Simona Colombelli

Early magnitude and potential damage zone estimates for the great Mw 9 Tohoku-Oki earthquake

The Mw 9.0, 2011 Tohoku-Oki earthquake has reopened the discussion among the scientific community about the effectiveness of earthquake early warning for large events. A well-known problem with real-time procedures is the parameter saturation, which may lead to magnitude underestimation for large earthquakes. Here we measure the initial peak ground displacement and the predominant period by progressively expanding the time window and distance range, to provide consistent magnitude estimates (M = 8.4) and a rapid prediction of the potential damage area. This information would have been available 35 s after the first P-wave detection and could have been refined in the successive 20 s using data from more distant stations. We show the suitability of the existing regression relationships between early warning parameters and magnitude, provided that an appropriate P-wave time window is used for parameter estimation. We interpret the magnitude under-estimation as a combined effect of high-pass filtering and frequency dependence of the main radiating source during the rupture process.

Twin ruptures grew to build up the giant 2011 Tohoku, Japan, earthquake.

The 2011 Tohoku megathrust earthquake had an unexpected size for the region. To image the earthquake rupture in detail, we applied a novel backprojection technique to waveforms from local accelerometer networks. The earthquake began as a small-size twin rupture, slowly propagating mainly updip and triggering the break of a larger-size asperity at shallower depths, resulting in up to 50 m slip and causing high-amplitude tsunami waves. For a long time the rupture remained in a 100–150 km wide slab segment delimited by oceanic fractures, before propagating further to the southwest. The occurrence of large slip at shallow depths likely favored the propagation across contiguous slab segments and contributed to build up a giant earthquake. The lateral variations in the slab geometry may act as geometrical or mechanical barriers finally controlling the earthquake rupture nucleation, evolution and arrest.

Evidence for a difference in rupture initiation between small and large earthquakes

The process of earthquake rupture nucleation and propagation has been investigated through laboratory experiments and theoretical modelling, but a limited number of observations exist at the scale of earthquake fault zones. Distinct models have been proposed, and whether the magnitude can be predicted while the rupture is ongoing represents an unsolved question. Here we show that the evolution of P-wave peak displacement with time is informative regarding the early stage of the rupture process and can be used as a proxy for the final size of the rupture. For the analysed earthquake set, we found a rapid initial increase of the peak displacement for small events and a slower growth for large earthquakes. Our results indicate that earthquakes occurring in a region with a large critical slip distance have a greater likelihood of growing into a large rupture than those originating in a region with a smaller slip-weakening distance.

Test of a Threshold-Based Earthquake Early Warning Method Using Japanese Data

Most of existing earthquake early‐warning systems are regional or on‐site systems. A new concept is the integration of these approaches for the definition of alert levels and the estimation of the earthquake potential damage zone (PDZ). The key element of the method is the real‐time, simultaneous measurement of initial peak displacement (P_d) and period parameter (τ_c) in a 3‐s window after the first P‐wave arrival time at accelerometer stations located at increasing distances from the epicenter. As for the on‐site approach, the recorded values of P_d and τ_c are compared to threshold values, which are set for a minimum magnitude M 6 and instrumental intensity I_MM VII, according to empirical regression analysis of strong‐motion data from different seismic regions. At each recording site the alert level is assigned based on a decisional table with four entries defined by threshold values of the parameters P_d and τ_c. A regional network of stations provides the event location and transmits the information about the alert levels recorded at near‐source stations to more distant sites, before the arrival of the most destructive phase. We present the results of performance tests of this method using ten M>6 Japanese earthquakes that occurred in the period 2000–2009 and propose a very robust methodology for mapping the PDZ in the first seconds after a moderate‐to‐large earthquake. The studied cases displayed a very good matching between the rapidly predicted earthquake PDZ inferred from initial P‐peak displacement amplitudes and the instrumental intensity map, the latter being mapped after the event, using peak ground velocity and/or acceleration, or from field macroseismic surveys.

Exploring the feasibility of a nationwide earthquake early warning system in Italy

When accompanied by appropriate training and preparedness of a population, Earthquake Early Warning Systems (EEWS) are effective and viable tools for the real-time reduction of societal exposure to seismic events in metropolitan areas. The Italian Accelerometric Network, RAN, which consists of about 500 stations installed over all the active seismic zones, as well as many cities and strategic infrastructures in Italy, has the potential to serve as a nationwide early warning system. In this work, we present a feasibility study for a nationwide EEWS in Italy obtained by the integration of the RAN and the software platform PRobabilistic and Evolutionary early warning SysTem (PRESTo). The performance of the RAN-PRESTo EEWS is first assessed by testing it on real strong motion recordings of 40 of the largest earthquakes that have occurred during the last 10 years in Italy. Furthermore, we extend the analysis to regions that did not experience earthquakes by considering a nationwide grid of synthetic sources capable of generating Gutenberg-Richter sequences corresponding to the one adopted by the seismic hazard map of the Italian territory. Our results indicate that the RAN-PRESTo EEWS could theoretically provide for higher seismic hazard areas reliable alert messages within about 5 to 10 s and maximum lead times of about 25 s. In case of large events (M > 6.5), this amount of lead time would be sufficient for taking basic protective measures (e.g., duck and cover, move away from windows or equipment) in tens to hundreds of municipalities affected by large ground shaking.

An Integrated Regional and On-Site Earthquake Early Warning System for Southern Italy: Concepts, Methodologies and Performances

We present an approach to Earthquake Early Warning for Southern Italy that integrates regional and on-site systems. The regional approach is based on the PRobabilistic and Evolutionary early warning SysTem (PRESTo) software platform. PRESTo processes 3-components acceleration data streams and provides a peak ground-motion prediction at target sites based on earthquake location and magni- tude computed from P-wave analysis at few stations in the source vicinity. On the other hand, the on-site system is based on the real-time measurement of peak dis- placement and dominant period, on a 3s P-wave time-window. These values are compared to thresholds, set for a minimum magnitude 6 and instrumental intensity VII, derived from empirical regression analyses on strong-motion data. Here we present an overview of the system and describe the algorithms implemented in the PRESTo platform. We also show some case-studies and propose a robust methodol- ogy to evaluate the performance of this Early Warning System.

A P wave‐based, on‐site method for earthquake early warning

A new strategy for a P wave-based, on-site earthquake early warning system has been developed and tested on Japanese strong motion data. The key elements are the real-time, continuous measurement of three peak amplitude parameters and their empirical combination to predict the ensuing peak ground velocity. The observed parameters are compared to threshold values and converted into a single, dimensionless variable. A local alert level is issued as soon as the empirical combination exceeds a given threshold. The performance of the method has been evaluated by applying the approach to the catalog of Japanese earthquake records and counting the relative percentage of successful, missed, and false alarms. We show that the joint use of three peak amplitude parameters improves the performance of the system as compared to the use of a single parameter, with a relative increase of successful alarms of about 35%. The proposed methodology provides a more reliable prediction of the expected ground shaking and improves the robustness of a single-station, threshold-based earthquake early warning system.

Rapid and reliable seismic source characterization in earthquake early warning systems: current methodologies, results, and new perspectives

In the present paper, we provided a review of the main principles and methodologies on which the current earthquake early warning systems are grounded and will also provide a perspective view for next future developments and improvements. First, we introduce the standard methodologies for the source characterization in earthquake early warning, with a special focus on the real-time earthquake magnitude determination. We discuss the suitability of existent methodologies and empirical regression laws for very large events. We then present the different approaches for the rapid prediction of the ground shaking and of the potential damaged zone, both based on traditional seismic data and on the use of continuous GPS data. Finally, the last part of the paper provides the perspective view toward a next generation of early warning systems, linking new research achievements about the earthquake rupture nucleation and the development of new methods/technologies aimed at a fast and high-resolution, real-time modeling of the ongoing source process and accurate prediction of the quake shaking at the regional and local scale.

A Threshold-Based Earthquake Early-Warning System for Offshore Events in Southern Iberia

The south of the Iberian Peninsula is situated at the convergence of the Eurasian and African plates. This region experiences large earthquakes with long separation in time, the best known of which was the great 1755 Lisbon Earthquake, which occurred SW of San Vicente Cape (SW Iberian Peninsula). The high risk of damaging earthquakes has recently led Carranzaet al. (Geophys. Res. Lett. 40, 2013) to investigate the feasibility of an EEWS in this region. Analysis of the geometry for the Iberian seismic networks and the San Vicente Cape area led the authors to conclude that a threshold-based approach, which would not require real-time location of the earthquake, might be the best option for an EEWS in SW Iberia. In this work we investigate this hypothesis and propose a new EEW approach that extends standard P-wave threshold-based single-station analysis to the whole network. The proposed method enables real-time estimation of the potential damage at stations that are triggered by P-waves and those which are not triggered, with the advantage of greater lead-times for release of alerts. Results of tests made with synthetic data mimicking the scenario of the great 1755 Lisbon Earthquake, and those conducted by applying the new approach to available recordings, indicate that an EEW estimation of the potential damage associated with an event in the San Vicente Cape area can be obtained for a very large part of the Iberian Peninsula.

Source and dynamics of a volcanic caldera unrest: Campi Flegrei, 1983–84

Despite their importance for eruption forecasting the causes of seismic rupture processes during caldera unrest are still poorly reconstructed from seismic images. Seismic source locations and waveform attenuation analyses of earthquakes in the Campi Flegrei area (Southern Italy) during the 1983–1984 unrest have revealed a 4–4.5 km deep NW-SE striking aseismic zone of high attenuation offshore Pozzuoli. The lateral features and the principal axis of the attenuation anomaly correspond to the main source of ground uplift during the unrest. Seismic swarms correlate in space and time with fluid injections from a deep hot source, inferred to represent geochemical and temperature variations at Solfatara. These swarms struck a high-attenuation 3–4 km deep reservoir of supercritical fluids under Pozzuoli and migrated towards a shallower aseismic deformation source under Solfatara. The reservoir became aseismic for two months just after the main seismic swarm (April 1, 1984) due to a SE-to-NW directed input from the high-attenuation domain, possibly a dyke emplacement. The unrest ended after fluids migrated from Pozzuoli to the location of the last caldera eruption (Mt. Nuovo, 1538 AD). The results show that the high attenuation domain controls the largest monitored seismic, deformation, and geochemical unrest at the caldera.