Seismic hazard estimates rely on interevent time distributions between earthquakes of a given magnitude. In the Himalaya, recurrence intervals are usually modeled as cyclic or quasiperiodic, whereas globally, they range from periodic and clustered to random. Statistical analyses of a 6000-year lake-sediment seismic record, calibrated against regional instrumental data, worldwide paleoseismic records, and synthetic seismic catalogs, demonstrate that time intervals between large earthquakes (M ≥ 6.5, based on shaking intensity thresholds calibrated locally) robustly follow a Poisson distribution. Second-order fluctuations indicate event clustering. These observations contradict periodic or quasiperiodic recurrence models. Comparisons with paleoseismic data from other tectonic settings and realistic synthetic catalogs confirm the robustness and broad applicability of these findings. Thus, major earthquakes appear as stochastic as smaller ones, challenging recurrence models derived from limited datasets and substantially increasing seismic hazard estimates.

SUMMARY The Lacq area in southwest France has been associated with continuous moderate induced seismic activity since 1969. However, the mechanisms driving this induced seismicity are not fully understood: reservoir depletion has been proposed as the main factor, and more recently wastewater injection has been suggested to play a more important role. The interpretation of these mechanisms relies heavily on the quality of earthquake locations, which we prove to be weak due to a lack of local instrumentation for several years. In order to provide the most complete and reliable induced event catalogue for the studies of the Lacq induced seismicity mechanisms and seismic hazard, we made an exhaustive compilation, analysis and improvement of all available catalogues. We also provided new earthquake detections and relocations in a 3-D velocity model from past and present temporary deployments never used for studying the Lacq area. Important remaining location uncertainties lead us to also carefully sort the events according to their location confidence, defining 3 classes of events (unconstrained location, location constrained within 2–3 km and 1–2 km, respectively). This new harmonized catalogue and the identification of well-constrained events, covering 50 yr of induced seismicity, allow us to propose that wastewater injection is almost certainly the main mechanism driving the seismicity, with (i) most of the constrained events located within the reservoir boundaries and (ii) the released seismic energy variations following variations in injection operations at different scales. In particular, we have also highlighted a change in the injection–seismicity relationship around 2010–2013. From 2013, despite lower injection volumes, seismicity remained persistent and some clusters of earthquakes were detected predominantly in spring, summer and early autumn, except in winter periods. From 2016, we observed a strong temporal relationship between days with higher rate/volume injections (approximately above 400 m3 d−1) and both clustered events and higher magnitude earthquakes (greater than 2.4).

On November 11, 2019, an unusually damaging Mw4.9 earthquake occurred in the south east of France within the lower Rhône river valley, an industrial region hosting several operating nuclear power plants. This event is exceptional considering its very shallow depth (<1 km). Based on farfield seismological observations, we demonstrate that the rupture properties are consistent with the ones commonly observed for large deeper earthquakes, implying that the near-surface faulting generated strong high-frequency seismic waves. In the absence of strong motion sensors in the fault vicinity, we perform numerical predictions of the ground acceleration on a virtual array of near-fault stations, that matches with the locations of independent quantitative estimations from in-situ observations of displaced objects (natural and anthropic). Both numerical and in-situ analyses converge toward an exceptional level of ground acceleration in the fault vicinity, exceeding gravity, and at the origin of the damage. This dramatically changes the perception of the impacts of superficial moderate earthquakes on seismic hazard assessment.

ABSTRACTThe activation of tectonics and anthropogenic swarms in time and space and size remains challenging for seismologists. One remarkably long swarm is the Lacq swarm. It has been ongoing since 1969 and is located in a compound oil–gas field with a complex fluid manipulation history. Based on the overlap between the volumes where poroelastic model predicts stresses buildup and those where earthquakes occur, gas reservoir depletion was proposed to control the Lacq seismic swarm. The 2016 Mw 3.9, the largest event on the site, is located within a few kilometers downward the deep injection well. It questions the possible interactions between the 1955–2016 wastewater injections and the Lacq seismicity. Revisiting 60 yr of fluid manipulation history and seismicity indicates that the impacts of the wastewater injections on the Lacq seismicity were previously underevaluated. The main lines of evidence toward a wastewater injection cause are (1) cumulative injected volume enough in 1969 to trigger Mw 3 events, onset of Lacq seismicity; (2) 1976 injection below the gas reservoir occurs only a few years before the sharp increase in seismicity. It matches the onset of deep seismicity (below the gas reservoir, at the injection depth); (3) the (2007–2010) 2–3 folds increase in injection rate precedes 2013, 2016 top largest events; and (4) 75% of the 2013–2016 events cluster within 4–8 km depths, that is, close to and downward the 4.5 km deep injection well. As quantified by changepoint analysis, our results suggest that timely overlaps between injection operations and seismicity patterns are as decisive as extraction operations to control the Lacq seismicity. The seismicity onset is contemporary to cumulative stress changes (induced by depletion and injection operations) in the 0.1–1 MPa range. The interrelation between injection and extraction is the most probable cause of the Lacq seismicity onset and is sustenance over time. The injected volume–largest magnitude pair for Lacq field is in the same range (90% confidence level) than wastewater volume–magnitude pairs reported worldwide, in a wide variety of tectonic settings.

On November 11, 2019, a M$_{\mathrm{w}}$ 4.9 earthquake hit the region close to Montelimar (lower Rhône Valley, France), on the eastern margin of the Massif Central close to the external part of the Alps. Occuring in a moderate seismicity area, this earthquake is remarkable for its very shallow focal depth (between 1 and 3 km), its magnitude, and the moderate to large damages it produced in several villages. InSAR interferograms indicated a shallow rupture about 4 km long reaching the surface and the reactivation of the ancient NE–SW La Rouvière normal fault in reverse faulting in agreement with the present-day E–W compressional tectonics. The peculiarity of this earthquake together with a poor coverage of the epicentral region by permanent seismological and geodetic stations triggered the mobilisation of the French post-seismic unit and the broad French scientific community from various institutions, with the deployment of geophysical instruments (seismological and geodesic stations), geological field surveys, and field evaluation of the intensity of the earthquake. Within 7 days after the mainshock, 47 seismological stations were deployed in the epicentral area to improve the Le Teil aftershocks locations relative to the French permanent seismological network (RESIF), monitor the temporal and spatial evolution of microearthquakes close to the fault plane and temporal evolution of the seismic response of 3 damaged historical buildings, and to study suspected site effects and their influence in the distribution of seismic damage. This seismological dataset, completed by data owned by different institutions, was integrated in a homogeneous archive and distributed through FDSN web services by the RESIF data center. This dataset, together with observations of surface rupture evidences, geologic, geodetic and satellite data, will help to unravel the causes and rupture mechanism of this earthquake, and contribute to account in seismic hazard assessment for earthquakes along the major regional Cévenne fault system in a context of present-day compressional tectonics.

Mining, water-reservoir impoundment, underground gas storage, geothermal energy exploitation and hydrocarbon extraction have the potential to cause rock deformation and earthquakes, which may be hazardous for people, infrastructure and the environment. Restricted access to data constitutes a barrier to assessing and mitigating the associated hazards. Thematic Core Service Anthropogenic Hazards (TCS AH) of the European Plate Observing System (EPOS) provides a novel e-research infrastructure. The core of this infrastructure, the IS-EPOS Platform (tcs.ah-epos.eu) connected to international data storage nodes offers open access to large grouped datasets (here termed episodes), comprising geoscientific and associated data from industrial activity along with a large set of embedded applications for their efficient data processing, analysis and visualization. The novel team-working features of the IS-EPOS Platform facilitate collaborative and interdisciplinary scientific research, public understanding of science, citizen science applications, knowledge dissemination, data-informed policy-making and the teaching of anthropogenic hazards related to georesource exploitation. TCS AH is one of 10 thematic core services forming EPOS, a solid earth science European Research Infrastructure Consortium (ERIC) (www.epos-ip.org.).

Seismology and paleoseismology seem to be two distant sisters when we address earthquake time-interval distributions. One observation stands out; an apparent discrepancy in time-interval models, i.e. periodic to cluster, within similar tectonic context. As a departure point, we will use the Himalayan context where according to instrumental or paleoseismic catalogues, time-interval distributions are presented as Poisson to periodic. We report on a new 6000-year lake-sediment seismic record and perform statistical analyses to show that time intervals between large (M≥6.5) earthquakes are robustly described by a Poisson distribution, while second-order fluctuations imply event clustering. These patterns are calibrated against an instrumental catalogue for the entire Himalaya; we show that both catalogues are inconsistent with periodic models. Throughout this presentation, we will compare the Himalayan results with paleoseismic catalogues from three distinct tectonic settings (Indonesia, New-Zealand and Jordan). Each of them displays a close to Poisson distribution, in consonance with instrumental catalogues results. Our results imply that the occurrence of major seismic events is as uncertain as smaller events on any time scale, increasing drastically previous estimate of the seismic hazard.

The EPOS TCS-AH brings together a broad community interested in Anthropogenic Hazards (AH) related to induced seismicity. It is designed as a functional e-research infrastructure that provides access to a large set of relevant data and allows free experimentations in a virtual laboratory, promoting interdisciplinary collaborations between stakeholders (the scientific community, industrial partners and society). The platform provides datasets as Episodes, which comprehensively describe AH cases for infrastructures, people and/or environment. They are grouped in several categories of subsurface exploitations: CO2 sequestration, conventional hydrocarbon extraction, geothermal energy production, reservoir impoundment, unconventional hydrocarbon extraction, underground gas storage, underground mining, and wastewater injection. They gather datasets relevant for the considered hazards (e.g. seismic, air/water quality), industrial data (e.g. well path, injection rates, mining front advance, gas production, water level), and other geodata (e.g. geological section, velocity model, faults, shear wave velocity, bathymetric map). Two local data centers (eNodes: IG-PAS/Poland and CDGP-EOST/France) provide metadata and data to the TCS-AH platform in commonly used standards and formats (e.g. miniSEED, GeoTIFF, and .mat). A registration/authorization is mandatory to access some data covered by restriction imposed by data industry providers or shared data embargoed by running projects. The platform grants access to an application portfolio, designed for the AH area, and addressing: (1) basic services for data integration and handling; (2) services for physical models of stress/strain changes over time and space as driven by geo-resource production; (3) services for analyses of geophysical signals; (4) services to extract the relation between technological operations and observed induced seismic/deformation; (5) services to quantitative probabilistic assessments of anthropogenic seismic hazard - statistical properties of anthropogenic seismic series and their dependence on time-varying anthropogenesis; ground motion prediction equations; stationary and time-dependent probabilistic seismic hazard estimates, related to time-changeable technological factors inducing the seismic process; (6) simulator for multi-hazard/multi-risk assessment in exploration/exploitation of georesources (MERGER) - numerical estimate of the occurrence probability of chains of events or processes impacting the environment.

SUMMARYDeveloping a model for anthropogenic seismic hazard remains an open challenge whatever the geo-resource production. We analyse the (Mmax) largest reported magnitude on each site where (RTS) Reservoir Triggered Seismicity in documented (37 events, 1933–2008), for aftershocks of reservoir impoundment loading. We relate each reservoir impoundment to its magnitude-equivalent M*reservoir = M*(Lr). We use (Lr) the reservoir length as a proxy for a rupture length of the reservoir main shock-equivallent. This latter is derived from the empirical relationship that exists for tectonic earthquake among magnitude and rupture length. We resolve (i) Mmax for RTS are bounded by M*reservoir at a 95 per cent confidence level; (ii) in average Mmax are smaller than M*reservoir by 2.2 units (iii) 50 per cent of the Mmax occurrence is within 2 ± 1 yr from the reservoir impoundment. These triggering patterns support the signature of fluid driven seismicity during the slow reservoir impoundment emerges as a weaker efficiency (larger ΔM = M*reservoir – Mmax) to trigger Mmax events than from earthquake interactions.

The ability to predict the magnitude of an earthquake caused by deep fluid injections is an important factor for assessing the safety of the reservoir storage and the seismic hazard. Here, we propose a new approach to evaluate the seismic energy released during fluid injection by integrating injection parameters, induced aseismic deformation, and the distance of earthquake sources from injection. We use data from ten injection experiments performed at a decameter scale into fault zones in limestone and shale formations. We observe that the seismic energy and the hydraulic energy similarly depend on the injected fluid volume ( V ), as they both scale as V 3/2 . They show, however, a large discrepancy, partly related to a large aseismic deformation. Therefore, to accurately predict the released seismic energy, aseismic deformation should be considered in the budget through the residual deformation measured at the injection. Alternatively, the minimal hypocentral distance from injection points and the critical fluid pressure for fault reactivation can be used for a better prediction of the seismic moment in the total compilation of earthquakes observed during these experiments. Complementary to the prediction based only on the injected fluid volume, our approach opens the possibility of using alternative monitoring parameters to improve traffic-light protocols for induced earthquakes and the regulation of operational injection activities.

We analyzed the impact of the 26 largest impounded reservoirs on reservoir-triggered seismicity (RTS) patterns in the low-seismicity region of continental France. We treat reservoir-triggered earthquakes as tectonic earthquakes and apply similar concepts in our analysis. Generally, the spatial extent of an aftershock zone is controlled by the mainshock rupture length. In a similar manner, we use reservoir length as an equivalent length to the rupture length to assess the spatial extent of reservoir-triggering earthquakes and one to three reservoir lengths as a proxy for the nearfield distance where the stress change induced by reservoir impoundment may trigger seismicity. Accordingly, we define the 1L(r) distance as the near-field reservoir effect on seismicity and the 10L(r) distance as the far field, null effect of reservoir stress change on background seismicity. We find that (1) about a quarter of the reservoirs trigger M-max = 2.5-4.7 within the 1L(r) distance in a 15 yr space time window, and (2) as tested against a randomized series, superposed epoch analysis demonstrates a robust increase in the average seismicity rate within 2 yrs for the 1-3L(r) distance from reservoirs. The reservoirs that trigger in the (1L(r)) near-field distance are significantly larger than the nontriggering ones. While considering the distance of triggering of earthquakes from the reservoir, it is more appropriate to consider the normalized distance (the distance normalized by the reservoir length) to identify earthquake triggering reservoirs at a 1L(r) distance. While considering reservoir dimensions, the reservoir length appears to be a more important parameter than the reservoir depth, as the length is proportional to the area of significant stress change. Our results suggest that the RTS mimics the aftershock sequence of a slow reservoir-impoundment loading, with a corresponding M*reservoir = M(L-r) mainshock magnitude. Further, when considering mainshock-aftershock interactions, our analysis and observations support that the M-max for RTS for a given reservoir remains, on average, smaller than the reservoir magnitude equivalent.

The rank-sum multiple change-point method is a robust statistical procedure designed to search for the optimal number and the location of change points in an arbitrary continue or discrete sequence of values. As such, this procedure can be used to analyse time-series data. Twelve years of robust data sets for the Sechilienne (French Alps) rockslide show a continuous increase in average displacement rate from 50 to 280 mm per month, in the 2004-2014 period, followed by a strong decrease back to 50 mm per month in the 2014-2015 period. When possible kinematic phases are tentatively suggested in previous studies, its solely rely on the basis of empirical threshold values. In this paper, we analyse how the use of a statistical algorithm for change-point detection helps to better understand time phases in landslide kinematics. First, we test the efficiency of the statistical algorithm on geophysical benchmark data, these data sets (stream flows and Northern Hemisphere temperatures) being already analysed by independent statistical tools. Second, we apply the method to 12-yr daily time-series of the Sechilienne landslide, for rainfall and displacement data, from 2003 December to 2015 December, in order to quantitatively extract changes in landslide kinematics. We find two strong significant discontinuities in the weekly cumulated rainfall values: an average rainfall rate increase is resolved in 2012 April and a decrease in 2014 August. Four robust changes are highlighted in the displacement time-series (2008 May, 2009 November-December-2010 January, 2012 September and 2014 March), the 2010 one being preceded by a significant but weak rainfall rate increase (in 2009 November). Accordingly, we are able to quantitatively define five kinematic stages for the Sechilienne rock avalanche during this period. The synchronization between the rainfall and displacement rate, only resolved at the end of 2009 and beginning of 2010, corresponds to a remarkable change (fourfold increase in mean displacement rate) in the landslide kinematic. This suggests that an increase of the rainfall is able to drive an increase of the landslide displacement rate, but that most of the kinematics of the landslide is not directly attributable to rainfall amount. The detailed exploration of the characteristics of the five kinematic stages suggests that the weekly averaged displacement rates are more tied to the frequency or rainy days than to the rainfall rate values. These results suggest the pattern of Sechilienne rock avalanche is consistent with the previous findings that landslide kinematics is dependent upon not only rainfall but also soil moisture conditions (as known as being more strongly related to precipitation frequency than to precipitation amount). Finally, our analysis of the displacement rate time-series pinpoints a susceptibility change of slope response to rainfall, as being slower before the end of 2009 than after, respectively. The kinematic history as depicted by statistical tools opens new routes to understand the apparent complexity of Sechilienne landslide kinematic.

(1) Institute of Geophysics Polish Academy of Sciences, Warsaw, Poland (orlecka@igf.edu.pl), (2) Helmholtz-Zentrum Potsdam Deutsches GeoForschungsZentrum GFZ, Potsdam, Germany, (3) Center for the Analysis and Monitoring of Environmental Risk (AMRA) Università di Napoli "Federico II", Napoli, Italy, (4) Sodankyla Geophysical Observatory/Oulu Unit, University of Oulu, Finland, (5) Institut de Physique du Globe de Strasbourg, Ecole et Observatoire des Sciences de la Terre, Strasbourg, France, (6) Physicien du Globe, Isterre, Grenoble Observatory, Grenoble, France, (7) INERIS Ecole des Mines de Nancy Campus ARTEM, France, (8) Istituto Nazionale di Geofisica e Vulcanologia Direttore Sezione di Pisa, Italy, (9) School of Physical and Geographical Sciences, Keele University, UK, (10) Central Mining Institute, Katowice, Poland, (11) ACK Cyfronet, AGH, Poland

There is striking evidence that the dynamics of the Earth crust is controlled by a wide variety of mutually dependent mechanisms acting at different spatial and temporal scales. The interplay of these mechanisms produces instabilities in the stress field, leading to abrupt energy releases, i.e., earthquakes. As a consequence, the evolution towards instability before a single event is very difficult to monitor. On the other hand, collective behavior in stress transfer and relaxation within the Earth crust leads to emergent properties described by stable phenomenological laws for a population of many earthquakes in size, time and space domains. This observation has stimulated a statistical mechanics approach to earthquake occurrence, applying ideas and methods as scaling laws, universality, fractal dimension, renormalization group, to characterize the physics of earthquakes. In this review we first present a description of the phenomenological laws of earthquake occurrence which represent the frame of reference for a variety of statistical mechanical models, ranging from the spring-block to more complex fault models. Next, we discuss the problem of seismic forecasting in the general framework of stochastic processes, where seismic occurrence can be described as a branching process implementing space time-energy correlations between earthquakes. In this context we show how correlations originate from dynamical scaling relations between time and energy, able to account for universality and provide a unifying description for the phenomenological power laws. Then we discuss how branching models can be implemented to forecast the temporal evolution of the earthquake occurrence probability and allow to discriminate among different physical mechanisms responsible for earthquake triggering. In particular, the forecasting problem will be presented in a rigorous mathematical framework, discussing the relevance of the processes acting at different temporal scales for different levels of prediction. In this review we also briefly discuss how the statistical mechanics approach can be applied to non-tectonic earthquakes and to other natural stochastic processes, such as volcanic eruptions and solar flares. (C) 2016 Elsevier B.V. All rights reserved.