Cretaceous Research 30 (2009) 575–586

Contents lists available at ScienceDirect

Cretaceous Research

journal homepage: www.elsevier.com/locate/CretRes

Age constraints on the Late Cretaceous alkaline magmatism on the West Iberian Margin

Rui Miranda a,b,d,*, Vasco Valadares c,a, Pedro Terrinha c,b, Joa˜o Mata a,d, Maria do Rosa´rio Azevedo e, Miguel Gaspar a,f, Jose´ Carlos Kullberg g, Carlos Ribeiro h a Fac. de Cieˆncias da Univ. de Lisboa, Depto. Geologia , Campo Grande, 1749-016 Lisboa, Portugal b LATTEX, IDL, Univ. de Lisboa, Campo Grande, 1749-016 Lisboa, Portugal c INETI, Depto. Geologia Marinha, Estrada da Azambuja, 2720-866 Amadora, Portugal d Centro de Geologia da Univ. de Lisboa, Campo Grande, 1749-016 Lisboa, Portugal e Depto. Geocieˆncias da Univ. de Aveiro, Santiago Campus, 3810-003 Aveiro, Portugal f CREMINER, Campo Grande, 1749-016 Lisboa, Portugal g Depto. Cieˆncias Terra, Fac. Cieˆncias Tecnologia., Univ. Nova de Lisboa and CICEGe Quinta da Torre, 2829-516 Caparica, Portugal h Depto. Geocieˆncias Univ. E´vora and Centro de Geofı´sica de E´vora, Rua Roma˜o Ramalho, 59, 7000 E´vora, Portugal article info abstract

Article history: The onshore sector of the West Iberian Margin (WIM) was the locus of several cycles of magmatic activity Received 21 November 2007 during the Mesozoic, the most voluminous of which was of alkaline nature and occurred between 70 and Accepted in revised form 100 Ma. This cycle took place in a post-rift environment, during the 35 counter-clockwise rotation of 13 November 2008 Iberia and initiation of the alpine compression. It includes the subvolcanic complexes of Sintra, Sines, and Available online 30 November 2008 Monchique, the volcanic complex of Lisbon and several other minor intrusions, covering an area of approximately 325 km2. Previous cycles were tholeiitic and transitional in nature, occuring around Keywords: 200 Ma and 130–135 Ma, respectively. Geochronology 40 39 Alkaline magmatism New LA-ICP-MS U-Pb, Ar/ Ar, K-Ar and Rb-Sr ages on several intrusions distributed along the onshore West Iberian Margin WIM are presented, which combined with previously published data allows us to constrain the duration Late Cretaceous of the Late Cretaceous alkaline cycle to circa 22 Ma (94–72 Ma) and define two pulses of magmatic activity. The first one (94–88 Ma) occurred during the opening of the Bay of Biscay and consequent rotation of Iberia and clusters above N38200. The second pulse (75–72 Ma) has a wider geographical distribution, from N37 to N39. This final pulse occurred during the initial stages of the Alpine orogeny in Iberia that led to the formation of the Pyrenees and Betics and to tectonic inversion of the Mesozoic basins. Isotope and trace element geochemistry point to a sublithospheric source for the alkaline magmatism that clearly distinguishes it from the previous cycles which had an important lithospheric mantle component. Also, it allows the discrimination between the two different alkaline pulses in terms of trace element abundance and residual mantle minerology. It is speculated that these differences might be the result of distinct magma ascent rates due to either more or less favourable tectonic settings that avoided or allowed the interaction with metasomatized lithosphere and equilibration with K rich minerals like amphibole and/or phlogopite. Ó 2008 Elsevier Ltd. All rights reserved.

1. Introduction 130–135 Ma (Ferreira and Macedo, 1979), respectively. The last cycle was the most voluminous, shows an alkaline nature, and took The onshore sector of the West Iberian Margin (WIM) was the place between 70 and 100 Ma (Ferreira and Macedo, 1979). locus of several cycles of magmatic activity during the Mesozoic. This cycle includes the NNW-SSE aligned subvolcanic complexes Occurrences related to the ﬁrst two cycles display tholeiitic and of Sintra, Sines, and Monchique, the volcanic complex of Lisbon and transitional afﬁnities (Martins, 1991; Martins et al., 2008) and ages several other minor intrusions (Fig. 1). These rocks are discontin- around 200 Ma (e.g. Dunn et al., 1998; Verati et al., 2007) and uously exposed from parallels 39 Nto37 N and cover an area of approximately 325 km2 (Fig. 1). With the exception of the Mon- chique alkaline complex, all the alkaline rocks were emplaced * Corresponding author. within the Mesozoic Lusitanian and Algarve rift basins, developed E-mail address: [email protected] (R. Miranda). in relation to the opening of the Atlantic.

Fig. 1. The Late Cretaceous alkaline magmatism in the onshore sector of the West Iberian Margin. Circles indicate all of the sampled intrusions. Portuguese Geology and batimetry adapted from data available at http://www.iambiente.pt/atlas/est/index.jsp. Spanish geology adapted from the 1:1000000 geological map of the Iberian peninsula (Alvaro et al., 1994). Batimetry contours every 100 m. Geographic coordinates, WGS 84.

The Cretaceous alkaline rocks were grouped along with other Ribeiro et al. (1979) suggested the massifs were emplaced along smaller intrusions from the Pyrenees in the Late Cretaceous Alka- a dextral strike-slip fault during the rotation of Iberia and Mouge- line Igneous Province of Iberia (Rock, 1982). Despite their older age, not (1980) proposed that the massifs intruded along crustal pull- Lustrino and Wilson (2007) included these occurrences in the apart strain domains between en echelon faults during tectonic Circum-Mediterranean Cenozoic Anorogenic Igneous Province. inversion of the WIM. Terrinha (1998) and Kullberg (2000) showed During the Late Cretaceous, the offshore sector of the WIM and the existence of Early-Middle Jurassic age syn-rift faults along the adjacent oceanic crust were also the site of important alkaline lineament and proposed that these faults controlled the magma magmatic activity that produced a series of seamounts like the ascent in the Late Cretaceous. Ormonde peak of the Gorringe Bank (Auzende et al., 1978; Cornen, The main goal of this work is to better constrain the chronology 1982; Fe´raud et al., 1982; Bernard-Grifﬁths et al., 1997) and the of emplacement of the onshore Late Cretaceous intrusions of the Northern section of the Madeira-Tore rise (Geldmacher et al., 2006; WIM by presenting new LA-ICP-MS U-Pb, 40Ar/39Ar, K-Ar, and Rb-Sr Merle et al., 2006). ages. These data, combined with new geochemical analyses and The alignment of the Sintra, Sines, and Monchique Late Creta- other geological information, bring new insights on the tectono- ceous alkaline massifs has been a matter of debate in the past. magmatic processes of the WIM during the Late Cretaceous. R. Miranda et al. / Cretaceous Research 30 (2009) 575–586 577

2. Geological setting 3. The Late Cretaceous alkaline cycle in the WIM: previous geochronological work In the Mesozoic Lusitanian and Algarve Basins, the last phases of rifting took place during the late Early Cretaceous (Rasmussen et al., The Late Cretaceous alkaline igneous activity in the onshore 1998; Terrinha, 1998) while the initiation of seafloor spreading in section of the WIM took place in a post rift setting, 30 My after the the Tagus an Iberian abyssal plains started at 133 Ma–128 Ma beginning of oceanization in the Tagus Abyssal Plain and, partially (Pinheiro et al., 1996; Russell and Whitmarsh, 2003)(Fig. 2). The contemporaneous with the Pyreneean continental collision in tectonic inversion of these basins was a consequence of the Northern Iberia and the onset of tectonic inversion on the Mesozoic contemporaneous onset of rapid collision and even subduction basins (Fig. 2), thus making the Late Cretaceous alkaline magma- between the Iberian, European, and African plates. This inversion tism a key episode in the history of this passive margin. These rocks started in the Late Cretaceous (Mougenot, 1980; Terrinha, 1998; have been object of several previous dating attempts, summarized Rosembaum et al., 2002) but reached its climax during the Miocene in Table 1. in the Lusitanian Basin and offshore the WIM (Ribeiro et al., 1990; As shown in Table 1, the chronology of the main magmatic Kullberg et al., 2000; Neves et al., in press). However, the Algarve events associated with the Late Cretaceous alkaline cycle has been Basin experienced the compressive effects of the Africa-Iberia established on the basis of a large number of K-Ar ages and Rb-Sr convergence mainly during the latest Cretaceous through Oligo- whole rock isochrons. These two isotopic systems are extremely cene-early Miocene times (Terrinha, 1998). sensitive to post-emplacement disturbance which may explain, at The opening of the Bay of Biscay and consequent rotation of least to some extent, the significant scatter of the published Iberia started between magnetic anomalies M3 and M0 (130 and geochronological data (sometimes exceding 15 Ma for the same 118 Ma, respectively) and ended at magnetic chron A33o (80 Ma) occurrence using one single method, Table 1). The use of more (Sibuet et al., 2004)(Fig. 2). Based on paleomagnetic data, this reliable dating techniques (e.g. incremental release 40Ar/39Ar dating counter-clockwise rotation of Iberia is estimated at about 35 (e.g. and/or U-Pb zircon geochronology) is therefore fundamental to Storetvedt et al., 1987; Galdeano et al., 1989; Moreau et al., 1997; obtain precise age constraints. In order to refine the geochronology Juarez et al., 1998; Ma´rton et al., 2004). It created a sinistral of the Late Cretaceous alkaline cycle, and if possible, detect any transtensive regime along the North Pirenaic Fault, synchronous geochemical variation with age and/or location, several of the with high temperature/low pressure metamorphism, formation of exposed occurrences were selected for further geochronological new sedimentary basins and magmatic activity. This tectono- studies and/or whole-rock geochemistry. These include: magmatic event culminated with the emplacement of mantle lherzolite bodies and minor alkaline intrusions between 105 and (A) The layered Paço d’Ilhas sill, comprising monzogabroic, mon- 85 Ma, in the Pyrenees (Fabrie`s et al., 1998; Montigny et al., 1986) zonitic, and rarer syenitic beds (Mahmoudi, 1991) that intruded and 82 and 69 Ma, in the Catalan Coastal Ranges (Sole´ et al., Early Cretaceous sediments (Barremian-Albian, Zbyszewski 2003). et al., 1955) and was, prior to this work, the oldest known In the northern Bay of Biscay - Cantabrian region, subduction alkaline intrusion in the WIM. It had been previously dated by took place between the Campanian and the Miocene (Sibuet et al., the K-Ar method using biotite and K-feldspar separates, with 2004), while in the Pyrenees, subduction ensued in two different the obtained ages corresponding to 88 2.7 and 86.8 2.5 Ma, phases, from the Aptian to the Santonian in the south and from the respectively (Mahmoudi, 1991). Campanian to the Miocene, in the north (Sibuet et al., 2004). (B) The granite from the Sintra laccolith body (Terrinha et al., Around the same time, Central Iberia was the locus of an 2003.) that was emplaced into an Upper Jurassic carbonate important thermal event, resulting in low grade thermal meta- sequence consequently forming a hornfels rim that surrounds morphism in the Cameros basin between 105 and 86 Ma (Casquet the intrusion. The granite had systematically yielded ages older et al., 1992) and in the formation of Hg hydrothermal deposits at than the suite of gabbro-diorite-syenite and associated breccias 85 Ma (Tritlla and Sole´, 1999). present in the complex (see Table 1).

Fig. 2. Summary of the most important geodynamic events in Iberia during the Cretaceous. Dates for the limits of the epochs and ages of the Cretaceous period from Gradstein et al. (2004). Ages from geochronological data in full lines. Geologically interpreted ages in dashed lines. 578 R. Miranda et al. / Cretaceous Research 30 (2009) 575–586

Table 1 (F) The Monchique alkaline complex is formed by small gabbro Age summary for the main Late Cretaceous alkaline occurrences in the West Iberian outcrops within two larger concentric nepheline syenite Margin. 1 – Ferreira and Macedo, 1979;2–MacIntyre and Berger, 1982;3–Store- bodies, with associated lamprophyre dykes and breccias (Rock, tvedt et al., 1987;4–Canilho and Abranches, 1982;5–Bernard-Griffiths et al., 1997; 6–Rock, 1976;7–Mahmoudi, 1991 1978; Gonzalez-Clavijo and Valadares, 2003). It intrudes the Carboniferous shales and quartzitic sandstones of the Brejeira Occurrence Age Reference Method Material Formation (Middle Namurian–Early Westphallian, Oliveira, Lisbon volcanic 72.6 3.1 1 K-Ar whole rock 1990) producing a 200 m thick contact metamorphism aureole. complex isochron Sintra (granite) 81.9 2 2 K-Ar biotite This is the most voluminous of the southern alkaline intrusions 84.0 1.1 3 K-Ar k feldspar and the only one intruding exclusively rocks of Paleozoic age. 79.2 0.8 this study U-Pb zircon The Monchique complex has been dated with both the K-Ar Sintra (gabbro) 74.9 1 3 K-Ar whole rock and the Rb-Sr method (table 1), with the obtained ages clus- (syenite) 76.1 1.1 3 K-Ar whole rock tering around 72 Ma (72.0 2.0 mean K-Ar ages on mineral 76.4 1.4 3 K-Ar whole rock 78.3 1.8 3 K-Ar whole rock separates, MacIntyre & Berger, 1982; 72.0 Ma Rb-Sr whole rock Sines 72.0 1.5 4 Rb-Sr whole rock isochron, Rock, 1976; Bernard-Griffiths et al., 1997), although isochron the K-Ar ages obtained for K-feldspar and nepheline separates 78.9 1.5 3 K-Ar diabase dyke show some spread, attributed to Ar loss (MacIntyre & Berger, 62.0 1.3 3 K-Ar diabase dyke 75.2 0.8 3 K-Ar k feldspar 1982). 63.8 0.8 3 K-Ar k feldspar (G) The ultrabasic lamprophyre dykes (camptonites) that occur 75.4 0.6 this study U-Pb zircon within the Loule´ salt diapir in the Mesozoic Algarve basin. Monchique 72.0 2 2 K-Ar mineral These rocks are representatives of the minor alkaline igneous separates activity hosted in the Algarve Basin. 72.0 1.5 5 Rb-Sr whole rock isochron 72.0 2 6 Rb-Sr whole rock isochron 4. Analythical methods 72.7 2.7 this study Ar-Ar amphibole 71.5 3.6 this study Rb-Sr whole rock isochron The samples were prepared for analysis at the Departamento de Paço d’Ilhas 88.0 2.7 7 K-Ar biotite Geologia da Universidade de Lisboa. They were hydraulically 86.8 2.5 7 K-Ar k feldspar crushed, in order to ensure the removal of all the signs of alteration, Loule´ dykes 71.8 1.9 this study K-Ar biotite Foz da Fonte sill 93.8 3.9 this study Ar-Ar amphibole and then reduced in size by a jaw crusher and powdered in an agate swing mill. Concentrations for major and trace elements were obtained at Actlabs Ancaster, Ontario, Canada. Major elements were analyzed using Inductively Coupled Plasma Mass Spectrometry (ICP-MS) (C) The Foz da Fonte microgranular tephritic sill, exposed along after alkaline melting with lithium metaborate/tetraborate fol- the N-S coast of the Setu´ bal peninsula (30 km SW of Lisbon). lowed by nitric acid dissolution (except for Ni measurements that The 8 m thick sill, corresponds to the southernmost intrusion were performed after acid digestion). of the Lusitanian basin and intrudes Early Cretaceous sedi- Isotopic analysis of Sr were performed on a VG SECTOR 54 ments (Aptian; Kullberg, 2000). multicollector thermal ionization mass spectrometer (TIMS) in the (D) The syenites of the Sines sub-volcanic intrusion, located half- Laboratory of Isotope Geology of the University of Aveiro, Portugal. way between the Sintra and Monchique massifs. The intrusion Rb and Sr contents were determined by ICP-MS at Actlabs Ancaster, includes a gabbro-diorite-syenite suite and associated dyke Ontario, Canada. The analytical data were corrected for mass frac- swarm that intruded and metamorphosed Jurassic sediments tionation using a exponential law (86Sr/88Sr ¼ 0.1194). Average to the north and Carboniferous shales and greywackes from the levels of blanks for Sr were in the range 0.25–1.0 ng. The standard Mira Formation (Namurian, Oliveira, 1990) to the south. Of all NBS-987 gave a repeated value of 0.710254 þ/ 34 for (2sd of 49 the known alkaline intrusions, this is the occurrence that analyses) during the period of analysis. Regression lines have been shows a wider range of ages (Table 1). calculated using the least-squares method of York (1969) as (E) The basalts and basanites from the Volcanic Complex of Lisbon implemented in the Isoplot program (Ludwig, 2000). that includes lava flows, vents, plugs, sills, and dykes that range For 40Ar/39Ar dating purposes whole rock samples from the Foz from basanitic to ryolithic in composition (Palaćios, 1985). It is da Fonte sill and a Monchique lamprophyre were sent to Actlabs one of the youngest known occurrences in the Lusitanian basin Ancaster, Ontario, Canada, where amphibole separates were z 2 and also the largest, covering an area of 200 km . However, obtained by crushing several kg of sample, sieving, washing and remnants of this complex indicated an original surface that separation using magnetic methods and heavy liquids. The sepa- might have been four times larger. It was dated beforehand rates were wrapped in Al foil and loaded in an evacuated and sealed using a 5 point K-Ar isochron, yielding an age of 72 3.1 Ma quartz vial with K and Ca salts and flux monitors (LP-6 biotite, with (Ferreira and Macedo, 1979, Table 1). and assumed age of 128.1 Ma, Ingamells and Engels, 1976; Odin

Table 2 40Ar/39Ar analytical data for the Foz da Fonte amphiboles. J ¼ 0.007063 0.000130 P T0C t (min) 40Ar(STP) 40Ar/39Ar 1s 38Ar/39Ar 1s 37Ar/39Ar 1s 36Ar/39Ar 1s Ca/K 39Ar (%) Age (Ma) 1s 550 10 5.53*e9 22.36 0.08 0.125 0.004 1.06 0.01 0.0516 0.0033 3.8 1.2 88.5 12.0 700 10 8.61*e9 14.13 0.03 0.034 0.001 0.53 0.01 0.0226 0.0018 1.9 4.3 92.6 6.7 850 10 8.43*e9 13.06 0.02 0.049 0.001 0.24 0.01 0.0196 0.0016 0.9 7.5 90.2 6.0 1000 10 9.12*e9 9.09 0.01 0.105 0.001 0.57 0.01 0.0053 0.0012 2.1 12.5 93.5 4.6 1200 10 143.6*e9 8.19 0.01 0.613 0.001 4.46 0.005 0.0019 0.0008 16.0 99.9 94.7 3.2 1250 10 2.50*e9 127.68 8.51 1.008 0.136 22.82 1.60 0.3966 0.0345 82.2 100.0 128.9 78.4 R. Miranda et al. / Cretaceous Research 30 (2009) 575–586 579 et al., 1982). They were irradiated in the McMaster University, Ontario, Canada, nuclear reactor for 48hours. After the flux monitors were run, J values were calculated, using the measured flux gradient. Neutron gradients did not exceed 0.27% on sample size. The Ar isotope composition was measured in a Micromass 5400 static mass spectrometer. 1200 C blank of 40Ar did not exceed 2– 5*1010 cc STP. Heating time for each step was 10 minutes. After each analysis, the extraction temperature is elevated to 1800 C for few minutes and the furnace is prepared for next analysis. The reported errors correspond to 1s (Table 2). The K-Ar age determination was also performed by Actlabs Ancaster, Ontario, Canada on a biotite separate from an ultrabasic lamprophyre obtained after crushing, washing and sieving the sample, and then using heavy liquids to obtain the mineral separate. The determination of the K concentration was obtained by ICP on Thermo Jarrell Ash Enviro II ICP Spectrometer after an aliquot of the sample was weighted into graphite crucible with lithium metaborate/tetraborate flux and fused using LECO induction Fig. 3. Age temperature spectra diagram for the Foz da Fonte amphibole separate. furnace. For the Ar analysis an aliquot of the sample was weighted Weighted Mean Plateau age, WMPA ¼ 93.6 2.3 Ma (including J). Total fusion age, TFA ¼ 94.4 2.9 Ma (including J). into an Al container, loaded into the sample collector of the extraction system, and degassed at z100 C for two days to remove the surface gases. Argon was extracted from the sample in a double 5.2. Sintra granite vacuum furnace at 1700 C. The Ar concentration was then deter- 38 mined using isotope dilution with a Ar spike, which is introduced The Sintra subvolcanic complex has been the target of several to the sample system prior to each extraction. The extracted gases age determinations. According to the published data, the are cleaned up in a two step purification system. Pure Ar was then emplacement of the granite (81.9 0.4 Ma, K-Ar in biotite, MacIn- introduced into a custom built magnetic sector mass spectrometer tyre & Berger, 1982; 84.0 1.1, K-Ar in K-feldspar, Storetvedt et al., (Reynolds type) with a Varian CH5 magnet. The ion source has an 1987) occurred before the intrusion of the gabbro and syenite axial design (Baur-Signer source), that provides more than 90% magmas (76.1 1.1 to 78.3 1.9 Ma, K-Ar in whole rock syenite, transmission and extremely small isotopic mass discrimination. Ar Storetvedt et al., 1987; 74.9 1.0 Ma whole-rock K-Ar in gabbro, isotope ratios measurements were corrected for mass discrimina- Storetvedt et al., 1987). 36 tion and atmospheric argon was removed assuming that Ar is of In order to obtain a reliable crystallization age for the Sintra 40 atmospheric origin. The concentration of radiogenic Ar was granite laccolith and to assess whether the emplacement of this 38 calculated using a Ar spike. After each analysis the extraction magma occurred before or after the intrusion of the other units of temperature is elevated to 1800 C for few minutes and the furnace the complex, a zircon separate from the granite was dated with the is prepared for the next analysis. U-Pb method, using LA-ICP-MS. The results are shown in Fig. 5 and Zircon separates for U-Pb analyses were obtained after crushing, Table 3. washing and sieving the rock, and then using a standard gravimetric The 79.2 0.8 Ma age obtained for the zircons from the Sintra (Wifley and heavy liquids) and magnetic (Frantz magnetic sepa- granite (Fig. 5), although younger than most of the previously rator) separation techniques. Zoning patterns and inclusions were published ages (Table 1), is still older than the central values for the investigated by cathodoluminiscence and transmitted light known ages for the gabbro and syenite (although within the error microscopy prior to LA-ICP-MS analyses. Analyses were performed limits o some of these) and close to the K-Ar age determined in in the Geoanalytical Laboratory at Washington State University biotite separates from that same granite (Table 1). This is compat- using a New Wave UP213 Nd-YAG (213 nm) laser ablation system, ible with: (i) field observations that show that the gabbroic and and a ThermoFinnigan Element2 single collector double focusing syenitic rocks define intrusive relationships with the granite and magnetic sector ICP-MS. The analytical parameters included a repetition rate of 10 Hz, 40 and 30 microns spot sizes, and a total analysis time of 30 sec per spot. See Chang et al. (2006) for complete analytical and data reduction procedures. Two zircon samples with ages of 564 Ma and 55.5 Ma were used as external standards to correct for elemental fractionation, and mass bias. Accuracy, based on age determinations for standards, is generally 2–3% or better.

5. Geochronology

5.1. Foz da Fonte sill

The 40Ar/39Ar isotopic analyses for amphibole separates from the Foz da Fonte sill give a plateau age of 93.6 Ma 2.3 Ma, corresponding to 98% of released 39Ar (Fig. 3). The reverse isochron diagram results in an age of 93.8 3.9 Ma, MSWD ¼ 0.47 and 40 36 ( Ar/ Ar)0 ¼ 293 11 (Fig. 4). The plateau age appears to correspond to the closing of the amphibole and provides a good estimate for the rock crystallization age, which corresponds to the oldest known age obtained for the entire alkaline magmatism of the Fig. 4. Inverse isochron diagram for the Foz da Fonte amphibole. Inverse isochron onshore WIM. age ¼ 93.8 3.9 Ma. (MSWD ¼ 0.47; 40Ar/36Ar ¼ 293 11). 580 R. Miranda et al. / Cretaceous Research 30 (2009) 575–586

5.3. Sines syenite

The Sines complex was also previously dated, using the K-Ar and Rb-Sr methods. The wide range of ‘‘ages’’ displayed by these rocks (78.9 1.5 to 62.0 1.3 Ma K-Ar in whole rock diabase dykes; 75.2 0.8 to 63.8 0.8 Ma K-Ar age in K-feldspar from the syenite; 72.0 1.5 Rb-Sr whole rock isochron, Canilho and Abranches, 1982), particularly the values below 70 Ma are probably the result of a late thermal overprint that may have caused argon loss (Storetvedt et al., 1987). With the aim of constraining the age of installation for this subvolcanic complex, a zircon separate from the syenite facies was dated using U-Pb geochronology by LA-ICP-MS. The obtained results are plotted in Fig. 6 and Table 3. The 75.4 0.6 Ma U-Pb age acquired for the zircons from the Sines syenites (Fig. 6) is concordant, within analytical error, with the oldest previously obtained K-Ar age on K-feldspar separates (Storetvedt et al.,1987) but older than the whole rock Rb-Sr isochron (Canilho and Abranches, 1982) and the ages obtained by the K-Ar method on diabase dykes and other feldspar separates (Table 1), Fig. 5. U-Pb Tera-Wasserburg plot for the analysed zircons from the Sintra granite. which may be an indicator of significant Ar loss since cooling. Error ellipses 2s error. Ages at 95% confidence level. 5.4. Monchique (ii) recently obtained gravity data showing that in this composite massif vertical gabbro plugs intrude an older granite laccolith The Monchique plutonic complex has been previously dated (Terrinha et al., 2003). Also, the known K-Ar ages for the syenite and using both the K-Ar and the Rb-Sr methods. The obtained ages gabbro suite can be subject to previously mentioned problems, cluster around 72 Ma (Rock, 1976; MacIntyre and Berger, 1982; such as excess Ar, which may exaggerate their age, as it appears to Bernard-Griffiths et al., 1997). be the case with the 84 1.1 age obtained in a K-feldspar separate Recent revision of the geological mapping of this complex allowed (Storetvedt et al., 1987) from the Sintra granite. the identification of two concentric units of different nepheline

Table 3 U-Pb isotopic data obtained by LA-ICP-MS for the analysed zircons

Locality Sample name_ 207Pb/235U2s 206Pb/238U2s RHO 207Pb/206Pb 2s 207Pb/235U2s 206Pb/238U2s 207Pb/206Pb 2s

Zircon grain intercept abs err intercept abs err average abs err age abs err age abs err age abs err Sintra STR1_18A 0.081046 0.004036 0.012522 0.000438 0.876 0.048043 0.001227 79.1 3.8 80.2 2.8 101.4 59.9 STR1_17B 0.080892 0.003152 0.012305 0.000320 0.948 0.047683 0.000787 79.0 3.0 78.8 2.0 83.6 38.9 STR1_17A 0.099415 0.004586 0.012567 0.000330 0.906 0.057378 0.001432 96.2 4.2 80.5 2.1 506.1 54.4 STR1_16A 0.083967 0.003863 0.012212 0.000368 0.916 0.049872 0.001098 81.9 3.6 78.2 2.3 189.0 50.9 STR1_15A 0.088568 0.004780 0.012583 0.000304 0.900 0.051053 0.001730 86.2 4.5 80.6 1.9 243.2 77.2 STR1_14A 0.087453 0.003836 0.012632 0.000356 0.922 0.050215 0.001054 85.1 3.6 80.9 2.3 205.0 48.3 STR1_13A 0.077031 0.003387 0.012064 0.000312 0.915 0.046311 0.001057 75.3 3.2 77.3 2.0 13.8 54.4 STR1_12A 0.078235 0.003456 0.011889 0.000282 0.915 0.047728 0.001165 76.5 3.3 76.2 1.8 85.8 57.4 STR1_11B 0.085289 0.003818 0.012380 0.000273 0.920 0.049967 0.001296 83.1 3.6 79.3 1.7 193.5 59.7 STR1_11A 0.082720 0.003554 0.012413 0.000256 0.934 0.048334 0.001201 80.7 3.3 79.5 1.6 115.7 58.1 STR1_10A 0.088589 0.005076 0.012256 0.000423 0.879 0.052428 0.001657 86.2 4.7 78.5 2.7 304.1 71.2 STR1_9A 0.082863 0.004183 0.012581 0.000433 0.910 0.047770 0.001140 80.8 3.9 80.6 2.8 87.9 56.1 STR1_8A 0.095096 0.004702 0.012267 0.000396 0.905 0.056225 0.001374 92.2 4.4 78.6 2.5 461.3 53.7 STR1_7B 0.086724 0.003844 0.012884 0.000349 0.915 0.048822 0.001091 84.4 3.6 82.5 2.2 139.3 52.0 STR1_5C 0.082794 0.003626 0.012279 0.000350 0.923 0.048907 0.001007 80.8 3.4 78.7 2.2 143.4 48.0 STR1_5B 0.089706 0.005239 0.012079 0.000508 0.907 0.053863 0.001451 87.2 4.9 77.4 3.2 365.3 60.2 STR1_5A 0.090422 0.004089 0.012413 0.000304 0.910 0.052834 0.001324 87.9 3.8 79.5 1.9 321.7 56.4 Wtd avg (95%) 82.7 2.7 79.2 0.8 199.0 72.0 Sines SNS_14A 0.077714 0.003117 0.011530 0.000287 0.860 0.050030 0.001131 76.0 2.9 73.9 1.8 196.4 52.1 SNS_13A 0.078990 0.003922 0.011615 0.000325 0.831 0.050480 0.001546 77.2 3.7 74.4 2.1 217.2 70.1 SNS_12A 0.075151 0.003611 0.012040 0.000351 0.843 0.046330 0.001308 73.6 3.4 77.1 2.2 14.9 67.1 SNS_11B 0.083217 0.004146 0.012036 0.000404 0.864 0.051321 0.001377 81.2 3.9 77.1 2.6 255.3 61.1 SNS_11A 0.076927 0.003837 0.011630 0.000395 0.867 0.049098 0.001301 75.3 3.6 74.5 2.5 152.5 61.5 SNS_10A 0.076682 0.003800 0.011922 0.000360 0.842 0.047743 0.001389 75.0 3.6 76.4 2.3 86.5 68.3 SNS_8B 0.076987 0.003433 0.012038 0.000323 0.847 0.047470 0.001239 75.3 3.2 77.1 2.1 72.9 61.5 SNS_8A 0.076631 0.003570 0.011861 0.000330 0.843 0.047955 0.001322 75.0 3.4 76.0 2.1 97.0 64.6 SNS_7A 0.078468 0.003410 0.011491 0.000307 0.852 0.050687 0.001267 76.7 3.2 73.7 2.0 226.6 57.3 SNS_6B 0.079020 0.005899 0.011353 0.000431 0.805 0.051661 0.002556 77.2 5.5 72.8 2.7 270.4 111.5 SNS_5B 0.080210 0.003594 0.011607 0.000301 0.841 0.051292 0.001382 78.3 3.4 74.4 1.9 254.0 61.4 SNS_5A 0.079097 0.003491 0.011910 0.000314 0.846 0.049293 0.001279 77.3 3.3 76.3 2.0 161.8 60.1 SNS_4B 0.079000 0.003815 0.011852 0.000328 0.835 0.049475 0.001456 77.2 3.6 76.0 2.1 170.4 68.0 SNS_4A 0.077659 0.003277 0.011657 0.000290 0.848 0.049449 0.001230 75.9 3.1 74.7 1.9 169.2 57.6 SNS_3C 0.075436 0.003415 0.011865 0.000298 0.835 0.047190 0.001319 73.8 3.2 76.0 1.9 58.9 65.9 SNS_3B 0.076110 0.003547 0.011674 0.000321 0.841 0.048392 0.001346 74.5 3.3 74.8 2.0 118.5 64.9 SNS_3A 0.080066 0.003886 0.011762 0.000318 0.831 0.050527 0.001520 78.2 3.6 75.4 2.0 219.3 68.9 SNS_1A 0.076458 0.003464 0.011894 0.000300 0.836 0.047715 0.001331 74.8 3.3 76.2 1.9 85.2 65.5 Wtd avg (95%) 77.2 2.6 75.4 0.5 214.7 94.3 R. Miranda et al. / Cretaceous Research 30 (2009) 575–586 581

Fig. 6. U-Pb Tera-Wasserburg plots for the analysed zircons from the Sines syenite. Error ellipses 2s error. Ages at 95% confidence level. syenites, the inner of which encloses various bodies of basic and ultra-basic rocks (Gonzalez-Clavijo and Valadares, 2003). We dated some of the nepheline syenites and ultrabasics identified by these authors with the aim of clarifying some intriguing ages obtained previously that may be related to the previously identified problem with Ar loss and its effect on some of the K-Ar ages obtained even in feldspar and nepheline separates (MacIntyre and Berger, 1982). Representative samples from the different magmatic units of the Monchique massif were selected for Rb-Sr whole-rock dating. The Rb-Sr isotope data together with the Rb and Sr concentrations are given in Table 4. As shown in Fig. 7a, the 5 nepheline syenite samples are roughly co-linear on the 87Sr/86Sr vs. 87Rb/86Sr diagram yielding a model 1 type IsoplotÒ Rb-Sr whole-rock isochron of 72.3 4.2 Ma 87 86 (MSWD ¼ 0.98; Sr/ Sri ¼ 0.703258 0.000049). Due to the low MSWD, the 72.3 4.2 Ma age appears to reflect the real intrusion age of the nepheline syenite magma. A similar age was obtained by adding two samples from the 87 86 mafic intrusives (71.8 5.8 Ma, initial Sr/ Sri ¼ 0.703244 0.000073 and MSWD ¼ 3.5, Fig. 7b) suggesting a coeval Fig. 7. Rb-Sr isochrons for the Monchique samples: (a) 5 nepheline syenites; (b) all emplacement and a genetic link between all units. The reported samples. ages are concordant, within analytical error, with the 72 2 Ma and 72 1.5 Rb-Sr whole-rock isochrons determined by Rock (1976) value and suggests presence of excess Ar. The total fusion age and Bernard-Griffiths et al. (1997), respectively. (75.6 1.3 Ma) shows a good agreement with the plateau age The amphibole separate obtained from an ultrabasic lamprophyre (74.4 2.0 Ma) but the presence of excess Ar suggests that the reverse in the Monchique complex defines a weighed mean plateau age of isochron should reflect the crystalization age for these amphiboles. 39 74.4 2.0 Ma, corresponding to 87% of released Ar (Table 5, The general concordance between the results of 40Ar/39Ar Figs. 8 ,9). On the reverse isochron plot, the plateau points define amphibole dating for the ultrabasic lamprohyres and the Rb-Sr a linear trend with an age of 72.7 2.7 Ma, MSWD ¼ 0.16 and an whole-rock age of the nepheline syenites appear to indicate that 40 36 initial Ar/ Ar of 410 103, which is higher than the atmospheric the different units of the Monchique massif may have been intruded within a short time interval, centered around 72–73 Ma. Table 4 This is also consistent with the geochronological data obtained in Rb and Sr isotopic data for the Monchique rocks. Rb and Sr concentrations obtain by previous studies. ICP-MS, estimated error for the values of 2%. Sample location can be found in Gonzalez-Clavijo and Valadares, 2003 5.5. Loule´ Sample Lithology Sr Rb 87Rb/86Sr 87Sr/86Sr Error (2s) 468B ultrabasic lamprophyre 2240 28 0.036 0.703205 0.000052 ´ 629 syeno-diorite 1580 87 0.159 0.703447 0.000052 The Loule lamprophyre dykes belong to a small group of alkaline 598 nepheline syenite 1290 214 0.480 0.703783 0.000055 intrusives exposed in the Algarve basin, showing strong petro- 745 nepheline syenite 291 189 1.881 0.705189 0.000056 graphical and chemical similarities with their equivalents in the 746 nepheline syenite 593 160 0.781 0.704078 0.000048 nearby Monchique complex (Martins, 1991). 747 nepheline syenite 999 223 0.646 0.703898 0.000046 A biotite separate from the Loule´ lamprophyre dyke was 749 nepheline syenite 1310 190 0.420 0.703674 0.000052 selected for K-Ar isotopic analysis. The results are summarized in 582 R. Miranda et al. / Cretaceous Research 30 (2009) 575–586

Table 5 40Ar/39Ar analythical data for the Monchique amphiboles. J ¼ 0.006625 0.000110 P T0C t (min) 40Ar(STP) 40Ar/39Ar 1s 38Ar/39Ar 1s 37Ar/39Ar 1s 36Ar/39Ar 1s Ca/K 39Ar (%) Age (Ma) 1s 500 10 3.03*e9 17.04 0.19 0.216 0.013 0.96 0.02 0.0289 0.0110 3.5 0.5 98.9 36.7 600 10 4.58*e9 12.63 0.06 0.115 0.004 1.59 0.01 0.0159 0.0047 5.7 1.6 92.4 15.9 700 10 8.44*e9 10.75 0.02 0.067 0.001 0.51 0.01 0.0134 0.0018 1.8 4.0 79.5 6.1 800 10 11.24*e9 10.22 0.02 0.066 0.001 0.26 0.01 0.0099 0.0015 0.9 7.3 85.1 5.1 900 10 8.10*e9 10.25 0.02 0.081 0.002 0.39 0.005 0.0107 0.0019 1.4 9.7 82.7 6.4 950 10 4.71*e9 10.58 0.04 0.103 0.001 0.56 0.01 0.0094 0.0036 2.0 11.1 90.8 12.2 1000 10 5.46*e9 10.23 0.03 0.143 0.004 1.07 0.01 0.0098 0.0025 3.9 12.7 85.6 8.5 1050 10 10.38*e9 8.22 0.01 0.275 0.001 2.70 0.01 0.0049 0.0013 9.7 16.5 79.3 4.4 1100 10 94.81*e9 6.67 0.01 0.238 0.001 3.47 0.01 0.0012 0.0006 12.5 59.6 74.0 2.0 1150 10 85.91*e9 6.47 0.01 0.245 0.001 4.13 0.01 0.0006 0.0006 14.9 99.8 73.8 2.1 1200 10 2.36*e9 40.73 1.27 0.512 0.034 21.77 0.680 0.1154 0.0313 78.4 100.0 77.6 105.1

Table 6. The biotite K-Ar age (71.8 1.9 Ma) is concordant, within patterns (not shown), suggesting the influence of residual garnet analytical error, with the 40Ar/39Ar age obtained for the Monchique and clinopyroxene during melting events Table 7. lamprophyre (74.4 2.0 Ma). This suggests that the K-Ar system Multielemental plots comparing intrusions of similar ages show was not disturbed by subsequent thermal episodes and that the age that some correlation exists between the shape of the pattern and of the biotite separates from this sample still dates the time of the age of intrusion, independently of their geographical location crystallization. From the obtained data, it is also possible to (Fig. 11). conclude that the lamprophyres intruding both the Monchique For instance, the approximately 72 Ma Loule´ dykes, the Lisbon complex and the Algarve basin are coeval. volcanic complex basalts and the less evolved rocks from Mon- chique show similar patterns, with negative anomalies in K, Zr and 6. Geochemistry Hf (Fig. 11a), despite different geographical locations and geological settings. Considering the uncontaminated nature of these mafic The ages reported above belong to Late Cretaceous alkaline magmas, demonstrated by the absence of Ti and Nb troughs, these magmatic rocks occurring either in southern Portugal (37 N; anomalies are probably caused by retention of these elements in Monchique and Loule´) and further north, around the Lisbon area residual mantle phases. K rich amphibole or phlogopite, for K, and (38 200 N; Foz da Fonte, Sintra). Some of the dated intrusives were eventually baddeleyite for, Zr and Hf, are the more probable also sampled for geochemical analysis (Foz da Fonte sill, Monchique retaining phases (Andronikov and Foley, 2002)] surviving during alkaline complex and Loule´ dykes), so that any possible geochem- the small degree of partial melting that characterizes the genera- ical variation with age and/or location could be detected. In tion of alkaline magmas. The samples from the northern Lisbon addition, the Paço d’Ilhas sill and the Lisbon volcanic complex were volcanic complex are, however, less enriched in incompatible also sampled as they represented, respectively, the oldest and elements than the ones from the south and have less prominent youngest known expressions of alkaline magmatism in the anomalies, probably reflecting larger amounts of partial fusion and/ northern region. or source heterogeneities. All of the sampled rocks plot on the alkaline field of the TAS The 94 Ma Foz da Fonte and 88 Ma Paço d’Ilhas sills show diagram (after recalculation on a volatile free basis, Fig. 10) except multielemental plots that are distinct from the ones seen on the c. for one sample of the Paço d’Ilhas basalts, probably reflecting its 72 Ma rocks (Fig. 11b), showing no systematic Zr or Hf anomalies mafic cumulate nature. The other samples classify as basanites and and less important K anomalies. The more evolved samples from tephrites while the lamprophyres fall on the foidite field. Several Paço d’Ilhas show negative Ti anomalies possibly due to the more evolved samples from the Paço d’Ilhas sill plot in the removal of titanomagnetite and clinopyroxene to the basaltic mugearite field reflecting their generation by accumulation of more layers. Note that the small negative anomalies observed for K and silica rich minerals by flotation (Mahmoudi, 1991). Sr in the Paço d’Ilhas basalt samples probably result from the removal of plagioclase and K feldspar into the leucocratic layers All the samples show strongly fractionated (LaN/ by flotation of these less dense crystals, as suggested by YbN ¼ 48.9 15.4) chondrite normalized rare-earth element (REE) Mahmoudi (1991).

Fig. 8. Age temperature spectra diagram for the Monchique amphibole separate. Weighted Mean Plateau age, WMPA ¼ 74.4 2.0 Ma (including J). Total fusion age, Fig. 9. Inverse isochron diagram for the Monchique amphibole. Inverse isochron TFA ¼ 75.6 1.3 Ma (including J). age ¼ 72.7 2.7 Ma. (MSWD ¼ 0.16; 40Ar/36Ar ¼ 410 103). R. Miranda et al. / Cretaceous Research 30 (2009) 575–586 583

Table 6 faults reactivated after the Jurassic rifting. In such a model K-Ar analytical data for the biotite from one of the Loule´ lamprophyres extension perpendicular to the N-S oriented compression, 40 40 Location Material % K Arrad, nl/g % Arair Age, Ma allowed for magma ascent through deep NNW-SSE dextral strike Loule´ biotite 6.75 þ 0.13 18.851 þ 0.310 6.5 71.8 þ 1.9 slip fractures generated by the rotation of Iberia, on the base of which magma had been generated by decompression on releasing bends. However, we emphasize that, contemporane- 7. Discussion and geodynamic implications ously with the onshore magmatism, large volumes of magma were erupted in the oceanic crust adjacent to the WIM (e.g. the The new age determinations, combined with previously pub- Madeira-Tore rise, Geldmacher et al., 2006; Merle et al., 2006). lished data, allow us to constrain the duration of the Late Creta- Therefore, although lithospheric anisotropies have influenced ceous alkaline cycle to circa 22 Ma and to define two distinct pulses the location of the onshore magmatism, the simultaneous of alkaline magmatic activity. The first one (94–88 Ma) occurred occurrence of continental and oceanic magmatism strongly during the opening of the Bay of Biscay and consequent rotation of suggests the presence of a regional mantle melting anomaly and Iberia (123–80 Ma; Sibuet et al., 2004) and clusters around the that horizontal plate kinematics and tectonic processes should Lisbon area (above N38 200) where it occurs mainly as sills (Foz da not be envisaged as the only contributing factor. In fact, the Fonte and Paço d’Ilhas). The second pulse lasted from 75 Ma to existence of uplift and erosion in the WIM (Proença da Cunha 72 Ma corresponds to a more widespread event with scattered and Pena dos Reis, 1995; Rasmussen et al., 1998) as well as occurrences from the Algarve Basin in southernmost Portugal thermal metamorphism and hydrothermal activity in the centre (37N) to the Lisbon area (N39) and comprises both intrusive and of Iberia (Casquet et al., 1992; Tritlla and Sole´ , 1999) pre-dating extrusive complexes. This final pulse is synchronous with the first and during the onset of the Late Cretaceous alkaline cycle pulses of tectonic inversion of the Mesozoic basins due to the onset support the hypothesis of the involvement of anomalous hot of rapid convergence between the African and Iberian plates (e.g. mantle related to this episode. z Mougenot, 1980; Terrinha, 1998; Rosembaum et al., 2002). The The uppermost Triassic - Lower Jurassic ( 200 Ma, Dunn et al., Sintra granite laccolith has an age (79 Ma) between these two 1998; Verati et al., 2007) tholeiitic cycle that occurred in the WIM pulses of magmatism. (Cebria´ et al., 2003; Martins et al., 2008) has been assigned high 87 86 z The similar ages obtained for the lower volume of alkaline initial Sr/ Sri (Messejana dolerite 0.705, Cebria´ et al., 2003; magmatism intruding the opposite rifted and thinned margins of Algarve volcanics > 0.70539, Martins et al., 2008), which coupled 143 144 z the Iberian plate (105–69 Ma in the Pyrenees and Catalan ranges, with low Nd/ Ndi ( 0.5124; Cebria´ et al., 2003) was inter- Fabrie`s et al., 1998; Montigny et al., 1986; Sole´ et al., 2003) suggest preted as pointing to an important contribution from enriched a relationship between the Late Cretaceous magmatism and the domains of the mantle lithosphere in their source (Cebria´ et al., changes in plate kinematics and seafloor spreading geometry 2003). On the other hand, the alkaline rocks from the Late Creta- contemporaneous with the opening of the Bay of Biscay. ceous cycle show isotopic signatures that are clearly distinct from 87 86 A mechanism involving alkaline magma ascent through the tholeiitic cycle (Monchique: Sr/ Sri ¼ 0.70324–0.70327, this 87 86 143 144 z ruptures in the lithosphere in zones of cortical relaxation and/or study; Sr/ Sri ¼ 0.70300–0.70330, Nd/ Ndi 0.51276 to thinning generated during the rotation episode of the Iberian 0.51283, Bernard-Griffiths et al., 1997). This clearly indicates a more plate was suggested (Ribeiro et al., 1979; Sole´ et al., 2003)in time integrated depletion in the source of these magmas which order to provide an explanation for these observations. A similar demonstrates the necessity of invoking a markedly different, sub- process was proposed by Terrinha (1998) for the generation and lithospheric, mantle source for the Late Cretaceous alkaline installation of the three NNW-SSE aligned subvolcanic magmas. complexes (Sintra, Sines and Monchique) along deep seated The anomalously hot sub-lithospheric mantle required to explain the Late Cretaceous alkaline magmatism at the WIM could correspond to actively upwelling, deep anchored, mantle plumes interacting with mid-ocean ridges and other major structures such as were invoked in order to explain the Late Cretaceous magmatism offshore of the WIM (104–65 Ma - Geldmacher et al., 2000, 2006; Merle et al., 2006). Alternatively, the existence of small passive convective diapiric instabilities in the upper mantle as suggested by several authors either in continental and oceanic settings (e.g. Granet et al., 1995; Mata et al., 1998; Lustrino and Wilson, 2007), could justify the genesis of the WIM alkaline magmatism and the existence of the widespread, long lived and yet relatively small volume Circum-Mediterranean Cenozoic Anorogenic igneous province. Hoernle et al. (1995) found geophysical evidence for a S-wave low velocity anomaly in the region between the Madeira and Canaries archipelagos and western and central Europe, covering an area of 2500 4000 km and extending to depths in excess of 500 km that could have provided such an anomalously hot sub- lithospheric mantle. These authors define a mantle component associated with this low velocity zone (the Low Velocity Compo- nent – LVC) that is traceable in rocks belonging to either the Central European Volcanic Province and the Canaries archipelago with ages ranging from Late Cretaceous/Eocene to recent times making this both a widespread and long lived anomaly. Fig. 10. TAS diagram for the studied samples. Alkaline-subalkaline series divide line as A sub-lithospheric origin for the WIM alkaline magmas is in MacDonald (1968). supported by the fact that the known isotopic ratios for rocks 584 R. Miranda et al. / Cretaceous Research 30 (2009) 575–586

Table 7 ICP-MS major and trace element data on Late Cretaceous alkaline rocks from the WIM. Mon –Monchique; FF – Foz da Fonte; PI – Paço d’Ilhas; Loule´ – Loule´ lamprophyres; LVC – Lisbon Volcanic Complex. n.d. non detected

Occurrence Mon Mon FF FF FF PI PI PI PI PI PI Loule´ Loule´ LVC LVC LVC

Sample b468B b468A FF 2 FF 3 FF 4 Pl02 PI03 PI05 PI06 PI08 PI09 C-20 C-12-1 RM66 RM92 RM80

SiO2 33.97 39.89 39.93 40.41 40.43 49.95 45.82 49.45 50.27 49.62 50.94 36.56 34.20 43.30 41.08 41.07 Al2O3 10.89 13.35 13.58 13.74 13.94 17.10 12.60 16.20 16.67 12.21 16.87 11.85 9.58 12.76 12.90 13.16 Fe2O3 16.60 12.72 11.77 12.21 12.22 9.02 13.94 7.84 8.84 13.31 8.21 11.880 11.820 11.930 13.120 12.850 MnO 0.250 0.279 0.146 0.139 0.157 0.143 0.16 0.179 0.142 0.164 0.182 0.18 0.23 0.17 0.19 0.17 MgO 7.87 5.23 6.13 8.54 8.17 3.27 3.95 2.42 3.26 4.31 3.04 7.68 12.48 9.98 9.75 10.07 CaO 14.97 11.72 12.34 8.72 8.91 5.75 7.50 6.66 5.48 7.93 5.80 14.04 14.89 11.45 12.02 10.89

Na2O 2.04 4.15 1.62 1.52 1.56 4.17 2.71 3.96 4.03 2.41 4.22 4.37 2.04 2.54 3.34 3.22 K2O 1.42 3.10 2.12 2.23 2.52 3.02 1.24 2.49 2.97 1.86 3.32 2.510 1.510 1.000 1.050 0.820 TiO2 7.48 4.53 4.45 4.53 4.46 2.20 3.95 2.01 2.19 4.10 2.01 3.88 3.46 3.00 3.92 3.98 P2O5 2.42 1.85 0.64 0.64 0.65 0.67 0.75 0.57 0.68 0.58 0.59 1.40 1.31 0.44 0.69 0.54 LOI 0.77 2.16 6.29 5.83 5.90 4.79 7.62 6.74 5.53 3.94 4.97 6.01 6.98 2.58 1.54 2.75 Total 98.7 99.0 99.0 98.5 98.9 100.1 100.2 98.5 100.1 100.4 100.1 100.4 98.5 99.1 99.6 99.5 Sc 24 12 25 26 24 9 25 9 9 24 9 21 24 31 30 28 V 462 303 362 364 339 123 394 111 122 340 109 297 276 328 365 340 Cr n.d. n.d. 66.0 70.5 58.7 7.0 8.0 13.0 n.d. 8.0 n.d. 110.0 410.0 400.0 280.0 210.0 Ga 21.54 21.67 21.24 20.88 20.98 21.00 19.00 20.00 19.00 21.00 20.00 19.00 16.00 19.00 21.00 20.00 Ge 1.01 1.24 1.61 1.34 1.35 1.50 1.40 1.10 1.50 1.40 1.30 1.30 1.20 1.30 1.20 1.30 Rb 28.3 84.4 43.6 40.4 46.0 76.0 64.0 65.0 67.0 81.0 83.0 58.0 64.0 26.0 43.0 54.0 Sr 2240 3280 1270 1040 1220 1099 577 1099 1054 760 977 1534 2149 763 1035 1043 Y 47.5 53.9 27.4 25.9 27.0 33.1 33.3 34.2 33.1 43.9 36.4 33.7 32.7 22.7 26.6 25.8 Zr 343.8 444.2 310.9 303.3 299.7 347.0 263.0 330.0 307.0 333.0 365.0 345.0 297.0 200.0 285.0 263.0 Nb 127.9 168.5 89.5 86.9 88.3 77.9 67.2 65.8 65.9 75.7 75.5 145.0 124.0 48.3 68.4 64.3 Cs 0.51 1.43 0.70 0.42 0.69 1.30 1.30 1.50 1.30 1.20 1.30 0.80 0.60 0.90 0.60 0.70 Ba 923 1430 680 1380 1030 903 902 807 844 682 894 1178 1230 440 585 613 La 127.6 204.4 60.0 57.7 59.5 60.5 60.3 60.4 62.1 64.7 70.0 124.0 165.0 41.5 57.4 52.7 Ce 265.7 396.6 118.6 115.8 118.9 122.0 123.0 124.0 126.0 134.0 140.0 243.0 293.0 85.2 119.0 108.0 Pr 31.4 42.1 14.6 14.2 14.7 14.0 14.0 14.1 14.2 15.1 15.7 28.1 32.7 10.2 14.3 12.9 Nd 128.1 159.1 60.2 57.6 58.9 56.7 56.1 55.8 56.5 59.6 61.8 105.0 115.0 42.1 59.5 52.6 Sm 24.08 26.50 11.37 11.20 11.17 10.40 10.50 10.30 10.80 11.00 11.40 18.20 18.70 8.03 11.00 9.91 Eu 7.00 7.66 3.59 3.41 3.47 3.40 3.47 3.30 3.41 3.58 3.48 5.61 5.81 2.66 3.62 3.27 Gd 16.58 17.58 8.65 8.15 8.18 8.39 8.75 8.57 8.55 9.00 9.38 14.50 13.90 6.80 9.32 8.46 Tb 2.28 2.45 1.26 1.19 1.21 1.27 1.29 1.26 1.30 1.35 1.44 1.76 1.61 1.02 1.30 1.19 Dy 10.65 11.66 6.07 5.69 5.86 6.58 6.74 6.59 6.79 7.05 7.37 7.94 7.71 4.98 6.17 5.75 Ho 1.79 1.99 1.03 1.00 1.00 1.19 1.25 1.25 1.25 1.31 1.38 1.29 1.26 0.86 1.04 0.97 Er 4.28 5.07 2.59 2.50 2.52 3.02 3.12 3.04 3.08 3.25 3.45 3.25 3.21 2.28 2.69 2.51 Tm 0.50 0.63 0.35 0.32 0.32 0.40 0.39 0.41 0.42 0.44 0.46 0.42 0.41 0.31 0.36 0.33 Yb 2.95 3.73 2.00 1.95 1.92 2.48 2.56 2.59 2.58 2.78 2.78 2.30 2.27 1.81 2.06 1.92 Lu 0.41 0.54 0.29 0.28 0.28 0.38 0.37 0.38 0.36 0.41 0.41 0.31 0.29 0.26 0.27 0.26 Hf 9.49 9.84 7.63 7.62 7.41 8.30 7.40 7.70 7.50 8.50 9.10 7.20 6.30 5.30 7.00 6.50 Ta 8.11 11.38 4.98 5.05 5.09 5.62 5.04 4.97 5.19 5.81 5.65 11.50 6.82 4.10 5.92 5.36 Th 6.61 18.81 6.33 6.23 6.35 8.01 7.19 7.50 7.41 8.45 9.07 7.89 13.50 4.42 5.66 5.55 U 1.58 4.89 1.74 1.73 1.72 2.36 2.24 2.18 2.28 2.53 2.75 5.64 2.96 1.20 1.50 1.36 Ni n.d. n.d. 42 41 57 2 2 3 2 2 2 61 190 143 154 126 #Mg 47.2 43.7 49.6 56.9 55.8 40.6 34.84 36.8 41.0 37.9 41.1 54.9 66.6 61.2 58.4 59.7

LaN/YbN 29.2 36.9 20.2 19.9 20.8 16.4 15.86 15.7 16.2 15.7 17.0 36.3 49.0 15.4 18.8 18.5

belonging to the alkaline cycle are within the compositional array to each pulse equilibrated with distinct residual mantle assem- defined for the component associated with the low velocity blages. In order to explain the important negative K anomalies on anomaly defined by Hoernle et al. (1995) (LVC: 87Sr/86Sr ¼ the rocks from the second alkaline pulse it is necessary to have 0.7030–0.7034; eNd ¼ 3.5–5.9; Hoernle et al., 1995) and are also magmas equilibrating in depth with a residual K-bearing phases, included in the compositional range defined by the Common such as K-rich amphiboles and/or phlogopite. Those minerals are Mantle Reservoir (CMR - 87Sr/86Sr ¼ 0.7030–0.7037; eNd ¼ 2.2– not stable at temperatures associated with the asthenospheric 5.8) of Lustrino and Wilson (2007). These components are mantle and upwelling mantle plumes (e.g. Class and Goldstein, thought to represent mostly sub-lithospheric upper mantle 1997), which argues for important magma-lithosphere interaction enriched either by ancient plume activity (LVC, Hoernle et al., during the second pulse of the Late Cretaceous alkaline magma- 1995) or by recycling of crust introduced in ancient subduction tism, in opposition to reduced or inexistent interactions in the first zones or through crustal delamination processes (CMR, Lustrino pulse, deduced from less significant K anomalies. A possible cause and Wilson, 2007). It is therefore suggested that the sub-litho- for the lack of interaction during the first pulse is a quicker ascent spheric Low Velocity Zone (Hoernle et al., 1995) could have due to a more favourable tectonic setting as a result of the provided the anomalously hot sub-lithospheric source of the WIM contemporaneous opening of the Bay of Biscay and consequent alkaline magmatism and possibly all-alkaline Late Cretaceous rotation of the Iberian plate which probably induced whole litho- rocks of Iberia as well as the heat source for the contemporaneous sphere fracturing (Terrinha, 1998). During the second pulse, thermal activity in Central Iberia. however, the previously mentioned onset of the rapid collision We demonstrated that the two pulses of Late Cretaceous between the African and Iberian plates created less favourable magmatic activity in the WIM are characterized by somewhat stress fields and restrained the opening of fractures and quick distinct elemental fingerprints. Such differences in trace element magmatic ascent, which favoured interaction with subcontinental geochemistry allow us to propose that the magmas corresponding lithosphere. R. Miranda et al. / Cretaceous Research 30 (2009) 575–586 585

(5) Differences in residual mantle mineralogy characterizing the two pulses of Late Cretaceous magmatic activity can be explained by differences in magma ascent rates due to more or less favourable tectonic settings that permitted or prevented the interaction with metassomatized lithosphere.

Acknowledgements

RM benefited from a BIC inserted in the project GEODYN (POCTI/ ISFL/5-32) and a PhD grant from Fundaçaõ para a Cieˆncia e a Tec- nologia (SFRH/BD/23028/2005). VV beneffited from PhD grant from Fundaçaõ para a Cieˆncia e a Tecnologia (SFRH/BD/17603/2004). Data acquisition and analytics were sponsored by FCT funded projects GEOSTES (PRAXIS/P/CTE/11052/1998), MATESPRO (PDCTM1999MAR15264) and MAGMAFLUX (POCTI/CTA/48450/ 2002). The authors would like to acknowledge an anonymous reviewer for critical comments and suggestions that improved the original manuscript.

References

Alvaro, M., Apalategui, O., Baena, J., Balcells, R., Barnolas, A., Barrera, J.L., Bellido, F., Cueto, L., Dıáz de Neira, A., Elı´zaga, E., Fernańdez-Gianotti, J.R., Ferreiro, E., Gabaldo´ n, V., Garcıá-Sansegundo, J., Gome´z, J., Heredia, N., Hernańdez-Urroz, J., Hernańdez-Samaniego, A., Lendıńez, A., Levya, F., Lo´ pez-Olmedo, F., Lorenzo, S., Martıń, L., Martıń Serrano, A., Matas, J., Monteserrıń, V., Nozal, F., Olive, A., Ortega, E., Piles, E., Ramı´rez, J., Robador, A., Roldań, F., Rodrı´guez, L., Ruiz, P., Ruiz, M.T., Sańchez-Carretero, R., Teixell, A., 1994. Mapa Geolo´ gico Nacional a escala 1:1.000.000. IGME, Madrid. Andronikov, A.V., Foley, S.F., 2002. Trace element and Nd-Sr isotopic composition of ultramafic lamprophyres from the East Antartic Beaver Lake area. Chemical Geology 175, 291–305. Auzende, J.-M., Olivet, J.-L., Charvet, J., Le Lann, A., Le Pichon, X., Monteiro, J.H., Fig. 11. Primitive upper mantle normalized (Palme and O’Neill, 2004) multielemental Nicolas, A., Ribeiro, A., 1978. Sampling and observation of oceanic mantle and plots for: a) the z72 Ma Lisbon volcanic complex, Monchique and Loule´ samples; b) crust on Gorringe Bank. Nature 273, 45–48. the z90 Ma Paço d’Ilhas and Foz da Fonte samples. Bernard-Griffiths, J., Gruau, G., Cornen, G., Azambre, B., Mace, J., 1997. Continental lithospheric contribution to alkaline magmatism: isotopic (Nd, Sr, Pb) and geochemical (REE) evidence from Serra de Monchique and Mout Ormonde 8. Conclusions complexes. Journal of Petrology 38, 115–132. Canilho, M.H., Abranches, M.C., 1982. Rb-Sr geochronology of the sines alkaline complex. Comuniço˜es dos Serviços Geolo´ gicos de Portugal, Lisboa, 68, 237–240. Our data allows the following conclusions: Casquet, A., Galindo, C., Gonzalez-Casado, J.M., Alonso, A., Mas, R., Rodas, M., Garcia, E., Barrenechea, J.F., 1992. El metamorfismo en la cuenca de los Cameros. Geocronologia e implicaciones tecto´ nicas. Geogaceta 11, 22–25. (1) Alkaline magmatic activity in the onshore part of the WIM is Cebria´, J.M., Lo´ pez-Ruiz, J., Doblas, M., Martins, L.T., Munha´, J.M., 2003. Geochem- divided in two pulses. The first one occurred between 94–88 Ma istry of the Early Jurassic Messejana-Plasencia dyke (Portugal-Spain); implica- and is limited to sills in the Lisbon region. The second pulse tons on the origin of the Central Atlantic magmatic province. Journal of apparently involved a much higher volume of magma and took Petrology 44, 547–568. Chang, Z., Vervoort, J.D., Knaack, C., McClelland, W.C., 2006. U-Pb dating of zircon by place between 75–72 Ma. It has a wider geographical distribu- LA-ICP-MS. Geochemistry, Geophysics, Geosystems 7, 1–14, doi:10.1029/ tion and more diverse modes of occurrence such as subvolcanic 2005GC001100. Q05009. complexes, flows, vents, plugs, dykes and sills. Class, C., Goldstein, S.L., 1997. Plume-lithosphere interections in the ocean basins: constrains from the source minerology. Earth and Planetary Science Letters 150, (2) Radiometric dating of the Sintra granite gives an 79.2 0.8 Ma 245–260. age, which although equivalent within error to the oldest age Cornen, G., 1982. Petrology of the alkaline volcanism of Gorringe Bank (southwest obtained for the syenite (78.3 1.9 Ma, whole rock K-Ar, Portugal). Marine Geology 47, 101–130. Dunn, A.M., Reynolds, P.H., Clarke, D., Ugidos, J., 1998. A comparison of the age and Storetvedt et al., 1987) its still older than the younger K-Ar ages composition of the Shelburne dyke, Nova Scotia, and the Messejana dyke, Spain. obtained for these rocks (76.4 1.4 and 76.1 1.1 Ma, Canadian Journal of Earth Sciences 35, 1110–1115. Storetvedt et al., 1987). This, combined with the observed field Fabrie`s, J., Lorand, J.-P., Bodinier, J.-L., 1998. Petrogenetic evolution of orogenic lherzolite massifs in the central and western Pyrenees. Tectonophysics 292, 145–167. cross-cutting relations and geophysical data (gravimetry and Fe´raud, G., Gastaud, J., Auzende, J.-M., Olivet, J.-L., Cornen, G., 1982. 40Ar/39Ar ages AMS, Terrinha et al., 2003), allows us to speculate that the for the alkaline volcanism and basement of the Gorringe Bank, North Atlantic granite corresponds to an older phase that is intruded by Ocean. Earth and Planetary Science Letters 57, 211–226. Ferreira, M.R.P., Macedo, C.R., 1979. K-Ar Ages of the Permian-Mesozoic Basaltic a more recent gabbro to syenite suite. Activity in Portugal. Abstracts VI. Europ. Col. Geochron., Cosmochron. and (3) As shown by geochronological and isotopic data, all the members Isotope Geology, Norway, p. 26–27. of the Monchique alkaline suite are co-magmatic and appear to Galdeano, A., Moureau, M., Pozzi, J., Berthou, P.Y., Malod, J.A., 1989. New paleo- have been emplaced within a short interval of time, within the magnetic results from Cretaceous sediments near Lisboa (Portugal) and implications for the rotation of Iberia. Earth and Planetary Science Letters 92, analytical error of the used geochronological methods. 95–106. (4) The Late Cretaceous alkaline magmatism of the WIM has Geldmacher, J., van den Bogaard, J.P., Hoernle, K., Schminke, H.U., 2000. The 40 39 a distinctive isotopic composition indicating a sub-lithospheric Ar/ Ar age dating of the Madeira Archipelago and hotspot track (eastern North Atlantic). Geochemistry, Geophysics, Geosystems 1, 2. source, noticeably different from the lithospheric signature Geldmacher, J., Hoernle, K., Klu¨ gel, A., Bogaard, P., Wombacher, F., Berning, B., 2006. associated with the Jurassic tholeiitic cycle. The widespread Origin and geochemical evolution of the Madeira-Tore Rise (eastern North and long lived low S-wave velocity anomaly identified by Atlantic). Journal of Geophysical Research 111 (B9), B09206. Gonzalez-Clavijo, E., Valadares, V., 2003. O Maciço Alcalino de Monchique Hoernle et al. (1995) or an actively upwelling plume are (SW Portugal): estrutura e modelo de instalaçaõ na crosta superior. Comuni- possible sub-lithospheric sources for the alkaline magmatism. caço˜es do Instituto Geolo´ gico e Mineiro 90, p. 43–64. 586 R. Miranda et al. / Cretaceous Research 30 (2009) 575–586

Gradstein, F.M., Ogg, J.G., Smith, A.G., Agterberg, F.P., Bleeker, W., Cooper, R.A., Odin, G.S., 35 collaborators, 1982. Interlaboratory Standards for Dating Purposes. Davydov, V., Gibbard, P., Hinnov, L.A., House, M.R., Lourens, L., Luterbacher, H.P., Numerical Dating in Stratigraphy. Wiley and Sons, Chichester, p. 123–149. McArthur, J., Melchin, M.J., Robb, L.J., Shergold, J., Villeneuve, M., Wardlaw, B.R., Oliveira, J.T., 1990. Stratigraphy and syn-sedimentary tectonism in the South Ali, J., Brinkhuis, H., Hilgen, F.J., Hooker, J., Howarth, R.J., Knoll, A.H., Laskar, J., Portuguese Zone. In: Dallmeyer, R.D., Martinez Garcia, E. (Eds.), Pre Mesozoic Monechi, S., Plumb, K.A., Powell, J., Raffi, I., Ro¨hl, U., Sadler, P., Sanfilippo, A., Geology of Iberia. Springer, pp. 334–347. Schmitz, B., Shackleton, N.J., Shields, G.A., Strauss, H., Van Dam, J., van Palaćios, T.P., 1985. Petrologia do Complexo Vulcaˆnico de Lisboa. Faculdade de Kolfschoten, T., Veizer, J., Wilson, D., 2004. A Geologic Time Scale 2004. Cieˆncias da Universidade de Lisboa, Portugal, Tese de Doutoramento, p. 260. Cambridge University Press, p. 589. Palme, H., O’Neill, H. St. C., 2004. Cosmochemical Estimates of Mantle Composition, Granet, M., Wilson, M., Achauer, U., 1995. Imaging a mantle plume beneath the vol. 2. Treatise on Geochemistry, p. 1–38. French Massif Central. Earth and Planetary Science Letters 136, 281–296. Pinheiro, L.M., Wilson, R.C.L., Pena dos Reis, R., Whitmarsh, R., Ribeiro, A., 1996. The Hoernle, K., Zhang, Y.-S., Graham, D., 1995. Seismic and geochemical evidence for western Ibe´ria margin: a geophysical and geological overview. In: Whitmarsh, large scale mantle upwelling beneath the eastern Atlantic and western and R., Sawyer, D., Klaus, A., Masson, D., (Eds.).Proceedings of the Ocean Drilling central Europe. Nature 374, 34–39. Program, Scientific Results 149. Ingamells, C.O., Engels, J.C., 1976. Preparation, analysis and sampling constants for Proença da Cunha, P., Pena dos Reis, R., 1995. Cretaceous sedimentary and tectonic a biotite. National Bureau of Standards Special Publication 422, 401–419. evolution of the northern sector of the Lusitanian Basin (Portugal). Cretaceous Juarez, M.T., Lowrie, W., Osete, M.L., Melendez, G., 1998. Evidence of wide- Research 16, 155–170. spread Cretaceous remagnetization in the Iberian Range and its relation Rasmussen, E.S., Lomholt, S., Andersen, C., Vejbaek, O., 1998. Aspects of the struc- with the rotation of Iberia. Earth and Planetary Science Letters 160, tural evolution of the Lusitanian Basin in Portugal and the shelf and slope area 729–743. offshore Portugal. Tectonophysics 300, 199–225. Kullberg, J., 2000. Evoluçaõ tecto´ nica Mesozo´ ica da Bacia Lusitaˆniana. FCT, UNL, Tese Ribeiro, A., Antunes, M.T., Ferreira, M.P., Rocha, R., Soares, A., Zbyszewski, G., Moi- de Doutoramento. tinho de Almeida, F., 1979. Introduction a` la geólogie geńe´rale du Portugal. Serv. Kullberg, M.C., Kullberg, J.C., Terrinha, P., 2000. Tecto´ nica da cadeia da Arra´bida. Geol. Portugal, p. 114. Memo´ rias de Geocieˆncias, 2. Museu Nac. Hist. Nat. Univ. Lisboa, p. 35–85. Ribeiro, A., Kullberg, M.C., Kullberg, J.C., Manuppella, G., Phipps, S., 1990. A review of Ludwig, K.R., 2000. Users Manual for Isoplot/Ex, Version 2.45, A Geochronological Alpine tectonics in Portugal: foreland detachment in basement and cover rocks. Toolkit for Microsoft Excel, vol. 1a. Berkeley Geochronology Center, Special Tectonophysics 184, 357–366. Publication, p. 43. Rock, N.M.S., 1976. The comparative strontium isotopic composition of alkaline Lustrino, M., Wilson, M., 2007. The circum-Mediterranean anorogenic Cenozoic rocks: new data from Southern Portugal and East Africa. Contributions to igneous province. Earth-Science Reviews 81, 1–65. Minerology and Petrology 56, 205–228. MacDonald, G.A., 1968. Composition and origin of Hawaiian lavas. In: Coats, R.R., Rock, N.M.S., 1978. Petrology and petrogenisis of the Monchique alkaline complex, Hay, R.L., Anderson, C.A. (Eds.), Studies in Volcanology: A Memoir in Honour Southern Portugal. Journal of Petrology 19, 171–214. of Howel Williams, vol. 116. Geological Society of America Memoir, pp. Rock, N.M.S., 1982. The Late Cretaceous alkaline igneous province in the Iberian 477–522. Peninsule, and its tectonic significance. Lithos 15, 111–131. MacIntyre, R.M., Berger, G.W., 1982. A note on the geochronology of the Iberian Rosembaum, G., Lister, G.S., Duboz, C., 2002. Relative motions of Africa, Iberia and Alkaline Province. Lithos 15, 133–136. Europe during Alpine orogeny. Tectonophysics 359, 117–129. Mahmoudi, A., 1991. Quelques intrusions alcalines et basiques du Cretace´ superieur Russell, S.M., Whitmarsh, R.B., 2003. Magmatism at the west Iberia non volcanic au Portugal. These de Doctorat, Univ. Nancy I. rifted continental margin: evidence from analyses of magnetic anomalies. Martins, L.T., 1991. Actividade I´gnea Mesozo´ ica em Portugal., Tese de Doutoramento, Geophysical Journal International 154, 706–730. Faculdade de Cieˆncias da Univeridade de Lisboa. Sibuet, J.C., Srivastava, S., Spakman, W., 2004. Pyrenean orogeny and plate Martins, L.T., Madeira, J., Youbi, N., Munha´, J., Mata, J., Kerrich, R., 2008. Rift-related kinematics. Journal of Geophysical Research 109, B8. magmatism of the Central Atlantic magmatic province in Algarve, Southern Sole´, J., Pi, T., Enrique, P., 2003. New data on the Late Cretaceous alkaline magma- Portugal. Lithos 101, 102–124. tism of the northeast Iberian Peninsula. Cretaceous Research 24, 135–140. Ma´rton, E., Abranches, M.C., Pais, J., 2004. Iberia in the Cretaceo s: new Storetvedt, K.M., Mogstad, H., Abranches, M.C., Mitchell, J.G., Serralheiro, A., 1987. paleomagnetic results from Portugal. Journal of Geodynamics 38, 209–221. Paleomagnetism and isotopic age data from Upper Cretaceous igneous rocks of Mata, J., Kerrich, R., MacRae, N.D., Wu, T.-W., 1998. Elemental and isotopic (Sr, Nd, W Portugal; geological correlation and plate tectonic aspects. Geophysical and Pb) characteristics of Madeira Island basalts: evidence for a composite Journal Royal Astronomical Society 88, 241–263. HIMU - EM I plume fertilizing lithosphere. Canadian Journal of Earth Sciences Terrinha, P., 1998. Structural Geology and Tectonic Evolution of theAlgarve Basin, 35, 980–997. South Portugal. Imperial College, University of London, PhD Thesis. Merle, R., Scha¨rer, U., Cornen, G., Girardeau, J., 2006. Cretaceous seamounts along Terrinha, P., Aranguren, A., Kullberg, M.C., Pueyo, E., Kullberg, J.C., Casas Sainz, A.M., the continent-ocean transition of the Iberian Margin: U-Pb ages and Pb-Sr-Hf Rillo, C., 2003. Complexo ı´gneo de Sintra - um modelo de instalaçaõ con- isotopes. Geochimica et Cosmochimica Acta 70, 4950–4976. strangido por novos dados de gravimetria e ASM. In: Cieˆncias da Terra, vol. E. V, Montigny, R., Azambre, B., Rossy, M., Thuizat, R., 1986. Kr-Ar study of cretaceous VI. Congresso Nacional de Geologia, p. 58–59. magmatism and metamorphism in the Pyrenees: age and length of rotation of Tritlla, J., Sole´, J., 1999. A newly dated Cretaceous hydrothermal event in the Iberian the Iberian Peninsula. Tectonophysics 129, 257–273. Ranges(Eastern Spain) and its significance within the Mesozoic thermal history Moreau, M.G., Berthou, J.Y., Malod, J.-A., 1997. New paleomagnetic data from the in the Iberian Peninsula. Ore Geology Reviews 15, 243–259. Algarve (Portugal): fast rotation of Ibe´ria between the Hauteverian and the Verati, C., Rapaille, C., Fe´raud, G., Marzoli, A., Bertrand, H., Youbi, H., 2007. 40Ar/39Ar Aptian. Earth and Planetary Science Letters 146, 689–701. ages and duration of the Central Atlantic Magmatic Province volcanism in Mougenot, D., 1980. Une phase de compression au Cre´tace´ terminal a` l’Ouest du Morocco and Portugal and its relation to the Triassic–Jurassic boundary. Portugal: quelques arguments. Boletim da Sociedade Geolo´ gica de Portugal Palaeogeography Palaeoclimatology Palaeoecology 244, 308–325. XXII, 233–239. York, D., 1969. Least squares fitting of a straight line with correlated errors. Earth Neves, M.C., Terrinha, P., Afilhado, A., Moulin, M., Matias, L., Rosas, F., Response of and Planetary Science Letters 5, 320. a multi-domain continental margin to compression: study from seismic Zbyszewski, G., Moitinho d’Almeida, F., Torre de Assunçaõ, C., 1955. Notıćia expli- reflection - refraction and numerical modelling in the Tagus Abyssal Plain. cativa da Carta geolo´ gica de Portugal, 1/50 000 Folha 30-C, Torres Vedras. Serv. Tectonophysics, in press. doi: 10.1016/j.tecto.2008.05.008. Geol. Portugal, p. 33.