Consistent assessment of global fluxes of blue and green virtual water


A standard European or American consumer nowadays knows about global connections in the energy sector. He sees clearly that for daily transportation or heating he consumes fossil fuels that are mostly stemming from elsewhere of the globe. Via his decision on a product (say natural gas, gasoline, …) from a specific providing country (say Russian Federation, Venezuela, United Kingdom, …) he is directly involved in the economy of another nation, depleting resources there and leaving his local sources untapped.

But what about his knowledge on global water teleconnections? How is a consumer in one country involved in the water balance of another country? Does daily food consumption inEuropehave an impact on the water balance of theAral SeainCentral Asiaor theDead Seain theMiddle Eastwhich are currently running dry? How much of the water needed for food production is used for highly productive irrigated agriculture? Is it more efficient from a global, regional or national perspective for the water balance to deplete local water resources under irrigated conditions or to buy food elsewhere where it was produced under rainfed conditions?

Up to now estimations on the global transfer of water that is indirectly contained in food mostly have been based on global mean consumptive estimates, such as 1 cubic metre of water for 1 kilogram of wheat, and often do not distinguish between sources of water (irrigation or rain). Even when this separation was made, it was done with mean national input data only, but not with spatially explicit data.


It is our intention to draw a spatially detailed picture of agricultural crop-specific water consumption representative for the period 1998-2002 at a spatial resolution of 5 arc-minutes by 5 arc-minutes, with a distinction of irrigated and non-irrigated areas. We will include trade statistics to deliver the world-wide fluxes of respective blue (provided by irrigation) and green (provided by precipitation) virtual water content. Via model inter-comparison we will validate the results. New indices of water scarcity that consider virtual water flows are developed.


  • A new global data set of monthly crop-specific growing areas of 26 distinct irrigated and rainfed crops at a spatial resolution of 5 arc-minutes by 5 arc-minutes for the period 1998-2002 was generated. The data set itself and comprehensive documentation related to data sources, methodology and results is available at the MIRCA2000 web-page.

  • The new Global Crop Water Model (GCWM) has been developed to simulate potential and actual evapotranspiration on a 5 arc-minute by 5 arc-minute resolution, separately for irrigated and non-irrigated conditions. Through the combination with respective information on production per crop this results in spatially explicit blue and green virtual water contents. The model and the results of its application are presented on the GCWM web-page.

Coming soon

  • Global fluxes of blue and green virtual water will be computed by combining virtual water contents of  the specific crops with global bi-national trade data.

  • We will compare our results systematically to results of a new version of the Lund-Potsdam-Jena (LPJ) Dynamic Global Vegetation Model developed at the Potsdam-Institute for Climate Impact Research (PIK).

  • New indicators of water stress will be defined and applied considering also blue and green virtual water fluxes.


Keywords: global hydrology, water use, water resources management, food trade, virtual water content, agriculture, irrigation, crop calendar

Participants: Felix Portmann, Stefan Siebert, Petra Döll

Cooperating institutions: Potsdam-Institute for Climate Impact Research (PIK), Global Change and Natural Systems (Dr. Dieter Gerten)

Duration: September 2005 – August 2008

Funding entity: German Research Foundation (DFG, Deutsche Forschungsgemeinschaft)