In the light of the Paris Agreement on Climate Change, the project "Challenges of Global Agriculture in a Climate Change Context by 2050" (AgCLIM50) assesses the impact of climate change on the agricultural sector by 2050, as well as the economic consequences of stringent global emission mitigation efforts under different socioeconomic and representative greenhouse gas concentration pathways.

The employed set of integrated assessment (IMAGE), partial equilibrium (CAPRI, GLOBIOM, MAgPIE) and computable general equilibrium (MAGNET) models ensures a good coverage of biophysical and economic features. Using different models and scenarios helps to explore a wide range of potential impacts, uncertainties, and how the modelling results are affected by data and methodological choices. Model inputs are harmonised by using the same projections for population and GDP growth, as well as relative biophysical crop yield changes due to climate change. Model results can differ depending on model characteristics and the specific quantitative implementations of the socioeconomic storylines.

Five global multi-region multi-sector models have been used for the analysis:

• CAPRI: Common Agricultural Policy Regionalised Impact Modelling System
• GLOBIOM: Global Biosphere Management Model
• IMAGE: Integrated Model to Assess the Global Environment
• MAGNET: Modular Applied GeNeral Equilibrium Tool
• MAgPIE: Model of Agricultural Production and its Impact on the Environment

The employed set of integrated assessment (IMAGE), partial equilibrium (CAPRI, GLOBIOM, MAgPIE) and computable general equilibrium (MAGNET) models ensures a good coverage of biophysical and economic features. Both the spatial resolution and the level of disaggregation of the agricultural sector are very different across these models.

Shared Socioeconomic Pathways (SSPs) were developed by the climate change research community to represent the socioeconomic dimension of the new climate scenarios (O’Neill et al. 2014; 2017). The SSPs contain narratives for future developments of demographics, economy and lifestyle, policies and institutions, technology, and environment and natural resources (O’Neill et al. 2017). Furthermore, the SSPs comprise quantitative projections of population and gross domestic product (GDP) at the country level (Crespo Cuaresma 2017; Dellink et al. 2017; KC and Lutz 2017; Leimbach et al. 2017).

In this project we focus on three SSPs: SSP1 (Sustainability) - featuring relatively high levels of economic growth, lower levels of demographic growth, high levels of education, international cooperation, fast technological growth, convergence between developed and developing countries, sustainability concerns in consumer behaviour…, SSP2 (Middle of the Road) - representing business as usual development, and SSP3 (Regional Rivalry/Fragmentation), featuring opposite tendencies to SSP1 – relatively slow economic growth, sustained population growth,…

Scenarios are implemented for the projection year 2050 and have global coverage with disaggregation into major world regions. Results are analysed with a focus on global implications of climate change and related policies. The focus of the analysis is on major crop groups (wheat, coarse grains, rice, sugar, oilseeds) and livestock products (meat from monogastrics, beef and milk).

Model inputs are harmonized by using the same projections for population and GDP growth over time, but model results differ depending on the specific quantitative implementations of the SSP storylines. The effects of ambitious mitigation with residual climate impacts, while stabilizing global warming at 2°C, is also systematically compared. The scenario setting is outlined in the next table, indicating also the adaptation challenge for agriculture within the different SSPs.

Global scenario results are presented with respect to the following variables: population, GDP, total agricultural production, production of ruminants and non-ruminants, land use (total, crops and livestock related), crop yields, producer prices (crops, livestock products), and emissions (CO₂ from land use, CH₄ and N₂O from agriculture). All results are presented as index changes for the projection year 2050 compared to 2010.

Global population in 2050

Changes in population are an exogenous driver in all models included in this study. All follow the general SSP storyline, with lower population growth in SSP1 than in SSP2 and SSP3. Population growth is assumed to be independent of the climate change and mitigation dimensions in scenarios.

Global GDP in 2050

GDP developments are exogenous in GLOBIOM, CAPRI, IMAGE and MAgPIE, and endogenous in MAGNET. Absolute numbers are slightly different across models, as they have different methods to convert the GDP Purchasing Power Parity (PPP) to GDP Market Exchange Rate (MER), which is reported here. However, the relative changes between SSPs are in line across models.

The SSP storylines are that economic growth is the highest in SSP1 and lowest in SSP3, i.e. GDP developments are opposite to population developments if one moves from SSP1 to SSP3. The implications for food demand are, therefore, uncertain as higher population means more people to feed, whereas lower total GDP means that there are less total resources to spend on food. In addition, assumptions about dietary preferences and waste management vary across the models, which makes it difficult to predict the implications for food demand directly from the population and GDP drivers.

Total global agricultural production in 2050

In general, total agricultural production in SSP1 is less than in SSP2 which in turn is less than in SSP3. This indicates that the demand for agricultural products is more influenced by the population developments and assumptions about waste and dietary preferences than GDP developments. CAPRI exhibits the opposite trend, indicating that GDP developments are a stronger driver than population and that the implementation of dietary changes has been more conservative than in the other models. SSP1 is lower in GLOBIOM as additional preference changes are assumed relative to MAGNET\IMAGE. In SSP3, MAGNET\IMAGE assume additional changes that increase demand and therefore also agricultural production. These additional changes in MAGNET\IMAGE include a 33% waste increase, 25% higher meat consumption and 10% higher import taxes of food. These shifts all induce additional production in MAGNET\IMAGE, but they are not included in GLOBIOM, which only considers a slower reduction in wastes compared to SSP2 and SSP1. In MAgPIE, higher production in SSP3 compared to SSP2 and SSP1 is mainly caused by population growth combined with SSP-specific income-demand responses (e.g., generally healthier diets in SSP1 compared to SSP2 and SSP3).

The impact of RCP6.0 climate forcing on agricultural production (comparing the first with the second scenario within an SSP) is quite similar to the impact of RCP2.6 climate forcing (comparing the third with the fourth scenario within an SSP). This is due to the selection of representative median scenarios as they actually imply rather similar yield impacts of the two RCPs in 2050. It has to be noted that crop model results show that climate impacts will increasingly differ between RCP2.6 and RCP6.0 after 2050.

Total production of ruminants in 2050

The additional cost of agricultural mitigation measures reduces production, most notably for rice and especially ruminant meat, in most models. In MAgPIE, final food demand for all products is driven by an exogenous trend at the regional level, and therefore regional demand is not influenced by mitigation policies. With global demand being exogenous, global production of ruminant meat does also not change in the mitigation scenarios. However, in MAgPIE there may be regional changes in production due to shifts in trade across regions. Moreover, production of feed crops changes if regional livestock production is changed due to mitigation policies.

The negative impact of mitigation policies on ruminant meat production is most pronounced in CAPRI. In CAPRI ruminant production in SSP3 is lower than in SSP1 and SSP2, indicating that GDP as a demand driver for meat, reinforced with a dependency of yields on GDP, has a stronger impact than population as demand driver. Moreover, there are no shifts in waste/meat preferences in CAPRI when comparing SSP2 and SSP3, which partially lead to an increase in ruminant production in MAGNET\IMAGE under SSP3.

Total production of non-ruminants in 2050

The production of non-ruminants also decreases due to the mitigation measures in most models. For CAPRI, an increase in production of some commodities (dairy and non-ruminants) is observed. This is due to the large decrease in ruminant meat production induced by the mitigation policies (as ruminant meats have the highest emission intensities their production decreases most). The decrease in production leads to a price increase for ruminant meat and therefore consumers reduce total consumption but also shift to cheaper non-ruminant meat (poultry and pork meat), which has lower emission intensities and therefore is less affected than the ruminant meats. From a technical perspective this is driven by higher cross price elasticities for CAPRI than for MAGNET and the other models do not include cross price elasticities.

Total land used by crops in 2050

Cropland area generally increases when moving from SSP1 over SSP2 to SSP3. Climate change increases cropland area in IMAGE\MAGNET, MAgPIE and CAPRI, whereas cropland area decreases in GLOBIOM. For the former four models the lower yield and an inelastic food demand induce the higher land use. For GLOBIOM the mechanism causing the negative impact on cropland is that grasslands are relatively favoured by climate change compared to crops, which leads in some regions to a small shift in the livestock production systems towards more grazing and less reliance on feed crops.

In all models except CAPRI, cropland area decreases due to mitigation measures. The decrease is caused by less available land due to afforestation and demand for bioenergy. In MAgPIE, reduced demand for livestock feed also contributes to this result. However, this does not hold for CAPRI, where mitigation was exclusively incentivised on non-CO₂ emissions. Hence, production shifts within agriculture, more specifically grassland being released from the decreasing ruminant production, explain why cropland expands in CAPRI in contrast to the other models.

Total land used by livestock in 2050

Mitigation measures, in particular afforestation, result in an even larger decrease in area used by livestock in the GLOBIOM, IMAGE and MAGNET models as compared to crops. This is because land is allocated (with imperfect substitution) according to its rental price and cultivating crops gives higher returns to land than livestock. The decrease in available land due to afforestation, therefore, impacts more on the livestock sector. In CAPRI this effect is not reflected as afforestation is not specifically considered. The decrease in SSP1 is higher in GLOBIOM due to the strong preference shifts away from ruminant meat in SSP1.

Real producer price of crops in 2050

Crop producer prices increase from SSP1 to SSP2 to SSP3 in all models. Compared to 2010, producer prices decrease in SSP1 in all scenarios, whereas they are stable or increase slightly in SSP2 and increase in SSP3. Important drivers on the production side are lower yields in SSP3 than in SSP2 and SSP1. The main demand drivers, population and income, and the interplay between demand and supply determine the prices, which are clearly different in the various models. Price changes are small in GLOBIOM and CAPRI, intermediate in MAgPIE and rather big in MAGNET. The endogenous calibration of technical change in MAGNET contributes to these bigger price effects. In MAgPIE, producer prices are higher in SSP3 due to increased production costs as a result of more restricted trade and augmented costs for additional technological change. Mitigation measures as well as climate impacts induce additional pressures, leading to even higher producer prices. As demand is exogenous in MAgPIE, all the adjustments to climate impacts and mitigation measures have to come from the production side, including reallocation of production through international trade. As agricultural land expansion is limited, especially with strong mitigation policies and restricted trade in SSP3, endogenous yield increase is the main mechanism to compensate.

Real producer price of livestock products in 2050

Developments in producer prices for livestock products are similar to those for crops in the models. However, climate change mitigation measures lead to an even higher increase in producer prices for livestock products than for crops. The impact is higher, because livestock is more emission intensive than crop production, and emission taxes, therefore, increase livestock production costs relatively more than the costs for crop production.

Total emissions of CH₄ and N₂O from agriculture in 2050

Mitigation measures strongly reduce agricultural non-CO₂ (i.e. methane and nitrous oxide) emissions in CAPRI, IMAGE and MAgPIE. As the latter two models use the same marginal abatement cost curves, the relative reduction in IMAGE and MAgPIE is similar, although slightly higher in IMAGE. In both models, the relative reduction is comparable across the different SSPs, as in all SSPs much of the mitigation potential is already applied early due to fast increasing carbon taxes. In CAPRI, the mitigation effort is similar across the SSPs as the same emission taxes and the same assumptions on mitigation technologies are applied across SSPs. In GLOBIOM, the emission reduction is much smaller than in the other three models, and differs across SSPs, with the lowest reduction in SSP3. This is related to the fact that mitigation in GLOBIOM is mostly based on GHG efficiency improvements through production system composition changes and production relocation across regions, both mediated through prices, but not via technological mitigation measures. Price-mediated consumption shifts are ignored in MAgPIE, and therefore, for example, also the pricing of methane emissions does not lead to consumption changes for livestock products, which dampens production decreases and hence limits related emissions mitigation in the mitigation scenarios. In IMAGE, technical mitigation measures are combined with system-wide effects due to GHG emission pricing (calculated via MAGNET). In CAPRI, the decline in agricultural non-CO₂ emissions is similar to the decline in IMAGE and MAgPIE as the same reference has been used for mitigation effects in non-European regions. CAPRI has a quite detailed non-CO₂ mitigation modelling for Europe, but the global results are dominated by other regions.