Reliable and rapid impact assessment for large earthquakes is a challenge because it is difficult to rapidly determine the fault geometry and thus the spatial distribution of shaking intensities. In this retrospective study of the M 7.8 Kahramanmara & scedil;- Pazarc & imath; k, T & uuml;rkiye, earthquake, we evaluate how eyewitness observations crowdsourced through the LastQuake system can improve such assessments. These data consist of felt reports describing the local level of shaking or damage and manually validated geolocated imagery. In the first part of this study, the methods used to derive macroseismic intensity values from felt reports, particularly for high values, are validated by comparison with independently determined intensities. This comparison confirms that the maximum intensity that can be derived from felt reports does not generally exceed VIII. A fatality estimate of 3000 could be made within a few hours by evaluating the number of people exposed to high intensities using the felt reports and assuming a point source. However, this estimate was known to be an underestimate because of the point-source approximation; this underestimate was also confirmed by the geolocated imagery showing high levels of damage at epicentral distances well beyond those predicted by circular isoseismals. However, improved estimates could have been derived from the event 's ShakeMaps using the U.S. Geological Survey Prompt Assessment of Global Earthquakes for Response (PAGER) fatality loss-modeling system, either by incorporating the felt reports into the ShakeMaps computation or using, in addition, a finite-source (here line-source) model derived from the felt reports using the Finite-fault rupture Detector software. The inclusion of fault geometry would have resulted in a fatality estimate with data collected within 10 min of the origin determination, which was consistent with the final PAGER alert level and the reported death toll that were both only known days later. Although more work would be helpful to assess the reliability of the derived fault geometry, in regions where they are collected in large numbers, felt reports collected within 10 min of the earthquake can be used to substantially improve current fatality estimates.

Public earthquake early warning systems have the potential to reduce individual risk by warning people of approaching tremors, but their development has been hampered by costly infrastructure. Furthermore, both users’ understanding of such a service and their reactions to actual warnings have been the topic of only a few surveys. The smartphone app of the Earthquake Network initiative utilizes users’ smartphones as motion detectors and provides the first example of a purely smartphone-based earthquake early warning system, without the need for dedicated seismic station infrastructure and operating in multiple countries. We demonstrate that this system has issued early warnings in multiple countries, including for damaging shaking levels, and hence that this offers an alternative to conventional early warning systems in the foreseeable future. We also show that although warnings are understood and appreciated by users, notably to get psychologically prepared, only a fraction take protective actions such as “drop, cover, and hold.”

@LastQuake is the official Twitter channel (200k followers) of the Euro-Med Seismological Centre. When an earthquake strikes, real-time information on the seismic event is automatically published via a Twitter-robot. This robot was developed in 2012 and its automatic tweets range from scientific information about earthquake location and magnitude, to accounts of shaking felt by earthquake eyewitnesses and safety guidelines and tsunami warnings. Although efficient and reliable, over the years the robot has shown margins for improvement: * after a large magnitude event, the tweets related to the aftershocks overshadow the information about the mainshock – how should the robot cluster and prioritize earthquake information? * a non-destructive earthquake currently generates interactions on Twitter for only twenty minutes, while a destructive event attracts the interest of various audiences (i.e., the affected population, the seismologists, the media) for a much longer period of time – how should the robot regulate the duration of information depending upon the earthquake? * although used world-wide, @LastQuake is still not well-known in certain countries – how could Twitter be used to reach out to a much greater number of earthquake eyewitnesses and better assess the earthquake effects? We renewed @LastQuake to better tailor the information to our different audiences and to make the most of the EMSC’s most recently-developed services. In order to cluster information on the same event, our new robot uses the Twitter-thread functionality, where information about the same event is gathered in a series of connected tweets. To regulate the duration of information, we classify earthquakes into five categories depending upon their magnitude, the interest they generate among the public, and their destructiveness potential: * small magnitude earthquakes with and without public interest; * larger magnitude earthquakes with and without public interest; * destructive earthquakes. Each class has different information displayed, and hence a different length of the thread. To increase awareness of the LastQuake service in a region where we are not well-known, and potentially gather a much greater number of felt reports after an earthquake, the robot is now equipped with an ‘invitation tweet’, a feature that allows, in compliance with the General Data Protection Regulation, to automatically respond to tweets from potential earthquake eyewitnesses in a specified region and invite them to report their experience in the case of an earthquake. During an earthquake, the affected population process information and act differently than they would do in times of non-crisis: our wording and tone are carefully chosen to provide reliable and empathetic communication, and we improved our illustrations to be accessible to users suffering from color-vision deficiency. To debunk misinformation and fake news, we prepared a series of educational tweets in collaboration with IRIS. The new robot is versatile, targeting not only the affected population, with urgent information, but also the seismologists, with technical information, and the general public and the media, with wrap-up information on what has just happened. We will present to you the renewed @LastQuake Twitter-bot environment and discuss our strategy for tailoring earthquake crisis communications via Twitter.

Earthquake early warning systems based on smartphone networks are emerging as complementary systems to the more expensive systems based on scientific-grade instruments. Hence, there is a need to better understand their detection capabilities. This article introduces a probabilistic framework for modeling the interaction between a smartphone network and seismic events to provide estimates of the detection probability for a given earthquake and to assess how the network geometry affects the detection delay. The framework was used to study the detection capability of the first operational smartphone-based earthquake early warning system implemented by the Earthquake Network (EQN) initiative, which started in 2013 and has issued more than 5500 warnings in 25 countries. The analysis showed that the probability of detection of an earthquake depends on the interaction between the network geometry and the earthquake parameters and that the detection probability is greatly affected by the population spatial distribution. Countries that benefit most from the EQN initiative are those without large gaps in the geographic distribution of their population.

Public earthquake early warning systems have the potential to reduce individual risk by warning people of an incoming tremor, but their development is hampered by costly infrastructure. Furthermore, users’ understanding of such a service and their reactions to warnings remains poorly studied. The Earthquake Network app turns users’ smartphones into motion detectors and provides rapid information about felt earthquakes in multiple countries. It offers an alternative without the need of dedicated infrastructure in the many regions unlikely to be covered by conventional early warning systems in the foreseeable future. We show here that it already provides an early warning service, including for damaging shaking levels and although warnings are appreciated and understood by users, only a fraction follow the “drop, cover and hold” advice.

In recent years, earthquake felt reports contributed via online systems have provided increasingly valuable sources of data to characterize earthquakes and their effects. Contributed felt reports are accompanied by increases in website traffic, which are themselves potentially useful for the early detection of seismic events. In February 2017 the European-Mediterranean Seismic Centre detected an unusual surge in traffic from the Punjab region in northwestern India, although no nearby seismic event was detected instrumentally. Had crowdsourcing detected a felt earthquake that instruments had missed? Or did Punjab cry wolf? In this Earthquake Lites report, we describe the sleuthing endeavor undertaken to find an answer.

We present a general-purpose numerical quantum mechanical solver using Schrödinger-Poisson equations called Aestimo 1D. The solver provides self-consistent solutions to the Schrödinger and Poisson equations for a given semiconductor heterostructure built with materials including elementary, binary, ternary, and quaternary semiconductors and their doped structures. The software can be used to calculate electronic band structures of heterostructures either using a single-band or multi-band k.p envelope function approximation. The software is fully open-source and it is released under the GNU general public license version 3 for full freedom of usage for applications in the fields of nano-electronics, optoelectronics, and solid-state device simulations.

The 'Earthquake Network’ (EQN) is an app which detects earthquakes by creating an ad-hoc network of smartphones' accelerometer sensors and provides early warnings for earthquakes via the same smartphone app. Detections are not due to individual smartphone measurements but due to near-simultaneous trigger signals from clusters of smartphones running the app. Therefore detections are normally located in the closest populated regions to an earthquake's epicentre. These datasets compare sets of detections with the earthquake parameters published by seismic institutes in order to analyse the performance of the EQN network. One dataset contains 550 detections made by EQN between 2017-12-15 and 2020-01-31 in Chile, USA and Italy. Wherever possible, each detection was associated with an earthquake from the parameter catalogue of each country's seismic institute (CSN for Chile, USGS for USA and INGV for Italy). Associations were carried out automatically but also checked manually. The other dataset contains 134 detections from around the world that could be associated to earthquakes with magnitude ≥ M5 or magnitude ≥ M4.5 in Italy and the USA. There are 68 detections that are common to the first dataset. All detections were associated to parameters from the the USGS earthquake parameter catalogue for consistency.

The 'Earthquake Network’ (EQN) is an app which detects earthquakes by creating an ad-hoc network of smartphone's accelerometer sensors and provides early warnings of earthquakes via the same smartphone app. The EMSC (Euro-Mediterranean Seismological Centre) and the University of Bergamo conducted an online survey, following an earthquake of magnitude M8 on 2019-05-26 07:41:13.6 UTC in Northern Peru with epicentre (5.81S, 75.27W). This survey targeted EQN users in the felt area of the earthquakes and was conducted from 2019-07-23 to 2019-08-18. It aimed at assessing users’ understanding and reaction to the EQN early warning for this specific earthquake. The questionnaire was written in Spanish since it is the most commonly spoken language in the studied area. Individuals who use the app in Spanish were invited to complete the survey via an advertisement on the Earthquake Network app. A PDF containing the questionnaire and the relationship between the questions is included in this archive. 3805 respondents took the survey, including 2 719 that were actually in the area at the time. The analysis: Results derived from this dataset will be included as part of a submitted Science article (Bossu et al. '“Shaking in 5 seconds!” A Voluntary Smartphone-based Earthquake Early Warning System', 2021) to show that respondents received notifications from the Earthquake Network App before feeling the shaking but also that many did not immediately “drop, hold and cover' and were too intent on warning those close to them of the impending danger. All respondents consented that their data could be used for research purposes. The EMSC and University of Bergamo made sure not to collect or diffuse personal data. The dataset is a zip-file that contains the questionnaire responses as a comma-separated text file (csv) and a pdf containing a representation of the questionnaire that was presented to respondents.

Over the past decade technologies and social media have been praised a revolution in the way seismic risk and information was communicated to the public. For instance, LastQuake system is crowdsourcing earthquake detections and providing timely information and safety tips to citizens through social media and a free app now used by over 900K users over the world. Through a series of other practical examples and case studies, among which Earthquake Network system and Raspberry Shakes use, we show that smartphones and social media along with other technologies have indeed shaped new ways to detect earthquakes. They also enable to collect key information in order to raise situational awareness and in the end, inform the public in a timely and geotargeted manner, passing from a top-down approach to a two-way communication. Technology use also led to a significant increase of citizens’ role and implication in seismology, not only raising their interest for the risk but also for the science that relies beneath it. Our analysis demonstrates that in order to be successful, and thus to reduce anxiety and create a trust relationship between scientists and citizens, efficient communication strategies must be based on a thorough knowledge of both risk culture and technological culture. Indeed, such assessment of the audience enables to better meet public’s various information needs in terms of content, format and time frame. However, technologies as well as their uses have kept evolving and seismologists are now facing new challenges to communicate key information. For instance, in an increasing number of cases, misinformation and rumours about earthquake predictions become viral on social media. This is a critical issue to be addressed, especially in a context of development of both earthquake early warning systems and earthquake operational forecast. Additionally, citizen communication and information routines are evolving, especially with the rise of messaging apps or the development of new social media. However, to date, messaging apps are not designed in a way that facilitates communication of critical seismic information by seismic institutions. We conclude that lessons learned from previous challenges, especially through a constant return of experience process, will be useful to address contemporary ones. Communicating towards the public is a necessity during all phases of the crisis cycle, from a scientific and risk perspective. Getting to know the audience’s needs, habits, language, emotions or cultural background and show them empathy is a critical part of efficient communication strategies. Finally, technologies should be seized as an opportunity to engage with citizens and build a lasting trust relationship.

The use of the LastQuake information system, its app, the associated Twitter account, and, to a lesser extent, the EMSC's websites have been analyzed for the 7 days following the November 26, 2019, M6.4 Albania destructive earthquake to evaluate what can be improved and how crowdsourcing of information and monitoring of both use and absence of use of the app can contribute to rapid situational awareness. The mainshock and its numerous felt aftershocks triggered a strong public desire for information, which in turn led to rapid and massive adoption of the LastQuake app by up to 5% of the country's population. The constant flow of new app users created a stress test of the app's crowdsourcing features and led to errors in the association of felt reports with their appropriate earthquake. However, these errors had no identifiable impact, supporting the conclusion that the curation mechanisms currently in place are efficient. The rapid succession of felt aftershocks contributed to these errors by making information related to the mainshock difficult to access within hours of its occurrence, especially for new users who were not attuned to the app, since more recent events pushed older ones down the timeline of presented information. This revealed that prioritization of information within the app layout was lacking and must be an important design objective, especially during aftershock sequences. LastQuake has been shown to be a powerful tool for rapid situational awareness. The possibility of damage was detected within 8 min of the mainshock earthquake by a lack of LastQuake app activity close to the epicenter. This possibility was then gradually strengthened as new data became available and was finally confirmed by the reception of the first geo-located pictures of structural damage and building collapse within 60-70 min. Direct exchanges on Twitter were appreciated by eyewitnesses and seemed to help to reduce their anxiety in some cases (based on the personal reports). Questions mainly focused on the possible evolution of the seismicity. Attempts to debunk prediction claims were difficult. We report on how this could be eased and possibly made more efficient by sharing among the different actors a clear, concise, pre-prepared statement in the local language, that explains the state of scientific knowledge and the difference between prediction, early warning, or forecasts.

We present a methodology that uses crowdsourced detections as an initial location to obtain fast and reliable hypocenter parameters for felt earthquakes using arrival-time data from the GEOFON Program. We derive selection criteria for issuing an alert message using a 3-year-long training set from the trial runs at the European-Mediterranean Seismological Centre (EMSC) to identify accurate event locations at a high confidence level. Since an event may have several crowdsourced detections, we also develop a methodology dealing with multiple triggers. We validate the selection criteria using real-time processing of recent data and demonstrate that 95% of the selected events are within 50 km distance from the traditional seismic location published by the EMSC. SinceCsLocremains essentially a seismic location algorithm, the selection criteria measure the quality of the seismological network coverage used in the location, not the method itself. We show that our methodology provides accurate locations much faster than those published by conventional seismic methods. On average, the EMSCCsLocservice can provide rapid and accurate locations within a minute after the occurrence of a felt earthquake, thus it can provide timely and accurate information on a felt earthquake to the civil protection services and the general public.

Starting the 10th May 2018, a series of earthquakes has hit Mayotte, a French island in the Indian Ocean. Facing a lack of seismic data, scientific information and communication from the authorities, the inhabitants took advantage of social media to develop, on their own, a citizen seismology group, composed of more than 10,000 people. Due to a particular cultural context, this was carried out mainly without the seismologist community. While some citizens did share seismological information (and eventually volcanology information when it was discovered that the earthquakes were caused by a new-born, undersea volcano), the lack of seismologists in the group also lead to the emergence of misinformation and even conspiracy theories. This mistrusting atmosphere had negative consequences for the way various seismological organizations were perceived, including LastQuake, a crowdsource-based earthquake information app which allows eyewitnesses to share information about earthquakes they felt, combined with seismic data. However, due to the lack of seismic data for these earthquakes, some were not displayed in the app. This lack of information and understanding of how the system functioned led to additional mistrust toward this citizen seismology tool. This paper combines sociological observations with an empirical approach. First, a sociological analysis of this independent citizen science network enables an identification of the reasons for its creation and the pitfalls caused by the absence of collaboration with the scientific community. Then, an empirical case study of the LastQuake system exposes how it has been improved to offer information, while admittedly more incomplete, is nevertheless closer to citizens' needs. It concludes that citizen seismology requires a stronger collaboration between citizens' and scientists' communities in order to be more efficient. It also advocates for scientific communication that takes into account cultural context from the beginning.

In many cases, it takes several minutes after an earthquake to publish online a seismic location with confidence. Via monitoring for specific types of increased website, app, or Twitter usage, crowdsourced detection of seismic activity can be used to “seed” the search in the seismic data for an earthquake and reduce the risk of false detections, thereby accelerating the publication of locations for felt earthquakes. We demonstrate that this low-cost approach can work at the global scale to produce reliable and rapid results. The system was retroactively tested on a set of real crowdsourced detections of earthquakes made during 2016 and 2017, with 50% of successful locations found within 103 s, 76 s faster than GEOFON and 271 s faster than the European-Mediterranean Seismological Centre’s publication times, and 90% of successful locations found within 54 km of the final accepted epicenter.