Carbon footprint of school lunch menus adhering to the Spanish dietary guidelines

Abstract

Current eating patterns are damaging the environment and there is a need for dietary guidelines to address this issue. Schools can act as springboards to promote sustainable food habits. However, unlike other European national dietary guidelines, the Spanish school dietary guidelines do not include environmental sustainability guidance. In this study, a life cycle assessment (LCA) approach was used to quantify the carbon footprint (CF) of school menus designed following the Spanish school dietary guidelines and adjusting to approximately 610 kcal/lunch-meal. To do so, the menus assessed were the baseline menu and six alternative menus served in schools (the menu without dairy and without legumes, the menu without meat, the menu without fish, the menu without eggs, the hypocaloric menu and the astringent menu designed to address intestinal illnesses). The CF of the baseline menu was 24.39 kg CO₂ eq/person/monthly lunch-meal which resembled, to some extent, the CF of the Mediterranean diet obtained for adults and approximately 700 kcal/lunch-meal. Regarding the alternative menus, the astringent menu and the menu without meat presented the lowest CF, being 14.77 kg CO₂ eq/person/month and 17.11 kg CO₂ eq/person/month, respectively. It is recommended that Spanish dietary guidelines include the environmental performance of foods to optimize school menus and favor a gradual transition towards a more environmentally-friendly diet.

Keywords:

• Carbon emissions
• eating patterns
• dietary guidelines
• meat
• sustainability
• school menu

Introduction

Nutrition is essential for human beings, and providing an adequate diet is a cornerstone of good health and well-being [1]. The world population has been estimated to reach 9.1 billion by 2050 and meeting the needs of this increasing human population means increasing food production by 70% [2]. However, in fulfilling the need for human nutrition, food production and consumption have been identified as key factors contributing to environmental degradation [3]. Food systems are major drivers of anthropogenic greenhouse gas (GHG) emissions, water pollution, and land use [4]. With this perspective, ensuring food security while transitioning towards sustainable food supply systems is a challenge that needs to be faced on the current global policy agenda [5]. Among the listed Sustainable Development Goals, the second goal refers to this issue and establishes the need to achieve food security and promote sustainable food production [6].

Different food consumption patterns can influence the generation of environmental burdens, as diets dictate the demand for food products [7]. As a mitigation measure, food cannot be eliminated, but modifying dietary patterns can result in the reduction of environmental impacts [8]. Current diets in Western countries are characterized mainly by the high consumption of proteins, predominately from animals [9]. Numerous studies have reported that animal-based food consumption exhibits higher environmental impacts than plant-based diets [10–12]. In addition, shifting from typical Western diets to more environmentally-friendly diets can reduce 70% of the emissions of GHG [13].

Within the debate on shifting towards more sustainable diets, the Mediterranean diet is regarded as a model of sustainable dietary pattern [14, 15]. Many studies have outlined that the Mediterranean diet has lower environmental impacts than other dietary patterns [10, 16, 17]. The evaluation of the environmental performance of diets is, in general, quantified based on the methodological framework of the life cycle assessment (LCA) [18–20]. LCA is an effective tool to evaluate the environmental performance of food products. The LCA framework accounts for the environmental impacts involved in the entire food supply chain, such as animal feed, agricultural operations, land-use change, and other processes [21].

Traditionally, Spain has followed the Mediterranean diet. However, in the past years, dietary habits of the Spanish population have shifted towards an increase of animal-based proteins and a reduction in the consumption of fruit and vegetables [22]. In light of these findings, including environmental information in the Spanish Dietary Guidelines could provide an extra stimulus for the population to shift to healthy choices in food consumption [23]. In this sense, special attention has been directed towards school menus [24]. In Spain, 73% of the children attending primary education consume meals prepared by the schools’ canteen services an average of 165 days per year [25]. Due to the growing number of children eating school meals, Spanish dietary guidelines for schools were established by the Spanish government to provide information to design school menus [26]. The menus designed for children should provide sufficient food quantity and quality to satisfy their nutritional needs, considering that school-aged children are in a stage of growth and development. In this sense, according to the Spanish guidelines, lunch meals should correspond to 35% of the child’s daily amount of energy (kcal) [27] and regarding the nutritional requirements, proteins should provide 12–15% of the total energy consumed, carbohydrates 50–60% of the total energy and lipids (fats) should not exceed 35% of the total energy consumed [28].

A step forward needs to be taken by addressing environmental aspects in school dietary guidelines. Many countries in Europe have policies to help schools provide nutritionally balanced lunch menus [29], and several studies have been conducted to evaluate the health aspects related to these menus [30, 31]. However, the assessment of environmental impacts caused by eating patterns in schools has not been extensively investigated. Research studies conducted quantifying environmental impacts of school lunch menus can be categorized as studies that quantify the environmental impacts of actual school lunch menus, and studies that elaborate hypothetical menus designed to optimize environmental impacts of schools’ lunch menus.

Studies quantifying the environmental impacts of actual school lunch menus have been carried out in Finland and England. In Finland, the impact categories analyzed were climate change and eutrophication potential [32], while in England, the environmental impacts of primary school meals were quantified using the CF and the Water Footprint [33, 34]. Several suggestions to improve the environmental performance of school lunches were highlighted from the quantification of these environmental impacts, such as limiting the number of meat dishes, including more vegetarian alternatives, and serving flexible portion sizes to minimize food waste, among other recommendations.

Other studies proposed different hypothetical school menus to optimize environmental impacts. Based on the Mediterranean diet guidelines in schools [35], Benvenuti et al. (2016) elaborated a procedure to design hypothetical menus minimizing the CF and water consumption [36]. A step forward was carried out by Ribal Sanchis et al. (2016), which developed a model based on the Mediterranean diet to optimize school menus considering not only health and environmental parameters, but also economic aspects [37]. The school environment can influence children’s early eating patterns. Therefore, educating young people at early stages will provide them with the knowledge and the skills needed to make better food choices and develop healthy eating patterns, such as the Mediterranean eating habits. In addition, schools can also provide a springboard for integrating nutritional health aspects with environmental sustainability.

Following this path, the goal of this research paper is to quantify the GHG emissions from an LCA approach, expressed as the carbon footprint (CF), of a school menu (baseline menu) designed following the guidelines established by the Spanish dietary guidelines for schools. Moreover, schools have to provide alternative menus to consider different eating habits and possible food allergies among children. Therefore, a comparison between the CF of the baseline menu and the different alternative menus is also evaluated.

The contribution of this study is to quantify the Carbon Footprint (CF) of school menus that have been designed following the Spanish school dietary guidelines. Previous studies have quantified the environmental impacts of diets following the Spanish dietary guidelines [23] and the environmental impacts of school meals considering the actual consumption patterns of different countries [32–34], following different school dietary guidelines and proposing mathematical models for designing hypothetical menus to minimize environmental impacts [36]. However, the CF of hypothetical school menus designed following the Spanish school dietary guidelines has not yet been quantified. Spanish school dietary guidelines claim to be advocating for healthy eating habits and lifestyle, and we aim to provide added value by incorporating an environmental aspect and compare these results with the existing literature on healthy diets, such as the Mediterranean diet.

The paper is organized as follows. Section 2 describes the baseline menu and the six alternative menus assessed and the data required to quantify the CF. The findings are shown in Section 3 and the main conclusions are discussed in Section 4.

Methodology

School menu design

To quantify the environmental performance of school menus, this research follows the Spanish schools' dietary guidelines [28]. Following these guidelines, a baseline menu was designed for the children of 3 to 8 years old, adjusting the daily energy intake per person to 610 kcal/lunch-meal with a maximum variation of 8%. The baseline menu was designed following the structure of a healthy menu according to the Spanish schools' dietary guidelines (Tables 1 and 2) [28]. The food quantity of each dish was estimated using a consensus report related to the nutrition in educative centers [26]. This report indicates the food portions in weight (grams) recommended depending on the age of the children and classified into main dishes or side dishes. For example, for the food group legumes, the total weight for a child between ages 3 to 8 is 60 g if it is the main dish and 30 g if it is a side dish. Once obtained the weight of the different food groups per lunch-meal, the calorie profile per lunch-meal (kcal/lunch-meal) was calculated assuming the following calorie factors: proteins have an energetic yield of 4 kcal/g, lipids of 9 kcal/g and carbohydrates 4 kcal/g [28]. The daily lunch meal caloric intake (kcal/lunch-meal) is the sum of the calories from these three nutrients.

Table 1. Recommended intake frequency of each type of food.

Food type	Weekly intake (times/week)
First course
Rice	1
Pasta	1
Legumes	1-2
Vegetables	1-2
Second course
Meat	1-3
Fish	1-3
Eggs	1-2
Side dish
Salads	3-4
Potatoes	1-2
Vegetables	1-2
Legumes	1-2
Dessert
Fruits	4-5
Other desserts	0-1

Table 2. Baseline school menu designed according to the Spanish school dietary guidelines.

Baseline menu 3 to 8 years old
Week 1
Monday	Tuesday	Wednesday	Thursday	Friday
646 kcal	684 kcal	692 kcal	603 kcal	689 kcal
Vegetable cream soup	Macaroni gratinated with cheese	Stewed fava beans	Ratatouille	Asian rice
Roast pork fillet	Battered hake fillet	Potato omelet	Roasted chicken leg	Tuna with tomato sauce
Roast potatoes	Salad	Green salad	Sauteed mushrooms	Baked vegetables with cheese
bread	bread	bread	bread	bread
fruit: orange	fruit: apple	fruit: pear	Canned peach in syrup	fruit: banana
Week 2
Monday	Tuesday	Wednesday	Thursday	Friday
566 kcal	732 kcal	521 kcal	655 kcal	610 kcal
Macaroni with tomato sauce	Zucchini and cheese cream soup	Baked potatoes with vegetables	Stewed lentils	Macaroni with vegetables
Omellete	Breaded escalope with cheese	Baked hake fillet	Beef burger	Battered calamari
Lettuce salad	Corn salad	Baked carrot	Pisto manchego	Lettuce salad
bread	bread	bread	bread	bread
fruit: apple	fruit: pear	Tiger nut milk	fruit: orange	fruit: banana
Week 3
Monday	Tuesday	Wednesday	Thursday	Friday
610 kcal	777 kcal	582 kcal	541 kcal	748 kcal
Vegetable cream soup	White beans with chorizo	Chicken soup	Asparagus cream	White beans with rice
Sausages	Potato omelet	Roasted chicken leg	Beef burger	Potato and zucchini omelet
Baked potatoes	Salad with olives	Baked vegetables	Roast potatoes	Carrot salad
bread	bread	bread	bread	bread
fruit: pear	fruit: orange	fruit: banana	fruit: pear	fruit: apple
Week 4
Monday	Tuesday	Wednesday	Thursday	Friday
541 kcal	561 kcal	674 kcal	670 kcal	605 kcal
Spaghetti bolognese	Stew	Potato salad	Vegetable cream	Noodle soup
Breaded hake with cheese	Stew soup	Chicken breast strips	Roast beef	Fried eggs
Roast potatoes	bread	Mixed salad	Mashed potatoes	Green salad
bread	fruit: banana	Bread	bread	bread
Yogurt		fruit: apple	fruit: pear	fruit: orange

The Spanish schools' dietary guidelines also establish that schools need to provide alternative menus for children suffering from food allergies or health issues. In relation to allergies and intolerances, alternative menus have been designed in this paper excluding dairy, legumes, eggs, and fish, as these food groups have been reported to be the main responsible for food allergies [28]. In addition, three more alternative menus were also designed. The hypocaloric menu and the astringent menu have been considered because they address health issues. The hypocaloric menu has the objective of reducing the total calorie intake for children with overweight. This menu was designed considering a maximum daily lunch meal caloric intake of 600 kcal, achieved by substituting starchy foods, such as potatoes, with vegetables and modifying the way of cooking some dishes, for example, frying has been changed for baking. The astringent menu is also related to health issues. In this case, this menu aims to palliate the effects of diarrhea and gastroenteritis and to void vomiting. The astringent menu was designed using the food dishes and culinary techniques of boiling and baking specifically recommended for this type of menu by the Spanish schools' dietary guidelines, using food portions in weight as established in the consensus report [26]. Some of the dishes indicated for the astringent menu are mashed potatoes and carrots, boiled chicken breast, zucchini soup, among other dishes. In addition to health aspects, we have also designed a menu without meat because schools include this type of menu to provide children who do not eat meat for ethical or religious reasons with an alternative menu. Considering these aspects, six alternative menus have also been designed to quantify each menu’s CF (Supplementary material SM Tables 4–10). These menus are summarized in Table 3.

Table 3. Description of the baseline menu and the six alternative menus.

Menu Type	Average energy intake	Average CO2 eq	Remarks
	(kcal/person/monthly lunch-meal)	(Kg CO2eq/person/monthly lunch-meal)
Baseline	2541	24.39	Generic menu.
A#1: w/o dairy or legumes	2524	22.26	Excludes dishes including dairy, i.e., milk-based, and legumes.
A#2: w/o fish	2543	23.41	All dishes including fish are substituted with other protein sources, e.g., meat and eggs.
A#3: hypocaloric	2379	20.72	The maximum calorie intake has been limited to approximately 600 kcal per day.
A#4: astringent	2210	14.77	Consists of food dishes specified by the Spanish school dietary guidelines to palliate the effects of intestinal illnesses.
A#5: w/o meat	2531	17.11	All dishes including meat, i.e., beef, poultry, and pork, are substituted with other protein sources, e.g., fish and eggs.
A#6: w/o egg	2540	22.33	All dishes including eggs are substituted with other protein sources, e.g., fish and meat.

Table 4. Daily carbon footprint and lunch carbon footprint of the Mediterranean diet and the school baseline menu and the alternative menus (shortened version of SM Table 13).

Diet/menu type	Daily carbon footprint	Lunch-meal carbon footprint	Reference
	kg CO2eq/person/day	kg CO2eq/person/ lunch-meal
Mediterranean diet	3.36	1.48	[47]
Mediterranean diet	2.86	1.26	[48]
Mediterranean diet	2.00	0.88	[49]
Mediterranean diet	3.40	1.50	[10]
Baseline	–	1.2
No milk, no legumes	–	1.1
No fish	–	1.2
Hypocaloric	–	1.0
No meat	–	0.9
No eggs	–	1.1
Astringent	–	0.7

A total of 88 dishes have been used to design the different menus (Supplementary material SM Tables 6–12). The daily menu consists of a first course, second course, side dish, dessert, and bread. The dessert is mainly fruit and it has been considered that olive oil is consumed in every dish as dressing or for cooking, except for the hypocaloric menu which has restricted its use.

Quantification of the carbon footprint

To quantify the CF, the different processes involved along the supply chain are converted into GHG emissions using carbon factors. These factors are relative to carbon dioxide and expressed in kg CO₂ equivalent per reference unit [38]. To allow comparisons of the global warming impacts of the different GHG, the Global Warming Potential (GWP) of each gas is used [39]. The main GHG are methane (CH4), which is estimated to have a GWP of 28–36 over 100 years, nitrous oxide (N2O), which has a GWP 265–298 over 100 years, and chlorofluorocarbons (CFCs), hydrofluorocarbons (HFCs), hydrochlorofluorocarbons (HCFCs), perfluorocarbons (PFCs), and sulfur hexafluoride (SF6), which have GWP between thousands or tens of thousands of CO₂ equivalent. For a detailed description of the GWP of the different GHG see Chapter 8 of the IPCC assessment report [40]. Regarding food products, the carbon factors are usually expressed as kg CO₂ equivalent per kg of product or per kcal of product.

In this study, the CF has been quantified for the stages of (1) food production, (2) transportation, and (3) cooking. Each dish has been disaggregated into its main ingredients and the CF factors for each ingredient and stage have been used. For food production, CF factors using a life cycle assessment framework (cradle to gate) have been obtained from the literature and can be found in the reference section of the Supplementary material (SM Table 2). Some of the CF factors from the literature have different system boundaries, such as cradle to retail or cradle to grave. In those cases, only the emissions generated in the cradle to gate boundaries have been considered to obtain the CF factors. For example, the CF factor of the food product mushrooms has been obtained from Gunady et al. (2012) study, in which the emissions from cradle (the pre-farm stage) to retail (distribution of mushrooms to grocer) are considered [41]. In our case, the transportation has been calculated using separate CF factors, so the transportation emissions have been excluded from the CF factor of mushrooms used in the reference research paper. Since the food weight from the consensus report is expressed as total food weight, which is the total weight including non-edible parts, conversion factors from Gustavsson et al. (2011) have been used to obtain the edible food weight, which excludes skin or pips of fruit and vegetables, bones in meat and fish (Supplementary material SM Table 1) [42]. These conversion factors account for the food losses in the consumption phase. For example, for the case of the apple, the conversion factor of 0.8 meaning that for every gram of apple consumed, 0.2 g will be food waste and will be lost, consuming 0.8 g of apple.

In relation to transportation, transport has been considered to be from the place where food is produced to Madrid, where it will be consumed in the school canteen (Supplementary material SM Table 3). The distances have been estimated using google maps [43]. The CF factors for transportation have been obtained from Ecoinvent [44] in ton per km (tkm), considering CF factors for truck (road) (0.25 kg CO₂ eq/tkm), ship (0.0061 kg CO₂ eq/tkm) and airplane (1.7 kg CO₂ eq/tkm) transportation.

Regarding the cooking stage, Sonesson approach has been used to calculate the CF from cooking the meals [80]. This report presents general models intended for use in LCA analyses to estimate the energy needed for food preparation. The models of three types of cooking have been used in this research: the frying, baking/roasting, and boiling models. The inputs needed in these models are the amount of product to cook (grams) and the time required (minutes). The amount of food corresponds to the food portions in weight and the time required was obtained from the recipes of a popular cooking website (Supplementary material SM Table 4) [45]. In this study, it is assumed that only one type of cooking method is used to cook each dish, meaning that a dish can only be boiled, fried or baked, but it cannot be boiled and then fried, for example. In addition, it is assumed that the energy used corresponds to electricity. In this way, the MegaJoules (MJ) calculated for each dish is converted to kilowatt-hour (kWh) and then by applying the CF factor of 0.385 kg CO₂ eq/kWh, the CF from cooking is obtained [46].

Results

The carbon footprint of the baseline school menu and the alternative school menus

The individual dishes in Supplementary material SM Table 1 have been used to design the baseline school menu for infant school children. Regarding the overall CF of the baseline menu and the six alternative menus, the total CF of the baseline menu is the highest compared to the rest of the menus, exhibiting a CF of 24.39 kg CO₂ eq/person/monthly lunch-meal, followed by the menu without fish (Figure 1). The menu without fish presents only a 4% reduction concerning the baseline menu due to the substitution of fish dishes by meat dishes in the majority of cases. There is still a reduction in the CF of the menu without fish due to the substitution of fish dishes by eggs, which have a CF factor lower than some types of fish, such as hake (70% lower) and salmon (9% lower). On the other hand, the astringent menu exhibits the lowest total CF (14.77 kg CO₂ eq/person/monthly lunch meal), followed by the menu without meat. Reductions in the CF of both menus with respect to the baseline menu, 40% for the astringent menu, and 30% for the menu without meat, are explained by the elimination of beef dishes. The astringent menu presents a lower CF than the menu without meat due to the restrictive calorie intake of the astringent menu, being the calorie intake 13% lower (Table 3).

Figure 1. Comparison of the Carbon footprint per month and per food stage of the school baseline menu and the six alternative menus.

Comparing the contribution of each stage to the overall CF, a similar contribution is observed for the baseline menu and the six alternative menus. On average, production CF contributes 15.2 kg CO₂ eq (73%), transport CF 2.2 kg CO₂ eq (11%) and cooking CF 3.3 kg CO₂ eq (17%). The menu without dairy and without legumes and the menu without eggs are the menus with the highest production CF (Supplementary material SM Table 14). This is partly due to the substitution of legumes and eggs with fish and meat, which have a higher CF factor (kg CO₂ eq/kg food product) than legumes and eggs. Focusing on the cooking CF, in the menu without fish and in the menu without meat, fish and meat dishes are, generally, substituted by dishes with similar cooking methods and time, therefore, the cooking CF does not vary significantly. The hypocaloric menu designed to restrict calories presents a low CF for the three stages explained by the use of low-fat dishes eliminating beef dishes and salads as side dishes. These facts explain the 15% reduction observed in the CF from the hypocaloric menu with respect to the baseline menu.

The carbon footprint of school dishes

In general, the dishes which present higher CF are the second-course dishes, ranging from 0.2 kg CO₂ eq to 0.41 kg CO₂ eq (Supplementary material SM Table 5). In particular, the second-course dishes which contain beef, like roast beef present higher CF. The substitution of beef, which has a very high CF factor (26.6 kg CO₂ eq/kg food product), by other protein sources, like chicken and eggs that have lower CF factors (1.8 kg CO₂ eq/kg food product and 2 kg CO₂ eq/kg food product, respectively), leads to a reduction in the overall CF of the dishes. Dishes containing fish as the main ingredient are also found to exhibit high carbon emissions, especially the ones which require several ingredients in their preparation. For example, hake presents a high CF factor (6.7 kg CO₂ eq/kg food product), and the dish battered hake fillet, which uses more ingredients apart from hake (Supplementary material SM Table 1), has a CF 12% higher than the dish baked hake fillet (Supplementary material SM Table 3). While the CF of the baked hake fillet is generated only by the CF of hake, the CF of milk (0.013 kg CO₂ eq) and egg (0.019 kg CO₂ eq) have to be considered in the total CF of the dish battered hake fillet. On the other hand, the dishes which generate lower total CF are the side dishes, ranging from 0.012 kg CO₂ eq to 0.21 kg CO₂ eq, among which, salad dishes, such as the mixed salad (0.014 kg CO₂ eq) and the carrot salad (0.012 kg CO₂ eq), can be highlighted. A possible explanation for this CF reduction is that the production of the foods included in these dishes (lettuce, tomato, carrots, etc.) is less carbon-intensive than the production of other food types. For example, comparing the production CF of lettuce with beef, the production CF is 99% lower (SM Table 2).

In relation to the cooking stage, salads are dishes that are consumed without the need to cook them, therefore, there are no carbon emissions from this stage. The cooking method and the cooking time also influences the carbon emissions. Foods that are boiled, such as legumes, rice and vegetables, exhibit a higher CF than fried food, such as fish and meat, as more energy is needed to boil water than to heat oil to fry in a pan. This is seen in the cooking CF of stewed fava beans and rice with vegetables with respect to chicken breast strips and cod fillet. The cooking CF of stewed fava beans is 92% higher than the cooking CF of chicken breast strips and 90% higher than the cod fillet. Similarly, the cooking CF of rice with vegetables is 95% higher than the cooking CF of chicken breast strips, and 94% higher than the cod fillet.

For the transport CF, bananas are the foods with the highest carbon emissions. The reason is that they are produced in the Canary Islands and transported by plane, which is the most pollutant way of transport, with a CF factor of 1.7 kg CO₂/tkm. Additionally, the dishes containing fish, such as tuna with tomato sauce and breaded fish with cheese, exhibit higher transport CF due to the long distance between the production location and the consumption place, in this case, Madrid. Additionally, as it is observed in the dish of tuna with tomato sauce, the mode of transport also influences the carbon impacts, having higher CF the food transported by truck than by ship.

Discussion

Comparison with results from the literature

The Spanish school dietary guidelines are developed based on the Mediterranean diet. Therefore, as expected, the CF of the baseline school menu is in line with the CF results obtained in previous studies regarding the Mediterranean diet (Table 4). Considering the three stages (food production, transport, and cooking), the CF of the baseline menu is similar to the carbon emissions of the Mediterranean diet in Spain [48], being this GHG emissions 1.26 kg CO₂ eq/person/lunch-meal. In another investigation carried out by Sáez-Almendros et al. (2013) it was seen that the GHG emissions for the lunch meal corresponded to 0.88 kg CO₂ eq/person/lunch-meal based on the Mediterranean dietary guidelines of Spain. This result is 27% lower than the baseline menu and approximates more to the astringent menu. The system boundaries were established from cradle to retail and did not consider the GHG emissions from cooking, thus exhibiting lower carbon emissions. Moreover, another country that follows the Mediterranean diet is Italy. Uulaszewska et al. (2017) and van Dooren et al. (2014) evaluated the GHG emissions generated following the Italian guidelines and obtained that the CF was 1.48 kg CO₂ eq/person/lunch-meal and 1.50 kg CO₂ eq/person/lunch-meal, respectively. These results differ 19% with respect to the CF of the baseline school menu. The CF of the actual Spanish food consumption was reported to be 2.64 kg CO₂ eq/person/lunch-meal [50]. The population’s food consumption patterns do not tend to adhere to the established dietary guidelines [51, 52], exhibiting higher carbon emissions than adhering to the Mediterranean diet guidelines [49]. The Mediterranean diet, and therefore the Spanish school dietary guidelines, are also characterized by recommending the consumption of a wide variety of foods, focusing on low consumption of meat, particularly beef. This aspect is noticeable in the CF generated by the baseline menu, which is comparable to the CF seen for Vegetarian and Vegan diets [10, 12, 48].

It has to be highlighted that these are tentative comparisons and the results can be influenced by different factors. For instance, age could determine the amount and type of food consumed. The number of calories recommended for children is different than for adults [9]. Previous studies, shown in Supplementary material SM Table 13, refer to a total consumption of 2000 to 2300 kcal/day, which approximate to the total calorie intake required for adults. This calorie intake can be estimated as a lunch energy intake of 700 to 805 kcal/lunch-meal, which is 19% higher than the established in this paper for children between 3 and 8 years old.

Policy and recommendations

The Spanish school dietary guidelines were elaborated on the basis of providing a useful tool for schools to guarantee a balanced diet adapted to the energy requirements of each children’s age group. This objective is confirmed by the findings from this study, which suggest that adhering to the Spanish school dietary guidelines results in a healthy and nutritious baseline menu. The school guidelines are based on the Mediterranean diet [28] and share the same principles of this well-recognized healthy diet, which advocates for high consumption of vegetables, fruits, nuts, legumes, and unprocessed cereals, and limits the consumption of meat and dairy products [53]. Therefore, in terms of health, it is evident that the Spanish school dietary guidelines succeed in providing information to encourage healthy choices about foods.

However, from an environmental perspective, there is still room for improvement. While a clear message regarding healthy eating patterns has been given by the Spanish school dietary guidelines, and the national dietary guidelines in general, there is no consideration of the environmental impacts generated. By incorporating explicit information on the environmental performance of foods, policy-makers will be contributing to a gradual transition towards more sustainable eating patterns [54]. As a matter of fact, the quantification of the CF of the baseline menu revealed that the Spanish school dietary guidelines are already promoting sustainable eating guidance for school menus. Incorporating environmental data will mean providing added value to the existing dietary guidelines. This is the case of the national dietary guidelines of European countries, such as Estonia, Germany, Finland, Denmark, the United Kingdom, the Netherlands, and Sweden, and other countries around the world, such as Australia, Brazil, Qatar, and Uruguay [55–65]. These national dietary guidelines establish recommendations to promote sustainability, for example, by reducing the consumption of meat products, reducing food waste, and choosing foods with minimal or no packaging, among others [54].

Furthermore, contemplating the environmental aspects in the Spanish school dietary guidelines could also help policy-makers re-evaluate the current guidelines to some extent to make them more environmentally-friendly. For instance, regarding the alternative menus, it can be seen that these are not aimed at reducing carbon emissions, as the reduction in CF is on average 18% compared to the baseline menu. In this case, the food recommendations indicated in the dietary guidelines to design the alternative menus are focused on different kinds of allergies and intolerances to specific food groups [66]. In order to improve the CF of the alternative menus, a possible recommendation could be to use plant-based protein sources as substitutes for fish, meat and eggs, such as tofu, soybeans, peanuts, and lentils [67]. The optimization of the menus by using plant-based replacements will, in principle, reduce their overall CF, resembling to some extent the Vegan diet [48].

Other policies could also be directed to improve the sustainability of food consumption. Food production has been identified to be the most carbon polluting stage and a possible measure to reduce its environmental impacts could be to promote the consumption of organic products. Several studies have highlighted lower environmental burdens from organic products [68]. For example, regarding fruit production, organic blueberries and strawberries have lower CF factors than conventional production, being 13% and 31% lower, respectively [69]. Organic cereal production has also been reported to have lower CF factors than conventional production. The organic CF factor for wheat is 60% lower than for conventional production, for winter barley it is 48% lower, for spring barley 38% lower, and for oats it is 32% lower [70]. The mitigation of the environmental impacts from food production could also be addressed by promoting the consumption of seasonal fruits and vegetables as alternatives to fruits and vegetables cultivated in heated greenhouses [71]. The consumption of seasonal foods reduces carbon emissions by removing the impacts associated with greenhouse electricity consumption. For instance, the CF factor used for the cucumber in this study is referred to greenhouse cucumber production, in which electricity contributed approximately 68% to the total carbon emissions generated. Therefore, these impacts could be avoided by consuming cucumber following its growing season and not requiring greenhouse cultivation [72].

In relation to the impacts associated with transport, promoting the consumption of local products reduces the known as food miles, and the pollution associated with them [73]. Following this path, a recent initiative, which could contribute to the reduction of carbon emissions generated by transport, is the implementation of school gardens [74]. This initiative encourages schools to grow vegetables and source their cafeteria food without the need for transportation. Apart from the distance, different types of transportation could have a significant impact on the CF. Transportation by truck is less polluting in terms of carbon emissions than by airplane [48].

In addition to the CF from the production and transport stages, the cooking process also impacts the total CF [75]. Cooking food is inevitable, and always using the same cooking method to prepare every dish is almost non-viable. For instance, frying presents a lower CF [48], but it cannot be used for cooking all the dishes in the school menus. For that reason, the reduction in the carbon impacts relies on searching for energy-efficiency solutions [76]. It has been recently investigated technologies related to solar cooking as an alternative to fossil fuel energy sources [77]. In addition to the development of new energy-efficient appliances, governments should also pay attention to incentives for consumers to purchase high-efficiency appliances [78].

Limitations of the study

Although a thorough study has been conducted in this research paper, some limitations need to be acknowledged that lead to an underestimation of the CF quantified.

Not considering food losses along the entire food supply chain, only assuming one cooking method per dish, and not accounting for the impacts from storage and refrigeration seem to be the main causes of this underestimation.

Food losses in the consumption stage have been used in the quantification of the CF. However, the factors of food losses considered in this study for Spain have been assumed to be equal to the food losses of the European region [42]. In addition, food losses along other stages in the supply chain have not been considered. When assessing the CF associated with the cooking stage, it has to be noted that only one type of cooking method (boiling, baking, or frying) per dish has been considered, whereas in reality several cooking methods could be needed. For example, the main cooking method for a dish could be frying, but then a further heating process in the microwave could be required increasing the energy utilized. We have not considered the CF associated with the storage and refrigeration of food. In another study, household storage had been identified to account for 10% of the total CF [79]. However, schools store food less time than households, thus contributing less to the total CF.

Further research could be directed towards obtaining specific CF factors, distinguishing between the food origin, agricultural practices and nature of food items. These specific CF factors will allow the assessment of more school menus, such as menus designed for diabetic children or lactose-intolerant children, which require CF factors for special products, such as bread without gluten or milk without lactose. Following this path, the accuracy of the CF of food consumption will be improved and more robust information will be available for governments for better-informed decision-making.

Conclusions

In this study, the CF of a baseline school menu designed following the Spanish school dietary guidelines has been quantified. These guidelines provide information regarding the nutritional profile, energy intake, food frequency intake and combinations of food products to structure the first course, second course, the side dish and the dessert. The total CF of the baseline menu was 24.39 kg CO₂eq/person/monthly lunch-meal. The outcomes of this study are in agreement with the CF quantified in previous studies for the Mediterranean diet, which presents a CF 5% higher on average than the school baseline menu. The Spanish school dietary guidelines are based on the Mediterranean diet so similar CF was expected to be obtained. Overall, adhering to the Spanish school dietary guidelines provides a healthy and sustainable menu for children. The wide range of food provides all the nutrients needed to fulfill children’s energy requirements. Additionally, the food variety prioritizes the consumption of fruits and vegetables and limits the consumption of carbon-intensive foods, such as red meat, leading to an overall reduction in the CF.

In addition, the Spanish school dietary guidelines establish that it is compulsory to include alternative menus for allergies and intolerances. Therefore, six menus have also been designed including the premises indicated by the guidelines concerning these alternative menus. The CF of the menus was 22.26 kg CO₂ eq/person/monthly lunch-meal for the menu without dairy and without legumes, 17.11 kg CO₂ eq/person/monthly lunch-meal for the menu without meat, 23.41 kg CO₂ eq/person/monthly lunch-meal for the menu without fish, 22.33 kg CO₂ eq/person/monthly lunch-meal for the menu without eggs, 20.72 kg CO₂ eq/person/monthly lunch-meal for the hypocaloric menu and 14.77 kg CO₂ eq/person/monthly lunch-meal for the astringent menu. From the results, the CF of the alternative menus exhibited on average 18% reduction with respect to the CF of the baseline menu. The guidelines provided to design the alternative menus are not focused on improving the CF, but only aiming at substituting foods which could cause allergies and intolerances. In light of these findings, a possible optimization of the alternative menus could be to use plant-based protein sources as substitutes for meat fish and eggs.

A series of recommendations have been presented to improve the sustainability of food consumption. In particular, policy-makers are encouraged to consider environmental aspects in the school dietary guidelines and in the national dietary guidelines in general. It is believed that the knowledge of the environmental performance of foods and their inclusion in the dietary guidelines could potentially lead to the implementation of more sustainable eating habits.

Acknowledgements

The authors heartily thank Adrian Jesus Garcia for his helpful suggestions and valuable support.

Disclosure statement

No potential conflict of interest was reported by the author(s).