Modelling the Hazard Footprint and Consequences of Lava Flows in an Urban Environment

Abstract

Nearly 10% of the world’s population lives within 100 km of a volcano that has been active in the Holocene; living so close to a volcano means potentially being affected by volcanic hazards such as ashfall, pyroclastic density currents, volcanic ballistics, and lava flow Numerous studies have examined the likely extent and impacts of ashfall, pyroclastic density currents, and ballistics. In contrast, very little detailed research has investigated lava flow hazards or impacts. Lava flows are commonly considered to fully destroy any structure with which they come in contact, i.e. cause binary damage. Anecdotes from modern eruptions suggest, however, that lava flows may cause a spectrum of damage and that there may be options for mitigation. An impact assessment framework is used in this study to determine the impact of lava flows on the built environment. First, a collation of modern effusive eruptions that have threatened and/or inundated inhabited areas and their supporting networks is created. These events are analysed for research gaps and lessons learned using a deductive thematic analysis based on eleven data types that could aid in lava flow hazard and risk studies. Second, interviews and focus groups with scientists, emergency managers, their industry partners, and community members help fill the research gaps identified for four of the studied eruptions. These discussions have also revealed that research stakeholders are especially interested in how buried infrastructure would fare under a lava flow. To address this, a series of laboratory molten rock flows were created. The temperature profiles under the molten rock were recorded as the rock cooled and serve as the basis of a computational heat transfer model. To verify the constrained heat transfer model, magnetic properties of soil samples from under the June 27th Lava Flow (2014- 2015; Hawaii, USA) were analysed to determine the peak temperatures to which they were heated while the flow was active. The heat transfer model recreated the peak temperature profile without requiring any alterations. To apply the resulting hazard model in an impact assessment, Birkenhead scenario of the Determining Volcanic Risk in Auckland (DEVORA) scenarios was selected as an Auckland Volcanic Field (AVF; New Zealand) case study. Lava flow simulation modelling was required to determine the elements of the built environment that would be exposed to the lava flow. As of January 2020, there are 31 lava flow simulation models, each with different strengths and weaknesses. A selection framework was created to aid in determining which model(s) are best suited to different purposes. It takes required outputs, available inputs, and stakeholder requirements into account. In the DEVORA case study, the selection framework indicated that MOLASSES would be the most appropriate model for the AVF buried infrastructure impact assessment. The lava flow simulation modelling was undertaken twice, once on a digital elevation model and once on a digital surface model. The results were compared using a modified Jaccard coefficient and serve as a reminder that the built environment can have a substantial impact on where lava flows advance. The overlap of 40% between the two areal footprints indicated that the digital surface model results more accurately reflect the AVF situation than the digital elevation model results. Then, the thermal vulnerability of buried infrastructure was gathered from conversations with stakeholders at utility companies in Auckland and Hawaii. A deterministic method to assess the lava flow thermal impacts to buried infrastructure utilising simulation modelling, thermo-rheological modelling, and heat transfer modelling is introduced and applied to the Birkenhead scenario. The heat transfer modelling revealed that the electric transmission cable running on the eastern side of Birkenhead will likely be operable for more than a week after the ground above is first inundated. Several mitigation measures are suggested to prolong the operability of the electric cable. One month after the lava has been emplaced, the substrate surrounding the cable is too warm to continue operations. Due to the highly elevated temperatures to which the cable is exposed, the cable will likely need to be replaced after the lava cools. While the damage in this case study is likely complete, it does not follow the assumed binary damage paradigm as the damage is progressive and is not necessarily always complete in every eruption. This suggests that, similar to other volcanic hazards, lava flows can cause a spectrum of damage and that mitigation is possible.