Nova Terra 52

&KDSWHU by which eroded continental crust and oceanic sediments enter the mantle. Seismological studies (e.g. Scholl and van Huene, 2007) suggest that tectonic erosion and sediment-trapped subduction must have operated transporting grabens that were filled by trench-turbidite near the trench, and the surface granitoid (e.g. TTG, Tonalite-Trondhjemite-Granodiorite) material into the deep mantle. Crustal materials may be added to the mantle through sub- duction erosion mechanisms which could play an important role in the melt generation, plate dynamics and the generation of enriched mantle reservoirs, especially during supercontinent assemblages (e.g. Sun and Stern, 2001; Stern and Scholl, 2010; Stern, 2020; Straub et al., 2020), but whose implication is scarcely traceable with conventional methods. When the evolution of a long-lived arc system (e.g. Cadomian arc) is studied in detail, distinctive maturation stages can some- times be constrained. The thermodynamic features of the oceanic lithosphere and its influence on the crust/mantle geothermal struc- ture plays a role in the opening and closing of marginal basins (Arenas et al., 2021 and references therein), varying cyclically the incorporation of materials to the subduction zone and determining the magma geochemistry produced at active margins (Yamamoto et al., 2009). This crust/mantle interplay results in more than one primitive melt type for most arcs, which testify for heterogeneity and complexity beneath the upper plate between the mantle wedge and added slab component (Schmidt and Jagoutz, 2017; Stern, 2020). Indeed, granitoid magmas represent mixtures of mantle and crust-derived melts of various compositions and in diverse proportions, being hardly derived from a single source (Frost et al., 2001; Sun and Stern, 2001). This continuous reworking makes it difficult to interpret the age of crystallization of melts, which can lead to errors in constraining the arc evolutionary his- tory. However, the detection of certain geochemical and isotopic patterns, widely suggested as subduction initiation signature, can greatly facilitate the interpretation in ancient magmatic arcs, for which essential parameters such as subduction polarity, plate cou- pling or interference with other orogenic systems (among other) are not available. The generation of adakites (Kay, 1978; Defant and Drummond, 1990) has been classically linked to subduction initiation of hot and young oceanic lithosphere, on many occasions associated spatially and temporally with the appearance of boni- nites and/or Nb-enriched basalt (Polat and Kerrich, 2004; Manikyamba et al., 2009). This term remains the subject of an extensive debate due to the unclear genesis mechanism of these rocks. An equivalent hot thermal regime can also be achieved by other mechanisms (for example, slab failure, development of slab windows or even zones of rapid subduction) and even generated through crustal subduction at great depths or even in combination with metasediments (Wang et al., 2008; Stern, 2020). This deriva- tion and division into subgroups (Martin et al., 2005; Moyen, 2009; Hastie et al., 2010; Zhang et al., 2019) represents the substitution of the initial term for adakitic rocks or rocks with adakitic signa- ture, emphasizing the geochemical similarities with these rocks. The Northern Gondwana margin, around the West African Cra- ton (WAC) was formed by subduction and accretion during the Proterozoic (Cadomian Orogeny). This pre-Paleozoic basement is formed by thick sequences of metasediments derived in part from cratonic areas as well as from the magmatic activity of the so- called long-lived Avalonian-Cadomian magmatic arc, developed between c. 750–500 Ma (e.g. Dalziel, 1997; Linnemann et al., 2008; Pereira et al., 2006, 2012; Andonaegui et al., 2016). Sections of this margin appear dismembered and were incorporated in late Paleozoic times to the innermost sections of the Variscan Orogen (Matte, 2001; Martínez Catalán, 2011; Díez Fernández and Arenas, 2015; Arenas et al., 2016; Díez Fernández et al., 2016), which extends from North Africa (e.g. Errami et al., 2009, 2021; El Hadi et al., 2010), eastward through the Alpine-Himalaya oro- genic belt (e.g. Moghadam et al., 2016; Moradi et al., 2020) and through the Appalachian Orogen. There is general consensus that the periphery of Gondwana was an active margin during the Neo- proterozoic, but the pulses and tectono-magmatic events that con- tributed to its building and evolution remain poorly understood (Ordóñez-Casado, 1998; Sánchez-Lorda et al., 2016; Arenas et al., 2018; Sarrionandia et al, 2020), and in most cases there are no recorded events prior to c. 580–560 Ma. Identification of specific magma patterns and eventually obtaining information about their tectonic setting generation needs to use whole-rock geochemistry along with isotope (Sr-Nd) geochemistry. In this regard, complementation with zircon geochronology can help to unravel in part the complexity of the processes that operated in this section of the crust between the Proterozoic and Early Paleozoic (e.g Arenas et al., 2018; Díez Fernández et al., 2019; Fuenlabrada et al., 2021). This complete methodology may provide relevant keys in the study of the Cado- mian arc evolution in many cases not very accessible due to the penetrative deformation, resulting from tectonic processes related to the orogenic Variscan (late Paleozoic) and Alpine (Cenozoic) cycles. Specific scientific contributions on tectono-magmatic evolution of the Cadomian arc cycle-steps, including subduction initiation, are scarce in the whole European Variscan Belt. In this research, we present a new complete study of the magmatic evolution that occurred in one of the best-preserved arc crustal sections of the Cadomian peri -Gondwana arc (Upper Schist-Metagranitoid Unit of the Mérida Massif, SW Iberia). This study includes a detailed whole-rock geochemical and isotopic Nd-Sr analysis, along with new U-Pb laser ablation zircon ages of five metaigneous com- plexes, which along with its sedimentary host recorded the main magmatic events happened in the arc section preserved in the SW Iberian Massif. Igneous rocks from this suite span from initial stages of subduction to more mature periods of the Cadomian arc and can contribute to the knowledge about magma sources and processes in the active margin from Ediacaran to Lower Cambrian times. 2. Geological setting The southern branch of the European Variscan Orogen (Fig. 1a), the Iberian Massif, contains a wide representation of the northern Gondwana margin. The most external sections of this margin define an Upper Allochthonous Terrane (Díez Fernández and Arenas, 2015) which is well exposed and preserved in the Ossa- Morena Complex (OMC, SW Iberian Massif) (Fig. 1b). The basement of the OMC contains a Neoproterozoic to Lower Cambrian metasedimentary sequence, traditionally referred to as Serie Negra Group (Fig. 1c; Black Series; Carvalhosa, 1965). This metasedimen- tary series has been traditionally divided into two different forma- tions, from bottom to top: the Montemolín Formation which consists of metagreywackes, metasandstones, schists, micaschists, quartzschists and black quartzites, with levels of amphibolites and amphibolic gneisses; whose maximum depositional age has been estimated at c. 600 Ma (U-Pb geochronology in detrital zircons; Ordóñez-Casado, 1998). The Tentudía Formation which consists of metasandstones, volcanogenic metagreywackes, slates and phyllites, black quartzites, metacherts and layers of micaschists and limestones, and whose maximum depositional age has been calculated at c. 565–541 Ma (U-Pb geochronology in detrital zir- cons; Schäfer et al., 1993; Linnemann et al., 2008; Pereira, 2015). The metamorphic grade of both formations varies along the differ- ent regions of OMC, ranging between chlorite zone to regional migmatization (Eguíluz and Quesada, 1981; Montero et al., 1999). The whole-rock geochemistry and Nd isotopic sources of E. Rojo-Pérez, U. Linnemann, M. Hofmann et al. Gondwana Research 109 (2022) 89–112