Forecasting volcanic eruptions
A central goal for the current generation of volcano researchers is to provide tools for predicting when volcanoes will erupt. These tools are of key importance to those making decisions about what action to take when, (for example, when volcanic seismicity increases). However, tools for predicting eruptions are in their infancy. We can’t always successfully predict an eruption as we lack the understanding of how these signals leading to catastrophe are generated.
Jeremie Vasseur and colleagues have taken the simple rules for forecasting eruptions and applied them to mini-catastrophes in the laboratory. These volcanologists carefully track the failure of samples of synthetic magmas under pressure and record the micro-signals that herald the final event. Analyzing these precursory signals and how they evolve is akin to analyzing the seismic signals prior to eruption. So, what can lab-scale experiments tell us about why we can’t always forecast eruptions?
Vasseur’s team, under the supervision of Prof Dingwell, find that the more heterogeneous the material is, the better we can predict its failure. On the scale of a volcano this is vital as it implies that the key to forecasting eruptions is a good knowledge of how heterogeneous the magma is. In magmas, heterogeneity is known to arise from gas-bubbles and crystals in an otherwise homogeneous liquid.
This study is one of the first of the next generation of forecasting tools that will accommodate the complex materials involved in eruptions that we call magmas. Jeremie thinks, “this approach unites material science with the volcano-forecasting effort,” adding that “it is surely multidisciplinary approaches like this where the most significant advances are to be made.” Dingwell’s LMU group will continue the effort in the hope that a comprehensive forecasting tool will some day be within our grasp
Vasseur, J., Wadsworth, F.B., Lavallée, Y., Bell, A.F., Main, I.G., Dingwell, D.B. Heterogeneity: The key to failure forecasting. Scientific Reports 5, 13259; doi:10.1038/srep13259 (2015).