EN中文

Journal of Resources and Ecology >

2025 , Vol. 16 >Issue 1: 93 - 104

DOI: https://doi.org/10.5814/j.issn.1674-764x.2025.01.009

Resource Economy

Shifting Diets for Low-carbon Patterns Based on Multi-objective Optimization in China’s Megacities: A Case Study of Beijing and Shanghai

  • ZHANG Yan , 1 ,
  • ZHU Yuanyuan 1 ,
  • ZHU Xiaohua , 2, * ,
  • LI Yan 1
Expand
  • 1. Key Laboratory of Geographic Process Analysis & Simulation of Hubei Province, Central China Normal University, Wuhan 430079, China
  • 2. Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China
*ZHU Xiaohua, E-mail:

ZHANG Yan, E-mail:

Received date: 2024-06-10

  Accepted date: 2024-09-09

  Online published: 2025-01-21

Supported by

National Natural Science Foundation of China(42171230)

Fold

Abstract

Dietary modification has been suggested as a viable path to mitigate climate change. This study explored the evolution of dietary structures in China’s megacities and quantified food-related carbon emissions using the carbon footprint method. Then, we developed a multi-objective optimization model to identify low-carbon dietary options that align with cultural preferences, economic feasibility, and nutritional requirements. We found that from 1992 to 2021, the total food consumption of residents in Beijing and Shanghai experienced a period of fluctuating decline and slow rise, respectively, with their dietary structures both shifting toward animal-based diets. Additionally, the significant increase in meat consumption led to increasing food carbon emissions in Beijing and Shanghai, with per capita emissions rising by 37.2% and 25.2%, respectively. According to the multi-objective optimization model, shifting towards reduced consumption of animal products, particularly meat, would align with culturally acceptable, economically feasible, and low carbon emission goals in Beijing and Shanghai, potentially leading to reductions of 92.14 t and 212.65 t in total food carbon emissions compared with the actual in 2021, respectively. This study enhances understanding of the changing dietary patterns in urban areas and their associated carbon emissions, and emphasizes the significance of optimizing diets as a key priority for advancing global sustainable development.

Key words： carbon footprint; diet optimization; evolutionary characteristics; China

Cite this article

ZHANG Yan , ZHU Yuanyuan , ZHU Xiaohua , LI Yan . Shifting Diets for Low-carbon Patterns Based on Multi-objective Optimization in China’s Megacities: A Case Study of Beijing and Shanghai[J]. Journal of Resources and Ecology, 2025 , 16(1) : 93 -104 . DOI: 10.5814/j.issn.1674-764x.2025.01.009

1 Introduction

Dietary, health and environmental dilemmas are major threats and challenges to the Anthropocene in the 21st century (Clark et al., 2019; Willett et al., 2019). Studies have indicated that the food system already contributes 34% of global greenhouse gas (GHG) emissions (Crippa et al., 2021). Some recent studies have focused on calculating food carbon emissions (Temme et al., 2015; Clune et al., 2017; Beylot et al., 2019; Sala et al., 2019), and most have shown that the food consumption process contributes the most to GHG emissions, while food choice contributes more (Eshel et al., 2014; Poore and Nemecek, 2018; Wang et al., 2022). Animal-based diets generate twice as much GHG emissions as plant-based diets (Xu et al., 2021). Recently, the global diet has shifted to become more animal-based and society has entered a phase of widespread degenerative diseases, in which non-communicable diseases such as obesity and chronic diseases are prevalent (Popkin, 2004). This is especially true in high-income countries (Mazzetto et al., 2023), which will irreparably threaten planetary boundaries. Human activities are already having a progressively greater impact on climate change than other activities (IPCC, 2022).
Fortunately, dietary adjustments have been identified as a feasible way to mitigate climate change (Wang et al., 2021). One approach compares GHG emissions between the current dietary structure and various models, considering potential reductions through adjustments in food consumption (Green et al., 2015; Tom, 2016; Springmann et al., 2018). The designs of some adjustment schemes also consider the goal of reducing land and water resource use (Song et al., 2019; Wu et al., 2022; Zhu et al., 2024b). Some studies have compared GHG emissions under balanced dietary patterns recommended by national dietary guidelines (Springmann et al., 2020). However, most studies have indicated that dietary guidelines, apart from those in countries such as India, fail to consider environmental impacts and do not align with the principles of sustainable development (FAO, 2016). Furthermore, a comparative analysis of representative global diets such as the Mediterranean diet (Ndlovu et al., 2019), the Lancet Diet and the flexitarian diet, revealed that the flexitarian diet brought more substantial low-carbon benefits (Zhu et al., 2023c). Recently, some studies have concentrated on dietary modification schemes that rely either on single and multi-objective functions, or on linear and non-linear optimization (Donati et al., 2016; Larrea-Gallegos and Vazquez-Rowe, 2020). These strategies consider factors such as water consumption, land usage, and carbon emissions, along with the nutritional needs of residents. They have explored dietary schemes that meet both nutritional requirements and environmental sustainability goals, and shown that the adjustments of plant-based diets can bring dual benefits of environmental optimization and nutritional improvement (Chaudhary and Krishna, 2019; Abejón et al., 2020). In addition, cultural acceptability has also been considered in some studies as a constraint of dietary adjustment to ensure that dietary changes aligning with various environmental objectives are feasible in practice (Yin et al., 2021; Muñoz-Martínez et al., 2023).
However, previous studies were limited to specific countries or small groups, and neglected the evolution of regional dietary structures (Sievert et al., 2019; Miller et al., 2022; Chen et al., 2023; Zhu et al., 2023a; Zhu et al., 2023b). Currently, more than 55% of the global population resides in urban areas (WBG, 2017), where rapid dietary changes are exacerbating food-related challenges. Although existing research has emphasized the impact of the food environment in urban areas, there has been limited exploration of dietary changes, with dietary adjustment programs largely centered on representative diets (Xiong et al., 2020). Importantly, regional dietary differences complicate unified dietary adjustments, and the economic burden remains unresolved. The application of multi-objective optimization methods at more microscopic scales may be the key to solving this problem. It is important to study past dietary changes and make differentiated adjustments to achieve sustainable diets based on the income levels and nutritional needs of regional residents.
To bridge this gap, this study analyzed the evolution of the dietary structures over the past 30 years in China’s first-tier cities, which are already world cities. Then we calculated the carbon emission effects of their food consumption and constructed a dietary adjustment model to meet both nutritional needs and low-carbon patterns in the highly urbanized areas under multi-objective optimization. The main contributions of this study are threefold. 1) We systematically explored the evolution of the dietary structure in China’s first-tier cities (Beijing and Shanghai) from multiple perspectives, including changes in food consumption and proportions, dietary diversity level, and food nutrient intake. 2) We estimated the carbon emissions effect driven by the evolution of the dietary structure. 3) We constructed a multi- objective optimization model by integrating nutritional and health needs, environmental sustainability goals, and food consumption costs. Then we explored sustainable dietary patterns that can achieve a win-win situation between nutritional health and environmental friendliness.

2 Materials and methods

2.1 Study area

Urbanization level influences the overall quantity and composition of regional food demand, with megacities serving as the focal point for national food demand. In 2021, Beijing’s urban population was projected to reach 17.75 million, with an 89.3% urbanization rate. Similarly, Shanghai was expected to have an urban population of 19.7 million, and an 87.5% urbanization rate. Beijing and Shanghai hold the top positions as the “First-Tier Cities” in China. Recently, the diet of residents in Beijing has tended to be primarily animal-based, leading to a sharp rise in diet-related carbon emissions (Xiong et al., 2020). The ongoing population growth and shift towards high-carbon dietary patterns will inevitably exacerbate environmental threats. In the future, China’s urbanization will continue to rise, with over 70% of the population living in cities. Cities with high urbanization levels are the primary political and economic hubs in China, so using them as a case study can, to a certain extent, serve as a representative model for guiding the dietary structure shift toward sustainability. This approach also holds certain reference significance for other cities.

2.2 Analyzing the characteristics of dietary structural evolution

2.2.1 Dietary structure

Dietary structure refers to the proportions of food group consumption in the diet. In this study, food types were classified into eight groups according to the Food and Agriculture Organization (FAO): grains (cereals, beans and potatoes), vegetables, fruit, oilseeds, eggs, dairy, meat (poultry, pork, beef and mutton), and aquatic products. Among them, animal-based foods include eggs, dairy, meat, and aquatic products. We adopted the concept of dysprosium (E) to quantify the degree of dietary diversification, which can provide a clearer reflection of the evolution of the dietary structure of residents (Liu et al., 2014). Food consumption data were derived from the National Statistical Yearbook (http://www.stats.gov.cn/tjsj/), Beijing Statistical Yearbook and Shanghai Statistical Yearbook (https://tjj.beijing.gov.cn/, https://tjj.sh.gov.cn/). The food consumption data used in this study only included food consumption at home, and according to the statistical standard, the data refer to the per capita food consumption, and it does not distinguish the edible parts and the waste parts of food. The calculation formulas are:
${{z}_{i}}=\frac{{{x}_{i}}}{\underset{i=1}{\overset{n}{\mathop{\mathop{\sum }^{}}}}\,{{x}_{i}}}$
$DDI=\underset{i=1}{\overset{n}{\mathop{\mathop{\sum }^{}}}}\,{{z}_{i}}\times \ln \left( \frac{1}{{{z}_{i}}} \right)$
where the value interval of DDI is (0, 1), and a larger value indicates a higher degree of dietary diversification; ${{z}_{i}}$ is the proportion of the consumption of type i food; n represents the number of food types; and xi represents the annual food consumption, and the same definitions were applied below.

2.2.2 Dietary nutrients

We analyzed the residents’ daily per capita intake of energy, protein, and fat to follow the changes in dietary structure in China’s first-tier cities. We referred to the China Food Composition Table (https://nlc.chinanutri.cn/fq/) and calculated the total energy contents and macronutrient intake (Table 1). Meanwhile, we compared nutrient intake against the recommended values outlined in the Chinese Dietary Guidelines 2022 (CNS, 2022), using the balanced dietary pattern to represent nutritious diets (Table 2). The calculation formula is:
$N{{f}_{j}}=\underset{i=1}{\overset{n}{\mathop{\mathop{\sum }^{}}}}\,{{y}_{ij}}\times {{x}_{i}}$
Table 1 Nutrients per unit mass of food
Food type Energy
(kcal kg-1)		 Protein
(g kg-1)		 Fat
(g kg-1)		 Carbohydrates
(g kg-1)
Grain 2999.5 104.3 39.8 545.8
Vegetables 487.6 32.0 4.0 83.0
Fruit 685.9 5.0 6.0 152.0
Oilseeds 8828.9 0.0 998.0 2.0
Eggs 2000.5 147.0 146.0 29.0
Dairy 628.6 29.0 31.0 59.0
Meat 2782.0 157.0 218.0 52.0
Aquatic products 1720.8 194.0 91.0 30.0
Table 2 The balanced dietary pattern recommended by Chinese dietary guidelines 2022
Food type Lower value
(g day-1)		 Upper value
(g day-1)		 Average
(g day-1)
Grain 200.0 300.0 250.0
Vegetables 300.0 500.0 400.0
Fruit 200.0 350.0 275.0
Oilseeds 25.0 30.0 27.5
Eggs 40.0 75.0 57.5
Dairy 300.0 500.0 400.0
Meat 40.0 75.0 57.5
Aquatic products 40.0 75.0 57.5
where Nfj represents the dietary intake of the j-th nutrient; and yij denotes the amount of the j-th nutrient available per unit quantity of the i-th food, and the same definitions were applied below.

2.3 Analysis of the carbon emission effect driven by dietary structural evolution

The carbon footprint (CF) is widely used to measure the total carbon emissions caused by direct or indirect activities during the product life cycle (Bastianoni et al., 2004). In this study, the CFs were calculated using life-cycle assessment (LCA) methods (Tilman and Clark, 2014), which included the carbon emissions produced in the entire food supply chain (including crop cultivation, breeding, industrial processes, transportation and storage). In order to measure the food carbon emissions of residents and explore the carbon emission effects under the evolution of the dietary structure in China’s first-tier cities from 1992 to 2021, we calculated the CF produced by each food consumed by multiplying the consumption by its corresponding footprint coefficient, as referenced in a study by Tilman and Clark (2014). The CF coefficients in that document were adopted from a recent meta-analysis of the emissions of the 8 basic food items (Table 3). The calculation formula is:
$Cf=\underset{i=1}{\overset{n}{\mathop{\mathop{\sum }^{}}}}\,c{{f}_{i}}\times {{x}_{i}}$
Table 3 Carbon emission coefficients of various types of food
Food type Carbon emission coefficient
(kg CO2 eq kg-1)
Grain 0.28
Vegetables 0.22
Fruit 0.07
Oilseeds 1.38
Eggs 0.86
Dairy 0.32
Meat 6.39
Aquatic products 1.74
where Cf represents the carbon emission of food; and cfi is the carbon emission coefficient of each food type.

2.4 Multi-objective optimization model

This study constructed a multi-objective optimization model to further explore strategies for adjusting the dietary structure of residents to achieve a low-carbon consumption model that meets their nutritional and health needs. Multi- objective optimization is often used to find one or more solutions that minimize or maximize several specified objectives while satisfying all constraints. In this study, we employed the nondominated sorting genetic algorithm version II (NSGA-II) (Deb et al., 2002) performed by Matlab 9.4 as our multi-objective optimization method to find the optimal food combination that would meet the residents’ food nutrient intake standards, while being culturally acceptable, having low carbon emissions, and minimizing the economic cost associated with food consumption. Each simulation was repeated 100 times to enhance the likelihood of identifying the globally optimal solution rather than a local optimum.

2.4.1 Objective function setting

(1) Dietary acceptability
A diet similar to the current diet structure can be considered culturally acceptable, while a diet with a larger deviation from current diet would be less acceptable (Perignon et al., 2016; Yin et al., 2020). To avoid unrealistic dietary shifts, we included dietary acceptability as one of the key factors to consider when optimizing the diet. The formula is:
$f(DA)=\frac{1}{8}\underset{i=1}{\overset{8}{\mathop{\mathop{\sum }^{}}}}\,ABS\left( \frac{{{{{x}'}}_{i}}-{{x}_{i}}}{{{x}_{i}}} \right)$
where f(DA) represents the acceptability of food groups after food adjustment; and ${{{x}'}_{i}}$ represents the food consumption after adjustment (for the diets based on the multi-objective optimum).
(2) Affordable cost
Affordability refers to the ability of residents to purchase food that meets the basic energy needs of the existing population based on the minimum standard of living. The economic cost of the adjusted dietary structure should not exceed the minimum standard of living of the residents. The economic cost was calculated as the product of food consumption and food unit prices. The prices of various foods were derived from the summary of the national statistical data on agricultural products (NDRC, 2021). The minimum living standards of Beijing and Shanghai in 2021 were obtained from the People’s governments of Beijing municipality and Shanghai municipality (https://www.beijing.gov.cn/, https://www.shanghai.gov.cn/). The formula is:
$F{{P}_{i}}={{{x}'}_{i}}\times {{p}_{f}}$
where FPi represents the cost of food groups after food adjustment; ${{{x}'}_{i}}$ represents the adjusted food consumption; and pf represents the unit price of food.
(3) Objective function setting
To ensure that the optimized diet is environmentally friendly, culturally acceptable and affordable for residents, we established the optimization objective as minimizing both carbon footprint and economic cost, as well as minimizing the deviation of each food and food group consumed. The objective equations are:
$Minimizef(1)=f(DA)$
$Minimizef(2)=\underset{i=1}{\overset{8}{\mathop{\mathop{\sum }^{}}}}\,F{{P}_{i}}$
$Minimizef(3)=\underset{i=1}{\overset{8}{\mathop{\mathop{\sum }^{}}}}\,C{{f}_{i}}$
where Minimizef (1) refers to the objective function with the minimum difference in consumption; Minimizef (2) refers to the objective function with the lowest economic cost; and Minimizef (3) refers to the objective function with minimum carbon emissions.

2.4.2 Constraint setting

We referenced the recommendations for the intake of essential nutrients provided by the Dietary reference intakes for China 2023 (CNS, 2023) to constrain the nutrient intake of the food groups (Table 4). The minimum and maximum calorie requirements for Chinese adults engaged in light physical labor are 1700 kcal and 2150 kcal, respectively. Therefore, we imposed a total energy constraint ranging from 1700 to 2150 kcal. Protein, fat, and carbohydrate intakes were confined to 10%-20%, 20%-30%, and 50%- 65% of total energy, respectively. Furthermore, this study applied constraints on total food quantities to ensure minimal deviation from the current diet. The upper and lower limits for these constraints were set at 120% and 80% of the total weight of the baseline model, respectively. Specifically, the lower limit for grain was set at 90% in order to align with the plant-based diet recommended in the Chinese Dietary Guidelines 2022.
Table 4 Dietary nutrient intake recommendations by the dietary reference intakes for China 2023
Nutrient type Lower value (kcal %E) Upper value (kcal %E) Average
(kcal %E)
Calories 1700 kcal 2150 kcal 1925 kcal
Protein 10%E 20%E 15%E
Fat 20%E 30%E 25%E
Carbohydrate 50%E 65%E 55%E

Note: E stands for energy value where 1 g of protein=4 kcal; 1 g of fat=9 kcal; and 1 g of carbohydrate=4 kcal.

3 Results

3.1 Characteristics of dietary structural evolution in Beijing and Shanghai

Over the past 30 years, the dietary structures of residents in Beijing and Shanghai have reflected a trend of “diversified consumption”, with a notable increase in the consumption of animal-based food that has been particularly pronounced in Beijing (Figure 1). This change in Beijing could be divided into three phases according to the variation trajectory of each food type: “continuous decline”, “fluctuating change” and “slow rise” (Figure 1a-1g). During the first stage, from 1992 to 2004, market economy reforms greatly expanded the food supply types that were available. This led to greater dietary diversity among Beijing residents and a notable rise in animal-based food consumption from 41.01 kg to 63.37 kg. Caloric intake, protein, and carbohydrates showed decreasing trends, while fat intake increased due to higher meat consumption. However, between 2003 and 2004, food consumption in Beijing experienced a brief, sharp decline, likely due to the regulation of refined grain during special sanitary events in 2004. From 2004 to 2016, Beijing residents consciously increased their consumption of vegetables, fruits, and other foods. However, the trend in nutrient intake from various foods showed only minimal changes. Specifically, there was a temporary decline in food consumption by residents during 2015-2016, which was attributed to fluctuating vegetable prices influenced by the weather conditions in 2016. Since 2016, grain consumption has steadily decreased, reaching only 109.9 kg in 2021. However, the greater consumption of vegetables, fruits, and meat has contributed to a gradual increase in total food consumption. In addition, the consumption of animal-based foods has risen to 89.5 kg, representing an increase from 10.0% to 22.2%. Nutrient intake from food also increased gradually during this period, while remaining within a reasonable range.
Figure 1 Evolution of the dietary structures of residents in China’s two first-tier cities

Note: c-g. Changes in food nutrients for residents of the two first-tier cities in China; h. Comparison of dietary structures and balanced dietary patterns of the residents in China’s two first-tier cities.

The total food consumption of Shanghai residents has undergone two stages of evolution: a “fluctuating decline” and a “slow rise” (Figure 1a-1g). The first stage occurred from 1992 to 2011. As in Beijing, the dietary diversity of Shanghai residents increased notably due to the reform of the market economic system, which improved the variety of food that was available. However, there was a deliberate shift towards a balanced diet, accompanied by a decline in the consumption of animal-based foods, from 79.8 kg to 72.5 kg. Caloric, protein, fat and carbohydrate nutrient intakes were on a downward trend towards a reasonable range. From 2011 to 2012, food consumption in Shanghai notably increased due to the establishment of the Shanghai refined grain base. Since 2012, residents have increasingly consumed plant-based products such as grains and fruits, resulting in an overall increase in food consumption. However, the consumption of animal-based foods increased to 114.2 kg, representing an increase from 19.0% to 27.7%. The intake levels of various nutrients also increased slowly. This indicates that the dietary patterns of residents in both Beijing and Shanghai have changed from staple food to non-staple food and from plant-derived food to animal-based food.
Compared with the balanced dietary pattern (Figure 1h), the calorie intake in Beijing is below 187.5 kcal, and the intake of protein, fat, and carbohydrates remains insufficient by 9.0 g, 14.5 g, and 6.2 g, respectively. Shanghai has been more aligned with the balanced dietary pattern, in which calorie intake exceed 81.3 kcal, carbohydrates are less than 1.9 g, and protein and fat intakes are higher than the balanced dietary pattern by 6.6 g and 6.5 g, respectively. Nonetheless, there remains a significant disparity between the actual diet and the balanced dietary pattern, so both cities should prioritize the principle of dietary balance in the future by focusing on increasing their proportions of low- consumption foods.

3.2 Carbon emission effects driven by dietary structural evolution in Beijing and Shanghai

The diversification of dietary structures in the study area from 1992 to 2021 contributed to an increasing trend of food carbon emissions, and Shanghai’s food carbon emissions were always at a high level (Figure 2). Specifically, per capita carbon emissions from food in Beijing increased from 238.1 kg to 326.7 kg (Figure 2a). Among the food types, per capita food carbon emissions from grains decreased from 53.9 kg to 30.4 kg, those from vegetables decreased from 34.0 kg to 26.7 kg, and those from fruits, oilseeds, eggs, dairy, aquatic products, and meat decreased from 1.3 kg, 8.6 kg, 4.6 kg, 3.5 kg, 9.63 kg and 122.6 kg, respectively, to 5.4 kg, 9.4 kg, 14.0 kg, 9.5 kg, 17.44 kg and 214.0 kg. The per capita carbon emissions from meat showed the most significant increase, contributing to a 103.1% change in the per capita food carbon emissions in Beijing. In Shanghai, per capita food carbon emissions increased from 361.1 kg to 452.1 kg, with per capita food carbon emissions from grains decreasing from 49.4 kg to 32.4 kg, while per capita food carbon emissions from fruits, oilseeds, eggs, dairy, aquatic products, and meat increased from 2.8 kg, 13.2 kg, 10.4 kg, 3.4 kg, 40.0 kg and 217.0 kg to 4.2 kg, 13.8 kg, 12.1 kg, 7.6 kg, 48.7 kg and 308.6 kg, respectively (Figure 2b). Among them, per capita carbon emissions from meat showed the most significant increase, contributing to a 100.7% change in the per capita food carbon emissions in Shanghai.
Figure 2 Food carbon emissions of residents in China's two first-tier cities (per capita and total)

Note: Cf refers to food carbon emissions.

Moreover, the total food carbon emission in Beijing increased from 262.4 t to 715.3 t, and the contributions of each food type to the carbon emissions increased by 22.6% (grain), 14.3% (vegetables), 0.5% (fruit), 3.6% (oilseeds), 2.0% (eggs), 1.5% (dairy), 4.0% (aquatic product), and 51.5% (meat), representing changes to 9.3%, 8.2%, 1.7%, 2.9%, 4.3%, 2.9%, 5.3%, and 65.6%, respectively (Figure 2a). Similarly, the total carbon emissions from food in Shanghai increased from 485.7 t to 1125.2 t, and the contributions of the individual food types to the total carbon emissions changed from 13.7%, 6.9%, 0.8%, 3.6%, 2.9%, 0.9%, 11.1%, and 60.1% to 7.2%, 5.5%, 0.9%, 3.1%, 2.7%, 1.7%, 10.8%, and 68.3%, respectively (Figure 2b). The contribution of meat to the total food carbon emissions has been gradually increasing, indicating that meat consumption is the key to influencing the change in food carbon emissions. Additionally, the per capita meat carbon emissions in Shanghai reached 308.6 kg by 2021, which is close to the total per capita food carbon emissions in Beijing. Moreover, the total food carbon emissions in Shanghai exceeded those in Beijing, reflecting the effect of affluence in dietary structures that lead to increased food carbon emissions in urban systems, thereby posing a threat to environmental sustainability.

3.3 Dietary structure adjustment plans for Beijing and Shanghai under multi-objective optimization

The results show that it is feasible to adjust the dietary structures to dietary patterns that are culturally acceptable, affordable and environmentally sustainable (Figure 3). The results indicated that, when considering the optimization objective of cultural acceptability (f(1)), the dietary structure of the residents in Beijing would tend to reduce food consumption (Figure 3a). This would entail consuming 271.0 g of grains, 293.5 g of vegetables, 196.1 g of fruits, and 22.4 g of oilseeds, along with 40.8 g of eggs, 65.3 g of dairy products, 21.9 g of aquatic products, and 90.5 g of meat. This structure would lead to food carbon emissions of approximately 313.1 kg annually, with an economic cost of 3114.09 yuan per year. There is no significant change in food consumption when the economic cost is increased to the minimum, with the food carbon emissions reaching about 312.0 kg, and an economic cost of 3111.56 yuan per year (f(1)&f(2)). When considering the environmental burden (f(1)&f(2)&f(3)), the consumption of plant-based foods increased while the consumption of meat further declined. Specifically, in this scenario, the consumption amounts are grains 280.3 g, vegetables 302.5 g, fruit 203.3 g, oilseeds 22.0 g, eggs 40.6 g, dairy 68.7 g, aquatic products 22.7 g, and meat 77.2 g. If the cultural acceptability, affordability, and environmental sustainability needs are all met, the annual food carbon emissions will reach 284.6 kg and the estimated economic cost would be about 2991.68 yuan per year (Figure 3c, d).
Figure 3 Dietary structure adjustment plans for the two first-tier cities in China. (a) Dietary structure adjustment plans in Beijing; (b) Dietary structure adjustment plans in Shanghai; (c) Food carbon emissions under the proposed dietary adjustment programs; d. Economic costs under the proposed dietary adjustment programs
With cultural acceptability as the optimization goal (f(1)), the dietary structure in Shanghai will tend to reduce food consumption, particularly meat (Figure 3b). Specifically, the consumption amounts are grains 289.0 g, vegetables 361.7 g, fruit 144.5 g, and oilseeds 21.9 g, in addition to 30.8 g of eggs, 78.2 g of dairy products, 61.6 g of aquatic products, and 105.94 g of meat. The estimated food carbon emissions are approximately 378.5 kg, with an economic cost of 3514.63 yuan per year. With a minimal increase in the economic cost, the consumption of animal-based food further declines, resulting in food carbon emissions of 368.7 kg and an economic cost of 3369.33 yuan per year (f(1)&f(2)). Considering the environmental impact (f(1)&f(2)&f(3)), there is an increase in plant-based food consumption and a further reduction in meat consumption. Specifically, this scenario includes 290.3 g of grains, 241.8 g of vegetables, 145.5 g of fruits, 22.0 g of oilseeds, 30.9 g of eggs, 61.4 g of dairy products, 61.2 g of aquatic products, and 105.9 g of meat. If the cultural acceptability, affordability, and environmental sustainability needs are all met, the annual emissions would be about 366.6 kg and the estimated economic cost would be about 3346.27 yuan per year (Figure 3c, d).

4 Discussion

4.1 Comparisons of urban dietary changes in Beijing and Shanghai

In this study, we observed a shift in the dietary patterns in Beijing and Shanghai from over-consumption towards balanced consumption, with Shanghai residents adopting balanced diets earlier than residents in Beijing. This disparity could be attributed to Shanghai's consistently faster urbanization compared to Beijing. In 1993, Shanghai had achieved an urbanization rate of 76.94%, which then increased to 89.3% by 2012. In contrast, Beijing reached only 86.5% by 2016 (http://www.stats.gov.cn/tjsj/). Rapid socioeconomic development has contributed to improvements in dietary habits. The regional dietary structure is also greatly influenced by the market. Since the market economy reform in 1992, influenced by the variety and convenience of market foods (Zhang et al., 2018; Lu et al., 2020), the dietary patterns of residents have transitioned from relatively monotonous to more diversified. However, meat consumption in Shanghai remains high, potentially influenced by the regional food culture.
The overall trend indicates a shift from plant-based to animal-based foods in Beijing and Shanghai. One issue is that the per capita food carbon emissions in Beijing and Shanghai have exceeded the average level for China (Zhu et al., 2024a). In addition, the risk of diet-related diseases among populations has been gradually increasing (GBD 2017 Diet Collaborators, 2019; Freund and Springmann, 2021). Thus, adapting the diets in regions that are receptive to cultures and resources is crucial for enhancing both population health and environmental sustainability. This can offer a feasible way to adjust diets while considering economic and cultural factors, with a focus on using the first-tier cities as pioneers for reshaping the national dietary structure.

4.2 Comparisons of global representative diets and dietary structure adjustment plans for the residents in Beijing and Shanghai

We compared the dietary structure adjustment programs in Beijing and Shanghai with global representative diets. The results indicate that the dietary structure adjustments for Beijing and Shanghai remain in a state of nutritional balance (Table 5). Compared with the current dietary pattern in 2021, the adjusted food consumption in these cities changes very little, is more balanced, and includes nutrient intakes that fall within reasonable ranges, with less food carbon emissions, and lower economic cost. In addition, compared to the balanced dietary pattern, the diet adjustment plan still shows low fruit consumption, while the consumption of oilseeds and meat are moving towards more reasonable levels. Furthermore, compared to the globally representative dietary patterns (such as the Lancet diet), the differences in grain, fruit, and meat consumption between the dietary adjustment program and these patterns would decrease, with the differences in grain reduced to 1.7 g and 7.7 g for Beijing and Shanghai, and 3.3 g for fruit consumption in Beijing. The differences in meat consumption would also decrease by 14.6 g and 26.4 g for Beijing and Shanghai.
Table 5 Dietary structure adjustment plans for residents in the first-tier cities in China and global representative diets
Food type Actual in 2021 Adjusted The balanced
diets		 The Lancet
diet		 Mediterranean
diet		 Flexitarian diet
Beijing Shanghai Beijing Shanghai
Grain (g) 301.1 320.8 280.3 290.3 250.0 282.0 240.0 325.0
Vegetables (g) 326.0 301.9 302.5 241.8 400.0 300.0 600.0 500.0
Fruit (g) 211.5 166.0 203.3 145. 5 275.0 200.0 376.0 350.0
Oilseeds (g) 18.6 27.4 22.0 22.0 27.5 47.0 47.0 28.0
Eggs (g) 44.4 38.4 40.6 30.9 57.5 13.0 21.0 45.0
Dairy (g) 81.6 65.8 68.7 61.4 400.0 250.0 250.0 300.0
Aquatic product (g) 27.4 76.4 22.7 61.2 57.5 28.0 70.0 58.0
Meat (g) 91.8 132.3 77.2 105.9 57.5 43.0 34.0 16.0
Calories (kcal) 1814.2 2083.0 1700.4 1782.9 2001.7 1895.2 2099.5 2128.8
Protein (g) 71.5 87.1 64.4 73.6 80.5 61.4 75.4 80.7
Fat (g) 64.7 85.8 62.5 70.4 79.2 82.1 85.7 69.6
Carbohydrate (g) 235.3 239.6 218.9 212.5 241.5 227.5 257.3 293.7
Cf (kg yr-1) 326.7 452.1 284.6 366.6 313.9 232.9 266.0 219.9

Note: Globally representative dietary data refer to the study of Zhu et al. (2023c).

As the objective function increases, the dietary patterns in Beijing and Shanghai exhibit a tendency towards greater consumption of plant-based foods, particularly grains, and a decline in meat consumption, which lead to the lower carbon emissions and economic costs in both cities. Moreover, the dietary adjustment program would reduce per capita food carbon emissions by 42.1 kg and 85.5 kg in Beijing and Shanghai, respectively, compared to the current diet model. Specifically, Beijing’s adjustment plan would reduce per capita food carbon emissions by 29.3 kg compared to the balanced diet model. However, due to significantly higher-than-average meat consumption, Shanghai would remain above the balanced diet model, which can also be seen in comparison with the global representative diet. In addition, the diet adjustment plans in these cities are more affordable in relation to the disposable incomes of Beijing and Shanghai residents (75002 yuan and 78027 yuan, respectively) as well as their minimum living security standards (1320 yuan month-1 and 1330 yuan month-1, respectively). Given the substantial populations of these international metropolitan areas in China, the adjustments of dietary for residents in Beijing and Shanghai could potentially lead to reductions of 92.14 t and 212.65 t of total carbon emissions compared with the actual in 2021, respectively.

4.3 Policy implications

Given the rapid urbanization and fast-paced lifestyle in the first-tier cities, implementing healthier, fast-paced diets can mitigate the occurrence of unbalanced diets. The disposable incomes of residents in Beijing (75002 yuan) and Shanghai (78027 yuan) consistently exceed the national average income standard for Chinese residents (35128 yuan), and they are approximately nine times higher than those of lower-income residents (8333 yuan). The total food consumption and animal-based food consumption in first-tier cities also consistently exceed the Chinese averages, which has also happened in other Chinese megacities such as Tianjin (Gao et al., 2019). The significant food consumption demands in first-tier cities will lead to heightened environmental pressures in food-producing regions, particularly those supplying animal-based products. If this dietary trend continues along a similar trajectory, the dual pressures of carbon reduction and resource input are likely to intensify in food- source regions such as Henan and Inner Mongolia (Xiong et al., 2018), thereby hindering coordinated regional development.
As great model systems, it has been found that substituting the excessive consumption of meat and dairy products with plant-based foods and shellfish can effectively control diseases and enhance the environmental conditions (Xu et al., 2021; Woodside et al., 2022; Kawasaki et al., 2024). In this context, the sustainable importance of plant-based diets should be highlighted considering the increasing demand for meat in China’s first-tier cities (Sahlin et al., 2020; Sun et al., 2022). On one hand, there should be a conscious adjustment of consumer behavior. In addition to developing national dietary guidelines, enhancing information on rational diets and their environmental impact is crucial for increasing public awareness about transitioning to sustainable diets, while also considering the nutritional needs of diverse populations. On the other hand, promoting urban agriculture and encouraging cities to collaborate in providing essential financial and technical support to their food-supply areas can facilitate the adoption of low-carbon and resource-efficient agricultural practices (Ferreira et al., 2018; Hu and Wang, 2020; Xiong et al., 2020; Chen et al., 2022). Moving forward, there is an urgent need for multifaceted efforts to promote the transition of dietary patterns to meet nutritional needs while aligning with environmental sustainability.

4.4 Limitations

This study has several limitations. First, the categorization and nutrient compositions of the diets are not sufficiently detailed to encompass specific food types and variations in micronutrient intake due to data limitations. Sugars were also not included in the dietary analysis. Second, the data in this study have some uncertainties, particularly regarding statistics from the statistical yearbook and the carbon emission coefficients of different foods, and uncertainty estimation was not conducted. Third, the multi-objective optimization model could also account for the realization of sustainable consumption. Future enhancements could include integrating constraint functions for micronutrient intake and defining food prices based on regional wholesale prices. Furthermore, sustainable dietary strategies can be extended to support other environmental goals, such as enhancing the sustainable use of land and water resources. Future research can also examine dietary variations among socioeconomic groups within a given country or city to develop more targeted adjustment programs.

5 Conclusions

The food system has become an important factor influencing global climate change, and food consumption has become an important medium and carrier that links population health and the environment. Shifting dietary patterns towards sustainability is the key to realizing global sustainability goals. In this study, we assessed the changes in dietary structure and their carbon emissions effects in China’s first-tier cities from 1992 to 2021. We constructed a multi- objective optimization model to investigate sustainable dietary adjustment strategies that would be aligned with cultural acceptance, affordable nutrition, health, and environmental friendliness. Based on this analysis, several conclusions can be drawn.
With urbanization and increasing affluence, the dietary structures of residents in both Beijing and Shanghai have shifted towards “diversified consumption”, primarily characterized by increased consumption of animal-based foods that now exceed the recommended dietary guidelines. This trend has resulted in increasing food carbon emissions, with both per capita and total food carbon emissions increasing in Beijing and Shanghai, and Shanghai exhibits relatively higher levels of these emissions. Additionally, we found that the contribution of meat to the total food carbon emissions is gradually increasing, making it a crucial factor driving food-related carbon emissions. By developing the multi- objective optimization, we found that adjusting the consumption of food, especially grains and meat, can effectively meet the requirements for obtaining a nutritionally balanced and healthy diet that aligns with cultural preferences, affordability, and environmental sustainability. The diets of residents in China’s two first-tier cities have shown trends that pose risks to both public health and environmental sustainability. Consequently, the transformation of their dietary structures will be the key to achieving sustainable consumption. In the future, different regions should proactively implement policies aimed at encouraging rational eating habits among the general public. This approach will not only promote the stable development of food and environmental systems, but it will also address the demand for nutritious, wholesome, and environmentally friendly food among residents. Such endeavors hold substantial importance for achieving global sustainability objectives.

References

Publishing order | Descend order by publishing year | Descend order by cited within
[1]
Abejón R, Batlle-Bayer L, Laso J, et al. 2020. Multi-objective optimization of nutritional, environmental and economic aspects of diets applied to the Spanish context. Foods, 9(11): 1677. DOI: 10.3390/foods9111677.

[2]
Bastianoni S, Pulselli FM, Tiezzi E. 2004. The problem of assigning responsibility for greenhouse gas emissions. Ecological Economics, 49 (3): 253-257.

[3]
Beylot A, Secchi M, Cerutti A, et al. 2019. Assessing the environmental impacts of EU consumption at macro-scale. Journal of Cleaner Production, 216: 382-393.

DOI PMID

[4]
Chaudhary A, Krishna V. 2019. Country-specific sustainable diets using optimization algorithm. Environmental Science & Technology, 53(13): 7694-7703.

[5]
Chen C, Yang J, Gao J L, et al. 2022. An observation over the rural-urban re-connecting process based on the Alternative Food Network (AFN) in China—From the perspective of ‘social capital’. Habitat International, 130: 102708. DOI: 10.1016/j.habitatint.2022.102708.

[6]
Chen Q R, Wu M Y, Zeng H C, et al. 2023. The effect of upgrading the dietary structure of residents on non-grain production of cultivated land. Journal of Resources and Ecology, 14(6): 1350-1360.

DOI

[7]
Clark M A, Springmann M, Hill J, et al. 2019. Multiple health and environmental impacts of foods. Proceedings of the National Academy of Sciences of the USA, 116: 23357-23362.

[8]
Clune S, Crossin E, Verghese K. 2017. Systematic review of greenhouse gas emissions for different fresh food categories. Journal of Cleaner Production, 140: 766-783.

[9]
CNS (Chinese Nutrition Society). 2022. Chinese dietary guidelines 2022. Beijing, China: People’s Medical Publishing House. (in Chinese)

[10]
CNS (Chinese Nutrition Society). 2023. Dietary reference intakes for China 2023. Beijing, China: People’s Medical Publishing House. (in Chinese)

[11]
Crippa M, Solazzo E, Guizzardi D. et al. 2021. Food systems are responsible for a third of global anthropogenic GHG emissions. Nature Food, 2: 198-209.

DOI PMID

[12]
Deb K, Pratap A, Agarwal S. 2002. A fast and elitist multi-objective genetic algorithms: NSGA-II. IEEE Transactions on Evolutionary Computation, 6(2): 182-197.

[13]
Donati M, Menozzi D, Zighetti C, et al. 2016. Towards a sustainable diet combining economic, environmental and nutritional objectives. Appetite, 106: 48-57.

DOI PMID

[14]
Eshel G, Shepon A, Makov T, et al. 2014. Land, irrigation water, greenhouse gas, and reactive nitrogen burdens of meat, eggs, and dairy production in the United States. Proceedings of the National Academy of Sciences of the USA, 111(33): 11996-12001.

[15]
FAO. 2016. Plates, pyramids, planet—Developments in national healthy and sustainable dietary guidelines: A state of play assessment. Rome, Italy: FAO.

[16]
Ferreira A J D, Guilherme R I M M, Ferreira C S S, et al. 2018. Urban agriculture, a tool towards more resilient urban communities? Current Opinion in Environmental Science & Health, 5: 93-97.

[17]
Freund F, Springmann M. 2021. Policy analysis indicates health-sensitive trade and subsidy reforms are needed in the UK to avoid adverse dietary health impacts post- Brexit. Nature Food, 2: 502-508.

DOI PMID

[18]
Gao B, Huang W, Wang L, et al. 2019. Driving forces of nitrogen flows and nitrogen use efficiency of food systems in seven Chinese cities, 1990 to 2015. Science of the Total Environment, 676: 144-154.

[19]
GBD 2017 Diet Collaborators. 2019. Health effects of dietary risks in 195 countries, 1990-2017: A systematic analysis for the global burden of disease study 2017. Lancet, 393(10184): 1958-1972.

DOI PMID

[20]
Green R, Milner J, Dangour A D, et al. 2015. The potential to reduce greenhouse gas emissions in the UK through healthy and realistic dietary change. Climatic Change, 129: 253-265.

[21]
Hu Q Y, Wang C. 2020. Quality evaluation and division of regional types of rural human settlements in China. Habitat International, 105: 102278. DOI: 10.1016/j.habitatint.2020.102278.

[22]
IPCC. 2022. Climate change 2022: Mitigation of climate change. https://www.ipcc.ch/report/ar6/wg3/. Viewed on 2024-06-05.

[23]
Kawasaki Y, Nagao-Sato S, Shimpo M, et al. 2024. Understanding sustainable dietary behaviors in Japanese and German adults: A cross-cultural comparison. Resources, Conservation and Recycling, 207: 107330. DOI: 10.1016/j.resconrec.2023.107330.

[24]
Larrea-Gallegos G, Vázquez-Rowe I. 2020. Optimization of the environmental performance of food diets in Peru combining linear programming and life cycle methods. Science of the Total Environment, 699: 134231. DOI: 10.1016/j.scitotenv.2019.134231.

[25]
Liu J, Shively G E, Binkley J K. 2014. Access to variety contributes to dietary diversity in China. Food Policy, 49: 323-331.

[26]
Lu C X, Liu A M, Xiao Y, et al. 2020. Changes in China’s grain production pattern and the effects of urbanization and dietary structure. Journal of Resources and Ecology, 11(4): 358-365.

DOI

[27]
Mazzetto A M, Falconer S, Ledgard S. 2023. Carbon footprint of New Zealand beef and sheep meat exported to different markets. Environmental Impact Assessment Review, 98: 106946. DOI: 10.1016/j.eiar.2022.106946.

[28]
Miller V, Webb P, Cudhea F, et al. 2022. Global dietary quality in 185 countries from 1990 to 2018 show wide differences by nation, age, education, and urbanicity. Nature Food, 3: 694-702.

DOI PMID

[29]
Muñoz-Martínez J, Elías R A, Batlle-Bayer L, et al. 2023. Optimizing sustainable, affordable and healthy diets and estimating the impact of plant-based substitutes to milk and meat: A case study in Spain. Journal of Cleaner Production, 424: 138775. DOI: 10.1016/j.jclepro.2023.138775.

[30]
Ndlovu T, van Jaarsveld F, Caleb O J. 2019. French and Mediterranean-style diets: Contradictions, misconceptions and scientific facts—A review. Food Research International, 116: 840-858.

[31]
NDRC (Department of Price, National Development and Reform Commission). 2021. Data compilation of cost and income of national agricultural products. Beijing, China: China Statistics Press. (in Chinese)

[32]
Perignon M, Masset G, Ferrari G, et al. 2016. How low can dietary greenhouse gas emissions be reduced without impairing nutritional adequacy, affordability and acceptability of the diet? A modelling study to guide sustainable food choices. Public Health Nutrition, 19(14): 2662-2674.

DOI PMID

[33]
Poore J, Nemecek T. 2018. Reducing food’s environmental impacts through producers and consumers. Science, 360: 987-992.

DOI PMID

[34]
Popkin B M. 2004. The nutrition transition: An overview of world patterns of change. Nutrition Reviews, 62: S140-S143.

[35]
Sahlin K R, Röös E, Gordon L J. 2020. ‘Less but better’ meat is a sustainability message in need of clarity. Nature Food, 1: 520-522.

[36]
Sala S, Serenella S, Lorenzo B L, et al. 2019. Consumption and consumer footprint: Methodology and results. Luxembour, Luxembourg: Publications Office of the European Union.

[37]
Sievert K, Lawrence M, Naika A, et al. 2019. Processed foods and nutrition transition in the Pacific: Regional trends, patterns and food system drivers. Nutrients, 11(6): 1328. DOI: 10.3390/nu11061328.

[38]
Song G, Gao X, Fullana-I-Palmer P, et al. 2019. Shift from feeding to sustainably nourishing urban China: A crossing-disciplinary methodology for global environment-food-health nexus. Science of the Total Environment, 647: 716-724.

[39]
Springmann M, Spajic L, Clark M A, et al. 2020. The healthiness and sustainability of national and global food based dietary guidelines: Modelling study. British Medical Journal, 370: m2322. DOI: 10.1136/bmj.m2322.

[40]
Springmann M, Wiebe K, Mason-D’Croz D, et al. 2018. Health and nutritional aspects of sustainable diet strategies and their association with environmental impacts: A global modelling analysis with country- level detail. Lancet. Planetary Health, 2(10): e451-e461.

[41]
Sun Z, Scherer L, Tukker A, et al. 2022. Dietary change in high-income nations alone can lead to substantial double climate dividend. Nature Food, 3: 29-37.

DOI PMID

[42]
Temme E H, Toxopeus I B, Kramer G F, et al. 2015. Greenhouse gas emission of diets in the Netherlands and associations with food, energy and macronutrient intakes. Public Health Nutrition, 18(13): 2433-2445.

DOI PMID

[43]
Tilman D, Clark M. 2014. Global diets link environmental sustainability and human health. Nature, 515(7528): 518-522.

[44]
Tom M S, Fischbeck P S, Hendrickson C T. 2016. Energy use, blue water footprint, and greenhouse gas emissions for current food consumption patterns and dietary recommendations in the US. Environment Systems and Decisions, 36: 92-103.

[45]
Wang L, Cui S H, Hu Y C, et al. 2021. The co-benefits for food carbon footprint and overweight and obesity from dietary adjustments in China. Journal of Cleaner Production, 289: 125675. DOI: 10.1016/j.jclepro.2020.125675.

[46]
Wang X, Qiang W, Liu X, et al. 2022. The spatiotemporal patterns and network characteristics of emissions embodied in the international trade of livestock products. Journal of Environmental Management, 322: 116128. DOI: 10.1016/j.jenvman.2022.116128.

[47]
WBG. 2017. World development indicators. Washington DC, USA. http://data.worldbank.org/data-catalog/world-development-indicators.

[48]
Willett W, Rockstr¨om J, Loken B, et al. 2019. Food in the Anthropocene: the EAT-Lancet Commission on healthy diets from sustainable food systems. Lancet, 393(10170): 447-492.

DOI PMID

[49]
Woodside J, Young I S, McKinley M C. 2022. Culturally adapting the Mediterranean diet pattern—A way of promoting more “sustainable” dietary change? The British Journal of Nutrition, 128(4): 693-703.

[50]
Wu H, MacDonald G K, Galloway J N, et al. 2022. A new dietary guideline balancing sustainability and nutrition for China’s rural and urban residents. iScience, 25: 1120. DOI: 10.1016/j.isci.2022.105048.

[51]
Xiong X, Zhang L, Hao Y, et al. 2020. Urban dietary changes and linked carbon footprint in China: A case study of Beijing. Journal of Environmental Management, 255: 109877. DOI: 10.1016/j.jenvman.2019.109877.

[52]
Xiong X, Zhang L, Zhang P, et al. 2018. Dynamics of urban food metabolism with water-carbon footprint respond: A case study of Beijing. Journal of Natural Resources, 33(11): 1886-1896.. (in Chinese)

DOI

[53]
Xu X, Sharma P, Shu S, et al. 2021. Global greenhouse gas emissions from animal-based foods are twice those of plant-based foods. Nature Food, 2: 724-732.

DOI PMID

[54]
Yin J, Yang D, Zhang X, et al. 2020. Diet shift: Considering environment, health and food culture. Science of the Total Environment, 719: 137484. DOI: 10.1016/j.scitotenv.2020.137484.

[55]
Yin J, Zhang X, Huang W, et al. 2021. The potential benefits of dietary shift in China: Synergies among acceptability, health, and environmental sustainability. Science of the Total Environment, 779: 146497. DOI: 10.1016/j.scitotenv.2021.146497.

[56]
Zhang S, Min S, Ma R. 2018. Urbanization, food consumption transformation and its impact on ecological environment. Urban Development Studies, 25(3): 13-20. (in Chinese)

[57]
Zhu X H, Zhang Y, Zhu Y Y, et al. 2024a. The shift to plant-based dietary patterns in China would have significant effects on cultivated land resources. Science Bulletin, 69(6): 737-740.

[58]
Zhu Y Y, Wang Z W, Zhu X H. 2023a. New reflections on food security and land use strategies based on the evolution of Chinese dietary patterns. Land Use Policy, 126: 106520. DOI: 10.1016/j.landusepol.2022.106520

[59]
Zhu Y Y, Wang Z W, Zhu X H. 2023b. Utilization effect of water-land resources under the evolution of Chinese dietary patterns. Journal of Geographical Sciences, 33(4): 741-759.

[60]
Zhu Y Y, Zhang Y, Zhu X H. 2023c. The evolution process, characteristics and adjustment of Chinese dietary structure guidelines: A global perspective. Resources, Conservation & Recycling, 193(6): 106964. DOI: 10.1016/j.resconrec.2023.106964.

[61]
Zhu Y Y, Zhang Y, Zhu X H. 2024b. Carbon emission effects driven by evolution of Chinese dietary structure from 1987 to 2020. Chinese Geographical Science, 34(1): 181-194.

Options
Outlines

/

〈 〉