Exposure Database

GED in a nutshell

The database structure is built to contain information on buildings and people from the country-level all the way down to the individual building. The first version of the Global Exposure Database (GED) will contain aggregate information on population and the number/built area/reconstruction cost of residential and non-residential buildings at a 1km resolution. Detailed datasets on single buildings will be available for a selected number of areas and will increase over time.

During the first three years it will not be possible to collect data for the whole globe for all levels in a homogeneous manner, which is tested and verified by (local) experts. To complete the project, we count on governments, companies, engineers, researchers, other organisations and individuals to continue the project, feed data and implement the dataset, so that its value continues to increase and it can become a resource for many types of applications worldwide, beyond earthquakes and natural hazard.

How is it of relevance?

Calculating loss and damage worldwide

(Earthquake) Risk = Vulnerability x Hazard x Exposure

Or in other words, you need to know what the elements are that are exposed to earthquake hazard and what their value is, to be able to calculate and understand what the likelihood is of  loss and damage due to earthquakes in a certain area.

The main elements at risk from earthquakes are people and buildings, but also infrastructure and other property play an important role. Currently there is no global database that captures all this information, as a basis for risk estimation on global, national and local level.

Because collecting and collating all this information for the entire globe will take many years, the first version of the global exposure database developed within GEM will focus on people and residential buildings with global coverage at national and sub-national (province, municipality) level. For some areas there will be datasets available of single buildings, but the database is built to capture all data, and by using the inventory data capture tools, it will be easy to enhance the database with new, more locally detailed data.  

It can then be used in many ways, for example to improve urban planning, to decide on retrofit strategies and for risk mapping and management.  

Multi Hazard

The database is developed as part of GEM and is optimized for earthquake risk assessment. However it can be used for many different types of hazards, natural and man-made, because people, buildings (and in a later stage perhaps infrastructure) are always the key elements being exposed to them.

What are its characteristics?

1. It is multi-scale:

The GED will feature data at 4 different geographical scales:

  • At country-level, the GED will contain statistical information about the total number of and distribution of dwellings and/or buildings fractions in a country, divided by building type. A distinction is made between data for rural and urban areas, and data on residential buildings or other types of buildings (public, commercial use, etc.) that can be grouped together as ‘non-residential. Information on the value of buildings per square meter will be provided where possible. 
  • At sub-country level (province/city), the GED will contain statistical information about the total number of and distribution of dwellings and/or buildings fractions in a region or province of a country. A distinction is made between data for rural and urban areas, and data on residential buildings or other types of buildings (public, commercial use, etc.) that can be grouped together as ‘non-residential. Information on the value of buildings per square meter will be provided where possible. 
  • At the local level, the GED divides the globe in a grid, with blocks of roughly 1x1km. The database will contain aggregated statistical information about each grid cell, in terms of the buildings, building distribution.
  • Detailed information on single buildings will be provided when available.

2. It is a unique product of data-fusion:

Data from household surveys, ground surveys, maps and GIS and from remote sensing (satellite imagery) is being combined to obtain information on:

  • population: totals, number of people living in rural / urban areas, behaviour (night time and day time)
  • building types and value of the building: based on information of roofs, walls and floors of household surveys
  • building density and building types: satellite imagery, maps, ground surveys

Who is developing it?

GED4GEM is a 3-year project that is carried out by a variety of leading institutions from around the globe working together to define best practices and homogenise data for all countries in the world. Project partners are the University Pavia (lead), UN-HABITAT, CIESIN, Joint Research Centre and ImageCat inc. Geoscience Australia, Eucentre and USGS are advising partners.

How can I use it?

The database will become available to all as part of the OpenQuake platform towards the end of 2014. You will be able to visualize the data in a GIS-environment, you will be able to download subsets of the dataset that you need for your work or analysis. In case you would like access to the full database you can contact us to explore options to collaborate.  

Intuitive Visualisation of Building Exposure

Urban, Rural, Residential and Non-residential building types all in one view

How can I contribute?

  • We are very much interested in (small) building datasets you would be willing to share with GEM, or to receive other types of information you possess on the buildings and their value in your country, region or city.

  • For Level 0/1 data you can compile the Level 0/1 data contribution form, whilst for Level 0/2 data we request gridded data in .csv format, and we provide an Excel template for this data. Please contact inventory@globalquakemodel.org for a copy of the Global Exposure Database grid in your region of interest. All data contributions can be uploaded through the following web form