Optimizing barrier placement for lava flow hazard and risk mitigation

Mitigating hazards when lava flows threaten infrastructure is one of the most challenging fields of volcanology, and has an immediate and practical impact on society. Lava flow hazard is determined by the probability of inundation, and essentially controlled by the topography of the area of interest. The most common actions of intervention for lava flow hazard mitigation are therefore the construction of artificial barriers and ditches that can control the flow direction and advancement speed. Estimating the effect a barrier or ditch can have on lava flow paths is non-trivial, but numerical modelling can provide a powerful tool by simulating the eruptive scenario and thus assess the effectiveness of the mitigation action. We present a numerical method for the design of optimal artificial barriers, in terms of location and geometric features, aimed at minimizing the impact of lava flows based on the spatial distribution of exposed elements. First, an exposure analysis collects information about elements at risk from different datasets: population per municipality, distribution of buildings, infrastructure, routes, gas and electricity networks, and land use; numerical simulations are used to compute the probability for these elements to be inundated by lava flows from a number of possible eruptive scenarios  (hazard assessment) and computing the associated economic loss and potential destruction of key facilities (risk assessment). We then generate several intervention scenarios, defined by the location, orientation and geometry (width, length, thickness and even shape) of multiple barriers, and compute the corresponding variation in economic loss. Optimality of the barrier placement is thus considered as a minimization problem for the economic loss, controlled by the barrier placement and constrained by the associated costs. We demonstrate the operation of this system by using a retrospective analysis of some recent effusive eruptions at Mount Etna, Sicily.