The original model covering the Central America and the Caribbean region (v2018.0.0) was initially developed in the framework of the CCARA project, a GEM collaboration project funded by USAID. Cuba and Puerto Rico were included a posteriori by the GEM hazard team. During the CCARA project and after it, some local organizations and experts were involved and this models benefits of it: University of Costa Rica, Costa Rica - Costa Rican Institute of Electricity, Costa Rica - Nicaraguan Institute of Territorial Studies, Nicaragua - Catholic University of El Salvador, El Salvador - Ministry of Environment and Natural Resources, El Salvador - Panama University, Panama, Puerto Rico Seismic Network, Puerto Rico, National Center for Seismological Research, Cuba. The model was built as a combination of a shallow model, where active faults and distributed seismicity sources were integrated, and a subduction model, divided into its main components (i.e. interface and in-slab). The interface seismicity was modelled using complex faults, while for the in-slab region, the typology of source preferred was the non-parametric source. Since the original version, several significant updates have been made to the model by the GEM hazard team, leading to the latest release (v2019.1.0), which was used in the computation of the 2023 Global Earthquake Hazard and Risk Maps. Differences in the versions can be consulted in the model documentation.

The Global Exposure Model is a mosaic of local and regional models with information regarding the residential, commercial, and industrial building stock at the smallest available administrative division of each country and includes details about the number of buildings, number of occupants, vulnerability characteristics, average built-up area, and average replacement cost. The dataset is developed and maintained by the GEM Foundation, using a bottom-up approach at the global scale, using national statistics, socio-economic data, and local datasets. This model allows the identification of the most common types of construction worldwide, regions with large fractions of informal construction, and areas prone to natural disasters with a high concentration of population and building stock. The Central Asia region of the model includes the information pertaining the following countries/territories: Kazakhstan, Kyrgyzstan, Tajikistan, Turkmenistan, Uzbekistan

The Global Hazard Mosaic coverage of Central Asia (Kazakhstan, Kyrgyzstan, Tajikistan, Uzbekistan, and Turkmenistan) comes from the Earthquake Model Central Asia (EMCA; Ullah et al., 2013 and www.emca-gem.org), a Global Earthquake Model regional program coordinated by the German Research Centre for Geosciences (GFZ). The model was originally developed for the OpenQuake (OQ) engine.

The China financial loss model has been developed by GEM using public sources of information, such as past seismicity, and geodetic and geologic data for the hazard component, combined with exposure and vulnerability data. The hazard component incorporates both faults and area sources. Earthquake occurrence rates on active faults are based on a new tectonic block model derived from the joint inversion of geodetic and geologic data. The model provides estimates of financial risk to residential, commercial and industrial buildings using GEM’s vulnerability models appropriate to Chinese construction practice. GEM has also developed an exposure model that can be used to estimate total losses to the building stock in addition to portfolio losses. Further technical information can be found in the documentation. GEM extensively collaborated with its private and public partners to test and validate the model against industry standards in order to produce a new model that represents GEM’s view of risk. The model is available in Oasis and Touchstone formats, as well as through the NASDAQ platform.

The Global Hazard Mosaic coverage of China was developed by the GEM Secretariat. The model covers mainland China and is composed of smooth seismicity sources and fault sources, the latter of which are mostly located in the western half of the territory. Further information on the model, as well as different versions available, can be consulted in the model documentation.

The GEM Colombia model is based upon GEM’s regional model for South America first developed from 2013 to 2015 within the South America Risk Assessment (SARA) project funded by the Swiss Re Foundation and updated in 2018. The development process involved dozens of local scientists and engineers from earthquake-prone countries in South America, resulting in an open product for analyzing ground-up loss from ground shaking for residential, commercial, and industrial buildings. This model has been further improved with local information, such as local construction practices, replacement costs and level of enforcement of the seismic regulations. Further technical information can be found in the documentation. GEM extensively collaborated with its private and public partners to test and validate the model against industry standards in order to produce a new model that represents GEM’s view of risk. The model is available in Oasis and Touchstone formats, as well as through the NASDAQ platform.

The 2018 United States National Seismic Hazard Model (Petersen et al., 2020), covers the conterminous United States and was developed by the United States Geological Survey (USGS) within the National Seismic Hazard Model Project (NSHMP). The California portion of the the model is covered by the time-independent version of the Uniform California Earthquake Rupture Forecast version 3 (UCERF3; Field et al., 2015). The model has been translated from its original format into the OpenQuake engine by GEM.

The GEM Foundation has produced a collection of Country/Territory Seismic Risk Profiles that summarize key metrics of seismic risk, to provide stakeholders in risk management an overview of the risk in a region at-a-glance. Each profile presents the following relavant information: *Social indicators, which provide context to the region in question *Risk indicators, detailing an occupancy breakdown of exposed value and losses *A list of the major earthquakes that have impacted the region *Loss per region, providing a breakdown of average annual losses per Administrative level 1 *Building classes, depicting the major construction materials used in the region *Loss curves, which provide expected losses per different return periods *Maps depicting the geographical distribution of hazard, exposure and losses The risk results are the results of an event-based risk analysis, where 100,000 years of earthquakes are simulated. Three lines of business are considered: residential, commercial, and industrial. Therefore, value or earthquake losses to other building occupancies (e.g., schools, healthcare) and infrastructure are not included.

The Dominican Republic probabilistic seismic hazard model (v2021.2.0) was developed as part of the USAID-funded TREQ project.

The Global Exposure Model is a mosaic of local and regional models with information regarding the residential, commercial, and industrial building stock at the smallest available administrative division of each country and includes details about the number of buildings, number of occupants, vulnerability characteristics, average built-up area, and average replacement cost. The dataset is developed and maintained by the GEM Foundation, using a bottom-up approach at the global scale, using national statistics, socio-economic data, and local datasets. This model allows the identification of the most common types of construction worldwide, regions with large fractions of informal construction, and areas prone to natural disasters with a high concentration of population and building stock. The East Asia region of the model includes the information pertaining the following countries/territories: China, Hong_Kong, Japan, Macao, North_Korea, South_Korea, Taiwan

The Global Exposure Model is a mosaic of local and regional models with information regarding the residential, commercial, and industrial building stock at the smallest available administrative division of each country and includes details about the number of buildings, number of occupants, vulnerability characteristics, average built-up area, and average replacement cost. The dataset is developed and maintained by the GEM Foundation, using a bottom-up approach at the global scale, using national statistics, socio-economic data, and local datasets. This model allows the identification of the most common types of construction worldwide, regions with large fractions of informal construction, and areas prone to natural disasters with a high concentration of population and building stock. For the case of Europe, the exposure model is built upon the dataset developed by the European Facilities for Earthquake Hazard and Risk (EFEHR). The original information can be consulted in the repository for the European Seismic Risk Model (ESRM20) in [this link](https://gitlab.seismo.ethz.ch/efehr/esrm20_exposure) The Europe region of the model includes the information pertaining the following countries/territories: Albania, Andorra, Austria, Belarus, Belgium, Bosnia_and_Herzegovina, Bulgaria, Croatia, Cyprus, Czechia, Denmark, Estonia, Finland, France, Germany, Gibraltar, Greece, Hungary, Iceland, Ireland, Isle_of_Man, Italy, Kosovo, Latvia, Liechtenstein, Lithuania, Luxembourg, Malta, Moldova, Monaco, Montenegro, Netherlands, North_Macedonia, Norway, Poland, Portugal, Romania, Serbia, Slovakia, Slovenia, Spain, Sweden, Switzerland, Turkey, Ukraine, United_Kingdom

The Global Hazard Mosaic uses the 2020 European Seismic Hazard Model (ESHM20) to cover the Euro-Mediterranean region. The ESHM20 model uses the same principles as the Seismic Hazard Harmonization in Europe (SHARE) Project model (Woessner et al., 2015), with state-of-the art procedures homogeneously applied for the pan-European region. A fully probabilistic framework was adopted in the hazard model implementation, using input datasets that are harmonized across national borders. ESHM20 was developed by the Seismology and Earthquake Engineering Research Infrastructure Alliance for Europe (SERA) and funded by the European Union’s Horizon 2020 research and innovation programme. For more details, refer to http://www.sera eu.org/en/home/ and http://www.hazard.efehr.org/en/home/ The model was originally implemented in the OpenQuake engine. In addition to the conterminous pan-European region, this model was used to compute hazard for the Kaliningrad Region of Russia, Iceland, and the Canary Islands. Athough ESHM20 covers Turkey, the Global Hazard Mosaic currently uses the EMME model.