Basement evolution of the Sierra de la Ventana fold belt: New evidence for Cambrian continental rifting along the southern margin of Gondwana

Journal of the Geological Society, Jul 2003 by Rapela, C W, Pankhurst, R J, Fanning, C M, Grecco, L E

The closest sector to the Sierra de la Ventana belt is that of the Saldania and Cape Fold Belts in South Africa (Figs 2 and 10). Apart from the well-established similarities between the post-Cambrian geology, the Cambrian magmatic events identified in South Africa are also represented in the basement of the South American counterpart of the Cape Fold Belt. The Cape Granite Suite is slightly older (536-547 Ma, Armstrong et al. 1998; Da Silva et al. 2000) than the Sierra de la Ventana Granite Suite (524-531 Ma) but has similar Mesoproterozoic Nd model ages (Fig. 8). Geochemically, the San Mario granite of Sierra de la Ventana (524 or - 5 Ma) is very similar to the Phase II, Ia association of the Cape Fold Belt (520-540 Ma, Da Silva et al. 2000). However, the early Cambrian Cerro Colorado (531 or - 4 Ma) and Agua Bianca A^sub 2^-type granites do not have coeval equivalents in the Saldania belt, in which alkaline granites and syenites are restricted to the Phase III event (500-520 Ma, Da Silva et al. 2000). The Mid-Cambrian peralkaline La Ermita rhyolite (509 or - 5 Ma) is coeval with the Phase III alkaline event in South Africa, indicating that similar tectonomagmatic conditions, typical of those found in modern continental rifts, prevailed in both belts. Cambrian trachytes and rhyolitic rocks with alkalic affinities in the Sierra de Animas of southern Uruguay (Fig. 2) may be related to the same continental rifting event (520 or - 26 Ma, Rb-Sr, MSWD = 9.6; Bossi et al. 1993; recalculated with error at the 2[sigma] level).

Farther east in the margin, volcanic rocks of Mid-Cambrian age in the Ellsworth microplate have been correlated with rift-related sedimentary rocks underlying the Table Mountain Group in South Africa, and interpreted as having been formed in a continental rift environment; mid-ocean ridge basalts (MORB) were erupted near the rift axis and lithospheric mantle melts were emplaced on the rift shoulder (Curtis et al. 1999). Furthermore, U-Pb ages of granite cobbles and boulders from Lower to Middle Cambrian formations in the Ellsworth Mountains fall in the range 525-532 Ma (Rees et al. 1998; Randall et al. 2000), suggesting that they may have been derived from unexposed granite bodies in the Cambrian rift flanks (Curtis 2001). Remarkably, this is the same interval of crystallization ages as reported here for the Sierra de la Ventana Granite Suite. If the A^sub 2^-type granites and peralkaline event of the Sierra de la Ventana belt are correlated with those in the contiguous Saldania belt and the Ellsworth microplate, it would imply that continental rifting affected an impressive length of more than 3000 km of the proto-Pacific margin (Fig. 10, Table 3). This event could have profound implications for the initiation of Gondwana. The age of the A^sub 2^-type granites of the Sierra de la Ventana suggests that continental extension may have started during the Early Cambrian, which is consistent with the inferred timing of rifting in the Cape Fold Belt and the Ellsworth Mountains (Barnett et al. 1997; Curtis et al. 1999). Underplating of mafic magma may have triggered a high-temperature partial melting of an already dehydrated lower crust, producing the A^sub 2^-type fluorite-bearing granites of the Sierra de la Ventana belt. The eruption of peralkaline A^sub 1^-type rhyolites derived from an undepleted lithospheric mantle source is regarded as the final event of the continental rifting. Compared with the Ellsworth Mountains, the rift-related igneous rocks of the Sierra de la Ventana lack the Mid-Cambrian MORB-like depleted basaltic rocks and any evidence of exposed oceanic lithosphere. However, strong positive magnetic anomalies between the Sierra de la Ventana and the Tandilia belt (Fig. 2) have been interpreted as due to the presence of basaltic rocks beneath the thickest part of the folded Palaeozoic sequence (Kohn et al. 2002). Hence the possibility cannot be discarded that some sectors of the basin were floored by a transitional thin crust, mostly composed of oceanic basalts.