Origin of pre-mesozoic xenocrystic zircons in cretaceous sub-volcanic rocks of the northern Andes (Colombia): paleogeographic implications for the region

This paper reports the presence of xenocrystic zircons in sub-volcanic rocks (Guarco Andesite and Porphyritic Intrusives units) exposed in the northeast of the Western Cordillera of Colombia (WCC) and provides a new contribution to the understanding of the tectono-magmatic evolution of the area during the Upper Cretaceous associated with the geological history of the Caribbean Large Igneous Province (CLIP). The study area is located in an accretionary terrane constituted by a basement of oceanic affinity (Barroso Formation and San José de Urama Diabases unit), amalgamated against the northwestern South American margin and associated with the development of CLIP. The elemental and isotopic geochemical data (with an initial ratio of 87Sr/86Sr between 0.70334 and 0.70543 and εΝd between +6.25 and + 7.42) indicate an island arc environment for the formation of sub-volcanic rocks and a mantle wedge source area (and/or the arc itself), which presents subordinate levels of crustal material. U-Pb zircon ages by SHRIMP IIe and LA-ICP-MS show ages from the Mesoarchean to the Upper Cretaceous, with representative peaks of approximately 2700, 2100, 600 and 90 Ma. The younger age population is interpreted as the crystallization age of the studied rocks and indicates a spatial and temporal relationship with the basement rocks associated with the CLIP. The older zircons are interpreted as xenocrysts coming from terrigenous sediments that were dragged down into the subduction channel (reflecting crustal recycling processes through the mantle in subduction zones). Alternatively, the xenocrystic zircons could have come from the assimilation of crustal sediments that were previously deposited over the primitive arc. We suggest that the source areas of zircon xenocrysts probably are located close to the Arequipa Massif and the Marañon Complex inliers, and close to the ensialic Basement of the Central Cordillera of Colombia (CCCB). Using these source areas and paleogeographic/paleomagnetic models published by other authors, we propose two possible paleogeographic scenarios differentiated by the polarity of the subduction zone.