Aqueduct Global Flood Risk Maps
Downloads and Resources
For the current scenario, we used hydrological data from 1960 through 1999 for generating flood inundations for 9 return periods, from 2-year flood to 1000-year flood, and 2010 GDP, population, and land use data for assessing flood impacts.
For future projections, we used 5 GCMs (Global Climate Models) from CMIP5 (Coupled Model Intercomparison Project Phase 5) projecting future flood inundations under two climate scenarios, RCP4.5 (Representative Concentration Pathway) and RCP8.5, and projected socio-economic changes using SSP2 (Shared Socio-economic Pathway) and SSP3, from the Intergovernmental Panel on Climate Change Assessment Report 5.
For more information on uses, methodology, and limitations, visit wri.org/floods
A number of caveats should be taken into account while using the hazard and risk data from the Aqueduct Flood Analyzer:
Types of flooding. In this initial version of the Analyzer, we only simulate large-scale river flooding, and not coastal flooding, flash flooding, or pluvial flooding. Flood protection. The flood hazard maps represent situations with no flood protection measures (established for example through dikes and water retention areas). Hence, where such measures already exist, the flood extent (and therefore the affected population) will be overestimated. In the results displayed in the online Analyzer, the user can include a protection standard (expressed in terms of a return period in years). Damage estimate assumptions. To estimate urban damage, we assumed that the relationship between the inundation depth and the actual damage is the same throughout the entire world. Additional limitations. Further uncertainties in the hazard and risk estimates may be due to inaccuracies in the climate data used to estimate river discharges, inaccuracies in the elevation data used to simulate inundation, and model simplifications with respect to the physics.
Given the limitations described above, the results are meant to provide a first impression of the distribution of risk among countries, provinces, and basins. This provides an indication of risk magnitude, and an impression of the magnitude of future change in risk that can be expected. The results should be used to focus attention on particular vulnerable areas and open dialogues on the risks and how they can be managed. The results can certainly not be used for the dimensioning of specific flood protection measures. This would require more detailed and locally calibrated models that include additional information on local conditions, including more accurate river profiles, structures, existing flood protection, reservoir conditions and management during floods, and more accurate information on exposure and vulnerability. It would also require thorough engagement with local experts and stakeholders.
Winsemius, H., Ward, P., Luo, T. 2015. 'Aqueduct Global Flood Risk Maps.' Washington, DC: World Resources Institute. Available online at:
Access & Use Information
License: Creative Commons Attribution 4.0 International License. Full license text available at Creative Commons Attribution 4.0
Page Last Updated: June 28, 2016