Scientific uncertainties: A perspective from probabilistic seismic hazard assessments for low-seismicity areas

Introduction: Earthquakes have claimed several million human lives over the last two millennia (Dunbar et al., 1992), as well as unquantifiable numbers of casualties and limitless grief. A large, and increasing, proportion of the world’s population nowadays resides in poorly constructed dwellings that are high susceptibility to seismic damage; building collapses account for more than 75% of human casualties in earthquakes (Coburn and Spence, 2002). On top of this, increasing urbanisation means that half of the world’s population now lives in cities (United Nations, 2002), and a significant number of these cities are located in earthquake-prone regions (Bilham, 2004). In recent times, the average global fatality count from earthquakes has been about 100000 per year, a rate that is determined mainly by the larger but less frequent disasters. Given the inexorable growth in the global population, an increasing worldwide earthquake fatality rate might be expected, but recent urban expansion has seen a decline in the fatality rate when expressed as a percentage of the instantaneous population (Bilham, 2004). There are two possible explanations for this. One is that the application of earthquake codes and better construction techniques – coupled to improved hazard assessments – are engendering real beneficial outcomes; while this may be true in some countries, it is more questionable for many cities in the developing world (Ambraseys and Bilham, 2011), especially mega-cities. An alternative reading is that the apparent decline in risk has arisen simply from a short-term fluctuation in large-magnitude earthquake activity. Major earthquakes recur, on average, at intervals of the order of decades to centuries, whereas the world population has dramatically increased only in the last few decades. Just one or two extreme earthquake disasters in the near future – affecting some of the world’s mega-cities – could easily reverse the current downturn in apparent risk.