jump to navigation

Stress change may provide clues to possible eruption locations 27 September 2010

Posted by admin in Africa, current research, Ethiopia, geoscience, volcano monitoring, volcanology.

It’s all rifts, dykes and magmatic intrusions at Nature Geoscience right now. Along with the paper by Pallister et al on the Saudi quake swarms of 2009, the journal is hosting advanced online publication of a paper on a recent episode of dyke emplacement in the Afar region of north-eastern Africa: ‘Stress transfer between thirteen successive dyke intrusions in Ethiopia‘, by Ian J. Hamling et al.

The study looks at the emplacement of thirteen magmatic dykes in north-eastern Ethiopia between 2006 and 2009. A rift zone produced by the spreading boundary between the African and Arabian plates runs through this region; most such rift zones are situated on the ocean floor, so this remote area provides a valuable opportunity to study the processes associated with spreading plate boundaries without getting one’s feet wet. A team led by Ian Hamling of Leeds University measured changes in ground tension associated with each successive dyke emplacement, and found that subsequent eruptions were most likely in locations where the tension had been increased. Although the initial level of stress along a rift zone that becomes active is unknown, measurements of stress transfer will reveal whether eruptions in one location cause compressive stress change (clamping) or tensile stress change (unclamping) elsewhere. New dyking would be expected in locations subject to unclamping – in other words, where the ground has been stretched and is under increased tension – and the study shows that such is indeed the case: ‘the mean percentage of opening in unclamped sections of the rift has been 70%, with seven of 12 dykes having over 75% of their opening in regions unclamped by the previous intrusion’. The study concludes: ‘This result indicates that the stress change, induced by a new dyke, is a controlling factor on the location of future events and should therefore be incorporated into routine volcanic hazard monitoring’.

  • Ian J. Hamling, Tim J. Wright, Eric Calais, Laura Bennati & Elias Lewi, ‘Stress transfer between thirteen successive dyke intrusions in Ethiopia’, Nature Geoscience, published online: 26 September 2010 | doi:10.1038/ngeo967 [abstract]

Pinpointing where volcanic eruptions could strike – EurekAlert, 26 September 2010

The Volcanism Blog



1. bruce stout - 27 September 2010

Brilliant, now we are getting close to a question I have long harboured. Do intrusive events follow fault lines purely for mechanical reasons (i.e. gaps/ cracks/ weaknesses in the country rock) or is there some geochemical element involved as well (i.e. the stresses on a fault alter the rocks chemically (or even simply thermically) in a way that is conducive to fault propagation? any takers?

2. Raving - 27 September 2010

“Stress change may provide clues to possible eruption locations”

– Isn’t this like saying … the interference pattern caused by the superposition of multiple anisotropies at assorted scales.

3. admin - 27 September 2010

That’s a good question, Bruce, and I just don’t know what work has been done on it. My first thought is that any geochemical alteration to the rock caused by the stresses of a fault alone (whatever form such changes took) would be a minor factor in determining magma paths compared to the more straightforward mechanical factors. In the case of this study the issue is a mechanical one of tension.

Raving: well yes, but I prefer my title.

4. bruce stout - 27 September 2010

Thanks Ralph, I am sure you are quite right that mechanical factors play by far the greater role in fault propagation. Callan had something on stoping not long ago that even illustrated it in Patagonia. Nevertheless I am still intrigued as to what goes on on a fault line at a chemical level when you think about hydration and an influx of heat to rocks that are already under stress. The only problem with the question is that I am almost certainly not going to be able to understand the answer!!

5. Gijs de Reijke - 29 September 2010

Very interesting. This process, although it hasn’t lead to volcanism for more than 15 million years, also takes place in Southern Germany, where the Kaiserstuhl volcano was formed in several phases between 19 and 16 million years ago. The formation of the Kaiserstuhl was part of the rifting process that formed the Upper Rhine Graben, a process that has a lot of similarities with the East African Rift.

Sorry comments are closed for this entry