Sinabung & dangerous lava domes

Sinabung in Sumatra has been erupting for the last few months, prompting regular evacuations of people living near to it.

(Incandescent lava dome, flow deposit and burning vegetation - Reuters/YT Haryono)

Currently, the style of activity is producing a lava dome: where viscous (less fluid) magma builds up on top of the vent. The lava, although seemingly solid, can have a lot of gas trapped inside at high pressures, and is also still very hot. Sometimes chunks of it fall off the dome, these blocks or slabs break apart, releasing the gas, mixing with and heating the surrounding air, forming pyroclastic flows and surges.

Depending on where the lava is being emplaced on the volcano, material may fall off in different directions, thus the areas most at risk can change quite quickly. Dome building eruptions often produce a LOT of flows - the size of which can often follow a power law relationship: many small events, with a diminishing number of larger events*. Lava domes make both managing and communicating the risk very difficult. Often people are evacuated in the anticipation of larger flows, which may not happen for a while (if at all). After a time of small flows, many people naturally want to return to their homes. Then they may be evacuated again, and subsequently return. This process can occur many times, and ultimately people can become very reluctant to leave.

Unfortunately, this kind of relationship is very different to grasp.  In Montserrat, 1997, this process (among other factors) occured - the danger perhaps obvious to the scientists, but people became used to where flows were going and how big they were. Many thought that they understood the speed of them, and unfortunately thought they could escape in time. On June 25th 1997, people on the slopes of the Soufrière Hills, in areas that they were advised not to visit, were caught off guard (despite numerous warnings from scientists) by a sudden increase in the magnitude and energy of the flows during a partial dome collapse, which lead to a tragic loss of life*. 

This has unfortunately been the case in many similar eruptions, from Soufrière Hills Volcano to Merapi. An added danger is that apart from larger flows related to small collapses...lava domes can also produce large vulcanian explosions - which create even more energetic flows, that can sweep down all sides of the volcano at once.

The key thing is to not expect a volcano to always behave in the same way, but rather to think "what could it do to surprise me?".

Despite the fact that the Indonesian scientists are very capable volcanologists and communicators, we just have to hope that the eruption calms down again.




*Loughlin, S. & Baxter, P., 2002. Eyewitness accounts of the 25 June 1997 pyroclastic flows and surges at Soufrière Hills Volcano, Montserrat, and implications for disaster mitigation. Geological Society of London. 
*Loughlin, S., Calder, E. & Clarke, A., 2002. Pyroclastic flows and surges generated by the 25 June 1997 dome collapse, Sonfière Hills Volcano, Montserrat. Geological Society of London. 

4 years ago, in a far away land...

Four years ago today, an earthquake that will be remembered as one of the worst in our generation shook the caribbean country of Haiti. The world watched in horror as the death toll rose to hundreds of thousands.

Construction in Haiti was known to have been generally poor. Even two years before the earthquake, a school collapsed killing nearly 100 children. When the earthquake struck many other buildings, and human lives, had the same fate.

Experts had warned of the major seismic hazard that the island was exposed to just two years before the event, so it was known that the region could be subject to seismic shaking, as is the case for many places. However, it so often remains a a low priority in favour of more pressing matters - the economy, trade, hurricanes, poverty - viewed as risk for the distant future. Ignored in Haiti, post-disaster the major concerns for the country remain as before: the economy, trade, hurricanes, poverty, with construction standards still not enforced properly. 

I can't help but worry that this devastation could have occurred in so many places in the Caribbean and throughout the world; Haiti was simply the 'unlucky' one that day. But what can be done to avoid a similar disaster elsewhere?

Below is a diagram highlighting the three possible ways forward.

The way forward...

Option 1 is asking for trouble.

Option 2 is the bare minimum.

Option 3 will be hard work and costly; if successful will avoid a disaster on the Haitian scale.

I worry that more needs to be done. That authorities need to be pressured to not take each day for granted. In some places, large earthquake will happen, the only question is whether it's tomorrow or in a 100 years. Lessons can be learnt from others mistakes. Learn this Haitian hotel owners lesson now. 

”This is something that we had forgotten for a long time — that Haiti was subject to earthquakes. Now we know. Now we know that we need to build better.” 

Risk perception…not the only important thing?

Why would you live here?! Nestled near the bottom of the Vazcun valley, these homes are in a very high risk area on the slopes of Tungurahua, Ecuador.  Built on top of pyroclastic flow deposits, which have regularly impacted the area in historic times. There hasn't been one to this location for almost 100 years - so despite heightened activity since 1999, people are still here. Sadly - unless the volcano becomes extinct (not likely any time soon) - history will inevitably repeat itself. I just hope not in the near future....

What if I told you that my friend lives there...and his family extended their house...only a few years ago - i.e during a time when the volcano was erupting? What would you think? Would you suppose that my friend and his family don't really know how dangerous the volcano is? What if you then found out that my friend is a very promising volcanologist...? Why on earth would they live there and invest money in their property as they surely know the volcano is very dangerous?? 

I regularly find myself telling people that I am a little dismayed that a ‘risk perception study’ is often the first social science approach taken off the shelf by volcanologists. When we are considering risk reduction, what we want to know is how people might respond to a hazard or forecasted hazard, and what steps they are likely to take to reduce the risk to themselves and their family. How people perceive risk or the ‘potential danger’ from a volcano is important, but it shouldn’t be the first thing we investigate and for me, it doesn’t explain adequately why people might take certain decisions or actions when confronted by risk.

Perhaps I trivialise the issue, but here is one interpretation of what a typical risk perception study is probably looking at:

Survey question: How dangerous do you think the volcano is? 

Answer: not very

Solution? Educate them about volcanic hazards. If they knew how dangerous it was, they wouldn’t live or work there, or they would at least make sensible decisions when we tell them something.

We could even add in a quantitative element – because that of course allows us to really understand something:

Survey question: on a scale of 1- 10 , how dangerous do you think the volcano is? 

Answer: 5

Solution: oooh – if we can educate them so that their answer is the same as our answer (about 7) then they will be safer. Risk Reduction ✓

In reality, the problem here isn’t actually about risk perceptions – it is about what we think they might tell us:

Thinking that people can have ‘bad’, ‘incorrect’ or ‘wrong’ risk perceptions isn’t helpful – it assumes that we all have the same way of calculating risk, or that there is some objective ‘true’ risk.

Assuming that we can ‘change’ or ‘improve’ people’s risk perceptions, to bring them more in line with scientists’ perceptions is a concept from straight out of the idiots guide to educating knowledge deficient publics:

The deficit model  of risk communication– suggests that the lay-public will make irrational decisions based on limited information about a problem.  This comes about because scientists often only consider objective science as the most important information. Whilst the amount of knowledge that the lay public has is a factor in their response, we have a responsibility to not disregard other factors as irrelevant. I very much doubt that we deliberately do this…but our obsessive focus on people’s risk perception doesn’t pay much attention to other things that might influence their decisions – as we know that the public’s judgments of risk aren’t necessarily based on the amount of information that they have, but more often than not on their ‘world view’, their social or political views and their circumstances*.  None of us, even scientists, simply process information with associated heuristics and biases, and then make a decision. We aren’t machines. Rather we like to attach meaning to issues. Further to this…much of how we make sense of the world is actually not individualistic, but a socially constructed reality. Particularly in volcanic areas, knowledge and meaning about risk is transferred between social groups, often passed down between generations. 

People create social representations about risk – reaching a consensual understanding of what did or could happen – we build common sense about an issue by anchoring and objectifying it. We anchor by drawing on shared experiences from the past, making an unfamiliar issue familiar amongst our group. Then we often objectify things, by representing them in a way that is easier to grasp, using more familiar terms. For example, people often objectify ash plumes, which are lit up and incandescent as “smoke and fire”. It may be different to the scientific reality, but anchoring and objectifying is the way in which social groups make sense of new or unfamiliar situations. For example, the way that a social group might have been affected in a previous evacuation and how they have made sense of it, may have a far greater effect on decisions they will make in a future risky situation. Thus, their risk perception of a volcano might suggest that they know it is incredibly dangerous, but if the community only talks about how last time there was an evacuation, they were looted or lost their animals, then how dangerous they perceive the volcano to be may have little baring on their decisions. If we simply asked them how dangerous the volcano is, we will get a false positive answer.

Where risk perceptions focus on knowledge and information, risk representations focus on meaning and understanding. What is more useful for us as volcanologists to know? How or why people might behave in a certain way, or what they know, which may or may not then affect how they behave?

I'm not suggesting that we don't try to understand risk perceptions - but let's try to not make it the first thing that we do. Instead what we could be looking for isn’t how dangerous people think a volcano is, but what do they think about it in relation to other hazards or life situations. When we frame the problem like this, we are able to attach meaning to people’s views about volcanic hazards - and then you can understand the factors determining why my friend and his family live where they live. 

  

To read more about social representations of risk, have a look at the paper below as a starter by Helene Joffe. Sorry if you can't access it...

Joffe, H., 2003. Risk: From perception to social representation. British Journal of Social Psychology, 42(1), pp.55–73.

*This is based on the work of Paul Slovic (among others)

Would you send your kids to this primary school?

As part of my research I am reviewing past mission reports complied by the Earthquake Engineering Field Investigation Team (EEFIT), a group of expert engineers who visit countries shortly after seismic disasters to assess the engineering damage.

Over and over again the reports state that even if adequate building engineering codes have been introduced and enforced, damage and collapse of new design structures continues to be caused by the poor quality of construction. Seismic design is one issue, good quality construction is the next.

I came across this newly constructed concrete column (see photo below) whilst assessing the seismic capacity of a primary school in the Caribbean. This column to beam connection (along with those in the distance) has substantial performance requirements as they are responsible for supporting lateral loads applied to the building. Simply, lateral loads arise primarily though wind pressure, pushing agains the building and during seismic ground shaking.

During an earthquake this structural 'weak-link' may cause the column to fail and as you may be able to imagine, if a column fails it is no longer able to support the beams/floor/roof connected to it and the building is only going one way - down.

But education is important, and this developing country has large but very infrequent earthquake (around every 150 years). The last major earthquake was 158 years ago posing a very real risk decision taken by many people in seismically active areas.

Would you live in this house?

Granted, it is on the idilic Caribbean island of St. Vincent, with an envious climate, glorious beaches and stunning landscapes, but it is also subject to infrequent but large earthquakes. 

As a structural engineer, my job is to ensure that structures withstand the forces of nature, whether that is wind, rain, snow, people, bathtubs full of water or seismic shaking. There are design rules, codes and standards, guidelines and common structural principles which apply to different loading scenarios throughout the world - for example snow loading in Cameroon will be different to that in Canada. 

For a structure to withstand seismic shaking there are a set of principles too. Survey of damage after destructive earthquakes can clearly highlight the reasons for failure, and so as each earthquake passes we learn more. One of these general principles is that, put simply, columns should be bigger than beams. Imagine a column failing, the floor structure will fall and likely bring the whole structure down with it. However if a beam fails, a localised section of floor might fall down but the columns are still intact, hence the structure may not undergo catastrophic collapse.

Another seismic design principle is to avoid 'soft storeys'. These are storeys of a building that have significantly less structural strength and stiffness, e.g. a tall office block with an open ground floor car park with less columns and no walls. When the building shakes, the soft storey is likely to give way and collapse and the building ends up a storey shorter. See here.

So look again at this house.

What do you think?

P.s. Other things that may be a worry: landslides, volcanic hazards, tsunamis, hurricanes, etc.