How are `Super' eruptions preserved in the past? Volcanological features of Large Volume Silicic Eruptions of the Paraná-Etendeka

The eruption and emplacement mechanisms of the Parana-Etendeka (132 Ma) silicic eruptive units have long been a contentious issue. Field relationships and textural evidence observed in the well-exposed sections in the Etendeka region (NW Namibia) indicate they have features associated with both lava flows and ignimbrites, therefore posing an enigma with respect to emplacement mechanisms. This is of particular interest given the dimensions of some of the erupted units, e.g. the Springbok is the largest unit with an estimated erupted volume of 6340km3 whilst the Grootberg (3775km3) and Goboboseb (2320km3) are also voluminous. If these units were emplaced as rheoignimbrites then explosive eruptive activity may have ejected large volumes of ash and aerosols into the stratosphere. We present initial results from detailed field studies on the Goboboseb Quartz Latite (QL) erupted from the Messum Complex, NW Namibia. The Goboboseb QL exhibits topography-filling facies that resemble the classical deposit geometry of valley-filling ignimbrites. Vertical sections indicate a decrease in flow shear vertically as deposit facies change from platy jointed and intensely sheared lower facies to a thick, massive to crudely columnar jointed and amygdaloidal QL. Some quartzite clasts were observed within the QL (~30km from Messum), and are interpreted to have been vent-derived as the substrate across which the Goboboseb flowed was covered by flood basalt lavas. The amygdaloidal upper parts of the QL eruptive units show prominent shear defining flow lineation`s, and variations up sequence for the Goboboseb QL from E-W to N-S; this may reflect a change in eruptive source location or changes in substrate topography as the QL buried surface irregularities. At distances ~30km from the source, the Goboboseb QL was coherent, exhibiting no basal breccias as expected for lava flows. However, basal breccias have been described at more distal sections and this lateral facies transition may reflect cooling of QL material with run-out. Current inferences from evidence collected from the recent field-season and described above, indicate that these units may well have been emplacement as rheoignimbrites