RICERCA ITALIANA     

Relationships between anorogenic and orogenic magma sources and geodynamic implications for the Europe-Africa system.

Università degli Studi di Firenze

Abstract

Petrological, geochemical and isotopic studies on lavas and hosted mantle xenoliths from Cenozoic key magmatic provinces, from subduction related, collisional and within plate geodynamic settings, are proposed. This research aim to contribute to the understanding of the compositional evolution of upper mantle sources of magmas in complex geodynamic settings. Origin of orogenic and anorogenic signatures will be therefore investigated to elucidate their dependence on possible processes of mantle metasomatism, condition of partial melting, and heritage by previous histories of mantle wedges. The acquired data will be discussed in the framework of the complex geodynamic evolution of the Central, Western Mediterranean Regions, which is driven by Africa-Eurasia plate convergence, and subsequent extension due to slab roll-back processes. The transition from orogenic to anorogenic magmatisms in the Central, Western Mediterranean Regions will be also addressed to better clarify relationships between mantle metasomatic processes and geodynamic evolution. On the other hand, the Madagascan volcanic areas will be investigated in order to understand the effects of late-Precambrian lithospheric resetting. Emphasis wil be given to the compositional and isotopic relationships between the mantle sources of the Madagascan alkaline volcanic rocks and those with similar serial affinities found in the Central- Western Mediterranean area, in order to distinguish possible hot-spot influence in both magmatic environments. The research will be complemented by structural, geological and paleomagnetic studies to finely define the geodynamic evolution of the segments studied and by detailed mineralogical and petrological studies of mantle xenoliths from Na-alkaline lavas from SE Spain, Italy and North Africa. <<<

Principal Investigator

Sandro CONTICELLI Università degli Studi di FIRENZE

Research Objectives

The goal of this research is to establish petrological, geochemical and isotopic characteristics of lithospheric and sub-lithospheric mantle sources belonging to the European and North-Africa domains, facing one another, with microplates in between, in the Mediterranean.
A complete comparative study between mafic lavas and hosted xenoliths from key localities in the Western Mediterranean and those from other typical magmatic associations related to European and African domains, will be also carried out. Because of the very complex geodynamic setting of the Cenozoic Western and Southern Mediterranean magmatic provinces and because of the existence of mutually excluding petrogenetic and geodynamic models proposed to explain the magmatic activity in this area, this project enlarges its interests to areas away from any subduction system in within-plate geodynamic systems such as Madagascar.
The aim of this study is to define the possible role of pristine lithospheric signatures which might be overprinted by both orogenic and anorogenic metasomatic events controlling the final petrological and geochemical features of generated magmas. In detail the proposed research project will be addressed to:
1) recognise the nature of mantle sources prior to metasomatism in both anorogenic and orogenic settings;
2) define the metasomatic components which affected the concerned mantle sections in both the European and African/Adriatic lithospheres;
3) constrain the existence and nature of mass transfer as fluxes between different geochemical reservoirs (i.e., asthenospheric mantle, lithospheric mantle and crust);
4) define the petrological and geochemical effects of lithospheric resetting in continetal within-plate areas in African domains (Madagascar), in contraposition to asthenospheric magma production in segments African and European domains from Western Mediterranean.

We will focus data acquisition, i.e. major and trace element abundances, and Sr, Nd, Pb, Hf, O and U-series isotopic compositions on i) calc-alkaline basalts, shoshonites and lamproites of Western Mediterranean magmatism (i.e., Tuscany, Corsica, Western Alps, SE Spain), ii) calc-alkaline basalts, shoshonites, and leucitites from Southern Italy, iii) calc-alkaline, tholeiitic, and Na-alkaline igneous rocks from Sardinia, iv) mantle xenoliths hosted by orogenic and anorogenic magmas of Western Mediterranean regions; v) alkaline and strongly alkaline mafic lavas of potassic and sodic affinity and associated mantle-derived xenoliths from Madagascar. Comparative data will be acquired from the typical orogenic magmas from European segments (Aegean, Serbia, and Macedonia).
These data should provide a basis for interpreting either lithospheric or sub-lithospheric provenance of magmas, the lithospheric control in producing structural discontinuities and loci of continental break-up, the geochemical and petrological signatures of magma sources and their thermobarometric histories of mantle melts, the effect of enrichment (crustal contamination) and depletion (extraction of melts) processes in the mantle wedge, and the potential role of deep-rooted mantle plumes vs. passive raise of asthenospheric mantle in back-arc settings.

Combination of petrological results with paleomagnetic and structural geological data will provide new insights for regional geodynamic models of the studied areas. In details the paleomagnetic and volcano-tectonic data set to be integrated with petrological, geochemical and isotopic in order to define: i) the paleomagnetic rotations occurring in each sedimentary basin and in the whole Southern Apennines and the Gibraltar arc, ii) the timing of the formation of the Calabrian and Gibraltar arcs and their relationships with the the geodynamics of the Central- Western Mediterranean, iii) the relationships between the arc formation and the development of the Cenozoic to Neogenic-Quaternary magmatism, with emphasis on the timing and the spatial relationships with the main structural elements, iv) the modelling of the geodynamic evolution of the Central- Western Mediterranean regions. <<<

First Results

At the end of this phase we will expect as results:
- to have analytical procedures running;
- to have databases available;
- to have the samples to be studied collected;
- to have preliminary petrological, geochemical, isotopic data on mafic lavas and xenoliths obtained;
- to have paleomagnetic data partially obtained;
- to have possible preliminary scientific notes written.At the end of this phase, which coincide with the end of the project, we will expect to have performed the overall scientific work scheduled. With this respect, results obtained can be grouped in several categories as follow:
a) analytical facilities. We will have new and up to date analytical methodologies running in each lab of the different Research Units beside the already present laboratories. Namely: U-series and Pb isotope analytical procedures (TIMS), along with trace element high resolution analytical procedures (HR ICP-MS) running at the labs of Reserch Unit n. 1; in situ trace element analytical procedures (LAM ICP-MS) running at laboratories of Research Unit n° 5;
b) databases. We will have different databases containing geochronological, geochemical, petrological, isotopic, paleomagnetic, structural, geological data of the studied areas, which will be available upon request;
c) analytical data. Major and trace alement data will be available both for whole rocks and minerals from lavas and xenoliths, along with isotope data on whole rocks and separated minerals. Beside also paleomagnetic data will be available.
d) maps. Thematic maps, structural and paleomagnetic, and geochemical contours maps showing distribution of isotopes and ratios between trace elements in the lithospheric mantle of the Mediterranean and Madacascan areas.
e) scientific papers. Papers reporting discussion of data, and petrologic and geodynamic modeling of the Africa-Europe system. <<<

Timescale

24 months

National and international background

Subduction-related magmas range in composition from tholeiitic to calc-alkaline, alkali potassic, and alkaline ultrapotassic. Alkalic potassic and ultrapotassic magmas are usually though to be collision-related, and therefore restricted either to the latest stages of the subduction processes or to post-orogenic collapse, post-dating convergent tectonics and active margin processes (e.g., Prelevic et al., 2004). They are found in many collisional orogens related to Alpine orogenesis (e.g., Venturelli et al., 1984a, 1984b; Conticelli et al., 1992; Conticelli, 1998; Duggen et al., 2004), but they also occurr in some continental arcs such as Andes, Western Mexico, and Eastern Sunda Arc (e.g., van Bergen et al., 1992; Carmichael et al. 1996; Righter and Carmichael, 1996; Carlier et al., 1997). The observed differences in alkali enrichment and in silica-saturation degree, shown by orogenic primary mantle magmas, are though to be variably related to: i) different depth of partial melting, ii) different partial melting degrees, iii) variable fertility of the pristine mantle wedge (prior to subduction); iv) variable degrees of metasomatic enrichments induced by subduction-related fluids/melts (e.g., Foley, 1992; Conticelli et al., 2002).
High Field Strength Elements (HFSE) depletion, usually observed in most orogenic magmas, could be related either to partial melting conditions and persistence of refractory accessory phases, or to the peculiar compositions of slab-derived metasomatic agents (e.g., Green and Pearson, 1986; Reagan and Gill, 1989; Foley and Wheller 1990; Thirlwall et al., 1994; Tatsumi and Kogiso, 1997; Elliott et al., 1997). In these magmas, incompatible element fractionation may be also observed among Large Ion Litophile Elements (LILE) in response to melt/fluid component of the metasomatic agent. Moreover, Sr, Nd and Pb isotopes represent a powerful tool in defining either mantle or crustal signatures, which might have been either acquired by magmas during ascent to surface or as primary characteristics of their mantle sources. All the above geochemical characteristics may be largely ascribed to metasomatic processes which affected the mantle sources in the various geodynamic settings.
Slab-derived metasomatic agents may show different isotopic (Sr, Nd, Pb, U, Th, Hf) signatures depending on their melt or fluid nature (e.g., Hawkesworth et al., 1997, Elliott et al. 1997, Turner et al., 2003) as well as their provenance (Bourdon and Sims, 2003). In particular, possible U/Th disequilibria in young (<100 ka) magmatic rocks can pinpoint the specific role of fluids opposite to adiabatic decompression in triggering mantle partial melting (Chabaux and Allegre, 1994; Elliott, 1997; Condomines and Sigmarsson, 2000; McKenzie, 2000).
In this frame, some questions still remain controversial, particularly with regard to mode and scale of the processes:
1) What is the petrological and physical meaning of the envisaged geochemical features?
2) What is the role of melts vs fluids as subduction-related metasomatic agents, also with respect to HFSE and LILE fractionation?
3) What is the role of pre-subduction geochemical and isotopic signatures of the mantle wedge?
4) How metasomatism (either cryptic or patent) is recorded in the upper mantle, and how newly stabilised metasomatic phases control trace element partitioning and isotope fractionation (if any)?
5) What is the role of prior depletion history in subduction- and collision-related magmatic sources (e.g., genesis of lamproitic magmas)?
6) Are the anorogenic features - sometimes detected in subduction environment- due to the pre-existing inherited signature in the mantle wedge or due to slab break-off?

Recent studies have shown that metasomatic processes on the lithospheric mantle are of fundamental importance in controlling the composition of mantle sources and the generated magmas in all geodynamic settings (e.g., Wilson and Downes, 1991; O'Reilly and Griffin, 1996; Beccaluva et al., 1998). Moreover, studies on mantle xenoliths have shown the strong variability and peculiarity of metasomatic processes affecting the lithospheric mantle (e.g., Menzies et al., 1987; Siena et al., 1991; Neumann & Wulff-Pedersen, 1997; Beccaluva et. al., 2001). Metasomatism causes generalized enrichment processes which over-impose on already depleted mantle (DM) sections due to previous extractions of mafic melts.
The proposed project will focus attention on the magmatism generated in the Central-Western Mediterranean during the Cenozoic Africa-Europe collisional events (e.g., Lustrino, 2000).This area, in fact, is particularly suitable for a comparative study of the mantle sources, since it is characterized by the presence of anorogenic and orogenic magmas, closely associated in space and time. Orogenic magmatism is basically related to Cenozoic subduction beneath the southern paleo-European continental margin at least since late Eocene, with generation of calc-alkaline to ultrapotassic volcanism in two domains: a) Provence-Sardinia-Tyrrhenian up to the Eolian Arc and the Roman Magmatic Province (Beccaluva et al., 1994), b) the Betic Cordillera (Southern Spain)(Benito et al., 1999; Duggen et al, 2004). Anorogenic magmatism is represented by important Cenozoic intraplate volcanic complexes located both on the African-Adriatic lithosphere (Veneto Volcanic Province and Etna-Iblei-Sicily Channel-North Africa volcanic districts)(Beccaluva et al., 1998; Bonadiman et al., 2001) and the European lithosphere (Sardinia and Olot-Calatrava-Tallante volcanic districts in Spain)(Cebrià et al, 2000).
Particular attention will be devoted to the petrological and isotopic features of the most primitive basic lavas, which may represent melts of direct mantle derivation. For comparison typical examples of both anorogenic and orogenic magmatic associations will be studied in well defined geodynamic settings related to the European and African domains, i.e. the Cenozoic volcanism of Madagascar, Macedonia and Canary Islands. The magmatism of northern Madagascar is emplaced above a lithosphere similar to that found in the Northern edge of the African plate which spreads also to the Italian regions and Adriatic area. This has an additional value to reconstruct the sources of the Italian anorogenic and orogenic rocks.. In addition, a petrogenetic study of mantle xenoliths hosted in the alkali basalts and basanites from northern Madagascar will be also carried out.
Petrological and isotopic studies will be also carried out on mantle xenolith populations from the same volcanic associations in the European (Sardinia and Spanish volcanic Districts) and African/Adriatic (Veneto volcanic province, Iblean and North-African volcanic districts) lithospheres. It should be remarked that, preliminary studies seem to indicate a different compositional evolution for the African and European lithospheric mantle sections (Beccaluva et al., 2001). The European lithosphere seems to be affected by multiple metasomatic processes with EM1 (Enriched Mantle) in addition to HIMU (high 238U/204Pb) components, overprinting an ancient depleted mantle. On the other hand, in the North African lithosphere the HIMU component predominates. Recent crustal derived metasomatic agents have also affected these two lithospheric mantle sources. Of particular interest are the mantle xenoliths included in anorogenic magmas which post-date orogenic associations (e.g.: Sardinia, Betic Cordillera, Morocco), in order to investigate the possible persistence of subduction-related components in the concerned mantle materials.
Petrological data will be compared with paleomagnetic and tomographic studies to be integrated in a coherent geodynamic model for the Mediterranean area.
In summary, the proposed project is addressed:
1) to recognise the nature of mantle sources prior to metasomatism in both anorogenic and orogenic settings;
2) to define the metasomatic components which affected the concerned mantle sections in both the European and African/Adriatic lithospheres;
3) to constrain the existence and nature of mass transfer as fluxes between different geochemical reservoirs (i.e., asthenospheric mantle, lithospheric mantle and crust);
4) to define paleomagnetic setting of studied areas and possible rotation in response to geodynamic evolution;
Geochemical and petrological results will be discussed in terms of large scale geodynamic processes, which occurred along the Africa-Europe boundary during the Tertiary period.
The Mediterranean region consists of an assemblage of relatively small lithospheric blocks, having a large variety of rheologies and thickness. These blocks are trapped between the Europa and Africa plates, and resulted from the North-Westwards convergence of the Africa plate respect to Europe during the Tertiary. This convergence has been absorbed during the Tertiary by plate-subduction and collision zones in both western and eastern portions of the Mediterranean (e.g., Rosenbaum et. al., 2002).
Seismic tomography and deep seismicity show that the present-day structure of the upper mantle in the Mediterranean region, is characterized by the presence of well defined subducting slabs below the Calabrian and the Aegean arcs, and (possibly) below the Gibraltar arc (Spakman et al., 1993; Piromallo and Morelli, 2003). The correspondence among narrow tongues of subducting lithosphere, arcuate mountain chains (i.e., Calabrian arc and Gibraltar arc) and back-arc basins opening (Tyrrhenian Sea and Alboran Sea) has suggested that differential roll-back of retreating slabs could be the common geodynamic process responsible for this complex tectonic setting (Malinverno and Ryan, 1986; Royden, 1993; Lonergan and White, 1997).
Paleomagnetic rotations about vertical axis are typical features of arcuate orogenic belts and back-arc basins (Jarrard and Sasajima,1980; Lonergan and White 1997). The precise definition of the amount and the timing of paleomagnetic rotations can allow to define the tectonic evolution of the arcs and consequently of the kinematic evolution of the subducting slabs roll-back. In particular huge paleo-magnetic rotations are expected to occurr along the lateral boundaries of the subducting slab, and therefore, may indicate regions of lateral inflow of astenospheric mantle into to the convective mantle wedge.
The Meso-Cenozoic volcanism of northern Madagascar is bimodal in age (late Cretaceous, mostly tholeiitic and late Cenozoic, strongly alkaline), and was emplaced above reworked Pan-African lithosphere (>510-550 Ma)(Melluso et al., 2003). The sources of the alkaline volcanics are close in space with those of presently active hotspots, like those of the Comores, but the influence of this plume is highly controversial (e.g., Melluso and Morra 2000), and the sources seem to have been lastly modified during the Precambrian orogeny (Lustrino et al., 2002). These sources were subsequently melted during intracontinental rift settings, like those in Sardinia and Southern Sicily. <<<