Search

During the Cenozoic widespread anorogenic magmatism, unrelated to recent supra-subduction zone modification of its mantle source, developed within the Mediterranean and surrounding regions; this is referred to collectively as the CiMACI (CircumMediterranean Anorogenic Cenozoic Igneous) province. On the basis of a comprehensive review of published and new major and trace element and Sr-Nd-Pb isotopic data (more than 7800 samples) for the magmatic rocks, a common sub-lithospheric mantle source component is identified for most of the region. This has geochemieal affinities to the source of HIMU oceanic island basalts and to the European Asthenospheric Reservoir (EAR) and the Low Velocity Component (LVC) of previous workers; we refer to this as the Common Mantle Reservoir (CMR). Global and local seismic tomography studies of the mantle beneath the CiMACI province have revealed a range of P- and Swave velocity anomalies, some of which have been related to the presence of mantle plumes. Detailed local tomography experiments in the Massif Central of France and the Eifel region of central Germany suggest that, locally, there are diapiric upwellings rooted within the upper mantle which induce adiabatic decompression melting and magma generation. These velocity anomalies can be interpreted as evidence of mantle temperatures up to 150 [degree]C hotter than the ambient mantle. However, the seismic attenuation could also be attributable to the presence of fluid or partial melt and significant thermal anomalies are not necessary to explain the petrogenesis of the magmas. A model is proposed in which the geochemical and isotopic characteristics of the sub-lithospheric mantle beneath the CiMACI province reflect the introduction of recycled crustal components (derived from both oceanic and continental lithosphere) into the ambient depleted upper mantle. This sub-lithospheric mantle is subsequently partially melted in a variety of geodynamic settings related to lithospheric extension, continental collision and orogenic collapse, and contemporaneous subduction, slab roll-back and slab-window formation. On the basis of this in-depth geochemical and petrological study of the CiMACI province, we consider that there is no need to invoke the involvement of anomalously hot mantle (i.e., the presence of a single or multiple deep mantle plumes) in the petrogenesis of the magmas. If, however, we adopt a more permissive definition of 'mantle plume', allowing it to encompass passive, diapiric upwellings of the upper mantle, then we can relate the CiMACI province magmatism to multiple upper mantle plumes upwelling at various times during the Cenozoic. To avoid confusion we recommend that such upper mantle plumes are referred to as diapiric instabilities. Keywords: Cenozoic; magmatism; Europe; mantle plumes; HIMU; basalt

The Earth's lithosphere can focus basaltic magmatism along pre-existing weakness zones or discontinuities. However, apart from the influence on the geochemistry of magmas emplaced in subduction tectonic settings (mantle wedge metasomatism related to dehydration of the subducting plates) the role of lithosphere as a magma source for intra-plate (both oceanic and continental), continental margin, and mid-ocean ridge magmatism is not yet fully understood. In many cases intra-plate magmatism has been explained with the existence of deep thermal anomalies (mantle plumes) whose origin has been placed near the upper-lower mantle transition zone (660 km discontinuity) or even deeper, near the mantle--core boundary (~2900 km). Also in many continental flood basalt provinces (mostly initiated at craton margins) an active role for mantle plumes has been invoked to explain the high melt productivity. In these cases, no active role for melt production has been attributed to the lithospheric mantle. Potential contaminations of asthenospheric or even deeper mantle melts are often considered the only influence of the lithosphere (both crust and mantle) in basalt petrogenesis. In other cases, an active role of the lithospheric mantle has been proposed: the thermal anomalies related to the presence of mantle plumes would trigger partial melting in the lithospheric mantle. At present there is no unequivocal geochemical tracer that reflects the relative role of lithosphere and upper/ lower mantle as magma sources. In this paper another role of the lithosphere is proposed. The new model presented here is based on the role of lower crustal and lithospheric mantle recycling by delamination and detachment. This process can explain at least some geochemical peculiarities of basaltic rocks found in large and small volume igneous provinces, as well as in mid-ocean ridge basalts. Metamorphic reactions occurring in the lower continental crust as a consequence of continent-continent can lead to a density increase (up to 3.8 g/[cm.sup.3]) with the appearance of garnet in the metamorphic assemblage (basalt [right arrow] amphibolite [right arrow] garnet clinopyroxenite/eclogite) leading to gravitative instability of the overthickened lithospheric keel (lower crust + lithospheric mantle). This may detach from the uppermost lithosphere and sink into the upper mantle. Accordingly, metasomatic reactions between Si[O.sub.2]-rich lower crust partial melts and the uprising asthenospheric mantle (replacing the volume formerly occupied by the sunken lithospheric mantle and the lower crust) lead to formation of orthopyroxene-rich layers with strong crustal signatures. Such metasomatized mantle volumes may remain untapped also for several Ma before being reactivated by geological processes. Partial melts of such sources would bear strong lower crustal signatures giving rise to Enriched Mantle type 1 (EMI)-like basaltic magmatism. Basaltic magmatism with such a geochemical signature is relatively scarce but in some cases (e.g., Indian Ocean) it can be a geographically widespread and long-lasting phenomenon. Keywords: delamination; detachment; lithosphere; mantle plume; petrology; lower crest: eclogite; pyroxenite