New 40Ar/39Ar analyses of borehole samples from the Parana-Etendeka continental flood basalt province provide critical evidence of eruption rates in unexposed regions. When combined with surface samples, the new data clarify the duration of this major volcanic province. New ages from both surface and borehole samples confirm a 10-12 million year duration of magmatism, which post-dates the Tithonian mass extinction, and provide a unique 3-dimensional picture of the spatial and temporal eruption of the lava pile, Chemically defined magma types are diachronous and the onset of magmatism occurred 500-1000 km inland, migrating southeast towards the incipient South Atlantic ocean. Calculated eruption rates increased from 0.03 km(3) yr(-1) between 138 and 135 Ma, through 0.13 km(3) yr(-1) between 135 and 133 Ma, to 0.21 km(3) yr(-1) between 133 and 131 Ma, supporting the notion that the rate of magmatism increased with time and proximity to the developing South Atlantic, After 131 Ma the continental eruption rates dropped dramatically to 0.01 km(3) yr(-1), presumably because rifting had occurred, and eruption centred in the oceanic Rio Grande Rise with eruption rates of around 0.3-0.5 km(3) yr(-1). Given the available constraints on the potential temperature of the plume, thickness of the continental lithosphere and degrees of extension through time, predicted melt production rates within the plume seriously underestimate the calculated eruption rates on the continent. However, the eruption rates can be reconciled if melting occurred by conductive heating of volatile-enriched mantle, The diachronous nature of the magma types supports the idea that melting occurred over a wide area and that the different magma types reflect different source regions rather than the temporal evolution of magmas from a single source. Given that 10 million years is appropriate to the timescales for conductive heating above the mantle plume, it is suggested that these distinct source regions were located within the lithospheric mantle, consistent with geochemical studies. New 40Ar/39Ar analyses of borehole samples from the Parana-Etendeka continental flood basalt province provide critical evidence of eruption rates in unexposed regions. When combined with surface samples, the new data clarify the duration of this major volcanic province. New ages from both surface and borehole samples confirm a 10-12 million year duration of magmatism, which post-dates the Tithonian mass extinction, and provide a unique 3-dimensional picture of the spatial and temporal eruption of the lava pile, Chemically defined magma types are diachronous and the onset of magmatism occurred 500-1000 km inland, migrating southeast towards the incipient South Atlantic ocean. Calculated eruption rates increased from 0.03 km(3) yr(-1) between 138 and 135 Ma, through 0.13 km(3) yr(-1) between 135 and 133 Ma, to 0.21 km(3) yr(-1) between 133 and 131 Ma, supporting the notion that the rate of magmatism increased with time and proximity to the developing South Atlantic, After 131 Ma the continental eruption rates dropped dramatically to 0.01 km(3) yr(-1), presumably because rifting had occurred, and eruption centred in the oceanic Rio Grande Rise with eruption rates of around 0.3-0.5 km(3) yr(-1). Given the available constraints on the potential temperature of the plume, thickness of the continental lithosphere and degrees of extension through time, predicted melt production rates within the plume seriously underestimate the calculated eruption rates on the continent. However, the eruption rates can be reconciled if melting occurred by conductive heating of volatile-enriched mantle, The diachronous nature of the magma types supports the idea that melting occurred over a wide area and that the different magma types reflect different source regions rather than the temporal evolution of magmas from a single source. Given that 10 million years is appropriate to the timescales for conductive heating above the mantle plume, it is suggested that these distinct source regions were located within the lithospheric mantle, consistent with geochemical studies.
