Real‐time forecasting following a damaging earthquake

Abstract

[1] We describe the results of a prospective, real-time earthquake forecast experiment made during a seismic emergency. A M w 6.3 earthquake struck the city of L'Aquila, Italy on April 6, 2009, causing hundreds of deaths and vast damage. Immediately following this event, we began producing daily earthquake forecasts for the region, and we provided these forecasts to Civil Protection -the agency responsible for managing the emergency. The forecasts are based on a stochastic model that combines the Gutenberg-Richter distribution of earthquake magnitudes and power-law decay in space and time of triggered earthquakes. The results from the first month following the L'Aquila earthquake exhibit a good fit between forecasts and observations, indicating that accurate earthquake forecasting is now a realistic goal. Our experience with this experiment demonstrates an urgent need for a connection between probabilistic forecasts and decisionmaking in order to establish -before crises -quantitative and transparent protocols for decision support.

Control of pore fluid pressure diffusion on fault failure mode: Insights from the 2009 L'Aquila seismic sequence

“…[63] The L'Aquila seismic sequence was studied by means of other approaches, and apparent discrepancies are evident between our results of Figure 5, and the pure Omori-like behavior described by the ETAS model [Kagan and Knopoff, 1981;Ogata, 1988;Kagan, 1991;Ogata, 1998] shown in the study by Marzocchi and Lombardi [2009].…”

“…When we compare the absolute number of aftershocks/day that were recorded between 18 and 75 h after the main quake (i.e., between the third and the fourth data point in the diagrams of Figure 5), we see that the SE data set contains roughly 2.5 times the aftershocks/day that are contained in the NW data set. Due to such difference in population, when the aftershocks decay rate is computed over the entire set of earthquakes (as done by Marzocchi and Lombardi [2009]), the diffusion signal seen to the NW gets wiped out by the regular Omori-like behavior that dominates the SE-marked frame of Figure 5. Due to the strong diffusion phenomenon that dominated the seismicity on the Campotosto fault for the first six-eight days, if we were to run the ETAS model over the NW part of the data set (ETAS does not incorporate diffusion terms), we expect the results to be problematic in the first day of the sequence (W. Marzocchi, personal communication, 2011).…”

Strong foreshock signal preceding the L'Aquila (Italy) earthquake (<i>M</i><sub>w</sub> 6.3) of 6 April 2009

“…1c). Marzocchi and Lombardi (2009) reported on the near realtime aftershock forecasting based on the ETAS model.…”

Significant improvements of the space-time ETAS model for forecasting of accurate baseline seismicity

“…Namely, μ(x, y) of the background seismicity is useful for long-term prediction of large earthquakes (Ogata, 2008). Also, the model with normalized aftershock productivity K (x, y) could possibly be more useful for immediate aftershock probability forecast than the one implemented in Marzocchi and Lombardi (2009), especially in the case where the anisotropic features are not neglected. The reasons and their utility of the basic structure of the model in (2) are demonstrated in Ogata (1998).…”