- Slides: 1
Density analysis of magmatic and phreatomagmatic phases of the 934 AD Eldgjá eruption, southern Iceland William M. Moreland* and Þorvaldur Þórðarson Institute of Earth Sciences, University of Iceland *presenting author, email: wmm [email protected] is • • • 1. Introduction to Eldgjá is a ~75 km long fissure in southern Iceland (Fig. 1) which erupted in the 10 th century and may have lasted as long as 8 years (Thordarson et al. 2001). It is the largest fissure eruption in Iceland in historic times. It is of the mixed-cone row type and produced two large lava fields (18. 3 km 3) and a widespread basaltic tephra deposit (1. 3 km 3 DRE). The total erupted volume of magma was 19. 6 km 3 (cf. 1783 AD Laki, 15. 1 km 3) Proximal tephra deposits are up to 5. 5 m thick at ~5 km distance from source and contain many sub-units which vary in thickness parallel to the fissure, indicating origin at multiple source vents. As well as producing a large volume of products, the Eldgjá explosive phases were very intense, even the purely magmatic ones (Fig. 2). In order to examine the processes leading to fragmentation, occurring in the shallow conduit, the density (i. e. vesicularity) of the eruptive products will be examined. Figure 1. Map showing location of the Eldgjá (red) and Laki (yellow) lava fields. The Eldgjá fissure is composed of several segments: the Western, W; Central, C; Eldgjá, E; and Northern, F. The dashed line marks the extent of which the lava field margins are covered by sediment (Sigurðardóttir 2014). After Thordarson et al. (2001). Thickness (cm) 1000 3. Results • All units sampled at Skælingar are magmatic. They produce unimodal, normal to moderately log-normal density distributions with a mean density of 790 -750 kg m-3 (60 -70% vesicularity) and a range of around 860 kg m-3 (40% vesicularity). Layer 8 and lower layer 9 produce broad density distributions distinct from the other layers. • The Stóragil section contains both magmatic and phreatomagmatic products. The phreatomagmatic samples have characteristic plateau-like distributions with a mean density of 720 -750 kg m-3 (70 -80% vesicularity) and a range of around 842 kg m-3 (45% vesicularity). Layer 5 b to 5 c shows a change from phreatomagmatic- to magmatic-type density distributions. Figure 3. Stratigraphic log of the key-sections: Skælingar and Stóragil. Magmatic sub-units are depicted by white with black pyroclasts, corresponding to clast size – from medium ash to bombs, or black layers representing fine, well-sorted sub-units. Phreatomagmatic sub-units are depicted by the brown, dashed layers. A thin silicic layer is present within the Stóragil section. 4. Discussion • The change in width of distribution between lower and upper Skælingar layer 5 b (yellow star) is due to an increase in the fraction of outgassed clasts. This is an expected result as the gas from each parcel of magma producing a sub-unit will rise to the upper part of that parcel leaving the lower part both degassed and outgassed. • Although broad density distributions are characteristic of phreatomagmatic deposits, the distribution shown by Skælingar layer 8 through 9 (red star) may be due to several separate parcels of magma, at varying stages of degassing, fragmenting at the same time. • The change in density distribution shown by layer 5 a to 5 b (green star) in Stóragil seems to record the transition from phreatomagmatic to magmatic. Given that L 5 a is the last phreatomagmatic sub-unit in the section, this may be due to activity on the fissure moving out from under the glacier. 5. Future Work • Thin-sections of representative clasts from each sample have been made. These will be imaged, both by desktop scanner and scanning electron microscope, and the images analysed (Fig. 4) as outlined by Shea (2010). • The results of this will give variables such as vesicle number densities, vesicle size distributions, and vesicle volume distributions. • These variables will then be used to investigate the fragmentation mechanism in the two eruption style end-members: magmatic and phreatomagmatic. 25 x Eldgja Laki 100 10 1 0. 0 5. 0 10. 0 15. 0 Distance, km 20. 0 25. 0 Figure 2. Plot showing change in deposit thickness with distance from source for Eldgjá and Laki. • • 2. Methods Two key-sections, representing magmatic and phreatomagmatic tephra, were logged (Fig. 3) 100 pyroclasts 8 -32 mm in diameter were collected from <5 cm layers within the sub-units following the method outlined by Houghton and Wilson (1989) The clasts were then weighed, sealed in parafilm and weighed again to calculate volume. Using 2. 85 g. cm-3 as melt density, the density and vesicularity of the clasts (Fig. 3, histograms). 100 x Figure 4. SEM images of bubbles. Three types of bubbles discernable: (1) largest, spherical ones; (2) smaller spherical ones, with thick walls; and (3) small mature, polygonal ones. Bubble nucleation can be very complex. 6. References Figure 3 (cont. ) The sub-units found in Stóragil are found at the base of Skælingar (L 1 -4). Activity began in the west and progressed to the north-east, this is why these layers thin in this direction. The histograms plot number of clasts within bins of density. Each sample set contains 100 clasts. Samples were collected so as to best represent a sub-unit. If a sub-unit changed character with thickness that unit was sampled at least twice to record that change. Houghton B, Wilson C (1989) A vesicularity index for pyroclastic deposits. Bulletin of volcanology 51: 451– 462. Shea T, Houghton BF, Gurioli L, et al. (2010) Textural studies of vesicles in volcanic rocks: An integrated methodology. Journal of Volcanology and Geothermal Research 190: 271– 289. doi: 10. 1016/j. jvolgeores. 2009. 12. 003 Sigurdardottir, S (2014) The Eldgjá lava flow beneath Mýrdalssandur, S-Iceland. Mapping with magnetic measurements. Unpublished MSci. Thesis, University of Iceland. Thordarson T, Miller D, Larsen G, et al. (2001) New estimates of sulfur degassing and atmospheric mass-loading by the 934 AD Eldgjá eruption, Iceland. Journal of Volcanology and Geothermal Research 108: 33– 54.