MAGMA MIXING ORIGIN FOR THE FAMATINIAN MAGMATISM. EASTERN MAGMATIC BELT, PUNA, NW ARGENTINA

SUZAÑO, N.; BECCHIO, R. A.; SOLA, A. M.; ORTIZ, A.; NIEVES, A.; QUIROGA, M.; FUENTES, M.G.

Tucumán

Congreso; XX CONGRESO GEOLÓGICO ARGENTINO; 2017

Magma mixing has been mentioned to explain the compositional range of high-K calc-alkaline suites, mainly intermediate to acid composition (SiO2=60?75 wt.%) that constitute the Western and Eastern magmatic belts in Puna and Tastil batholith in Eastern Cordillera (e.g. Otamendi et al. 2010, Suzaño et al. 2015). They are constituted by mafic to peraluminous-weakly metaluminous granitic plutonic, subvolcanic and eruptive rocks. Plutonic rocks from Eastern belt ranging in age from Late Cambrian to Early Silurian (500?440 Ma; Viramonte et al. 2007, Bahlburg et al. 2016) with a main magmatic peak at ca. 470 Ma. On the other hand, Hongn et al. (2010) determined that the called ?Tilcarean arc? (included Chañi, Tipayoc Cañañí intrusives and Tastil batholith, ca. 540-514 Ma, e.g. Hongn et al. 2010 and references therein, Hauser et al. 2011) should be considered as Famatinian granites and not be integrated in the Pampean magmatic arc. Alternatively, intermediate and acid plutonic rocks might be the result of crustal recycling processes with minor contribution of juvenile mantle-derived magmas (Lucassen et al. 2000, Viramonte et al. 2007) and then, undergo assimilation of larger amounts of sedimentary or local metamorphic country rocks (e.g. Bahlburg and Berndt 2016). Also, they could represent fractionated liquids derived from a parental magma of broadly diorite composition (e.g. Bellos et al. 2015). In order to account the high FeOt-MgO-K2O content, Fernández et al. (2008) suggest a ?ferrosilicic magmatism? origin related to particular petrogenetic model with near-total melting (80?90%) of crustal sources under very high P-T conditions (1000 °C?1200 °C and 1.0 to 2.0 GPa). Thus, considerable uncertainties and controversies remain inconclusive regarding the petrogenesis of the magmatic belts in Puna, particularly on the geochemistry features and inherited zircon distribution pattern. In this contribution, we evaluate the genesis of Cambrian-Ordovician intermediate-acid rocks from the Diablillos Intrusive Complex (CID, 501±17 Ma, Suzaño et al. 2015) and the Cerro Bayo area (CBA), through detailed geological, petrographic, mineralogical, geochemical, and Nd isotopic analyses in combination with published data. The CID is located on the eastern edge of the southern Puna and four intrusive units have been distinguished: (1) two-mica syeno- and monzogranite, (2) granodiorite and tonalite, (3) diorite and Qz-diorite, and (4) late mafic dykes (Suzaño et al. 2015). The contact relationships between granite bodies vary from sharp to transitional types. The intermediate-acid rocks contain abundant microgranular enclaves ranging in composition from granodiorite to diorite (MME) and are intruded by diorite and Qz-diorite stocks and syn- to post-magmatic mafic dikes (symbols from Whitney y Evans 2010). Overall, granites consist of gray, porphyritic with Kfs and Pl phenocrysts immersed in an essentially homogeneous medium-grained Bt-rich (Qz+Kfs+Pl±Ms) matrix. Mafic to intermediate enclaves show various shapes (rounded, ellipsoidal) and diameter (up to 2 m). Granites and mafic rocks show several disequilibrium textures such as quartz ocellar, rapakivi and Pl phenocrysts with complex internal structures (see Nieves el al. this congress). Pl grains from the most intermediate-acid granites show core?An-rich mantle?outer rim structures in normal and reverse compositional zoning pattern. For instance, An values from granites both increasing and decreasing from the core to mantle and then decreasing to the rim. Also, some Pl contains cores with contrasting An content, Albite (An6-9) and Andesine (An46). Some plagioclases of mafic rocks (MME and intrusive body) show An-rich cores (An41 to An60) surrounded by an An-poor outer rim (An31 to An43) compatible with intermediate-acid magmas composition. The CID granites yield εNdt values from −4.7 to −7.5 and two-stage Nd model ages (TDM) of 1.5?1.8 Ma. This values are consistent with observed in Eastern belt and the Tastil batholith granites and overlapping with western Puna-Eastern Cordillera metamorphic basement. Granodiorite and tonalite yield values from −2.4 to −3.9 and TDM values of 1.363?1.456 Ma. The diorite body sample yields εNd value of +1.7 and TDM value of 1.040 Ma, whereas MMEs show strongly negative εNd values from −4.3 to −7.4. The dioritic intrusive body from the CID is consistent with the mafic isotopic composition of the Western and Eastern belts. Cerro Bayo area is located in the SW of the CID and it is constituted by an elongate synplutonic composite dyke body enclosed within granites (two-mica monzogranite), granodiorites and tonalities fo the Eastern belt. Two-mica monzogranite is a distinctive porphyritic granite with k-feldspar megacrysts up to 7 cm in size and encloses scarce rounded dark-gray microgranular enclaves. Some Kfs megacrysts develop a thin outer rim composed by Pl (rapakivi texture). Synplutonic composite dyke includes Qz-diorite (Bt-Qz-diorite and Amp-Qz-diorite) and a wide range of leucogranites (syeno to monzogranite) with abundant MME and acid enclave inclusions. All these magmatic units have transitional contacts with one another. Diorites are dark-gray, equigranular fine-grained and the mineral association is Amp+Bt+Pl+Qz±Bt+Ttn +Ep+Aln+Ap+Zrn and opaque minerals. Bt-Qz-diorite exhibits conversion of Amp to Bt texture. Successive acid dykes cut mafic ones and generate variable leucogranites types (Bt-Ttn syenogranite and Bt-Ttn monzogranite) and may include both previous mafic magmas (MME swarm) and leucogranites as enclaves. Overall, leucogranites are white to dark-gray, equigranular fine to medium-grained constituted by Qz+Kfs+Pl+Bt±Ms±Ttn+Zrn+Ep +Aln+Ap and opaque oxides. Leucogranites show progressive variation of Bt (9 to 20%), Ms (disappear in hybrid ones) and Ttn (2.5 to 4.3%) modal proportion. Collectively, both complexes have pronounced-linear trends of decreasing FeOt, CaO, and MgO, V, near-lineal TiO2, Sr and near-lineal increasing K2O, Rb, Ba with increasing silica content. However, the P2O5, Al2O3, and Na2O, Cr values exhibit greater scattering (mainly in mafic members) and less correlation with SiO2. Intermediate to acid rocks (59?69 wt.%, SiO2) always plotting in the middle between the two-mica syenogranite and mafic rocks. In addition, intermediate to acid rocks from the studied localities, Eastern belt and the Tastil batholith (average n=180; SiO2=68.63 wt.%) show enrichment in K2O, FeOt, MgO and minor Na2O content than typical calc-alkaline values (e.g. Sierra Nevada batholith). These distinctive geochemical features remain along the Famatinian-Puna paleo-arc (e.g. Otamendi et al. 2010, Suzaño et al. 2015). In CBA, the field relationship indicates that successive acid pulses cut the previous mafic dyke and undergo progressive hybridization in dyke interiors. Hybrid leucogranites are featured by progressive enrichment in major elements such as FeOt (3.98 wt.%), CaO (2.33 wt.%), TiO2 (0.72 wt.%) and MgO (0.94 wt.%) when are compared with Ms-bearing leucogranite end-member. Major, trace element and whole-rock Nd isotope modeling and petrological data, suggests that magma mixing between hydrous juvenile mantle- and crustal-derived magmas contributed significantly to the geochemical variation of granites from studied localities. Our geochemical modeling reveals that fractional crystallization from a parental mafic magma involving the hypothetical fractionation of Amp/Cpx and Pl cannot explain the compositional variations in the trace elements. Langmuir et al. (1978) showed that on a ratio?element plot data consistent with mixing, it would lie along a hyperbolic curve between two contrasting end-members. In our study, these characteristic hyperbolic mixing arrays are observed for CID, CBA rocks in plots for K2O/MgO, CaO/MgO and Al2O3/FeOt vs SiO2. Most intermediate-acid plutonic rocks, inclusive Tastil batholith and Eastern belt, fall into or close to the mix line predicted by different two-component mixing hypotheses. Also, hypothetical εNdt mixing lines were computed for CID and Puna magmatic belt suites. We used local end-members for CID and potential primitive mafic rock and two crustal-derived leucogranites from Puna and Eastern Cordillera. The intermediate and acid plutonic rocks from the Eastern belt (intrusive and volcanic equivalent rocks), Tastil batholith together with studied rocks appear close to the SiO2 vs. εNd(t) hyperbolic array. It suggests at least two kinds of magmas with contrasting Nd isotope compositions were mixed. Assimilation processes involving a large amount of eastern Puna basement (e.g. Viramonte et al. 2007; Bahlburg and Berndt 2016) are also unlikely to explain the compositional variability of the studied samples. For instance, to explain the high MgO contents on intermediate rocks is necessary more than 40% of assimilation (Bellos et al. 2015), which is hard to be accepted in petrological constrained terms (see Reiners et al. 1995). On the other hand, isotopic systematics of Nd-Sr and inherited zircon pattern in particular, provided evidence for the bulk assimilation of crustal xenoliths. However, Sola et al. (2013) showed the inherited zircons ages in the crustal-derived melt exhibit a large proportion of inherited zircon from the source (Puncoviscana Formation). According to our study, the MME in the granites from the CID and CBA are generated through both injections of a mafic/acid magmatic component into a granitic magma chamber (e.g. Vernon 1984) or an early mafic dyke (e.g. Collins 2000, synplutonic composite dyke) via a mixing/mingling process. Several lines of evidence lead us to conclude that the MME were generated via these mechanisms or a combination of them (Barbarin 2005). Enclaves with ellipsoidal?ovoidal shapes, transitional zones in the contacts, multi-component enclaves, and xenocrysts exchanges are consistent with the MME representing globules of near-liquidus hybridized mafic melt. Furthermore, MME have disequilibrium textures such as resorbed surface and zoning in Pl, quartz ocelli and acicular apatite reflecting physic-chemical and/or thermal changes (e.g. Pietranik and Koepke 2014). The lowest initial 87Sr/86Sr ratio and the high εNd(t) values isotope compositions indicate the contribution of the mantle and crustal contamination to the generation of the mafic magmas of the CID and others mafic rocks along the Western- and Eastern belt. The variable Ni, Co, Cr and Y values suggests that at least some early fractionation process could be involved in the first stages of evolution of mafic rocks. Aluminum-in-hornblende geobarometer (Anderson and Smith 1995) suggests that the MME and Qz-diorites bodies from CID crystallized under the similar P-T conditions (average 2.21?2.48 kbar), shallow depths emplacement (~8-9 km, assuming that 1 kbar=3.5 km of the crust) and at 677?727°C (Holland and Blundy 1994). According to Ridolfi et al. (2010), the Qz-diorites intrusive bodies yield water contents of 4.37?5.91 wt.% (error of ca. 15%). We propose that partial melting of the crust and the main magma mixing event occurred in the lower crust and was possibly triggered by underplated and intraplated hydrous mafic magmas. Underplating of the crust by mantle-derived mafic magmas has been regarded as important heat source factors in the metamorphism (HT/LP, Lucassen et al. 2000) and origins of the widespread Lower Paleozoic granitoids. Extensive and variable interaction degree between the crustal-derived- and juvenile mafic melts may have produced large bodies of hybrid magmas (monzogranite to Qz-diorite) at depth level crust. The produced hybrid intermediate to acid melts then, were emplaced at shallow depth and occasionally injected either by synplutonic- to late successive pulses of mafic magmas. Isotopic and geochemical comparison between the studied rocks and the magmatic belts in Puna and Tastil batholith suites reveals a marked resemblance. The data suggests that those rocks were probably generated by magma mixing and denotes a regionally and continuous process in a long-lasting (~540-440 Ma) active continental margin. This widespread magmatic activity includes contemporaneous mafic and dominate peraluminous acid magmatic activity. For a wide distribution of a long-lasting magmatic activity, it is necessary that this magmatism develops in a back-arc environment such as proposed by Hyndman et al. (2005) where the high heat flow is independent of compression or extension setting. Overall, the estimated proportion of mafic component 0.40-0.67 and 0.14-0.35 in the mixtures may produce the typical tonalite-granodiorites and monzogranites magmas. Thus, acid magmas contain a significant amount (up to 35%) of mafic magma. The widespread regional distribution of acid rocks suggests a significant volume of mafic magma that has not been considered before in the literature.