bCL-SZ-C

−21

0.0518

10

0.38

10

0.05314

1.7

0.163

16.1

bCL-CZ to UZ-C

6

0.0536

5.2

0.39

5.4

0.05281

1.4

0.259

17.1

SZ

1

0.0533

14

0.394

14

0.05368

1.7

0.122

18.1

dCL-UZ-R

−6

0.0526

2.3

0.3828

2.5

0.05275

1.1

0.447

19.1

UZ

−12

0.0525

6.6

0.396

6.8

0.05467

1.5

0.214

20.1

UZ

−4

0.0531

4.4

0.401

4.6

0.05477

1.3

0.289

21.1

dCL-UZ-R

−12

0.05257

0.95

0.3999

1.4

0.05517

1.1

0.744

21.2

BZ-C

4

0.0535

4.7

0.395

4.9

0.05358

1.5

0.309

21.3

dCL-UZ-R

1

0.05344

0.33

0.4022

1.1

0.05459

1

0.950

22.1

dCL-UZ-R

−1

0.05338

0.63

0.4087

1.2

0.05552

1

0.853

23.1

dCL-UZ-R

4

0.0536

2.6

0.401

2.9

0.05418

1.1

0.392

24.1

SZ to CZ-C

−23

0.0518

5.8

0.387

6

0.05417

1.4

0.231

25.1

SZ to CZ

−1

0.0532

14

0.396

15

0.054

2.4

0.162

26.1

BZ-C

−8

0.0526

3

0.39

3.2

0.05373

1.3

0.399

27.1

dCL-UZ-R

6

0.05368

0.72

0.3982

1.3

0.0538

1

0.822

28.1

dCL-UZ-R

−2

0.05318

0.58

0.4036

1.2

0.05504

1

0.870

29.1

CZ

−5

0.053

2.2

0.4

2.6

0.05471

1.3

0.497

(b)

1.1

dCL-CZtoUZ-R

−5

0.0526

0.7

0.38

1.3

0.0523

1.1

0.851

1.2

dCL-CZtoUZ-R

−2

0.0530

1.6

0.39

1.9

0.0534

1.1

0.566

2.1

dCL-CZtoUZ-R

−6

0.0529

1.4

0.40

1.8

0.0543

1.1

0.624

3.1

dCL-CZtoUZ-R

−3

0.0531

2.4

0.40

2.7

0.0548

1.2

0.449

3.2

dCL-CZtoUZ-R

−22

0.0523

1.7

0.42

1.9

0.0581

1.0

0.527

4.1

dCL-CZtoUZ-R

−15

0.0524

1.1

0.40

1.6

0.0551

1.1

0.704

5.1

dCL-CZ-R

−17

0.0526

0.7

0.42

1.3

0.0580

1.1

0.829

6.1

dCL-CZ-R

−22

0.0519

5.4

0.39

5.6

0.0542

1.1

0.204

7.1

dCL-CZ-R

−23

0.0528

1.3

0.46

1.7

0.0627

1.1

0.641

8.1

CZ Int.R

18

0.0545

3.2

0.39

3.4

0.0515

1.1

0.334

9.1

dCL-UZ Int.R

−7

0.0528

1.0

0.40

1.5

0.0548

1.1

0.740

10.1

dCL-UZ Int.R

−19

0.0521

0.9

0.39

1.4

0.0544

1.1

0.764

10.2

BZ to UZ-C

−54

0.0506

3.2

0.38

3.3

0.0544

1.1

0.339

11.1

UZ-R

−28

0.0515

1.6

0.38

1.9

0.0534

1.0

0.553

11.2

UZ-R

−33

0.0519

0.9

0.43

1.4

0.0598

1.0

0.747

11.3

bC-CZ to UZ-C

−70

0.0501

7.5

0.37

7.6

0.0537

1.4

0.180

12.1

dCL-CZtoUZ-R

5

0.0536

1.6

0.40

1.9

0.0537

1.1

0.555

13.1

dCL-CZtoUZ-R

−12

0.0525

0.7

0.40

1.3

0.0546

1.0

0.816

14.1

dCL-CZtoUZ-R

−10

0.0526

2.1

0.40

2.3

0.0549

1.1

0.453

15.1

dCL-CZ Int.R

4

0.0536

2.8

0.40

3.0

0.0545

1.1

0.357

16.1

dCL-UZ-R

−14

0.0525

1.7

0.40

2.0

0.0559

1.0

0.522

16.2

BZ-C

−10

0.0527

4.0

0.40

4.1

0.0550

1.1

0.270

17.1

dCL-CZtoUZ-R

7

0.0534

3.8

0.38

3.9

0.0511

1.1

0.270

18.1

dCL-CZ

−5

0.0532

1.4

0.41

1.7

0.0565

1.0

0.598

19.1

dCL-CZtoUZ-R

12

0.0544

0.6

0.41

1.2

0.0542

1.0

0.847

20.1

dCL-CZ Int.R

−2

0.0531

1.6

0.39

1.9

0.0539

1.0

0.555

21.1

dCL-UZ-R

−28

0.0521

0.9

0.43

1.4

0.0593

1.0

0.740

22.1

dCL-UZ-R

−18

0.0521

1.7

0.39

2.0

0.0547

1.1

0.523

Errors are 1-sigma; Pbc and Pb* indicate the common and radiogenic portions, respectively

Error in Standard calibration was (a) 0.35 % and (b) 0.50 % (not included in above errors but required when comparing data from different mounts)

dCL dark (low luminescence), bCL bright (high luminescence), CZ concentric zoned, UZ unzoned, BZ banded zoned, SZ sector zoned, R rim; C core; Int.R intermediate rim

aCommon Pb corrected using measured 204Pb

Fig. 4

a, c Concordia diagrams for samples Fg and Pg; b, d weighted mean of 206Pb/238U ages for samples Fg and Pg, giving an Early Carboniferous age for magmatism

Fg—foliated granitoid (Almansor–Montemor-o-Novo)

Twenty-five U–Th–Pb isotopic analyses of twenty-three zircons are listed in Table 1 and plotted in Fig. 4. Zircons are medium- to coarse-grained (65–350 µm in diameter), most of them stubby to equant euhedral crystals. Only 3 % of zircons are needle-shaped acicular crystals (Fig. 3a). CL imaging shows that the internal pattern of morphologically complex zircon consists basically of a variable-width dark-CL rim surrounding a bright-CL core. There are cores with variable internal patterns (Fig. 3a): (1) bright-CL unzoned to weakly concentric zoned (grains 8, 18, 22 and 23); (2) banded zoned (grains 21 and 26); and (3) sector zoned (grain 15). These cores are surrounded by dark-CL external rims with high U content that are unzoned to weakly concentrically zoned (grain 14, 18, 21, 22, 23, 27 and 28; Fig. 3a). In a few grains, the interface between bright-CL cores and dark-CL rims is characterized by transgressive bulbs (grain 23), and others have possible mineral inclusions surrounded by recrystallization halos (grain 26). The population of zircons is characterized by a wide range of U (80–4,808 ppm) and Th (25–603 ppm) contents. The average Th/U value for the rims (0.12) is lower than that estimated for cores (0.34). Four rims have very low Th/U ratios (<0.06; analyses 13.1, 14.1, 21.3 and 22.1; Table 1a).

The 25 U–Th–Pb analyses are concordant within analytical uncertainty, with a very small range in 206Pb/238U (Fig. 4a). The 206Pb/238U ages obtained range from ca. 348 to ca. 331 Ma, yielding a weighted mean age of 341.4 ± 2.0 Ma (95 % CI; MSWD = 1.3; Probability = 0.18; Fig. 4b), which seems to represent the best estimate of the crystallization age of Fg. However, this range of ages can be divided into two age clusters (Fig. 4b): (1) one including the sixteen older analyses with 206Pb/238U ages, yielding a weighted mean 206Pb/238U age of 344.1 ± 2.1 Ma (95 % CI; MSWD = 0.4; Probability = 0.98) and (2) another containing the nine youngest ages, giving a weighted mean 206Pb/238U age of 335.6 ± 3.0 Ma (95 % CI; MSWD = 0.45; Probability = 0.89). These two age clusters, which are equal within analytical uncertainty, probably represent two transient stages of zircon crystallization. The average age of the cores calculated on the basis of fourteen analyses (340.5 ± 2.6 Ma; 95 % CI; MSWD = 1.03) falls within the error range, with the age obtained from eleven analyses of rims (342 ± 3.3 Ma; 95 % CI; MSWD = 1.6), suggesting that cores and rims crystallized without much time lag. The ages obtained for grain 21 are anomalous (Table 1a); the 206Pb/238U age of the core is 336.5 ± 5 Ma (analysis 21.2; Th/U = 0.33), whereas those of the rim are 346.2 ± 3.6 Ma (analysis 21.1; Th/U = 0.17) and 342.6 ± 3.4 Ma (analysis 21.3; Th/U = 0.01). This analysis, and all others with U content above 2,500 ppm, should be corrected for the U-related matrix effect of about 2 % per 1,000 ppm over 2,500 ppm (Williams and Hergt 2000) that will bring the age of analysis 21.3 back to same age within the error of the core and remove much of the scatter in the analyses from the Pg. These results may also reflect the exchange of radiogenic isotopes between the core and the rim, with the ‘younger age’ of the core indicating radiogenic Pb loss.

Pg—porphyritic granitoid (Alto de São Bento-Évora)

Twenty-eight U–Th–Pb isotopic analyses of twenty-two representative zircon grains are listed in Table 1 and plotted in Fig. 4. Most zircons are medium- to coarse-grained euhedral prisms (130–470 µm in diameter) (Fig. 3b). Stubby elongated grains predominate, but only 8 % of the zircon population consists of needle-shaped acicular crystals. The overwhelming majority of zircons show a low-CL response. Analyzed morphologically complex zircon shows cores with variable size (Fig. 3b): (1) bright-CL sub-euhedral cores with concentric zoning and wavy extinction (grain 11); (2) sub-euhedral cores with a patchy to irregular texture with linear or wavy dark and bright-CL bands (grains 1 and 15); (3) sub-rounded dark-CL and unzoned cores (grain 9); and (4) needle-shaped cores with distinct parallel banded zoning (grain 16). Some crystals show discordant zircon additions or recrystallization structures (grains 9 and 10). These cores are surrounded by texturally disconformable concentric zoned to unzoned well-developed rims (Fig. 3b): (1) thin (30 to 50 µm) dark-CL unzoned rims (grain 11) developed on pyramidal crystal faceswith very low Th/U ratios (0.02–0.004) and (2) variable thickness of rims (80–150 µm) with weakly concentric zoning to almost invisible concentric zoning (grain 19). Apart from one rim with Th/U < 0.02 (analyses 11.1 and 11.2), a core with Th/U = 0.78 (analysis 11.3) and a rim with concentric zoning and Th/U = 1.04 (analysis 19.1), the other 24 analyses reveal a wide range of U (791–4,706 ppm) and Th (121–1,160 ppm) contents, and a uniform Th/U ratio ranging from 0.13 to 0.48 (with an average value of 0.31) (Table 1b). Dark-CL and rounded unzoned cores are often separated from more highly developed external rims by a thin, diffuse bright-CL internal rim (grain 9) (Fig. 3b).

The twenty-eight U–Th–Pb isotopic compositions obtained are distributed along a discordia line with a lower intercept at 330.7 ± 8.6 Ma (MSWD = 1.3) (Fig. 4c). This set of ages with significant scattering yielded a weighted mean 206Pb/238U age of 345.0 ± 5.7 Ma (95 % CI; MSWD = 16; Probability < 0.001; Fig. 4d). The oldest and youngest ‘ages’ suggest radiogenic Pb gain or loss from a primary isotopic composition close to zircon crystallization age. It could be argued that some zircon growths (with U content above 2,500 ppm) were display an example of U-dependent Pb/U bias in SHRIMP analyses (Williams and Hergt 2000). The seven ‘oldest ages’ are probably associated with Pb gain (analyses 3.2, 5.1, 7.1, 11.2, 16.1, 18.1 and 21.1), while the three ‘youngest ages’ may indicate partial radiogenic Pb loss (analyses 1.1, 8.1 and 17.1) (Fig. 5d; Table 1b). Apart from these ten analyses, the best crystallization age of Pg is estimated, more conservatively, using the remaining eighteen analyses, giving a weighted mean 206Pb/238U age of 341.1 ± 1.7 Ma (95 % CI; MSWD = 0.79; Fig. 4d). On the basis of these eighteen 206Pb/238U ages, two age clusters (Fig. 4d) were defined: (1) one comprising the thirteen oldest ages, giving a weighted mean 206Pb/238U age of 342.8 ± 2.0 Ma (95 % CI; MSWD = 0.24; Probability = 0.99) and (2) another comprising the five youngest ages (analyses 1.2, 12.1, 20.1 and 11.1) with a weighted mean 206Pb/238U age of 336.7 ± 3.2 Ma (95 % CI; MSWD = 0.115; Probability = 0.98).
Fig. 5

Variation of the Th/U ration with time and having in regard the fields of distinct source compositions that reflect different sources of melt from which zircon crystallized

Discussion

Internal structures and Th/U ratio changes as indicators of multistage zircon crystallization

The CL images obtained in this study show that morphologically complex zircons are dominant in Fg and Pg. The refractory nature of zircon and the resultant difficulty in full recycling by dissolution led to the preservation of older components (cores) surrounded by relatively young overgrowths (rims). Fg composite zircons show variable size bright-CL sub-euhedral and sub-rounded cores surrounded by texturally disconformable variable-width dark-CL concentric zoned to unzoned rims. Pg composite zircons are characterized by sub-euhedral and sub-rounded dark-CL cores with smoothly rounded corrosion surfaces surrounded by very thin bright-CL rims (too thin to be analyzed using SHRIMP). These very thin bright-CL rims together with the more developed external rims observed in the two samples suggest transient periods of thermally activated solid-state recrystallization that may have occurred after corrosion or resorption during the multistage evolution of a zircon crystal (Corfu et al. 2003). Alternating growth and corrosion of zircon can be caused by transient heating of resident felsic magma by the influx of hot mafic magma, which increases the temperature and causes transient zircon undersaturation by changing the melt composition during relatively short periods of time (Miller and Wooden 1994). These rapid compositional changes may be associated with magma mixing (Castro et al. 1991; Hoskin and Schaltegger 2003), which may be the cause of the simultaneous growth of: (1) many cores with euhedral concentric zoning commonly found in zircon precipitated from felsic to intermediate granitic magmas and (2) a few other cores which present banded zoning typical of dioritic magmas or rapidly crystallized zircon, sector zoning found in zircon from more highly mafic magmas, or granulites and unzoned zircons found in dacites (Corfu et al. 2003).

In Fig. 5, most zircon analyses of Fg and Pg (in morphologically simple grains and in cores and rims of morphologically complex grains) are plotted in the short interval of 0.25 < Th/U average ratio < 0.31, indicating an igneous origin (Heaman et al. 1990; Hanchar and Miller 1993). Most cores and simple zircons fall within the range 0.2 < Th/U < 0.5, suggesting crystallization from a relatively chemically homogeneous source. Th/U ratios <0.5 are typical of igneous zircon (Hoskin and Schaltegger 2003) consistent with precipitation from a melt phase of a felsic-intermediate metaluminous composition, close to the field of felsic peraluminous sources, where a number of rims are plotted (Th/U < 0.1). There are few rims with very low Th/U < 0.06, which is commonly found in zircons crystallized during the partial melting of peraluminous rocks, usually metasedimentary rocks, in the presence of a mineral with high Th/U, for example, monazite (Williams and Claesson 1987; Williams 2001).

In Fig. 5a, each zircon crystal has a different distribution of U and Th contents. Pg has a higher Th content in relation to U content and thus higher Th/U than Fg. In both granitoids, the Th/U ratio decreases from the cores to the rims, reflecting a more important increase in U than in Th content. These variations in the U and Th contents of zircons do not necessarily indicate two different stages of crystallization separated in time. There are also cores and rims of the same age with different Th/U ratios. These Th/U changes in zircon from core to rim can be attributed to different physical and chemical conditions during crystallization. In boundary conditions between metamorphic and magmatic systems, partial melting, magma mixing/mingling and zircon growth versus dissolution may occur during complex histories of interaction between melts with distinct compositions. The Th/U differences between cores and rims found in each sample may reflect variations in growth rates relative to diffusion-controlled dissolution rates affected by rapid variations in temperature and in Zr saturation of the melt (Wang et al. 2011). At relatively high-temperatures, melt/zircon diffusion rate is higher (Miller et al. 2003). Accordingly, U and Th with partition coefficients lower than Zr tend to be excluded from the crystal lattice of zircon. Moreover, as temperature decreases and Zr becomes oversaturated, there may be a progressive concentration of U and Th at the melt–zircon interface, creating suitable conditions for the growth of rims enriched in U and Th. These transient temperature fluctuations required for explaining a model of multistage recrystallization of zircon are likely to occur in a system dominated by anatexis and the voluminous formation of felsic magmas that was periodically injected with magmas at a higher temperature (Miller and Wooden 1994), like intermediate to mafic magmas. In addition, the influence of variations in the chemical conditions of the melt is also essential for explaining differences in the cores and rims of morphologically complex zircon that crystallized as a result of the interaction between magmatic and metamorphic systems. Comparing an apparent systematic distinction between igneous zircon and metamorphic-melt zircon based on the Th/U ratio, which is very low for the latter (Th/U < 0.07; Rubatto 2002), it was additionally found that low Th/U can also indicate competition for Th with high Th minerals (monazite and allanite), whereas high Th/U may indicate the breakdown of minerals with high Th/U, or competition with high U minerals (xenotime) (Moller et al. 2003).

U–Pb age constraints on the timing of Variscan tectonics in SW Iberia

Taking into account the complete range of U–Pb data (of morphologically simple grains and of cores and rims of morphologically complex grains), it should be noted that the reported weighted mean 206Pb/238U ages of ca. 345 Ma (Pg, n = 28) and ca. 341 Ma (Fg, n = 25) are slightly different from, or match, the crystallization age of ca. 341 Ma estimated on the basis of the two main clusters of each sample. The coherent age of the two age clusters of Pg was estimated after excluding the most discordant ages. The ‘youngest and oldest ages’ are distributed along a discordia line, which intercepts at ca. 330 Ma, approaching the crystallization age of ca. 328–323 Ma granitoids from the Pavia pluton (Lima et al. 2012), suggesting that these discordant ages are probably related to disturbances in the U–Th–Pb systems caused by the emplacement of later intrusions. Even if discordant ages are discounted, both samples show a degree of scattering revealed by the dispersed distribution of individual ages obtained from different zircons and/or zircon sub-domains in composite grains.

The U–Pb data obtained from composite zircons of Fg and Pg indicate that most of the cores and rims are very similar in age, ranging from ca. 350 Ma to ca. 335 Ma (Early Carboniferous), a relatively short time interval, approximately 15 million years (Fig. 5b).

Considering the set of analyses with no isotopic disturbance, there are cores which have the same 206Pb/238U age within the analytical error of their respective rims and other individual cores or rims. Taken together, the evidence presents indicates a complex magmatic evolution centered on two main stages of zircon growth with overlapping ages in both samples: (1) an earlier stage at 342.8 ± 2 Ma (Pg) and 344.1 ± 2.1 Ma (Fg) and (2) a later stage at 336.7 ± 3.2 Ma (Pg) and 335.6 ± 3 Ma (Fg), approximately 6–10 million years later. Figure 5b shows that morphologically simple zircons and cores or rims of morphologically complex zircon were able to crystallize at the same time and that this stage of crystallization is repeated later. Pg has a simple zircon that was incorporated from a previous crystallization stage (ca. 354 Ma), and Fg contains simple zircons that crystallized directly from the accumulating magma associated with the earlier and later stages of crystallization (ca. 343–337 Ma). Fg includes morphological complex zircons with core and rim ages in the range ca. 348–331 Ma, and Pg shows core ages in the interval ca. 344–335 Ma and rim ages range from ca. 350 Ma to ca. 335 Ma. The age distribution of morphological simple and complex grains suggests multiple stages of zircon growth and corrosion, which could have lasted ca. 10 Ma, with intervals recorded for the same complex grain from ca. 2 Ma (Pg, grain 11) to 8 Ma (Fg, grain 14) (Fig. 5b).

A salient feature of the U–Pb data obtained is the absence of older inherited zircons in Fg and Pg. The absence of inherited zircons contrasts with in the results obtained for the Arraiolos biotitic granite (Évora Massif), which resulted from the anatexis of inherited zircon-rich Ediacaran pelites and graywackes (Pereira et al. 2008, 2009). One possible explanation for the absence of inherited zircons in Fg and Pg could be that the melts from which they crystallized resulted also from the anatexis of inherited zircon-rich Ediacaran pelites and graywackes, but with an initial magma temperature at the source higher than 837 °C (granulite facies). This high temperature is required in order to explain the almost complete dissolution of inherited zircon (‘hot granite’; Miller et al. 2003). However, their weakly peraluminous compositions, as well as their 87Sr/86Sri (0.707184; 0.707321) and εNdi (−3.9; −5.5), and trace element patterns show that they are not simply products of the melting of metapelitic rocks (Moita et al. 2009). Another explanation could be that the melt was the product of anatexis of zircon-poor felsic igneous rocks (Valverde and Alcáçovas felsic orthogneisses; Chichorro et al. 2008) and MORB-like amphibolites (Pereira et al. 2007). A more likely scenario is that Fg and Pg resulted from the chemical interaction of products of anatexis with products of fractional crystallization from primitive mafic magmas, both with minimal inheritance.

In this study, the assumption was made that mean ages of the youngest age clusters represent the best estimate for igneous crystallization: ca. 336 Ma (Pg) and ca. 335 Ma (Fg). This clarifies the following points: (1) the ca. 335 Ma age obtained for Fg constrains the timing of shearing that is coeval with the cooling of granitic magma from which Fg crystallizes (Pereira et al. 2013a, b); thus, the age of shearing observed in migmatites and granitoids at Almansor–Montemor-o-Novo matches the age of shearing that was dated in the detachment (Boa Fé shear zone) that separates Évora high-grade metamorphic terrains (footwall) from the Montemor-o-Novo shear zone (hanging hall). In the Boa Fé shear zone, mylonitic paragneisses were dated at ca. 341 Ma (U–Pb; zircon) and ca. 337 Ma (Ar–Ar; biotite) (Pereira et al. 2012a); (2) the ca. 336 Ma age obtained for Pg reveals that undeformed granitoids and foliated granitoids in the Évora Massif may have the same ages, as already suggested by Moita et al. (2009), which means that granitoids emplaced during Variscan (‘synorogenic’) tectonic events may show in some circumstances, isotropic textures. Similar field relationships with ca. 346.345 Ma unfoliated and foliated granites were described by Gladney et al. (2014) in the Sierra Norte Batholith (South Portuguese Zone). Therefore, the use of presence versus absence of deformational features as the only criterion to establish a relative age between granitoids may lead to erroneous interpretations, such as the classification of unfoliated granitoids as ‘late-orogenic’, as is often the case in the Geological Map of Portugal, scale 1:500,000; (3) the ca. 336–335 Ma ages obtained for Fg and Pg coincide with the age interval obtained for the calc-alkaline intermediate granitoids of the Évora Massif (ca. 337–336 Ma) for a tonalite and mafic microgranular enclave of the Hospitais pluton (Moita et al. 2013), and ca. 335 Ma for a quartz-dioritic microgranular enclave hosted in the ca. 325 Ma granodiorite of the Pavia pluton (Lima et al. 2012), and other spatially related calc-alkaline granitoids (ca. 338–337 Ma for the Reguengos de Monsaraz granitoids; Antunes et al. 2011). Thus, the coexistence in time and space of magmas of different compositions in the Évora Massif is corroborated.

All zircon ages of Fg and Pg in the range ca. 350–335 Ma match plutonism, high-grade metamorphism, thermal relaxation and rapid exhumation ages interpreted to be caused by wrenching in the Évora Massif and south of it (Pereira et al. 2009, 2012a) (Fig. 6): (1) the crystallization ages of the Beja gabbros and diorites (Pin et al. 2008; Azor et al. 2008); (2) the ages of metamorphism estimated in the range ca. 342–336 Ma of the Ventosa and São Brissos amphibolites and the Alcáçovas gneiss (Dallmeyer et al. 1993; Pereira et al. 2009); and 3) the cooling ages of ca. 339–337 Ma of the Beja gabbros (amphibole, Ar–Ar; Dallmeyer et al. 1993). Figure 6 shows the distribution of U–Pb zircon ages of calc-alkaline magmatism and metamorphism in SW Iberia. In the Ossa-Morena Zone, calc-alkaline magmas were emplaced at different crustal levels, in association with migmatites and gneisses of the same age in both the Coimbra–Cordoba shear zone and the Évora-Aracena-Lora del Rio metamorphic belt. The distribution of ages of calc-alkaline magmatism reveals that older ages (ca. 353–347 Ma) are located further south (in present day coordinates) than younger ages (ca. 340–335 Ma), suggesting a northward migration of magmatism. This alleged migration is also found with the volcanism of the South Portuguese Zone, over a previous time interval (ca. 374–349 Ma; Rosa et al. 2008), but whether the two are related or not has yet to be explored. It has been proposed that the calc-alkaline magmatism of the two zones is associated with the rollback of the Rheic oceanic lithosphere (Pin et al. 2008). An alternative model suggests that the emplacement of ultramafic and mafic sills into the mid-crust under both zones was the main cause for the voluminous calc-alkaline magmatism (the IBERSEIS Iberian Reflective Body, IBR; Simancas et al. 2003).
Fig. 6

Schematic geological map of the Ossa-Morena Zone, which includes the Coimbra–Cordoba shear zone and the Evora–Aracena-Lora del Río metamorphic belt with high-grade metamorphic rocks, showing the location of Tournaisian–Visean granitoids and migmatites (adapted from geological map of the Iberian Peninsula, Balearic and Canary Islands, Romeo et al. 2006; Pereira et al. 2009, 2012a; Cambeses et al. 2014; Gladney et al. 2014)

In the Early Carboniferous (Fig. 7), following the collision between Laurussia and Gondwana, the steps leading to the emplacement of voluminous calc-alkaline felsic and mafic magmatism (ca. 353–335 Ma) in SW Iberia may be summarized as follows: (1) the upwelling of the asthenosphere was probably responsible for the decompressional melting of the lithospheric mantle, which had already been metasomatized by the subducted slab, and this most likely led to the generation of the mafic parental magmas of the Early Carboniferous calc-alkaline suites and (2) the underplating of mantle-derived magmas and the intra-orogenic extension created the right conditions for the partial melting of crustal materials.
Fig. 7

Schematic cross-section (not to scale) illustrating the structure of the basement of SW Iberia and relationship with basins, volcanism and plutonism during the Early Carboniferous (adapted from Pereira et al. 2009, 2012a)

Conclusions

The main conclusions of this study are the following:
  1. 1.

    Both calc-alkaline granitoids of the Évora Massif have zircon populations dominated by morphologically complex grains with cores and rims presenting ages in the range ca. 350–335 Ma (Early Carboniferous); two main zircon growth stages can be defined, separated by 6–10 million years (ca. 344–342 Ma and ca. 336–335 Ma), testified by the development of single grains showing cores and rims with distinct ages, which either present or do not present different Th/U ratios, indicating the multistage crystallization of zircon.

     
  2. 2.

    A salient feature of both calc-alkaline granitoids of the Évora Massif is the fact that they do not include inherited zircons (pre-Carboniferous), probably because they were formed at the highest grade of metamorphism (T > 837 °C; granulite facies) and/or because they derived from inheritance-poor felsic anatectic crustal melts or mafic magmas.

     
  3. 3.

    Zircon from the two studied calc-alkaline granitoids of the Évora Massif have cores with internal structures suggesting crystallization from melts with different compositions (felsic peraluminous to felsic-intermediate metaluminous; 0.001 < Th/U ratio < 0.5) and/or under different temperature conditions (granulite facies); they include composite zircons with cores and rims with different ages and different Th/U ratios, indicating an increase in Th and U contents over time from the core to the rim that can be explained by transient temperature fluctuations in a system where anatectic felsic magmas periodically underwent injection of more mafic magmas at a higher temperature.

     
  4. 4.

    This study of zircons provides a contribution toward our understanding of the field relations, petrography and geochemistry of the calc-alkaline granitoids of the Évora Massif, which together indicate that they were formed as a result of the interaction of magmas of different compositions (mixing/mingling) under conditions at the interface of metamorphic and magmatic systems.

     
  5. 5.

    The most probable crystallization ages for both calc-alkaline granitoids of the Évora Massif were estimated at ca. 336–335 Ma based on their youngest age clusters; therefore, they are classified as Variscan magmatic rocks and are spatially and temporally related to high-temperature metamorphism, anatexis, processes of interaction between crustal and mantle-derived magmas, and extensional tectonics that acted in SW Iberia during the Early Carboniferous (Fig. 7).

     

Acknowledgments

G. Gutierrez-Alonso and J. B. Murphy are gratefully acknowledged for detailed reviews of the manuscript. This paper is a contribution to research projects: GOLD-PTDC/GEO-GEO/2446/2012; FCOMP-01-0124-FEDER-029192 (Portugal) and IGCP 597—Amalgamation and Breakup of Pangaea: the type example of the supercontinent cycle (UNESCO-IUGS).

Copyright information

© Springer-Verlag Berlin Heidelberg 2015

Authors and Affiliations

  • M. F. Pereira
    • 1
  • M. Chichorro
    • 2
  • P. Moita
    • 3
  • J. F. Santos
    • 4
  • A. M. R. Solá
    • 5
  • I. S. Williams
    • 6
  • J. B. Silva
    • 7
  • R. A. Armstrong
    • 6
  1. 1.IDL/Departamento de Geociências, ECTUniversidade de ÉvoraÉvoraPortugal
  2. 2.CiCEGE/Departamento de Ciências da TerraUniversidade Nova de LisboaLisbonPortugal
  3. 3.Centro de Geofísica de Évora/Departamento de Geociências, ECTUniversidade de ÉvoraÉvoraPortugal
  4. 4.Geobiotec, Departamento de Geociências daUniversidade de AveiroAveiroPortugal
  5. 5.LNEG, Unidade de Geologia e Cartografia GeológicaPortoPortugal
  6. 6.Research School of Earth SciencesThe Australian National UniversityActonAustralia
  7. 7.IDL/Departamento de Geologia, Faculdade de CiênciasUniversidade de LisboaLisbonPortugal

Personalised recommendations