Greening School Meals: Sustainability Analysis and CO₂e Mitigation Potential in Spanish Publicly Funded Primary Schools

Greening School Meals: Sustainability Analysis and CO₂e Mitigation Potential in Spanish Publicly Funded Primary Schools | 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 Article Greening School Meals: Sustainability Analysis and CO₂e Mitigation Potential in Spanish Publicly Funded Primary Schools Melina Lima This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8723135/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 Shifting the global food system away from overconsuming animal-based foods and toward plant-rich diets is crucial to tackling the world’s most urgent environmental problems, including climate change. National governments have a key role to play in this process, as public institutions like schools and hospitals are significant food consumers. This article aims to understand the extent to which current school meal menus in Spanish publicly funded primary schools align with sustainability and low-emissions standards. We analyzed 163 primary school menus from across all 17 Spanish autonomous communities. Among the results, we found that 79.1% of schools had unsustainable menus and that 87.7% emitted more CO2e than the proposed benchmark, suggesting considerable potential for policy actions. Earth and environmental sciences/Environmental social sciences/Sustainability Scientific community and society/Social sciences/Climate change/Climate-change mitigation Earth and environmental sciences/Environmental social sciences/Climate-change policy Figures Figure 1 Figure 2 Figure 3 Main Strong scientific evidence shows that addressing climate change and urgent environmental issues requires transforming the global food system. This transformation must include shifting toward plant-rich diets, as the livestock sector is the primary contributor to environmental problems in the food system. 1 2 3 4 5 Failing to do so will cause us to miss the temperature limit targets of 1.5°C and 2°C, 6 and will result in a violation of the Paris Agreement. 7 Despite this, the livestock industry and meat consumption have been overlooked in environmental policies at both the national and international levels. Achieving the necessary transformation of the global food system will be complex and require involvement from a wide range of stakeholders, including individuals, local and national governments, companies, and international organizations. Regarding countries, they are legally obliged to act – especially if they are signatories to the Paris Agreement – and effective actions rely mainly on political will. In this context, extensive literature 8 9 10 11 indicates that countries can adopt initiatives across many areas to efficiently promote food system transformation, including subsidies, public procurement, taxes, dietary guidelines, and regulations. This article aims to understand the extent to which current school meal menus in Spanish publicly funded primary schools align with sustainability and low-emissions objectives. By focusing on publicly funded schools, the study examines a segment of the food system where public authorities have significant influence over food provisioning. Therefore, while the study does not directly assess specific regulatory mechanisms, it provides empirical evidence of the sustainability outcomes currently observed in Spanish publicly funded schools, indicating the potential of regulation-based interventions, such as public procurement. In the global context of meat consumption, developed countries generally consume significantly more than the recommended amount for human and planetary health, which is about 26 kg per person per year. 4 5 While the world average is 44.5 kg 12 per person annually – already well above the recommended amount – the European Union’s consumption reaches 66 kg 1 3 per person per year. When it comes to Spain, the country ranks as the top meat consumer in Europe, with an average of 94.04 kg per person per year. 14 Despite recent progress in food policy and school meal legislation, there remains significant potential to reduce meat consumption. Spain’s 2025 National Food Strategy (ENA) 15 , for example, does not consider meat reduction as part of its overall sustainability goal or as a factor in public procurement (p. 230). The country also issued a new decree 16 to promote healthy and sustainable school meals. On the one hand, it recognizes parents' right to request 100% plant-based menus, promotes the inclusion of plant foods rich in protein, and limits red meat consumption to 1 serving per week. On the other hand, the decree still allows schools to serve six portions of all meat (red, poultry, and fish) in a five-day week (Article 9), exceeding health and planetary limits 4 5 . Moreover, it is unclear about the dishes where meat is present but isn’t the main ingredient, which are very common in Spanish school meals. In Spain, there are 2.715.544 students 17 enrolled in primary school, of which 47.4% 18 use the canteen services. 96% 13 (p. 6) of primary school students attend a public school or a state-funded private school ( colegio concertado hereafter), which are the focus of this study, as they represent a food environment largely shaped by public funding and regulation and are therefore particularly relevant for sustainability-oriented policy discussions. Considering the context described earlier, this article aims to evaluate the sustainability of menus randomly collected from primary Spanish schools and estimate their CO2e mitigation potential. Our hypothesis is that most Spanish primary school canteens serve menus with unsustainable amounts of meat and that there is a significant potential to reduce CO2e emissions. The study employed a quantitative cross-sectional approach with time-stratified sampling. To assess sustainability, our primary reference was the Planetary Health Diet (EAT-Lancet Reports of 2019 and 2025) 4 5 . To calculate the CO2e emissions of each school and then extrapolate them to estimate national emissions and mitigation potential, our primary reference was Poore and Nemecek. 19 The detailed methodology is described in the Methods section. Results 3.1 General Characteristics and Sustainability: In 77.3% of schools (126), menus did not include any fully plant-based day, meaning that foods of animal origin were served every day. Accordingly, 22.7% of the schools (37) offered a plant-based menu at least once a month. Contrary to our expectations, 89% of schools (145) offered a vegetarian menu at least once per month, with the maximum being 8 days. 11% of the schools (18) have served meat every day. 79.1% of schools (129) have served more than 71.5 grams of meat per student daily, making their menus unsustainable according to the EAT-Lancet’s meat consumption recommendations (100 g weekly: 1 of beef or pork, 2 of fish, and 2 of poultry). Meanwhile, 20.9% of schools (34) have offered sustainable menus. All the schools that served meat every day (18) have been classified as unsustainable, as their daily meat intake exceeded 71.5 grams per student. 90.2% of schools (147) exceeded the limit for beef and pork. 61 of them – 37.4% of schools – offered double or more than double the recommended limit of beef and pork. Fewer schools exceeded the limit for chicken and fish, totaling 49.7% (81 schools). Moreover, the excess was usually not high, with many schools (30) exceeding the 57-gram daily limit by less than 3 grams. The most sustainable school – with the lower meat intake – consumed an average of 47.6 grams of meat per student per day. The least sustainable school – with the largest meat intake – consumed an average of 105.4 grams of meat per student per day, which is more than twice the amount of the most sustainable school and approximately 47.4% above the sustainability benchmark of 71.5 grams. Out of the 17 Spanish autonomous communities, 5 had all their schools serving unsustainable menus: Balearic Islands, Cantabria, Castile and León, Extremadura, and La Rioja. The other 12 autonomous communities had schools serving both sustainable and unsustainable menus. Public Schools x Colegios Concertados : They display very similar profiles regarding meat provision and menu sustainability. 80.2% of public schools (73) have unsustainable menus, while 19.8% (18) have sustainable menus. 77.8% of colegios concertados (56) have unsustainable menus, while 22.2% (16) have sustainable menus. Major Caterers x Local/regional Caterers: The proportion of schools classified as sustainable was higher among those whose menus were provided by major catering companies. 71.6% of schools using major catering services (48) have unsustainable menus, while 28.4% (19) have sustainable menus. 84.4% of schools that do not use a major catering service (81) have unsustainable menus, while 15,6% (15) have sustainable menus. Although one might expect that winter menus would feature more red meats and be less sustainable, while spring menus would be lighter in red meats and more sustainable, this was not observed in the collected sample. There were 6 menus from December, January and February, and 3 were sustainable while 3 were unsustainable. Concerning the months of April, May and June, there were 7 menus. Of those, 5 were unsustainable and 2 were sustainable. We observed that meat is frequently featured on menus throughout the school year, regardless of the month or season. 3.2 CO2e mitigation potential The average excess emission of the analyzed sample is 103.28 kg CO2e per student per year. 87.7% of the schools (143) emitted more CO2e than the proposed benchmark of 1226 grams (1.2 kg) of CO2e per day per student. 78 were public schools, while 65 were colegios concertados . Accordingly, 12.3% of the schools (20) emitted less CO2e than the proposed benchmark, being 12 public schools and 8 colegios concertados . Regarding catering companies, 11 of the 20 menus within the emission limit were from major catering companies, while 9 weren’t. Contrary to what might be inferred from common sense, 50% of the schools (10) that emitted CO₂e below the proposed benchmark had previously been classified as unsustainable. However, compared to the schools that were both sustainable and emitted less than the proposed benchmark, they performed worse in terms of potential CO2e emissions mitigation. While the unsustainable schools that emitted less than the benchmark saved an average of 14.8 kg of CO2e per student per year, the sustainable schools saved an average of 54 kg of CO2e per student per year. Out of 34 schools classified as sustainable, 24 emitted more CO2e than the proposed benchmark. The school with the best results saved 84.11 kg of CO2e per student annually, while the average emission reduction from schools that emitted below the proposed benchmark was 34.42 kg of CO2e per student annually. The school with the worst results exceeded the proposed benchmark by 347.15 kg of CO2e per student each year. By scaling up sample-based excess emission estimates, we have obtained the following country-level results: considering 1.235.680 primary students using catering services in public schools and colegios concertados , 13 the average total excess emission is ≈ 122.583 tons of CO₂e/year. If schools with high-emission menus matched the average performance of low-emission schools, the estimated reduction in emissions would be about 137.7 kg CO₂e per student annually. Matching the most efficient observed menu would increase this potential to nearly 187.4 kg CO₂e per student per year. At scale, this would represent an annual mitigation potential of roughly 169.800 tons of CO₂e under the average low-emission scenario, and up to 231.000 tons if the best-performing menus were adopted. 3.3 Exploratory findings on early regulatory implementation In addition to the main results, an exploratory empirical observation emerged from the data. One of the catering companies included in the sample, Aramark, formally announced its intention to begin implementing the new Spanish Decree on healthier, more sustainable school meals as of October 2025, despite the regulation becoming mandatory only in April 2026. Of the 13 menus collected from this provider, two correspond to the period prior to October 2025, while 11 were collected from October 2025 onwards. Among these eleven menus, five were classified as sustainable according to the sustainability criteria applied in this study, while the remaining six were classified as unsustainable. Moreover, only three of the eleven menus met or fell below the proposed CO₂e emission benchmark. Although based on a very limited sample and not a primary objective of this study, this observation suggests that the early implementation of the new decree does not automatically translate into sustainable menus or compliance with the proposed emission benchmark. Discussion The results confirm our hypothesis that most menus in Spanish primary schools are unsustainable (79.1%) and emit more CO2e than the proposed benchmark (87.7%), indicating substantial room to reduce CO2e emissions. By analyzing the data, we could not identify any specific factor that determined good and bad results. There were very good performers in terms of sustainability and CO2e emissions in both public schools and colegios concertados . The same is true for the worst performers. In addition, while schools served by major catering companies performed generally better on sustainability, all companies except Seral offered both very sustainable and very unsustainable menus, indicating no consistent pattern even within the same catering company. Possible explanations for the absence of a clear pattern include schools' autonomy in selecting suppliers and defining the menu profiles served, as well as the lack of regulatory definitions of food types and amounts of unsustainable foods that can be served. The Spanish Decree on healthy and sustainable school meals has not yet entered into force. However, we emphasize that the Spanish decree, although undeniably positive, is not as strict as our main reference (EAT-Lancet’s Planetary Health Diet) regarding meat portions. Additionally, it also lacks clarity on several important topics, which may explain why the catering provider that has already started complying with it still offers meals with widely varying sustainability standards. Finally, the main takeaway of this study is that the solution to the sustainability problem in Spanish school canteens has already been adopted by 20% of the schools analyzed. Regarding CO2e emissions, 12% of the schools already have a fairly reasonable emissions profile. Therefore, there is no need for disruptive measures or unknown abrupt changes. Regulating the sector to align with the sustainable standards already in place in many schools across the 17 Spanish autonomous communities should not, thus, entail major cultural changes nor lead to social or political unrest. We emphasize that, although our sample shows that a minority of Spanish autonomous communities only had schools with unsustainable menus, this should not be interpreted as evidence that sustainable menus are absent in those regions. Given the non-exhaustive nature of the sample and the known heterogeneity in school food provision within autonomous communities, it is highly likely that schools offering sustainable menus exist in these regions but were not captured in the present dataset. Therefore, the results should be understood as reflecting observed practices within the sampled schools rather than as a regional assessment. According to a recent study 20 in 171 countries, purchasing locally available, cheaper products from each of the main food groups generally leads to a healthy, low-emission diet. This indicates that a planet-friendly, healthy diet would generally not require significant lifestyle changes or expensive substitutes. Our results suggest this might also be true for Spanish school menus, which could improve their current poor sustainability and emissions profile by reducing meat consumption and increasing the use of healthy, sustainable ingredients already present in every analyzed menu. We emphasize that all the menus examined, including the unsustainable ones, featured dishes made with nutritious, relatively low-cost ingredients traditional to Spanish cuisine, such as chickpeas, lentils, beans, and vegetables. Moreover, we highlight that five of the six major catering companies, which account for nearly 60% of the market share, already know how to provide highly sustainable, low-emission menus, as evidenced by the sample. In this context, the policy component can help ensure that best practices become the norm rather than the exception, as they are now. Ultimately, the results suggest that implementing sustainable menus in the Spanish educational system is operational and feasible within the existing institutional and political context. 4.1. Limitations: a. Limitations potentially leading to an overestimation of unsustainability - Although we are aware of special menus for allergies or religious restrictions, we used the regular menu as a reference for the calculations. We acknowledge that this could affect some of the results, but we believe the differences would be minor for the following reasons: 1. There is usually no special menu for allergies related to beef, pork, or chicken; 2. A restriction concerning an animal-protein food is likely to be replaced by another animal-protein food, as observed in some schools in the sample; 3. Estimates suggest that only about 2.8% 21 of the menus are for all special diets (16 categories, including allergies and religious/ethical reasons); 4. Vegetarian and vegan menus, which would have the biggest impact on the analysis, are not available at every school and would represent only a small fraction of the 2.8% of special diets; 5. Lastly, the total sample size would be much smaller because most schools only publish the regular menus online, not the special ones. - Since the menus were collected randomly, only 41.1% came from schools served by large catering companies. Statistics estimate that up to 58% 2 2 of schools are served by market-dominating companies, and when comparing the two groups – schools served and not served by these companies – it was found that menus provided by large companies were more often sustainable than those not managed by them (28.4% of schools managed by large companies had sustainable menus, versus 15.6% of schools not managed by large companies). Therefore, if the sampled schools truly reflected the proportion served by large catering companies, there might have been slightly more sustainable schools. b. Limitations potentially leading to an underestimation of the unsustainability - We adopted a very cautious approach when interpreting dishes and always assumed they didn’t contain meat when in doubt. For example, we analyzed a selected school’s regular menu and then compared it with its vegetarian menu. The regular menu had some dishes whose names didn’t suggest they contained meat, so we assumed they didn’t. However, the vegetarian menu had many dishes with identical names, some labeled “meatless” or “with plant protein”, which led us to believe that the regular menu featured meat in those dishes. As a result, we probably underestimated the amount of meat in the overall sample, but it was a deliberate choice aimed at minimizing false positives and avoiding over-attribution of meat content in the absence of explicit information. This approach reduces the risk of systematic overestimation and makes any subsequent findings regarding meat presence more, rather than less, robust. - We only analyzed primary schools, but potential regulations would likely apply to the entire mandatory education system, including early childhood education and secondary school. Early childhood education in Spain has about 1.576.090 17 students and significant use of catering services, while secondary schools have over 2 million students 17 , but catering services are more common only in colegios concertados and private schools. Therefore, the potential for sustainability and CO2e emissions is certainly greater for the entire Spanish education system than what this paper predicts, which only includes primary education. Methods 5.1. General a. Primary data source for this study: school menus obtained from primary public schools and colegios concertados in Spain. Both types of schools account for about 96% of students enrolled in Spain and are more likely to be affected by public policies than private schools since they are either public institutions or publicly funded. The sample included public schools and colegios concertados from all 17 Spanish autonomous communities. Regarding the education level, we decided to focus on primary education because it includes the largest number of students who use school meal services in both public schools and colegios concertados . In secondary school, catering services are common in colegios concertados , but not in public schools, where attendance rates are quite low. In total, we have collected 163 menus. 91 were from public schools and 72 from colegios concertados . Regarding food service management, 67 menus were provided by one of the major catering companies in Spain, while 96 were not. b. Sample Size: Although the quantitative analysis is based on a corpus of 163 school menus – each collected, archived, and included in the supplementary materials – this figure should be understood as a conservative unit of analysis. A significant number of these menus were standardized, caterer-provided menus explicitly identified as applicable to multiple schools within a city or region (e.g., “menus for schools in Toledo” or “central kitchen services”). As a result, while the calculations consistently rely on the 163 menus for reasons of transparency and reproducibility, the empirical material likely reflects food provision practices affecting a substantially larger number of schools. c. Menu Collection Procedures: To assess the sustainability and carbon footprint of Spanish school menus, we needed to gather actual menus. However, these menus are not available through a centralized list or website. Most of the time, they are not publicly accessible on schools’ websites and are shared only with families via private channels. Because of this, we first tried to contact the major catering companies, which serve up to 58% of the school meals 21 , and asked them for menu samples. Since they either ignored or declined our messages, and it was impossible to research the websites of over 13.000 primary schools individually, we used a Large Language Model (LLM) – ChatGPT 5.1 – to identify schools that publicly displayed their menus online. We asked the LLM to list public schools and colegios concertados with online menus in each autonomous community of Spain. We usually moved on to the next community when the LLM indicated it could find no more menus. We ensured we gathered more menus from the most populous communities (Andalusia, Madrid, Catalunya, and the Valencian Community) to improve accuracy and reduce potential bias in our results. Therefore, when the LLM didn’t list enough schools for a larger community, we changed the question to include specific cities within that community. By “not enough schools,” we mean the same number or fewer than the ones listed for less populated communities. We had to use this strategy for Andalusia and Catalunya. As a result, the menus were selected at random based on their online availability, and all data was collected directly from school websites, local government websites, or the online pages of catering services. 1 d. Menu exclusion criteria: We used most of the menus found online and only excluded samples for the following reasons: 1. The menu was not from the calendar year 2025; 2. It was the third identical menu, despite being served in different schools. We observed that public schools in some cities and/or communities likely have contracts with a catering company that provides the same menu across all public schools we found. In these cases, we kept a maximum of 2 identical menus, which appeared to be representative of the community, and discarded the others to reduce potential bias in the results. We must highlight that in these communities or cities, we tried different research strategies with the LLM, but they all resulted in the same outcome: a list of schools with 100% identical menus. This situation with identical menus across all schools listed by the LLM did not occur with colegios concertados , only public ones. Identical menus from the same catering companies are common even in colegios concertados , but the full list always included schools with different menus and catering companies. e. Major catering companies versus smaller or independent companies: InSpain, there is reportedly significant market concentration among school catering providers. The menus were collected randomly, and by the end of the collection, 67 of 163 menus were supplied by one of the major companies (Serunion, Aramark, Ausolan, Mediterranea, Scolarest, and Seral). In addition to analyzing the data as a whole, we also disaggregated it into two categories: public schools and colegios concertados , and major catering companies versus non-major companies. f. Structure of school menus and rationale for using daily calculations: Most menus are scheduled monthly. A few follow a trimester system or are cycle menus that rotate every six weeks, as is the case in the Murcia community. Regarding the monthly menus, most schools’ websites had the October menu available, which was the current month at the time the research was conducted. Other websites provided more than one menu, including the October one, or a random menu not from October. Whenever menus for other months were accessible, we selected those to create the most diverse sample possible. Out of the 163 menus collected, 40 are not from October. Aside from July and August, which are vacation months, menus for all other months were analyzed. Since months and cycles have different lengths, we used the average school day as a reference instead of the month or cycle. Therefore, it didn’t matter if the menu had 15, 20, or 30 days. All the meat on each menu was accounted for and divided by the number of school days in the respective month or cycle, yielding an average amount per day. g. Rationale for Focusing on Meat as the Primary Indicator: Regarding the food, our main focus was to understand how much animal-based products were used and how they affected the sustainability of the menus. Initially, this included meat, dairy, and eggs. However, we noticed a discrepancy across the menus regarding the presence of eggs and dairy, which could have biased our results. Some menus listed all the allergens in each dish, while others did not. The count of eggs and dairy was generally much higher on menus that displayed allergens. As a result, we decided not to assess the sustainability of the menus based on these products. They were only considered to determine how many days featured plant-based meals. The calculations were based solely on the amount of meat present on the menus. Every time we use the word “meat”, it refers to all types of meat. h. Approach to Classifying Meat Content in Regional and Traditional Dishes: When interpreting dishes from Spain or its regions and cities, we researched their names on Google and reviewed the first five recipes to understand the ingredients. If there was a discrepancy between the recipes, and at least 2 of the five did not include meat, we took a cautious approach and assumed there was no meat in the dish. Additionally, when we could identify or infer that a dish traditionally made with meat was meatless, we did not count it as containing meat. The identification was made objectively, for example, when the menu described the dishes’ ingredients. The inference occurred when we had strong clues suggesting the dish was meatless. For example, the dish “lentil estofado ” or “chickpea estofado” usually contains pork meat, and the name of the dish is exactly as mentioned (in Spanish: estofado de lentejas/estofado de garbanzos). When the dish was named “lentil/chickpea estofado with vegetables,” we assumed it contained no meat, and the vegetables or any specific vegetable in the dish name replaced the pork. 5.2. Sustainability of the Menus We assessed the sustainability of the menus based on how much they exceeded the meat consumption recommended by the Planetary Health Diet (EAT-Lancet reports 2019 and 2025) 4 5 . To determine the amount of meat on each menu, we used the Spanish Consensus Document on Food Provision in Educational Institutions 23 (hereafter, the Consensus Document). Below, we outline each step taken to obtain the results. 1. We counted how many dishes had beef, pork, poultry, fish, or any other type of meat. We grouped each type of meat into two categories: dishes featuring a full meat portion and dishes containing only a partial meat portion. 2. According to the Consensus Document mentioned above, for students aged 7 to 12 (primary school), full portions of beef, pork, and poultry ranged from 80 to 160 grams, while full portions of fish ranged from 100 to 120 grams (p. 15). In this article, we used 100 grams for beef, pork, and poultry, and 110 grams for fish. 3. Regarding the partial meat portions in dishes like paella and pasta, the Consensus document specifies a range of 20 to 30 grams. We used 25 grams for partial portions. However, such dishes often included two or three types of meat. In these cases, we used the 30-gram reference and divided it by the number of different meats. For example, in a paella made with pork, poultry, and fish, we considered 10 grams per type of meat. 4. A small number of the menus (6 out of 163) featured rabbit meat. Since the EAT-Lancet reports don’t include any recommendations on this type of meat, we treated it similarly to poultry. In 5 of 6 cases where rabbit meat was offered, it was served in a partial portion. 5. To determine the daily amount of each type of meat, we multiplied the number of times each meat was served by its respective portion size (full or partial) and divided by the number of school days listed in the menu, which varied by month and school. Then, we grouped the meat types according to the EAT-Lancet report: beef with pork, and poultry with fish. 6. The daily recommended amounts of meat changed slightly between the 2019 and 2025 EAT-Lancet reports: from 14 to 15 grams for beef and pork, from 28 to 30 grams for fish, and from 29 to 30 grams for poultry. However, both reports refer to weekly portions of 100 grams 2 – one portion for beef, pork, or lamb, and two portions for fish and poultry – which match the daily amounts in the 2019 report (14, 28, and 29 grams, respectively). Therefore, we assumed that the 2025 report rounded the figures for daily servings, while keeping the weekly serving sizes unchanged (one or two of 100 grams, depending on the type of meat). Because of that, we used daily reference values of 14.5 grams for beef and pork and 28 and 29 grams for fish and poultry, respectively. Accordingly, any menu exceeding 71.5 grams per day (14.5 + 28 + 29) was classified as unsustainable. 5.3. Calculating the CO2e mitigation potential of Spanish primary school menus Climate change and CO2e emissions are just one of the nine planetary boundaries that determine sustainability. Therefore, the Planetary Health Diet (2019 and 2025) 4 5 , which established recommendations for meat consumption and was used to determine the menu’s sustainability, is based on a much more complex approach that includes all nine planetary boundaries and human health indicators. Nevertheless, given the urgent need to reduce CO2e emissions and address climate change, we find it useful to understand the emissions of school menus and their mitigation potential. Since some factors influencing CO2e emissions differ from those affecting sustainability, we calculated emissions for all collected menus, regardless of whether they were classified as sustainable or unsustainable. We chose to do so because different criteria for determining sustainability and emissions could lead to a school being labeled sustainable while emitting CO2e above the limit, and vice versa. Therefore, in this section, we used indicators from the Planetary Health Diet when pertinent, but we also included others that were exclusively pertinent to climate change and CO2e calculations. To estimate the CO2e mitigation potential of the menus, we primarily relied on Poore and Nemecek. 19 However, in Poore and Nemecek, the metric used is kg CO2e per 100 grams of protein, whereas the Consensus Document 23 and the Planetary Health Diet 4 5 use grams of food type rather than protein. Therefore, we used the numbers converted to kg CO2e per kg of food, featured on the Our World in Data 24 website, which sources its data from Poore and Nemecek. We outline below each step taken to obtain the results: 1. We initially calculated the emission limit for a sustainable school menu. To do this, we considered the sustainability limit of 71.5 grams of meat per student daily, consisting of 29 grams of poultry, 28 grams of fish, and 14.5 grams of beef or pork. The Planetary Health Diet does not specify how the 14.5 grams should be divided between pork and beef. While both meats have a similar impact on many planetary boundaries and human health, this is not true regarding CO2e emissions. Beef produces an average of 7.8 times more CO2e than pork. The impact of beef on CO2e emissions and climate change is well documented, as is the need to reduce its consumption more than that of other meats. Therefore, to divide the 14.5-gram portion, we used a ratio of 3/7, meaning three parts beef for every seven parts pork. As a result, the daily limit for beef was 4.35 grams, while for pork it was 10.15 grams. 2. With the portions for each type of meat determined, we multiplied them by their respective CO2e emissions: 99.48 kg CO2e per kg of beef; 12.31 kg CO2e per kg of pork; 9.87 kg CO2e per kg of poultry; and 13.63 kg CO2e per kg of fish. We then found that the daily emission limit for a sustainable menu (71.5 grams of meat) was 1226 grams, or around 1.2 kg, of CO2e per day per student – the sustainable menu benchmark. 3. Later, we calculated the proportion attributable to each meat type for each school. For example, at school N.° 1, beef consumption was 175 grams per student per month, while pork consumption was 625 grams per student per month. This shows that pork consumption was roughly 3.57 times higher than beef consumption. The average daily consumption of both meats per student was 36.3 grams. Based on the 3.57 ratio, the average daily intake of beef was about 7.94 grams per student, and pork was approximately 28.36 grams per student. We repeated this process for each school in both groups (pork/beef and poultry/fish). 4. As we identified the consumption of each type of meat, we multiplied each by its respective CO2e emissions: 99.48 kg CO2e per kg of beef; 12.31 kg CO2e per kg of pork; 9.87 kg CO2e per kg of poultry; and 13.63 kg CO2e per kg of fish. Then, we summed all the values to determine the total emissions. For example, school n° 20 had the following average daily consumption (in grams) per student of beef, pork, poultry, and fish, respectively: 5.94 g, 15.46 g, 26.17 g, and 47.83 g. The calculations were: (5.94 x 99.48) + (15.46 x 12.31) + (26.17 x 9.87) + (47.83 x 13.63) ≈ 1691.44 g CO2e. After calculating the total daily emissions per student for each school, we subtracted the sustainable menu benchmark (1226 g) from it to determine whether there was an excess of emissions. In the case of school N.° 20, the excess was 465.44 g CO2e. Finally, we multiplied the excess by 175 (the number of school days in a year) to determine the annual emissions exceedance relative to the benchmark. For school N.° 20, it was 81.452 g or 81.4 kg of CO2e per student per school year. 5. We repeated this process for all schools in the sample. Those below the sustainable menu benchmark had negative numbers, indicating they emitted less than the maximum set as ideal. 6. Then, by summing the results of all schools, including those with negative emissions, and dividing by the total number of schools analyzed (163), we calculated the average excess emissions across the entire sample. 7. Next, we calculated national estimates using an extrapolation method in which the average annual excess emissions per student at the sample level were multiplied by the estimated number of primary school students enrolled in public schools and colegios concertados who use the school canteen service in Spain. This figure was determined by combining official statistics 17 on total primary enrollment (2,707,302 students), percentage of students enrolled in public schools and colegios concertados (96% of the total), with the reported proportion of canteen users (47.4%). Because statistics on school meal service usage are not disaggregated by school type, we assumed a similar rate of use across school types. 8. Finally, we tested two counterfactual scenarios by estimating: (1) the mitigation potential if all schools performed at the level of the average school whose emissions were below the benchmark; and (2) the mitigation potential if all schools performed at the level of the most efficient school in the sample. These estimates assume full convergence of high-emission menus toward lower-emission benchmarks and homogeneous student participation in school meal services. 1 Before resorting to AI, the author attempted to collect menus by messaging Facebook groups of expatriates living in Spain and reaching out to acquaintances who have children enrolled in Spanish schools. After it only returned 5 menus, we tried the AI approach. Since the menus obtained through this first approach fitted the criteria, they were included in the analysis and comprise the sample of 163 menus. 2 P. 52 EAT-Lancet 2019 4 ; pp. 1632 and 1635 EAT-Lancet 2025 5 . References Yanxian Li et al., ‘Reducing Climate Change Impacts from the Global Food System Through Diet Shifts’ (2024) 14 Nat Clim Chang 943–53. https://doi.org/10.1038/s41558-024-02084-1. Marco Springmann et al., ‘Options for Keeping the Food System Within Environmental Limits’ (2018) 562 Nature 519–25. IPCC, ‘Climate Change and Land: an IPCC Special Report on Climate Change, Desertification, Land Degradation, Sustainable Land Management, Food Security, and Greenhouse Gas Fluxes in Terrestrial Ecosystems’ (2019). https://www.ipcc.ch/site/assets/uploads/sites/4/2022/11/SRCCL_Full_Report.pdf. W. Willett et al., ‘Food in the Anthropocene: the EAT-Lancet Commission on Healthy Diets from Sustainable Food Systems’ (2019) 393 Lancet 447–92. P. 52 https://doi.org/10.1016/S0140-6736(18)31788-4. Rockström, Johan et al., The EAT– Lancet Commission on healthy, sustainable, and just food systems, (2025) The Lancet, Volume 406, Issue 10512, 1625 – 1700. Pp. 1632 and 1635. 10.1016/S0140-6736(25)01201-2 Michael A. Clark et al., ‘Global Food System Emissions Could Preclude Achieving the 1.5° and 2°C Climate Change Targets’ (2020) 370 Science 705–8. https://doi.org/10.1126/science.aba7357. Author, 2024. Simona Vallone & Eric F. Lambin, Public Policies and Vested Interests Preserve the Animal Farming Status Quo at the Expense of Animal Products Analog , 6 One Earth 1213, 1213–26 (2023). Swensson, Luana F J et al., Public food procurement as a game changer for food system transformation, The Lancet Planetary Health, Volume 5, Issue 8, e495 - e496 Luana.F.J. Swensson; Florence Tartanac. Public food procurement for sustainable diets and food systems: The role of the regulatory framework, Global Food Security Volume 25, June 2020, 100366, https://doi.org/10.1016/j.gfs.2020.100366 Springmann, M., Dinivitzer, E., Freund, F. et al. A reform of value-added taxes on foods can have health, environmental and economic benefits in Europe. Nat Food 6, 161–169 (2025). https://doi.org/10.1038/s43016-024-01097-5 Our World in Data, Per capita meat and fish consumption, 2022, https://ourworldindata.org/grapher/per-capita-meat-type#:~:text=Per%20capita%20meat%20and%20fish,/meat%2Dproduction%20%7C%20CC%20BY Our World in Data, Per capita meat consumption in the EU28, https://ourworldindata.org/grapher/per-capita-meat-eu28 Euromeat, Spain and Austria have the highest rate of meat consumption in Europe (Feb, 2025) https://www.euromeatnews.com/Article-Spain-and-Austria-have-the-highest-rate-of-meat-consumption-in-Europe/2558#:~:text=Spaniards%20are%20leading%20the%20top,according%20to%20a%20UN%20chart Spain, Estrategia Nacional de Alimentación (2025), p. 230. Spain, Real Decreto 315/2025, de 15 de abril, por el que se establecen normas de desarrollo de la Ley 17/2011, de 5 de julio, de seguridad alimentaria y nutrición, para el fomento de una alimentación saludable y sostenible en centros educativos, Article 9. https://www.boe.es/eli/es/rd/2025/04/15/315 Spain, Datos y Cifras: Curso Escolar 2024-2025. https://www.lamoncloa.gob.es/serviciosdeprensa/notasprensa/educacion-fp-deportes/Documents/2024/110924-%20datos-y-cifras-curso-escolar-2024-2025.pdf?utm_source=chatgpt.com STECyL-i, La mitad del alumnado de primaria accede al comedor escolar, pero 1,3 millones de niños y niñas en pobreza siguen sin beca (Sep 2025), https://stecyl.net/la-mitad-del-alumnado-de-primaria-accede-al-comedor-escolar-pero-13-millones-de-ninos-y-ninas-en-pobreza-siguen-sin-beca/ Joseph Poore & Thomas Nemecek, Reducing Food’s Environmental Impacts Through Producers and Consumers , 360 Sci. 987, 991 (2018). Bai, Y., Martinez, E.M., Yamanaka, M. et al. Environmental impacts and monetary costs of healthy diets worldwide. Nat Food 6, 1176–1185 (2025). https://doi.org/10.1038/s43016-025-01270-4 J. R. Martínez Álvarez et al.,National survey on school lunches and demand of special diets, Nutr. Hosp. vol.27 no.1 Madrid ene./feb. 2012. https://scielo.isciii.es/scielo.php?pid=S0212-16112012000100032&script=sci_arttext&utm_source=chatgpt.com Del Campo al Cole, Los Comedores Escolares en España (2018). https://www.observatoriodelainfancia.es/ficherosoia/documentos/5587_d_Informe-Comedores-Escolares.pdf?utm_source=chatgpt.com Spain. Documento de Consenso Sobre la Alimentación em los Centros Educativos (2010). Our World in Data, Greenhouse gas emissions per kilogram of food product, https://ourworldindata.org/explorers/food-footprints?Commodity+or+Specific+Food+Product=Commodity&Environmental+Impact=Carbon+footprint&Kilogram+%2F+Protein+%2F+Calories=Per+kilogram&By+stage+of+supply+chain=false&country=Almonds~Bacon~Bananas~Beans~Beef+%28beef+herd%29~Beef+%28dairy+herd%29~Beefburger~Cheese~Cow%27s+milk~Eggs~Lamb+%26+Mutton~Maize~Milk~Peas~Penne+pasta~Pig+Meat~Pizza~Poultry+Meat~Prawns+%28farmed%29~Rice~Steak+pie~Tofu~Tomatoes~Vegetable+lasagne~Wheat+%26+Rye Additional Declarations There is NO Competing Interest. Supplementary Files DataXCalculationsSustainabilityandCO2e.xlsx Data x calculations Sustainability and CO2e emissions SchoolMenusandcateringcompanyletterAramarkcompressed.pdf School menus and Aramark Letter ExtendedData.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-8723135","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":581931566,"identity":"d32154a5-e11b-44df-a9ff-de14c4321806","order_by":0,"name":"Melina Lima","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA+ElEQVRIiWNgGAWjYJCCAw/YgCQ78wEGxgaoEJyBS0sCSAszWwLxWhggWngMiNPCL3b44YGEMrtofmaebxI/d9jIMbAfPrqBccc9nFokZ6cZHEg4l5w7s5l3m2TvmTRjBp60tBuMZ4pxajG4nWBwILGNOXfDYd5tErxthxMbJHjMbjC2JeDUYn87/QNQS33u/sM8zyT/EqPFQDoHZMvh3A3MPGzSRNkicTunAOiX47kzDrMZW8u2pRmzgfySeAa3Fv7Z6Zs/fCirzu1vb354822bjRw/++FjNz7uwK0FGbBIgEhQHDEQpwEYmR+IVDgKRsEoGAUjDAAAO/1Y951IW08AAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0001-7600-8402","institution":"IBMEC University","correspondingAuthor":true,"prefix":"","firstName":"Melina","middleName":"","lastName":"Lima","suffix":""}],"badges":[],"createdAt":"2026-01-28 16:05:30","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8723135/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8723135/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":101943518,"identity":"11030200-f094-4436-a572-e47eee90fea5","added_by":"auto","created_at":"2026-02-05 09:42:11","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":36571,"visible":true,"origin":"","legend":"\u003cp\u003eLegend not included with this version\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-8723135/v1/89ba4e2765213cd8ed0b7b8c.png"},{"id":101943434,"identity":"6e1768f6-ae87-4123-9f3a-090d14453002","added_by":"auto","created_at":"2026-02-05 09:41:56","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":34774,"visible":true,"origin":"","legend":"\u003cp\u003eLegend not included with this version\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-8723135/v1/99a9896ba4f7988909be5433.png"},{"id":101943498,"identity":"98e21c97-0693-4c57-b6c4-0102f52aa045","added_by":"auto","created_at":"2026-02-05 09:42:06","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":112616,"visible":true,"origin":"","legend":"\u003cp\u003eLegend not included with this version\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-8723135/v1/1402f81af5bcebe7f900ca28.png"},{"id":105033557,"identity":"120a0341-f115-489a-a7a0-824aef4f94b5","added_by":"auto","created_at":"2026-03-20 07:19:46","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":829814,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8723135/v1/1caf1f5e-cf08-4913-9871-0b664bba0cdf.pdf"},{"id":101943588,"identity":"9a7e39de-abd4-4c59-bd35-8e36faef68d4","added_by":"auto","created_at":"2026-02-05 09:42:26","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":85426,"visible":true,"origin":"","legend":"Data x calculations Sustainability and CO2e emissions","description":"","filename":"DataXCalculationsSustainabilityandCO2e.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-8723135/v1/9afe02bf705e4ad705541767.xlsx"},{"id":101943329,"identity":"f32ae711-2334-47f4-a70f-3f3bd7ec5c96","added_by":"auto","created_at":"2026-02-05 09:41:38","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":19876731,"visible":true,"origin":"","legend":"School menus and Aramark Letter","description":"","filename":"SchoolMenusandcateringcompanyletterAramarkcompressed.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8723135/v1/49ed5474ea1892faa62cf67a.pdf"},{"id":101924169,"identity":"0e82c4c5-02cf-4332-8c6a-faa55a8ebfb4","added_by":"auto","created_at":"2026-02-05 05:45:35","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":30096,"visible":true,"origin":"","legend":"","description":"","filename":"ExtendedData.docx","url":"https://assets-eu.researchsquare.com/files/rs-8723135/v1/122379d267d828c3f2738114.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Greening School Meals: Sustainability Analysis and CO₂e Mitigation Potential in Spanish Publicly Funded Primary Schools","fulltext":[{"header":"Main","content":"\u003cp\u003eStrong scientific evidence shows that addressing climate change and urgent environmental issues requires transforming the global food system. This transformation must include shifting toward plant-rich diets, as the livestock sector is the primary contributor to environmental problems in the food system.\u003csup\u003e1\u003c/sup\u003e\u003csup\u003e2\u003c/sup\u003e\u003csup\u003e3\u003c/sup\u003e\u003csup\u003e4\u003c/sup\u003e\u003csup\u003e5\u003c/sup\u003e Failing to do so will cause us to miss the temperature limit targets of 1.5°C and 2°C,\u003csup\u003e6\u003c/sup\u003e and will result in a violation of the Paris Agreement.\u003csup\u003e7\u003c/sup\u003e Despite this, the livestock industry and meat consumption have been overlooked in environmental policies at both the national and international levels. \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAchieving the necessary transformation of the global food system will be complex and require involvement from a wide range of stakeholders, including individuals, local and national governments, companies, and international organizations. Regarding countries, they\u0026nbsp;are legally obliged to act – especially if they are signatories to the Paris Agreement – and effective actions rely mainly on political will. In this context, extensive literature\u003csup\u003e8\u003c/sup\u003e\u003csup\u003e9\u003c/sup\u003e\u003csup\u003e10\u003c/sup\u003e\u003csup\u003e11\u003c/sup\u003e indicates that countries can adopt initiatives across many areas to efficiently promote food system transformation, including subsidies, public procurement, taxes, dietary guidelines, and regulations.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThis article aims to understand the extent to which current school meal menus in Spanish publicly funded primary schools align with sustainability and low-emissions objectives. By focusing on publicly funded schools, the study examines a segment of the food system where public authorities have significant influence over food provisioning. Therefore, while the study does not directly assess specific regulatory mechanisms, it provides empirical evidence of the sustainability outcomes currently observed in Spanish publicly funded schools, indicating the potential of regulation-based interventions, such as public procurement.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn the global context of meat consumption, developed countries generally consume significantly more than the recommended amount for human and planetary health, which is about 26 kg per person per year.\u003csup\u003e4\u003c/sup\u003e\u003csup\u003e5\u003c/sup\u003e While the world average is 44.5 kg\u003csup\u003e12\u003c/sup\u003e per person annually – already well above the recommended amount – the European Union’s consumption reaches 66 kg\u003csup\u003e1\u003c/sup\u003e\u003csup\u003e3\u003c/sup\u003e per person per year.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWhen it comes to Spain, the country ranks as the top meat consumer in Europe, with an average of 94.04 kg per person per year.\u003csup\u003e14\u003c/sup\u003e Despite recent progress in food policy and school meal legislation, there remains significant potential to reduce meat consumption. Spain’s 2025 National Food Strategy (ENA)\u003csup\u003e15\u003c/sup\u003e, for example, does not consider meat reduction as part of its overall sustainability goal or as a factor in public procurement (p. 230). The country also issued a new decree\u003csup\u003e16\u003c/sup\u003e to promote healthy and sustainable school meals. On the one hand, it recognizes parents' right to request 100% plant-based menus, promotes the inclusion of plant foods rich in protein, and limits red meat consumption to 1 serving per week. On the other hand, the decree still allows schools to serve six portions of all meat (red, poultry, and fish) in a five-day week (Article 9), exceeding health and planetary limits\u003csup\u003e4\u003c/sup\u003e\u003csup\u003e5\u003c/sup\u003e. Moreover, it is unclear about the dishes where meat is present but isn’t the main ingredient, which are very common in Spanish school meals.\u003c/p\u003e\n\u003cp\u003eIn Spain, there are 2.715.544 students\u003csup\u003e17\u003c/sup\u003e enrolled in primary school, of which 47.4%\u003csup\u003e18\u003c/sup\u003e use the canteen services. 96%\u003csup\u003e13\u003c/sup\u003e (p. 6) of primary school students attend a public school or a state-funded private school (\u003cem\u003ecolegio concertado\u003c/em\u003e hereafter), which are the focus of this study, as they represent a food environment largely shaped by public funding and regulation and are therefore particularly relevant for sustainability-oriented policy discussions.\u003c/p\u003e\n\u003cp\u003eConsidering the context described earlier, this article aims to evaluate the sustainability of menus randomly collected from primary Spanish schools and estimate their CO2e mitigation potential. Our hypothesis is that most Spanish primary school canteens serve menus with unsustainable amounts of meat and that there is a significant potential to reduce CO2e emissions.\u0026nbsp;The study employed a quantitative cross-sectional approach with time-stratified sampling. To assess sustainability, our primary reference was the Planetary Health Diet (EAT-Lancet Reports of 2019 and 2025)\u003csup\u003e4\u003c/sup\u003e\u003csup\u003e5\u003c/sup\u003e. To calculate the CO2e emissions of each school and then extrapolate them to estimate national emissions and mitigation potential, our primary reference was Poore and Nemecek.\u003csup\u003e19\u003c/sup\u003e The detailed methodology is described in the Methods section. \u0026nbsp;\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003e3.1 General Characteristics and Sustainability:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003cul class=\"decimal_type\"\u003e\n \u003cli\u003eIn 77.3% of schools (126), menus did not include any fully plant-based day, meaning that foods of animal origin were served every day. Accordingly, 22.7% of the schools (37) offered a plant-based menu at least once a month. \u0026nbsp;\u003c/li\u003e\n \u003cli\u003eContrary to our expectations, 89% of schools (145) offered a vegetarian menu at least once per month, with the maximum being 8 days. 11% of the schools (18) have served meat every day.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003e79.1% of schools (129) have served more than 71.5 grams of meat per student daily, making their menus unsustainable according to the EAT-Lancet\u0026rsquo;s meat consumption recommendations (100 g weekly: 1 of beef or pork, 2 of fish, and 2 of poultry). Meanwhile, 20.9% of schools (34) have offered sustainable menus.\u003c/li\u003e\n \u003cli\u003eAll the schools that served meat every day (18) have been classified as unsustainable, as their daily meat intake exceeded 71.5 grams per student.\u003c/li\u003e\n \u003cli\u003e90.2% of schools (147) exceeded the limit for beef and pork. 61 of them \u0026ndash; 37.4% of schools \u0026ndash; offered double or more than double the recommended limit of beef and pork.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eFewer schools exceeded the limit for chicken and fish, totaling 49.7% (81 schools). Moreover, the excess was usually not high, with many schools (30) exceeding the 57-gram daily limit by less than 3 grams. \u0026nbsp;\u003c/li\u003e\n \u003cli\u003eThe most sustainable school \u0026ndash; with the lower meat intake \u0026ndash; consumed an average of 47.6 grams of meat per student per day.\u003c/li\u003e\n \u003cli\u003eThe least sustainable school \u0026ndash; with the largest meat intake \u0026ndash; consumed an average of 105.4 grams of meat per student per day, which is more than twice the amount of the most sustainable school and approximately 47.4% above the sustainability benchmark of 71.5 grams. \u0026nbsp;\u003c/li\u003e\n \u003cli\u003eOut of the 17 Spanish autonomous communities, 5 had all their schools serving unsustainable menus: Balearic Islands, Cantabria, Castile and Le\u0026oacute;n, Extremadura, and La Rioja. The other 12 autonomous communities had schools serving both sustainable and unsustainable menus. \u0026nbsp;\u003c/li\u003e\n \u003cli\u003ePublic Schools x \u003cem\u003eColegios Concertados\u003c/em\u003e: They display very similar profiles regarding meat provision and menu sustainability. 80.2% of public schools (73) have unsustainable menus, while 19.8% (18) have sustainable menus. 77.8% of \u003cem\u003ecolegios concertados\u003c/em\u003e (56) have unsustainable menus, while 22.2% (16) have sustainable menus.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eMajor Caterers x Local/regional Caterers: The proportion of schools classified as sustainable was higher among those whose menus were provided by major catering companies. 71.6% of schools using major catering services (48) have unsustainable menus, while 28.4% (19) have sustainable menus. 84.4% of schools that do not use a major catering service (81) have unsustainable menus, while 15,6% (15) have sustainable menus. \u0026nbsp;\u003c/li\u003e\n \u003cli\u003eAlthough one might expect that winter menus would feature more red meats and be less sustainable, while spring menus would be lighter in red meats and more sustainable, this was not observed in the collected sample. There were 6 menus from December, January and February, and 3 were sustainable while 3 were unsustainable. Concerning the months of April, May and June, there were 7 menus. Of those, 5 were unsustainable and 2 were sustainable. We observed that meat is frequently featured on menus throughout the school year, regardless of the month or season. \u0026nbsp;\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003e\u003cstrong\u003e3.2 CO2e mitigation potential\u003c/strong\u003e\u003c/p\u003e\n\u003cul class=\"decimal_type\"\u003e\n \u003cli\u003eThe average excess emission of the analyzed sample is 103.28 kg CO2e per student per year.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003e87.7% of the schools (143) emitted more CO2e than the proposed benchmark of 1226 grams (1.2 kg) of CO2e per day per student. 78 were public schools, while 65 were \u003cem\u003ecolegios concertados\u003c/em\u003e. Accordingly, 12.3% of the schools (20) emitted less CO2e than the proposed benchmark, being 12 public schools and 8 \u003cem\u003ecolegios concertados\u003c/em\u003e. Regarding catering companies, 11 of the 20 menus within the emission limit were from major catering companies, while 9 weren\u0026rsquo;t.\u003c/li\u003e\n \u003cli\u003eContrary to what might be inferred from common sense, 50% of the schools (10) that emitted CO₂e below the proposed benchmark had previously been classified as unsustainable. However, compared to the schools that were both sustainable and emitted less than the proposed benchmark, they performed worse in terms of potential CO2e emissions mitigation. While the unsustainable schools that emitted less than the benchmark saved an average of 14.8 kg of CO2e per student per year, the sustainable schools saved an average of 54 kg of CO2e per student per year.\u003c/li\u003e\n \u003cli\u003eOut of 34 schools classified as sustainable, 24 emitted more CO2e than the proposed benchmark.\u003c/li\u003e\n \u003cli\u003eThe school with the best results saved 84.11 kg of CO2e per student annually, while the average emission reduction from schools that emitted below the proposed benchmark was 34.42 kg of CO2e per student annually.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eThe school with the worst results exceeded the proposed benchmark by 347.15 kg of CO2e per student each year. \u0026nbsp;\u003c/li\u003e\n \u003cli\u003eBy scaling up sample-based excess emission estimates, we have obtained the following country-level results: considering 1.235.680 primary students using catering services in public schools and \u003cem\u003ecolegios concertados\u003c/em\u003e,\u003csup\u003e13\u003c/sup\u003e the average total excess emission is \u0026asymp; 122.583 tons of CO₂e/year.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eIf schools with high-emission menus matched the average performance of low-emission schools, the estimated reduction in emissions would be about 137.7 kg CO₂e per student annually. Matching the most efficient observed menu would increase this potential to nearly 187.4 kg CO₂e per student per year. At scale, this would represent an annual mitigation potential of roughly 169.800 tons of CO₂e under the average low-emission scenario, and up to 231.000 tons if the best-performing menus were adopted.\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003e\u003cstrong\u003e3.3 Exploratory findings on early regulatory implementation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn addition to the main results, an exploratory empirical observation emerged from the data. One of the catering companies included in the sample, Aramark, formally announced its intention to begin implementing the new Spanish Decree on healthier, more sustainable school meals as of October 2025, despite the regulation becoming mandatory only in April 2026.\u003c/p\u003e\n\u003cp\u003eOf the 13 menus collected from this provider, two correspond to the period prior to October 2025, while 11 were collected from October 2025 onwards. Among these eleven menus, five were classified as sustainable according to the sustainability criteria applied in this study, while the remaining six were classified as unsustainable. Moreover, only three of the eleven menus met or fell below the proposed CO₂e emission benchmark.\u003c/p\u003e\n\u003cp\u003eAlthough based on a very limited sample and not a primary objective of this study, this observation suggests that the early implementation of the new decree does not automatically translate into sustainable menus or compliance with the proposed emission benchmark.\u0026nbsp;\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe results confirm our hypothesis that most menus in Spanish primary schools are unsustainable (79.1%) and emit more CO2e than the proposed benchmark (87.7%), indicating substantial room to reduce CO2e emissions.\u003c/p\u003e\n\u003cp\u003eBy analyzing the data, we could not identify any specific factor that determined good and bad results. There were very good performers in terms of sustainability and CO2e emissions in both public schools and \u003cem\u003ecolegios concertados\u003c/em\u003e. The same is true for the worst performers. In addition, while schools served by major catering companies performed generally better on sustainability, all companies except Seral offered both very sustainable and very unsustainable menus, indicating no consistent pattern even within the same catering company. Possible explanations for the absence of a clear pattern include schools' autonomy in selecting suppliers and defining the menu profiles served, as well as the lack of regulatory definitions of food types and amounts of unsustainable foods that can be served.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe Spanish Decree on healthy and sustainable school meals has not yet entered into force. However, we emphasize that the Spanish decree, although undeniably positive, is not as strict as our main reference (EAT-Lancet’s Planetary Health Diet) regarding meat portions. Additionally, it also lacks clarity on several important topics, which may explain why the catering provider that has already started complying with it still offers meals with widely varying sustainability standards.\u003c/p\u003e\n\u003cp\u003eFinally, the main takeaway of this study is that the solution to the sustainability problem in Spanish school canteens has already been adopted by 20% of the schools analyzed. Regarding CO2e emissions, 12% of the schools already have a fairly reasonable emissions profile. Therefore, there is no need for disruptive measures or unknown abrupt changes. Regulating the sector to align with the sustainable standards already in place in many schools across the 17 Spanish autonomous communities should not, thus, entail major cultural changes nor lead to social or political unrest.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe emphasize that, although our sample shows that a minority of Spanish autonomous communities only had schools with unsustainable menus, this should not be interpreted as evidence that sustainable menus are absent in those regions. Given the non-exhaustive nature of the sample and the known heterogeneity in school food provision within autonomous communities, it is highly likely that schools offering sustainable menus exist in these regions but were not captured in the present dataset. Therefore, the results should be understood as reflecting observed practices within the sampled schools rather than as a regional assessment.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAccording to a recent study\u003csup\u003e20\u003c/sup\u003e in 171 countries, purchasing locally available, cheaper products from each of the main food groups generally leads to a healthy, low-emission diet. This indicates that a planet-friendly, healthy diet would generally not require significant lifestyle changes or expensive substitutes. Our results suggest this might also be true for Spanish school menus, which could improve their current poor sustainability and emissions profile by reducing meat consumption and increasing the use of healthy, sustainable ingredients already present in every analyzed menu. We emphasize that all the menus examined, including the unsustainable ones, featured dishes made with nutritious, relatively low-cost ingredients traditional to Spanish cuisine, such as chickpeas, lentils, beans, and vegetables. Moreover, we highlight that five of the six major catering companies, which account for nearly 60% of the market share, already know how to provide highly sustainable, low-emission menus, as evidenced by the sample. In this context, the policy component can help ensure that best practices become the norm rather than the exception, as they are now. Ultimately, the results suggest that implementing sustainable menus in the Spanish educational system is operational and feasible within the existing institutional and political context. \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.1.\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp; Limitations:\u003c/strong\u003e\u003c/p\u003e\n\u003ch4\u003ea. \u003cem\u003eLimitations potentially leading to an overestimation of unsustainability\u0026nbsp;\u003c/em\u003e\u003c/h4\u003e\n\u003cp\u003e- Although we are aware of special menus for allergies or religious restrictions, we used the regular menu as a reference for the calculations. We acknowledge that this could affect some of the results, but we believe the differences would be minor for the following reasons: 1. There is usually no special menu for allergies related to beef, pork, or chicken; 2. A restriction concerning an animal-protein food is likely to be replaced by another animal-protein food, as observed in some schools in the sample; 3. Estimates suggest that only about 2.8%\u003csup\u003e21\u003c/sup\u003e of the menus are for all special diets (16 categories, including allergies and religious/ethical reasons); 4. Vegetarian and vegan menus, which would have the biggest impact on the analysis, are not available at every school and would represent only a small fraction of the 2.8% of special diets; 5. Lastly, the total sample size would be much smaller because most schools only publish the regular menus online, not the special ones. \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e- Since the menus were collected randomly, only 41.1% came from schools served by large catering companies. Statistics estimate that up to 58%\u003csup\u003e2\u003c/sup\u003e\u003csup\u003e2\u003c/sup\u003e of schools are served by market-dominating companies, and when comparing the two groups – schools served and not served by these companies – it was found that menus provided by large companies were more often sustainable than those not managed by them (28.4% of schools managed by large companies had sustainable menus, versus 15.6% of schools not managed by large companies). Therefore, if the sampled schools truly reflected the proportion served by large catering companies, there might have been slightly more sustainable schools.\u003c/p\u003e\n\u003ch4\u003e\u003cem\u003eb.\u0026nbsp; \u0026nbsp;\u003c/em\u003e\u003cem\u003eLimitations potentially leading to an underestimation of the unsustainability\u003c/em\u003e\u003c/h4\u003e\n\u003cp\u003e- We adopted a very cautious approach when interpreting dishes and always assumed they didn’t contain meat when in doubt. For example, we analyzed a selected school’s regular menu and then compared it with its vegetarian menu. The regular menu had some dishes whose names didn’t suggest they contained meat, so we assumed they didn’t. However, the vegetarian menu had many dishes with identical names, some labeled “meatless” or “with plant protein”, which led us to believe that the regular menu featured meat in those dishes. As a result, we probably underestimated the amount of meat in the overall sample, but it was a deliberate choice aimed at minimizing false positives and avoiding over-attribution of meat content in the absence of explicit information. This approach reduces the risk of systematic overestimation and makes any subsequent findings regarding meat presence more, rather than less, robust.\u003c/p\u003e\n\u003cp\u003e- We only analyzed primary schools, but potential regulations would likely apply to the entire mandatory education system, including early childhood education and secondary school. Early childhood education in Spain has about 1.576.090\u003csup\u003e17\u003c/sup\u003e students and significant use of catering services, while secondary schools have over 2 million students\u003csup\u003e17\u003c/sup\u003e, but catering services are more common only in \u003cem\u003ecolegios concertados\u003c/em\u003e and private schools. Therefore, the potential for sustainability and CO2e emissions is certainly greater for the entire Spanish education system than what this paper predicts, which only includes primary education. \u0026nbsp; \u0026nbsp;\u0026nbsp;\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003e5.1.\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;General\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ea. \u003cstrong\u003ePrimary data source for this study:\u003c/strong\u003e school menus obtained from primary public schools and \u003cem\u003ecolegios concertados\u003c/em\u003e in Spain. Both types of schools account for about 96% of students enrolled in Spain and are more likely to be affected by public policies than private schools since they are either public institutions or publicly funded. The sample included public schools and \u003cem\u003ecolegios concertados\u003c/em\u003e from all 17 Spanish autonomous communities. Regarding the education level, we decided to focus on primary education because it includes the largest number of students who use school meal services in both public schools and \u003cem\u003ecolegios concertados\u003c/em\u003e. In secondary school, catering services are common in \u003cem\u003ecolegios concertados\u003c/em\u003e, but not in public schools, where attendance rates are quite low. In total, we have collected 163 menus. 91 were from public schools and 72 from \u003cem\u003ecolegios concertados\u003c/em\u003e. Regarding food service management, 67 menus were provided by one of the major catering companies in Spain, while 96 were not.\u003c/p\u003e\n\u003cp\u003eb. \u003cstrong\u003eSample Size:\u003c/strong\u003e Although the quantitative analysis is based on a corpus of 163 school menus \u0026ndash; each collected, archived, and included in the supplementary materials \u0026ndash; this figure should be understood as a conservative unit of analysis. A significant number of these menus were standardized, caterer-provided menus explicitly identified as applicable to multiple schools within a city or region (e.g., \u0026ldquo;menus for schools in Toledo\u0026rdquo; or \u0026ldquo;central kitchen services\u0026rdquo;). As a result, while the calculations consistently rely on the 163 menus for reasons of transparency and reproducibility, the empirical material likely reflects food provision practices affecting a substantially larger number of schools.\u003c/p\u003e\n\u003cp\u003ec. \u003cstrong\u003eMenu Collection Procedures:\u003c/strong\u003e To assess the sustainability and carbon footprint of Spanish school menus, we needed to gather actual menus. However, these menus are not available through a centralized list or website. Most of the time, they are not publicly accessible on schools\u0026rsquo; websites and are shared only with families via private channels. Because of this, we first tried to contact the major catering companies, which serve up to 58% of the school meals\u003csup\u003e21\u003c/sup\u003e, and asked them for menu samples. Since they either ignored or declined our messages, and it was impossible to research the websites of over 13.000 primary schools individually, we used a Large Language Model (LLM) \u0026ndash; ChatGPT 5.1 \u0026ndash; to identify schools that publicly displayed their menus online. We asked the LLM to list public schools and \u003cem\u003ecolegios concertados\u003c/em\u003e with online menus in each autonomous community of Spain. We usually moved on to the next community when the LLM indicated it could find no more menus. We ensured we gathered more menus from the most populous communities (Andalusia, Madrid, Catalunya, and the Valencian Community) to improve accuracy and reduce potential bias in our results. Therefore, when the LLM didn\u0026rsquo;t list enough schools for a larger community, we changed the question to include specific cities within that community. By \u0026ldquo;not enough schools,\u0026rdquo; we mean the same number or fewer than the ones listed for less populated communities. We had to use this strategy for Andalusia and Catalunya. As a result, the menus were selected at random based on their online availability, and all data was collected directly from school websites, local government websites, or the online pages of catering services.\u003csup\u003e1\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003ed. \u003cstrong\u003eMenu exclusion criteria:\u003c/strong\u003e We used most of the menus found online and only excluded samples for the following reasons: 1. The menu was not from the calendar year 2025; 2. It was the third identical menu, despite being served in different schools. We observed that public schools in some cities and/or communities likely have contracts with a catering company that provides the same menu across all public schools we found. In these cases, we kept a maximum of 2 identical menus, which appeared to be representative of the community, and discarded the others to reduce potential bias in the results. We must highlight that in these communities or cities, we tried different research strategies with the LLM, but they all resulted in the same outcome: a list of schools with 100% identical menus. This situation with identical menus across all schools listed by the LLM did not occur with \u003cem\u003ecolegios concertados\u003c/em\u003e, only public ones. Identical menus from the same catering companies are common even in \u003cem\u003ecolegios concertados\u003c/em\u003e, but the full list always included schools with different menus and catering companies.\u003c/p\u003e\n\u003cp\u003ee. \u003cstrong\u003eMajor catering companies versus\u003c/strong\u003e \u003cstrong\u003esmaller or independent companies:\u0026nbsp;\u003c/strong\u003eInSpain, there is reportedly significant market concentration among school catering providers. The menus were collected randomly, and by the end of the collection, 67 of 163 menus were supplied by one of the major companies (Serunion, Aramark, Ausolan, Mediterranea, Scolarest, and Seral). In addition to analyzing the data as a whole, we also disaggregated it into two categories: public schools and \u003cem\u003ecolegios concertados\u003c/em\u003e, and major catering companies versus non-major companies. \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003ef. \u003cstrong\u003eStructure of school menus and rationale for using daily calculations:\u003c/strong\u003e Most menus are scheduled monthly. A few follow a trimester system or are cycle menus that rotate every six weeks, as is the case in the Murcia community. Regarding the monthly menus, most schools\u0026rsquo; websites had the October menu available, which was the current month at the time the research was conducted. Other websites provided more than one menu, including the October one, or a random menu not from October. Whenever menus for other months were accessible, we selected those to create the most diverse sample possible. Out of the 163 menus collected, 40 are not from October. Aside from July and August, which are vacation months, menus for all other months were analyzed. Since months and cycles have different lengths, we used the average school day as a reference instead of the month or cycle. Therefore, it didn\u0026rsquo;t matter if the menu had 15, 20, or 30 days. All the meat on each menu was accounted for and divided by the number of school days in the respective month or cycle, yielding an average amount per day. \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eg. \u003cstrong\u003eRationale for Focusing on Meat as the Primary Indicator:\u003c/strong\u003e Regarding the food, our main focus was to understand how much animal-based products were used and how they affected the sustainability of the menus. Initially, this included meat, dairy, and eggs. However, we noticed a discrepancy across the menus regarding the presence of eggs and dairy, which could have biased our results. Some menus listed all the allergens in each dish, while others did not. The count of eggs and dairy was generally much higher on menus that displayed allergens. As a result, we decided not to assess the sustainability of the menus based on these products. They were only considered to determine how many days featured plant-based meals. The calculations were based solely on the amount of meat present on the menus. Every time we use the word \u0026ldquo;meat\u0026rdquo;, it refers to all types of meat. \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eh. \u003cstrong\u003eApproach to Classifying Meat Content in Regional and Traditional Dishes:\u0026nbsp;\u003c/strong\u003eWhen interpreting dishes from Spain or its regions and cities, we researched their names on Google and reviewed the first five recipes to understand the ingredients. If there was a discrepancy between the recipes, and at least 2 of the five did not include meat, we took a cautious approach and assumed there was no meat in the dish. Additionally, when we could identify or infer that a dish traditionally made with meat was meatless, we did not count it as containing meat. The identification was made objectively, for example, when the menu described the dishes\u0026rsquo; ingredients. The inference occurred when we had strong clues suggesting the dish was meatless. For example, the dish \u0026ldquo;lentil \u003cem\u003eestofado\u003c/em\u003e\u0026rdquo; or \u0026ldquo;chickpea \u003cem\u003eestofado\u0026rdquo;\u0026nbsp;\u003c/em\u003eusually contains pork meat, and the name of the dish is exactly as mentioned (in Spanish: \u003cem\u003eestofado de lentejas/estofado de garbanzos).\u0026nbsp;\u003c/em\u003eWhen the dish was named \u0026ldquo;lentil/chickpea \u003cem\u003eestofado\u003c/em\u003e with vegetables,\u0026rdquo; we assumed it contained no meat, and the vegetables or any specific vegetable in the dish name replaced the pork. \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e5.2.\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Sustainability of the Menus\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe assessed the sustainability of the menus based on how much they exceeded the meat consumption recommended by the Planetary Health Diet (EAT-Lancet reports 2019 and 2025)\u003csup\u003e4\u003c/sup\u003e\u003csup\u003e5\u003c/sup\u003e. To determine the amount of meat on each menu, we used the Spanish Consensus Document on Food Provision in Educational Institutions\u003csup\u003e23\u003c/sup\u003e (hereafter, the Consensus Document). Below, we outline each step taken to obtain the results.\u003c/p\u003e\n\u003cp\u003e1.\u0026nbsp; \u0026nbsp;We counted how many dishes had beef, pork, poultry, fish, or any other type of meat. We grouped each type of meat into two categories: dishes featuring a full meat portion and dishes containing only a partial meat portion.\u003c/p\u003e\n\u003cp\u003e2.\u0026nbsp; \u0026nbsp;According to the Consensus Document mentioned above, for students aged 7 to 12 (primary school), full portions of beef, pork, and poultry ranged from 80 to 160 grams, while full portions of fish ranged from 100 to 120 grams (p. 15). In this article, we used 100 grams for beef, pork, and poultry, and 110 grams for fish. \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e3.\u0026nbsp; \u0026nbsp;Regarding the partial meat portions in dishes like paella and pasta, the Consensus document specifies a range of 20 to 30 grams. We used 25 grams for partial portions. However, such dishes often included two or three types of meat. In these cases, we used the 30-gram reference and divided it by the number of different meats. For example, in a paella made with pork, poultry, and fish, we considered 10 grams per type of meat. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e4.\u0026nbsp; \u0026nbsp;A small number of the menus (6 out of 163) featured rabbit meat. Since the EAT-Lancet reports don\u0026rsquo;t include any recommendations on this type of meat, we treated it similarly to poultry. In 5 of 6 cases where rabbit meat was offered, it was served in a partial portion. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e5. \u0026nbsp; To determine the daily amount of each type of meat, we multiplied the number of times each meat was served by its respective portion size (full or partial) and divided by the number of school days listed in the menu, which varied by month and school. Then, we grouped the meat types according to the EAT-Lancet report: beef with pork, and poultry with fish. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e6. \u0026nbsp; The daily recommended amounts of meat changed slightly between the 2019 and 2025 EAT-Lancet reports: from 14 to 15 grams for beef and pork, from 28 to 30 grams for fish, and from 29 to 30 grams for poultry. However, both reports refer to weekly portions of 100 grams\u003csup\u003e2\u003c/sup\u003e \u0026ndash; one portion for beef, pork, or lamb, and two portions for fish and poultry \u0026ndash; which match the daily amounts in the 2019 report (14, 28, and 29 grams, respectively). Therefore, we assumed that the 2025 report rounded the figures for daily servings, while keeping the weekly serving sizes unchanged (one or two of 100 grams, depending on the type of meat). Because of that, we used daily reference values of 14.5 grams for beef and pork and 28 and 29 grams for fish and poultry, respectively. Accordingly, any menu exceeding 71.5 grams per day (14.5 + 28 + 29) was classified as unsustainable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e5.3. \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Calculating the CO2e mitigation potential of Spanish primary school menus\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eClimate change and CO2e emissions are just one of the nine planetary boundaries that determine sustainability. Therefore, the Planetary Health Diet (2019 and 2025)\u003csup\u003e4\u003c/sup\u003e\u003csup\u003e5\u003c/sup\u003e, which established recommendations for meat consumption and was used to determine the menu\u0026rsquo;s sustainability, is based on a much more complex approach that includes all nine planetary boundaries and human health indicators. Nevertheless, given the urgent need to reduce CO2e emissions and address climate change, we find it useful to understand the emissions of school menus and their mitigation potential. Since some factors influencing CO2e emissions differ from those affecting sustainability, we calculated emissions for all collected menus, regardless of whether they were classified as sustainable or unsustainable. We chose to do so because different criteria for determining sustainability and emissions could lead to a school being labeled sustainable while emitting CO2e above the limit, and vice versa. Therefore, in this section, we used indicators from the Planetary Health Diet when pertinent, but we also included others that were exclusively pertinent to climate change and CO2e calculations. \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo estimate the CO2e mitigation potential of the menus, we primarily relied on Poore and Nemecek.\u003csup\u003e19\u003c/sup\u003e However, in Poore and Nemecek, the metric used is kg CO2e per 100 grams of protein, whereas the Consensus Document\u003csup\u003e23\u003c/sup\u003e and the Planetary Health Diet\u003csup\u003e4\u003c/sup\u003e\u003csup\u003e5\u003c/sup\u003e use grams of food type rather than protein. Therefore, we used the numbers converted to kg CO2e per kg of food, featured on the Our World in Data\u003csup\u003e24\u003c/sup\u003e website, which sources its data from Poore and Nemecek. We outline below each step taken to obtain the results: \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e1.\u0026nbsp; \u0026nbsp;We initially calculated the emission limit for a sustainable school menu. To do this, we considered the sustainability limit of 71.5 grams of meat per student daily, consisting of 29 grams of poultry, 28 grams of fish, and 14.5 grams of beef or pork. The Planetary Health Diet does not specify how the 14.5 grams should be divided between pork and beef. While both meats have a similar impact on many planetary boundaries and human health, this is not true regarding CO2e emissions. Beef produces an average of 7.8 times more CO2e than pork. The impact of beef on CO2e emissions and climate change is well documented, as is the need to reduce its consumption more than that of other meats. Therefore, to divide the 14.5-gram portion, we used a ratio of 3/7, meaning three parts beef for every seven parts pork. As a result, the daily limit for beef was 4.35 grams, while for pork it was 10.15 grams. \u0026nbsp; \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e2.\u0026nbsp; \u0026nbsp;With the portions for each type of meat determined, we multiplied them by their respective CO2e emissions: 99.48 kg CO2e per kg of beef; 12.31 kg CO2e per kg of pork; 9.87 kg CO2e per kg of poultry; and 13.63 kg CO2e per kg of fish. We then found that the daily emission limit for a sustainable menu (71.5 grams of meat) was 1226 grams, or around 1.2 kg, of CO2e per day per student \u0026ndash; the sustainable menu benchmark.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e3.\u0026nbsp; \u0026nbsp;Later, we calculated the proportion attributable to each meat type for each school. For example, at school N.\u0026deg; 1, beef consumption was 175 grams per student per month, while pork consumption was 625 grams per student per month. This shows that pork consumption was roughly 3.57 times higher than beef consumption. The average daily consumption of both meats per student was 36.3 grams. Based on the 3.57 ratio, the average daily intake of beef was about 7.94 grams per student, and pork was approximately 28.36 grams per student. We repeated this process for each school in both groups (pork/beef and poultry/fish).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e4.\u0026nbsp; \u0026nbsp;As we identified the consumption of each type of meat, we multiplied each by its respective CO2e emissions: 99.48 kg CO2e per kg of beef; 12.31 kg CO2e per kg of pork; 9.87 kg CO2e per kg of poultry; and 13.63 kg CO2e per kg of fish. Then, we summed all the values to determine the total emissions. For example, school n\u0026deg; 20 had the following average daily consumption (in grams) per student of beef, pork, poultry, and fish, respectively: 5.94 g, 15.46 g, 26.17 g, and 47.83 g. The calculations were: (5.94 \u0026nbsp;x 99.48) + (15.46 x 12.31) + (26.17 x 9.87) + (47.83 x 13.63) \u0026asymp;\u0026nbsp;1691.44 g CO2e.\u0026nbsp;After calculating the total daily emissions per student for each school, we subtracted the sustainable menu benchmark (1226 g) from it to determine whether there was an excess of emissions. In the case of school N.\u0026deg; 20, the excess was 465.44 g CO2e. Finally, we multiplied the excess by 175 (the number of school days in a year) to determine the annual emissions exceedance relative to the benchmark. For school N.\u0026deg; 20, it was 81.452 g or 81.4 kg of CO2e per student per school year. \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e5.\u0026nbsp; \u0026nbsp;We repeated this process for all schools in the sample. Those below the sustainable menu benchmark had negative numbers, indicating they emitted less than the maximum set as ideal.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e6.\u0026nbsp; \u0026nbsp;Then, by summing the results of all schools, including those with negative emissions, and dividing by the total number of schools analyzed (163), we calculated the average excess emissions across the entire sample.\u003c/p\u003e\n\u003cp\u003e7.\u0026nbsp; \u0026nbsp;Next, we calculated national estimates using an extrapolation method in which the average annual excess emissions per student at the sample level were multiplied by the estimated number of primary school students enrolled in public schools and \u003cem\u003ecolegios concertados\u003c/em\u003e who use the school canteen service in Spain. This figure was determined by combining official statistics\u003csup\u003e17\u003c/sup\u003e on total primary enrollment (2,707,302 students), percentage of students enrolled in public schools and \u003cem\u003ecolegios concertados\u003c/em\u003e (96% of the total), with the reported proportion of canteen users (47.4%). Because statistics on school meal service usage are not disaggregated by school type, we assumed a similar rate of use across school types.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e8.\u0026nbsp; \u0026nbsp;Finally, we tested two counterfactual scenarios by estimating: (1) the mitigation potential if all schools performed at the level of the average school whose emissions were below the benchmark; and (2) the mitigation potential if all schools performed at the level of the most efficient school in the sample. These estimates assume full convergence of high-emission menus toward lower-emission benchmarks and homogeneous student participation in school meal services.\u003c/p\u003e\n\u003cdiv id=\"ftn1\"\u003e\n \u003cp\u003e\u003csup\u003e1\u003c/sup\u003e Before resorting to AI, the author attempted to collect menus by messaging Facebook groups of expatriates living in Spain and reaching out to acquaintances who have children enrolled in Spanish schools. After it only returned 5 menus, we tried the AI approach. Since the menus obtained through this first approach fitted the criteria, they were included in the analysis and comprise the sample of 163 menus.\u0026nbsp;\u003c/p\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003csup\u003e2\u003c/sup\u003e P. 52 EAT-Lancet 2019\u003csup\u003e4\u003c/sup\u003e; pp. 1632 and 1635 EAT-Lancet 2025\u003csup\u003e5\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eYanxian Li et al., \u0026lsquo;Reducing Climate Change Impacts from the Global Food System Through Diet Shifts\u0026rsquo; (2024) 14 Nat Clim Chang 943\u0026ndash;53. https://doi.org/10.1038/s41558-024-02084-1.\u003c/li\u003e\n\u003cli\u003eMarco Springmann et al., \u0026lsquo;Options for Keeping the Food System Within Environmental Limits\u0026rsquo; (2018) 562 Nature 519\u0026ndash;25.\u003c/li\u003e\n\u003cli\u003eIPCC, \u0026lsquo;Climate Change and Land: an IPCC Special Report on Climate Change, Desertification, Land Degradation, Sustainable Land Management, Food Security, and Greenhouse Gas Fluxes in Terrestrial Ecosystems\u0026rsquo; (2019). https://www.ipcc.ch/site/assets/uploads/sites/4/2022/11/SRCCL_Full_Report.pdf.\u003c/li\u003e\n\u003cli\u003eW. Willett et al., \u0026lsquo;Food in the Anthropocene: the EAT-Lancet Commission on Healthy Diets from Sustainable Food Systems\u0026rsquo; (2019) 393 Lancet 447\u0026ndash;92. P. 52 https://doi.org/10.1016/S0140-6736(18)31788-4.\u003c/li\u003e\n\u003cli\u003eRockstr\u0026ouml;m, Johan et al., The EAT\u0026ndash;\u003cem\u003eLancet\u003c/em\u003e Commission on healthy, sustainable, and just food systems, (2025) The Lancet, Volume 406, Issue 10512, 1625 \u0026ndash; 1700. Pp. 1632 and 1635. 10.1016/S0140-6736(25)01201-2\u003c/li\u003e\n\u003cli\u003eMichael A. Clark et al., \u0026lsquo;Global Food System Emissions Could Preclude Achieving the 1.5\u0026deg; and 2\u0026deg;C Climate Change Targets\u0026rsquo; (2020) 370 Science 705\u0026ndash;8. https://doi.org/10.1126/science.aba7357.\u003c/li\u003e\n\u003cli\u003eAuthor, 2024. \u003c/li\u003e\n\u003cli\u003eSimona Vallone \u0026amp; Eric F. Lambin, \u003cem\u003ePublic Policies and Vested Interests Preserve the Animal Farming Status Quo at the Expense of Animal Products Analog\u003c/em\u003e, 6 One Earth 1213, 1213\u0026ndash;26 (2023).\u003c/li\u003e\n\u003cli\u003eSwensson, Luana F J et al., Public food procurement as a game changer for food system transformation, The Lancet Planetary Health, Volume 5, Issue 8, e495 - e496\u003c/li\u003e\n\u003cli\u003eLuana.F.J. Swensson; Florence Tartanac. Public food procurement for sustainable diets and food systems: The role of the regulatory framework, Global Food Security Volume 25, June 2020, 100366, https://doi.org/10.1016/j.gfs.2020.100366\u003c/li\u003e\n\u003cli\u003eSpringmann, M., Dinivitzer, E., Freund, F. \u003cem\u003eet al.\u003c/em\u003e A reform of value-added taxes on foods can have health, environmental and economic benefits in Europe. \u003cem\u003eNat Food\u003c/em\u003e 6, 161\u0026ndash;169 (2025). https://doi.org/10.1038/s43016-024-01097-5\u003c/li\u003e\n\u003cli\u003eOur World in Data, Per capita meat and fish consumption, 2022, https://ourworldindata.org/grapher/per-capita-meat-type#:~:text=Per%20capita%20meat%20and%20fish,/meat%2Dproduction%20%7C%20CC%20BY \u003c/li\u003e\n\u003cli\u003eOur World in Data, Per capita meat consumption in the EU28, https://ourworldindata.org/grapher/per-capita-meat-eu28 \u003c/li\u003e\n\u003cli\u003eEuromeat, Spain and Austria have the highest rate of meat consumption in Europe (Feb, 2025) https://www.euromeatnews.com/Article-Spain-and-Austria-have-the-highest-rate-of-meat-consumption-in-Europe/2558#:~:text=Spaniards%20are%20leading%20the%20top,according%20to%20a%20UN%20chart \u003c/li\u003e\n\u003cli\u003eSpain, Estrategia Nacional de Alimentaci\u0026oacute;n (2025), p. 230.\u003c/li\u003e\n\u003cli\u003eSpain, Real Decreto 315/2025, de 15 de abril, por el que se establecen normas de desarrollo de la Ley 17/2011, de 5 de julio, de seguridad alimentaria y nutrici\u0026oacute;n, para el fomento de una alimentaci\u0026oacute;n saludable y sostenible en centros educativos, Article 9. https://www.boe.es/eli/es/rd/2025/04/15/315 \u003c/li\u003e\n\u003cli\u003eSpain, Datos y Cifras: Curso Escolar 2024-2025. https://www.lamoncloa.gob.es/serviciosdeprensa/notasprensa/educacion-fp-deportes/Documents/2024/110924-%20datos-y-cifras-curso-escolar-2024-2025.pdf?utm_source=chatgpt.com\u003c/li\u003e\n\u003cli\u003eSTECyL-i, La mitad del alumnado de primaria accede al comedor escolar, pero 1,3 millones de ni\u0026ntilde;os y ni\u0026ntilde;as en pobreza siguen sin beca (Sep 2025), https://stecyl.net/la-mitad-del-alumnado-de-primaria-accede-al-comedor-escolar-pero-13-millones-de-ninos-y-ninas-en-pobreza-siguen-sin-beca/ \u003c/li\u003e\n\u003cli\u003eJoseph Poore \u0026amp; Thomas Nemecek, \u003cem\u003eReducing Food\u0026rsquo;s Environmental Impacts Through Producers and Consumers\u003c/em\u003e, 360 Sci. 987, 991 (2018).\u003c/li\u003e\n\u003cli\u003eBai, Y., Martinez, E.M., Yamanaka, M. \u003cem\u003eet al.\u003c/em\u003e Environmental impacts and monetary costs of healthy diets worldwide. \u003cem\u003eNat Food\u003c/em\u003e 6, 1176\u0026ndash;1185 (2025). https://doi.org/10.1038/s43016-025-01270-4\u003c/li\u003e\n\u003cli\u003eJ. R. Mart\u0026iacute;nez \u0026Aacute;lvarez et al.,National survey on school lunches and demand of special diets, Nutr. Hosp. vol.27 no.1 Madrid ene./feb. 2012. https://scielo.isciii.es/scielo.php?pid=S0212-16112012000100032\u0026amp;script=sci_arttext\u0026amp;utm_source=chatgpt.com\u003c/li\u003e\n\u003cli\u003eDel Campo al Cole, Los Comedores Escolares en Espa\u0026ntilde;a (2018). https://www.observatoriodelainfancia.es/ficherosoia/documentos/5587_d_Informe-Comedores-Escolares.pdf?utm_source=chatgpt.com \u003c/li\u003e\n\u003cli\u003eSpain. Documento de Consenso Sobre la Alimentaci\u0026oacute;n em los Centros Educativos (2010). \u003c/li\u003e\n\u003cli\u003eOur World in Data, Greenhouse gas emissions per kilogram of food product, https://ourworldindata.org/explorers/food-footprints?Commodity+or+Specific+Food+Product=Commodity\u0026amp;Environmental+Impact=Carbon+footprint\u0026amp;Kilogram+%2F+Protein+%2F+Calories=Per+kilogram\u0026amp;By+stage+of+supply+chain=false\u0026amp;country=Almonds~Bacon~Bananas~Beans~Beef+%28beef+herd%29~Beef+%28dairy+herd%29~Beefburger~Cheese~Cow%27s+milk~Eggs~Lamb+%26+Mutton~Maize~Milk~Peas~Penne+pasta~Pig+Meat~Pizza~Poultry+Meat~Prawns+%28farmed%29~Rice~Steak+pie~Tofu~Tomatoes~Vegetable+lasagne~Wheat+%26+Rye \u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"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":"","lastPublishedDoi":"10.21203/rs.3.rs-8723135/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8723135/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Shifting the global food system away from overconsuming animal-based foods and toward plant-rich diets is crucial to tackling the world’s most urgent environmental problems, including climate change. National governments have a key role to play in this process, as public institutions like schools and hospitals are significant food consumers. This article aims to understand the extent to which current school meal menus in Spanish publicly funded primary schools align with sustainability and low-emissions standards. We analyzed 163 primary school menus from across all 17 Spanish autonomous communities. Among the results, we found that 79.1% of schools had unsustainable menus and that 87.7% emitted more CO2e than the proposed benchmark, suggesting considerable potential for policy actions.","manuscriptTitle":"Greening School Meals: Sustainability Analysis and CO₂e Mitigation Potential in Spanish Publicly Funded Primary Schools","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-05 05:45:29","doi":"10.21203/rs.3.rs-8723135/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":"3c2af337-6e01-4f58-876b-d72cfbcae60a","owner":[],"postedDate":"February 5th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":61913384,"name":"Earth and environmental sciences/Environmental social sciences/Sustainability"},{"id":61913385,"name":"Scientific community and society/Social sciences/Climate change/Climate-change mitigation"},{"id":61913386,"name":"Earth and environmental sciences/Environmental social sciences/Climate-change policy"}],"tags":[],"updatedAt":"2026-03-03T06:00:42+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-05 05:45:29","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8723135","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8723135","identity":"rs-8723135","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}