Commentary The aviation sagas: geographies of volcanic risk AMY R DONOVAN AND CLIVE OPPENHEIMER Department of Geography, University of Cambridge, Downing Place, Cambridge CB2 3EN E-mail: ard31@cam.ac.uk This paper was accepted for publication in November 2011 This commentary discusses several recent incidents involving intersecting human and physical geographies. It examines several volcanic crises that demonstrate the multiple uncertainties inherent in volcanic risk analysis, and the impact of these events on volcanology as a discipline. Its intertwined human and physical epistemologies are discussed. The commentary argues that the geography of risk is an important organising concept in a more detailed and layered assessment of the risk from geophysical hazards such as earthquakes and volcanic eruptions. KEY WORDS: geography of risk, natural hazards, policy relevance I n May 2011, the eruption of Grímsvötn volcano in South Iceland demonstrated once more the increasing vulnerability of the modern world to relatively small volcanic eruptions. In March, the dev- astating M9 earthquake and subsequent tsunami in Japan showed that even the best prepared nations face massive losses as a result of geophysical hazards, par- ticularly where technological development in one area (such as aviation or nuclear power, in these cases) exceeds scientific ability to model accurately the hazards that may be encountered. At the time when the power stations were built, seismic hazard assessment was less advanced than today and the need for nuclear energy to meet Japan’s increasing consumption outweighed the prevailing view of seismic and tsunami risk. While the death toll was significantly reduced by engineering solutions such as warning systems, sea walls and building codes, models underestimated the tsunami height. This error has effectively led to a public crisis of confidence in nuclear power – notably in Germany, where large earthquakes do not occur (Balser 2011)! While the nuances of public perceptions of nuclear energy are beyond the scope of this article (see Poortinga et al. 2006), this event demonstrates the complex interface between the physical trigger and the human response. Preparedness in no small part depends on the nature of the physical system involved and how knowable it is: are there parameters that can be identified, mea- sured and modelled? Or are there too many uncer- tainties involved? There are also questions about how uncertainty is represented and handled, not only within models but, crucially, in their use by scientists, policymakers and the public. On Montserrat, for example, where a volcanic eruption has occurred epi- sodically since 1995 and has devastated two-thirds of the island (Figure 1), models of pyroclastic density currents are used to aid the identification of safe and unsafe areas for habitation and labour. When shown to the public, however, one interviewee commented, ‘well, you get out of your computer what you put in’: the implication being that model outputs will reflect the risk perception of the operator. Physical risk assessments are nuanced by opinions, networks (and increasingly social media), knowledge economies and beliefs within communities. Earthquakes and volcanoes behave in stochastic ways – uncertainty cannot be completely reduced by knowing more about them. While investment in sci- entific research is important, it has to be accompanied by better popular and political understanding of the uncertainties, and how we learn to accept potential losses of life and property. We inhabit a lively earth that has the capacity to make and unmake our worlds. Risk is a matter of uncertainty management (Morgan et al. 2009; Eiser 2004) – at the societal and govern- mental level, but also at the emotional and attitudinal level, and across diverse and evolving cultures. One of the challenges in managing the ash crises, then, is understanding the social and personal fears that underlie public response – as well as increasing our capacity to monitor and model volcanic eruptions. This commentary uses recent geophysical events to explore this intersection between the human and The Geographical Journal, Vol. 178, No. 2, June 2012, pp. 98–103, doi: 10.1111/j.1475-4959.2011.00458.x The Geographical Journal Vol. 178 No. 2, pp. 98–103, 2012 © 2012 The Authors. The Geographical Journal © 2012 Royal Geographical Society (with the Institute of British Geographers)