The dietary footprints and transition priorities of over 12,000 cities

The dietary footprints and transition priorities of over 12,000 cities | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article The dietary footprints and transition priorities of over 12,000 cities Hongyi Cai, Shruti Jain, Daniel Moran, Oliver Taherzadeh This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8830739/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Urban consumption plays a central role in driving global carbon emissions and is associated with a wide range of environmental pressures, including deforestation, water use, and biodiversity loss. As cities and networks of cities work to define and achieve quantitative sustainability goals, the lack of insight into their unique food footprints, shaped by each city’s specific dietary patterns and demographic composition, remains a significant obstacle. Here we combine demographic, group-specific diet, and supply-chain environmental footprint data to estimate the dietary footprints in 12,683 cities across 159 countries. We show that diet-related greenhouse gas emissions (GHGE) are highly concentrated in a small number of megacities. The 200 cities with the largest footprints account for approximately 15% of global population but nearly half of total global urban dietary emissions. Adoption of the EAT–Lancet 2.0 diet in the studied cities could modestly reduce total footprints (by 5–15%, depending on impact category). High-footprint cities have the greatest potential reductions from reducing meat and dairy consumption. However, these gains are largely matched by the potential rises in footprints from lower-income cities presently consuming below the EAT-Lancet recommended levels. Across transnational city networks (C40 Cities, Eurocities, and Milan Urban Food Policy Pact members) implementation of the EAT–Lancet 2.0 diet delivers substantial reductions 22–50% in GHGE, underscoring the policy relevance of city-level dietary footprinting. Our results highlight the need for context-specific urban dietary strategies. The resulting database, MATILDA-City, provides a globally harmonised evidence base to benchmark, rank, and monitor city-level dietary footprints, supporting municipalities to coordinate actions toward sustainable food systems. dietary footprints sustainable food systems EAT–Lancet diet 2.0 urban sustainability Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction Cities play a critical role in achieving ambitious sustainability goals, as key drivers of global resource consumption and emissions ( 1 ). Infrastructure decisions are critical to curtailing these impacts in the buildings and transport sectors ( 2 , 3 ). Yet, dietary change is also a key pillar to urban sustainability. Unlike the lock-in of infrastructure, dietary patterns and their impacts can be influenced by short-term, large-scale policy interventions. Several transnational city networks have been established to this end. The C40 Good Food Cities initiative, 100 climate neutral cities mission of the EU, and the Milan Urban Food Policy Pact (signed by 330 cities) were set up to align and even surpass local food strategies with national and international sustainability goals. These include near-term goals for wide-scale adoption of a planetary health diet (e.g. Barcelona, New York, Milan), food waste reduction targets, and fiscal instruments to limit sugar-sweetened beverages. Despite these commitments, progress on these action plans remains hampered by data to benchmark, compare and rank city-level food consumption, impacts, strategies and goals ( 4 , 5 ). Data collection and modelling efforts designed to guide sustainable development routinely focus on consumption patterns and impacts at a national scale, overlooking the uneven contribution of urban regions to anthropogenic environmental change. Where cities are the focus of concern, models typically support an aggregated or partial assessment of consumption activities. Early studies captured direct, production-based emissions of cities ( 6 , 7 ). Although instructive, these approaches ignore the global impacts of urban consumption that extend beyond city borders and can account for around 50% of emissions-related impacts ( 8 ). Global, consumption-based footprinting analyses of cities have emerged in recent years in response to the growing need for cities to report direct and indirect emissions ( 9 ). These models have sought to use economic data to decompose national expenditure patterns by region, using household expenditure surveys ( 10 ) or spending data ( 11 ), to better capture supply-chain related environmental impacts of city consumption ( 12 , 13 ). However, these are limited by coarse food categorisation and poorly reflect actual physical food intake, preventing actionable hotspotting of food consumption patterns and impacts across urban regions. Bottom-up data collection on urban food consumption patterns offers another avenue for city-level dietary environmental footprinting. Such an approach relies on localised dietary recall surveys and urban assessment of food provisioning ( 14 , 15 ). Whilst this provides superior detail of urban food habits, such data collection is not amenable to global comparison and remains impractical at a global scale due to time and resource constraints. Many cities lack funding to establish a comprehensive assessment of city-level consumption and monitoring against a wide suite of indicators ( 16 ). Among the G20, only around 40% of cities have adequate data to track climate pledges among, an issue that is more pronounced in low-income countries ( 17 ). Although national food intake surveys collect information on sub-regions, such information is either aggregated for privacy reasons or too incomplete to enable robust assessment of city-specific consumption patterns ( 18 ). Yet, by coupling information on population size and structure, food consumption patterns and impacts of cities can be derived, based on nationally-derived food consumption profiles of socio-demographic subgroups. We follow such an approach, to construct a global database for comparable, city-level dietary footprint estimates: MATILDA-City, building on MATILDA (Micro-Macro Assessment Tool to Identify Low-impact Dietary Actions), a global, consumer-level dietary footprinting tool ( 19 ). The harmonisation of food consumption microdata, environmental impact information and demographic data provides a basis to estimate city-level footprints ( 20 ). Within this study, we couple MATILDA ( 19 ) with spatially-derived city population and economic data to estimate the food consumption patterns and environmental footprint of 12,683 cities by greenhouse gas emission (GHGE), water, and land footprint (see Materials and Methods). The resultant database, MATILDA-City, enables comparison, ranking and scenario modeling of dietary environmental footprints across different cities and can flexibly accommodate new city consumption or food supply data. We illustrate the usefulness of this tool by calculating the environmental gains of adopting the EAT-Lancet 2.0 diet across each city and broad transnational city networks, including C40 cities, EU climate cities and the Milan Urban Food Policy Pact (MUFPP). Our analysis shows a highly uneven distribution of city dietary environmental footprints worldwide. The 200 largest cities account for nearly half of the total 5.16 Gt CO₂-eq of global city dietary emissions, while megacities such as Jakarta, São Paulo, Guangzhou, and Tokyo, each exceed 70 Mt CO₂-eq. Across cities, animal-source foods dominate environmental pressures, with meat alone contributing 49% of dietary GHGE, 38% of water use, and 54% of land use. If the EAT-Lancet 2.0 diet were fully adopted, global city dietary greenhouse gas emissions would decrease by 14%, land use by 16%, and water use by 5%. These aggregate reductions arise because the EAT–Lancet diet leads to heterogeneous changes in diet-related environmental footprints across cities, with reductions in high-footprint cities but increases in other cities. Overall, 56% of cities experience increases in at least one environmental footprint indicator, resulting in a modest global net reduction. Within transnational city networks, adoption of the EAT-Lancet 2.0 diet yields net reductions in dietary environmental footprints, amounting to 22% in C40 Cities, 50% in Eurocities, and 30% across the Milan Urban Food Policy Pact members. The underlying database, MATILDA-City, provides a scalable database to support city-level decision support towards sustainable and equitable urban food systems. Results Our results indicate that urban diet-related greenhouse gas emissions (GHGE) are driven by a limited number of megacities. Globally, city-level diet-related GHG emissions total 5.16 Gt CO₂-eq per year (Fig. 1 A). Asian cities contribute to nearly half of global dietary GHGE (51.4%), followed by South America (16.6%), Africa (11.9%), and Europe (10.7%) (Fig. 1 B). As expected, total city-dietary GHGE show a strong positive correlation with total population size (R² = 0.810), indicating that larger cities generally have higher absolute emissions (Fig. 2 A). The 200 most populous cities collectively account for 39.8% of the total urban population and 45.5% of the total GHGE across all cities. Collectively, the dietary GHGE from the 200 cities amount to 2.35 Gt CO₂-eq, comparable to diet-related emissions of Brazil (2.05 Gt CO₂-eq) ( 21 ). Among the 12,683 cities analysed, only four megacities exceed 70 Mt CO₂-eq - Jakarta (84.4 Mt CO₂-eq), São Paulo (84.1 Mt CO₂-eq), Guangzhou (81.9 Mt CO₂-eq), and Tokyo (73.5 Mt CO₂-eq), each comparable to the diet-related emissions of a small- to medium-sized country such as the Philippines (81.7 Mt CO₂-eq) or South Sudan (71.8 Mt CO₂-eq) ( 21 ). Cities with populations below 10 million contribute less than 5% of their national dietary GHGE, whereas several cities account for substantial shares of their respective national dietary GHGE footprints, such as Lima (9.9 million people, 45.9 Mt CO₂-eq, 59.5% of national total) and Moscow (15.6 million people, 45.9 Mt CO₂-eq, 23.8%). Regarding population distribution, most cities (12,234; 96.4%) represent less than 5% of their national population, whereas only 255 cities (2.0%) account for more than 10%. City-level diet-related GHGE are unevenly distributed across food groups and cities. Meat (processed meats & unprocessed meats), dairy products, and refined grains account for the largest shares of city-level GHGE (Supplementary Fig. 1A). In high-emission cities such as Jakarta, São Paulo, and Guangzhou, meat alone accounts for over one-third of total dietary GHGE, while refined grains and dairy products together account for approximately one-quarter. In contrast, fruits, vegetables, and legumes contribute less than 10% on average to city-level GHGE and other impacts. Per-capita consumption patterns across cities (Supplementary Fig. 1B) show marked heterogeneity. Refined grains dominate diets in most cities (> 400 g/day), especially across Africa and Asia, whereas dairy intake varies widely, from low levels in many African and Asian cities (about 120 g/day) to very high levels in Europe, North America, and Oceania (> 900 g/day). Meat consumption is particularly high in several European and North American cities. Cities with comparable population sizes exhibit markedly different dietary compositions and associated diet-related GHGE. For example, despite comparable total populations, Lima (9.9 million) and Paris (10.1 million) exhibit distinct dietary patterns and associated environmental impacts. In Lima, plant-based foods account for 59.6% of total dietary intake, while in Paris the corresponding share was 37.1%. Nevertheless, the total diet-related GHGE in Lima (45.9 Mt CO₂-eq) are approximately four times higher than those in Paris (15.4 Mt CO₂-eq) (Supplementary Fig. 1). Despite a higher plant-based share, total dietary GHGE in Lima are driven by higher emission intensities per unit of agricultural output, reflecting lower production efficiency. City-level diet-related environmental impacts show substantial reductions when urban populations transition toward EAT-Lancet dietary patterns, with varying mitigation potential across GHGE, water use, and land use (Fig. 3 ). We estimate that full adoption of the EAT-Lancet 1.0 and 2.0 diets reduces total city-level dietary GHGE by 13.6% and 14.1%, respectively (approximately 0.7 Gt CO₂-eq, comparable to Mexico’s total annual GHGE ( 22 )) (Supplementary Fig. 2). For dietary water use, the same transitions lead to modest reductions of 5.6% and 4.8% relative to the baseline levels under the EAT-Lancet 1.0 and 2.0 diets, respectively. These savings roughly correspond to the entire dietary water footprint of Indonesia (230 km³) ( 23 ). Patterns for land use mirror those observed for GHGE, resulting in a 16.1% reduction in total city-level dietary land use under the EAT-Lancet 2.0 diet, equivalent to restoring an area of farmland roughly the size of India’s cropland (156 Mha) ( 24 ). When the EAT-Lancet 2.0 diet is applied within specific transnational city networks, dietary environmental footprints decline by 22.3% across C40 member cities, 50.0% across Eurocities, and 29.8% across Milan Urban Food Policy Pact (MUFPP) cities. Across all environmental indicators, animal-based foods drive most of the global reduction in city-level diet-related environmental impacts under the EAT-Lancet 2.0 diet (Supplementary Fig. 3). At baseline, meats (processed meat and unprocessed meat) are the largest contributors to city-level dietary GHGE (2.55 Gt CO₂-eq), water use (1,419 km³), and land use (479.8 Mha). Following full adoption of the EAT-Lancet 2.0 diet, meat-related GHGE decline by 1.15 Gt CO₂-eq (-45.3%), water use by 729 km³ (-51.4%), and land use by 236 Mha (-49.2%), relative to baseline levels. Eggs account for a smaller baseline footprint (0.18 Gt CO₂-eq, 123 km³, and 20.5 Mha) yet show comparable sector-specific impact reductions across GHGE, water use, and land use of 57.6%, 52.8%, and 54.0%, respectively. Seafood contributes 0.03 Gt CO₂-eq at baseline and decreases by 0.01 Gt CO₂-eq following adoption of the EAT–Lancet 2.0 diet, equivalent to a 39.9% reduction. Although adoption of the EAT–Lancet 2.0 diet by cities leads to a net reduction in global diet-related environmental footprints, sustainability outcomes for individual cities are highly heterogeneous (Fig. 4 A). Overall, 55.9% of cities show higher GHGE, 57.0% higher water use, and 56.4% higher land use under the EAT–Lancet 2.0 diet compared with baseline diets. In high-footprint cities such as São Paulo, Moscow, Guangzhou, Lima, and Tokyo (upper rows), total GHGE typically fall by one-third to two-thirds, driven by sharp declines in meat and dairy, partly offset by higher consumption of whole grains, legumes, nuts, fruits and vegetables. By contrast, in several lower-footprint cities, particularly in South Asia, adoption of the EAT–Lancet 2.0 diet increases environmental pressures. In Dhaka and Delhi, where current diets are dominated by cereals and starchy staples with low intake of animal-source foods, GHGE almost double (89% increase in Dhaka; 108% increase in Delhi), accompanied by increases of 61–65% in water footprints and 64–92% increase in land footprints of these cities. Discussion This study offers a foundational assessment of dietary environmental footprints for over 12,000 cities, providing a new empirical basis for understanding how urban food consumption drives environmental pressures. Our findings have important implications for how cities can contribute to global sustainability transitions. We find that diet-related impacts are highly concentrated in a relatively small set of megacities. The top 50 cities drive more than a quarter of urban dietary environmental impacts (Fig. 1 ). Yet, over a quarter (28%) of these cities are not members of transnational food sustainability networks and fewer than half appear to have any city-level food plans or commitments to tackle these impacts (Supplementary Table 1), underscoring an urgent need for policy action on their dietary sustainability goals. The strong concentration of dietary footprints in a limited number of large metropolitan areas implies that targeted interventions in these cities could yield disproportionate global benefits. For example, our analysis for C40, Eurocities, and MUFPP shows that even partial adoption of sustainable reference diets within these networks can deliver notable aggregate reductions, with 9% of total dietary environmental impact reductions in dietary environmental impacts, despite covering only 2.5% of cities worldwide (Supplementary Fig. 2). Moreover, the variation in dietary patterns among similarly sized cities shows that population growth alone does not determine dietary environmental pressures; rather, the interaction of local food environments and supply chains shapes dietary footprints ( 25 ). The considerable heterogeneity in city dietary footprints and responses to adoption of the EAT-Lancet 2.0 diet highlights the need to account for equity when promoting dietary transitions. We find that shifts towards the EAT-Lancet 2.0 diet can deliver meaningful aggregate reductions in GHGE and Land use, with smaller benefits for water use, in a subset of individual cities (44%) and across transnational city networks (Fig. 4 , Supplementary Fig. 3). This is because high-footprint cities can realise substantial reductions by markedly lowering their intake of meat and dairy. However, in 56% of cities, adoption of the EAT-Lancet diet may lead to increased dietary environmental footprints across one or more environmental impacts. Specifically, cities in South Asia and parts of sub-Saharan Africa show substantial increases in GHGE, water use, and land use under the EAT-Lancet 2.0 diet, driven by their already low consumption of animal-based foods. Dietary recommendations designed for global health and sustainability objectives may not uniformly reduce environmental pressures within all contexts ( 26 ). Yet the substantial variation in dietary impacts among cities of comparable size shows that population alone does not dictate environmental pressure; rather, the interaction of urban food consumption patterns and supply-chain emission intensities also shapes city-level dietary sustainability ( 27 , 28 ). This highlights the importance of aligning global dietary recommendations with region-specific nutritional needs and environmental contexts ( 29 , 30 ). Beyond identifying the scale and distribution of dietary environmental footprints, these results point to concrete policy levers that cities can deploy to influence food consumption patterns and their associated environmental impacts. Several C40 and MUFPP cities have increasingly used regulatory, fiscal, and procurement-based interventions to reshape local food environments ( 31 , 32 ). Cities such as Milan and Barcelona have introduced measures including restrictions on high-sugar beverages, mandatory food-waste separation, and incentives for healthier retail environments ( 33 ), while Paris and London have advanced urban food strategies that integrate zoning rules for food retail, fiscal support for alternative proteins, and public information campaigns ( 34 , 35 ). The heterogeneity of dietary footprints revealed in our analysis suggests that interventions must be tailored: high-impact cities may prioritise reducing red and processed meat consumption through procurement and pricing policies ( 36 ), whereas low-footprint cities may focus on improving nutritional adequacy and food access ( 37 ). Regionalising dietary environmental footprints presents challenges for precise mapping of consumption patterns and production sources owing to the aggregated nature of food system data and footprint information. The accuracy of city-level food consumption estimates remains a major constraint of this study. Our model infers dietary patterns from population size, gender and age structure, national food intake patterns of sociodemographic subgroups, and city GDP as a proxy for energy intake, without capturing within-city heterogeneity in dietary habits. This approach overlooks regional, cultural, and socioeconomic differences that shape food choices and food access. While household budget surveys could provide income-based dietary differentiation, they often come at the expense of sectoral detail and spatial resolution. Future extensions could integrate alternative data sources such as food delivery platforms, restaurant menu datasets, and supermarket transactions, which can offer more spatially explicit information on consumption preferences ( 38 ). City-level dietary patterns have also been inferred from street-view analyses of obesogenic advertising ( 39 ) and studies of food deserts ( 40 ), which reveal spatial inequities in food access. Regionally disaggregated food intake surveys, facilitated by national statistical agencies, would substantially improve data accuracy while maintaining privacy safeguards. Using city GDP as a proxy for energy intake introduces additional uncertainty because it does not capture income inequality or related differences in calorie intake, food waste, and dietary composition ( 3 ). Finally, the geospatial datasets used to map urban populations vary in their reliance on census interpolation or building-density models, which may affect spatial accuracy, especially in megacities. For large metropolitan areas, bottom-up data integration is needed to better assess the implications of different input datasets for spatially explicit footprint estimates ( 41 ). Several uncertainties also affect the estimation of production-side impacts associated with city-scale food consumption. Environmental impacts from livestock, irrigation, and land systems can vary by several-fold within large countries due to climate, technology, and management differences ( 42 ). The FABIO framework cannot capture differences between subnational production zones or identify specific sourcing regions for individual cities. However, we recognise that regionalised sourcing patterns and cross-border linkages can connect cities to production systems with varying levels of efficiency ( 43 ). Future work should prioritise resolving city-specific sourcing heterogeneity. Accessibility-based spatial proxies, such as road connectivity ( 44 ) or port proximity ( 45 ), may help distinguish city-level sourcing patterns, but their empirical validity as predictors of import dependence remains limited ( 46 ). Furthermore, household-level food waste falls outside the MATILDA system boundaries, as the FABIO framework tracks environmental impacts only up to the point of food purchase. If post-consumer waste were incorporated, overall dietary GHGE, water use, and land use would rise—especially among younger, urban, and higher-income populations, who typically discard more food. Evidence indicates that the greatest losses occur for highly perishable plant-derived items, with waste rates commonly reaching 40–50% for fruits and vegetables, whereas animal-source foods are generally discarded at lower rates of around 20–30%. Nonetheless, because animal-based foods have substantially higher environmental intensities per kilogram, accounting for household food waste would raise the absolute magnitudes of impacts but would not alter the relative contributions of major food groups or the comparative patterns revealed by MATILDA-City. These findings highlight both the responsibility and the opportunity for cities to drive global food-system transformations. As local governments increasingly translate national and global sustainability goals into urban policy, networks such as ICLEI (International Council for Local Environmental Initiatives), which represents more than 2,500 cities that identify food, water, and energy as core pillars of their transitions, illustrate the growing momentum for integrated urban action ( 47 ). MATILDA-City provides one such rapid appraisal database, delivering globally harmonized and spatially resolved estimates. Yet these model-based assessments are most effective when complemented by targeted bottom-up data collection, strengthened local monitoring capacity, and investment in food-system analytics to capture city-specific consumption patterns and inequities. Materials and Methods This study builds upon the MATILDA (Micro–Macro Assessment Tool to Identify Low-impact Dietary Actions) framework, a harmonised global modelling system that integrates dietary intake, supply-chain environmental footprinting, and population structure data ( 19 ). MATILDA is designed to bridge the micro-level detail of individual food consumption with the macro-level representation of global food production and trade. It combines three complementary datasets: the Global Dietary Database (GDD) ( 48 ), which provides harmonised intake data for 16 food and beverage groups across 36 socio-demographic subgroups (by age, sex, education, and urban/rural area) in 165 countries; the Food and Agriculture Biomass Input–Output model (FABIO) ( 49 ), which traces food-related supply chains across 187 regions and quantifies country-specific GHGE, water use, and land use; and GLOPOP-S, a synthetic global population dataset used to scale per-capita results to population-level estimates. At its core, MATILDA establishes a concordance between the 16 GDD food groups and 72 primary commodity sectors in FABIO, enabling estimation of environmental footprints (GHGE, water use, and land use) associated with different levels of national food consumption in 165 countries. FABIO uses a Leontief demand-pull analysis to trace both direct and indirect environmental impacts linked to national food consumption, providing full supply-chain footprint of dietary patterns ( 50 ). The environmental impacts of food groups are calculated while accounting for product-level differences in food group composition (e.g. the relative shares of apples and oranges in fruit consumption) to ensure that heterogeneous national food consumption patterns in each food group were represented. Intensity coefficients calculated in FABIO (e.g. m³ of water use per kg of food intake) are then linked with GDD subgroup dietary intakes (standardised to 2,000 kcal per capita per day) to estimate per-capita environmental footprints of demographic subgroups by food group. Outlier detection and interquartile range adjustments ensure stability across products with small trade volumes or extreme coefficients, as described in Taherzadeh and colleagues ( 19 ). Finally, by multiplying per-capita footprints by subgroup population counts from GLOPOP-S or other external population datasets, MATILDA scales individual dietary impacts to total subgroup and national levels, capturing heterogeneity across 1,152 demographic–geographic combinations. Additional information on the model design and implementation is available in the Supplementary Material (Section 1 and 2). Building on MATILDA, this study developed MATILDA-City, a spatially explicit extension of the modelling framework designed to assess urban dietary footprints. MATILDA-City integrates city-level demographic data with the MATILDA framework to quantify dietary environmental impacts across 12,683 cities worldwide. For each city, GDD-based dietary intake profiles are combined with age- and sex-resolved population distributions and country-specific FABIO coefficients to estimate GHGE, water use, and land-use footprints. The scalable nature of this model also helps contextualise city-level dietary footprints relative to national and international environmental targets ( 51 ). City-level populations of different socio-demographic groups and GDP were obtained from publicly available geospatial datasets. Gridded data for population in 2025, disaggregated by age and sex bins, were obtained from WorldPop ( 52 ). Gridded GDP data for 2020 were obtained from Kummu et al. (2025) ( 53 ). We obtained city boundaries from the Global Urban Polygons and Points Dataset ( 54 ). This dataset contains a hierarchy of 123,034 urban settlements – 11% of these are urban centres, 75% are dense urban clusters, and 14% are semi-dense urban clusters. We restricted the analysis to 13,135 urban centres, i.e., top-tier cities, which together are inhabited by approximately 3.2 billion people. Gridded datasets were aggregated within each city polygon to obtain city-level totals (population by age, sex and GDP). We then excluded cities with missing boundary identifiers and retained only those located in countries covered by the MATILDA framework (165 countries), resulting in a harmonised dataset of 12,683 cities across 159 countries. Energy intake was adjusted in two steps. First, energy-intake scaling was introduced to adjust GDD’s 2,000 kcal reference values to realistic energy levels based on city-level economic status. We estimated a global log-linear relationship between national GDP per capita (PPP, constant 2021 USD) and per-capita calorie supply from the FAO food balance sheets and applied this relationship to downscaled gridded GDP data. The predicted energy intake for each city was then used to rescale all dietary intakes prior to environmental impact calculation. Second, dietary intakes were further rescaled using anthropometry-based, age- and sex-specific energy intake estimates from Springmann ( 55 ), allowing energy intake to vary simultaneously across cities and demographic groups prior to environmental impact calculations. We apply the EAT-Lancet reference diet to study the environmental impacts of cities shifting towards healthier and sustainable diets. The first version of the EAT-Lancet diet (1.0, published 2019) ( 56 ) recommends a daily intake of approximately 2,503 kcal, whereas the updated EAT-Lancet 2.0 (2025) ( 57 ) refines nutrient and environmental targets and is standardised to 2,395 kcal per day. Both guidelines promote predominantly plant-based diets rich in whole grains, fruits, vegetables, legumes, and nuts, and lower intakes of animal-source foods, added sugars, and refined grains. Since the EAT-Lancet diets do not specify quantities for coffee, tea, or refined grains, these food groups were held constant across all analyses to ensure comparability with current consumption. The benchmarking framework compares six cases: (i) the baseline reflecting current urban diets; (ii) full adoption of EAT-Lancet 1.0; (iii) full adoption of EAT-Lancet 2.0; and (iv–vi) partial adoption where only selected transnational city networks transition toward the EAT-Lancet 2.0 pattern. These networks include C40 Cities ( 58 ), the Eurocities Network ( 59 ), and the Milan Urban Food Policy Pact (MUFPP) ( 60 )—each comprising municipalities with explicit climate and food-system commitments. Under partial adoption, member cities adopt the EAT-Lancet 2.0 diet, while all other cities retain baseline consumption profiles. Declarations Data, Materials, and Software Availability All datasets used in this study are publicly available. No proprietary or access-restricted data was used. All model inputs, intermediate datasets, and outputs are documented in the Supplementary Material, and the full methodological description is provided in the manuscript. Dietary intake data: Global Dietary Database ( https://globaldietarydatabase.org/ ); Supply-chain environmental footprint data: Food and Agriculture Biomass Input–Output model (FABIO v1.2) ( https://doi.org/10.5281/zenodo.3551067 ); Population and demographic data: WorldPop Global 1 km Population Grids ( https://doi.org/10.5258/SOTON/WP00647 ); City boundary data: Global Urban Polygons and Points Dataset (GUPPD v1) ( https://doi.org/10.7927/BRQ1-XC29 ); Economic data: Global Downscaled GDP per capita (1990–2022) (Kummu et al. 2025, Scientific Data ( https://doi.org/10.1038/s41597-025-04487-x ); World Bank World Development Indicators (GDP per capita, PPP)(worldbank.org/en/programs/icp/data). All computations were performed in R (v4.3) and Stata (v18). CREDIT authorship O.T., H.C., and S.J. designed the research. H.C., S.J., and O.T. performed the research. H.C., O.T., and S.J. analysed data. H.C. performed validation. H.C. and O.T. prepared the figures. O.T. led the research, and O.T., H.C., S.J. and D.D.M. wrote the paper. References Wiedmann T, Allen C (2021) City footprints and SDGs provide untapped potential for assessing city sustainability. Nat Commun 12:3758 Jones C, Kammen DM (2014) Spatial Distribution of U.S. Household Carbon Footprints Reveals Suburbanization Undermines Greenhouse Gas Benefits of Urban Population Density. Environ Sci Technol 48:895–902 Goldstein B, Birkved M, Fernández J, Hauschild M (2017) Surveying the Environmental Footprint of Urban Food Consumption. J Ind Ecol 21:151–165 Chu EK et al (2023) U.S. Global Change Research Program,., Chapter 12 : Built Environment, Urban Systems, and Cities. Fifth National Climate Assessment Song K, Farr KB, Hsu A (2024) Assessing subnational climate action in G20 cities and regions: Progress and ambition. One Earth 7:2189–2203 Huo D et al (2022) Carbon Monitor Cities near-real-time daily estimates of CO2 emissions from 1500 cities worldwide. Sci Data 9:533 D. Moran, et al., Estimating CO 2 emissions for 108 000 European cities. Earth System Science Data 14, 845–864 (2022) Completing the picture: How the circular economy tackles climate change (2021) Wilmsen F, Gesing F (2016) The Global Protocol for Community-Scale Greenhouse Gas Emission Inventories (GPC) – A New Passage Point on an Old Road? [Preprint] Available at: https://papers.ssrn.com/abstract=2874704 Zezza A, Carletto C, Fiedler JL, Gennari P, Jolliffe D (2017) Food counts. Measuring food consumption and expenditures in household consumption and expenditure surveys (HCES). Introduction to the special issue. Food Policy 72:1–6 Trendl A et al (2023) Estimating carbon footprints from large scale financial transaction data. J Ind Ecol 27:56–70 Ivanova D et al (2017) Mapping the carbon footprint of EU regions. Environ Res Lett 12:054013 Engström RE et al (2021) Succeeding at home and abroad: accounting for the international spillovers of cities’ SDG actions. npj Urban Sustain 1:18 Stelwagen RE, Slegers PM, de Schutter L, van Leeuwen ES (2021) A bottom-up approach to model the environmental impact of the last-mile in an urban food-system. Sustainable Prod Consum 26:958–970 Riveros F, López-Eccher C, Muñoz E (2024) Life cycle assessment of food consumption in different cities: Analysis of socioeconomic level and environmental hotspots. Clean Environ Syst 13:100190 Carey J, Cook B (2021) The Milan Urban Food Policy Pact monitoring framework WHO (2023) The State of Food Security and Nutrition in the World 2023. Urbanization, agrifood systems transformation and healthy diets across the rural–urban continuum. FAO; IFAD; UNICEF; WFP Goldstein BP et al (2025) The carbon hoofprint of cities is shaped by geography and production in the livestock supply chain. Nat Clim Chang 15:1190–1197 Taherzadeh O, Schoorlemmer A, Behrens P, Bruckner M, Cai H (2025) Mapping high-impact groups across dietary sustainability and nutrition. [Preprint] Available at: https://www.researchsquare.com/article/rs-7188924/v1 Caleffi S, Micha R, Mozaffarian D, Springmann M (2025) The socio-demographic characteristics of food-related environmental impacts. [Preprint] Available at: https://www.researchsquare.com/article/rs-5434310/v1 FAO (2022) Greenhouse Gas Emissions from Agri-food Systems—Global, Regional and Country Trends, 2000–2020. FAOSTAT Anal Br Ser 50:1–12 European Parliament. Directorate General for Internal Policies of the Union., The COP27 climate change conference: status of climate negotiations and issues at stake. (2022) National water footprint explorer – Water Footprint Network Available at: https://www.waterfootprint.org/resources/interactive-tools/national-water-footprint-explorer/ [Accessed 21 January 2026] Countries ranked by Arable land (hectares) Available at: https://www.indexmundi.com/facts/indicators/ag.lnd.arbl.ha/rankings?utm_source=chatgpt.com Schröder P et al (2019) Advancing sustainable consumption and production in cities - A transdisciplinary research and stakeholder engagement framework to address consumption-based emissions and impacts. J Clean Prod 213:114–125 Springmann M et al (2018) Health and nutritional aspects of sustainable diet strategies and their association with environmental impacts: a global modelling analysis with country-level detail. Lancet Planet Health 2:e451–e461 Hu Y et al (2020) Transboundary Environmental Footprints of the Urban Food Supply Chain and Mitigation Strategies. Environ Sci Technol 54:10460–10471 Nakayama S, Yan W (2025) Unravelling food carbon footprint heterogeneity in metropolitan areas using Tokyo as a case study. Sustainable Cities Soc 121:106236 James-Martin G et al (2022) Environmental sustainability in national food-based dietary guidelines: a global review. Lancet Planet Health 6:e977–e986 Springmann M et al (2020) The healthiness and sustainability of national and global food based dietary guidelines: modelling study. BMJ 370:m2322 Candel JJL (2020) What’s on the menu? A global assessment of MUFPP signatory cities’ food strategies. Agroecology Sustainable Food Syst 44:919–946 Nguyen TMP, Davidson K, Coenen L (2020) Understanding how city networks are leveraging climate action: experimentation through C40. Urban Transform 2:12 EUROCITIES COM (2017) Food in cities: study on innovation for a sustainable and healthy production, delivery, and consumption of food in cities Moragues-Faus A (2021) The emergence of city food networks: Rescaling the impact of urban food policies. Food Policy 103:102107 Baker L, de Zeeuw H (2015) Urban food policies and programmes: An overview. Cities Agric 44–73 Trewern J, Chenoweth J, Christie I (2022) Does it change the nature of food and capitalism? Exploring expert perspectives on public policies for a transition to ‘less and better’ meat and dairy. Environ Sci Policy 128:110–120 Berti C et al (2025) Climate Change and Consumers’ Food Choices towards Sustainability: A Narrative Review. Nutr Rev nuaf151 https://doi.org/10.1093/nutrit/nuaf151 Tufano M et al (2025) Data-driven nutritional assessment of urban food landscapes: insights from Boston, London, and Dubai. Sci Rep 15:24453 Egli V et al (2019) Viewing obesogenic advertising in children’s neighbourhoods using Google Street View. Geographical Res 57:84–97 Chenarides L, Cho C, Nayga RM, Thomsen MR (2021) Dollar stores and food deserts. Appl Geogr 134:102497 Świąder M, Schafer LJ, Lysák M, Henriksen CB (2025) How data collection may affect the carbon footprint – The case of carbon foodprint accounting for cities. Ecol Ind 172:113256 Cai H et al (2024) How do regional and demographic differences in diets affect the health and environmental impact in China? Food Policy 124:102607 Clark M, Tilman D (2017) Comparative analysis of environmental impacts of agricultural production systems, agricultural input efficiency, and food choice. Environ Res Lett 12:064016 Cooper GS, Shankar B (2024) Mapping coexisting hotspots of multidimensional food market (in)accessibility and climate vulnerability. Environ Res Lett 19:054055 Brown ME, Silver KC, Rajagopalan K (2013) A city and national metric measuring isolation from the global market for food security assessment. Appl Geogr 38:119–128 Kummu M et al (2020) Interplay of trade and food system resilience: Gains on supply diversity over time at the cost of trade independency. Global Food Secur 24:100360 Frantzeskaki N, Buchel S, Spork C, Ludwig K, Kok MTJ (2019) The Multiple Roles of ICLEI: Intermediating to Innovate Urban Biodiversity Governance. Ecol Econ 164:106350 Miller V et al (2021) Global Dietary Database 2017: data availability and gaps on 54 major foods, beverages and nutrients among 5.6 million children and adults from 1220 surveys worldwide. BMJ Glob Health 6 Bruckner M et al (2019) FABIO—The Construction of the Food and Agriculture Biomass Input–Output Model. Environ Sci Technol 53:11302–11312 Kanemoto K, Lenzen M, Peters GP, Moran DD, Geschke A (2012) Frameworks for Comparing Emissions Associated with Production, Consumption, And International Trade. Environ Sci Technol 46:172–179 Rubiconto F, Halleck Vega SM, van Leeuwen ES (2024) Cross-scale consumption-based simulation models can promote sustainable metropolitan food systems. npj Urban Sustain 4:44 WorldPop (2018) Deposited, Global 1km Population. University of Southampton. https://doi.org/10.5258/SOTON/WP00647 Kummu M, Kosonen M, Masoumzadeh S, Sayyar (2025) Downscaled gridded global dataset for gross domestic product (GDP) per capita PPP over 1990–2022. Sci Data 12:178 Center For International Earth Science Information Network-CIESIN-Columbia University, Joint Research Centre-JRC-European Commission (2024) Global Urban Polygons and Points Dataset (GUPPD), Version 1. Palisades, NY: NASA Socioeconomic Data and Applications Center (SEDAC). https://doi.org/10.7927/BRQ1-XC29 Springmann M (2025) Estimates of energy intake, requirements and imbalances based on anthropometric measurements at global, regional and national levels and for sociodemographic groups: a modelling study. bmjph 3 Willett W et al (2019) Food in the Anthropocene: the EAT–Lancet Commission on healthy diets from sustainable food systems. Lancet 393:447–492 Rockström J et al (2025) The EAT–Lancet Commission on healthy, sustainable, and just food systems. Lancet 406:1625–1700 C40 Cities Transforming urban systems for climate action. C40 Cities . Available at: https://www.c40.org/ Eurocities - Home Available at: https://eurocities.eu/ Milan Urban Food Policy Pact Milan Urban Food Policy Pact . Available at: https://www.milanurbanfoodpolicypact.org/ Additional Declarations The authors declare no competing interests. Supplementary Files SupplementaryMaterial.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8830739","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":588317530,"identity":"d5309b41-c496-4d14-87b2-6865c701a279","order_by":0,"name":"Hongyi Cai","email":"","orcid":"https://orcid.org/0000-0003-3767-0465","institution":"Institute of Environmental Sciences (CML), Leiden University, PO Box 9518, Leiden 2300 RA, The Netherlands","correspondingAuthor":false,"prefix":"","firstName":"Hongyi","middleName":"","lastName":"Cai","suffix":""},{"id":588318513,"identity":"94f87b4e-500b-4737-a9d6-0e2978c0710d","order_by":1,"name":"Shruti Jain","email":"","orcid":"https://orcid.org/0000-0002-8697-2673","institution":"Institute of Environmental Sciences (CML), Leiden University, PO Box 9518, Leiden 2300 RA, The Netherlands","correspondingAuthor":false,"prefix":"","firstName":"Shruti","middleName":"","lastName":"Jain","suffix":""},{"id":588318514,"identity":"fb5aebfb-2bc5-4aee-8a36-eaa24787bd1a","order_by":2,"name":"Daniel Moran","email":"","orcid":"https://orcid.org/0000-0002-2310-2275","institution":"NILU, Environmental Impacts and Sustainability, Kjøpmannsgata 8, 7013 Trondheim, Norway","correspondingAuthor":false,"prefix":"","firstName":"Daniel","middleName":"","lastName":"Moran","suffix":""},{"id":588318515,"identity":"d1d0730f-043d-4154-8150-ba0c6239401a","order_by":3,"name":"Oliver Taherzadeh","email":"data:image/png;base64,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","orcid":"https://orcid.org/0000-0001-6144-9483","institution":"Institute of Environmental Sciences (CML), Leiden University, PO Box 9518, Leiden 2300 RA, The Netherlands","correspondingAuthor":true,"prefix":"","firstName":"Oliver","middleName":"","lastName":"Taherzadeh","suffix":""}],"badges":[],"createdAt":"2026-02-09 13:00:18","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-8830739/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8830739/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":102276775,"identity":"79c5d322-f290-4435-9d2b-68499bb7f472","added_by":"auto","created_at":"2026-02-10 06:05:39","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":480880,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCumulative city-level diet-related GHGE by population rank and world region\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003ePanel (A) shows the cumulative distribution of total diet-related GHGE across all cities. Cities are ordered on the x-axis from largest to smallest by total population, and the y-axis shows the cumulative sum of city-level GHGE in Gt CO₂-equivalent. The right-hand y-axis expresses the corresponding percentage of city-level diet-related GHGE. Panel (B) presents the same cumulative GHGE curve, but the x-axis is organised by the world region (Asia, Africa, South America, Europe, North America, Oceania), with cities within each region sorted from largest to smallest by population size.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"figure1cumulativeenvironmentalimpact.png","url":"https://assets-eu.researchsquare.com/files/rs-8830739/v1/57ca26bbcd2abc2e0c38fb89.png"},{"id":102276773,"identity":"b384e952-503b-4c99-8a1e-bb1bd0fc03cb","added_by":"auto","created_at":"2026-02-10 06:05:39","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1515071,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eRelationship between city-level GHGE and total population across 12,683 cities worldwide.\u003c/strong\u003e\u003cbr\u003e\n \u003cem\u003ePanel\u003c/em\u003e \u003cem\u003e(A) shows the full distribution of cities showing the relationship between total population (million) and total dietary GHGE (Mt CO₂-eq). Each point represents one city (n = 12,683). The x-axis shows total population (million), the y-axis indicates total city-level GHGE (Mt CO₂-eq). Point size reflects the city’s share of national dietary GHGE, as calculated by MATILDA (20) and point transparency corresponds to its share of the national population. The dashed line represents the fitted linear regression between variables (R² = 0.807). Panel (B) shows the same relationship after excluding cities with total population \u0026gt; 10 million or total dietary GHGE \u0026gt; 10 Mt CO₂-eq (12,633 cities remaining), to highlight the trends among small- and medium-sized cities. Cities labelled fall within the top 40 cities ranked by total GHGE in each panel.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"figure2scatterplotA.png","url":"https://assets-eu.researchsquare.com/files/rs-8830739/v1/ae997246ae0b911839a34461.png"},{"id":102298010,"identity":"2e8350d2-9407-497d-970d-27cd67b11839","added_by":"auto","created_at":"2026-02-10 10:30:06","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":588441,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCity-level changes in total dietary GHGE under the EAT-Lancet 2.0 diet\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e(A) Representative cities were ranked by percentage change in total dietary GHGE (Gt CO₂-eq), and the 10 labelled cities were selected at evenly spaced intervals across the lower and upper ends of the distribution. Total dietary GHGE were computed by aggregating food-group–specific emissions across all food groups, with EAT-Lancet 2.0 substitutions applied to plant-based foods, animal-source foods, and sugar, while food groups not specified in the EAT-Lancet recommendations were held constant. (B) The left bar represents baseline total GHGE, while the right bar shows total GHGE under EAT-Lancet 2.0. Middle bars illustrate the aggregated GHGE increases and decreases attributable to shifts in individual food groups, relative to the baseline. Positive and negative contributions are shown separately, stacked by food group.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"figure3waterfall.png","url":"https://assets-eu.researchsquare.com/files/rs-8830739/v1/edc2c182c29b1580b58af0cc.png"},{"id":102276772,"identity":"3f6367a8-a245-435b-a20b-13117a75bf39","added_by":"auto","created_at":"2026-02-10 06:05:39","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":946773,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eComparison of dietary (A) GHGE, (B) Water use, and (C) Land use between the Baseline and EAT-Lancet 2.0 diet across representative cities.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003ePanel (A) presents stacked bar charts of absolute changes in total diet-related greenhouse gas emissions (GHGE; Mt CO₂-eq/year) for selected cities, comparing current dietary patterns (Baseline) with the EAT–Lancet 2.0 diet. Panel (B) shows corresponding plots for diet-related water use (km³/year), and Panel (C) for land use (Mha/year). For each indicator, the upper row displays the five cities with the largest absolute reductions in environmental footprints following adoption of the EAT–Lancet 2.0 diet, while the lower row displays the five cities with the smallest absolute reductions (i.e., lowest absolute changes). Within each bar, total city-level changes are decomposed into contributions from 16 food groups.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"figure4ghgewaterlandproportion.png","url":"https://assets-eu.researchsquare.com/files/rs-8830739/v1/883f943303811fbaa69b7471.png"},{"id":102962360,"identity":"728b9e2b-9340-42da-b152-4f24907c3aed","added_by":"auto","created_at":"2026-02-19 04:07:28","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3644016,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8830739/v1/38b3f0bb-9060-44f4-a6f3-edc5cd6a9353.pdf"},{"id":102276774,"identity":"9e40eee2-e3a5-45ec-9252-89e62b134607","added_by":"auto","created_at":"2026-02-10 06:05:39","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":657052,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cbr\u003e\u003c/p\u003e","description":"","filename":"SupplementaryMaterial.docx","url":"https://assets-eu.researchsquare.com/files/rs-8830739/v1/a30a4c9aed747861c75f292e.docx"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eThe dietary footprints and transition priorities of over 12,000 cities\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eCities play a critical role in achieving ambitious sustainability goals, as key drivers of global resource consumption and emissions (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e). Infrastructure decisions are critical to curtailing these impacts in the buildings and transport sectors (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e). Yet, dietary change is also a key pillar to urban sustainability. Unlike the lock-in of infrastructure, dietary patterns and their impacts can be influenced by short-term, large-scale policy interventions. Several transnational city networks have been established to this end. The C40 Good Food Cities initiative, 100 climate neutral cities mission of the EU, and the Milan Urban Food Policy Pact (signed by 330 cities) were set up to align and even surpass local food strategies with national and international sustainability goals. These include near-term goals for wide-scale adoption of a planetary health diet (e.g. Barcelona, New York, Milan), food waste reduction targets, and fiscal instruments to limit sugar-sweetened beverages. Despite these commitments, progress on these action plans remains hampered by data to benchmark, compare and rank city-level food consumption, impacts, strategies and goals (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eData collection and modelling efforts designed to guide sustainable development routinely focus on consumption patterns and impacts at a national scale, overlooking the uneven contribution of urban regions to anthropogenic environmental change. Where cities are the focus of concern, models typically support an aggregated or partial assessment of consumption activities. Early studies captured direct, production-based emissions of cities (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e). Although instructive, these approaches ignore the global impacts of urban consumption that extend beyond city borders and can account for around 50% of emissions-related impacts (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e). Global, consumption-based footprinting analyses of cities have emerged in recent years in response to the growing need for cities to report direct and indirect emissions (\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e). These models have sought to use economic data to decompose national expenditure patterns by region, using household expenditure surveys (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e) or spending data (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e), to better capture supply-chain related environmental impacts of city consumption (\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e). However, these are limited by coarse food categorisation and poorly reflect actual physical food intake, preventing actionable hotspotting of food consumption patterns and impacts across urban regions.\u003c/p\u003e \u003cp\u003eBottom-up data collection on urban food consumption patterns offers another avenue for city-level dietary environmental footprinting. Such an approach relies on localised dietary recall surveys and urban assessment of food provisioning (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e). Whilst this provides superior detail of urban food habits, such data collection is not amenable to global comparison and remains impractical at a global scale due to time and resource constraints. Many cities lack funding to establish a comprehensive assessment of city-level consumption and monitoring against a wide suite of indicators (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e). Among the G20, only around 40% of cities have adequate data to track climate pledges among, an issue that is more pronounced in low-income countries (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e). Although national food intake surveys collect information on sub-regions, such information is either aggregated for privacy reasons or too incomplete to enable robust assessment of city-specific consumption patterns (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e). Yet, by coupling information on population size and structure, food consumption patterns and impacts of cities can be derived, based on nationally-derived food consumption profiles of socio-demographic subgroups. We follow such an approach, to construct a global database for comparable, city-level dietary footprint estimates: MATILDA-City, building on MATILDA (Micro-Macro Assessment Tool to Identify Low-impact Dietary Actions), a global, consumer-level dietary footprinting tool (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe harmonisation of food consumption microdata, environmental impact information and demographic data provides a basis to estimate city-level footprints (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e). Within this study, we couple MATILDA (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e) with spatially-derived city population and economic data to estimate the food consumption patterns and environmental footprint of 12,683 cities by greenhouse gas emission (GHGE), water, and land footprint (see Materials and Methods). The resultant database, MATILDA-City, enables comparison, ranking and scenario modeling of dietary environmental footprints across different cities and can flexibly accommodate new city consumption or food supply data. We illustrate the usefulness of this tool by calculating the environmental gains of adopting the EAT-Lancet 2.0 diet across each city and broad transnational city networks, including C40 cities, EU climate cities and the Milan Urban Food Policy Pact (MUFPP).\u003c/p\u003e \u003cp\u003eOur analysis shows a highly uneven distribution of city dietary environmental footprints worldwide. The 200 largest cities account for nearly half of the total 5.16 Gt CO₂-eq of global city dietary emissions, while megacities such as Jakarta, S\u0026atilde;o Paulo, Guangzhou, and Tokyo, each exceed 70 Mt CO₂-eq.\u0026nbsp;Across cities, animal-source foods dominate environmental pressures, with meat alone contributing 49% of dietary GHGE, 38% of water use, and 54% of land use. If the EAT-Lancet 2.0 diet were fully adopted, global city dietary greenhouse gas emissions would decrease by 14%, land use by 16%, and water use by 5%. These aggregate reductions arise because the EAT\u0026ndash;Lancet diet leads to heterogeneous changes in diet-related environmental footprints across cities, with reductions in high-footprint cities but increases in other cities. Overall, 56% of cities experience increases in at least one environmental footprint indicator, resulting in a modest global net reduction. Within transnational city networks, adoption of the EAT-Lancet 2.0 diet yields net reductions in dietary environmental footprints, amounting to 22% in C40 Cities, 50% in Eurocities, and 30% across the Milan Urban Food Policy Pact members. The underlying database, MATILDA-City, provides a scalable database to support city-level decision support towards sustainable and equitable urban food systems.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eOur results indicate that urban diet-related greenhouse gas emissions (GHGE) are driven by a limited number of megacities. Globally, city-level diet-related GHG emissions total 5.16 Gt CO₂-eq per year (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA). Asian cities contribute to nearly half of global dietary GHGE (51.4%), followed by South America (16.6%), Africa (11.9%), and Europe (10.7%) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB). As expected, total city-dietary GHGE show a strong positive correlation with total population size (R\u0026sup2; = 0.810), indicating that larger cities generally have higher absolute emissions (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). The 200 most populous cities collectively account for 39.8% of the total urban population and 45.5% of the total GHGE across all cities. Collectively, the dietary GHGE from the 200 cities amount to 2.35 Gt CO₂-eq, comparable to diet-related emissions of Brazil (2.05 Gt CO₂-eq) (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e). Among the 12,683 cities analysed, only four megacities exceed 70 Mt CO₂-eq - Jakarta (84.4 Mt CO₂-eq), S\u0026atilde;o Paulo (84.1 Mt CO₂-eq), Guangzhou (81.9 Mt CO₂-eq), and Tokyo (73.5 Mt CO₂-eq), each comparable to the diet-related emissions of a small- to medium-sized country such as the Philippines (81.7 Mt CO₂-eq) or South Sudan (71.8 Mt CO₂-eq) (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e). Cities with populations below 10\u0026nbsp;million contribute less than 5% of their national dietary GHGE, whereas several cities account for substantial shares of their respective national dietary GHGE footprints, such as Lima (9.9\u0026nbsp;million people, 45.9 Mt CO₂-eq, 59.5% of national total) and Moscow (15.6\u0026nbsp;million people, 45.9 Mt CO₂-eq, 23.8%). Regarding population distribution, most cities (12,234; 96.4%) represent less than 5% of their national population, whereas only 255 cities (2.0%) account for more than 10%.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eCity-level diet-related GHGE are unevenly distributed across food groups and cities. Meat (processed meats \u0026amp; unprocessed meats), dairy products, and refined grains account for the largest shares of city-level GHGE (Supplementary Fig.\u0026nbsp;1A). In high-emission cities such as Jakarta, S\u0026atilde;o Paulo, and Guangzhou, meat alone accounts for over one-third of total dietary GHGE, while refined grains and dairy products together account for approximately one-quarter. In contrast, fruits, vegetables, and legumes contribute less than 10% on average to city-level GHGE and other impacts. Per-capita consumption patterns across cities (Supplementary Fig.\u0026nbsp;1B) show marked heterogeneity. Refined grains dominate diets in most cities (\u0026gt;\u0026thinsp;400 g/day), especially across Africa and Asia, whereas dairy intake varies widely, from low levels in many African and Asian cities (about 120 g/day) to very high levels in Europe, North America, and Oceania (\u0026gt;\u0026thinsp;900 g/day). Meat consumption is particularly high in several European and North American cities. Cities with comparable population sizes exhibit markedly different dietary compositions and associated diet-related GHGE. For example, despite comparable total populations, Lima (9.9\u0026nbsp;million) and Paris (10.1\u0026nbsp;million) exhibit distinct dietary patterns and associated environmental impacts. In Lima, plant-based foods account for 59.6% of total dietary intake, while in Paris the corresponding share was 37.1%. Nevertheless, the total diet-related GHGE in Lima (45.9 Mt CO₂-eq) are approximately four times higher than those in Paris (15.4 Mt CO₂-eq) (Supplementary Fig.\u0026nbsp;1). Despite a higher plant-based share, total dietary GHGE in Lima are driven by higher emission intensities per unit of agricultural output, reflecting lower production efficiency.\u003c/p\u003e \u003cp\u003eCity-level diet-related environmental impacts show substantial reductions when urban populations transition toward EAT-Lancet dietary patterns, with varying mitigation potential across GHGE, water use, and land use (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). We estimate that full adoption of the EAT-Lancet 1.0 and 2.0 diets reduces total city-level dietary GHGE by 13.6% and 14.1%, respectively (approximately 0.7 Gt CO₂-eq, comparable to Mexico\u0026rsquo;s total annual GHGE (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e)) (Supplementary Fig.\u0026nbsp;2). For dietary water use, the same transitions lead to modest reductions of 5.6% and 4.8% relative to the baseline levels under the EAT-Lancet 1.0 and 2.0 diets, respectively. These savings roughly correspond to the entire dietary water footprint of Indonesia (230 km\u0026sup3;) (\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e). Patterns for land use mirror those observed for GHGE, resulting in a 16.1% reduction in total city-level dietary land use under the EAT-Lancet 2.0 diet, equivalent to restoring an area of farmland roughly the size of India\u0026rsquo;s cropland (156 Mha) (\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e). When the EAT-Lancet 2.0 diet is applied within specific transnational city networks, dietary environmental footprints decline by 22.3% across C40 member cities, 50.0% across Eurocities, and 29.8% across Milan Urban Food Policy Pact (MUFPP) cities.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eAcross all environmental indicators, animal-based foods drive most of the global reduction in city-level diet-related environmental impacts under the EAT-Lancet 2.0 diet (Supplementary Fig.\u0026nbsp;3). At baseline, meats (processed meat and unprocessed meat) are the largest contributors to city-level dietary GHGE (2.55 Gt CO₂-eq), water use (1,419 km\u0026sup3;), and land use (479.8 Mha). Following full adoption of the EAT-Lancet 2.0 diet, meat-related GHGE decline by 1.15 Gt CO₂-eq (-45.3%), water use by 729 km\u0026sup3; (-51.4%), and land use by 236 Mha (-49.2%), relative to baseline levels. Eggs account for a smaller baseline footprint (0.18 Gt CO₂-eq, 123 km\u0026sup3;, and 20.5 Mha) yet show comparable sector-specific impact reductions across GHGE, water use, and land use of 57.6%, 52.8%, and 54.0%, respectively. Seafood contributes 0.03 Gt CO₂-eq at baseline and decreases by 0.01 Gt CO₂-eq following adoption of the EAT\u0026ndash;Lancet 2.0 diet, equivalent to a 39.9% reduction.\u003c/p\u003e \u003cp\u003eAlthough adoption of the EAT\u0026ndash;Lancet 2.0 diet by cities leads to a net reduction in global diet-related environmental footprints, sustainability outcomes for individual cities are highly heterogeneous (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). Overall, 55.9% of cities show higher GHGE, 57.0% higher water use, and 56.4% higher land use under the EAT\u0026ndash;Lancet 2.0 diet compared with baseline diets. In high-footprint cities such as S\u0026atilde;o Paulo, Moscow, Guangzhou, Lima, and Tokyo (upper rows), total GHGE typically fall by one-third to two-thirds, driven by sharp declines in meat and dairy, partly offset by higher consumption of whole grains, legumes, nuts, fruits and vegetables. By contrast, in several lower-footprint cities, particularly in South Asia, adoption of the EAT\u0026ndash;Lancet 2.0 diet increases environmental pressures. In Dhaka and Delhi, where current diets are dominated by cereals and starchy staples with low intake of animal-source foods, GHGE almost double (89% increase in Dhaka; 108% increase in Delhi), accompanied by increases of 61\u0026ndash;65% in water footprints and 64\u0026ndash;92% increase in land footprints of these cities.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study offers a foundational assessment of dietary environmental footprints for over 12,000 cities, providing a new empirical basis for understanding how urban food consumption drives environmental pressures. Our findings have important implications for how cities can contribute to global sustainability transitions. We find that diet-related impacts are highly concentrated in a relatively small set of megacities. The top 50 cities drive more than a quarter of urban dietary environmental impacts (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Yet, over a quarter (28%) of these cities are not members of transnational food sustainability networks and fewer than half appear to have any city-level food plans or commitments to tackle these impacts (Supplementary Table\u0026nbsp;1), underscoring an urgent need for policy action on their dietary sustainability goals. The strong concentration of dietary footprints in a limited number of large metropolitan areas implies that targeted interventions in these cities could yield disproportionate global benefits. For example, our analysis for C40, Eurocities, and MUFPP shows that even partial adoption of sustainable reference diets within these networks can deliver notable aggregate reductions, with 9% of total dietary environmental impact reductions in dietary environmental impacts, despite covering only 2.5% of cities worldwide (Supplementary Fig.\u0026nbsp;2). Moreover, the variation in dietary patterns among similarly sized cities shows that population growth alone does not determine dietary environmental pressures; rather, the interaction of local food environments and supply chains shapes dietary footprints (\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe considerable heterogeneity in city dietary footprints and responses to adoption of the EAT-Lancet 2.0 diet highlights the need to account for equity when promoting dietary transitions. We find that shifts towards the EAT-Lancet 2.0 diet can deliver meaningful aggregate reductions in GHGE and Land use, with smaller benefits for water use, in a subset of individual cities (44%) and across transnational city networks (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, Supplementary Fig.\u0026nbsp;3). This is because high-footprint cities can realise substantial reductions by markedly lowering their intake of meat and dairy. However, in 56% of cities, adoption of the EAT-Lancet diet may lead to increased dietary environmental footprints across one or more environmental impacts. Specifically, cities in South Asia and parts of sub-Saharan Africa show substantial increases in GHGE, water use, and land use under the EAT-Lancet 2.0 diet, driven by their already low consumption of animal-based foods. Dietary recommendations designed for global health and sustainability objectives may not uniformly reduce environmental pressures within all contexts (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e). Yet the substantial variation in dietary impacts among cities of comparable size shows that population alone does not dictate environmental pressure; rather, the interaction of urban food consumption patterns and supply-chain emission intensities also shapes city-level dietary sustainability (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e). This highlights the importance of aligning global dietary recommendations with region-specific nutritional needs and environmental contexts (\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eBeyond identifying the scale and distribution of dietary environmental footprints, these results point to concrete policy levers that cities can deploy to influence food consumption patterns and their associated environmental impacts. Several C40 and MUFPP cities have increasingly used regulatory, fiscal, and procurement-based interventions to reshape local food environments (\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e). Cities such as Milan and Barcelona have introduced measures including restrictions on high-sugar beverages, mandatory food-waste separation, and incentives for healthier retail environments (\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e), while Paris and London have advanced urban food strategies that integrate zoning rules for food retail, fiscal support for alternative proteins, and public information campaigns (\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e). The heterogeneity of dietary footprints revealed in our analysis suggests that interventions must be tailored: high-impact cities may prioritise reducing red and processed meat consumption through procurement and pricing policies (\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e), whereas low-footprint cities may focus on improving nutritional adequacy and food access (\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eRegionalising dietary environmental footprints presents challenges for precise mapping of consumption patterns and production sources owing to the aggregated nature of food system data and footprint information. The accuracy of city-level food consumption estimates remains a major constraint of this study. Our model infers dietary patterns from population size, gender and age structure, national food intake patterns of sociodemographic subgroups, and city GDP as a proxy for energy intake, without capturing within-city heterogeneity in dietary habits. This approach overlooks regional, cultural, and socioeconomic differences that shape food choices and food access. While household budget surveys could provide income-based dietary differentiation, they often come at the expense of sectoral detail and spatial resolution. Future extensions could integrate alternative data sources such as food delivery platforms, restaurant menu datasets, and supermarket transactions, which can offer more spatially explicit information on consumption preferences (\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e). City-level dietary patterns have also been inferred from street-view analyses of obesogenic advertising (\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e) and studies of food deserts (\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e), which reveal spatial inequities in food access. Regionally disaggregated food intake surveys, facilitated by national statistical agencies, would substantially improve data accuracy while maintaining privacy safeguards. Using city GDP as a proxy for energy intake introduces additional uncertainty because it does not capture income inequality or related differences in calorie intake, food waste, and dietary composition (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e). Finally, the geospatial datasets used to map urban populations vary in their reliance on census interpolation or building-density models, which may affect spatial accuracy, especially in megacities. For large metropolitan areas, bottom-up data integration is needed to better assess the implications of different input datasets for spatially explicit footprint estimates (\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eSeveral uncertainties also affect the estimation of production-side impacts associated with city-scale food consumption. Environmental impacts from livestock, irrigation, and land systems can vary by several-fold within large countries due to climate, technology, and management differences (\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e). The FABIO framework cannot capture differences between subnational production zones or identify specific sourcing regions for individual cities. However, we recognise that regionalised sourcing patterns and cross-border linkages can connect cities to production systems with varying levels of efficiency (\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e). Future work should prioritise resolving city-specific sourcing heterogeneity. Accessibility-based spatial proxies, such as road connectivity (\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e) or port proximity (\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e), may help distinguish city-level sourcing patterns, but their empirical validity as predictors of import dependence remains limited (\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e). Furthermore, household-level food waste falls outside the MATILDA system boundaries, as the FABIO framework tracks environmental impacts only up to the point of food purchase. If post-consumer waste were incorporated, overall dietary GHGE, water use, and land use would rise\u0026mdash;especially among younger, urban, and higher-income populations, who typically discard more food. Evidence indicates that the greatest losses occur for highly perishable plant-derived items, with waste rates commonly reaching 40\u0026ndash;50% for fruits and vegetables, whereas animal-source foods are generally discarded at lower rates of around 20\u0026ndash;30%. Nonetheless, because animal-based foods have substantially higher environmental intensities per kilogram, accounting for household food waste would raise the absolute magnitudes of impacts but would not alter the relative contributions of major food groups or the comparative patterns revealed by MATILDA-City.\u003c/p\u003e \u003cp\u003eThese findings highlight both the responsibility and the opportunity for cities to drive global food-system transformations. As local governments increasingly translate national and global sustainability goals into urban policy, networks such as ICLEI (International Council for Local Environmental Initiatives), which represents more than 2,500 cities that identify food, water, and energy as core pillars of their transitions, illustrate the growing momentum for integrated urban action (\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e). MATILDA-City provides one such rapid appraisal database, delivering globally harmonized and spatially resolved estimates. Yet these model-based assessments are most effective when complemented by targeted bottom-up data collection, strengthened local monitoring capacity, and investment in food-system analytics to capture city-specific consumption patterns and inequities.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cp\u003eThis study builds upon the MATILDA (Micro–Macro Assessment Tool to Identify Low-impact Dietary Actions) framework, a harmonised global modelling system that integrates dietary intake, supply-chain environmental footprinting, and population structure data (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e). MATILDA is designed to bridge the micro-level detail of individual food consumption with the macro-level representation of global food production and trade. It combines three complementary datasets: the Global Dietary Database (GDD) (\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e), which provides harmonised intake data for 16 food and beverage groups across 36 socio-demographic subgroups (by age, sex, education, and urban/rural area) in 165 countries; the Food and Agriculture Biomass Input–Output model (FABIO) (\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e), which traces food-related supply chains across 187 regions and quantifies country-specific GHGE, water use, and land use; and GLOPOP-S, a synthetic global population dataset used to scale per-capita results to population-level estimates.\u003c/p\u003e \u003cp\u003eAt its core, MATILDA establishes a concordance between the 16 GDD food groups and 72 primary commodity sectors in FABIO, enabling estimation of environmental footprints (GHGE, water use, and land use) associated with different levels of national food consumption in 165 countries. FABIO uses a Leontief demand-pull analysis to trace both direct and indirect environmental impacts linked to national food consumption, providing full supply-chain footprint of dietary patterns (\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e). The environmental impacts of food groups are calculated while accounting for product-level differences in food group composition (e.g. the relative shares of apples and oranges in fruit consumption) to ensure that heterogeneous national food consumption patterns in each food group were represented. Intensity coefficients calculated in FABIO (e.g. m³ of water use per kg of food intake) are then linked with GDD subgroup dietary intakes (standardised to 2,000 kcal per capita per day) to estimate per-capita environmental footprints of demographic subgroups by food group. Outlier detection and interquartile range adjustments ensure stability across products with small trade volumes or extreme coefficients, as described in Taherzadeh and colleagues (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eFinally, by multiplying per-capita footprints by subgroup population counts from GLOPOP-S or other external population datasets, MATILDA scales individual dietary impacts to total subgroup and national levels, capturing heterogeneity across 1,152 demographic–geographic combinations. Additional information on the model design and implementation is available in the Supplementary Material (Section 1 and 2).\u003c/p\u003e \u003cp\u003eBuilding on MATILDA, this study developed MATILDA-City, a spatially explicit extension of the modelling framework designed to assess urban dietary footprints. MATILDA-City integrates city-level demographic data with the MATILDA framework to quantify dietary environmental impacts across 12,683 cities worldwide. For each city, GDD-based dietary intake profiles are combined with age- and sex-resolved population distributions and country-specific FABIO coefficients to estimate GHGE, water use, and land-use footprints. The scalable nature of this model also helps contextualise city-level dietary footprints relative to national and international environmental targets (\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eCity-level populations of different socio-demographic groups and GDP were obtained from publicly available geospatial datasets. Gridded data for population in 2025, disaggregated by age and sex bins, were obtained from WorldPop (\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e). Gridded GDP data for 2020 were obtained from Kummu et al. (2025) (\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e). We obtained city boundaries from the Global Urban Polygons and Points Dataset (\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e). This dataset contains a hierarchy of 123,034 urban settlements – 11% of these are urban centres, 75% are dense urban clusters, and 14% are semi-dense urban clusters. We restricted the analysis to 13,135 urban centres, i.e., top-tier cities, which together are inhabited by approximately 3.2\u0026nbsp;billion people. Gridded datasets were aggregated within each city polygon to obtain city-level totals (population by age, sex and GDP). We then excluded cities with missing boundary identifiers and retained only those located in countries covered by the MATILDA framework (165 countries), resulting in a harmonised dataset of 12,683 cities across 159 countries.\u003c/p\u003e \u003cp\u003eEnergy intake was adjusted in two steps. First, energy-intake scaling was introduced to adjust GDD’s 2,000 kcal reference values to realistic energy levels based on city-level economic status. We estimated a global log-linear relationship between national GDP per capita (PPP, constant 2021 USD) and per-capita calorie supply from the FAO food balance sheets and applied this relationship to downscaled gridded GDP data. The predicted energy intake for each city was then used to rescale all dietary intakes prior to environmental impact calculation. Second, dietary intakes were further rescaled using anthropometry-based, age- and sex-specific energy intake estimates from Springmann (\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e), allowing energy intake to vary simultaneously across cities and demographic groups prior to environmental impact calculations.\u003c/p\u003e \u003cp\u003eWe apply the EAT-Lancet reference diet to study the environmental impacts of cities shifting towards healthier and sustainable diets. The first version of the EAT-Lancet diet (1.0, published 2019) (\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e) recommends a daily intake of approximately 2,503 kcal, whereas the updated EAT-Lancet 2.0 (2025) (\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e) refines nutrient and environmental targets and is standardised to 2,395 kcal per day. Both guidelines promote predominantly plant-based diets rich in whole grains, fruits, vegetables, legumes, and nuts, and lower intakes of animal-source foods, added sugars, and refined grains. Since the EAT-Lancet diets do not specify quantities for coffee, tea, or refined grains, these food groups were held constant across all analyses to ensure comparability with current consumption.\u003c/p\u003e \u003cp\u003eThe benchmarking framework compares six cases: (i) the baseline reflecting current urban diets; (ii) full adoption of EAT-Lancet 1.0; (iii) full adoption of EAT-Lancet 2.0; and (iv–vi) partial adoption where only selected transnational city networks transition toward the EAT-Lancet 2.0 pattern. These networks include C40 Cities (\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e), the Eurocities Network (\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e), and the Milan Urban Food Policy Pact (MUFPP) (\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e)—each comprising municipalities with explicit climate and food-system commitments. Under partial adoption, member cities adopt the EAT-Lancet 2.0 diet, while all other cities retain baseline consumption profiles.\u003c/p\u003e\n\n\n\n"},{"header":"Declarations","content":"\n\u003ch3\u003eData, Materials, and Software Availability\u003c/h3\u003e\n\u003cp\u003eAll datasets used in this study are publicly available. No proprietary or access-restricted data was used. All model inputs, intermediate datasets, and outputs are documented in the Supplementary Material, and the full methodological description is provided in the manuscript. Dietary intake data: Global Dietary Database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://globaldietarydatabase.org/\u003c/span\u003e\u003cspan address=\"https://globaldietarydatabase.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e); Supply-chain environmental footprint data: Food and Agriculture Biomass Input–Output model (FABIO v1.2) (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5281/zenodo.3551067\u003c/span\u003e\u003cspan address=\"10.5281/zenodo.3551067\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e); Population and demographic data: WorldPop Global 1 km Population Grids (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5258/SOTON/WP00647\u003c/span\u003e\u003cspan address=\"10.5258/SOTON/WP00647\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e); City boundary data: Global Urban Polygons and Points Dataset (GUPPD v1) (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.7927/BRQ1-XC29\u003c/span\u003e\u003cspan address=\"10.7927/BRQ1-XC29\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e); Economic data: Global Downscaled GDP per capita (1990–2022) (Kummu et al. 2025, Scientific Data (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41597-025-04487-x\u003c/span\u003e\u003cspan address=\"10.1038/s41597-025-04487-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e); World Bank World Development Indicators (GDP per capita, PPP)(worldbank.org/en/programs/icp/data). All computations were performed in R (v4.3) and Stata (v18).\u003c/p\u003e\u003ch3\u003eCREDIT authorship\u003c/h3\u003e\u003cp\u003eO.T., H.C., and S.J. designed the research. H.C., S.J., and O.T. performed the research. H.C., O.T., and S.J. analysed data. H.C. performed validation. H.C. and O.T. prepared the figures. O.T. led the research, and O.T., H.C., S.J. and D.D.M. wrote the paper.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eWiedmann T, Allen C (2021) City footprints and SDGs provide untapped potential for assessing city sustainability. Nat Commun 12:3758\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJones C, Kammen DM (2014) Spatial Distribution of U.S. Household Carbon Footprints Reveals Suburbanization Undermines Greenhouse Gas Benefits of Urban Population Density. Environ Sci Technol 48:895\u0026ndash;902\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGoldstein B, Birkved M, Fern\u0026aacute;ndez J, Hauschild M (2017) Surveying the Environmental Footprint of Urban Food Consumption. J Ind Ecol 21:151\u0026ndash;165\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChu EK et al (2023) U.S. Global Change Research Program,., Chapter 12 : Built Environment, Urban Systems, and Cities. Fifth National Climate Assessment\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSong K, Farr KB, Hsu A (2024) Assessing subnational climate action in G20 cities and regions: Progress and ambition. One Earth 7:2189\u0026ndash;2203\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuo D et al (2022) Carbon Monitor Cities near-real-time daily estimates of CO2 emissions from 1500 cities worldwide. Sci Data 9:533\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eD. Moran, et al., Estimating CO\u003csub\u003e2\u003c/sub\u003e emissions for 108\u0026thinsp;000 European cities. \u003cem\u003eEarth System Science Data\u003c/em\u003e 14, 845\u0026ndash;864 (2022)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCompleting the picture: How the circular economy tackles climate change (2021)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWilmsen F, Gesing F (2016) The Global Protocol for Community-Scale Greenhouse Gas Emission Inventories (GPC) \u0026ndash; A New Passage Point on an Old Road? [Preprint] Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://papers.ssrn.com/abstract=2874704\u003c/span\u003e\u003cspan address=\"https://papers.ssrn.com/abstract=2874704\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZezza A, Carletto C, Fiedler JL, Gennari P, Jolliffe D (2017) Food counts. Measuring food consumption and expenditures in household consumption and expenditure surveys (HCES). Introduction to the special issue. Food Policy 72:1\u0026ndash;6\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTrendl A et al (2023) Estimating carbon footprints from large scale financial transaction data. J Ind Ecol 27:56\u0026ndash;70\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIvanova D et al (2017) Mapping the carbon footprint of EU regions. Environ Res Lett 12:054013\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEngstr\u0026ouml;m RE et al (2021) Succeeding at home and abroad: accounting for the international spillovers of cities\u0026rsquo; SDG actions. npj Urban Sustain 1:18\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStelwagen RE, Slegers PM, de Schutter L, van Leeuwen ES (2021) A bottom-up approach to model the environmental impact of the last-mile in an urban food-system. Sustainable Prod Consum 26:958\u0026ndash;970\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRiveros F, L\u0026oacute;pez-Eccher C, Mu\u0026ntilde;oz E (2024) Life cycle assessment of food consumption in different cities: Analysis of socioeconomic level and environmental hotspots. Clean Environ Syst 13:100190\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCarey J, Cook B (2021) The Milan Urban Food Policy Pact monitoring framework\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWHO (2023) The State of Food Security and Nutrition in the World 2023. Urbanization, agrifood systems transformation and healthy diets across the rural\u0026ndash;urban continuum. FAO; IFAD; UNICEF; WFP\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGoldstein BP et al (2025) The carbon hoofprint of cities is shaped by geography and production in the livestock supply chain. Nat Clim Chang 15:1190\u0026ndash;1197\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTaherzadeh O, Schoorlemmer A, Behrens P, Bruckner M, Cai H (2025) Mapping high-impact groups across dietary sustainability and nutrition. [Preprint] Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.researchsquare.com/article/rs-7188924/v1\u003c/span\u003e\u003cspan address=\"https://www.researchsquare.com/article/rs-7188924/v1\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCaleffi S, Micha R, Mozaffarian D, Springmann M (2025) The socio-demographic characteristics of food-related environmental impacts. [Preprint] Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.researchsquare.com/article/rs-5434310/v1\u003c/span\u003e\u003cspan address=\"https://www.researchsquare.com/article/rs-5434310/v1\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFAO (2022) Greenhouse Gas Emissions from Agri-food Systems\u0026mdash;Global, Regional and Country Trends, 2000\u0026ndash;2020. FAOSTAT Anal Br Ser 50:1\u0026ndash;12\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEuropean Parliament. Directorate General for Internal Policies of the Union., The COP27 climate change conference: status of climate negotiations and issues at stake. (2022)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNational water footprint explorer \u0026ndash; Water Footprint Network Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.waterfootprint.org/resources/interactive-tools/national-water-footprint-explorer/\u003c/span\u003e\u003cspan address=\"https://www.waterfootprint.org/resources/interactive-tools/national-water-footprint-explorer/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e [Accessed 21 January 2026]\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCountries ranked by Arable land (hectares) Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.indexmundi.com/facts/indicators/ag.lnd.arbl.ha/rankings?utm_source=chatgpt.com\u003c/span\u003e\u003cspan address=\"https://www.indexmundi.com/facts/indicators/ag.lnd.arbl.ha/rankings?utm_source=chatgpt.com\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchr\u0026ouml;der P et al (2019) Advancing sustainable consumption and production in cities - A transdisciplinary research and stakeholder engagement framework to address consumption-based emissions and impacts. J Clean Prod 213:114\u0026ndash;125\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSpringmann M et al (2018) Health and nutritional aspects of sustainable diet strategies and their association with environmental impacts: a global modelling analysis with country-level detail. Lancet Planet Health 2:e451\u0026ndash;e461\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHu Y et al (2020) Transboundary Environmental Footprints of the Urban Food Supply Chain and Mitigation Strategies. Environ Sci Technol 54:10460\u0026ndash;10471\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNakayama S, Yan W (2025) Unravelling food carbon footprint heterogeneity in metropolitan areas using Tokyo as a case study. Sustainable Cities Soc 121:106236\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJames-Martin G et al (2022) Environmental sustainability in national food-based dietary guidelines: a global review. Lancet Planet Health 6:e977\u0026ndash;e986\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSpringmann M et al (2020) The healthiness and sustainability of national and global food based dietary guidelines: modelling study. BMJ 370:m2322\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCandel JJL (2020) What\u0026rsquo;s on the menu? A global assessment of MUFPP signatory cities\u0026rsquo; food strategies. Agroecology Sustainable Food Syst 44:919\u0026ndash;946\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNguyen TMP, Davidson K, Coenen L (2020) Understanding how city networks are leveraging climate action: experimentation through C40. Urban Transform 2:12\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEUROCITIES COM (2017) Food in cities: study on innovation for a sustainable and healthy production, delivery, and consumption of food in cities\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoragues-Faus A (2021) The emergence of city food networks: Rescaling the impact of urban food policies. Food Policy 103:102107\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBaker L, de Zeeuw H (2015) Urban food policies and programmes: An overview. Cities Agric 44\u0026ndash;73\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTrewern J, Chenoweth J, Christie I (2022) \u003cem\u003eDoes it change the nature of food and capitalism?\u003c/em\u003e Exploring expert perspectives on public policies for a transition to \u0026lsquo;less and better\u0026rsquo; meat and dairy. Environ Sci Policy 128:110\u0026ndash;120\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBerti C et al (2025) Climate Change and Consumers\u0026rsquo; Food Choices towards Sustainability: A Narrative Review. \u003cem\u003eNutr Rev\u003c/em\u003e nuaf151 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/nutrit/nuaf151\u003c/span\u003e\u003cspan address=\"10.1093/nutrit/nuaf151\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTufano M et al (2025) Data-driven nutritional assessment of urban food landscapes: insights from Boston, London, and Dubai. Sci Rep 15:24453\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEgli V et al (2019) Viewing obesogenic advertising in children\u0026rsquo;s neighbourhoods using Google Street View. Geographical Res 57:84\u0026ndash;97\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChenarides L, Cho C, Nayga RM, Thomsen MR (2021) Dollar stores and food deserts. Appl Geogr 134:102497\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eŚwiąder M, Schafer LJ, Lys\u0026aacute;k M, Henriksen CB (2025) How data collection may affect the carbon footprint \u0026ndash; The case of carbon foodprint accounting for cities. Ecol Ind 172:113256\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCai H et al (2024) How do regional and demographic differences in diets affect the health and environmental impact in China? Food Policy 124:102607\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eClark M, Tilman D (2017) Comparative analysis of environmental impacts of agricultural production systems, agricultural input efficiency, and food choice. Environ Res Lett 12:064016\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCooper GS, Shankar B (2024) Mapping coexisting hotspots of multidimensional food market (in)accessibility and climate vulnerability. Environ Res Lett 19:054055\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrown ME, Silver KC, Rajagopalan K (2013) A city and national metric measuring isolation from the global market for food security assessment. Appl Geogr 38:119\u0026ndash;128\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKummu M et al (2020) Interplay of trade and food system resilience: Gains on supply diversity over time at the cost of trade independency. Global Food Secur 24:100360\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFrantzeskaki N, Buchel S, Spork C, Ludwig K, Kok MTJ (2019) The Multiple Roles of ICLEI: Intermediating to Innovate Urban Biodiversity Governance. Ecol Econ 164:106350\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMiller V et al (2021) Global Dietary Database 2017: data availability and gaps on 54 major foods, beverages and nutrients among 5.6 million children and adults from 1220 surveys worldwide. BMJ Glob Health 6\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBruckner M et al (2019) FABIO\u0026mdash;The Construction of the Food and Agriculture Biomass Input\u0026ndash;Output Model. Environ Sci Technol 53:11302\u0026ndash;11312\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKanemoto K, Lenzen M, Peters GP, Moran DD, Geschke A (2012) Frameworks for Comparing Emissions Associated with Production, Consumption, And International Trade. Environ Sci Technol 46:172\u0026ndash;179\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRubiconto F, Halleck Vega SM, van Leeuwen ES (2024) Cross-scale consumption-based simulation models can promote sustainable metropolitan food systems. npj Urban Sustain 4:44\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWorldPop (2018) Deposited, Global 1km Population. University of Southampton. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5258/SOTON/WP00647\u003c/span\u003e\u003cspan address=\"10.5258/SOTON/WP00647\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKummu M, Kosonen M, Masoumzadeh S, Sayyar (2025) Downscaled gridded global dataset for gross domestic product (GDP) per capita PPP over 1990\u0026ndash;2022. Sci Data 12:178\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCenter For International Earth Science Information Network-CIESIN-Columbia University, Joint Research Centre-JRC-European Commission (2024) Global Urban Polygons and Points Dataset (GUPPD), Version 1. Palisades, NY: NASA Socioeconomic Data and Applications Center (SEDAC). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.7927/BRQ1-XC29\u003c/span\u003e\u003cspan address=\"10.7927/BRQ1-XC29\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSpringmann M (2025) Estimates of energy intake, requirements and imbalances based on anthropometric measurements at global, regional and national levels and for sociodemographic groups: a modelling study. \u003cem\u003ebmjph\u003c/em\u003e 3\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWillett W et al (2019) Food in the Anthropocene: the EAT\u0026ndash;Lancet Commission on healthy diets from sustainable food systems. Lancet 393:447\u0026ndash;492\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRockstr\u0026ouml;m J et al (2025) The EAT\u0026ndash;Lancet Commission on healthy, sustainable, and just food systems. Lancet 406:1625\u0026ndash;1700\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eC40 Cities Transforming urban systems for climate action. \u003cem\u003eC40 Cities\u003c/em\u003e. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.c40.org/\u003c/span\u003e\u003cspan address=\"https://www.c40.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEurocities - Home Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://eurocities.eu/\u003c/span\u003e\u003cspan address=\"https://eurocities.eu/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMilan Urban Food Policy Pact \u003cem\u003eMilan Urban Food Policy Pact\u003c/em\u003e. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.milanurbanfoodpolicypact.org/\u003c/span\u003e\u003cspan address=\"https://www.milanurbanfoodpolicypact.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[{"identity":"1bcab4b1-51e4-4abc-bf80-b4058b365bae","identifier":"10.13039/100010661","name":"Horizon 2020 Framework Programme","awardNumber":"Project No: 101182025","order_by":0}],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"dietary footprints, sustainable food systems, EAT–Lancet diet 2.0, urban sustainability","lastPublishedDoi":"10.21203/rs.3.rs-8830739/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8830739/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eUrban consumption plays a central role in driving global carbon emissions and is associated with a wide range of environmental pressures, including deforestation, water use, and biodiversity loss. As cities and networks of cities work to define and achieve quantitative sustainability goals, the lack of insight into their unique food footprints, shaped by each city’s specific dietary patterns and demographic composition, remains a significant obstacle. Here we combine demographic, group-specific diet, and supply-chain environmental footprint data to estimate the dietary footprints in 12,683 cities across 159 countries.\u003c/p\u003e\n\u003cp\u003eWe show that diet-related greenhouse gas emissions (GHGE) are highly concentrated in a small number of megacities. The 200 cities with the largest footprints account for approximately 15% of global population but nearly half of total global urban dietary emissions. Adoption of the EAT–Lancet 2.0 diet in the studied cities could modestly reduce total footprints (by 5–15%, depending on impact category). High-footprint cities have the greatest potential reductions from reducing meat and dairy consumption. However, these gains are largely matched by the potential rises in footprints from lower-income cities presently consuming below the EAT-Lancet recommended levels. Across transnational city networks (C40 Cities, Eurocities, and Milan Urban Food Policy Pact members) implementation of the EAT–Lancet 2.0 diet delivers substantial reductions 22–50% in GHGE, underscoring the policy relevance of city-level dietary footprinting.\u003c/p\u003e\n\u003cp\u003eOur results highlight the need for context-specific urban dietary strategies. The resulting database, MATILDA-City, provides a globally harmonised evidence base to benchmark, rank, and monitor city-level dietary footprints, supporting municipalities to coordinate actions toward sustainable food systems.\u003c/p\u003e","manuscriptTitle":"The dietary footprints and transition priorities of over 12,000 cities","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-10 06:05:34","doi":"10.21203/rs.3.rs-8830739/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"937c9f2e-706e-42c3-9dc4-b10d43cf5193","owner":[],"postedDate":"February 10th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-02-10T06:05:34+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-10 06:05:34","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8830739","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8830739","identity":"rs-8830739","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}