In 2006 the OECD Global Science Forum established an International Steering Committee to develop through a public private partnership a not-for profit facility through which information on earthquake risk could be shared globally through the development of databases, educational programmes and models that could be used for estimating earthquake hazard risk and earthquake event loss risk in a uniform manner globally using open source software. Following a series of planning workshops in 2007 and 2008 the Global Earthquake Model (GEM) Foundation was formally established in March 2009 with its Secretariat located at the EUCENTRE in Pavia, Italy to oversee the initial 5 year phase of development which ran from the beginning of 2009 to the end of 2013. The project is a continuing one and has now entered its second 5 year phase.

During the 2009-2013 phase, GEM received €5.6 million (A$8.2 million) from 14 public participants and €14 million (A$20.6 million) from 11 private participants, and was supported by 10 international organisations. Australia through Geoscience Australia is one of the 14 OECD member countries supporting the project. The private participants include major reinsurance companies such as Munich Re and Swiss Re, major global insurance companies such as Zurich and FM Global, the global insurance and reinsurance broking firm Willis, and the international catastrophe loss risk modelling firm AIR. The International Association of Earthquake Engineering, of which AEES is a member, is one of the supporting organisations. The current Chairman of the GEM Foundation Board is John Schneider from Geoscience Australia.

A large effort during the first 5 year phase has been made on establishing databases of information relative to earthquake hazard and loss risk. A major outcome has been the development of the ISC-GEM database of instrumental earthquake records undertaken in conjunction with the International Seismological Centre in the UK which contains approximately 20,000 records of global earthquakes dating back to 1900. Other significant achievements have been the development of a taxonomy for recording exposure, the critical appraisal of currently available ground motion prediction equations and recommendations for their use globally, a global database of available information on active faults, a catalogue of major earthquakes known to have occurred over the past 1000 years, a geodetic strain rate model based on 18,000 GPS velocity measurements worldwide, standardised systems for collecting exposure data using hand held tablets and satellite imagery, and guidelines for the development of vulnerability and fragility functions. Much of this information is available through the GEM website www.globalquakemodel.org/gem/

The planned open source model, known as OpenQuake, is also well on the way with Version 1.0 expected to be released by the end of this year. The model essentially comprises two basic components, the Engine which embodies the calculations subroutines and is common to all users, and the Platform which inputs and outputs data to the Engine, and contains the subroutines to do this along with the user interface, and which will allow a considerable degree of flexibility for users in respect of how they use OpenQuake. The Engine is essentially complete and the Platform in the process of being completed. Currently there is a version of the model available on the website for public alpha testing which is known as OATS. Version 1.0 is being developed to undertake universal point seismic hazard assessment (PSHA) analysis, universal direct event loss risk analysis for use in management and mitigation of economic losses arising from earthquake events, and evaluation of integrated disaster impact including socio-economic factors for emergency risk management.

GEM has the potential to be a major vehicle in applying the extensive results of earthquake engineering research to the global community, and particularly to the less developed regions lacking the expertise available in well developed countries.