Reconstructing ice at Blahnukur

Reconstructing the ice thickness at Bláhnúkur

Bláhnúkur was a small, non-explosive eruption that occurred entirely beneath the ice sheet. The composition of the magma was rhyolite (a high silica content magma which makes it thick and gloopy by comparison to the more common basalt). As a result of these factors, two main products were formed: hyaloclastite (a fragmental material formed by magma-water interaction) and lava lobes (mounds of lava which protrude out of the hyaloclastite like warts.

Bláhnúkur is a small volcanic edifice within the Torfajökull complex in south Iceland (see map). She erupted about 100,000 years ago and therefore within the last glacial period (commonly referred to as the Ice Age). At this time Iceland was covered in a thick glacier but just how thick was the ice over Bláhnúkur when she erupted?

To answer this question we adopted a technique which involves collecting glassy (uncrystallised) samples and measuring their water content. This can then be used to determine the pressure that the lava felt when it cooled and therefore how thick the ice was when the eruption occurred (click here for more information on this technique).

A lava lobe protruding out of hyaloclastite

The outer surface of the lava lobes tends to be very glassy, suggesting that the magma cooled relatively quickly here. This 'glass', known as obsidian, provides an excellent material for reconstructing ice thicknesses and so most of our samples were collected from lava lobes, however, some dykes and sills were also sampled.

The map on the left shows where on Bláhnúkur samples were collected from. They have been colour coded according to sampling location.

Sampling locations (modified from Owen et al., (2012)

The water content of the samples was measured with FTIR and plotted against elevation. Various solubility pressure curves were then plotted. These show the water content expected at each elevation for various different ice thicknesses (thicker ice means more pressure means more water retained) - as shown by the numbers in brackets. The numbers without brackets represents the elevation of the surface of the glacier. (click here for more information on this technique.)

Three trends emerged, leading to three explanations...

Water data with solubility pressure curves (modified from Owen et al., (2012))

1) samples from the Top Ridge, Feeder Dyke, Northern Slope and Grænagil plot to a solubility pressure curve representing ice 400 m thick.

2) samples from the Lobe Slope and Brandsgil plot to a solubility pressure curve representing higher pressure. However, the data is more scattered and it would be unlikely that the ice thickness varied this much (100s of meters) during the eruption

3) samples from A-Ridge are relatively water-poor (particularly at low elevations) and show an increasing trend of water content with elevation which is the opposite trend we would expect for a subglacial eruption.

explanations

Bláhnúkur erupted under ice approximately 400 m thick, meaning that the glacier had a surface topography roughly 1000 m a.s.l. over this part of Iceland when Blahnukur formed.

The Lobe Slope and Brandsgil samples formed intrusively (within the fragmental material) where the hyaloclastite provided additional loading pressure, allowing more water to be retained. This is supported by field evidence (as shown below)

A-Ridge samples as well as being water-poor, also lack bubbles (unlike the other Bláhnúkur samples). This suggests that A-Ridge samples (particularly those that erupted near the edifice base) had low initial volatile content. If there wasn't much water in the magma to begin with, it could explain why there isn't much in there now.

The Lobe Slope has got significantly more lava lobes than other visible parts of Bláhnúkur (e.g. the slope to the right). It also has plenty of evidence of erosion e.g. gullies exposing the older, orange rhyolite underneath. Perhaps they are just as abundant on other slopes but simply not exposed...?

The Brandsgil sill, overlain by hyaloclasite. Perhaps there was more which has since eroded away...?

Conclusions

(Owen, 2013)

Bláhnúkur erupted under ice 400 m thick. The Lobe Slope and Brandsgil had additional loading from hyaloclastite, meaning that they formed intrusively within the fragmental deposits. Samples from low elevations on A-ridge were volatile undersaturated (i.e. they had a low initial water content) and so could not record the ice thickness.

This has led to the following new model for the construction of Bláhnúkur:

(Owen, 2013)

1) Magma rises to the bedrock surface under an ice sheet ~400 m thick. This melts a cavity into the glacier base. As the magma continues to rise, quench fragmentation occurs and the feeder dyke becomes surrounded by a growing mound of hyaloclastite.

2) Dykes and sills extend out from the feeder dyke producing ridges.

3) Lava lobes bud from the sills and dykes producing both intrusive and extrusive lava lobes, such are found on the Lobe Slope and A-Ridge, respectively.

4) The eruption terminates with intrusion of a large sill into the volcanic pile at Brandsgil.

5) After the eruption has ceased, the ice recedes, and considerable erosion from the Lobe Slope exposes the lava lobes.

If you have any comments or questions, please post them on the bottom of the page or contact me

The literature behind it:

This webpage is a summary of the findings of Owen, J., Tuffen, H. and McGarvie, D.W. (2012) Using dissolved H2O in rhyolitic glasses to estimate palaeo-ice thickness during a subglacial eruption at Bláhnúkur (Torfajökull, Iceland), Bulletin of Volcanology, 74(6), 1355-1378