Publications A Building Classification System for Multi-hazard Risk Assessment Share Facebook LinkedIn Download 2022 | Peer-reviewed A uniform and comprehensive classification system, often referred to as taxonomy, is fundamental for the characterization of building portfolios for natural hazard risk assessment. A building taxonomy characterizes assets according to attributes that can influence the likelihood of damage due to the effects of natural hazards. Within the scope of the Global Earthquake Model (GEM) initiative, a building taxonomy (GEM Building Taxonomy V2.0) was developed with the goal of classifying buildings according to their seismic vulnerability. This taxonomy contained 13 building attributes, including the main material of construction, lateral load-resisting system, date of construction and number of stories. Since its release in 2012, the taxonomy has been used by hundreds of experts working on exposure and risk modeling efforts. These applications allowed the identification of several limitations, which led to the improvement and expansion of this taxonomy into a new classification system compatible with multi-hazard risk assessment. This expanded taxonomy (named GED4ALL) includes more attributes and several details relevant for buildings exposed to natural hazards beyond earthquakes. GED4ALL has been applied in several international initiatives, enabling the identification of the most common building classes in the world, and facilitating compatibility between exposure models and databases of vulnerability and damage databases.

News GEM convenes key players in global exposure modelling at a workshop in Pavia By: Apr 24, 2026 Feb 5, 2026 Share Facebook LinkedIn GEM hosted a one-day Global Exposure Modelling Workshop in Pavia in February 2026, bringing together experts from Humanitarian OpenStreetMap Team, Overture Maps Foundation, the Joint Research Centre (JRC), German Aerospace Centre (DLR), Meta, Technical University of Munich, Oak Ridge National Laboratory and the U.S. Geological Survey, alongside the GEM exposure modelling team. The workshop focused on a central challenge in global risk modelling: how to build consistent exposure information that can describe buildings and infrastructure across countries while remaining useful for hazard, vulnerability and loss analysis. In GEM’s broader modelling framework, exposure data are a core component of risk assessment, providing information on building stock, occupancy, replacement cost and related characteristics needed to estimate potential impacts. GEM’s existing Global Exposure Model is a mosaic of national models covering residential, commercial and industrial building stock worldwide. The model includes information at the smallest available administrative level for each country or territory and is designed to support global and regional risk assessment. The workshop in Pavia built on this foundation by exploring how newer sources and methods could help move toward more detailed, individual building-level exposure information at a global scale. Participants contributed perspectives ranging from crowd-sourced mapping and open geospatial datasets to Earth observation, AI-based classification and building-level exposure models. Across the day, discussions moved from global imagery and map sources to the practical requirements of integrating detailed building information into multi-hazard risk models. A final roundtable considered shared priorities and possible frameworks for collaboration toward a building-level global exposure model, including the need for consistency, interoperability and open methodologies. The workshop highlighted how global exposure modelling increasingly depends on collaboration across communities that do not always work together directly, including geospatial data providers, Earth observation specialists and risk scientists. By convening these perspectives in one setting, GEM used the event not only to review current approaches, but also to help define the next steps in strengthening the exposure component of global risk modelling. No images found. GALLERY 1/0 VIDEO RELATED CONTENTS

The first global assessment of the carbon cost of earthquakes Products Global Seismic Embodied Carbon Risk The first global assessment of the carbon cost of earthquakes Share Facebook LinkedIn Description The v2026.1 Global Seismic Embodied-Carbon Risk Map shows the carbon footprint of earthquakes in terms of emissions from building damage due to ground shaking. The model combines seismic hazard, vulnerability, exposure and embodied carbon factors to estimate the average annual embodied carbon (AAEC) for residential, commercial and industrial buildings worldwide, corresponding to approximately 17 MtCO2e per year. AAEC is computed using the event-based calculator of the GEM OpenQuake Engine, which aggregates impacts from multiple earthquake scenarios into a long-term annualised metric. The underlying global exposure model describes building stocks using data on construction material, structural system, occupancy, replacement cost and other relevant vulnerability attributes. A newly developed harmonised embodied-carbon database is used to assign replacement carbon values to exposed buildings worldwide, capturing regional differences in production systems and supply chains. Overall, the current global building stock embodies approximately 102 GtCO2e. How to cite this work Caruso M, Aljawhari K, Nafeh AMB, Calderón A, Rao A, Yepes-Estrada C, Silva V (2026). Global Earthquake Model (GEM) Seismic Embodied-Carbon Risk Map (version 2026.1), https://doi.org/10.5281/zenodo.20208633. Available Versions A PDF poster, as well as a high-resolution PNG of embodied carbon risk calculated globally, is available for direct download under a CC BY-SA license. If your use case does not meet the open license requirements, or if you are interested in obtaining the underlying dataset as a shapefile, please submit a request in our system by clicking on the "License Request", where a specific license will be provided, depending on the use case. License information The downloadable resources are available under a Creative Commons CC BY-SA 4.0 license, which requires: Attribution (you must give appropriate credit, provide a link to the license, and indicate if changes were made) ShareAlike (derivatives created must be made available under the same license as the original) Any deviation from these terms incur in license infringement and requires a license agreement that must be made tailored to your use case, in such instance please click on "License Request". Facebook LinkedIn Share Latest Version v2026.1 License CC BY-SA 4.0 Available resources Geoviewer PNG Poster Download Acknowledgements License Request Contributors text Map View Search Popup title Close Country/Region Available Resources Afghanistan Exposure Africa Exposure Alaska Exposure Country/Region Available Resources Afghanistan Vulnerability Africa Vulnerability Alaska Vulnerability Country/Region Resource Url Afghanistan Risk Profile Africa Risk Profile Alaska Risk Profile Search Found Country/Region Resource Url Afghanistan Exposure Africa Exposure Alaska Exposure Preview Preview is not available. Search Found Country/Region Resource Url Afghanistan Vulnerability Africa Vulnerability Alaska Vulnerability Preview Preview is not available. Search Found Country/Region Resource Url Afghanistan Risk Profile Africa Risk Profile Alaska Risk Profile Preview Preview is not available. Related products Global Liquefaction Hazard Global Exposure Model Global Vulnerability Model Global Seismic Embodied Carbon Risk Global Seismic Risk Map Country-Territory Seismic Risk Profiles OpenQuake Engine Global Seismic Hazard Map Related publications For downloading or accessing detailed product information like PNG/PDF maps, datasets, license request, shapefiles and more, please switch to a desktop or laptop computer. Thank you for your understanding.

Publications Advancing nearshore and onshore tsunami hazard approximation with machine learning surrogates Share Facebook LinkedIn Download 2024 | Peer-reviewed Probabilistic tsunami hazard and risk assessment (PTHA and PTRA) are vital methodologies for computing tsunami risk and prompt measures to mitigate impacts. At large regional scales, their use and scope are currently limited by the computational costs of numerically intensive simulations behind them, which may be feasible only with advanced computational resources like high-performance computing (HPC) and may still require reductions in resolution, number of scenarios modelled, or use of simpler approximation schemes. To conduct PTHA and PTRA for large proportions of the coast, we therefore need to develop concepts and algorithms for reducing the number of events simulated and for more efficiently approximating the needed simulation results. This case study for a coastal region of Tohoku, Japan, utilises a limited number of tsunami simulations from submarine earthquakes along the subduction interface to build a wave propagation and inundation database and fits these simulation results through a machine learning-based variational encoder-decoder model. This is used as a surrogate to predict the tsunami waveform at the coast and the maximum inundation depths onshore at the different test sites. The performance of the surrogate models was assessed using a 5-fold cross validation assessment across the simulation events. Further to understand its real world performance and test the generalisability of the model, we used 5 very different tsunami source models from literature for historic events to further benchmark the model and understand its current deficiencies.

News News Briefs December 2023: Sponsor-supported projects By: Dec 14, 2023 Dec 14, 2023 Share Facebook LinkedIn Sponsor-supported projects Global Earthquake Scenarios (GEeSe) Project In the GEeSe project, a workflow has been developed for generating ground-motion fields for ISC-GEM catalogue events with Mw > 6.5. Models from the Global Hazard Mosaic are selected based on hypocenter proximity, then sources matching the events are filtered out. The rupture matching algorithm identifies the best-matched ruptures. These generate scenario hazard calculations, considering site conditions. The project aims to expand to consider regional conditioning of ground-motion fields and to provide an accessible dashboard output for users. Recent ad-hoc tests showed promising results, such as for the September 2023 Mw 6.8 Morocco event. Retrace3D Project with Italian Civil Protection Retrace3D, in partnership with the Italian Civil Protection, focuses on seismic hazard modelling in Central Italy, the site of the 2016–2017 earthquake sequence. Using RETRACE-3D's high-res fault model and ASPECT's geodynamics approach, the project characterises earthquake activity and computes seismic hazard using the OpenQuake Engine. The finite-element modelling of the fault system was recently completed. A workflow now calculates seismic moment rates and slip rates for each fault, forming source model files for OpenQuake Engine. COMET/UK Project Update The ongoing COMET collaboration has made significant progress. GEM's focus in this phase involves crafting a preliminary time-independent hazard model for Central Asia, leveraging a fault database from the previous collaboration phase. Progress depends on FERMI, GEM's new fault modelling tool. The next phase aims to develop a time-dependent model, a complex task in earthquake science. GEM is exploring innovative methods to modify earthquake timing within established fault slip rates and regional magnitude-frequency distributions. Initial experiments show promise, although scaling to larger fault networks presents challenges. Business Interruption Project Update Following a request for more work in this space from Guy Carpenter, this project led to the creation of new vulnerability functions for business disruption. These functions were based on results derived from simplified multi-degree-of-freedom systems and component-based models. This new modelling framework was developed to enhance the estimation of damage to nonstructural elements and contents. Real claim data from the 2017 Mexico earthquake was used to validate these models, leading to promising results. Liquefaction Module Upgrades Notable enhancements were made to the OpenQuake-engine's liquefaction module. Recent regional liquefaction occurrence models, like those utilised by the US Geological Survey's ground failure (GF) product and Todorovic and Silva's machine learning-based nonparametric model, were integrated. These models aim to predict ground failure using mapped information, offering improved national-scale liquefaction hazard and risk assessments. Additionally, the engine now supports machine learning-based models through the ONNX Runtime library integration. Risk Assessment and Mitigation Study for Earthquakes in the Yilgarn: Infrastructure, Community, Interdependency (RAMSEY) Project The RAMSEY project in Western Australia aims to understand infrastructure system risks in the Yilgarn region. Focusing on critical infrastructure networks, it assesses vulnerabilities, post-hazard dependencies, and recovery strategies. Despite challenges in formalising partnerships, progress continues. Assessments of community buildings' mitigation measures, impacts of scenario events, and substation vulnerabilities have been made. The next key step in the project is the formal request to Federal government by project partners to be able to share infrastructure information. Taiwan Earthquake Model (TEM) In the latter half of 2023, efforts were dedicated to three primary areas. Firstly, the focus was on developing techniques to detect building footprints via satellite imagery, aimed at refining the mapping of exposure data. Secondly, emphasis was on enhancing the understanding of active faults in continental Southeast Asia, specifically addressing the suggested slip rate heterogeneity along the Red River Fault. Lastly, comprehensive seismic models were proposed for the Myanmar region, incorporating area source geometries and fault alignments for Probabilistic Seismic Hazard Assessment (PSHA). Furthermore, the Japan-Taiwan Exchange Association gave the green light to the proposed Japan-Taiwan collaboration concerning Ryukyu PSHA during the Japan-Taiwan-New Zealand Seismic Hazard Assessment meeting held from November 29 to December 3 in Napier, Aotearoa, New Zealand. Collaboration with the Swiss Seismological Service (ETH Zürich) on the EMME24 model This ongoing work is a crucial part of the Middle East Earthquake Model (EMME24). Its objectives include implementing a probabilistic framework, with updated datasets, harmonised cross-border data, revised tectonics and seismogenic source models, as well as ground motion models. The latter involves using multiple GMPEs and backbone models within a logic tree, all implemented in the OpenQuake Engine. In this semester, progress focused mainly on advancing the EMME24 regional hazard model through ongoing efforts that involve 51 Experts from 12 countries. Weekly discussions have been conducted, addressing various components of the model, different toolkits, and advancements in OpenQuake Engine features. No images found. GALLERY 1/0 VIDEO RELATED CONTENTS

News PNG clears the ground for new building codes By: Jul 2, 2018 Share Facebook LinkedIn Papua New Guinea is situated at the edge of what is called the “ring of fire”, within the convergence zone of the Pacific and Australian tectonic plates. Its current seismic zoning map, which was developed in the early 1980s and is included in the national building code, needed a revision. In 2013 the PNG’s Department of Mineral Policy and Geohazards Management began a project in collaboration with Geoscience Australia to re-evaluate the level of seismic hazard in the region and to replace the existing model. Fully funded by the Australian Aid, this project leveraged the OpenQuake- Engine for the hazard assessment, selecting the software among others for its openness, transparency and scalability. Preliminary results of the study are now available (Fig.1); whereas the final results as well as the input models will be released after peer review and publication of the official report. OpenQuake is also used routinely by Port Moresby Geophysical Observatory to generate earthquake ShakeMaps for significant earthquakes in the region. No images found. GALLERY 1/0 VIDEO RELATED CONTENTS

License Request Form You have chosen to get more information about: Global homogenised earthquake catalogue Hazard Please check the link below to see if this product already meets your your requirements before submitting your request for a license. Thank you. DOWNLOAD THE OPEN VERSION Summary of steps to obtain a license for the requested product. Fill in the application form below. Click Submit. Please check your email Inbox or Spam folder for the summary of your request. You will then be contacted by the GEM Product Manager with either a request for more information, or a request to sign the license. If you do not hear from us within 2 weeks, please send an email to product@globalquakemodel.org . REQUEST DETAILS A. Requesting party information First Name Last Name Role/Job Email Business type Business type Other business Sector Sector Other sector B. License agreement signatory information The signatory must be someone who is authorised to sign license agreements on your behalf such as your immediate supervisor, manager or legal officer. If you’re a PhD student, the signatory must be your adviser or a university officer in charge of license agreements or similar legal documents. Full Name of Signatory Position Company Email of Signatory Organisation name Complete Address C. Purpose of request GEM is able to offer products for free because of the support of our project partners, national collaborators and institutional sponsors. All of GEM’s products are freely available for public good, non-commercial use, but with different license restrictions. In most cases we release products under an open license (e.g., CC BY-SA or CC BY-NC-SA), which permits (re)distribution. In this case, we are granting access under a more restricted license that forbids distribution or disclosure and requires signing by GEM and the licensee in order to better assure accountability for the confidentiality of the information. In order for GEM to properly assess your request, please answer the following questions below. 1. Explain briefly how will the GEM product be used e.g. project, research including the expected results and the foreseen public benefit. Minimum 60 words. 2. Will you be able to share the results of your work with GEM? YES NO 3. Will you be able to provide feedback to GEM on the quality and usefulness of this product via a survey? YES NO D. Privacy Policy By submitting this form, you consent to the processing of your personal data in accordance with our Privacy Policy and the EU General Data Protection Regulation (GDPR). We are committed to safeguarding your information and ensuring it is only used for the purpose outlined in this form. You have the right to access, rectify, or delete your data at any time. For more information, please refer to our Privacy Policy. I agree Words: 0 Email us at product@globalquakemodel.org if you're experiencing problems submitting your application. Thank you. Submit Thanks for submitting! You will be contacted as soon as possible Incomplete data. Please fill in all required fields. Thank you.

License Request Form You have chosen to get more information about: Pacific Ocean Hazard Hazard Please check the link below to see if this product already meets your your requirements before submitting your request for a license. Thank you. DOWNLOAD THE OPEN VERSION Summary of steps to obtain a license for the requested product. Fill in the application form below. Click Submit. Please check your email Inbox or Spam folder for the summary of your request. You will then be contacted by the GEM Product Manager with either a request for more information, or a request to sign the license. If you do not hear from us within 2 weeks, please send an email to product@globalquakemodel.org . REQUEST DETAILS A. Requesting party information First Name Last Name Role/Job Email Business type Business type Other business Sector Sector Other sector B. License agreement signatory information The signatory must be someone who is authorised to sign license agreements on your behalf such as your immediate supervisor, manager or legal officer. If you’re a PhD student, the signatory must be your adviser or a university officer in charge of license agreements or similar legal documents. Full Name of Signatory Position Company Email of Signatory Organisation name Complete Address C. Purpose of request GEM is able to offer products for free because of the support of our project partners, national collaborators and institutional sponsors. All of GEM’s products are freely available for public good, non-commercial use, but with different license restrictions. In most cases we release products under an open license (e.g., CC BY-SA or CC BY-NC-SA), which permits (re)distribution. In this case, we are granting access under a more restricted license that forbids distribution or disclosure and requires signing by GEM and the licensee in order to better assure accountability for the confidentiality of the information. In order for GEM to properly assess your request, please answer the following questions below. 1. Explain briefly how will the GEM product be used e.g. project, research including the expected results and the foreseen public benefit. Minimum 60 words. 2. Will you be able to share the results of your work with GEM? YES NO 3. Will you be able to provide feedback to GEM on the quality and usefulness of this product via a survey? YES NO D. Privacy Policy By submitting this form, you consent to the processing of your personal data in accordance with our Privacy Policy and the EU General Data Protection Regulation (GDPR). We are committed to safeguarding your information and ensuring it is only used for the purpose outlined in this form. You have the right to access, rectify, or delete your data at any time. For more information, please refer to our Privacy Policy. I agree Words: 0 Email us at product@globalquakemodel.org if you're experiencing problems submitting your application. Thank you. Submit Thanks for submitting! You will be contacted as soon as possible Incomplete data. Please fill in all required fields. Thank you.

You have chosen to request a demo on: Indonesia Hazard First Name Last Name Email Company or Institution Name Country Sector Provide Details I accept the license terms and the conditions of products use. Email us at product@globalquakemodel.org if you're experiencing problems submitting this form or if you have any inquiries. Thank you. Request Thanks for submitting! You will be contacted as soon as possible Add answer here

News Geospatial is the future By: Jul 2, 2018 Share Facebook LinkedIn The second edition of the European symposium on Free and Open Source Software for Geospatial, which took place on July 15-17 in Como (Italy), was a great occasion for GEM to relate with experts and leading organizations in the field of open data, but also to show progress in the development of its tools. The OpenQuake Platform and the Integrated Risk Toolkit were presented to a diverse and curious audience, impressed by the performance and scalability of the tools. Questions raised were mostly about the value of open data and the importance of communicating open licenses, as often the concept of licensing is associated with commercial activities rather than as a way to release resources to the public. The attendance was excellent, with more than 400 people – the maximum capacity of Aula magna and the presence of distinguished experts in geospatial data (Patrick Hogan from NASA, presenting a project developed in collaboration with the Politecnico di Milano and Georg Gartner, a master in server and operating systems among the others), and for GEM it was a good chance to meet new organizations and potential stakeholders. In the run up to FOSS4G Seoul, which will take place in September 2015, GEM is planning to strengthen the communication of its products and encourage the application of its resources worldwide. No images found. GALLERY 1/0 VIDEO RELATED CONTENTS