Perception of volcanic hazards in Iceland

The eruption may be subsiding a bit, but there is still a lot of discussion (and arguing) centered around the Eyafyallajökull event. It's not entirely surprising; most people in Europe don't have to deal with active volcanoes, and the last time an Icelandic one caused widespread trouble was in the 18th century. But what about the Icelandic response? One might assume, given the prevalence of volcanic and geothermal activity in Iceland, not to mention hazards caused by volcano-water interaction, that Icelanders might be better prepared than other Europeans to deal with natural hazards. But is that really the case?

In "Resident perception of volcanic hazards and evacuation procedures", published in 2009 in Natural Hazards and Earth Systems Science, Australian and Icelandic scientists set out to evaluate how Icelanders perceived risk and what their response would be to an evacuation drill for a jökulhlaup hazard. The study was conducted in March 2006 in the jökulhlaup hazard zone of Rangávallasýsla, a region immediately adjacent to the Mýrdalsjökull and Eyafyallajökull glaciers and the Katla volcano, which is notorious for producing jökulhlaups.

Figure 1 from Bird et al. (2009). The jökulhlaup hazard zone of Rangávallasýsla. The hazard zone is the maximum area that a  catastrophic jökulhlaup is expected to flood. Evacuation centers are represented by blue triangles.

The authors of the study used a combination of methods to assess the reactions of residents and emergency officials: they directly observed an evacuation drill, did face-to-face interviews with officials and residents, and distributed surveys to those involved in the drill. The authors also discuss the parameters of the drill:

If an eruption is imminent residents would be notified via a text message to their mobile phone. If residents do not have a registered mobile phone number a recorded message would call through to their landline. Upon receiving this message residents have 30 minutes to prepare to evacuate. However, if an eruption occurs without precursory activity, residents will be instructed to evacuate immediately. Before leaving, they are required to hang the evacuation sign outside their house to indicate that they have left. Certain residents in each region have volunteered to ‘sweep’ their local area to ensure their neighbours have left for the evacuation centres...

To test the proposed evacuation plan the ICP conducted a full scale evacuation exercise on 26 March 2006 in Rangávallasýsla. Approximately 1200 residents live within the hazard zone (K. Þorkelsson, personal communication, 2006) and for the purpose of fully testing the evacuation plan residents were not informed of the timing of the eruption scenario. Instead residents were instructed to go about their business as usual until they received an evacuation message (R. Ólafsson, personal communication, 2006). The mock eruption began at 10:55 local time (LT) and the first evacuation message was communicated to residents at 10:59 LT. Residents then had 30 minutes to complete the instructions on the hazard sign (Fig. 2) before evacuating their homes to their designated centre.

So what were the results of the evacuation drill and the study?

• Many residents did not receive notice of the evacuation, but about 65% of the local population still registered at evacuation centers. Some of the reasons cited by the remaining 35% for their non-participation included lack of communication from officials, reluctance to leave their livestock, or that they were simply not interested in the drill. The response from those who did participate, however, was overwhelmingly in favor of the drill.
• 71% of evacuation participants were able to correctly describe the evacuation procedures they were supposed to follow during the drill, and 94% were able to define what a jökulhlaup was (and knew that it was the major hazard associated with an eruption of Katla).
• Many of the residents of towns on higher ground stated that they would remain in their homes rather than evacuate, citing that it was safer there than on roads and that they thought the flood stage of the glacial drainage would be too low to reach them.
• Many of those surveyed - especially farmers - did not think that 30 minutes was enough time to prepare for an evacuation, since they had livestock to care for in addition to dealing with their homes and families.
• "None of the participants from the 18–30 year age group and very few from the 31–50 year age group could correctly describe a brief volcanic history of Katla." Some residents who had family members who had seen the 1918 Katla eruption had knowledge of what Katla was capable of, but this has apparently not been passed down to their children.

This paints an interesting picture. Most of the residents, even if they didn't participate in the drill or completely understand the potential hazards associated with an eruption of Katla, were still very well informed about what they should do in an emergency. Many evacuated even though they didn't receive a direct message from emergency officials (mostly because of community volunteers who helped spread the word of the drill). But it is troubling that a large number of people living near Katla (and Eyafyallajökull) knew very little about the past activity of the volcanoes. This is often the case when volcanic disasters have passed partially or completely out of living memory, but given that volcanic activity is extremely common in Iceland, it's not particularly reassuring. The authors suggest that a lack of outreach by public officials may be the cause for this:

Our participants are aware of jökulhlaup, tephra, lightning and rock fall hazards but they have not been provided with enough information to enable them to make an informed decision on whether to evacuate or take shelter in place and how to best protect their livestock.

Finally, the authors comment on some of the underlying problems with the evacuation itself, most having to do with communication issues (again):

Results from our study highlighted problems associated with communication during the evacuation exercise and the possible need to find alternative modes which do not rely so heavily on technology. In light of this, scientists and emergency management officials should collaborate with media agencies and the public in order to promote the use of media resources and, to ensure hazard information is accurately distributed in an understandable form. Furthermore, the importance of the sweepers’ role during an evacuation should be emphasised as they may provide the only communication link between emergency management and farming communities. Recent public meetings which involved residents in risk mitigation efforts are a positive step toward empowering residents with evacuation procedures and preparedness strategies.

What's the bottom line? It's an interesting one: people near this volcanic center in Iceland seemed to be fairly well informed about what they should do in an evacuation. But they weren't necessarily as knowledgeable about the hazards that necessitate the evacuations, even though they live very close to an active (and now erupting) volcanic center. (Given the recent eruptions, I suspect that a follow-up study would show a distinct change in this observation. If anyone comes up with one, I'd be interested to see it.) This study does emphasize again the importance of good communication between scientists, emergency officials and the public; in an emergency, if people are better informed about hazards and what they should do to avoid them, evacuations will run more smoothly and officials will waste less time dealing with confusion and misinformation.

Bird, D., Gisladottir, G., & Dominey-Howes, D. (2009). Resident perception of volcanic hazards and evacuation procedures Natural Hazards and Earth System Science, 9 (1), 251-266 DOI: 10.5194/nhess-9-251-2009