Geoscientist Online

Explosive volcanism in the Paraná- Etendeka Province, Namibia: Michael Mawby describes Society-funded research that is casting new light on some of the largest volcanic events ever to occur on Earth.

Geoscientist 18.10 October 2008

10,000BC, Flight of the Phoenix.... many recent Hollywood films have been made amid the spectacular scenery of Namibia, carved out through millions of years of geological turmoil. This land that time forgot possesses many geological wonders: including Fish River Canyon, the Hoba meteorite, “snowball Earth” deposits, Ediacaran fossils, and the little Mesosaurus bones that Alexander Logie du Toit and others used to prove that Africa and South America were once linked. Like them, the vast volcanic remnants in the north and north-western regions of Namibia originally spread across both the African and South American continent, representing massive eruptions caused by continental break-up.

Fig. 1 Location of the Etendeka study area. Landsat image of the features within Namibia Image reproduced from Mr Sid Images available via NASA

Current views

This is a topic of great current interest. Earth scientists are wondering what would happen if so called ‘mega-eruptions’ occurred today, and are presently monitoring and researching previous eruptions (at Yellowstone) to assess this. With so many important questions unanswered in this volcanic province, we begin to appreciate the scale and importance of the task that awaits us.

So - did these large volume (>1000 km³) eruptions come out as lava flows or as explosive volcanics (tephra)? The predominant view of the emplacement mechanism of Springbok and Goboboseb units of the Etendeka region depicts them as big, explosive ignimbrite flows - which enables them to travel far and blanket a large area¹. But debate still continues, as the units themselves show many features associated with silicic lava flows and because of their extreme volumes and areal coverage, they represent some of the most unusual silicic eruptive manifestations on the planet.

The term ‘rheomorphic ignimbrite’ has been used to describe the extensive welding on emplacement, which makes these units appear ‘lava like’. Voluminous explosive pyroclastic eruptions are believed to have deposited anomalously hot material over the region, which during or just after deposition underwent secondary flow. Such ignimbrites are not unusual and are commonly found in the Snake River Plain (USA) as well as Pantelleria Island (SW Italy). These units commonly have characteristics of both lavas and ignimbrites, and numerous techniques are needed to ascertain how the units where emplaced².

Fig. 2 Representation of the correlation between the Paraná-Etendeka province. The bimodal nature of the igneous activity can be seen, with the silicic units punctuated with the more mafic basaltic component of Paraná-Etendeka lavas. Notice how widespread the main silicic units such as the Goboboseb and the Springbok units cover both the Etendeka and the Paraná region. These silicic eruptive units cover a vast area and are estimated to have erupted 1000s km3 of material. The inset shows how much real-estate these units would have covered, with a comparison of one of the Paraná-Etendeka units with the largest mapped flood basalt unit (the Rosa member from the Columbia River Flood basalts). Skye and Mull, which contain smaller flood basalt units from the North Atlantic Igneous Province, for scale.

Fig. 4 Detailed logging of the Etendeka allows changes in the outcrop with height and distance to be analysed. Notice on the log how the facies change gradually with height. This may occasionally be a cooling pattern such as columnar or platy jointing, or the emergence of a new texture such as amygdales/vesicles or sheath folds. When a number of logs are collated a detailed 3D picture of textural and morphological variations can be made, helping us to determine the emplacement mechanism for the Paraná-Etendeka silicic units.

 

Current work

The work conducted at Durham involves detailed sampling of each unit for textural analysis and crystal populations on samples from proximal and distal units. This method aids our understanding of possible chemical changes through the sequence, or the crystal abundance through the units, as ignimbrites can have varying degrees of crystal populations from source to tip of the erupted unit(s). Detailed logging and measurements are also needed to understand the emplacement of these units (Figure 4). Unit logging (trying all the time to keep a safe distance from local wildlife) enables us to see these changes readily and easily once all the information is correlated together.

These logged units highlight not only the interplay of emplacement with topography, but also allow us to build up a picture of thickness from the source of the eruption to the edge of the package. The units collected as part of recent fieldwork will also enable us to build up a complex 3D image of the erupted units, permitting us to look at each facies in detail. The detailed measurements (e.g. amygdale/gas bubble or vesicle orientations) are taken to help constrain a direction/pattern of emplacement, therefore allowing us to give us an indication of sense of movement of the flows. This is an aid in locating the eruptive source, which for many units is unknown.

The collated logged sections reveal pinching and swelling of certain horizons as well as distinct horizons within topographic lows and slumping and folding structures at more distal locations that appear more indicative of ignimbrites.

Basal breccias can occur in lava flows and are best visualised as being similar to the tracks of a tank whereby the lava ‘breaks apart’ on the upper surface and gets ‘carried’ to the base and preserved. In the Namibian silicic units basal breccias have been found only at locations distant from the source. The inconsistent nature of the brecciation, and the fact that it is not found near the eruption source, leads us to believe that the brecciation is a response to cooling with distance - again a feature occasionally seen in rheomorphic ignimbrites. The vesicle measurements enable us to make a physical volcanological link to a proposed eruption source of the Etendeka silicic eruptions. The currently favoured source for the Goboboseb eruption , based on geochemical links, is the Messum crater (Figure 1).

Fig. 5 With 95% of the province on the Paraná side, we look at some of the highlights from the lava sequences on the wetter side. Great views are found at the Iguaçu falls where the Rio Iguaçu runs over thick flood basalt sequences. Large mega-vesicle horizons in some of the basalt flows in Southern Brazil have been filled with amethyst, agate and other gems. Some of the cavities are man-sized. Gems such as these are also found in the Etendeka - Namibia and are often sold at the roadside by local miners. Most outcrops are found in road cuts and quarries, making the mapping and correlation of the units a real challenge compared with Namibia.

Conclusions

Evidence collected so far indicates that the observed eruption style is consistent with an ignimbrite, and yes correlations with the Paraná so far seem justified. However that still leaves an enigma with respect to the lack of any ash horizon associated with these eruption(s).

Weissert and Erba3 have investigated environmental effects around this time (~132Ma) and have found that some nannofossils and certain faunal types experienced coincident extinction events that may be directly related to the Paraná-Etendeka volcanic episode. Currently we are aiming to use a number of techniques to quantify any possible environmental effects of these eruptions, and using FTIR (Fourier Transform Infra Red spectroscopy) to measure volatiles that cause explosive eruptions such as CO¬2 levels, as well as looking at apatite crystals to derive possible SO2 levels. Some cores recovered from the South Atlantic have silicic ash horizons that have been dated to an approximate age consistent with the Paraná-Etendeka eruption event. Detailed analysis of these may provide the missing link to Paraná-Etendeka ash horizons yet to be discovered.

When all the above information is collated we will more fully understand the emplacement mechanisms and environmental effects of these massive eruptive events that blanketed large parts of the Earth during the break-up of the ancient megacontinent of Gondwanaland.

Background

The Etendeka Province is situated in the Northwest of Namibia, formerly “Southwest Africa” until independence in 1990 (Figure 1). This vast province spreads across the arriviste ocean that now separates them, to Southern Brazil and Uruguay (Paraná). The Paraná-Etendeka is a classic bimodal province, exhibiting both mafic-silicic volcanic units erupted over a period of ~5Ma during the Early Cretaceous. The silicic eruptions have an age of ~132Ma in NW Namibia.

The silicic component, well exposed in the Etendeka, presents an enigma with respect to physical volcanological emplacement mechanisms - the debate centring on lava flow vs. explosive ignimbrite. The resolution of this debate could have profound repercussions for our understanding of volcanology. So why should this be the case, and what makes the Paraná-Etendeka so special?

The estimated eruptive volume of these units is significant - the largest units (Springbok) estimated at 6340km³; Grootberg, 3775km³, and Goboboseb 2320km³. (see Figure 2 for Etendeka-Parana correlation). Compare these numbers with the commonly cited largest silicic single eruption (Fish Canyon Tuff ~5000km³).

If the units are lava flows, the questions we need to resolve are: how do you erupt such sizable volumes of magma – and particularly thick, viscous silicic lava? How does this reflect magma origins and formation at depth? And how do you erupt a lava body that ‘usually’ does not cover such an areal extent? On the other hand, if the units are ignimbritic (explosive) in origin, why is no regional or global ash record associated with the eruptive packages? What climatic effects did the Paraná-Etendeka eruptions create?

 

Acknowledgements

Namibia fieldwork was made possible by the Elspeth Matthew grant of the Geological Society of London, which is gratefully acknowledged. Thanks go to Petrobras and Breno Waichel for funds for a short visit by Dougal Jerram to Brazil in April 2008. Michael Mawby thanks his supervisors and co-authors; Dougal Jerram, Jon Davidson and Scott Bryan. Field assistance was provided by Jen Waters, Petra Zippe and Matthews Nakalemo, as well as help from the Geological Survey of Namibia and Air Namibia. This article was compiled by Michael Mawby ([email protected]) with additional material by Dougal Jerram (www.dougalearth.com).

References cited

1. Milner, S C , Duncan, A R , & Ewart, A (1992). Quartz latite rheoignimbrite flows of the Etendeka-Formation, north-western Namibia. Bull Volcanol., 54, 200-219.
2. Henry, C D & Wolff, J A (1992). Distinguishing strongly rheomorphic tuffs from extensive silicic lavas. Bull Volcanol., 54, 171-186
3. Weissert, H , Erba, E , (2004). Volcanism and paleoclimate: a late Jurassic-Early Cretaceous carbon and oxygen isotope record. Journal of the Geological Society, London, Vol. 161, 695-702

Further reading/links

• Milner, SC, Duncan, AR, Whittingham, AM, Ewart, A, (1995). Trans-Atlantic correlation of eruptive sequences and individual silicic volcanic units within the Paraná-Etendeka igneous province. Journal of Volcanology and Geothermal Research, 69, 137-157
• Jerram, DA, (2002). Volcanology and facies architecture of flood basalts. In: Menzies, MA, Klemperer, SL, Ebinger, CJ, Baker, J (eds) Volcanic Rifted Margins. Geol. Soc. Amer. Spec. Pap. 362, 121-135.
• Author’s website: www.dur.ac.uk/m.r.mawby