What do we know about the future of land in relation to food system transformation?
By Ricky Robertson and Nicola Cenacchi
Food, land, and water systems face daunting challenges in the future, and the body of research exploring these challenges is growing rapidly. This note is part of a series developed by the CGIAR Foresight Initiative to summarize what we know today about the future of various aspects of food systems. The goal of these notes is to serve as a quick reference, point to further information, and help guide future research and decisions.
- The total amount of cropland across the globe is likely to expand over the next three decades due to the rising demand for food along with feed for livestock.
- Pasture will likely be stable or contract as livestock production continues to shift away from grazing and toward intensive use of feed and transported fodder.
- Climate change will increase the overall challenge and drive additional cropland expansion by generally reducing potential yields, although some crops and locations will benefit (typically higher latitudes and to a lesser extent higher elevations).
- Natural land will be displaced by cropland in some areas. In particular, forests in the tropics are at greater risk of conversion than in other regions. Beyond cropland incursion, the types, mixes, and footprints of natural vegetation may also be affected by climate change.
Recent trends and challenges
Land continues to be transformed by human activities and management. With the growing human population and higher incomes, the demand for land has risen for various purposes like habitation, crop cultivation, livestock rearing, and timber production. Within the global trend, local dynamics can feature diverse patterns of land use and change which partially cancel each other out when looking at world-wide figures (Winkler, et al., 2021).
Despite the increasing accuracy of Earth observation systems, estimates of the precise 2020 cropland area, or 2000-2020 expansion rates, are still uncertain, largely due to differing land-use definitions across maps (Tubiello et al., 2023). The FAOStat database continues to be the primary starting point for understanding global trends in cropland (Tubiello et al., 2023). The country-level values reported in the database are for harvested areas of crops during the reported year in each country which can then be aggregated to the entire world. The corresponding actual space taken up on the ground (the physical area) will be smaller because some locations support multiple harvests in a single year whether of the same crop or different crops using the same fields.
Reported harvested areas have been consistently trending upward. Currently, there are about 290 million hectares (Mha) of harvested area. Since 2002, the rate of change has been approximately 14.5 Mha/year of additional harvested area while prior to 2002, the rate was much lower at about 5.4 Mha/year. The main crops associated with the post-2002 acceleration are maize and soybeans, but they only account for 40% of the acceleration (Cassman and Grassini, 2020). Oil palm, sesame seed, cassava, and rape/colza seed have also made important contributions to the acceleration showing high area growth compared to their 2002 levels. Geographically, major cropland expansion has occurred in South America and Africa.
The trends in reported pasture areas are opposite to cropland and have been declining over the last two decades (Ritchie, 2022). The shift arises from the interplay between changes in demand for animal products, intensification of animal production (which decreases the need for pasture by shifting to other sources of feed/fodder instead of pasture-based forage, which in turn requires more cropland), and the challenges within the reporting process. The aggregate numbers show a global decline of about 4.5 Mha/year of pasture recently.
The overall combination of cropland and pasture has leveled off or is possibly meaningfully decreasing as a result of this intensification of livestock production based on extensification and intensification of cropland to produce the additional feed (Ritchie, 2022).
The most visible of the natural ecosystems are forests. Much like cropland and pastures, there are forest gains in some regions (primarily the temperate zones) and losses in others (largely in the tropics). The bulk of forest losses are associated with agricultural expansion (with cropland consistently expanding and pasture moving or expanding in forest regions despite its overall downward trend; Pendrill et al., 2022). Overall forest area appears to be declining at a rate of about 4.75 million hectares per year.
What is the latest foresight research on this topic, and what do those studies show?
Although there is a wealth of research available on topics such as land use, crop distribution, and natural ecosystem factors, global-scale analyses in these areas are relatively infrequent. Integrated assessment models that address these issues use various strategies to pull together climate, hydrology, natural ecosystems, agriculture, and economics. Typically, the economics and agriculture components largely determine the amount of cropland which then interacts with the natural ecosystem component to decide where and how much of the different types of land will occur. The effects of climate change and water availability exert their influence throughout the process.
Schmitz et al. (2014) brought together the results from eleven economic models of global agriculture to compare projections to 2050 using a particular shared-socioeconomic pathway (SSP2, as defined for climate change studies). Without climate change, they found a median increase (from 2005 to 2050) of around 193 Mha of cropland (although ranging from -100 to +420 Mha) implying an overall annual average increase of about 4.3 Mha/year. When applying climate change, this increased to 317 Mha (or 7 Mha/year) meaning climate change increased the amount of cropland needed by 125 Mha (an additional 2.7 Mha/year above the no-climate-change case).
Similarly, Popp, et al. (2017) brought five integrated assessment models together to evaluate the effects of climate change along with all five of the shared-socioeconomic pathways from 2005 to 2100. In addition to cropland, they report projections for all land types including pasture, forests, and other natural land. For the unmitigated climate change baseline case with SSP2, the median of the cropland projections showed an increase of about 150 Mha and 231 Mha by 2100 (about 3.3 Mha/year before 2050 and 1.6 Mha/year after). Pasture is projected to be fairly stable. For the baseline case, forest coverage continues to decline by about 83 Mha in 2050 (averaging 1.8 Mha/year), but then stabilize and somewhat recover by 2100. Of course, there are a great diversity of outcomes depending on the SSP and climate change assumptions, some of which result in much greater cropland expansion and forest losses and others which show the opposite.
Stehfest et al. (2019) similarly assess six models, focusing on cropland, pasture, and what aspects of the modeling process have the greatest influence on various outcomes. The models showed a range of increases in cropland area under SSP2 of about 200 Mha by 2050 (about 5 Mha/year between 2010 and 2050), but much less agreement concerning pasture ranging from roughly 200 Mha decreases to 200 Mha increases (in contrast to the fairly clear decline seen in the recent past).
The role of land in the food system is entangled in varying degrees with climate, the water cycle, biodiversity, technology developments, consumer preferences, and energy policy. Integrated assessment models (which include land use components) have been used in combination with sectoral models to explore 1) how carbon taxes on emissions from land use change may increase food prices and negatively impact food security (Hasegawa, et al., 2018), 2) the crucial effect of alternative land use policies on future biodiversity extinction risk and other indicators (Leclère et al., 2020), and 3) reveal the effects that changes in livestock species may have on emissions (Cheng et al., 2022). Further developing these links is an important activity in foresight research and requires refining the techniques employed in land use research.
What are key gaps, questions, and opportunities for further foresight research on this topic?
There are several key opportunities for refining foresight research on how cropland, pasture, and natural ecosystems will interact in the future.
Firstly, improvements can be made regarding the data that currently characterize the present situation. Disagreements persist between the reported cropland/pasture/natural areas from administrative sources and those derived from imagery. Both sources present their own challenges and are actively being addressed. Foresight modelers welcome further advancements in this area, as they can refine their understanding of crucial processes and provide more accurate initial conditions for their simulations. A recent exploration of these challenges can be found in the works of Potapov et al. (2023) and Potapov et al. (2022).
Secondly, it is essential to establish connections across scales to enable global models to benefit from local information and local models to benefit from the global context. This way, both scales can effectively inform policymakers.
The third opportunity involves the representation of natural ecosystems. Most integrated assessment models currently show that changes in forest area (and to a lesser extent, other natural ecosystems) closely mirror changes in cropland area. Similarly, strong relationships exist between non-forest ecosystems and assumptions about bioenergy crops, where appropriate. However, it is likely that the dynamics among land types in existing studies primarily occur through conversion to or from human-managed land, with limited interconversion between natural land types. But in the face of climate change-induced temperature variations and altered precipitation patterns, the stability of natural land appears uncertain (e.g., concerning boreal forests, refer to Reich et al., 2022). Consequently, understanding the evolution of natural vegetation patterns becomes crucial for assessing habitat loss, degradation, and ultimately the challenges and opportunities for future biodiversity. This area thus warrants further research.
Lastly, progress on the aforementioned points must be continually integrated to better capture the links between biophysical conditions and economic mechanisms.
The authors of this note are Ricky Robertson, a Research Fellow with Foresight and Policy Modeling Unit, International Food Policy Research Institute (IFPRI); and Nicola Cenacchi, a Senior Research Analyst, Foresight and Policy Modeling Unit, IFPRI.
If you have any feedback or questions about this note, please get in touch with Ricky Robertson (email@example.com).
Cassman, K.G. and Grassini, P., 2020. A global perspective on sustainable intensification research. Nature Sustainability, 3(4), pp.262-268.
Cheng, L., Zhang, X., Reis, S., Ren, C., Xu, J. and Gu, B., 2022. A 12% switch from monogastric to ruminant livestock production can reduce emissions and boost crop production for 525 million people. Nature Food, 3(12), pp.1040-1051.
d’Annunzio, R., Sandker, M., Finegold, Y. and Min, Z., 2015. Projecting global forest area towards 2030. Forest Ecology and Management, 352, pp.124-133.
FAOSTAT. Food and Agricultural Commodities Production (FAO, 2022). http://faostat.fao.org/
Hasegawa, T., Fujimori, S., Havlík, P., Valin, H., Bodirsky, B.L., Doelman, J.C., Fellmann, T., Kyle, P., Koopman, J.F., Lotze-Campen, H. and Mason-D’Croz, D., 2018. Risk of increased food insecurity under stringent global climate change mitigation policy. Nature Climate Change, 8(8), pp.699-703.
Leclère, D., Obersteiner, M., Barrett, M., Butchart, S.H., Chaudhary, A., De Palma, A., DeClerck, F.A., Di Marco, M., Doelman, J.C., Dürauer, M. and Freeman, R., 2020. Bending the curve of terrestrial biodiversity needs an integrated strategy. Nature, 585(7826), pp.551-556.
Pendrill, F., Gardner, T.A., Meyfroidt, P., Persson, U.M., Adams, J., Azevedo, T., Bastos Lima, M.G., Baumann, M., Curtis, P.G., De Sy, V. and Garrett, R., 2022. Disentangling the numbers behind agriculture-driven tropical deforestation. Science, 377(6611), p.eabm9267.
Popp, A., Calvin, K., Fujimori, S., Havlik, P., Humpenöder, F., Stehfest, E., Bodirsky, B.L., Dietrich, J.P., Doelmann, J.C., Gusti, M. and Hasegawa, T. 2017. Land-use futures in the shared socio-economic pathways. Global Environmental Change, 42, pp.331-345.
Potapov, P., Hansen, M.C., Turubanova, S. et al. Reply to: Measuring the world’s cropland area. Nat Food (2023). https://doi-org.ifpri.idm.oclc.org/10.1038/s43016-022-00668-8
Potapov, P., Turubanova, S., Hansen, M.C. et al. Global maps of cropland extent and change show accelerated cropland expansion in the twenty-first century. Nat Food 3, 19–28 (2022). https://doi.org/10.1038/s43016-021-00429-z
Reich, P.B., Bermudez, R., Montgomery, R.A., Rich, R.L., Rice, K.E., Hobbie, S.E. and Stefanski, A., 2022. Even modest climate change may lead to major transitions in boreal forests. Nature, 608(7923), pp.540-545.
Ritchie, H. 2022. After millennia of agricultural expansion, the world has passed ‘peak agricultural land’. Our World in Data. https://ourworldindata.org/peak-agriculture-land
Schmitz, C. et al. Land‐use change trajectories up to 2050: insights from a global agro‐economic model comparison. Agricultural Economics 45, 69-84 (2014).
Stehfest, E., et al. Key determinants of global land-use projections. 2019. Nature Communications. 10:2166
Tubiello, Francesco N, Giulia Conchedda, Leon Casse, Hao Pengyu, Chen Zhongxin, Giorgia De Santis, Steffen Fritz, and Douglas Muchoney. “Author Correction: Measuring the World’s Cropland Area.” Nature Food 4, no. 1 (2023): 125. https://doi.org/10.1038/s43016-023-00697-x
Winkler, K., Fuchs, R., Rounsevell, M. et al. Global land use changes are four times greater than previously estimated. Nat Commun 12, 2501 (2021). https://doi.org/10.1038/s41467-021-22702-2