<p>We present new geochemical, isotopic and geochronological analyses of Late Triassic-Jurassic volcanic and intrusive rocks of the Antarctic Peninsula and Patagonia. Whole-rock geochemical data suggest that all of these igneous units formed in an active margin setting. This conclusion challenges the current paradigm that Jurassic magmatism of the Chon Aike province formed by the migration of the Karoo mantle plume from Africa towards the Pacific margin (Pankhurst et al., 2000). KDE analysis of 98 crystallisation ages reveals four main pulses of magmatism (V0: ~223-200 Ma; V1: ~188-178 Ma; V2: ~173-160 Ma; V3: ~157-145 Ma), which are approximately coincident with the episodic nature of the Chon Aike Magmatic Province reported by Pankhurst et al. (2000). Some magmatic units in eastern Patagonia are distal to the hypothetical paleo-trench relative to most active margin magmatism. These rocks have geochemical and geochronological characteristics that are indistinguishable from active margin-related rocks located ~200km from the palaeo-trench. Thus, we propose that a segment of the slab formed a flat-slab along southwestern Gondwana during the Late Triassic-Jurassic. This flat-slab is probably a temporal extension of the flat-slab episode suggested by Navarrete et al. (2019) for the Late Triassic (V0 episode) in eastern Patagonia. The progressive migration of the flat-slab magmatism to the southwestern margin of Patagonia suggest an evolution of its architecture during the Jurassic. Further, we propose that the flat-slab magmatism present in eastern Patagonia was triggered by slab failure, where foundering of the slab drove upwelling of hot mantle, forming a broad arc in an inland position in eastern Patagonia. Flat-slab subduction finished during the V3 episode (~157-145 Ma), with a continuation of an active margin along the western margin of the Antarctic Peninsula and Patagonia. Coeval extension in the South Atlantic and in western Patagonia lead to sea floor spreading, the formation of the Weddell Sea (~155-147 Ma; e.g. Konig & Jokat. 2006) and the Rocas Verdes Basin (~150 Ma; e.g. Calderon et al., 2007), respectively. The paleogeographic reconstructions juxtapose the northern Antarctic Peninsula and southern Patagonia during the Late Jurassic (e.g. Jokat et al., 2003), which suggest that the Rocas Verdes Basin and the Weddell Sea are oriented by a ~120&#176; angle and potentially meet in southern Patagonia. This junction of sea-floor spreadings corresponds to the limits of the southern Rocas Verdes Basin with the eastern Weddell Sea oceanic lithosphere. We suggest that these rifts formed part of a triple junction, while the third rift arm should be located with a sub north-south orientation in the Antarctic Peninsula. Vast regions of the Antarctic Peninsula remain unexplored beneath the ice-cap, although we speculate that the third arm may correspond to the Eastern Palmer Land Shear Zone, which currently has a lateral extension of ~1500km (Vaughan & Storey, 2000). This new triple junction would be a Ridge-Ridge-Transform Fault intersection.</p><p>Calderon et al. 2007. JGS,164: 1011-1022.</p><p>Jokat et al. 2003. JGR, 108: 2428.</p><p>Konig & Jokat. 2006, 111: B12102.</p><p>Pankhurst et al. 2000. JP, 41(5): 605-625.</p><p>Navarrete et al. 2019. ESR, 194: 125-159.</p><p>Vaughan & Storey. 2000. JGS, 157: 1243-1256.</p>
A millennial summer temperature reconstruction for northeastern Canada using oxygen isotopes in subfossil trees