Item

Abstract

Strong earthquakes are followed by numerous aftershocks, which themselves can be strong and devastating, sometimes even stronger than the initial main event. Although earthquakes cannot be predicted, the occurrence of aftershocks in space, time, magnitude and numerosity follows several universally observed empirical laws. Operational earthquake forecasting (OEF) systems aim to provide a probabilistic assessment of expected earthquake occurrence in near-real time, enabling informed decision-making for civil protection, authorities, the public, or other user groups. While few agencies worldwide have OEF systems in place, most countries, even those in regions of high seismic hazard, only estimate the time-independent, long-term earthquake probability as part of their seismic hazard and risk assessment, which serves a different user group involved in long-term decision making. This lack of OEF systems may be due to a lack of data, knowledge, resources, or guidance required to build them. With a plethora of forecasting models and testing schemes published every year, it is difficult to choose the most appropriate ones. On top of technical choices involved in the development and testing of forecasting models comes the additional challenge of communicating forecasts to users. In our study, we conducted an elicitation of expert views on these topics, aiming to provide good practice recommendations for the development, testing and communication of earthquake forecasts. We applied the Delphi method – with two surveys and one workshop – to identify consensus and dissent among a group of 20 international earthquake forecasting experts on these issues, and here present the results of this elicitation.