Operational earthquake forecasting in Europe: progress, despite challenges

摘要

Is Europe-wide operational earthquake forecasting (OEF) possible? We discuss the myriad problems that prevent it today, many of which relate to heterogeneities in earthquake recording, processing, and reporting. We contemplate the difficulty of building models that cross political boundaries, and we consider the prospect of European OEF in light of recent efforts to harmonize long-term seismic hazard assessment among several nations. Emphasizing the Strategies and Tools for Real-time Earthquake Risk Reduction (REAKT) project, we report achievements related to short-term seismicity forecasting in Iceland and Italy that could apply elsewhere in Europe. In Iceland, collaboration fostered by REAKT resulted in a revised earthquake catalog and a prototype OEF system. We report results from an experiment conducted with this prototype; these results suggest ensemble models provide an information gain, updating models more frequently improves their forecast skill, and that OEF is computationally feasible. In Italy, REAKT supported the creation of an ensemble model that now issues weekly hazard forecasts. We present examples of these forecasts, highlighting the problem that OEF often yields low probabilities, which are difficult to interpret and convert into actionable decisions. Motivated by such low hazard probabilities, we highlight Europe's pioneering efforts in operational earthquake loss forecasting and mention solutions to problems that currently prevent OEF at the European scale.