| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Journal of Petrology | Volume 44 | Number 3 | Pages 547-568 | 2003
© Oxford University Press 2003
Geochemistry of the Early Jurassic Messejana–Plasencia dyke (Portugal–Spain); Implications on the Origin of the Central Atlantic Magmatic Province
1 DEPARTAMENTO DE GEOLOGÍA, MUSEO NACIONAL DE CIENCIAS NATURALES (CSIC), JOSÉ GUTIERREZ ABASCAL 2, 28006 MADRID, SPAIN
2 DEPARTAMENTO DE GEOLOGIA, UNIVERSIDADE DE LISBOA, CAMPO GRANDE C2, 1749-016 LISBON, PORTUGAL
Telephone: +34-914111328. Fax: +34-915644740. E-mail: cebria{at}mncn.csic.es
RECEIVED JULY 23, 2001; ACCEPTED SEPTEMBER 24, 2002
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONSAMPLING AND ANALYTICAL...PETROLOGICAL AND GEOCHEMICAL...PETROGENETIC MODELCAMP MANTLE SOURCES AND...DISCUSSION AND CONCLUSIONSREFERENCES |
|---|
The Messejana–Plasencia dyke (MPD) is one of the largest doleritic dykes of the so-called Central Atlantic Magmatic Province (CAMP). Emplaced at
200 Ma in the northernmost sector of this province, it crops out for some 530 km from South Portugal to Central Spain with variable width ranging from 5 to 200 m. The bulk-rock composition is that of a typical continental tholeiitic basalt and there is a remarkable geochemical homogeneity along the dyke, with only minor petrographic variations from the margin (microdolerite) to the centre (gabbro). Major and trace element and Sr–Nd–Pb isotopic compositions of the samples from the MPD suggest that they were the result of combined assimilation plus fractional crystallization (AFC). Quantitative modelling shows that the parental magma originated from a source that was 87Sr enriched and 143Nd depleted (87Sr/86Sr 0·7050 and 143Nd/144Nd 0·51244) relative to Primitive Mantle, which may represent an enriched portion of the mantle–lithosphere. The Pb isotope composition of the MPD is comparable with that of other CAMP dolerites and suggests that the enrichment of the mantle lithosphere occurred during a metasomatic event at 1 Ga. Before their emplacement at upper-crustal levels, the primitive magmas crystallized up to 30% and assimilated <10% of lower-crustal felsic granulitic rocks. Comparison of the MPD data with those from other parts of the CAMP suggests that their petrogenesis can be linked to the arrival of a Central Atlantic mantle Plume (CAP) at the base of the lithosphere before the onset of continental break-up. The presence of the CAP provides an explanation for several other geological and geochemical features of this province (e.g. the large-scale regional dyke orientations and the presence of magmatic rocks with plume-like signatures), and could have provided the heat source required for inducing melting of enriched, easily melted domains within the overlying refractory mantle lithosphere. However, the scarcity of rocks with plume-like signatures and the results of our petrogenetic modelling suggest that the CAP was only rarely the direct source of the magmas. The explanation for this is that the plume head was probably detached, cooling quickly after impingement and spreading or channelling in the form of a wide, hot sheet following favourable paths of thinned lithosphere. This scenario explains why the CAMP is mainly constituted by dykes whereas lava flows are scarce. KEY WORDS: assimilation and fractional crystallization; Central Atlantic Magmatic Province; Central Atlantic Plume; continental tholeiites; Messejana–Plasencia dyke
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONSAMPLING AND ANALYTICAL...PETROLOGICAL AND GEOCHEMICAL...PETROGENETIC MODELCAMP MANTLE SOURCES AND...DISCUSSION AND CONCLUSIONSREFERENCES |
|---|
The extensively studied Central Atlantic Magmatic Province (CAMP, Fig. 1), at present separated into a number of segments by the opening of the central Atlantic Ocean, has been identified recently as the biggest large igneous province (LIP) on Earth with an areal extent of up to 7 x 106 km2 (Marzoli et al., 1999
; Olsen, 1999), covering considerable portions of four tectonic plates. Although the distribution of dykes on a local scale shows more complex patterns (see, e.g. Bertrand, 1991; Hames et al., 2000), on a large scale it represents the best example on this planet of a complete giant radiating dyke swarm system (Greenough & Hodych, 1990; Ernst et al., 1995; Dalziel et al., 2000), comparable only with some of the radial dyke swarms of Venus (Ernst et al., 1995). It is also one of the few LIPs that is mainly constituted by dykes and sills whereas volcanic outflows are relatively scarce. Despite the obvious interest of this province and after many decades of discussion on a wide variety of CAMP-related tectonomagmatic topics, many aspects of the CAMP remain controversial (Deckart et al., 1997; Olsen, 1999; Hames et al., 2000; McHone, 2000; Puffer, 2001).
|
The study of regionally extensive dyke systems represents one of the fundamental tools in the analysis of LIPs to clarify their origin (e.g. plume vs non-plume) and geodynamic setting. In this context, the detailed geochemical study of the Messejana–Plasencia dyke (MPD) reported here may shed light on some of the unresolved questions linking LIPs and regional dyke swarms, such as (1) the association of radiating dykes with either sublithospheric plume impingement (Ernst et al., 1995
) or mantle insulation beneath highly refractory cratons (Yale & Carpenter, 1998) and (2) vertical emplacement vs far-reaching (>3000 km) and nearly instantaneous (in less than a few millionyears) lateral migration of magma from its source (Greenough & Hodych, 1990; Ernst et al., 1995; Elliot et al., 1999).
The Messejana–Plasencia dyke–fault system is one of the major representatives of the CAMP dyke swarm. It is also one of the most conspicuous geological features in the Iberian Peninsula, trending SW–NE and cutting the Hercynian basement for >530 km from Alentejo (South Portugal) to Avila (Central Spain), where it is covered by Tertiary sediments (Fig. 2). This lineament has had a complex tectonic history (Schermerhorn et al., 1978; Schott et al., 1981) involving an initial Late Variscan sinistral strike-slip fault, later reactivated as a transtensional fault allowing the emplacement of Triassic–Jurassic magmas, and finally reactivated as a major neotectonic transtensional lineament during the Alpine cycle.
|
Earlier K–Ar age determinations of the MPD on whole rocks and plagioclase separates (Soares de Andrade, 1972
; Teixeira & Torquato, 1975; Schermerhorn et al., 1978; Schott et al., 1981) resulted in a relatively wide range of ages from 156 ± 2 to 276 ± 13 Ma, which were interpreted as the result of the emplacement of the dyke by multiple pulses or intrusions of magma at different times. However, as some workers have indicated (e.g. Schermerhorn et al., 1978), perturbation of the K–Ar system makes these ages unreliable. More recent 40Ar/39Ar age determinations on plagioclase, amphibole, pyroxene and biotite mineral separates suggest that the MPD was emplaced during a brief period of time around 200 Ma (198·8 ± 1·7 to 204·7 ± 2·5 Ma; Sebai et al., 1991; Dunn et al., 1998). This age coincides with the peak of tholeiitic magmatism in the CAMP (Marzoli et al., 1999). In fact, the orientation of the MPD in the pre-drift Triassic–Jurassic continental reconstructions (Fig. 1) coincides with the radial pattern of dolerite dykes around the Central and North Atlantic, interpreted as the result of a stress field imposed on the lithosphere by the upper mantle before the break-up of the North Atlantic Ocean (May, 1971), and it is thus suggested that the MPD represents one of the farthest-north manifestations of the CAMP.
Although the MPD has been studied since the early 1920 s (assigned different names, such as Messejana, Alentejo–Plasencia, Odemira–Avila), from the petrological–geochemical point of view the dyke has been analysed mainly for major and trace element compositions (e.g. Dupuy & Dostal, 1984; Bertrand & Millot, 1987; Martins, 1991) and K–Ar and Ar–Ar age determinations (e.g. Ferreira & Portugal, 1977; Schott et al., 1981; Sebai et al., 1991; Dunn et al., 1998) with very few Sr–Nd isotopic analyses available (e.g. Alibert, 1985). Furthermore, several key questions concerning the origin and geodynamic setting of this dyke remain unanswered. For example, as most petrological–geochemical studies have focused on high-level magmatic differentiation processes (e.g. Alibert, 1985; Bertrand & Millot, 1987; Martins, 1991) there is no precise information on the composition and location of the mantle source of the primary magmas (asthenosphere, plume and/or lithosphere), and the implications of the relationship of this source with the CAMP. In this study we present the first combined major and trace element, and Sr–Nd–Pb isotopic study of the tholeiitic dolerites of the MPD. These data are used to assess the possible petrogenetic–geodynamic scenario, which leads to new constraints on the origin of the magmas from this magmatic province.
|
SAMPLING AND ANALYTICAL PROCEDURES |
|---|
|
TOPABSTRACTINTRODUCTIONSAMPLING AND ANALYTICAL...PETROLOGICAL AND GEOCHEMICAL...PETROGENETIC MODELCAMP MANTLE SOURCES AND...DISCUSSION AND CONCLUSIONSREFERENCES |
|---|
Samples were collected along the whole outcrop of the MPD in Spain and Portugal. Although thin sections were made from all collected samples, only the freshest were finally selected for analysis and are those summarized in Table 1 and shown in Fig. 2. As weathering is variable along the dyke, this selection procedure implies that more samples are available for the less altered sections. However, the results presented below are considered representative of the whole dyke length.
|
Mineral analyses were performed by electron microprobe ( Jeol JCXA 733) at the Centro de Geologia da Universidade de Lisboa. Whole-rock samples were analysed for Si, Al, total Fe, Mn, Mg, Ca, Na and K by atomic absorption spectrometry using standard techniques at the Serviços Geológicos de Portugal, where Ti, P, Rb, Ba, Sr, Y, Zr, V, Ni and Cu were also analysed by X-ray fluorescence on pressed powder discs using a Philips PW1410/00 spectrometer. Further details have been given by Martins (1991)
. Trace elements Hf, Ta, Th, U, Sc, Cr, Co, Zn, Cs and rare earth elements (REE) were determined by instrumental neutron activation analysis (INAA) at the Open University (UK) using a modified version of the technique of Potts et al. (1985).
Sr and Nd isotopic ratios were measured on 200 mg of whole-rock powders. To minimize the possible alteration of the original ratios by secondary alteration, the samples were leached for 1 h in warm 2·5M HCl in sealed PFA Teflon beakers. The leached residues were then dried down and digested in concentrated HNO3 + HF for 48 h minimum. Conventional cation exchange procedures were then followed to obtain Sr and Nd. Pb isotopic ratios were measured on aliquots of 200 mg of sample. Leaching procedures were the same as those followed for Sr and Nd, and standard cation exchange procedures were also used. Strontium was analysed on a VG 54E Isomass spectrometer, and Nd and Pb were analysed on a VG MM30 mass spectrometer, both at the Geochronology Laboratory of the School of Earth Sciences (University of Leeds, UK). Nd isotopic compositions were normalized to 146Nd/144Nd = 0·7219 relative to La Jolla standard (143Nd/144Nd = 0·511897 ± 18), and Sr isotopic compositions were normalized to 87Sr/86Sr = 0·1194 relative to SRM 987 standard (87Sr/86Sr = 0·7102927 ± 10). All Pb analyses were performed in temperature-controlled runs (1250°C) and corrected for mass fractionation by 0·47
per a.m.u., based on 10 replicate runs of SRM 981, which gave 208Pb/206Pb = 2·16475 ± 0·00071, 207Pb/206Pb = 0·914250 ± 0·000152 and 204Pb/206Pb = 0·059095 ± 0·000026.
|
PETROLOGICAL AND GEOCHEMICAL CHARACTERISTICS |
|---|
|
TOPABSTRACTINTRODUCTIONSAMPLING AND ANALYTICAL...PETROLOGICAL AND GEOCHEMICAL...PETROGENETIC MODELCAMP MANTLE SOURCES AND...DISCUSSION AND CONCLUSIONSREFERENCES |
|---|
The mineralogy of the MPD dolerites is typical of continental tholeiites, with plagioclase and clinopyroxene (augite and pigeonite) as major phases as well as minor olivine and titanomagnetite. In general it is difficult to observe if there are any textural variations from margin to centre as most outcrops are incomplete or heavily weathered. There are no noticeable petrological variations along the dyke and the only textural differences observed in better preserved outcrops are directly related to the thickness of the dyke (ranging from 5 to 75 m in Portugal to
200 m in Spain). In general, the MPD shows variations from microdoleritic texture at the margins to gabbroic textures at the centre whenever the thickness is >100 m (Bertrand & Millot, 1987; Martins, 1991; Martins et al., 1995). The centre of the dyke appears to be more differentiated than the margins, comprising 85% plagioclase + clinopyroxene (in similar proportions), and scarce olivine, Fe–Ti oxides and apatite as well as biotite and amphibole as accessory minerals. Clinopyroxene follows the variations usually found in tholeiitic intrusions mainly represented by augite (Ca37Mg54-Fe9–Ca39Mg14Fe47), with pigeonite (Ca9Mg61Fe30–Ca14Mg61Fe25) being present in lower proportions and usually as an interstitial phase. Plagioclase ranges from An77 to An68, reaching values of An20 when present as an interstitial phase. Olivine appears in low proportion in the more differentiated liquids and is usually Fo <74%. Quartz–feldspar granophyric intergrowths are also present usually within the central part of the dyke. The Fe–Ti oxides belong to the ilmenite–haematite and ulvöspinel–magnetite series.
The major element compositions of the MPD samples analysed (Table 1) are characteristic of relatively differentiated continental tholeiites with high contents of SiO2, Al2O3 and CaO and low P2O5 and TiO2, resembling the composition of the low-P2O5–TiO2 tholeiitic basalts of the southern part of the Paraná flood-basalt province (see, e.g. Fodor, 1987; Piccirillo et al., 1988). Additionally, the depletion in CaO, and enrichment in Al2O3, FeO, K2O, TiO2, Na2O and P2O5 with decreasing MgO (Fig. 3) are in agreement with typical trends produced by relatively small amounts of differentiation of a primary tholeiitic magma dominated by the extraction of clinopyroxene and olivine (see, e.g. Cox, 1980). However, as plagioclase megacrysts are a prominent feature of the MPD samples (estimations from thin sections indicate that they represent 40% of the mineral assemblage), plagioclase was clearly also likely to be a relatively important fractionating phase (e.g. Bertrand & Millot, 1987; Martins, 1991). It is possible that the high Al2O3 contents of the more SiO2-rich, MgO-poor samples (which argue against plagioclase as a significant fractionating phase) could be the result of a process superimposed upon simple fractional crystallization. As we will show below, assimilation of Al2O3-rich crustal rocks could account for this Al2O3 input as well as for the higher scattering in Al2O3 when compared with other elements such as TiO2 or CaO.
|
) are included for comparison. Although the data of Bertrand & Millot (1987)Trace element variations relative to Th (considered as a highly incompatible element, Fig. 4) are also in agreement with the petrological observations and major element trends, such as the depletion in Ni (olivine fractionation) and Sc (clinopyroxene fractionation). Sr or Eu increase in concentration in the more differentiated melts, which fits with the Al2O3 trend (Fig. 3), is also in contradiction with a simple fractional crystallization process involving significant plagioclase.
|
Primitive mantle-normalized trace element variation diagrams (Fig. 5) show similar patterns for all the MPD samples, characterized by relative enrichments in large ion lithophile elements (LILE) with a strong positive spike in Rb and negative troughs in Ta, P and Ti, whereas Sr shows a rather uniform normalized concentration clustered around SrN = 9. This pattern is similar to those of other contemporaneous dolerites belonging to the CAMP such as Foum-Zguid (Morocco) or Shelburne (Nova Scotia, Canada) and even at more distant locations such as Guinea (Fig. 6a). Although most CAMP tholeiites share this trace element signature, there are also some compositions within the province comprising both tholeiites and alkaline olivine dolerites that show a contrasting trace element pattern characterized by higher Nb–Ta, low Rb and variable Sr contents (Fig. 6b).
|
|
Sr–Nd–Pb isotopes
The Sr–Nd–Pb isotopic compositions of the MPD samples (at 200 Ma) are presented in Table 1 and Figs 7 and 8. Initial Sr and Nd isotopic compositions are located in the bottom right-hand quadrant of the 87Sr/86Sr–143Nd/144Nd diagram (Fig. 7), close to the present-day EM1 mantle component (Hart et al., 1992
), defining a linear array towards more radiogenic Sr isotope compositions. This trend is related to the degree of differentiation as shown in Fig. 9 (using MgO as a differentiation index), suggesting contamination by a more radiogenic component. Despite their apparent similarity in the normalized trace element diagrams, the CAMP magmatic rocks exhibit a wide range of Sr–Nd compositions and some dolerites that are usually considered as compositionally equivalent (e.g. the MPD and Shelburne dyke) show significant differences.
|
|
|
207Pb/204Pb and 208Pb/204Pb for the CAMP magmatic rocks correlate positively with 206Pb/204Pb (Fig. 8) defining a tight linear array subparallel to the Northern Hemisphere Reference Line (NHRL; Hart, 1984
) but displaced towards higher 207Pb/204Pb and 208Pb/204Pb compositions and showing relatively higher 207Pb/204Pb enrichments at lower 206Pb/204Pb ratios. The MPD data from this study show a relatively narrow range of 206Pb/204Pb, plotting slightly displaced towards the more radiogenic end of the CAMP array. Although the most radiogenic sample of the MPD in terms of Pb isotopes corresponds to the most differentiated rock (i.e. with lowest MgO), which could be interpreted as caused by crustal contamination, the remaining samples do not follow a clear variation of Pb isotopes relative to a differentitation index (e.g. MgO or Mg-number) and describe a narrow and poorly correlated field within analytical error, defined by 206Pb/204Pb = 18·45. If the available MPD Pb dataset is representative, this suggests that in general Pb isotopes are not strongly influenced by the contamination process inferred from the MgO–87Sr/86Sr variation. The general trend of the CAMP array in Fig. 8 is primarily defined by the Mesozoic Appalachian Tholeiites (MAT), which also lack a significant correlation between Mg-number values and Pb isotopes, and are interpreted to reflect partial melting of old enriched continental lithospheric mantle, with enrichment at 1000 Ma (Pegram, 1990). The only outliers from the main CAMP array are the Late Triassic–Early Jurassic lamprophyres and alkaline olivine dolerites from the Plymouth area (Nova Scotia, Canada), which have been attributed to the involvement of a plume-like mantle source (Pe-Piper & Reynolds, 2000). Clearly, the Pb isotope data arrays could also be explained in terms of two-component mixing, which will be considered subsequently. |
PETROGENETIC MODEL |
|---|
|
TOPABSTRACTINTRODUCTIONSAMPLING AND ANALYTICAL...PETROLOGICAL AND GEOCHEMICAL...PETROGENETIC MODELCAMP MANTLE SOURCES AND...DISCUSSION AND CONCLUSIONSREFERENCES |
|---|
The geochemical characteristics (Table 1) of the MPD dolerites (e.g. Ni <105 ppm, Cr <270 ppm, Mg-number <71) indicate that these magmas are not primary. The only sample with Mg-number >66 (sample 15), has abundances of Ni (105 ppm) and Cr (166 ppm) that, although relatively primitive, cannot be considered as characteristic of a primary magma.
Earlier studies by Alibert (1985), Bertrand & Millot (1987) and Martins et al. (1995) reported that differentiation took place mainly through fractional crystallization of plagioclase and clinopyroxene, and minor olivine and Ti-magnetite. However, in addition to the fractional crystallization process, which seems to have controlled some of the major and trace element variations, the enrichment in 87Sr/86Sr in the more evolved samples (see Fig. 9) suggests that contamination by a more radiogenic Sr component also contributed to the composition of these rocks. Coupling of the 87Sr/86Sr enrichment with a fractionation index such as MgO suggests that the most likely petrogenetic process that can explain the trace element and isotopic characteristics of the MPD dolerites is assimilation plus fractional crystallization (AFC). In this sense, it is also conceivable that some of the distinctive features of the MPD compositional variations (e.g. Al2O3 and Sr enrichment with progressive differentiation, despite the presence of significant amounts of plagioclase megacrysts) may be a consequence of the contamination process. If this inference is correct, then the apparent buffering of Sr contents observed in the mantle-normalized trace element variation diagrams would suggest that the Sr enrichment as a result of crustal assimilation nearly equals the depletion produced by the crystallization of plagioclase.
Constraints on the crustal contamination parameters
The relatively small variation in the Sr–Nd–Pb isotopic compositions of the MPD tholeiites suggests that the amount of assimilated material was not very significant and/or that the composition of the contaminant was not isotopically very different from that of the primitive magmas. The isotopic composition of the crustal contaminant can be partially constrained by the most 87Sr-enriched and 143Nd-depleted sample in the MPD series (i.e. 87Sr/86Sr >0·707 and 143Nd/144Nd <0·5119). Although the available data on the composition of the crust in the Iberian Peninsula are scarce, Villaseca et al. (1999) have provided valuable information on the nature and composition of the continental crust in Central Spain in a sector that is actually cut by the MPD. On the basis of a study of crustal xenoliths entrained within an Early Mesozoic alkaline ultramafic dyke swarm, Villaseca et al. (1999) have demonstrated that the lower crust (20–30 km depth) in this area is dominantly composed of fertile felsic metaigneous granulites, which could act as contaminants of the primitive MPD magmas. The composition of these granulites can then be used to model the crustal contamination process.
To test if the MPD rocks were indeed contaminated during their passage through the crust we have followed the classical formulation by DePaolo (1981) for modelling AFC processes. For calculation purposes we have considered only the Sr–Nd system, as Pb data are not available for the fertile granulite xenoliths used as a proxy for the composition of the lower crust. In addition to the composition of the primitive magma and the contaminant, to perform the AFC calculations two other important parameters must be estimated: the ratio of the mass of material assimilated to the mass of crystals fractionated (r) and the bulk distribution coefficients (D) for Sr and Nd.
As a first approximation to the actual values, DSr and DNd can be calculated from major-element-based mass-balance calculations, assuming that plagioclase and clinopyroxene are the only phases that can fractionate significant amounts of Sr and Nd and appropriate plagioclase–liquid and cpx–liquid distribution coefficients. For this calculation we adopted the mass-balance results by Bertrand & Millot (1987), which suggest residual liquid values (F) ranging from 0·72 to 0·35 with plagioclase and cpx fractionating at 13–30% and 14–29%, respectively. These ranges are in good agreement with mass-balance calculations based on our own data (see Table 3), which suggest a value of F = 0·54, with 22% of fractionated plagioclase and 19% of cpx. If we consider DPlg–liqSr = 1·639 and 
(Bindeman et al., 1998) and 
and 
(Cawthorn, 1996), then the bulk DSr = 0·220–0·506 and DNd = 0·015–0·033. However, it must be stressed that the values obtained this way, which overlook the effects caused by crustal assimilation and assume that one of the most MgO-rich samples represents the parental liquid, can only be considered as approximate estimates.
|
Although the available data cannot be used to constrain the values of r, we use as a reasonable approximation the range of values (r
0·3–0·5) suggested by Grove & Baker (1984) on the basis of thermal considerations for basaltic magmas assimilating crustal rocks. To estimate the amount of contamination and the geochemical characteristics of the contaminant we also need to estimate the composition of the starting magma. In the following sections we present the main results obtained from the calculation of AFC models involving the interaction of calculated MPD primitive magmas with the continental crust and test other possibilities that arise from the geodynamic models proposed for the CAMP magmatism.
Calculation of primary MPD magmas
As the MPD rocks show good correlations between both Sr and Nd content and their respective isotopic ratios (Fig. 10), a way to estimate the primary magma composition is to assume that it should be located along the linear arrays. If we consider the F value obtained from major element mass-balance calculations for the least differentiated liquids (F = 0·72; Bertrand & Millot, 1987), the corresponding calculated bulk distribution coefficients for Sr (0·220) and Nd (0·015), and the concentrations in the least contaminated sample (i.e. sample 16, with 150 ppm Sr and 10·6 ppm Nd), then for a simple fractional crystallization process 
and 
. From the linear regression in an Sr–(87Sr/86Sr)i diagram (Fig. 10), for C0Sr = 114 ppm, its 87Sr/86Sr ratio should be 0·7050. From this value, and assuming the linear regression between the initial Sr and Nd isotopic ratios [Fig. 7; (87Sr/86Sr)i = -0·09274 x (143Nd/144Nd)i + 0·57782, R = 0·615], 143Nd/144Nd should be 0·5124. Calculating this value from the linear fit in an Nd–(143Nd/144Nd)i diagram (Fig. 10) would produce a similar result, 0·51238, but in this case the correlation is low, R = 0·449, thus involving a larger error in the estimated values. This isotopic composition (87Sr/86Sr 0·705, 143Nd/144Nd 0·5124) is within the range assigned by Menzies (1990) to the Phanerozoic lithosphere, and strongly suggests the participation of a lithospheric mantle source component in the petrogenesis of the MPD rocks. Additionally, the similarity of this composition to that of the least differentiated samples and the relatively small variation observed in the MPD series suggest that the amount of crystallization is likely to be lower than previously estimated from major element calculations.
|
Calculation of the crustal contamination process
Considering the above calculated starting composition for the parental magma and testing AFC models with different crustal components [using actual compositions of granulite xenoliths as reported by Villaseca et al. (1999)
as contaminants (see Table 2)], we demonstrate that the only contaminants that appear to fit both the isotopic and trace element Sr–Nd data are the lower-crust metaigneous granulites. In the example shown in Fig. 11, the assimilation of granulite U-141 would require 30% crystallization (i.e. F 0·7) and an r value of 0·3. These results show that the assimilation of crustal rocks significantly reduces the amount of crystallization relative to the calculations performed from major element data considering a simple fractional crystallization process. The AFC results are in better agreement with the inference made above that the geochemical similarity of the MPD rocks suggests relatively small amounts of crystallization.
|
|
Although the main conclusions obtained from the numerical modelling are likely to be correct, this model should only be considered as an initial approximation until further data are available. For example, the use of Pb isotopes is at present hindered as very few data are available, especially concerning the contaminants and the plume-derived dolerites, and most data cannot be age-corrected to the time of magma emplacement. Nevertheless, the geochemical characteristics of the components involved in the contamination process can explain the absence of significant Pb isotopic variations induced by crustal assimilation. As the lower-crustal felsic granulites of the Spanish Central System are likely to be strongly Pb depleted owing to 29–33% extraction of granitic melts (Villaseca et al., 1999
), it is probable that contamination by these granulites cannot produce significant effects on the isotopic composition of relatively Pb-rich magmas derived from enriched mantle lithosphere sources. To summarize, the numerical modelling suggests that the primitive MPD magmas can be the result of partial melting of an enriched lithospheric mantle source. These parental magmas underwent <30% crystallization, which was coupled to the assimilation of small volumes (under 10%) of lower-crustal metaigneous granulites.
Could the primitive MPD magmas beplume derived?
Although the above calculations can explain the observed variation in Sr–Nd trace element and isotopic ratios, another possibility that should be explored is that the primitive MPD magmas could be derived from a plume-type source. As we have seen, it is generally agreed (e.g. Schermerhorn et al., 1978; Dunn et al., 1998; Marzoli et al., 1999) that the MPD tholeiites are an expression of the CAMP magmatism that is believed to be related to the earliest stages of the opening of the Central Atlantic and that may be plume related (Ernst et al., 1995; Oyarzun et al., 1997; Wilson, 1997). If the MPD primitive magmas originated directly from the so-called Central Atlantic Plume (CAP), they should inherit the CAP isotopic composition. The presence of a plume-related mantle source component in the CAMP is supported by recently published data for the oldest (160–120 Ma) Atlantic oceanic crust ( Janney & Castillo, 2001), which suggest the involvement of a plume-type mantle component during the early stages of Central Atlantic opening. Similarly, the primitive olivine dolerites and lamprophyre dykes from the Plymouth area (Nova Scotia, Canada) also suggest the presence of plume sources in the CAMP (Pe-Piper & Reynolds, 2000). Furthermore, as we have shown in Fig. 6b, the trace element signatures of some West African (Dupuy et al., 1988) and South American (Oliveira et al., 1990) CAMP basalts also support the participation of an ocean-island basalt (OIB)-type mantle source in this realm. Unfortunately, the available isotopic data are scarce (especially for Pb) but it is noticeable that the Sr–Nd ratios of those CAMP samples with OIB-like trace element signatures (Fig. 6b) are also the closest to the present-day HIMU mantle end-member composition (Fig. 7).
According to Oyarzun et al. (1997), the present sublithospheric thermal anomaly identified by Hoernle et al. (1995), which extends from the Eastern Atlantic to Western–Central Europe, may represent NE-channelled CAP material. The geochemical composition of this anomaly is also relatively well constrained and has been termed Low Velocity Component (LVC, Hoernle et al., 1995) or European Asthenospheric Reservoir (EAR, Cebriá & Wilson, 1995). We therefore could use the age-corrected (200 Ma) isotopic composition of primitive LVC or EAR melts as a proxy for the possible CAP plume starting composition. In contrast, the alkaline affinity of the unusual CAMP primitive olivine dolerites and lamprophyre dykes from the Plymouth area (Nova Scotia, Canada; Pe-Piper & Reynolds, 2000) disallows their use as a primitive starting composition of a contaminated tholeiitic suite. Finally, the pre-120 Ma mid-ocean ridge basalt (MORB) data of Janney & Castillo (2001) cannot be used to constrain the CAP starting composition, as these MORBs are 50 Myr younger than the MPD tholeiites, they are not primary magmas and their composition is likely to be the result of the variable addition of a plume-type component to a depleted N-MORB mantle source.
Although, according to the geodynamic model of Oyarzun et al. (1997), the present isotopic composition of the LVC or EAR calculated back to 200 Ma could be used as a proxy for the possible CAP starting magmas, the Cenozoic undersaturated alkali basalts that are assumed to derive from this reservoir cannot be considered as suitable primitive magmas for the tholeiitic MPD series. For calculation purposes we have assumed the trace element composition of a typical primitive plume-derived tholeiite, such as that of the Loihi Seamount [Rb 5·8 ppm, Sr 300 ppm, Sm 4·06 ppm and Nd 15·5 ppm, which corresponds to sample 29-10 of Frey & Clague (1983), with Ni 270 ppm]. This composition is very similar to the one that can be calculated adopting the melting model proposed by Cebriá & López-Ruiz (1996) for the LVC- or EAR-derived melts of Calatrava and assuming a degree of melting of 35% (Sr 314 ppm, Sm 3·94 ppm, Nd 18·65 ppm), which is slightly over the maximum estimated for the generation of primitive continental flood basalts (e.g. 10–30% for the Deccan traps; Sen, 2001).
As can be observed in Fig. 11, the assumed CAP-derived primitive tholeiitic magma is unlikely to produce liquids with the Sr–Nd geochemical characteristics observed in the MPD by assimilation of continental crust, as AFC curves require a improbable combination of both low F values (<0·4) and high r values (>0·5) to fit the data. Additionally, although under these unrealistic conditions it is possible to reproduce the observed Sr–Nd isotope variation, the calculated trace element variation completely fails to fit the data. These results confirm that the parent liquid of the MPD cannot be represented by the proposed sublithospheric magmas.
|
CAMP MANTLE SOURCES AND SOME GEODYNAMIC IMPLICATIONS |
|---|
|
TOPABSTRACTINTRODUCTIONSAMPLING AND ANALYTICAL...PETROLOGICAL AND GEOCHEMICAL...PETROGENETIC MODELCAMP MANTLE SOURCES AND...DISCUSSION AND CONCLUSIONSREFERENCES |
|---|
One of the key questions concerning the MPD and in general the CAMP giant dyke swarm is to determine if the dykes are the result of lateral injection of magmas radiating from a CAP or if they originate from other mantle sources. According to Ernst et al. (1995)
, the giant radiating dyke swarm of the Central Atlantic is linked to the formation and evolution of the CAP, and the major dykes, which are among the more distant ones, would correspond to lateral injection of magmas flowing away from the plume head. However, although dyke emplacement models confirm that magmas can travel laterally long distances (>3000 km) from their focal point [see also Elliot et al. (1999)], the geochemical differences between individual dykes suggest that in general they are not fed from a common source.
As we have seen, the geochemical homogeneity of the MPD relative to other tholeiites in the CAMP (e.g. the Mesozoic Appalachian tholeiites; Pegram, 1990) and our geochemical modelling argue against the possibility of the MPD resulting from the lateral injection of CAP-derived magmas and require a local 87Sr-enriched lithospheric mantle source and subsequent lower-crustal contamination. Although crustal contamination of enriched mantle lithosphere-derived magmas may be valid for other geochemically similar CAMP tholeiites, it is difficult to assess if this model is predominant in this realm. At a first glance, the scarcity of plume-like trace element and Sr–Nd–Pb compositions (Fig. 6b and 7) and the predominance of 87Sr-enriched compositions such as in the MPD would suggest that the mantle lithosphere is the main source of CAMP magmas. However, as most of the CAMP tholeiites are not primitive and the most abundant compositions are intermediate between the inferred composition of the CAP and crustal-like compositions (Fig. 7), those dolerites might also be the result of contamination of CAP-derived melts by a crustal component. It is thus suggested that, in each case, numerical approaches such as the one presented here should be applied to obtain useful constraints that help to discern between the possibilities.
Another important point on which geochemistry can shed some light is the compositional similarity observed in the northernmost representatives of the regional tholeiitic dykes such as MPD, Shelburne and Foum-Zguid. As mentioned above, the similarity in trace element patterns for these dykes has led some workers to suggest either that they might represent different manifestations of the same intrusion (e.g. Bertrand & Millot, 1987) or that they are magmas produced from similar mantle sources and processes related to similar geodynamic settings (Dunn et al., 1998). The MPD and Foum-Zguid dykes could indeed be part of the same intrusion as they seem to be coincident in the pre-drift reconstructions of the African and Iberian plates (see Fig. 1). Although these dykes have similar trace element patterns, the scarce (87Sr/86Sr)i data available for Foum-Zguid (0·706301–0·706942; Fiechtner et al., 1992) show that in general they are slightly more radiogenic than most MPD samples. If, as suggested by Ernst et al. (1995), magmatic differentiation and crustal assimilation processes occur before the emplacement of the dykes, it seems more likely that the MPD and Foum-Zguid dykes are the result of different batches of magma produced from similar lithospheric mantle sources that underwent analogous differentiation processes. This explanation can also be applied to other cases, such as the MPD and Shelburne dykes (see Dunn et al., 1998), which are obviously not directly linked in the pre-drift reconstructions as they are nearly parallel dykes located on different margins of the future Central Atlantic mid-ocean ridge.
The overall compositional similarity of the CAMP dolerites and basalts has been used to support the involvement of a common mantle source (Marzoli et al., 1999). These interpretations have serious implications as they require magmas to remain unmodified during their travel over distances >1500 km. However, the inferences made for many of these dolerites when considered individually imply the contamination of the original magmas by a crustal component (e.g. Shelburne dyke, Dunn et al., 1998; Foum Zguid dyke, Bertrand et al., 1982). Even if the original magmas were generated from a common source, to remain nearly identical after contamination, the crustal component should also be very similar in all cases (at least for trace element contents), thus implying the unlikely presence of a homogeneous crust over great distances and diverse geological settings.
|
DISCUSSION AND CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONSAMPLING AND ANALYTICAL...PETROLOGICAL AND GEOCHEMICAL...PETROGENETIC MODELCAMP MANTLE SOURCES AND...DISCUSSION AND CONCLUSIONSREFERENCES |
|---|
The participation of a mantle plume as the main cause of the CAMP is a matter of intense debate and there are many arguments that can be put forward to either support (e.g. Ernst et al., 1995
; Wilson, 1997) or reject (e.g. Bertrand, 1991; McHone, 2000) the participation of a Central Atlantic Plume (CAP) in the petrogenesis of the primary magmas.
Comparison of the normalized trace element patterns for CAMP magmatic rocks allows us to distinguish two main groups (Fig. 6): (1) rarer tholeiites and alkaline magmas characterized by positive anomalies in Nb–Ta, low Rb contents (usually showing negative spikes) and variable Sr contents, which might suggest derivation from a plume-like mantle source [e.g. some Liberian dolerites (Dupuy et al., 1988), Brazilian basalts (Bellieni et al., 1990) and the alkaline lamprophyres and olivine dolerites from the Plymouth area of Canada (Pe-Piper & Reynolds, 2000)]; (2) more numerous tholeiites characterized by a negative Nb–Ta trough, positive Rb anomalies and a restricted Sr concentration [e.g. the MPD; some of the MAT (Pegram, 1990); etc.], which are usually interpreted as derived from mantle lithosphere sources. Similarly, but from a broader point of view, Puffer (2001) has also proposed distinguishing between plume-derived continental flood basalts (P-CFBs) and arc-type CFBs (A-CFBs), which can be assigned respectively to the two groups described above for the CAMP. In addition to the characteristics based on normalized trace element diagrams, Puffer (2001) also pointed out that P-CFBs exhibit highly variable contents of MgO and TiO2 whereas A-CFBs have a rather constant MgO–TiO2 composition (MgO 5–8 wt % and TiO2 0·75–1·5 wt % for 90% of the samples considered). This is a geochemical characteristic that can also be observed in the two groups of CAMP magmas.
As shown in Figs 7 and 8, as well as in other compilations [see fig. 2 of Janney & Castillo (2001)], the isotopic signature of the CAMP tholeiites shows a substantial heterogeneity, with a full compositional range from OIB-like ratios in the 87Sr/86Sr–143Nd/144Nd diagram to highly radiogenic compositions. However, the distinction between plume derived and lithosphere derived as suggested by trace element data is not readily observed. This suggests that there is not a unique petrogenetic process that can explain the whole CAMP tholeiitic magmatism.
The simplest way to interpret the trace element data and the linear relationships observed in Sr–Nd–Pb isotopic space is to invoke a binary mixture between a plume-type component and the lithosphere. However, it is likely that the linear relationship between Pb isotopes is mainly the consequence of an old enrichment event of the mantle lithosphere, and crustal contamination may have also been a common process in the CAMP. This can be observed in the (143Nd/144Nd)i–206Pb/204Pb diagram (Fig. 12), where the distribution of the CAMP samples may be explained in terms of a single vector defined by the interaction between a plume component and a lithospheric 206Pb–143Nd-depleted component. However, this distribution can also result from melting of heterogeneous mantle lithosphere sources and additional 143Nd depletion induced by crustal contamination.
|
If we apply the conclusions of the geochemical modelling presented above for the MPD to other similar CAMP tholeiites, it would seem that most of these Jurassic dolerites are likely to be the result of partial melting of enriched lithospheric mantle sources, thus apparently making unnecessary the participation of a mantle plume. However, as we have seen, there are a small number of samples (e.g. some Liberian and Brazilian Amapa–Jari tholeiitic dykes and the alkaline olivine dykes of Pymouth in Canada) whose geochemical signature cannot be accounted for by this model and require the presence of a plume-like mantle source. Additionally, although the dyke patterns appear to be complex on a local scale (e.g. Bertrand, 1991
; Marzoli et al., 1999; Hames et al., 2000) the general dyke pattern as originally described by May (1971) suggests a focus zone located somewhere close to the junction between North America, Africa and South America in pre-drift reconstructions. Seismic images of the Atlantic margin of North America (Lizarralde & Holbrook, 1997) confirm that the distribution of the dykes at depth also corresponds to a radial stress field imposed from a focal point located under the lithosphere.
Another consequence of the geochemical modelling proposed for the MPD is that primitive magmas are modified by crustal contamination. If this is a common process in the CAMP, many plume-derived magmas could develop a lithosphere signature by contamination whereas the lithosphere-derived magmas would acquire more radiogenic compositions. This would explain in part the prevalence of magmas with lithospheric signature and the scarcity of plume-type magmas. In addition to the possibility that some CAP-derived magmas could develop a lithospheric signature by crustal contamination, the inference that most of the outcropping CAMP magmas are not plume related may also be the result of the special characteristics of the CAP. According to the numerical simulations of Leitch et al. (1998) for the Atlantic margin of North America, the impingement of a mantle plume in this realm requires that its geometry is characterized by a wide, thin and flat head that was detached from its tail. These characteristics imply that as the isolated plume head continued to spread, its potential temperature was quickly reduced. Lizarralde & Holbrook (1997) also reached a similar conclusion on the basis of subsidence calculations, also suggesting that the initial distribution of hot material in the upper mantle in this area was laterally extensive but not deep and that it cooled very quickly. Additionally, those workers pointed out the fact that under these conditions a plume head is likely to be rapidly channelled and to flow laterally following the relief at the base of the lithosphere, which in the case of pre-drift areas is usually thinner lithosphere acting as a ‘thin-spot’.
Considering these characteristics, it seems reasonable that the main effects of the CAP on the Central Atlantic lithosphere were (Fig. 13): (1) the observed radial stress field; (2) a spreading or channelling of the detached plume head along preferential structural zones of thinned lithosphere (see Lizarralde & Holbrook, 1997; Oyarzun et al., 1997; Wilson, 1997); (3) a generalized rise in the temperature of the overlying lithosphere, which would favour localized melting of enriched portions and the injection of the magmas along fracture zones induced by the localized stress field. This scenario is also in agreement with the geodynamic model proposed by Oyarzun et al. (1997) and Cebriá et al. (2000), which suggests that the present-day manifestation of the CAP is a sheet-like geochemical and thermal anomaly identified by seismic tomography (Hoernle et al., 1995), which, due to its present relatively low potential temperature, can produce magmas only under favourable geodynamic conditions (e.g. the extensional regime prevailing in the Tertiary–Quaternary European extension-related magmatic province).
|
This interpretation also allows us to explain the observation that the CAMP, although recently considered as an LIP (Marzoli et al., 1999
; Olsen, 1999), differs from other classical examples such as the Paraná or the Deccan (see Coffin & Eldholm, 1994), as it is mainly constituted by dykes, with lava flows being very scarce. Although, similar to other LIPs (e.g. Hooper, 1990), it is usually assumed that the CAMP dykes represent the feeders of lava flows that have now been eroded away (e.g. McHone, 2000; Hesselbo et al., 2002), the evidence in favour of this idea is only circumstantial (see Olsen, 1999). In fact, despite the high erosion rates calculated for some of the exposed rift basins (e.g. Steckler et al., 1993), it seems unlikely that the assumed largest LIP on Earth has lost virtually all its flood basalt cover in all three drifted portions of the formerly welded continent, whereas presumably smaller LIPs (e.g. the 200–240 Ma Siberian traps; Zolotukhin & Al'Mukhamedov, 1988) have survived to the present. If we accept the geodynamic model proposed here, it seems more likely that in most cases the CAMP dolerites do not represent feeder dykes and that subaerial volcanism was much less voluminous than currently assumed. This inference implies a further argument against the usually proposed relationship between the Triassic–Jurassic mass extinction and the presumed CAMP flood basalt volcanism. In addition to other observations, such as that the extinction event seems to take place before the CAMP magmatism (see Olsen, 1999; Olsen et al., 2002), the lack of massive flood basalt outpourings implies that this relationship should be reconsidered on the basis of alternative extinction mechanisms, as proposed by Tanner et al. (2001) or Olsen et al. (2002). If these conclusions are correct, we would propose the existence of two types of plume-related igneous provinces: (1) classical CFB LIPs (e.g. Siberian or Deccan traps), constituted primarily by plume-derived magmatic outflows with or without the addition of mantle lithosphere- or even crustally derived magmas, on top of a plume head swell; (2) non-CFB-producing LIPs (e.g. the CAMP), in which plume-induced phenomena are associated with the long-lived existence of a sublithospheric hot sheet resulting from the channelling of a detached plume head thousands of kilometres away from its original site of impingement.
To summarize, the general geodynamic model that we adopt here for the CAMP (Fig. 13) involves the arrival at the base of the lithosphere of a detached mantle plume head (the CAP) by the Late Triassic, which induced localized lithosphere uplift and a nearly radial stress field over a very large region. The CAP head was then preferentially channelled towards the NE, following structurally favourable paths (presumably of thinned lithosphere), and adopted a sheet-like elongated shape that would approximately correspond to the CAMP area. Although it seems that in most cases the CAP did not produce melts (or at least they did not reach upper-crustal levels), this anomalously hot material heated the overlying mantle lithosphere, thus triggering melting of old metasomatically enriched and easily melting domains and leading to the tholeiitic magmatism that intruded into the crust following the preferential distensive structures induced by the stress field. Before their emplacement at upper-crustal levels, the initial magmas underwent fractional crystallization coupled to the assimilation of small percentages of lower-crustal granulites.
|
ACKNOWLEDGEMENTS |
|---|
This work was financially supported by the Dirección General de Investigación Research Project PB98-0507. Special thanks are due to Bob Cliff, Jeff Rosenbaum and Rod Green from the Geochronology Laboratory of the School of Earth Sciences (University of Leeds, UK), for their help and support during the isotope work. We also thank Nick Rogers for the INAA work at Open University (UK). We also appreciate the constructive comments by Marjorie Wilson, Enzo Piccirillo, Peter Hooper and Andrea Marzoli, which greatly contributed to improve the original manuscript.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONSAMPLING AND ANALYTICAL...PETROLOGICAL AND GEOCHEMICAL...PETROGENETIC MODELCAMP MANTLE SOURCES AND...DISCUSSION AND CONCLUSIONSREFERENCES |
|---|
Alibert, C. (1985). A Sr–Nd isotope and REE study of late Triassic dolerites from the Pyrenees (France) and the Messejana Dyke (Spain and Portugal). Earth and Planetary Science Letters 73, 81–90.[CrossRef][Web of Science]
Bellieni, G., Piccirillo, E. M., Cavazzini, G., Petrini, R., Comin-Chiaramonti, P., Nardy, A. J. R., Civetta, L., Melfi, A. J. & Zantedeschi, P. (1990). Low- and high-TiO2 Mesozoic tholeiitic magmatism of the Maranhao basin (NE-Brazil): K/Ar age, geochemistry, petrology, isotope characteristics and relationships with Mesozoic low- and high-TiO2 flood basalts of the Paranà basin (SE-Brazil). Neues Jahrbuch für Mineralogie, Abhandlungen 162, 1–33.[Web of Science]
Bertrand, H. (1991). The Mesozoic tholeiitic province of Northwest Africa: a volcano-tectonic record of the early opening of Central Atlantic. In: Kampunzu, A. B. & Lubala, R. T. (eds) Magmatism in Extensional Structural Settings. Berlin: Springer, pp. 147–188.
Bertrand, H. & Millot, G. (1987). Le magmatisme tholéitique continental de la marge Ibérique, précurseur de l'overture de l'Atlantique Central: les dolérites du dyke de Messejana–Plasencia (Portugal–Espagne). Comptes Rendus de l'Académie des Sciences 304, 215–220.
Bertrand, H. & Villeneuve, M. (1989). Témoins de l'ouverture de l'Atlantique Central au début du Jurassique: les dolérites tholéiitiques continentales de Guinée (Afrique de l'Ouest). Comptes Rendus de l'Académie des Sciences 308, 93–98.
Bertrand, H., Dostal, J. & Dupuy, C. (1982). Geochemistry of Early Mesozoic tholeiites from Morocco. Earth and Planetary Science Letters 58, 225–239.[CrossRef][Web of Science]
Bindeman, I. N., Davis, A. M. & Drake, M. J. (1998). Ion microprobe study of plagioclase–basalt partition experiments at natural concentration levels of trace elements. Geochimica et Cosmochimica Acta 62, 1175–1193.[CrossRef][Web of Science]
Cawthorn, R. G. (1996). Models for incompatible trace-element abundances in cumulus minerals and their application to plagioclase and pyroxenes in the Bushveld Complex. Contributions to Mineralogy and Petrology 123, 109–115.[CrossRef][Web of Science]
Cebriá, J. M. & López-Ruiz, J. (1996). A refined method for trace element modelling of nonmodal batch partial melting processes: the Cenozoic continental volcanism of Calatrava, Central Spain. Geochimica et Cosmochimica Acta 60, 1355–1366.[CrossRef][Web of Science]
Cebriá, J. M. & Wilson, M. (1995). Cenozoic mafic magmatism in Western/Central Europe: a common European asthenospheric reservoir? Terra Nova Abstracts Supplement 7, 162.
Cebriá, J. M., López-Ruiz, J., Doblas, M., Oyarzun, J., Hertogen, J. & Benito, R. (2000). Geochemistry of the Quaternary alkali basalts of Garrotxa (NE Volcanic Province, Spain): a case of double enrichment of the mantle lithosphere. Journal of Volcanology and Geothermal Research 102, 217–235.[CrossRef][Web of Science]
Coffin, M. F. & Eldholm, O. (1994). Large igneous provinces: crustal structure, dimensions, and external consequences. Reviews of Geophysics 32, 1–36.[CrossRef][Web of Science]
Cox, K. G. (1980). A model for flood basalt volcanism. Journal of Petrology 21, 629–650.
Dalziel, I. W. D., Lawver, L. A. & Murphy, J. B. (2000). Plumes, orogenesis, and supercontinental fragmentation. Earth and Planetary Science Letters 178, 1–11.[CrossRef][Web of Science]
Deckart, K., Féraud, G. & Bertrand, H. (1997). Age of Jurassic continental tholeiites of French Guyana, Surinam and Guinea: implications for the initial opening of the Central Atlantic Ocean. Earth and Planetary Science Letters 150, 205–220.[CrossRef][Web of Science]
DePaolo, D. J. (1981). Trace element and isotopic effects of combined wallrock assimilation and fractional crystallization. Earth and Planetary Science Letters 53, 189–202.[CrossRef][Web of Science]
Dostal, J. & Durning, M. (1998). Geochemical constraints on the origin and evolution of early Mesozoic dikes in Atlantic Canada. European Journal of Mineralogy 10, 79–93.
Dunn, A. M., Reynolds, P. H., Clarke, D. B. & Ugidos, J. M. (1998). A comparison of the age and composition of the Shelburne dyke, Nova Scotia, and the Messejana dyke, Spain. Canadian Journal of Earth Sciences 35, 1110–1115.[CrossRef][Web of Science]
Dupuy, C. & Dostal, J. (1984). Trace element geochemistry of some continental tholeiites. Earth and Planetary Science Letters 67, 61–69.[CrossRef][Web of Science]
Dupuy, C., Marsh, J., Dostal, J., Michard, A. & Testa, S. (1988). Asthenospheric and lithospheric sources for Mesozoic dolerites from Liberia (Africa): trace element and isotopic evidence. Earth and Planetary Science Letters 87, 100–110.[CrossRef][Web of Science]
Elliot, D. H., Fleming, T. H., Kyle, P. R. & Foland, K. A. (1999). Long-distance transport of magmas in the Jurassic Ferrar Large Igneous Province, Antarctica. Earth and Planetary Science Letters 167, 89–104.[CrossRef][Web of Science]
Ernst, R. E., Head, J. W., Parfitt, E., Grosfils, E. & Wilson, L. (1995). Giant radiating dyke swarms on Earth and Venus. Earth-Science Reviews 39, 1–58.
Ferreira, M. & Portugal, C. A. R. (1977). Actividade basáltica Pérmico–Liássica no território português. Uma achega para a datação. Publicacoes do Museu e Laboratorio Mineralogico e Geologico da Universidade de Coimbra 83, 39–52.
Fiechtner, L., Friedrichsen, H. & Hammerschmidt, K. (1992). Geochemistry and geochronology of Early Mesozoic tholeiites from Central Morocco. Geologische Rundschau 81, 45–62.
Fodor, R. V. (1987). Low- and high-TiO2 flood basalts of southern Brazil: origin from picritic parentage and a common mantle source. Earth and Planetary Science Letters 84, 423–430.[CrossRef][Web of Science]
Frey, F. A. & Clague, D. A. (1983). Geochemistry of diverse basalt types from Loihi Seamount: petrogenetic implications. Earth and Planetary Science Letters 66, 337–355.[CrossRef][Web of Science]
Greenough, J. D. & Hodych, J. D. (1990). Evidence for lateral magma injection in the Early Mesozoic dykes of eastern North America. In: Parker, R. J., Rickwood, P. C. & Tucker, D. H. (eds) Mafic Dykes and Emplacement Mechanisms. Rotterdam: Balkema, pp. 35–46.
Grove, T. L. & Baker, M. B. (1984). Phase equilibrium controls on the tholeiitic vs calc-alkaline differentiation trends. Journal of Geophysical Research 89, 3253–3274.
Hames, W. E., Renne, P. R. & Ruppel, C. (2000). New evidence for geologically instantaneous emplacement of earliest Jurassic Central Atlantic magmatic province basalts on the North American margin. Geology 28, 859–862.
Hart, S. R. (1984). A large-scale isotope anomaly in the Southern Hemisphere mantle. Nature 309, 753–757.[CrossRef]
Hart, S. R., Hauri, E. H., Oschmann, L. A. & Whitehead, J. A. (1992). Mantle plumes and entrainment: isotopic evidence. Science 256, 517–520.
Hesselbo, S. P., Robinson, S. A., Surlyk, F. & Piasecki, S. (2002). Terrestrial and marine extinction at the Triassic–Jurassic boundary synchronized with major carbon-cycle perturbation: a link to initiation of massive volcanism. Geology 30, 251–254.
Hoernle, K., Zhang, Y. S. & Graham, D. (1995). Seismic and geochemical evidence for large-scale mantle upwelling beneath the eastern Atlantic and western and central Europe. Nature 374, 34–39.[CrossRef]
Hooper, P. R. (1990). The timing of crustal extension and the eruption of continental flood basalts. Nature 345, 246–249.[CrossRef]
Janney, P. E. & Castillo, P. R. (2001). Geochemistry of the oldest Atlantic oceanic crust suggests mantle plume involvement in the early history of the central Atlantic Ocean. Earth and Planetary Science Letters 192, 291–302.[CrossRef][Web of Science]
Leitch, A. M., Davies, G. F. & White, M. (1998). A plume head melting under a rifting margin. Earth and Planetary Science Letters 161, 161–177.[CrossRef][Web of Science]
Lizarralde, D. & Holbrook, W. S. (1997). U.S. mid-Atlantic margin structure and early thermal evolution. Journal of Geophysical Research 102, 22855–22875.[CrossRef]
Martins, L. T. (1991). Actividade Ġgnea Mesozóica em Portugal (contribuição petrológica e geoquĠmica). Ph.D. thesis, Universidade de Lisboa, 418 pp.
Martins, L. T., Cebriá, J. M., Doblas, M., López-Ruiz, J. & Munhá, J. (1995). Petrologia e geoquĠmica do dique de Messejana. Memorias del Museu e Laboratorio Mineralogico e Geologico da Universidade de Porto 4, 747–750.
Marzoli, A., Renne, P. R., Piccirillo, E. M., Ernesto, M., Bellieni, G. & De Min, A. (1999). Extensive 200-million-year-old continental flood basalts of the Central Atlantic Magmatic Province. Science 284, 616–618.
May, P. R. (1971). Pattern of Triassic–Jurassic diabase dikes around the North Atlantic in the context of predrift position of the continents. Geological Society of America Bulletin 82, 1285–1292.
McDonough, W. F. & Sun, S. S. (1995). The composition of the Earth. Chemical Geology 120, 223–253.[CrossRef][Web of Science]
McHone, J. G. (2000). Non-plume magmatism and rifting during the opening of the central Atlantic Ocean. Tectonophysics 316, 287–296.[CrossRef][Web of Science]
Menzies, M. (1990). Archean, Proterozoic and Phanerozoic lithospheres. In: Menzies, M. (ed.) Continental Mantle. Oxford: Clarendon Press, pp. 67–86.
Oliveira, E. P., Tarney, J. & João, X. J. (1990). Geochemistry of the Mesozoic Amapá and Jari dyke swarms, northern Brazil: plume-related magmatism during the opening of the central Atlantic. In: Parker, R. J., Rickwood, P. C. & Tucker, D. H. (eds) Mafic Dykes and Emplacement Mechanisms. Rotterdam: Balkema, pp. 173–183.
Olsen, P. E. (1999). Giant lava flows, mass extinctions and mantle plumes. Science 284, 604–605.
Olsen, P. E., Kent, D. V., Sues, H. D., Koeberl, C., Huber, H., Montanari, A., Rainforth, E. C., Fowell, S. J., Szajna, M. J. & Hartline, B. W. (2002). Ascent of dinosaurs linked to an iridium anomaly at the Triassic–Jurassic boundary. Science 296, 1305–1307.
Oyarzun, R., Doblas, M., López-Ruiz, J. & Cebriá, J. M. (1997). Opening of the central Atlantic and asymmetric mantle upwelling phenomena: implications for long-lived magmatism in western North Africa and Europe. Geology 25, 727–730.
Pe-Piper, G. & Reynolds, P. H. (2000). Early Mesozoic alkaline mafic dykes, southwestern Nova Scotia, Canada, and their bearing on Triassic–Jurassic magmatism. Canadian Mineralogist 38, 217–232.[CrossRef][Web of Science]
Pegram, W. J. (1990). Development of continental lithospheric mantle as reflected in the chemistry of the Mesozoic Appalachian tholeiites, U.S.A. Earth and Planetary Science Letters 97, 316–331.[CrossRef][Web of Science]
Piccirillo, E. M., Melfi, A. J., Comin-Chiaramonti, P., Bellieni, G., Ernesto, M., Marques, L. S., Nardy, A. J. R., Pacca, I. G., Roisenberg, A. & Stolfa, D. (1988). Continental flood volcanism from the Paraná Basin (Brazil). In: Macdougall, J. D. (ed.) Continental Flood Basalts. Dordrecht: Kluwer Academic, pp.195–238.
Potts, P. J., Williams-Thorpe, O., Issacs, M. C. & Wright, D. W. (1985). High precision instrumental neutron activation analysis of geological samples employing simultaneous counting with both planar and coaxial detectors. Chemical Geology 48, 145–155.[CrossRef][Web of Science]
Puffer, J. H. (2001). Contrasting high field strength element contents of continental flood basalts from plume vs reactivated-arc sources. Geology 29, 675–678.
Salters, J. M., Hart, S. R. & Pantó, Gy. (1988). Origin of Late Cenozoic volcanic rocks of the Carpathian Arc, Hungary. In: Royden, L. H. & Horvath, F. (eds) The Pannonian Basin. AAPG Memoir 45, 279–292.
Schermerhorn, L. J. G., Priem, H. N. A., Boelrijk, N. A. I. M., Hebeda, E. H., Verdurmen, E. A. Th. & Verschure, R. H. (1978). Age and origin of the Messejana dolerite fault–dike system (Portugal and Spain) in the light of the opening of the North Atlantic Ocean. Journal of Geology 86, 299–309.[Web of Science]
Schott, J. J., Montigny, R. & Thuizat, R. (1981). Paleomagnetism and potassium–argon age of the Messejana Dike (Portugal and Spain): angular limitation to the rotation of the Iberian Peninsula since the Middle Jurassic. Earth and Planetary Science Letters 53, 457–470.[CrossRef][Web of Science]
Sebai, A., Feraud, G., Bertrand, H. & Hanes, J. (1991). 40Ar/39Ar dating and geochemistry of tholeiitic magmatism related to the early opening of the Central Atlantic rift. Earth and Planetary Science Letters 104, 455–472.[CrossRef][Web of Science]
Sen, G. (2001). Generation of Deccan trap magmas. Proceedings of the Indian Academy of Sciences (Earth and Planetary Science) 110, 409–431.
Soares de Andrade, A. A. (1972). Sur l'âge Permien inférieur d'une intrusion doléritique à Portel, Alentejo (filon de Messejana). Revista da Faculdade de Ciências, Universidade de Lisboa 17, 237–242.
Steckler, M. S., Omar, G. I., Karner, G. D. & Kohn, B. P. (1993). Pattern of hydrothermal circulation within the Newark Basin from fission-track analysis. Geology 21, 735–738.
Sun, S. S. & McDonough, W. F. (1989). Chemical and isotopic systematics of oceanic basalts: implications for mantle composition of the Earth and mantle evolution. In: Saunders, A. D. & Norry, M. J. (eds) Magmatism in Ocean Basins. Geological Society, London, Special Publications 42, 313–345.
Tanner, L. H., Hubert, J. F., Coffey, B. P. & McInerney, D. P. (2001). Stability of atmospheric CO2 levels across the Triassic/Jurassic boundary. Nature 411, 675–677.[CrossRef][Medline]
Teixeira, C. & Torquato, J. R. (1975). Nouvelles données sur l'âge du grand filon doléritique de Vila do Bispo-Placência. Boletim da Sociedade Geologica de Portugal 19, 99–101.
Thompson, G. A. (1998). Deep mantle plumes and geoscience vision. GSA Today 8, 17–25.
Villaseca, C., Downes, H., Pin, C. & Barbero, L. (1999). Nature and composition of the lower continental crust in Central Spain and the granulite–granite linkage: inferences from granulitic xenoliths. Journal of Petrology 40, 1465–1496.[CrossRef][Web of Science]
Wilson, M. (1997). Thermal evolution of the Central Atlantic passive margins: continental break-up above a Mesozoic super-plume. Journal of the Geological Society, London 154, 491–495.
Yale, L. B. & Carpenter, S. J. (1998). Large igneous provinces and giant dike swarms: proxies for supercontinent cyclicity and mantle convection. Earth and Planetary Science Letters 163, 109–122.[CrossRef][Web of Science]
Zindler, A. & Hart, S. R. (1986). Chemical geodynamics. Annual Review of Earth and Planetary Sciences 14, 493–571.[CrossRef][Web of Science]
Zolotukhin, V. V. & Al'Mukhamedov, A. I. (1988). Traps of the Siberian platform. In: Macdougall, J. D. (ed.) Continental Flood Basalts. Dordrecht: Kluwer Academic, pp. 273–310.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J.-P. Burg, J.-L. Bodinier, T. Gerya, R.-M. Bedini, F. Boudier, J.-M. Dautria, V. Prikhodko, A. Efimov, E. Pupier, and J.-L. Balanec Translithospheric Mantle Diapirism: Geological Evidence and Numerical Modelling of the Kondyor Zoned Ultramafic Complex (Russian Far-East) J. Petrology, February 1, 2009; 50(2): 289 - 321. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Perini, J. M. Cebria, J. Lopez-Ruiz, and M. Doblas Carboniferous-Permian mafic magmatism in the Variscan belt of Spain and France: implications for mantle sources Geological Society, London, Special Publications, January 1, 2004; 223(1): 415 - 438. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||