Publicação

Petrogenesis and geochronology of the late cretaceous alkaline magmatism in the west iberiam margin

Detalhes bibliográficos
Resumo:	The Late Cretaceous alkaline magmatism is the most voluminous of the three cycles of magmatic activity that took place on the West Iberian Margin (WIM) during the Mesozoic. It was preceded by a tholeiitic cycle in the Early Jurassic (200-198 Ma) and by a transitional episode during the Late Jurassic-Early Cretaceous transition (147-142 Ma). The alkaline cycle took place in a post-rift setting, after oceanic break-up, and is contemporaneous of the opening of the Bay of Biscay and consequential 35º counterclockwise rotation of Iberia, as well as with the first phases of tectonic inversion in the Mesozoic Lusitanian and Algarve Basins, as a result of the onset of the convergence and collision between the Iberian, European and African plates. Onshore, this cycle includes the intrusive Sintra, Sines and Monchique igneous complexes, the extrusive Lisbon Volcanic Complex as well as several other minor intrusions scattered in the Lusitanian and Algarve basins, such as the Foz da Fonte sill, the Paço d’Ilhas sill, the Mafra Radial dyke complex, the Oeiras Cascais dykes and the lamprophyre-basanite suite of the littoral Algarve. It extends offshore where it is represented by the Madeira-Tore Rise and the Ormonde peak in the Gorringe Bank. In this study, the geochronology of the Late Cretaceous onshore occurrences was updated through the acquisition of new U-Pb, 40Ar/39Ar, K-Ar and Rb-Sr. The combination of these data with previously published ages allowed to better constrain the duration of this cycle (94-72 Ma), as well as for the identification of two separate magmatic pulses. The first pulse took place between 94 and 88 Ma and is restricted to the Foz da Fonte and Paço d’Ilhas sills, both of which occur in the Lusitanian basin, between N39º and N38º20'. This pulse is contemporaneous with the opening of the Bay of Biscay and consequent rotation of Iberia (118-80 Ma). The second pulse has a younger age (75-72 Ma,) wider geographical distribution, between the Lisbon region (≈ 39ºN) and the Algarve (≈ 37ºN), and occurred simultaneously with the initial stages of the Alpine orogeny in the WIM, which culminated in the formation of the Betics and Pyrenees as well as to the inversion of the Mesozoic basins. The age obtained for the Sintra granite (79.2 Ma ± 0.8), is equivalent within error to the oldest age obtained for the syenite (78.3 ± 1.9 Ma, Storetvedt et al., 1987) but older than the other K-Ar ages obtained for these rocks (76.4 ± 1.4 and 76.1 ± 1.1 Ma, Storetvedt et al., 1987) and for the gabbros (74.9 ± 1 Ma Storetvedt et al., 1987). The combination of the geochronological data with the observed field cross-cutting relations and geophysical data (gravimetry and AMS), appears to indicate that the Sintra granite corresponds to an older phase that was intruded by a more recent gabbro to syenite suite. Geochemical analysis of samples from both pulses revealed that these rocks derive from small degrees (1-6%) of partial melting of trace element enriched garnet peridotite sources, with fairly homogenous isotopic compositions reflecting low time-integrated Rb/Sr and high Sm/Nd and U/Pb ratios, which are compatible with a sub-lithospheric origin. However, there are clear differences in the trace element abundance patterns in rocks from the first and second pulses. Samples from the second pulse show negative K, Zr, Hf and sometimes even Ti anomalies, which are absent in the first pulse and that are probably related to a metasomatic event. The metasomatic agent is likely to correspond to a melt of carbonatitic to siliceous volatile (CO2+H2O) rich nature that promoted, for example the stability of a hydrated K rich phase such as amphibole. Amphibole is unstable at the temperatures that characterize the asthenosphere or mantle plumes and therefore the negative K anomalies should indicate interaction of the ascending sub-lithospheric magmas the metasomatized lithospheric mantle or melting of the latter. This metasomatic signature is more noticeable in the rocks which derived from smaller fractions of melting, becoming progressively diluted as melting degree increases. In terms of isotopic composition, no major differences are observed between the rocks of the second pulse, which show a clear metasomatic contribution, and the rocks of the first pulse which do not show such a signature. This indicates that the metasomatic agent probably originated from the same sublithospheric source which had earlier produced the magmas of the first pulse. The differences in trace element signatures between the two pulses created by this metasomatic episode can be related to their contemporaneous geodynamic context. During the first pulse, the rotation of Iberia created a stress regime favorable to quick magma ascent while from 80 Ma onwards, the onset of continental collision restricted the opening of fractures and magma ascent, favouring the interaction of the ascending magmas with the lithosphere, probably resulting in the metasomatism of the lithospheric mantle and the formation of amphibole during or previous to the second pulse. Uncontaminated samples show isotopic signatures (87Sr/86Sri 0.7030-0.7037; εNdi 5.7–3.7; 206Pb/204Pbi 19.564-19.20, 207Pb/204Pbi 15.609-15.580, 208Pb/204Pbi 39.245-39.00) that are distinct from the previous tholeiitic and transitional cycles and point towards a sublithospheric source, unlike the previous cycles, which show an important lithospheric contribution. These signatures are also very similar to the ones observed in the rocks derived from the melting of the Canaries mantle plume, which was located near the WIM during the Late Cretaceous, according to paleogeographic reconstitutions and paleomagnetic data, and is therefore considered the most likely source for the melts and metasomatic agents that were identified in this study. The large volume of magmatism that took place in the onshore and offshore WIM during the Late Cretaceous is also compatible with an origin from a deep seated hot mantle anomaly, such as a mantle plume. However, the differences between the volumes erupted onshore and offshore must be related to different lithospheric thicknesses in these two domains, since larger volumes of decompression melting are possible under the thinner oceanic and transitional lithospheres than under the thicker continental lithosphere. Also, the Iberian plate kinematics and pre-existing structure must have played a major role in controlling the timing and location of this magmatism, thus explaining the absence of a spatial age progression and the scattered location of the intrusive and extrusive features. Mineral chemistry data (e.g reverse zoning) indicates that magma mixing between liquids of basic/ultrabasic and intermediate natures took place at upper mantle/lower crustal depths but that does not seem to have greatly affected the composition of the resulting magmas. These magmas seem to have ascended to upper crustal levels by taking advantage of pre-existing fractures related with Mesozoic rifting and/or the Paleozoic Variscan orogeny. Many of these occurrences in the western Mesozoic basins seem to be related to WNW-ESE to E-W transfer faults that were active during rifting but, their exact location seems to be controlled by the intersection of these structures with other also previously existing fault systems with NE-SW and NNW-SSE orientations. After emplacement at upper crustal levels, the rocks from both pulses seem to have evolved through combined fractional crystallization of olivine, clinopyroxene, Fe-Ti oxides and more or less plagioclase and amphibole, with assimilation of crustal lithologies. Fractionation and accumulation of some of these minerals led to specific major and trace element characteristics of some of the intrusive rocks, such as the Sines and Sintra gabbros and the Paço d’Ilhas sill. Assimilation seems to have been an important factor in generating the observed petrological diversity and in creating the observed radiogenic Sr (87Sr/86Sri up to 0.7049), unradiogenic Nd (εNdi as low as 3.1) and variable Pb (206Pb/204Pbi and 208Pb/204Pbi decrease to 18.827 and 38.603, respectively, while 207Pb/204Pbi increases up to 15.634) signatures seen in the more evolved rocks. Two different contamination trends are observed and thus require the existence of two different contaminants. Increasing Th/La, Zr/Nb, 87Sr/86Sr, 207Pb/204Pb and K/Nb and decreasing Nb/U, 206Pb/204Pb and 208Pb/204Pb towards values closer to crustal compositions with increasing SiO2 are related to contamination by a siliceous upper crustal lithology. The other trend is characterized by only increasing 87Sr/86Sr without causing significant changes in other isotopic ratios and is probably related to a Sr rich contaminant that is simultaneously depleted in Nd and Pb, which is likely to be of carbonated nature.
Autores principais:	Miranda, Rui Miguel Leal, 1982-
Assunto:	Magmatismo Geocronologia Geoquímica Petrogénese Cretácico Superior Teses de doutoramento - 2010
Ano:	2010
País:	Portugal
Tipo de documento:	tese de doutoramento
Tipo de acesso:	acesso aberto
Instituição associada:	Universidade de Lisboa
Idioma:	inglês
Origem:	Repositório da Universidade de Lisboa