Modeling the future of food systems with IFPRI’s IMPACT modeling system
Grigory Bronevetsky
Grigory Bronevetsky
Summary:
Focus on entire global food system value cycle:
Resources
Production
Distribution
Consumption
Surrounding social, economic governance systems
Key challenges:
Will we have enough food for the entire global population?
Hunger and micronutrient deficiencies
Resource use
Slowing productivity growth
Climate change
Geopolitical instability
IFPRI: International Food Policy Research Center
Mapping/modeling food systems
Testing & scaling solutions
Food Policy
Key questions
Will we produce enough food?
Will people be able to afford it?
Will we do this nature-sustainably?
Answering these questions using models
Long-term prediction is very challenging because conditions, policies and technologies change
Past behavior informs causal drivers and relationships that drive food prices, supplies and systems
Many models of the global food system
IMPACT, AIM, CAPRI, ENVISAGE, GLOBIOM, MAGNET, MAgPIE, RIAPA
Driven by climate models, which drive biophysics (crops, hydrology, etc.), and ultimately to economic dynamics
AGMIP brings together agricultural modelers (https://agmip.org/)
IMPACT uses DSSAT (https://dssat.net/) as their crop growth model, population forecasts from IIASA (https://iiasa.ac.at/models-tools-data/ssp-2023), Climate forecasts from IPCC and other drivers from FAO, World Bank, USDA and others
Global economic model:
158 countries, 161 water basins and their intersections
60+ commodities (plants and animals)
Driven by: population growth, productivity growth (formerly driven by land and inputs, now more by tech innovation)
Warning: development of global agriculture occurred during past 10k years when temperatures were very stable; leaving this range likely to cause major disruptions to food systems
Key insights:
Maize production more vulnerable to climate change than wheat
Tropical regions more vulnerable to more temperate ones
Upto mid-century socioeconomic factors will drive yield more than climate change but afterwards, climate change will be the dominant driver
Population growth globally drives increases in food demand;
Rich country demand for cereals, roots and tubers will drop but demand for oils will keep increasing
Poorer countries will continue to demand more stable crops
Agriculture will use up more land area in poor countries due to poor per-hectare yield
Climate change will degrade yields, but the effect will be blunted by improving technology and farming practices (yields increase but less than they could have been without climate change)
Technology impacts are modeled based on estimates from domain experts
Incorporates elasticities of crop production based on price
Global cereals production will vary globally
Lower income regions will become importers
Rich and Latin America will export more
IMPACT can model the impacts of policies and investments
Has been used to evaluate risk of hunger under climate change or no climate change
Climate change will reverse a lot of progress in improving food security
Targeted investments in food systems will reverse these impacts to a point to where food security is even better than in the no-climate change scenario
Key findings:
Food demand will increase and will be met by more supply
Will become more expensive to provide due to climate change, which will have a cost in ecosystem services
Poor countries will depend more on food imports
Ongoing work on IMPACT
Land use change
Modeling water, climate variability, etc.
Modeling social outcomes (e.g. nutrition), accounting for gender and other social strata
Documentation: https://cgspace.cgiar.org/items/4a89cd40-7e86-4392-88dc-464bde71a5b8
Impact of technology on crops: https://cgspace.cgiar.org/items/ee6ed5de-ae98-4a0a-af14-57aa9d237be2
Other models:
RIAPA: national-scale: https://www.ifpri.org/project/riapa-model/
SPAM: Spatial Production Allocation Model: https://foresight.cgiar.org/spam/