Search

This report details the procedures that have been applied, developed or improved in Task 4.5 for the rapid estimation of damages and losses after an earthquake event. The report revolves around several actions and technical results, which are organized as follows: • Section 3 summarizes the state-of-the-art of current rapid response algorithms and systems, based on a review by Guérin-Marthe et al. (2021). The SELENA and Armargedom software, coupled with shake-map estimates from WP3, are put forward as one of the technical solutions to be implemented in the TURNkey platform. • Section 4 conducts a benchmark study of two damage estimation methods, namely SELENA (Molina et al., 2010), using fragility functions, and Armagedom (Sedan et al., 2013), using semi-empirical vulnerability indices. Both methods are applied on two case-study areas (Luchon area in TB-2 and Alicante area) and the damage distributions are compared. • Section 5 investigates the impact of various factors (resolution of soil amplification maps and building exposure datasets, influence of ground-motion variability) on the accuracy of the damage estimates when applying earthquake scenarios over built areas. Two case studies are considered, namely the Luchon area (TB-2) and the town of Hveragerði (TB3). • Section 6 introduces a probabilistic framework for near real-time seismic damage assessment that exploits heterogeneous sources of information about the seismic input and the structural response to the earthquake. The value of information provided by various combinations of observations and measurements is quantified. The approach is applied to an arbitrary roadway bridge near Patras (TB-4). • Section 7 introduces a sampling-based Bayesian updating method in order to refine damage and loss estimates from field observations. The method is applicable to large real-world systems, such as extended built areas or real-world infrastructure systems. It is applied to the Luchon area (TB-2), where the damage distribution of common buildings and the connectivity loss of the road network system are estimated in a rapid response context. The Bayesian approach for the updating of damage and loss estimated has been implemented in a R/Matlab code, which is briefly described in the companion Deliverable D4.9.

International audience; In recent years, the French seismological, geodetic and gravimetric community has been structured within RESIF (French seismological and geodetic network). In addition to instrumental developments, RESIF has structured the work on French seismicity (metropolitan and overseas) within the RESIF transverse seismicity action (ATS). The purpose of this article is to present the ATS and the way it is structured to propose to the community different products: seismicity bulletin and catalogue, historical and instrumental macroseismicity data, shakemaps. The places where these products can be found are indicated, as well as the way they are realized and the improvements in progress for a better realization and availability. The link with EPOS (European plate observing system) is also underlined.; Ces dernières années, la communauté sismologique, géodésique et gravimétrique française s'est structurée au sein de RESIF (Réseau sismologique et géodésique français). Outre les développements instrumentaux, RESIF a structuré les travaux sur la sismicité française (métropole et outre-mer) au sein de l'Action transverse sismicité (ATS) de RESIF. L'objet de cet article est de présenter l'ATS et la manière dont elle est structurée pour proposer à la communauté différents produits : bulletin et catalogue de sismicité, données historiques et instrumentales de macrosismicité, shakemaps. Les endroits où ces produits peuvent être trouvés sont indiqués, ainsi que la façon dont ils sont réalisés et les améliorations en cours pour une meilleure réalisation et disponibilité. Le lien avec EPOS (European plate observing system) est également souligné.

International audience; Rapid location and classification of data posted onsocial networks during time-critical situations such as naturaldisasters, crowd movement and terrorism is very useful wayto gain situational awareness and to plan response efforts.Twitter as successful real time micro-blogging social media, isincreasingly used to improve resilience during extreme weatherevents/emergency management situations, including earthquake.It being used during crises by communicating potential risksand their impacts by informing agencies and officials. Thegeographical location information of such events are vital torescue people in danger, or need assistance. However, only fewmessages contains there native geographical coordinates (GPS).So identifying location is a real challenge with Twitter data duringcritical situations. Identification of Tweets and their preciselocation are still inaccurate. In this work, we propose to usesemi-supervised technique to utilize unlabeled data, which is oftenabundant at the onset of a crisis event, along with fewer labeleddata. Specifically, we adopt an iterative Random Forest fittingpredictionframework to learn the semi-supervised model.

International audience; Based on our experience in the project REAKT, we present a methodologicalframework to evaluate the potential benefits and costs of using earthquake early warning(EEW) and operational earthquake forecasting (OEF) for real-time mitigation of seismicrisk at nuclear facilities. We focus on evaluating the reliability, significance and usefulnessof the aforementioned real-time risk-mitigation tools and on the communication of real-time earthquake information to end-users. We find that EEW and OEF have significantpotential for the reduction of seismic risk at nuclear plants, although much scientificresearch and testing is still necessary to optimise their operation for these sensitive andhighly-regulated facilities. While our test bed was Switzerland, the methodology presentedhere is of general interest to the community of EEW researchers and end-users and itsscope is significantly beyond its specific application within REAKT.

International audience; Earthquake early warning systems (EEWSs) that rapidly trigger risk-reduction actions after a potentially-damaging earthquake is detected are an attractive tool to reduce seismic losses. One brake on their implementation in practice is the difficulty in setting the threshold required to trigger pre-defined actions: set the level too high and the action is not triggered before potentially-damaging shaking occurs and set the level too low and the action is triggered too readily. Balancing these conflicting requirements of an EEWS requires a consideration of the preferences of its potential end users. In this article a framework to define these preferences, as part of a participatory decision making procedure , is presented. An aspect of this framework is illustrated for a hypothetical toll bridge in a seismically-active region, where the bridge owners wish to balance the risk to people crossing the bridge with the loss of toll revenue and additional travel costs in case of bridge closure. Multi-attribute utility theory (MAUT) is used to constrain the trigger threshold for four owners with different preferences. We find that MAUT is an appealing and transparent way of aiding the potentially controversial decision of what level of risk to accept in EEW. Keywords Earthquake early warning (EEW) Á Decision making Á End-user preferences Á Bridges Á Thresholds Á Multi-attribute utility theory (MAUT)

National audience; Experience of past earthquakes as well as "earthquake" civil-protection's exercises underlines the need for crisis managers to have at their disposal rapid-response tools able to assess consequences caused by earthquakes, even for moderate events. This paper presents a transposition of the American simplified approach "PAGER" for France, allowing the rapid assessment of losses resulting from earthquakes by crossing an intensity map with population considered as a spatialized proxy of vulnerable assets. Finally, while existing rapid response tools are based on the diptych instrumental measurements from seismic sensors / macroseismic individual testimonies collected online via dedicated forms, exploratory tracks of "crowd-sourcing" are under consideration in order to feed the models with additional citizen-based data.; L'expérience acquise après séismes ainsi que dans le cadre d'exercices « séisme » de sécurité-civile montre que les gestionnaires de crise ont besoin de disposer d'outils pour identifier rapidement l'ampleur des conséquences occasionnées, y-compris dans le cas d’événements modérés. Aussi, cet article présente une déclinaison française de l'approche américaine simplifiée « PAGER », permettant l'estimation des pertes consécutives à un séisme par le croisement d'une carte d'intensité avec une donnée de population considérée comme un proxy spatialisé des enjeux vulnérables. Enfin, et alors que les outils de réponse rapide existants reposent sur le diptyque mesures instrumentales en provenance des capteurs sismologiques / témoignages macrosismiques individuels recueillis en ligne via des formulaires dédiés, des pistes exploratoires de « crowd-sourcing » sont à l'étude, dans le but d'alimenter les modèles de pertes de données citoyennes complémentaires.

International audience; The Pyrenees are a 400-km-long mountain range located in southwest Europe along the French–Spanish border, and constitute one of the most earthquake-prone regions of mainland France and Spain.At the “observational” level, the Pyrenean region is monitored by several seismological networks on both sides of the French -Spanish border, counting in total around 120 seismic stations (of different types). Thanks to a progressive decrease of constraints associated to real-time seismology (generalization of low-cost robust data-transfer technologies, continuous increasing of data storage capacities, etc.), a growing proportion of these stations are progressively called to evolve toward real-time data transmission. Moreover, a recent project called “SISPyr” (www.sispyr.eu), involving the main owners of Pyrenean seismic stations, has notably allowed the establishment of a real-time pooling process of Pyrenean seismological data resulting in an improvement of the coverage of the massif.At the “operational” level also, each country has its own civil protection organization as well as specific earthquake crisis plans. However, big earthquakes in the Pyrenees can impact the two (or the three) borders. Moreover, systemic cross-border interactions are multiples (transport network, energy lifelines, hospitals access, cross-border populations, etc.).Rapid response overviewExperience of past earthquakes as well as "earthquake" civil-protection’s exercises underlines the need for crisis managers to have at their disposal rapid-response tools able to assess consequences caused by earthquakes, even for moderate events. SisPyr partnership has developed tools to meet these operational needs, in order to automatically and quickly (15 min) produce maps of seismic ground-motions. These “Shakemaps” integrate both seismological real-time data coming from observatories and internet citizen data (web-questionnaires). Exploratory tracks are also being considered in order to enrich the information feedback from the field by using techniques of "crowd-sourcing", thanks to the use of distributed "citizen" sensors or of social-networks.To go further in taking account operational requirements related to the management of seismic crisis, work is being done in order to provide the authorities with a quick assessment of the human tolls (potential victims or damages, needs for shelters) that may control their actions, structured within reports dedicated to civil-protection teams. Need for geospatial ICT supportMulti-actors contextPyrenean region disposes of several seismological networks on both sides of the Franco-Spanish border: CEA-LDG, OMP and BRGM for the French part (with stations belonging to the Seismic Monitoring National Network – RéNaSS, and to the French Permanent Accelerometric network – RAP), and the Spanish and the Catalan seismological surveys (respectively IGN and ICGC) for the Spanish part. It is also to notice the presence of a broad-band station in Andorra managed by Andorran Studies Institute (IEA). In case of earthquake, several of these institutes produce their own assessment of magnitude/location, while alert itself is assigned in France to CEA-LDG, and in Spain to IGN. At the same time, in case of great earthquakes, international organizations like JRC and CSEM (Europe), and even USGS (US), produces information bulletins, which are not really followed by national/regional crisis management community. Moreover in France Internet citizen data (macroseismic intensities) are collected by another institution, BCSF. In Spain these citizen data are collected both by ICGC and IGN. The Sispyr’s Shakemap system is triggered by alerts coming from IGN (disregarding ones coming from CEA-LDG), and uses IGN, ICGC, OMP and BRGM seismic data, as well as IGN, ICGC and BCSF citizen data. Input data interoperability Due to the multiplicity of data-producers, question of interoperability of input data is critical. Regarding real-time collection of instrumental data, different protocols are used (NAQS, Seedlink and Scream!), corresponding to well-known or accepted “standards” in the scientific community. In any case, the goal was to converge to shakemaps needed input files format and standards. Output data Shakemaps produces intensity maps, as image format or kmz files. This restitution format is not really adapted to crisis managers which work with own GIS platforms. Otherwise, intensity maps are still quite difficult to be interpreted in local-regional crisis management centers, for people which are not familiar with seismic risk. Thus, the incoming step is to produce automatically a “human tolls bulletins” estimating the level of the earthquake according to estimated potential victims or no-shelters populations. These bulletins follow a “light color code”. Red color signification for USGS, European Commission, Spanish and French Civil protection would be the same? Which coherence with other natural risks? (cf. discussions about natural risks zones maps in INSPIRE). Limits-Is Web diffusion of bulletins and maps well adapted for crisis management community?-How communicate in understandable and interoperable way information about outputs’ uncertainties? This issue also questions the notion of responsibility of broadcasters of these data with respect to crisis managers?-In case of significant earthquake, international organizations produce information bulletins, which are not really followed by national/regional crisis management community.-Input data interoperability: de facto standards are used, but are they international or own?-Output data: how to allow reuse of outputs?

International audience; Seismic risk in areas of moderate seismicity (AMS) is a reality and destructive earthquakes may occur there. This is why stakeholders and risk managers in these areas show an increasing interest in Earthquake Early Warning Systems (EEWS) and are wondering about the possibility of equipping their territories with such systems. This question of feasibility has to be broached globally: distinguishing technical feasibility from notions of opportunity and utility, an, thus, it has to involve both scientific experts and stakeholders. Moreover, it is necessary to consider each situation specifically without any preconceptions, specifically concerning potential uses that can be made of early warnings, in particular, it is clear that when applied to AMS, the EEWS principles translate into shorter lead times than the ones expected for more seismic areas and would, consequently, require considering new ways of using early warnings. Through the experience conducted for the Pyrenees (France–Spain border) as part of the SISPyr project, this study aims at identifying responses to the potential value of EEWS as real-time seismic risk mitigation tools for AMS as well as to offer a methodological framework to guide stakeholders in their assessment of opportunities to set up EEWS.

Practical implementations of operational earthquake forecasting (OEF) and earthquake early warning (EEW) require a decision (or many decisions) to be made in order to mitigate risk. For example, for OEF a decision may be made to remove hazardous material from an industrial site during a period of heightened seismic hazard to reduce the chance of a chemical spill in case of an earthquake. In the context of EEW, a gas network may be automatically shut down to decrease the risk of fire following an earthquake...

Supraglacial Lac de Rochemelon was formed 50 years ago behind an ice dam and grew steadily until 2004. In October 2004, the volume of the lake reached 650 000 m3, bringing its surface within 0.2 m of the top of the ice dam. To eliminate the threat to towns located below in the event of an overflow, the lake was drained artificially in October 2004 and during the summer of 2005. Once the volume had been reduced to 250 000 m3 by siphoning, a channel was dug with explosives and the remaining water overflowed naturally. This offered a very good opportunity to investigate the breaching of an ice dam accompanied by thermal erosion of the drainage channel. Extensive field measurements were carried out during drainage. Analysis of the energy dissipated in the channel reveals that only half of the available energy was used for breach erosion. A numerical model was used to simulate the evolution of a number of variables during drainage and to study the sensitivity of discharge and ice erosion to different parameters, revealing a high sensitivity to water temperature. Model simulations indicate that natural drainage of this lake at the beginning of October 2004 would have led to a peak discharge of 3 s−1.

Martinique is, with Guadeloupe, classified in seismicity zone III (decree n°2007-1467 of October 12, 2007) which is the highest level of seismic hazard for the French territory. From the point of view of intensities, with a maximum intensity of VI-VII, the earthquake of November 29, 2007 is the most important one felt in Martinique since the earthquake of June 8, 1999 (intensity VII). Macro-seismic intensity reached (in order of distance from the epicenter) VI-VII in the communes of Sainte-Marie, La Trinité, Fort-de-France, Le François, Trois Ilets, Le Marin and Sainte-Anne. The analysis of this earthquake mobilized many people, research laboratories and technical centers. The French Central Seismological Office has prepared a report based on the data processed by the Volcanological and Seismological Observatory of Martinique of the IPGP. The macro-seismic data were collected thanks to the survey forms distributed to the town halls and gendarmerie services by the SIDPC of the prefectures of Martinique and Guadeloupe, thanks to the testimonies filed on the BCSF website and thanks to the information collected during the BCSF field survey that took place from December 5 to 11, 2007. The BCSF's mission is to collect data on the earthquakes felt in France, to gather useful information and to facilitate its dissemination to the actors concerned by the seismic risk or conducting studies or research requiring the use of these observations. It is a member of the Transversal Seismicity Action of the Résif research infrastructure (French Seismological and Geodetic Network).; La Martinique est, avec la Guadeloupe, classée en zone de sismicité III (décret n°2007-1467 du 12 octobre 2007) qui est le niveau d'aléa sismique le plus élevé pour le territoire français. Du point de vue des intensités, avec une intensité maximale de VI-VII, le séisme du 29 novembre 2007 est le plus important ressenti en Martinique depuis le séisme du 8 juin 1999 (intensité VII). L’intensité macrosismique a atteint (par ordre de distance à l'épicentre) VI-VII dans les communes de Sainte-Marie, La Trinité, Fort-de-France, Le François, Trois Ilets, le Marin et Sainte-Anne. L’analyse de ce séisme a mobilisé de nombreuses personnes, laboratoires de recherche et centres techniques. Le Bureau Central Sismologique Français a élaboré un rapport qui s’est appuyé sur les données traitées par l’Observatoire Volcanologique et Sismologique de Martinique de l’IPGP. Les données macrosismiques ont été collectées grâce aux formulaires d’enquête distribués auprès des mairies et des services de gendarmerie par le SIDPC des préfectures de Martinique et de Guadeloupe, grâce aux témoignages déposés sur le site Internet du BCSF et grâce aux informations recueillies lors de l’enquête BCSF sur le terrain qui s'est déroulée du 5 au 11 décembre 2007. Le BCSF a pour mission de collecter les données sur les séismes ressentis en France, de rassembler les informations utiles et de faciliter leur diffusion vers les acteurs concernés par le risque sismique ou menant des études ou recherches nécessitant l’usage de ces observations. Il est membre de l’Action transverse sismicité de l’infrastructure de recherche Résif (Réseau sismologique et géodésique français).

National audience; Experience of past earthquakes as well as "earthquake" civil-protection’s exercises underlines the need for crisis managers to have at their disposal rapid-response tools able to assess consequences caused by earthquakes, even for moderate events. BRGM and its partners have developed tools to meet these operational needs, in order to automatically and quickly produce maps of seismic ground-motions, as well as sectorized assessments of damages. To go further in taking account operational requirements related to the management of seismic crisis, work is being done in order to provide the authorities with a quick assessment of the human tolls that may control their actions, structured within reports dedicated to civil-protection teams. Exploratory tracks are also being considered in order to enrich the information feedback from the field by using techniques of "crowd-sourcing", thanks to the use of distributed "citizen" sensors or of social-networks.; L’expérience acquise après séismes ainsi que dans le cadre d’exercices « séisme » de sécurité-civile montre que les gestionnaires de crise ont besoin de disposer d’outils pour identifier rapidement l’ampleur des conséquences occasionnées, y-compris dans le cas d’évènements modérés. Le BRGM et ses partenaires ont mis en place des outils visant à répondre à ces besoins opérationnels, permettant notamment la cartographie rapide et automatique de l’ampleur des secousses sismiques, ainsi que l’estimation sectorisée des dommages. Afin d’aller encore plus loin dans la prise en compte des besoins opérationnels liés à la gestion de crise sismique, des travaux sont menés dans le but de fournir aux autorités une évaluation rapide de bilans essentiels au dimensionnement de leur réponse opérationnelle, structurée selon des communiqués directement appropriables par les services concernés. Des pistes exploratoires sont par ailleurs à l’étude dans le but d’enrichir la remontée d’information terrain en ayant recours aux techniques de « crowd-sourcing », par la mobilisation de capteurs « citoyens » distribués ou par l’usage des réseaux sociaux.

In November 2021, a large-scale crisis exercise was organized in the heart of the Pyrenees. The main objective of this cross-border exercise between France, Spain and Andorra was to allow practitioners to prepare for a largescale earthquake, by testing the contribution of the tools developed within the framework of the European project POCRISC. Among the different functions activated during the exercise, this article focuses on the particularly critical function of emergency assessment of building damage. It analyses the feedback from the exercise participants asked to evaluate the deployment, coordination and operation of the damage assessment function, including the use of a tool developed specifically for these activities.

National audience; Public information on major risks is a civil right in France. It is, therefore, important to provide data on the exposure of the citizen to seismic risk, notably by providing information on historical seismicity. BRGM is, therefore, developing a database providing the macroseismic intensities that are estimated to have occurred in every French community during historical earthquakes but for which insufficient documentary proof is currently available within SisFrance (www.sisfrance.net, BRGM/IRSN/EDF). A macroseismic intensity is assessed for every community either based on documentary information for that location (SisFrance) or via spatial interpolation. This development firstly consisted of a feasibility study for the creation of such a database by using the statistical technique known as kriging. Following this, methodological improvements were made and the procedure was automated and applied to all of the earthquakes in SisFrance (both for mainland and overseas communities). Maps generated for some events are shown here as a validation of the procedure and the developed tools. Finally, the database of communal intensities has been structured and a tool created to facilitate the publication of these results via a cartographic server on a website.; L'information des citoyens sur les risques majeurs est un droit et il est ainsi important de mettre à disposition des données sur l'exposition d'un territoire vis-à-vis du risque sismique, notamment sur la sismicité historique. Le BRGM a ainsi mis en place une base de données permettant de faire état de niveaux d'intensités présumés qui seraient occasionnés par des séismes historiques, pour les communes qui sont actuellement insuffisamment documentées dans SisFrance (www.sisfrance.net, BRGM/IRSN/EDF). A chaque commune du territoire des évènements sismiques historiques sont associés avec un niveau d'intensité ressenti (SisFrance) ou présumé (interpolation). Cette étude a consisté à étudier la faisabilité d'une telle base de données, à valider l'utilisation d'un code de calcul réalisant des interpolations des intensités observées à l'aide d'une technique de krigeage. Ensuite des développements méthodologiques ont été apportés et la procédure de calcul a été automatisée et appliquée pour tous les séismes de la base de données SisFrance (métropole et DROM-COM). Des cartes ont été automatiquement générées pour quelques événements sélectionnés et permettent de valider la procédure et les outils mis en place. Enfin une base de données communale des intensités extrapolées est structurée et un outil est développé pour faciliter la mise à disposition des résultats obtenus via un serveur cartographique sur un site internet.

Rapid estimation of the intensity of seismic ground motions is crucial for an effective rapid response when an earthquake occurs. To this end, maps of updated grond-motion fields (orshakemaps) are produced by using observations or measurements in near real-time to better constrain initial estimates. In this work, two types of observations are integrated to generate shakemaps right after an earthquake: the common type of data recorded by physicalsensors(seismic stations) and the data extracted from social sensors (Twitter), or the combination of both. We investigate an approach to extract an approximation of the macroseismic intensity from social sensors 10min after the earthquake; the approach relies on Twitter feeds to define the “felt area” where the earthquake was felt by the population, and the “unfelt locations” where the earthquake was not reported. Two recent earthquakes in France of moderate magnitude are studied and the results are compared to the official macroseismic intensity maps for validation. For the two studied cases, we note that Peak Ground Acceleration recordings far from theepicentertend to underestimate the entire macroseismic field, and that the tweets from “felt areas” are complementary for a better estimation of the intensity shakemap. We highlight the importance and the limits of each type of observations when generating the seismic shakemaps.

Systems for Rapid Response to Earthquakes (RRE) aim at providing reliable and near-real time ground-motion and loss estimates following an earthquake, in order to help authorities taking appropriate actions when needed. The purpose of this study is to review the state-of-the-art for RRE systems, focusing first on the tools and methods that have been developed for shake-map computation (i.e., taking into account the seismic recordings as well as macroseismic observations when available), in order to provide a map of the ground shaking intensity rapidly after an earthquake event. The second part focuses on the different approaches taken for rapid loss assessment, the ones based on shake-maps and the others. We give an overview of the current operating systems with emphasis on European ones, and we highlight their differences and identify the current gaps and pending issues. Finally, we call attention for the need to treat carefully uncertainties propagated at each calculation step, which added up are non-negligible, and are an important part of the result itself. Considering the full statistical distribution of loss and damage estimates provides more information than average or median values, and such statistics should therefore be provided by RRE systems and taken into account by decision-makers in order to take informed actions following an earthquake.

During their postseismic interventions, Urban Search and Rescue (USAR) teams are particularly exposed to the risk of collapse of buildings due to aftershocks. Risk reduction requires the capacity building of rescuers to anticipate incoming ground motions and to quickly protect themselves, which makes the concept of earthquake early warning (EEW) particularly interesting. In parallel to the recent scientific advances in EEW, it is crucial to understand what are the real expectations and needs of USAR teams in terms of protection against this risk, and to what extent EEW solutions could meet them. In this study, we conduct a survey to collect insights from 104 USAR rescuers (32 teams represented from 11 countries). Results highlight that aftershocks are a major concern for rescuers, and that the current tools and procedures do not allow an effective protection. In this context, we find that the concept of EEW is very favorably received by the respondents, who consider different types of possible actions upon receipt of an early warning. Depending on the time available, the perceive benefits range from simply reducing the surprise effect to moving teams into safe areas and stopping dangerous operations. Combined with the opportunities offered by the miniaturization of seismic monitoring means and by the high level of interoperability of USAR teams on an international scale, this study also provides a basis for the functional specifications of future solutions of EEW useful to all USAR teams, as well as for the definition of their modalities of engagement on the field.