Climate change affects agricultural crops in large part through shifts in three core climate change factors: growing season temperature, precipitation (rainfall) amounts and the ground-level concentration of carbon dioxide ([CO2], which benefits plant growth, photosynthesis and water retention). The impact of these climate change factors depends on the crop type, management system, and geographic location — with maize tending to have a more pessimistic response than wheat, rice or soybean due to a reduced CO2 effect, for example. Farms that use higher nitrogen fertilizer amounts are also better able to keep up with higher growth rates under elevated carbon dioxide concentrations, while low fertilizer farms may face nitrogen limitations that obscure climate response. Regions that are already warm or dry may also be the first to cross dangerous thresholds as the climate changes, leading to strong geographic patterns in climate response that highlight areas of heightened vulnerability and exposure to climate hazards that can inform adaptation planning priorities.

The maps in this crop model Impacts Explorer show long-term average crop yield changes for maize, wheat, rice, and soybean in response to systematic local changes in temperature, precipitation, CO2 concentration and fertilizer application as simulated by the AgMIP Global Gridded Crop Model Intercomparison Project (GGCMI). GGCMI is a collaborative effort across more than a dozen process-based crop modeling groups around the world that each simulate crop growth over the course of a season as determined by plant genetics, farm management and the soil and weather environment in which plants grow. Inputs and assumptions are harmonized across the teams to reduced overall uncertainty, but the sparsity of available field measurements and required simplifications render global yield estimates far into the future still highly uncertain. Here, an ensemble of 6 crop models is used to represent core climate factor responses that are considered robust by the crop modeling community.

You can use the viewer’s interactive sliders to visualize responses to isolated and combined changes in temperature, precipitation (rainfall), carbon dioxide concentration and fertilizer application. The current situation (corresponding to the 1980-2010 period) is indicated for each climate factor and all changes are relative to that period’s conditions. Yield responses of rainfed crops are shown as relative changes [%] compared with the current situation. The irrigated response can be seen when moving the precipitation perturbation slider to the far right (irrigated). To highlight areas of concentrated production and observed yield levels, you can also see the observational reference yield map [t/ha] by ticking the reference yield box.

The reference for the crop model simulations can be found here:
Franke, J., C. Müller, J. Elliott, A. C. Ruane, J. Jägermeyr, J. Balkovic, P. Ciais, M. Dury, P. Falloon, C. Folberth, L. François, T. A. M. Pugh, A. Reddy, X. Wang, K. Williams, and F. Zabel, 2020, The GGCMI Phase 2 experiment: global gridded crop model simulations under uniform changes in CO2, temperature, water, and nitrogen levels (protocol version 1.0). Geosci. Model Dev. 13, 2315–2336 (2020).

Population at risk of hunger (in millions)

The hunger metric is based on a calorie relationship between per capita availability and minimum required dietary energy for the average consumer in a country.  It gives the number of people in a country who face chronic hunger in a year.  It is calculated from the share of population at risk of hunger and population projections.

Share of population at risk of hunger (% of total population)

The share of population at risk of hunger is based on a calorie relationship between per capita availability and minimum required dietary energy for the average consumer in a country.  It gives the share of the population in a country that faces chronic hunger in a year.  It is equivalent to FAO’s prevalence of undernourishment (PoU).
 

KCAL availability per capita

Shows the total calorie availability for the average consumer in each country.  Recommended levels are usually between 2100 and 2600 KCAL per person per day.  Below these levels, hunger issues are likely across a population.  Above these levels, overnutrition will become more prevalent.  Distribution issues at subnational levels will mean that not everyone in a country will have access to this level of calories (some will have more, some less).
 

Food availability per capita, by commodity

Shows the food availability for the average consumer in each country for four major commodities (maize, rice, wheat, and soybean).  Diets will vary at the subnational level, but general trends in levels of consumption of different types of foods (meat, staples, fruits & vegetables, pulses, oils, etc) will indicate how diets are changing over time.  Distribution issues at subnational levels will mean that not everyone in a country will have access to these levels of food (some will have more, some less).

Net trade / Total demand: Net trade as proportion of total demand, by commodity

Shows how much of each commodity is imported or exported in each country relative to how much is being consumed.  Values less than zero show where national consumption needs to be met with imports (a value of -1 indicates all of national consumptions is met through imports).  Values greater than zero show countries that will be exporting their surplus production. 

Total production / total population: production per capita, by commodity

Shows how much of each commodity is produced in each country relative to the population size.  By looking at production per capita, we see a coarse but better indication of productivity across countries (which will not simply be dominated by the largest countries).
 

World commodity prices in constant 2005 US$

Gives an overall indication of the trends in prices for each of the commodities at the global level.  While prices encountered by consumers in the marketplace and by farmers at the farmgate (and other actors along the value chain) will differ from global prices, overall trends will eventually reach consumers and producers who will react accordingly.