Home Page Methodology Data limitations The Scenarios Query the data Web resources CIESIN home page SEDAC HOME
Potential Impacts of Climate Change on World Food Supply
Data Sets from a Major Crop Modeling Study
by Cynthia Rosenzweig and Ana Iglesias


Note: This data set has now been superseded by a new data set on the effects of climate change on global food production under SRES emissions and socio-economic scenarios produced by the same authors. Users are recommended to use the new data instead. SEDAC is maintaining this web site for reference purposes.

Field Research Sites Used in the Crop Modeling Study

Map showing Sites

The outputs from three major atmosphere-ocean General Circulation Models (GCMs - also known as Global Climate Models) were used as inputs for crop simulation models at 125 agricultural sites representing major world agricultural regions. These process-based dynamic crop growth models simulate the effects of meteorological variables (e.g., temperature, precipitation, and solar radiation), environmental modifications (e.g. CO2 enrichment), and crop management strategies (e.g., fertilization, irrigation, and timing of labors) on crop physiology and yields. Models were used for the three most significant cereal crops - wheat, rice, and coarse grains (e.g., maize, barley, etc)  - which collectively account for over 85% of global cereal exports, and for soy bean, which accounts for 67% of trade in protein cake equivalent. For each GCM scenario (GISS, GFDL and UK Met Office), three levels of farmer adaptation to climate change were assumed:

  • No adaptation;
  • Level 1 adaptation: changes that imply small additional cost to the farmers and not necessary policy changes, such as shifts in planting dates, variety and crop, and increases in water application to irrigated crops; and
  • Level 2 adaptation: higher order adaptations that imply significant additional costs to the farmers, such as large shifts in crop production timing, increased fertilizer application, installation of irrigation systems, and development of new varieties, and/or changes in policy.
Crop simulation models were run based on the temperature and precipitation changes projected by the GCMs for an increased CO2 driven climate change, at the different adaptation levels described above, both with and without the direct effects of CO2 enrichment on crop physiology. A full list of the 46 scenarios used in this set of studies can be found under the Scenario section of this web resource. Once the crop simulation models were run, the estimated change in yields for each crop at specific field stations were used to estimate changes at sub-national or national levels (depending on the number of agricultural sites per country), and these changes were then aggregated to the trade economic regions of the economic model used (BLS). If a crop is simply not present in a country, then the estimated change in yields was not calculated (e.g., rice in Canada). For those countries that did not include agricultural sites, the results were extrapolated on the basis of their crop-climate profiles. This web resource makes available the estimated percentage change in yield for the four commodities (wheat, rice, coarse grains and soy bean) as well as aggregates for grains and all four commodities at the BLS region level.

Note that the estimated changes in yield under the no adaptation scenario do not consider additional changes that are already anticipated to take place, such as projected future increases in yields due to changes in technology. The adaptation scenarios include different levels of agricultural technology and inputs. The rate of technological innovations considered in adaptation Level 2 is correlated to the economic regional development. For other caveats and to better understand the limitations of the data, read through the data limitations section of this web resource.


Changes in crop yield by the 2080s, under scenarios of unmitigated emissions (top),
and those which lead to stabilization of CO2 at 750 ppm (middle) and 550 ppm (bottom).

More on the Basic Linked System

The Basic Linked System (BLS) consists of a set of linked national agricultural sector models. It is comprised of 16 national (including EU) models with a common structure, 4 models with country-specific structure and 14 regional group models. The political changes as well as changes in national boundaries of the very recent past are not in the BLS, although the model formulation has been adjusted away from centrally planned economies to more market-oriented behavior. The 20 models in the first two groups cover approximately 80 percent of world agricultural production; the remaining 20% is covered by the 14 regional models for countries with broadly similar attributes (for example African oil-exporting countries or Latin American high-income exporting countries). The BLS is a general equilibrium model system, with representation of all economic sectors, empirically estimated parameters and no unaccounted supply sources or demand sinks. Countries are linked through trade, world market prices and financial flows.

Introduction | Methodology | Data Limitations | The Scenarios | Query the Data | Web Resources | CIESIN home page | SEDAC home page
CIESIN Home Page
Need help or information? Contact SEDAC User Services
About SEDAC    Acknowledgments   Privacy Policy and Important Notices
SEDAC Home Page
Copyright© 2001. The Trustees of Columbia University in the City of New York. 

SEDAC Home CIESIN Home Web Resources Query the Data The Scenarios Data Limitations Introduction