As per the usual pattern of AGU blogging, I’ve been trying to get caught up with other things after returning from AGU, so naturally I’m only getting to writing about the meeting a week after it happened. I’d better finish this up at some point, so I’ll combine Thursday and Friday’s activities into one post (and then move on to posting photos from San Francisco, which has some fabulous geology that I finally got to check out in my spare time).

On Thursday I went to my second press conference, a joint session between three researchers looking at the influence of weather on earthquakes and eruptions. Fabien Albino presented his research about how the loss of ice caps on Icelandic volcanoes can cause load and stress changes which affect magmatic reservoirs; these changes could lead to a number of things, including propagation of the magma upward in a conduit, failure of the reservoir, or even an increase in the production of magma. He linked the results of models investigating these stress changes to the seasonal and long-term weather effects on the ice cap at Katla (where there have been no observed historical eruptions during colder months, when there is more ice, and more eruptive events seen in warmer months, when the ice cap is thinner). Thomas Ader investigated earthquake occurrences in Nepal, linking an increase in earthquakes on the main Himalayan fault to the stress changes caused by loading and unloading on the Indian plate due to rain from monsoons. The load from monsoon water apparently bends the Indian plate and creates stresses that oppose tectonic stress, relieving the stress on the Himalayan fault; when the water goes away in the dry season, the Indian plate “unbends” and earthquakes become more common. Finally, Shimon Wdowinski suggested that large earthquakes in Haiti and Taiwan could be linked to cyclones in the months before the earthquake; the cyclones are thought to increase erosion and unloading on faults, which might allow the fault to slip sooner than it might otherwise have done.

In the afternoon, I attended the Daly Lecture, which this year was given by Colin Wilson, a volcanologist at the Victoria University of Wellington. Dr. Wilson discussed explosive rhyolitic eruptions – big ones – and in particular those occurring in the Taupo Volcanic Zone in the central North Island of New Zealand. These eruptions are of particular interest to volcanologists because there have been very few historic ones (Novarupta in 1912, Tuluman in 1953, and Chaiten last year are three examples), and we know very little about what happened during any of the prehistoric eruptions. Some questions that Dr. Wilson discussed included how long the eruptions take, whether they are fueled by one magma body or many, what controls the start or stop of an eruption, and how we can constrain the timescales. Erik Klemetti has a great summary of the lecture over at Eruptions, so I won’t repeat it here, but it was a really intriguing session.

On Friday morning I spent most of my time in a volcanology session listening to talks about cyclicity in eruptive activity. One talk I listened to with some interest was given by Jenny Suckale of Harvard, who suggested that eruptions of lava at Stromboli could be better explained by a plug-filled conduit than the generally accepted bubble-burst model of eruption. In the plug model, a crystalline mush underlies the summit of Stromboli (which has several active vents), and explosions of lava occur when gas pressure underneath the plug overcomes the yield strength of the plug. Episodicity in the eruptions is explained by cycles of draining and recharging gas beneath the plug, rather than the idea of gas slugs rising through a magma-filled conduit at regular intervals. I find the model interesting, as it also provides an explanation for the crystal-rich scoria produced by these explosions, but it also seems to require that a very large single conduit, with a plug wide enough to encompass multiple vents which are tens of meters apart, has to exist beneath the summit of Stromboli. This seems a little implausible to me, but perhaps I was misreading the presentation. It will be interesting to see the reaction to this idea, at any rate!

Friday afternoon was about the point when my brain shut down and refused to absorb any more science, an inevitability at any conference as massive as AGU. I wandered back to my hotel for a bit of a rest and some packing, and joined a number of other conference-goers on a flight out of San Francisco the next morning. Traveling aside, I’ve really enjoyed both of my trips to AGU, and I’m definitely looking forward to attending the meeting again next year!