Thursday, August 2, 2018

Communicating about Agung, how did that even happen?

At the end of 2017 I dropped everything in my life to communicate the Agung crisis online. Just me, my two cats, and my laptop at the kitchen table. All night, every night, for just over three months.

I am going over the notes that I took during the Agung crisis in order to prepare a talk that I will be giving at the Cities on Volcanoes conference next month. From the beginning, it was very clear to me that this was something that we, as a global volcanology community, need to learn from. During the crisis I took a lot of notes. I took screenshots of thank you messages (showing what was helpful), critics, hard questions, points of confusion, evolution of the eruption, and much more. This was all with the goal to one day learn from this experience so that the next time this happens, we will be better prepared (turns out it is already happening with the Kilauea eruption and USGS is doing a great job!). 

So, what the heck happened?

On 21 September, 2017, I was living in Pittsburgh trying to find a job so that I could stay in the country. I had just completed my volcanology PhD, and had been to a conference in Portland. I had no idea what I was going to do. I was watching meteorologists communicate on Twitter about the 2017 hurricanes and I was both impressed and concerned. Impressed at how well they were doing, and concerned that volcanology would have a similar situation and we would not be prepared for it. The situation I worried about was a volcanic crisis (increased unrest in a populated area) with intense international interest in the age of social media and 24-7 online media. This is something I had spent quite a lot of time thinking about and I had started to talk to other volcanologists about it. I did not, however, have any idea that Agung was about to happen. There is no way I could have.

On 22 September this was about to change. On the 20th I posted my first tweet about Agung. I did not think much of it. On the 21st I tweeted about Agung another two times. Just like I had many times with many other volcanoes since joining Twitter in 2013. On the 22nd I tweeted that the Alert Level had increased to IV out of IV This caught my attention. I knew this was potentially bad news for the locals, but I didn’t put a lot of thought into it beyond that. 

Note the 17 retweets. This was pretty normal attention for a tweet back then.

A little before midnight that night I received a Facebook message from a Kiwi friend in Bali:

Is the Bali volcano going to explode?

I'm in Bali now


I'd rather rely on a better source than random mates/article on FB

I'm weird like that

He just told me: The volcano is imminent alert of eruption. Flights are getting canceled

I had not seen an alert for an imminent eruption... I started looking for more information. I could easily find fear-mongering articles that spelled out imminent doom, but nothing to back that up. I realized that in a potentially dangerous situation people could not find the right information. I was very, very concerned by this. I cared. A lot.

Agung had produced a deadly eruption a few decades earlier and no one knew what it was about to do (hindsight is great). People needed information. People in Bali needed information and people about to travel to Bali needed information. The media needed information. The world needed information. As experts, we have the information they need. As far as activity specifics go, that is all up to the official agency (CVGHM and BNPB in this case), but we can direct people to potentially life-saving information. What on Earth is a pyroclastic flow? A lahar? What is volcanic ash? What do people do to stay safe? Who should they be listening to? What do the words in the official warnings mean? As a trained volcanologist sitting there with this information, I felt a responsibility to share it. Even to my small Twitter following at the time.

There were two large issues with this situation: 1) Unless you are Indonesian, chances are you don’t know where to get the official information or what websites to trust (USGS, GNS, IVHHN); 2) for the local websites, we can use online translation services to translate the information into English but there were some key words that did not translate correctly, changing the meaning of the message.
The Magma Indonesia website, and important source of official information.

Using Google translate (I do not speak Indonesian) and volcanology experience (not specifically on Agung), I started tweeting the official information in English, with the links to the original information. No big deal, right? Wrong. Big deal. Very big deal.

People were searching online for information. This was a big deal in terms of global media. A volcano that had killed over 1,100 people in 1963-1964 waking up in a tourism hotspot. Very. Big. Deal.

People found my little volcano-based Twitter account. The questions started flooding in, then the media requests. I am still in awe of how polite, kind, and patient everyone on Twitter was through this entire event. Thank you.

Those next few weeks are now a blur, the main things that happened were: my waking hours switched from day to night (as any shift worker can tell you, this feels awful); I reached out to a bunch of specialists (volcanologists, social scientists, emergency mangers, science communicators) around the world to make sure I was doing everything right (in the first month this list grew to 60, then I lost count); I started working with media to explain basic volcano terms, history, and processes so that reports were accurate; I wrote a blog post giving links to official information so that people could actually find it; I made it very clear that I was not an official and I was not interpreting the data in any way – just passing on the official word and general volcano information; I assembled an online support group of mostly volcanologists different time zones; I worked with several people in Bali who were working to help disseminate the right information (like this) to locals.
When I started getting emails telling me about this article, that was a big 'wait, what is happening??' moment. I was so focused on the volcano that I completely missed the growing interest in who I was.

Every waking moment I was on high alert, double checking everything I said and trying to make sure I stuck very carefully to my self-made rules that I had written out in front of me.

I was very well supported behind the scenes and I frequently asked experts to check what I was doing. Input from volcanologists around the world was crucial. There was one instance where the feedback was not so great. People who I highly respect got the idea that I was putting my own interpretations of the monitoring data out there – a very big no no and I knew this. Doing that can undermine the hard-working official volcanologists who I was working so hard to direct people towards. I was devastated. If I am honest, I still haven’t fully recovered from it. After seeking a lot of reassurance that the claims were unfounded I pushed forward and keep doing what I could to help. Sometimes in life we are given a turning point where we can take the easier path or do what we know needs to be done to help others. This was that moment.

So, this is how this all started. I saw that people needed help to get important information and I dove in without knowing where it would lead. Agung erupted about two months into this crisis but damage to the tourism industry had already been done by fabricated and fear-mongering tabloid headlines. It was not a huge eruption but as we know in volcanology, it is not only the eruption itself that does the damage. Today Agung still sits on Alert Level 3 out of 4. It is still active. The authorities are still working very hard to monitor the situation and to keep people informed and safe.

I will be giving a talk “#AgungErupts: How Twitter and 24-hour media are changing the roles of volcanologists” in a month to discuss this event with the broader community. The way people search for information during an event has changed. We need to evolve with it.

Abstract for my upcoming Cities on Volcanoes talk.

Thank you to everyone around who helped me through this. Thank you to my dedicated online support team. Thank you to everyone who shared the right information (there are also plenty of other scientists online!). Thank you to everyone in media who I worked with, you were all fantastic. Thank you, most of all, to the local volcanologists and authorities who did, and continue to do, all the real work.

Monday, July 2, 2018

Crystal Clocks: How minerals in magmas can be used to unravel what happens before an eruption (Guest Blog by Dawn Ruth)

- Guest blogger Dawn C.S. Ruth @rockdoc11

Keeping time with volcanoes

Hi everyone. I’m neither Alison nor Janine. My name is Dawn C.S. Ruth and, like our fearless leaders, I also study volcanoes. However, where Alison uses experiments to delve deeper into volcanic processes, and Janine uses satellites to spy on volcanoes, I look at the minerals to see how magma moves and behaves before an eruption. One of the reasons I think volcanoes are so cool is because there are so many different ways to study volcanoes (and geology in general).

You want to know a little more about me, the blogging interloper? Well, a long time ago….

No really, my geologic history spans 2.54 billion years. I actually started doing research on impact ejecta from the Archean; this work formed the foundation of my undergraduate research project. When a chance to go to Antarctica and study Erebus volcano came up, I took it immediately. So I went from studying 2.54 billion year old rocks to studying the gasses coming out of Erebus. Funny story – my sister still calls me a gas sniffer. Ahhh, siblings.

Anyway, after my masters I worked for the state of New Mexico as a petrographer and then moved to the University at Buffalo to study volcanoes again. It just so happened that Llaima volcano (Chile) started erupting in January 2008, right before I began my PhD. As a result, Llaima became the focus of my doctoral research. Ok enough about me.

How does one, anyone, study volcanoes using minerals? Oh, let me count the ways. There are many blog posts I could provide, but I’ll focus on one. I use the chemistry of minerals to get at time.

Volcanoes erupt periodically, and we as volcanologists want to know when the next eruption is going to occur. By looking at the mineral chemistry we can get at the timing of processes before the eruption, which in the future we might be able to link to unrest like earthquakes and deformation.

I use element diffusion in minerals to get at the time of magmatic processes. Before I go there, let’s chat a little about diffusion.

Diffusion is a process that occurs wherever there is an unequal distribution of chemicals in a given space/volume/area. Here’s a classic analogy. Imagine you are sitting on one side of a room and some sprays perfume on the other side of the room. The chemicals that make the perfume are unevenly distributed in the room. Over time the perfume will diffuse (spread out) throughout the room, and eventually you will smell it. A similar process occurs in minerals.

In a volcano, minerals are sitting around in magmas. These minerals will have one chemical composition. We think that eruptions are triggered by injections of new magma into the shallow parts of volcanoes. When this happens the minerals that were already present start to grow new material, with a different chemistry. Now the crystal has an unequal distribution of elements (i.e. chemical gradient), which will now diffuse.

Ok, that’s a lot of background. So what do I really do?

I collect samples at volcanoes, and make thin sections, which are very thin slices of rock that allow folks, like myself, to investigate the chemistry and textures of a rock. I use a special microscope called an electron probe microanalyser (EPMA) with a backscattered detector to find minerals with the uneven element distribution. Using the EPMA, I collect a chemical profile across the chemical gradient and then use numerical models to model how long it took that chemical gradient to form. We call this a timescale. I try to analyze as many samples as possible, because nature can be messy and there will never be single timescale.
 I collected rock samples from volcanoes, and brought them back to (often very cold) labs, where I collected the images.  The different grays in the image tell me something about the chemistry and that there is an uneven distribution of chemistry in the minerals. I collected actual chemical data (see the red and yellow lines on the greyscale images) and compared them with a numerical model.  The shape of the profile gives me time.

Once I have collected these, I then try to interpret (or develop a scientific story) that explains the timing and magma movement before an eruption. By looking at data from multiple eruptions, I can get a general sense of the timing of magma movement before an eruption. At Llaima volcano, the one I studied for my PhD, I found that magma movement and injections are occurring frequently and they seem to pick up before an eruption. This makes sense because Llaima is a persistently active volcano that needs lots of magma to stay active. This accelerating behavior seems similar to that observed at other volcanoes like Etna, Mayon (another volcano I studied), and even Mount St. Helens.

We are only beginning to understand how minerals can record these events. We even have a cool name for it; we call them crystal clocks. As we move forward, we hope to take these timescales and then compare them to the seismic records and see if there is a link between was the crystals records and the seismic activity. Hopefully, we can then move toward a forecasting model to help folks plan around a pending volcanic eruption.

For more information on:

Crystal Clocks:

2018    Ruth, D.C.S., Costa, F., Bouvet de Maisonneuve, C., Franco, L., Cortés, J.A., Calder, E.S. Crystal and melt inclusion timescales reveal the evolution of magma migration before eruption. Nature Communications. doi:10.1038/s41467-018-05086-8 
Open Access link:

2016    Ruth, D.C.S., Cottrell, E., Cortés, J.A., Kelley, K., Calder, E.S. From passive degassing to violent Strombolian eruptions: Deciphering the triggering processes of the 2008 eruption of Llaima Volcano Chile. Journal of Petrology, 57(9), 1833-1864, doi: 10.1093/petrology/egw063. 

Cool things at Llaima: 
2014    Ruth, D.C.S. and Calder, E.S. Plate tephra: Preserved bubble walls from large slug bursts during violent Strombolian eruptions. Geology.  doi:10.1130/G34859.1. 

Gas sniffing at Erebus:
2008    Sweeney, D.C., Oppenheimer, C., and Kyle, P.R. Sulfur dioxide emissions and degassing behavior of Erebus volcano, Antarctica. Journal of Volcanology and Geothermal Research, 177, 725-733.

2005    Oppenheimer, C, Kyle, P.R., Tsanev, V.I., McGonigle, A.J.S., Mather, T.A. and Sweeney, D. Mt. Erebus, the largest point source of NO2 in Antarctica. Atmospheric Environment, 39 (32), 6000-6006.