ARTICLE https://doi.org/10.1038/s41467-021-22581-7 OPEN Energy implications of the 21st century agrarian transition Lorenzo Rosa 1,2, Maria Cristina Rulli 3, Saleem Ali4,5 ✉, Davide Danilo Chiarelli 3, Jampel Dell’Angelo6, Nathaniel D. Mueller7,8, Arnim Scheidel 9, Giuseppina Siciliano10 & Paolo D’Odorico 1 1234567890():,; The ongoing agrarian transition from small-holder farming to large-scale commercial agriculture is reshaping systems of production and human well-being in many regions. A fundamental part of this global transition is manifested in large-scale land acquisitions (LSLAs) by agribusinesses. Its energy implications, however, remain poorly understood. Here, we assess the multi-dimensional changes in fossil-fuel-based energy demand resulting from this agrarian transition. We focus on LSLAs by comparing two scenarios of low-input and high- input agricultural practices, exemplifying systems of production in place before and after the agrarian transition. A shift to high-input crop production requires industrial fertilizer application, mechanization of farming practices and irrigation, which increases by ~5 times fossil-fuel-based energy consumption compared to low-input agriculture. Given the high energy and carbon footprints of LSLAs and concerns over local energy access, our analysis highlights the need for an approach that prioritizes local resource access and incorporates energy-intensity analyses in land use governance. 1 Department of Environmental Science, Policy, and Management, University of California, Berkeley, Berkeley, CA, USA. 2 Institute of Energy and Process Engineering, ETH Zurich, 8092 Zurich, Switzerland. 3 Department of Civil and Environmental Engineering, Politecnico di Milano, Milan, Italy. 4 Department of Geography and Spatial Sciences, University of Delaware, Newark, DE, USA. 5 Sustainable Minerals Institute, University of Queensland, St Lucia, Australia. 6 Institute for Environmental Studies (IVM), Vrije Univeristeit Amsterdam, Amsterdam, The Netherlands. 7 Department of Ecosystem Science and Sustainability, Colorado State University, Fort Collins, CO, USA. 8 Department of Soil and Crop Sciences, Colorado State University, Fort Collins, CO, USA. 9 Institut de Ciència i Tecnologia Ambientals (ICTA-UAB), Universitat Autònoma de Barcelona, Bellaterra, Spain. 10 Centre for Development, Environment and Policy, SOAS, University of London, London, UK. ✉email: saleem@alum.mit.edu NATURE COMMUNICATIONS | (2021)12:2319 | https://doi.org/10.1038/s41467-021-22581-7 | www.nature.com/naturecommunications 1 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22581-7 T he historical limits of traditional agriculture, namely, soil dispossession and livelihood loss16–18. There is strong evidence fertility, water, and energy have been overcome by the that large-scale investments in agriculture by agribusiness cor- process of industrialization implemented through the porations have been implemented at the cost of smallholders, green revolution1. The adoption of fossil-fuel-based industrial traditional users, Indigenous people, and more vulnerable seg- fertilizers, new cultivars, and machineries—including motorized ments of the rural population12,16. LSLAs could paradoxically pumps—permitted to remove fundamental limitations associated increase crop production, while undermining local food security with soil nutrients, water resources, and labor. Advances in the through the production of energy-rich but nutrient-poor crops nitrogen fertilizer industry, irrigation, and other technologies for export markets14. While this radical transformation of increased land efficiency in agriculture2, but did not necessarily agrarian systems through LSLAs has been extensively investigated lead to inputs’ savings3–5. As a result, global crop production in relation to its political implications19,20 and its impacts on has tripled in the last 50 years, causing strong environmental property systems21, rural livelihoods22, crop yields14, water use impacts3. and redistribution23, food security24–26, environmental impacts27, A crucial challenge for humanity is to sustainably meet future and carbon emissions28, however, with few exceptions29, its food demand6, while limiting agriculture’s environmental energy implications remain poorly understood. footprint7. Assessments on how to reduce the environmental Global food systems are major energy users and contributors to footprint of agriculture commonly point to a combination of climate change. Food systems consume 15–30% of global primary measures, including the promotion of dietary shifts, reduction energy30 and emit 25–34% of global total greenhouse gas (GHG) of food waste, and halting agricultural expansion by closing emissions31–33. Although 35% of these emissions are caused by yield gaps—the difference between maximum attainable yield dairy and meat production, the remaining share is from activities and current yield for specific crops—through agricultural pertaining to crop production for direct human consumption31. intensification8. While, GHG emissions associated with deforestation and land use Increasing yields do not guarantee local hunger alleviation or change over 40 million hectares of LSLAs have been recently reduced pressure on local natural resources3, though evidence estimated to 8 gigatons CO2-equivalent in the 2000–2016 supports benefits to global health9 and land use2. Agricultural period28, the energy and related GHG emissions associated with intensification rarely produces win-win situations10; only a few the agrarian transition induced by LSLAs and other dynamics cases have demonstrated the contribution of agricultural inten- favoring the expansion of commercial farming have often been sification to multiple Sustainable Development Goals (SDGs), overlooked. such as ending hunger (SDG2) and sustainable land-use Here, we evaluate the energy and fossil-fuel implications of the (SDG15)10. Controversies over the role of agricultural intensifi- agricultural transition that is being globally promoted by LSLAs. cation for land use and users have intensified with the recent In particular, we focus on the energy-related systematic changes unprecedented rise in large-scale land acquisitions (LSLAs) that happen when moving from traditional low-input, labor- globally11,12. LSLAs refer to long-term and large-scale acquisi- intensive agricultural systems to intensified systems of crop tions of land property or use rights through domestic and foreign production with high-input of fertilizers, pesticides, machineries, actors, which have been sparked by—among other factors—food mechanized irrigation, and comparatively less labor associated security concerns and the rediscovery of agriculture as a key with the transition. Using a variety of data34–36, we estimate pre- investment sector following the 2008 food, financial, and energy and post-LSLA energy inputs in crop production over acquired crises13. According to the Land Matrix—a joint international lands. In particular, we assess the energy intensity that LSLAs initiative collecting data on LSLAs since 2000—90 million hec- would have in the case a low-input or a high-input farming were tares of land (about the surface area of Venezuela) have been developed. We also calculate the energy intensity of irrigation acquired globally by investors since 200014. over land deals considering different irrigation technologies. A few studies have discussed the potential opportunities of Finally, we discuss strategies to reduce the energy intensity of LSLAs for agricultural development, for instance, their potential farming, decrease reliance on fossil-fuel-based technologies, and to increase yields13 and economic benefits for the country provide policy recommendations that might help to lower the through large-scale investments into farmland15. However, other carbon emissions. For our analysis, we consider 197 land deals studies have described these land deals as land grabs that entail a with size greater than 200 hectares, obtained from the Land vast range of adverse impacts on local users, including land Matrix dataset36 (Fig. 1). Fig. 1 Geographical distribution of large-scale land acquisitions (LSLAs) considered in this study. We consider 197 land deals for agricultural use for which the geographic coordinates were available in the Land Matrix database36. These land deals are located in 39 countries and account for 4.07 million hectares of acquired land across Africa (73 deals, 2.4 Mha), Asia (43 deals and 0.58 Mha), Europe (33 deals and 0.54 Mha), and Latin America (11 deals and 0.55 Mha). 2 NATURE COMMUNICATIONS | (2021)12:2319 | https://doi.org/10.1038/s41467-021-22581-7 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22581-7 ARTICLE Fig. 2 Nitrogen application rates and yield gap closure levels before large-scale land acquisition (LSLA). a Shows average synthetic nitrogen application per harvested hectare in each deal (kg N per ha). An application smaller than 30 kg ha−1 is considered low-input agriculture; between 30 and 100 kg ha−1 medium-input agriculture; and greater than 100 kg ha−1 fairly high-input agriculture. b Agricultural productivity levels measured in terms of yield-gap fraction of major crops before LSLAs. Yield-gap fractions close to zero show low-yielding croplands with high yield gaps. Yield-gap fractions lower than 0.3 show land deals involving lands with high agricultural productivity. Data source: Mueller et al.35. The energy implications of LSLAs have not yet been Energy usage of low- and high-input agriculture. Figure 3 investigated in depth. Given that many existing policy shows the energy intensity per area for the main crops intended mechanisms support LSLAs and other pathways to agricultural by LSLAs at the farm level under low- and high-input agriculture intensification37, this study provides a first examination of the scenarios. Fossil-fuel-based energy inputs are from labor, fossil energy and associated energy implications of agricultural machinery, fertilizers, chemicals, fuels, and seeds. The figure also transitions following LSLAs. shows the energy intensity of oil palm and jatropha mills for biodiesel production. Among the four main staple crops (rice, wheat, maize, and soybean, which account for 70% of global food Results production), high-input rice production is the most energy- Agricultural productivity before LSLAs. To ascertain whether intensive agricultural practice. Because soybean and pulses have the development of commercial crop production intended the ability to fix nitrogen directly from the atmosphere—requir- by the land deals would entail agricultural intensification, we ing lower amounts of nitrogen fertilizers—their energy intensity is considered fertilizer application rates in areas prior to land relatively low compared to other staple crops. Interestingly, low- acquisition, as well as the fertilizer applications needed to reach input agriculture with sorghum—a widespread crop in Sub- maximum attainable yields35. Figure 2a shows that 80% of Saharan Africa—is the least energy-intensive agricultural practice. the deals involve land that was previously characterized by We find that a transition from sorghum and millet to cash crops low-input agriculture. Only 1% of the deals had high-input or staple crops might have profound energy implications. Cotton application of synthetic fertilizers before LSLAs. Figure 2b has the highest energy-intensity even for low-input options, shows that 85% of the deals occurred on land with low agri- which also raises important questions about non-food crop choice cultural productivity and high yield gaps. This implies that, for land use as well as broader consumer options for end-use for the purpose of developing large-scale commercial crop materials. Groundnut crops show the largest variation in energy production, additional inputs, such as fertilizers and water, are intensity for low- and high-input cultivation forms. needed for most of the land acquired through large-scale land We further analyzed the aggregate area of intended crops investments. to estimate fossil-fuel energy requirements and related carbon NATURE COMMUNICATIONS | (2021)12:2319 | https://doi.org/10.1038/s41467-021-22581-7 | www.nature.com/naturecommunications 3 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22581-7 Fig. 3 Fossil-fuel-based energy intensity at the farm level of low- and high-input agriculture. Fossil-fuel-based energy inputs are from labor, machinery, fertilizers, chemicals, fuels, and seeds. See Supplementary Table 1 for data sources. emissions in the context of global land-use change triggered by as well as lower attainable yields, and agroecological practices LSLAs (Fig. 4). Oil palm is the most frequently proposed crop of such as crop diversification, natural fertilizer use, biological pest LSLAs, accounting for 25% (1.02 Mha) of the total area of land control, deficit irrigation, and other soft-path water-management deals considered here (Fig. 4a). This raises concerns about high practices are expected to reduce the amount of fertilizers and levels of GHG emissions from palm oil plantations, particularly irrigation water, and therefore reduce energy usage. We tested the when they are associated with forest conversion38, as well as sensitivity of our results by assuming that 75% of the transacted about the high fossil energy requirements in palm oil milling for deal area is cultivated, and considering an 80% attainable yield. biodiesel production (see Fig. 3). Other important crops intended Moreover, considering global primary energy demand statistics, by LSLAs are other tree plantations, followed by a range of food we assumed that 80% of energy inputs are powered by fossil and fodder crops (Fig. 4a). fuels and the remaining fraction by renewable energy40. This We assumed that the whole deal area cultivated in low-input conservative scenario would lower GHG emissions of high-input agriculture transitioned to high-input agriculture. Developing farming by 3 million tons of CO2 per year, or half of the GHG the 4.07 Mha of intended crops under a scenario of low-input emissions of a business-as-usual strategy. agriculture would require an annual amount of 3 million barrels of oil equivalent as energy input (Fig. 4b). However, given that many land deals seek large-scale commercial crop production, Energy usage for irrigation on LSLAs. In addition to nutrients, water is another important input necessary to increase yields and high-input, capital-intensive production can be expected to avoid crop growth under water stress41. The use of irrigation prevail in many land deals28,39. Cultivating the intended crops enables reliable water supply while boosting crop productivity42. with high-input agriculture would require ~5 times more fossil- However, in many geographical settings, irrigation practices are fuel-based energy (or 15 million barrels of oil equivalent unsustainable because their water requirements exceed local per year). renewable water availability and irrigation induces the depletion We assessed the additional GHG emissions that would be of environmental flows and groundwater stocks43. Using recent generated through high-input farming over LSLAs. Assuming irrigation sustainability assessments44, we find that 20% of land that modern agricultural technologies are all powered with oil deals were located in regions where water resources are insuffi- and considering an average GHG emission intensity of oil equal cient to sustainably meet irrigation demand. to 492.6 kg CO2 equivalent per barrel of oil (see “Methods”), we Meeting crop water requirements has further energy find that high-input farming over LSLAs would emit an implications45. Irrigation requires energy to transfer water from additional 6 million tons of CO2 per year (approximately the the withdrawal source to the field—unless the local topography emission of Uganda or Uruguay in the year 2017). Under a business-as-usual strategy, large-scale farms are allows for gravity irrigation. We use the WATNEEDS expected to be established on the entire transacted area and crop water model46 to assess the energy intensity of irrigation are expected to be cultivated and used for intensified crop (i.e., the irrigation energy demand per unit of area) of each land production28 with high-inputs to maximize agricultural produc- deal, considering two of the most widespread irrigation systems: tivity. This implementation strategy yields the highest potential sprinkler and surface irrigation (Fig. 5). Because sprinkler energy use and provides an upper-bound estimate. Similarly, an irrigation has a higher operating pressure than surface upper-bound GHG emissions’ estimate is provided assuming that irrigation, our results show that sprinkler irrigation over LSLAs additional energy demand is met entirely with fossil-fuel-based requires more energy than surface irrigation—though surface energy sources. Different pathways of development can fuel high- irrigation requires larger volumes of water because of its lower input agriculture through less carbon-intensive energy sources efficiency (Fig. 5). Furthermore, the combination of crop type and therefore reduce carbon emissions. Inclusion of these factors, and climate in Eastern Europe, Asia, and Latin America is such 4 NATURE COMMUNICATIONS | (2021)12:2319 | https://doi.org/10.1038/s41467-021-22581-7 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22581-7 ARTICLE Fig. 4 Intended crops for agricultural use over large-scale land acquisitions (LSLAs) and their aggregate fossil energy footprint under low- and high- input agriculture scenarios. a Area (Mha) of intended crops for the sample of 197 land deals intended for agricultural use with size greater than 200 hectares, obtained from the Land Matrix dataset36. b Aggregated fossil-fuel-based energy input (million GJ yr−1) for the intended crop area under low- and high-input agricultural scenarios. c Aggregated fossil energy intensity of low- and high-input agriculture over LSLAs. Energy inputs from oil palm and jatropha milling were not accounted for in this assessment. Source: Estimates based on data from Supplementary Table 1 and Land Matrix36). Note: one barrel of oil equivalent is equal to 6.1 GJ. that land acquisitions display lower irrigation water require- irrigation water requirements and consequently energy demand ments and therefore smaller specific energy needs (Fig. 5a). for water pumping47,48. Conversely, Africa has the highest energy requirement for irrigation purposes, due to the combination of dry climate Discussion conditions and cultivated crops with high water needs (includ- Our research focuses on the neglected connection between the ing oil palm and sugarcane). Because of the perennial features fossil energy requirements of intensive agriculture and a fun- of these biofuel crops, results also show that land deals for damental component of the global agrarian transition, the biofuel are in general the most energy-intensive land uses in one produced by LSLAs. With particular reference to fertilizer relation to the energy required for irrigation (Fig. 5b). We usage and irrigation in low- and middle-income countries, we also show that if acquired lands were irrigated with sprinkler find that a transition to high-input agriculture produced by irrigation, an additional 4.3 million barrels of oil equivalent LSLAs would require 5 times more fossil energy than low-input per year (or 26 million GJ) of fossil-fuel-based energy compared agriculture. The higher energy-intensity of high-input agri- to non-irrigated conditions would be required to meet culture over the considered LSLAs (4.07 million hectares or full crop water demand (Fig. 5c). Improving crop water ~0.27% of global croplands extent) would translate into 15 management, by applying farming practices that reduce soil million barrel of oil equivalent per year (~0.04% of global evaporation, retaining more rainwater in the soil, growing more annual oil consumption) and increase GHG emissions by 6 water-saving crops, or applying deficit irrigation41 can reduce million tons of CO2 per year (or 0.04% of global total GHG NATURE COMMUNICATIONS | (2021)12:2319 | https://doi.org/10.1038/s41467-021-22581-7 | www.nature.com/naturecommunications 5 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22581-7 Fig. 5 Energy demand from irrigation over large-scale land acquisitions (LSLAs). a, b Average irrigation energy-intensity per region and per crop, respectively. c Aggregated irrigation energy requirements (million GJ yr−1) for the intended crop area to biofuels and food crop plantations. Irrigation scenarios consider one with sprinkler irrigation systems and one with surface irrigation systems. Table 1 Comparison of energy- and greenhouse gas (GHG)-intensities over large-scale land acquisitions (LSLAs). Activity Fossil-fuel-based energy GHG intensity (tons CO2- GHG emission over 197 considered Reference intensity (GJ ha−1) eq yr−1 ha−1) LSLAs (Mtons CO2-eq yr−1) Low-input 4.1 0.3 1.5 This study High-input 19.1 1.6 7.2 This study N2O emissions from high-input – 0.3 1.3 This study synthetic fertilizers Irrigation 3.2 0.3 2.1 This study Land use change – 14.1 – Ref. 28 Low-input and high-input show fossil-fuel-based energy use from labor, machinery, fertilizers, chemicals, fuels, and seeds. Land use change shows GHG intensity from deforestation and loss of soil carbon28. N2O emissions show GHG emission intensity from synthetic N fertilizer application in high-input farming. emissions from agriculture). While here we assessed agri- and have an important role in supporting food security of local cultural energy-related GHG emissions needed to close crop- populations50. Assuming that 360 million hectares of global small yield gaps, non-energy-related activities, such as land use farms will make a transition from low-input to high-input change, are estimated to have a GHG emission intensity of 14.1 farming, an additional 0.6 gigatons CO2-equivalent per year tons CO2-equivalent yr−1 ha−1 and therefore might be the would be emitted, or 4% of current global total GHG emissions main drivers of GHG emissions over LSLAs (Table 1). from agriculture31. The global agrarian transition is a broader global phenomenon The Haber-Bosch process to manufacture synthetic fertilizers is that far exceeds the footprint of LSLAs considered in this study. estimated to use ~1.2% of energy production and carbon emission Driven by economies of scale, the ongoing industrialization of worldwide51. Many of the major LSLAs will utilize ammonia agriculture is promoting a transition from smallholder to large- produced from the energy-intensive Haber-Bosch process. scale commercial farming in low- and middle-income countries. Moreover, after farmers apply synthetic fertilizers to crops, che- Globally, small farms (under 2 hectares) produce 30% of food mical reactions lead to the formation and emission of nitrous supply over 24% of agricultural areas (~360 million hectares)49 oxide, a potent GHG with 265 times more global warming 6 NATURE COMMUNICATIONS | (2021)12:2319 | https://doi.org/10.1038/s41467-021-22581-7 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22581-7 ARTICLE potential than carbon dioxide52,53. Over the considered LSLAs, large-scale agricultural investments or in specific land-deal deci- we estimate that synthetic fertilizer application in high-input sions. Specifically, LSLAs have impacts on the energy security of farming would release 1.3 million tons CO2-equivalent per year, local communities which have not been carefully addressed and with a GHG intensity of 0.3 tons CO2-equivalent yr−1 ha−1 which should be considered. Given the high energy demand of (Table 1). However, less carbon-intensive options to industrial fertilizer production and the associated GHG emissions, there is a fertilizers are available, such as using recycled organic matter, plausible case for more integrated monitoring of the changes in nutrients recovered from wastewater-treatment plants or manure, embedded energy resulting from crop choices and LSLAs. In a more efficient application of fertilizers53, or the deployment of order to evaluate future directions and options of land-use policy cover crops like soy and other legumes that fix atmospheric and rural development, the energy-budgeting approach should be nitrogen. Even though these less carbon-intensive agroecological integrated to give a more comprehensive account of the impli- alternatives to industrial fertilizers are available, they might be cations and tradeoffs, and inform decisions for longer term technically and economically difficult to adopt over large farms. sustainability. The production of fossil-fuel-free ammonia to supplement ferti- The speed with which LSLAs and agrarian transitions are lizers derived from organic materials54 is also an option to lower taking shape globally may have profound and lasting con- the carbon emissions in industrial fertilizer production55,56. sequences. While LSLAs might intensify agricultural production, In terms of the energy associated with irrigation, it is clear that achieve economies of scale, provide infrastructures, and increase LSLAs are often established on land where the closure of the yield agricultural productivity for specific crops, its impacts on socio- gap of specific crops for commercial agriculture would require environmental sustainability and the distribution of potential irrigation in addition to fertilizer applications. Except for the benefits between agricultural corporations and local farmers cases of land suitable for gravity-based irrigation, the withdrawal remains highly questionable16. There is consolidated knowledge and delivery of irrigation water from streams, aquifers, or other of the long-term unsustainability of energy-intensive agricultural water bodies to cultivated land typically requires substantial practices because of the massive reliance on fossil-fuel-based amounts of energy. While energy for fertilizer usage may be energy. Especially for non-industrialized countries, fossil-fuel- employed in supply chains distant from land acquisition, the fuel based agricultural transitions have been usually funded through needed for water delivery must be sourced from the local agrarian public subsidies, and historically suffered from high vulnerability economy, where it might compete with demand from other because of their sensitivity to energy crises and volatility of fossil- livelihood uses. fuel prices62,63. Lack of energy security is still a structural The LSLA component of the global agrarian transition has condition in large parts of the rural areas in low- and middle- clear implications on multiple dimensions and scales. Focusing income countries, where there is little and inconsistent access to on the energy side of transnational land investments provides affordable and reliable energy for agricultural production64. analytical insight into a series of interdependencies and Both fertilizer and irrigation energy usage should be evaluated tradeoffs that emerge from the integrated perspective of the for transnational and localized impacts, respectively, and LSLA energy–water–food nexus. For example, synergies and tradeoffs expansion needs to be placed under scrutiny following an between land use for food crops, water use, and energy infra- integrated food–energy–water approach that prioritizes issues of structure should be considered. The additional energy required by local resource access and justice. The energy variable in this a transition to high-input farming might have direct impact on conversation is therefore not only limited to issues of fossil fuel energy poverty and fairness in access to local energy resources by consumption and its implications for climate change, but it also marginalized rural communities, especially for those communities directly relates to issues of energy access and justice, the relying on the collection of local scarce natural resources for food–water–energy nexus, livelihood choices, property rights, and energy production57. These processes of agrarian transformation quality of life in the context of smallholder farmers. can be highly energy-intensive and should be evaluated to avoid further competition with scarce local energy resources. It is Methods fundamental to apply a nexus approach when assessing the sus- In this paper, we consider the energy investment of moving from a range of tainability and distribution of benefits of rural transitions, and to existing land-uses to the intended uses of LSLAs. Using estimates of annual energy input per area of crops from a range of sources (Fig. 3 and Supplementary Table 1), understand who are the winners and losers in such transforma- we compare the energy investment for some of the largest crop plantations being tions. For instance, under suitable environmental and social proposed for LSLAs based on the Land Matrix dataset36. We consider a sample of conditions, small-scale energy infrastructures such as renewable 197 land deals from the Land Matrix dataset36 and covering an area of about 4.07 mini-grid58 can be used by local farmers to access water through, million hectares across 39 countries (Fig. 1). The Land Matrix is a joint interna- tional initiative collecting data on transnational land acquisitions since 2000. This for instance, solar water pumps, thereby allowing for improved dataset reports only land investments with areas greater than 200 ha. Following energy security with limited land competition with food crops, Müller et al.14, we selected land deals: (1) having their status updated to “con- and at the same time reduce the use of fossil fuel for agricultural tracted”, “in start-up phase”, or “in production”; (2) intended for agricultural use; production and water use59. In this context, the development of (3) for which the coordinates of the deal location were reported in the Land Matrix small renewable-energy infrastructures at the community level for database. agricultural production, including energy infrastructure for irri- gation purposes, has been proven to be a viable solution to Energy usage of low- and high-input agriculture. To our knowledge, this is the first systematic evaluation of the energy requirements of agricultural intensification support the livelihood of smallholder farmers, mitigate climate through LSLAs. We use large-scale agricultural land-deals as a case study of the change, reduce energy poverty, and improve local energy access, energy implication of a new green revolution across LSLAs in low- and middle- with positive effects on social justice, such as better distribution of income countries. We estimated nitrogen application rates and yield gap-closure benefits and improved gender equality60. For example, small-scale levels before LSLAs using the seminal work of Mueller et al.35. We consider low- input versus high-input agricultural practices and use the seminal work from and community agrivoltaics—combining agriculture with solar Pimentel and Pimentel34, as well as our new large-scale estimates of energy con- photovoltaics—can increase crop productivity, while at the same sumption by irrigation to draw energy comparison estimates (Supplementary time generating renewable energy for local populations61. Materials). Energy inputs are at the farm level and consider fossil-fuel-based energy The energy footprint of LSLAs and their implication for cli- from labor, machinery, fertilizers, chemicals, fuels, and seeds. Irrigation water energy inputs were assessed considering irrigation water demand volume46 and mate change have not received sufficient attention in the scholarly energy requirements for pumping (Supplementary Materials). There are, of course, literature and in public assessments of their potential benefits and practical limitations to energy calculations in life cycle assessment methods65. For perils13,15, nor are they considered in national policies promoting oil palm and jatropha we also show the fossil-fuel-based energy required in the mill NATURE COMMUNICATIONS | (2021)12:2319 | https://doi.org/10.1038/s41467-021-22581-7 | www.nature.com/naturecommunications 7 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22581-7 to produce biodiesel (Fig. 3). We then calculate the aggregate energy investment for 19. Borras, S. M. Jr & Franco, J. C. Global land grabbing and trajectories of the crops proposed by current major LSLA investments globally (Figs. 4 and 5) in agrarian change: a preliminary analysis. J. Agrar. Change 12, 34–59 (2012). terms of their energy investment and present potential shifts that could be con- 20. Hall, R. et al. Resistance, acquiescence or incorporation? An introduction to sidered from low-input to high-input agricultural land-use practices. land grabbing and political reactions ‘from below’. J. Peasant Stud. 42, 467–488 (2015). Assessment of GHG emissions. GHG emissions from high-input LSLAs were 21. Wily, L. A. ‘The law is to blame’: the vulnerable status of common property assessed considering emissions from oil combustion and upstream emissions rights in sub‐Saharan Africa. Dev. Change 42, 733–757 (2011). during crude oil production. According to the US Environmental Protection 22. Oberlack, C., Tejada, L., Messerli, P., Rist, S. & Giger, M. Sustainable Agency, average emissions during oil combustion are 429.61 kg CO2 per 42-gallon livelihoods in the global land rush? Archetypes of livelihood vulnerability and barrel of oil66. Average global emissions from upstream crude oil production sustainability potentials. Glob. Environ. Change 41, 153–171 (2016). processes are 63.03 kg CO2 per 42-gallon barrel of oil emitted in upstream crude 23. Rulli, M. C., Saviori, A. & D’Odorico, P. Global land and water grabbing. Proc. oil67. Therefore, we assume a GHG emission intensity of oil equal to 492.6 kg CO2 Natl Acad. Sci. USA 110, 892–897 (2013). equivalent per barrel of oil. GHG emissions from N2O from synthetic nitrogen 24. D’odorico, P. & Rulli, M. C. The fourth food revolution. Nat. Geosci. 6, application in high-input farming were assessed considering the Food and Agri- 417–418 (2013). culture Organization emissions factor of 0.0132 kg N2O per kg synthetic N, and 25. Rulli, M. C. & D’Odorico, P. Food appropriation through large scale land assuming a global warming potential from N2O equal to 265. Nitrogen application acquisitions. Environ. Res. Lett. 9, 064030 (2014). rates (kg N per hectare) were taken from the corresponding data source used to 26. Mechiche-Alami, A., Yagoubi, J. & Nicholas, K. A. Agricultural land determine energy intensities of high-input farming (See Supplementary Table 1). acquisitions unlikely to address the food security needs of African countries. World Dev. 141, 105384 (2021). 27. Davis, K. F., Yu, K., Rulli, M. C., Pichdara, L. & D’Odorico, P. Accelerated Data availability deforestation driven by large-scale land acquisitions in Cambodia. Nat. Geosci. Data used to support findings of the study are available in the reference list and 8, 772–775 (2015). Supplementary Materials. 28. Liao, C. et al. Carbon emissions from the global land rush and potential mitigation. Nat. Food 2, 15–18 (2021). Received: 11 December 2020; Accepted: 19 March 2021; 29. Scheidel, A. & Sorman, A. H. Energy transitions and the global land rush: ultimate drivers and persistent consequences. Glob. Environ. Change 22, 588–595 (2012). 30. Schramski, J. R., Woodson, C. B. & Brown, J. H. Energy use and the sustainability of intensifying food production. Nat. Sustain. 3, 257–259 (2020). 31. Hong, C. et al. Global and regional drivers of land-use emissions in 1961–2017. Nature 589, 554–561 (2021). References 32. Crippa, M. et al. Food systems are responsible for a third of global 1. Pingali, P. L. Green revolution: impacts, limits, and the path ahead. Proc. Natl anthropogenic GHG emissions.Nat. Food 2, 198–209 (2021). Acad. Sci. USA 109, 12302–12308 (2012). 33. Rosenzweig, C. et al. Climate change responses benefit from a global food 2. Burney, J. A., Davis, S. J. & Lobell, D. B. Greenhouse gas mitigation by system approach. Nat. Food 1, 94–97 (2020). agricultural intensification. Proc. Natl Acad. Sci. USA 107, 12052–12057 34. Pimentel, D. & Pimentel, M. H. (eds.) Food, Energy, and Society (CRC Press, (2010). 2007). 3. Pellegrini, P. & Fernández, R. J. Crop intensification, land use, and on-farm 35. Mueller, N. D. et al. Closing yield gaps through nutrient and water energy-use efficiency during the worldwide spread of the green revolution. Proc. Natl Acad. Sci. USA 115, 2335–2340 (2018). management. Nature 490, 254–257 (2012). 36. Land Matrix. The Land Matrix Database (Land Matrix, accessed May 2019). 4. Hamant, O. Plant scientists can’t ignore Jevons paradox anymore. Nat. Plants http://landportal.info/landmatrix. Accessed April 2021. 6, 720–722 (2020). 37. FAO. Guidelines for Sustainable Large-Scale Land Deals in Africa (FAO, 2017) 5. Grafton, R. Q. et al. The paradox of irrigation efficiency. Science 361, 748–750 (2018). http://www.fao.org/3/i7548e/i7548e.pdf. 38. Carlson, K. M. et al. Carbon emissions from forest conversion by Kalimantan 6. Beltran-Peña, A., Rosa, L. & D’Odorico, P. Global food self-sufficiency in the oil palm plantations. Nat. Clim. Change 3, 283–287 (2013). 21st century under sustainable intensification of agriculture. Environ. Res. Lett. 39. Nolte, K., Chamberlain, W. and Giger, M., 2016. International Land Deals for 15, 095004 (2020). 7. Godfray, H. C. J. et al. Food security: the challenge of feeding 9 billion people. Agriculture: Fresh Insights from the Land Matrix: Analytical Report II. Bern: Bern Open Publ. https://doi.org/10.7892/boris.85304. Science 327, 812–818 (2010). 40. International Energy Agency. Global Primary Energy Demand by Fuel, 8. Springmann, M. et al. Options for keeping the food system within 1925–2019 (IEA, accessed February 2021). https://www.iea.org/data-and- environmental limits. Nature 562, 519–525 (2018). statistics/charts/global-primary-energy-demand-by-fuel-1925-2019. Accessed 9. von der Goltz, J. et al. Health impacts of the Green Revolution: evidence from April 2021. 600,000 births across the Developing World. J. Health Econ. 74, 102373 41. Rosa, L., Chiarelli, D. D., Rulli, M. C., Dell’Angelo, J. & D’Odorico, P. Global (2020). agricultural economic water scarcity. Sci. Adv. 6, eaaz6031 (2020). 10. Rasmussen, L. V. et al. Social-ecological outcomes of agricultural 42. Rosa, L. et al. Potential for sustainable irrigation expansion in a 3 °C warmer intensification. Nat. Sustain. 1, 275–282 (2018). 11. Borras, S. M. Jr, Hall, R., Scoones, I., White, B. & Wolford, W. Towards a climate. Proc. Natl Acad. Sci. USA 117, 29526–29534 (2020). 43. Rosa, L., Chiarelli, D. D., Tu, C., Rulli, M. C. & D’Odorico, P. Global better understanding of global land grabbing: an editorial introduction. J. unsustainable virtual water flows in agricultural trade. Environ. Res. Lett. 14, Peasant Stud. 38, 209–216 (2011). 114001 (2019). 12. Dell’Angelo, J., D’Odorico, P., Rulli, M. C. & Marchand, P. The tragedy of the grabbed commons: coercion and dispossession in the global land rush. World 44. Rosa, L. et al. Closing the yield gap while ensuring water sustainability. Environ. Res. Lett. 13, 104002 (2018). Dev. 92, 1–12 (2017). 45. Rothausen, S. G. & Conway, D. Greenhouse-gas emissions from energy use in 13. World Bank. Rising Global Interest in Farmland: Can it Yield Sustainable and the water sector. Nat. Clim. Change 1, 210–219 (2011). Equitable Benefits? (World Bank, 2010) http://www.donorplatform.org/ content/view/457/2687. 46. Chiarelli, D. D. et al. The green and blue crop water requirement WATNEEDS model and its global gridded outputs. Sci. Data 7, 1–9 (2020). 14. Müller, M.F. et al. Impact of transnational land acquisitions on local food 47. Jägermeyr, J. et al. Integrated crop water management might sustainably halve security and dietary diversity. Proc. Natl Acad. Sci. USA 118, e2020535118 the global food gap. Environ. Res. Lett. 11, 025002 (2016). (2021). 48. Graves, A., Rosa, L., Nouhou, A. M., Maina, F. & Adoum, D. Avert 15. Cotula, L. et al. Land Grab or Development Opportunity? Agricultural catastrophe now in Africa’s Sahel. Nature 575, 282–286 (2019). Investment and International Land Deals in Africa. IIED Report (IIED/FAO/ 49. Ricciardi, V., Ramankutty, N., Mehrabi, Z., Jarvis, L. & Chookolingo, B. How IFAD, 2009). http://www.ifad.org/pub/land/land_grab.pdf. much of the world’s food do smallholders produce? Glob. Food Secur. 17, 16. De Schutter, O. How not to think of land-grabbing: three critiques of large- 64–72 (2018). scale investments in farmland. J. Peasant Stud. 38, 249–279 (2011). 17. White, B., Borras, S. M. Jr, Hall, R., Scoones, I. & Wolford, W. The new 50. Samberg, L. H., Gerber, J. S., Ramankutty, N., Herrero, M. & West, P. C. Subnational distribution of average farm size and smallholder contributions to enclosures: critical perspectives on corporate land deals. J. Peasant Stud. 39, global food production. Environ. Res. Lett. 11, 124010 (2016). 619–647 (2012). 51. Smith, C., Hill, A. K. & Torrente-Murciano, L. Current and future role of 18. Dell’Angelo, J., D’Odorico, P. & Rulli, M. C. Threats to sustainable development posed by land and water grabbing.Curr. Opin. Environ. Sustain. Haber–Bosch ammonia in a carbon-free energy landscape. Energy Environ. Sci. 13, 331–344 (2020). 26, 120–128 (2017). 8 NATURE COMMUNICATIONS | (2021)12:2319 | https://doi.org/10.1038/s41467-021-22581-7 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22581-7 ARTICLE 52. Galloway, J. N. et al. The nitrogen cascade. Bioscience 53, 341–356 (2003). received from the National Science Foundation DBI-1639145. J.D.A., M.C.R., and P.D.O. 53. Gerber, J. S. et al. Spatially explicit estimates of N2O emissions from croplands acknowledge support from the European Union’s Horizon 2020 research and innovation suggest climate mitigation opportunities from improved fertilizer programme under the Marie Skłodowska-Curie Action (MSCA) Innovative Training management. Glob. Change Biol. 22, 3383–3394 (2016). Network (ITN) grant agreement No. 861509 - NEWAVE. A.S. acknowledges funding 54. Chen, J.G. et al. Beyond fossil fuel-driven nitrogen transformations.Science from a Beatriu de Pinós grant, Government of Catalonia’s Secretariat for Universities and 360, eaar6611 (2018). Research of the Ministry of Economy and Knowledge (2017 BP 00023). 55. Gabrielli, P., Gazzani, M. & Mazzotti, M. The role of carbon capture and utilization, carbon capture and storage, and biomass to enable a net-zero-CO2 emissions chemical industry. Ind. Eng. Chem. Res. 59, 7033–7045 (2020). Author contributions L.R., M.C.R., S.A., D.D.C., J.D.’A., N.D.M., A.S., G.S., and P.D.’O. conceived, designed, 56. Davis, S.J. et al. Net-zero emissions energy systems.Science 360, eaas9793 and wrote the study; L.R., M.C.R., S.A., D.D.C., and N.M. performed the analyses; L.R. (2018). and S.A. analyzed the data. 57. Guta, D. D. Effect of fuelwood scarcity and socio-economic factors on household bio-based energy use and energy substitution in rural Ethiopia. Energy Policy 75, 217–227 (2014). Competing interests 58. Schmitt, R.J.P., Kittner, N., Kondolf, G.M. & Kammen, D.M. Deploy diverse The authors declare no competing interests. renewables to save tropical rivers.Nature 569, 330–332 (2019). 59. IFAD. Renewable Energy for Smallholder Agriculture (RESA) (IFAD, 2020). https://www.ifad.org/documents/38714170/41937394/resa.pdf/715e1a75-35df- Additional information bafc-f491-7effde867517. Supplementary information The online version contains supplementary material 60. Burney, J., Woltering, L., Burke, M., Naylor, R. & Pasternak, D. Solar-powered available at https://doi.org/10.1038/s41467-021-22581-7. drip irrigation enhances food security in the Sudano–Sahel. Proc. Natl Acad. Sci. USA 107, 1848–1853 (2010). Correspondence and requests for materials should be addressed to S.A. 61. Barron-Gafford, G. A. et al. Agrivoltaics provide mutual benefits across the Peer review information Nature Communications thanks Michael Clark and other food–energy–water nexus in drylands. Nat. Sustain. 2, 848–855 (2019). anonymous reviewers for their contributions to the peer review of this work. Peer review 62. Pimentel, D. et al. Food production and the energy crisis. Science 182, 443–449 reports are available. (1973). 63. Giampietro, M., Mayumi, K., & Sorman, A. H. Energy Analysis for a Reprints and permission information is available at http://www.nature.com/reprints Sustainable Future. Multi-Scale Integrated Analysis of Societal and Ecosystem Metabolism (Routledge, 2013). 64. IEA. Africa Energy Outlook 2019. World Energy Outlook special report (IEA, Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in 2019). published maps and institutional affiliations. 65. Zhang, Y. & Colosi, L. M. Practical ambiguities during calculation of energy ratios and their impacts on life cycle assessment calculations. Energy Policy 57, 630–633 (2013). Open Access This article is licensed under a Creative Commons 66. United States Environmental Protection Agency. Greenhouse Gases Equivalencies Attribution 4.0 International License, which permits use, sharing, Calculator - Calculations and References (USEPA, accessed November 2020). adaptation, distribution and reproduction in any medium or format, as long as you give https://www.epa.gov/energy/greenhouse-gases-equivalencies-calculator- appropriate credit to the original author(s) and the source, provide a link to the Creative calculations-and-references#:~:text=The%20average%20carbon%20dioxide% Commons license, and indicate if changes were made. The images or other third party 20coefficient,gallon%20barrel%20 (EPA%202018). Accessed April 2021. material in this article are included in the article’s Creative Commons license, unless 67. Masnadi, M. S. et al. Global carbon intensity of crude oil production. Science indicated otherwise in a credit line to the material. If material is not included in the 361, 851–853 (2018). article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/ Acknowledgements licenses/by/4.0/. Publication was made possible in part by support from the Berkeley Research Impact Initative (BRII), sponsored by the UC Berkeley Library. This work was partially sup- ported by the National Socio-Environmental Synthesis Center (SESYNC) under funding © The Author(s) 2021 NATURE COMMUNICATIONS | (2021)12:2319 | https://doi.org/10.1038/s41467-021-22581-7 | www.nature.com/naturecommunications 9