The objective of the paper is to evaluate the long-term prospects of sustainable productivity growth linked to plausible assumptions on public agricultural R&D investments as the key productivity driver. Second, it investigates the role of changing R&D focus from yield maximization to input saving technologies (fertilizers and pesticides). The projections using CGE model MAGNET identify China, India and Brazil as regions with high productivity growth from agricultural R&D while Sub-Saharan Africa regions will struggle with low productivity growth rates and substantial increase in GHG emissions. Overall, GHG emissions are projected to grow more than agricultural output. However it is found that sustainable agricultural productivity growth could become feasible under the reorientation of R&D programs in high income countries (with use of chemical inputs declining as much as 30%) where these policies can make an important contribution to sustainability while food security concerns are limited and spillover effects in terms of higher food prices are low.

The global poor are expected to suffer most from the impact of climate change, in particular the increasing frequency of extreme weather events. To develop targeted climate adaptation strategies, national decision makers need to have detailed information on the quantity, location and profile of the people that are most vulnerable to climate hazards. This study presents an innovative spatial microsimulation modelling framework for projecting subnational income distribution and poverty trends under different scenarios that can be combined with spatial data on climate hazards to support climate risk assessments. The model combines household survey data with subnational projections on key drivers of income to simulate how the distribution of income changes as a consequence of economic development and structural transformation. To illustrate our modelling framework, we provide an application to Ethiopia. We projected changes in poverty headcount and income distribution for 60 different zones and three different socio-economic scenarios for the period 2020–2050. We combined the subnational income projections with heat stress maps to identify the number and profile of the population that are most vulnerable to climate change and found that, depending on the scenario, between 1.4 and 9.4 million poor people (1-5% of the population) will be at risk of heat stress in Ethiopia in 2050. The modelling framework can be combined with spatial data of additional climate hazards, such as floods and droughts, and be applied to other countries and regions, to support national climate information systems and inform climate adaptation strategies and policies.

Detailed spatial information on food consumption and diet quality is essential to develop effective policies for a successful transition towards sustainable and healthy diets. However, fine-scale maps of food consumption patterns are not frequently available. To address this, we used a machine learning framework in combination with household survey information and spatial variables to generate high-resolution maps (30 arcsec; ~1 km) for 25 different food groups. The models explain more than 50% of the variance in food consumption of ten distinct food groups, which account for 71.7% of total energy consumption (in kcal). We applied this approach to calculate an indicator showing the distance to the EAT-Lancet diet, the global reference for a healthy diet. This resulted in insightful high-resolution diet quality maps. Altogether, we demonstrate that machine learning models combined with household survey give insight into the complex interplay between food environment-related and individual-based motivational factors influence food consumption.

The global poor are expected to suffer most from the impact of climate change, in particular the increasing frequency of extreme weather events, such as floods, heat waves and fores fires as well as changes in food supply. These events are very uncertain as they depend on different climate scenarios and are highly local, only affecting a certain part of the population. To develop targeted climate adaptation strategies national decision makers need to have detailed information on the quantity, location and profile of the people that are most vulnerable to climate impacts. This study presents an innovative spatial microsimulation modelling framework for projecting subnational income distribution and poverty trends under different scenarios that can be combined with spatial data sets on extreme climate events to support climate risk assessments. The Microsimulation of Income DynamicS (MIDS) model, combines household survey data with subnational projections on key drivers of income, including demographic change, urbanization and shifts in skills and occupation, to simulate how the distribution of income changes as a consequence of economic development and structural transformation. To account for impact of global linkages and shocks, MIDS incorporates labor income projections from a global general computable equilibrium model. To illustrate the model we provide an application to Ethiopia, one of the largest countries in Africa in terms of population size, which is characterized by high poverty levels and deep inequality. We used MIDS to project changes in income distribution and the poverty headcount for 60 different zones (administrative level 2 regions) and three different Shared Socio-economic Pathways (SSP) scenarios for the period 2018-2050. We extended the SSPs with additional assumptions on regional development to make them suitable for subnational assessments. We combined the subnational income projections with heat stress maps to identify the share of the population that is most vulnerable to climate change. We found that, depending on the scenario, between 1.4 and 9.4 million people will be at risk of heat stress in 2050.

This study proposes a monitoring framework for European agriculture within the Safe and Just Operating Space (SJOS) concept, integrating environmental thresholds, socio-economic considerations, and the objectives outlined in European Union (EU) policy frameworks. A thorough review of EU policies, including the European Green Deal (EGD), the Common Agricultural Policy (CAP), and related strategies, informed the adaptation of global SJOS concepts defined for the Planet and Humanity to the specific requirements of the EU agricultural sector. This downscaling to the regional contexts of the EU and the disaggregation to agricultural processes was central to ensuring the framework’s relevance and precision. The methodology identified over 30 indicator domains across 12 thematic areas, aligning global SJOS principles with EU policy targets. These indicators were refined through collaboration with modelling teams and external consultations, ensuring feasibility for use in existing simulation models. Workshops with experts and stakeholders revealed critical gaps, such as in biodiversity and animal welfare, highlighting areas for further model development. The framework enables projections of EU agriculture’s status relative to SJOS across various future scenarios, helping to analyse interdependencies, trade-offs, and synergies among objectives and indicators. The study emphasizes the importance of aligning scientific benchmarks with EU policy targets, including CAP and EGD objectives, while addressing challenges in data integration and interdisciplinary modelling.

The transition to healthier diets might be accompanied by trade-offs that occur in other parts of the food system. In this study the trade-offs between socio-economic, environmental, and health indicators were analyzed in different dietary scenarios for Bangladesh between 2022 and 2050. We used a global economic simulation model with updated national food consumption data, extended with a footprint module to track environmental impacts through the food value chain in Bangladesh and its trading partners. This study compares a business-as-usual (BAU) diet with the EAT-Lancet diet and the Bangladesh food-based dietary guidelines (FBDGs). The BAU diet has a higher intake of animal products and sugar, and a lower intake of vegetables, fruits, legumes, and nuts than the EAT-Lancet and FBDG diets. We found that promoting a diet with more plant-based proteins has a strong positive impact on dietary health and an overall positive impact on the environment compared to the BAU scenario. This is due to the reduced impact of animal protein production on greenhouse gas emissions and the reduced impact of rice production on water use and nitrogen application. In addition, the transition to sustainable and healthy diets had minor impacts on the wages of low-skilled workers, Bangladesh's self-sufficiency, and the affordability of food and cereals. In particular, the FDBG diet scenario scored best on diet and cereal affordability, as well as freshwater use compared to the other two scenarios. The decrease in the self-sufficiency ratio was comparable to the BAU diet scenario and smaller compared to the EAT-Lancet diet.

This paper proposes a novel method for mapping livestock distribution in Africa using the Living Standards Measurement Study-Integrated Surveys on Agriculture (LSMS-ISA). Using a Bayesian spatial statistical model, we produce maps of livestock distribution at a resolution of 1 km2. Our case study in Malawi, covering 2010 and 2019, demonstrates the effectiveness of the method in mapping five livestock species. We compare our results with the Gridded Livestock of the World (GLW) database and use the maps to assess livestock vulnerability to climate-related flood risks under different climate scenarios. This approach provides a rapid, data-rich tool for policy makers to assess climate risks to livestock, which is critical for sustainable agricultural development and environmental management in data-poor regions.

The transition to healthier diets might be accompanied by trade-offs that occur in other parts of the food system. In this study the trade-offs between socio-economic, environmental, and health indicators were analyzed in different dietary scenarios for Bangladesh between 2022 and 2050. We used a global economic simulation model with updated national food consumption data, extended with a footprint module to track environmental impacts through the food value chain in Bangladesh and its trading partners. The study compares a business-as-usual (BAU) diet with the EAT-Lancet diet and the Bangladesh food-based dietary guidelines (FBDG). The BAU diet has a higher intake of animal products and sugar, and a lower intake of vegetables, fruits, legumes and nuts than the EAT-Lancet and FBDG diets. It was found that promoting a diet with more plant-based proteins would have a strong positive impact on dietary health and an overall positive impact on the environment compared to the BAU scenario, due to the reduced impact of animal protein production on greenhouse gas emissions and the reduced impact of rice production on water use and nitrogen application. In addition, the transition to sustainable and healthy diets had minor impacts on the wages of low-skilled workers, Bangladesh’s self-sufficiency, and the affordability of food and cereals. In particular, the FDBG diet scenario scored best on self-sufficiency and cereal affordability compared to the other two scenarios, and the increase in low-skilled wages was comparable to the BAU diet scenario.

Worldwide hundreds of millions of people suffer from water, food and energy insecurity in transboundary river basins, such as the Zambezi River Basin. The interconnected nature of nexus is often not recognized in investment planning and many regional policymakers lack adequate tools to tackle it. Future growing demands and climate change add an additional challenge. In this study, we combine policy relevant co-developed stakeholder scenarios and integrated nexus modeling tools to identify key solutions to achieve sustainable development in the Zambezi. Results show that siloed development without coordination achieves the least economic and social benefits in the long term. Prioritizing economic benefits by maximizing the use of available natural resources results in the expansion of irrigated areas by more than a million hectares and increase in hydropower production by 22,000 GWh/year in the coming decades, bringing significant economic benefits, up to $12.7 billion per year, but causes local water scarcity and negative impacts on the environment. Combining environmental protection policies with sustainable investments of $7.2 billion per year (e.g. groundwater pumping and wastewater treatment and reuse, irrigation efficiency improvements, and farmer support aimed to improve food security and productivity) results in significantly higher social benefits with economic benefits that still reach $11.7 billion per year.

Detailed data on the location of crops is essential to inform national food and agricultural policies. A key source of information on the spatial distribution of crops are the global datasets produced with the Spatial Production Allocation Model (SPAM). SPAM uses an optimization approach to allocate national and subnational crop statistics for four production systems, informed by spatial information on both biophysical (e.g. suitability and potential yield) and socio-economic (e.g. market access and population density) drivers of crop location. The SPAM crop distribution maps are produced at a resolution of 5 arc minutes, which is often too coarse for detailed country and subnational assessments, which require higher resolution products and the flexibility to subsume more detailed information from national sources. The aim of this paper is to demonstrate an extended and improved version of SPAM, which is able to (a) produce maps at a higher resolution than the current existing global maps; (b) incorporate additional and more detailed information on the location of crops (i.e. from OpenStreetMap); and (c) create linked maps that can be compared over time. The model is applied to seven countries in Southern African region. Detailed results are presented for maize and irrigated wheat production in Southern Africa. Maize is widespread in Mozambique, Malawi and Zimbabwe, Angola and Zambia, while very little maize is grown in Botswana and Namibia. Wheat is predominantly located in the Central and Southern provinces of Zambia, which overlaps with the location of commercial farm blocks, and is widespread in Zimbabwe. A validation using information on the location of a large number of crops from household surveys for Malawi and Zambia showed 75%-100% true positives for most crops. The model is performing less well for a few marginal crops that are only grown in specific regions and several crops that are observed throughout the country. The crop distribution maps can be used to support regional crop monitoring systems, guide investment decisions and inform national assessments of food security under socio-economic and climate change.

We examine spillovers from agricultural estates to Malawian smallholders within an econometric counterfactual framework. We consider economic spillovers such as income, as well as agrarian spillovers such as yields, harvests, and crop diversity. We identify long-run effects of large agricultural investments on small-scale farmers. For the location of large estates, we use a novel OpenStreetMap dataset, while data on smallholder’s stems from a household survey. We provide evidence for the importance of the distance threshold for spillovers, and explore multiple thresholds. In proximity to estates we find higher groundnut and pigeon pea yields and increased crop diversity. In very close proximity, incomes are also higher. Area under cultivation in total and for maize are smaller for nearby households, while maize yields are not significantly different. Overall, our results suggest that policies should aim to leverage the increased crop diversity and groundnut yields while mitigating potential detrimental effects arising from reduced cultivated land.

In China, irrigation plays a fundamental role in food production, which is hampered by water security, rising population and climate change. To ensure food security and formulate agricultural and irrigation policies, decision-makers need detailed grid-level information on the location of irrigated areas. Unfortunately, this information is not easily available as national irrigation maps are frequently outdated and often have a coarse spatial resolution. In this paper, we present new irrigation maps for China at a spatial resolution of 30 arc seconds (~1 × 1 km) that cover the period between 2005 and 2015. The maps were created using a synergy approach, which combines and integrates regional and global irrigation maps, cropland maps and subnational statistics. The maps were calibrated to subnational irrigation statistics and validated using an external dataset with geocoded information on the location of irrigated areas. The maps show, at the detailed spatial level, how much irrigation expanded over the period of 2005–2015. The proposed synergy approach is flexible and can easily be applied to create irrigation maps for other regions.

Abstract Spatial information on the location of crops is key to inform agricultural policies, and are an important input for global and national land use change models. The global crop distribution maps produced with the Spatial Production Allocation Model (SPAM) are widely used by researchers, policy makers and business for this purpose. SPAM uses a downscaling approach to allocate national and subnational crop statistics to a 5 arc minutes grid, informed and constrained by spatial information on biophysical and socio-economic drivers. This study introduces the R mapspamc package that allows users to create crop distribution maps for single countries using the SPAM cross-entropy crop allocation algorithm as well as an alternative approach to create maps at higher spatial resolution. It presents a six-step approach and a detailed example for Malawi to illustrate how the package can be used and what type of outcomes can be produced.

Detailed and accurate labor statistics are fundamental to support social policies that aim to improve the match between labor supply and demand, and support the creation of jobs. Despite overwhelming evidence that labor activities are distributed unevenly across space, detailed statistics on the geographical distribution of labor and work are not readily available. To fill this gap, we demonstrated an approach to create fine-scale gridded occupation maps by means of downscaling district-level labor statistics, informed by remote sensing and other spatial information. We applied a super-learner algorithm that combined the results of different machine learning models to predict the shares of six major occupation categories and the labor force participation rate at a resolution of 30 arc seconds (~1x1 km) in Vietnam. The results were subsequently combined with gridded information on the working-age population to produce maps of the number of workers per occupation. The super learners outperformed (n = 6) or had similar (n = 1) accuracy in comparison to best-performing single machine learning algorithms. A comparison with an independent high-resolution wealth index showed that the shares of the four low-skilled occupation categories (91% of the labor force), were able to explain between 28% and 43% of the spatial variation in wealth in Vietnam, pointing at a strong spatial relationship between work, income and wealth. The proposed approach can also be applied to produce maps of other (labor) statistics, which are only available at aggregated levels.

We review consumer-side interventions and their effectiveness to support a transition to healthier and more environmentally sustainable diets and identify taxes/subsidies as relevant instruments. To quantify the scope of necessary tax levels to achieve dietary recommendations on EU average, we apply three established economic models. Our business-as-usual food intake projections stress the need for policy intervention to resolve continued divergence from nutrition guidelines. Our findings suggest that food group specific taxes are effective in reaching nutrition and environmental sustainability targets. However, considerable tax levels are required to achieve the targeted consumption shifts, inducing a discussion about alternative policy designs and current model limitations. A coherent policy package is suggested to approach nutrition and sustainability objectives simultaneously.

Ending hunger and achieving food security - one of the UN sustainable development goals - is a major global challenge. To inform the policy debate, quantified global scenarios and projections are used to assess long-term future global food security under a range of socio-economic and climate change scenarios. However, due to differences in model design and scenario assumptions, there is uncertainty about the range of food security projections and outcomes. We conducted a systematic literature review and meta-analysis to assess the range of future global food security projections to 2050. We reviewed 57 global food security projection and quantitative scenario studies that have been published over the last two decades and discussed the methodology, underlying drivers, indicators and projections. We harvested quantitative information from 26 studies to compare future trends of the two most used global food security indicators: per capita food demand (593 projections) and population at risk of hunger (358 projections). We found that across five representative scenarios that span divergent but plausible socio-economic futures total global food demand is expected to increase by +35% to +56% between 2010 and 2050, while population at risk of hunger is expected to change by -91% to +8% over the same period. If climate change is taken into account the range changes slightly (+30% to +62% for total food demand and -91% to +30% for population at risk of hunger) but overall we do not find statistical support for differences in projections with and without climate change. Finally, our review suggests that current modeling approaches can be improved by better incorporating several options that have been proposed to tackle global food security, in particular aquaculture and ‘future foods’, and expand the number of indicators to better cover the multiple dimensions of food security. The results of our review can be used to benchmark new global food security projections and quantitative scenario studies and inform policy analysis and the public debate on the future of food.

Rice is one of the staple food crops and is a profitable smallholder cash crop in Zambia. It has the potential to contribute significantly to increased incomes and employment among rural producers. However, rice is the only staple crop in the country for which domestic production does not meet or exceed domestic demand. Low productivity is one of the factors that contribute to this. One necessary step towards addressing this problem is the identification of land with greatest potential for rice production, as well as the identification of land-based limitations which might be overcome by improved management. The aim of this study was to develop a land suitability index for rainfed paddy rice production reflecting expert opinion and published studies based on climatic, topographic and soil properties. Land suitability was evaluated using a method which accounts for important multiple factors, and which considers their joint effect in terms of a hierarchical model of constraints. The suitability classes were ranked according to the FAO land suitability classification as: Highly Suitable (S1), Moderately Suitable (S2), Marginally Suitable (S3), Currently Not Suitable (N2), and Permanently Not Suitable (N1). Results showed that there is limited potential for rainfed paddy rice production in Zambia with <20% of the land classified as either highly or moderately suitable. Therefore, the potential of irrigated and upland rice production in Zambia needs to be assessed as this would help expand the potential production area of rice.

Satisfying China’s food demand without harming the environment is one of the greatest sustainability challenges for the coming decades. Here we provide a comprehensive forward-looking assessment of the environmental impacts of China’s growing demand on the country itself and on its trading partners. We find that the increasing food demand, especially for livestock products (~16%–30% across all scenarios), would domestically require ~3–12 Mha of additional pasture between 2020 and 2050, resulting in ~−2% to +16% growth in agricultural greenhouse gas (GHG) emissions. The projected ~15%–24% reliance on agricultural imports in 2050 would result in ~90–175 Mha of agricultural land area and ~88–226 MtCO2-equivalent yr−1of GHG emissions virtually imported to China, which account for ~26%–46% and ~13%–32% of China’s global environmental impacts, respectively. The distribution of the environmental impacts between China and the rest of the world would substantially depend on development of trade openness. Thus, to limit the negative environmental impacts of its growing food consumption, besides domestic policies, China needs to also take responsibility in the development of sustainable international trade. Meeting China’s growing demand for food, especially for livestock products, will have huge environmental impacts domestically and globally. This study finds large increases in land, water, fertilizer and greenhouse gas emissions that vary based on openness of trade.

Quantified global scenarios and projections are used to assess long-term future global food security under a range of socio-economic and climate change scenarios. Here, we conducted a systematic literature review and meta-analysis to assess the range of future global food security projections to 2050. We reviewed 57 global food security projection and quantitative scenario studies that have been published in the past two decades and discussed the methods, underlying drivers, indicators and projections. Across five representative scenarios that span divergent but plausible socio-economic futures, the total global food demand is expected to increase by 35% to 56% between 2010 and 2050, while population at risk of hunger is expected to change by -91% to +8% over the same period. If climate change is taken into account, the ranges change slightly (+30% to +62% for total food demand and -91% to +30% for population at risk of hunger) but with no statistical differences overall. The results of our review can be used to benchmark new global food security projections and quantitative scenario studies and inform policy analysis and the public debate on the future of food.

Achieving climate neutrality in the European Union (EU) by 2050 will require substantial efforts across all economic sectors, including agriculture. At the same time, an ambitious unilateral EU agricultural mitigation policy is likely to have adverse effects on the sector and may have limited efficiency at global scale due to emission leakage to non-EU regions. To analyse the competitiveness of the EU's agricultural sector and potential non-CO2 emission leakage conditional on mitigation efforts outside the EU, we apply three economic agricultural sector models. We find that an ambitious unilateral EU mitigation policy in line with efforts needed to achieve the 1.5 °C target globally strongly affects EU ruminant production and trade balance. However, since EU farmers rank among the most greenhouse gas efficient producers worldwide, if the rest of the world were to start pursuing agricultural mitigation efforts too, economic impacts of an ambitious domestic mitigation policy get buffered and EU livestock producers could even start to benefit from a globally coordinated mitigation policy.