Climatic Change

, Volume 116, Issue 2, pp 249–262 | Cite as

The global warming potential of two healthy Nordic diets compared with the average Danish diet

  • Henrik Saxe
  • Thomas Meinert Larsen
  • Lisbeth Mogensen
Article

Abstract

The potential greenhouse gas (GHG) emissions from the production of food for three different diets are compared using consequential Life Cycle Assessment. Diet 1 is an Average Danish Diet (ADD); diet 2 is based on the Nordic Nutritional Recommendations (NNR), whilst diet 3 is a New Nordic Diet (NND) developed by the OPUS project. The NND contains locally produced Nordic foods where more than 75 % is organically produced. NNR and NND include less meat and more fruit and vegetables than the ADD. All diets were adjusted to contain a similar energy and protein content. The GHG emissions from the provision of NNR and NND were lower than for ADD, 8 % and 7 % respectively. If GHG emissions from transport (locally produced versus imported food) are also taken into account, the difference in GHG emissions between NND and ADD increases to 12 %. If the production method (organic versus conventional) is taken into account so that the ADD contains the actual ratio of organically produced food (6.6 %) and the NND contains 80 %, the GHG emissions for the NND are only 6 % less than for the ADD. When the NND was optimised to be more climate friendly, the global warming potential of the NND was 27 % lower than it was for the ADD. This was achieved by including less beef, and only including organic produce if the GHG emissions are lower than for the conventional version, or by substituting all meat with legumes, dairy products and eggs.

1 Introduction

The consumption of food and beverages in the EU is responsible for 22–31 % of total greenhouse gas (GHG) emissions from overall private consumption (Tukker et al. 2009, Weidema et al. 2008). Obviously we have to eat and drink, but what we choose to eat and drink has a great impact on the GHG emissions from food production. Meat and dairy production typically cause greater GHG emissions than the production of fruit and vegetables (e.g. Audsley et al. 2009, Stehfest et al. 2009, Vieux et al. 2012, Williams et al. 2006) and therefore it is logical to assume that reducing meat and increasing fruit and vegetables in the typical Western diet would decrease the global warming potential (GWP) of food and beverage consumption. However, substituting animal produce with fruit, vegetables and legumes, results in a reduced intake of protein and some essential nutrients. To compensate for this, plant-based diets must be varied and include greater volumes of food.

This study compares the GWP of two healthy Nordic diets with the average Danish diet (ADD). For a fair comparison, all diets were modified to contain the same total amount of energy and protein per person per year. With a comparable content of energy and protein, environmentally sustainable diets typically reduce the overall GWP by 5–10 % below that of existing diets (Saxe 2011, Tukker et al. 2009, Wallén et al. 2004). Macdiarmid et al. (2011) found a 25 % reduction in the GWP of a diet that met the UK’s nutritional recommendations compared to the average UK diet.

This study is a small part of the Danish OPUS project (Mithril et al. 2012), the aim of which is to develop, introduce and test a science-based NND which is simultaneously palatable, healthy and environmentally sustainable.

The rationale for the present study is to establish whether the NND is indeed environmentally sustainable, and if it can be improved for further mitigation. The rationale is concretised in the form of three aims for this paper: (1) to evaluate the GWP of two healthy Nordic diets (NND and NNR) relative to the Average Danish Diet (ADD); (2) to look for further reductions in the GWP of the basic NND; and (3) to provide specific recommendations for the design of healthy and climate-friendly diets.

This study used the GWP as an indicator for the environmental sustainability of diets as it is the most common environmental indicator. But it must be emphasised that the GWP is only one of several impact categories used for evaluating environmental effects. The GWP does not by itself give a complete picture of all the environmental effects related to diets, though Saxe (2011) found that many other impact categories point in the same direction as the GWP.

2 Methods and materials

2.1 Calculation of GHG emissions

The GHG emissions of food and beverages were calculated by consequential life cycle assessment (cLCA). The functional unit was 1 kg of produced food or beverage available at the local supermarket. The scope of the calculations included the GHG emissions associated with all the activities and use of materials from soil to supermarket. All anthropogenic GHGs were converted to CO2-equivalents according to the IPCC (2007).

The carbon footprint of the food items used in this study was mainly based on the Danish LCA Food Database (2004), using the SimaPro® software, and calculated with the Stepwise2006 method (Weidema 2009). When the LCA-Food database lacked information, supplementary data which best fitted the Danish production conditions, were taken from the literature (references quoted below in Table 1). The Potential GHG emissions were all calculated according to the ISO standard 14040, but limited to GHG emissions.
Table 1

Composition of the 3 diets ADD (Average Danish Diet), NNR (a diet based on the Nordic Nutritional Recommendations) and NND (New Nordic Diet), and emissions of GHGs from the diets in the three basic scenarios. The diets were normalised to have an equal energy and protein content

Product categories

Diet composition (kg/person/year)

<–––––––––––––––––––– Emission of greenhouse gases ––––––––––––––––––––> (Kg CO2-equivalent/person/year)

Mass distribution of the 30 OPUS categories of food and beverage (excl. water).

Scenario 1: Effects of altered content on GHG emissions

Scenario 2: Adding effects of international trans-portation on GHG emissions(3)

Scenario 3: Adding effects of substituting conv. with org. produce on GHG emissions

 

ADD

NNR

NND

ADD

NNR

NND

ADD

NNR

NND

ADD

NNR

NND

Beer, wine, alcoholabcd

115

45.9

57.2

154

62.0

52.8

177

71.1

52.3

177

71.1

52.3

Berriese

3.5

6.4

65.1

2.4

4.4

44.6

2.8

5.3

44.6

2.8

5.3

44.6

Butterf

2.6

0.5

1.3

16.7

3.1

8.3

16.9

3.2

8.3

16.6

3.1

7.5

Cabbageg

6.1

11.6

18.3

1.4

2.6

4.1

2.5

4.7

4.1

2.5

4.7

4.1

Candy, sweets, etc.e

20.3

11.1

10.1

124

67.8

61.8

142

87.8

61.8

142

87.8

61.8

Cheeseagh

13.4

15.2

21.9

154

174

251

155

174

251

154

174

248

Coffee, tea, cocoa(1)bij

15.7

8.0

7.8

119

60.1

6.8

119

60.1

6.8

119

60.1

6.8

Conveniencea

5.2

4.6

2.6

4.0

3.5

2.0

4.6

4.0

2.0

4.6

4.0

2.0

Dairy productsadefjklmn

138

167

198

167

206

238

167

206

238

162

199

214

Eggsabehj

8.0

19.8

23.8

16.0

39.5

47.4

16.1

39.8

47.4

17.3

42.7

60.1

Fruit. excl. berriesehk

85.6

150

129

46.0

81.5

43.2

68.9

122

43.2

69.6

123

117

Herbsb

1.8

1.7

5.2

1.6

1.5

14.1

1.9

1.8

14.1

1.9

1.8

14.1

Jame

3.8

6.1

0.1

2.0

3.0

0.1

3.5

5.4

0.1

3.5

5.4

0.1

Juice(2) dh

45.5

24.9

143

45.5

24.9

62.4

50.7

27.7

62.3

51.7

28.2

27.7

Legumesabe

3.6

5.5

15.2

1.7

2.5

7.5

2.3

3.5

7.5

2.3

3.5

7.5

Meataefjloprrstu

74.9

61.6

53.5

739

563

514

743

566

514

743

566

576

Mushrooms, lettucee

7.8

12.0

7.8

8.8

13.6

8.7

10.2

15.7

8.7

10.2

15.7

8.7

Nutse

1.6

1.4

11.0

0.8

0.7

4.7

2.0

1.8

4.7

2.0

1.8

4.7

Oils excl. rapee

11.7

16.4

0.0

29.1

52.6

0.0

29.4

55.7

0.0

29.4

55.6

0.0

Oils of rapeev

0.0

0.0

11.7

0.1

0.0

41.4

0.1

0.0

41.4

0.1

0.0

35.2

Pasta, industrialbs

6.2

5.9

0.0

5.6

5.3

0.0

7.1

7.2

0.0

6.8

7.0

0.0

Potatoesahjs

58.0

94.3

87.0

12.3

19.8

18.5

15.6

24.9

18.5

15.4

24.7

17.3

Roots, excl. potatoese

19.7

31.1

98.6

3.7

5.8

18.3

7.3

11.5

18.3

7.6

11.8

22.9

Ricebs

3.0

4.7

0.0

10.4

16.5

0.0

11.8

18.7

0.0

11.8

18.7

0.0

Seafood and fisha

11.2

21.2

25.1

35.6

67.6

81.7

38.1

72.3

81.7

38.1

72.3

81.7

Soft drinksbd

119

30.9

0.0

16.6

4.3

0.0

19.3

5.3

0.0

19.3

5.3

0.0

Sugaraw

4.9

3.4

2.5

4.8

3.3

2.4

4.8

3.3

2.4

4.8

3.3

2.4

Vegetables (other)abfhjsx

45.5

61.3

61.5

136

194

184

143

202

184

148

209

237

Wheat, proc. products.ab

39.0

35.4

0.0

33.0

29.3

0.0

33.5

29.7

0.0

33.3

29.7

0.0

Whole grain products.ab

39.2

66.3

84.0

30.5

50.7

65.3

35.3

61.6

65.3

34.9

61.3

60.4

Other productsay

1.5

0.6

3.3

1.8

0.5

3.4

1.8

0.5

3.4

1.8

0.5

3.4

Total

910

920

1140

1920

1760

1790

2030

1890

1790

2030

1900

1920

References:a.LCA Food (2004); b. Saxe et al.(2006); c. Ardente et al. (2006); d.Hanssen et al. (2007); e. Audsley et al. (2009, values excluding LUC); f. Personal estimate involving new data from DJF; g. Berlin (2002): h. Wallén et al. (2004); i. Salomone (2003); j. Williams et al. (2006); k. Foster et al. (2006); l. Weidema et al. (2008); m. Weiske et al. (2006); n. Thomassen et al. (2008); o. Halberg et al. (2010); p. Halberg et al. (2008); q. Dalgaard et al. (2007); r. Nguyen et al. (2010); s. Carlsson-Kanyama (1998); t. Cederberg (2003); u. Kilic et al. (2010); v. Schmidt (2010); w. Renouf et al. (2008); x. Halberg et al. (2006); y. Anderson et al. (1998).Comments:(1) Dry mass; (2) 84 % apple juice +16 % vegetable juice; (3) Cooling and freezing during transportation of any product was based on data for a small diesel generator taken from the database Ecoinvent®, the ratio of countries of origin, where import distances, transfer time, and means of transport were supplied by Statistics Denmark (2011) or derived from personal estimates. For most product categories, several emission factors were employed since they each contained many different products.

2.2 The three diets

This study assesses the GWP associated with three diets. The Average Danish Diet (ADD) is the reference against which the GWP of the two healthy Nordic diets (NNR and NND) are measured. The ADD constitutes more than 300 food and beverage items or categories supplied to the average Dane for private consumption (Saxe et al. 2006, Table 5 of that reference).

The NNR diet was designed by modifying the ADD according to the Nordic Nutrition Recommendations (Nordic Council 2004). The exact composition of the NNR diet is available from Saxe et al. (2006, Table 8 of that reference).

The New Nordic Diet (NND) was designed to be a healthy, palatable, and environmentally friendly diet of Nordic origin in accordance with the Danish dietary guidelines (Astrup et al. 2005) and the OPUS dietary recommendations (Mithril et al. 2012) supported by a few hundred recipes for all seasons. That the NND is of Nordic origin means that it is inspired by the Nordic diet of the olden days, with a higher content of, e.g. roots, berries, nuts, fish, and whole grain products, and a lower content of animal produce than the ADD. The NND recommendations can be distilled into three guiding principles. First of all, an altered content of major groups of food and beverages compared with the ADD (Table 1, third and first columns). Secondly, all food and beverages are locally produced, i.e. no international transport. Thirdly, at least 75 % of the food products are organically produced. These three NND recommendations are analysed consecutively as illustrated by the three scenarios in Table 1 and Fig. 1.
Fig. 1

Content of the three diets (kg/person/year) (Columns 1–3) and GHG emissions (kg CO2-eq/person/year) from the diets throughout the three basic scenarios when the diets are balanced to have equal energy and protein content. ADD: Average Danish Diet. NNR: Diet based Nordic Nutritional Recommendations. NND: New Nordic Diet

2.3 Balancing the energy and protein content in the 3 diets

The energy and protein content of the 3 diets was intentionally balanced to secure the most reasonable foundation for comparing the relative GWP of the diets. The overall energy and protein content of the three diets was calculated using the DANKOST3000® software. Before balancing the healthy diets, the NNR and the NND were a little lower in energy and protein compared to the ADD. By adding 6 kg of both cheese and eggs per year to the NNR and 12 kg of both cheese and eggs per year to the NND plus 120 kg apples/year (as 0.3 l of apple juice/day), the healthy diets obtained an energy and protein content equal to the ADD.

2.4 Three basic scenarios

Scenario 1 describes the GWP of the three diets according to their differing composition of food items.

Scenario 2 and later scenarios add the GHG emissions associated with the international transportation of imported products in the ADD and the NNR, while the NND only includes the GHG emissions associated with local transportation. The GHG emissions associated with the international transportation were calculated by combining data on: (i) the share of each import-country for each imported food product (Statistics Denmark 2011); (ii) the distance transported measured from the midpoint of each producer country to the midpoint of Denmark (http://www.viamichelin.com); (iii) the means of transportation and associated GHG emissions: truck, ship or both; and (iv) contributions from cooling and freezing during transportation simulated by data for electricity use (Ecoinvent 2010).

Scenario 3 and later scenarios add the GHG emissions associated with the production method (organically versus conventionally produced food). The GHG emissions of the NND were reduced by 11 ‘positive’ (means lower GHG from organic compared to conventional products) organic food items or categories (beer, coffee, dairy products, lamb, pasta, rapeseed oil, pork, potato, oats, rye bread, and whole wheat products), while the GHG emissions were increased by seven ‘negative’ organic food items or categories (apples, beef, carrots, chicken, eggs, non-alcoholic beverages, tomatoes). For the ADD and the NNR, the actual national ratio between organically and conventionally produced food and beverages of each item was applied, which resulted in the diets having an overall content of 6.6 % organic products (Økologisk Landsforening 2009). For the NND, all of the 18 above mentioned organically produced food and beverage products were included as such, which resulted in the diet having an overall content of 80 % organic products.

2.5 Scenarios for further mitigation

In scenarios 4–8 (Table 2, Fig. 2), the NND is modified for further mitigation of the GHG emissions using the ADD of scenario 3 as the ongoing reference.
Table 2

Emissions of GHGs associated with the New Nordic Diet (NND) in scenarios 3–8 compared with the emissions from the reference Average Danish Diet (ADD) in scenario 3. In scenario 4, the content of chicken, beef and other meat in the NND of scenario 3 was multiplied by 2, 0.5 and 0.67 respectively, and in scenario 5 they were multiplied by 3.0, 0.2 and 0.2 respectively. In scenario 6, the content of organics was reduced to the ADD level in scenario 3 (6.6 %). In scenario 7, only the ‘positive’ organics were included. Scenario 8 is an ovo-lacto-vegetarian version of NND. All scenarios contain equal energy and protein levels

Product categories

<––––––––––––––––––– Emission of greenhouse gases –––––––––––––––––––> (Kg CO2-equivalent/person/year)

 

Scenario 3

Scenario 4

Scenario 5

Scenario 6

Scenario 7

Scenario 8

 

ADD

NND

NND

NND

NND

NND

NND

Beer, wine, alcoholabcd

177

52.3

52.3

52.3

52.8

52.3

0.0

Berriese

2.8

44.6

44.6

44.6

44.6

44.6

44.6

Butterf

16.6

7.5

7.5

7.5

8.2

7.5

0.0

Cabbageg

2.5

4.1

4.1

4.1

4.1

4.1

4.1

Candy, sweets, etc.e

142

61.8

61.8

61.8

61.8

61.8

0.0

Cheeseagh

154

248

248

248

251

248

373

Coffee, tea, cocoa(1)bij

119

6.8

6.8

6.8

6.8

6.8

6.8

Conveniencea

4.6

2.0

2.0

2.0

2.0

2.0

0.0

Dairy productsadefjklmn

162

214

214

214

230

214

429

Eggsabehj

17.3

60.1

60.1

60.1

50.8

50.8

81.3

Fruit. excl. berriesehk

69.6

117

117

117

85.7

85.7

85.7

Herbsb

1.9

14.1

14.1

14.1

14.1

14.1

14.1

Jame

3.5

0.1

0.1

0.1

0.1

0.1

0.1

Juice(2) dh

51.7

27.7

27.7

27.7

23.2

23.2

23.2

Legumesabe

2.3

7.5

7.5

7.5

7.5

7.5

74.8

Meat(3) aefjloprrstu

743

576

411

311

237

235

0.0

Mushrooms, lettucee

10.2

8.7

8.7

8.7

8.7

8.7

8.7

Nutse

2.0

4.7

4.7

4.7

4.7

4.7

4.7

Oils excl. rapee

29.4

0.0

0.0

0.0

0.0

0.0

0.0

Oils of rapeev

0.1

35.2

35.2

35.2

41.4

35.2

35.2

Pasta, industrialbs

6.8

0.0

0.0

0.0

0.0

0.0

0.0

Potatoesahjs

15.4

17.3

17.3

17.3

18.3

17.3

17.3

Roots, excl. potatoese

7.6

22.9

22.9

22.9

18.8

18.8

18.8

Ricebs

11.8

0.0

0.0

0.0

0.0

0.0

0.0

Seafood and fisha

38.1

81.7

81.7

81.7

81.7

81.7

0.0

Soft drinksbd

19.3

0.0

0.0

0.0

0.0

0.0

0.0

Sugaraw

4.8

2.4

2.4

2.4

2.4

2.4

0.0

Vegetables (other)abfhjsx

148

237

237

237

190

190

190

Wheat, proc. products.ab

33.3

0.0

0.0

0.0

0.0

0.0

0.0

Whole grain products.ab

34.9

60.4

60.4

60.4

64.7

60.4

60.4

Other productsay

1.8

3.4

3.4

3.4

3.4

3.4

3.4

Total

2030

1920

1750

1650

1510

1480

1480

References a-y: see Table 1

Fig. 2

GHG emissions (kg CO2-eq/person/year) from the New Nordic Diet (NND) throughout scenarios 4–8 of further mitigation relative to the average Danish Diet (ADD) of scenario 3. In scenario 4, the content of chicken, beef and other meat in the NND of scenario 3 was multiplied by 2, 0.5 and 0.67 respectively, and in scenario 5 they were multiplied by 3.0, 0.2 and 0.2 respectively. In scenario 6, the content of organics was reduced to the ADD level in scenario 3 (6.6 %). In scenario 7, only the ‘positive’ organics were included. Scenario 8 is a vegetarian version of NND. All scenarios contain equal energy and protein levels

In scenarios 4 and 5, the composition of meat is changed, whilst maintaining the weight, energy and protein content constant. The content of chicken, beef and other types of meat was multiplied by the factors 2.0, 0.5 and 0.67, respectively (scenario 4), or 3.0, 0.2 and 0.2, respectively (scenario 5).

In scenarios 6 and 7, the content of organic produce is changed. Scenario 6 is like scenario 5, but with the overall content of organic produce in the NND reduced to 6.6 %, i.e. to the level of the ADD reference in scenario 3. Scenario 7 is like scenario 5, but with the ‘negative’ organic products being excluded.

Scenario 8 is an ovo-lacto vegetarian version of scenario 7. Meat, fish and seafood were substituted with other protein-rich foods by including two times more dairy products, 1.6 times more eggs, 1.5 times more cheese, and 10 times more legumes. Alcoholic beverages, sweets, butter or convenience foods are omitted for improved health.

3 Results

3.1 The three basic scenarios

Table 1 gives the composition of the three diets (kg/person/year) and the emissions of GHG (Kg CO2-equivalent/person/year) based on the composition of the three diets (scenario 1), added emissions due to international transport (scenario 2), and the added effect of substituting a percentage of the conventional produced products with organically produced products (scenario 3). For Table 1, the 300 foods and beverages included in this study were condensed into the 30 food categories defined by the OPUS New Nordic Diet. All items which did not fit into these groups were pooled under category 31, ‘other products’. Including wasted or spoiled food, all diets in all scenarios contained 13.2 MJ/person/day and 137 g protein/person/day.

Figure 1 shows the GWP hotspots of all diets and scenarios, where the number of food categories was reduced to 11 by only including the categories which contributed more than 5 % to the total GHG emissions. The remaining food and beverage items were pooled in category 12, the ‘sum of others’. Figure 1 illustrates that the animal produce categories are the hotspots with a high contribution to the overall GHG-emission in all diets and scenarios.

Scenario 1 shows that the altered composition of the NNR and the NND reduced the GHG emissions by 8 % and 7 % respectively, compared to the ADD.

Scenario 2 shows that the inclusion of GHG emissions from transport related to imports reduces the GHG emissions of the NNR and the NND to 7 % and 12 % respectively, compared to the ADD. Only the GWP of the NND was reduced more than in scenario 1

Scenario 3 shows that substituting 6.6 % of the conventional produce with organic produce in both the ADD and the NNR makes little difference to GHG emissions of both diets (Table 1). This is because the inclusion of the positive and negative effects on the GWP counterbalanced each other in these diets. However, substituting 80 % of the conventional produce with organic produce in the NND increased the emission from this diet by 130 kg to 1920 kg CO2-eq/person/year (Table 1), or by 7 % relative to scenario 2. Thereby, the GWP of the NND is increased in scenario 3 so that it is only 5.7 % better than the ADD.

3.2 Further mitigation through modified diets

Scenarios 4–8 investigate possibilities for the further mitigation of GHG emissions from diet choice than the basic NND offered in scenario 3. As previously mentioned, all diets and scenarios contained 13.2 MJ/person/day and 137 g protein/person/day.

Figure 2 shows that the animal produce categories are still the hotspots with the highest contribution to the overall GHG-emission in all scenarios, typically making up just over half of the total GHG-emissions.

Scenarios 4 and 5 show the significant effect on the GWP of changing the type of meat eaten. If less beef and more chicken and other types of meat are included in the NND, the GHG emissions can be reduced by 170 kg or 270 kg in scenarios 4 and 5 respectively (Table 2) compared to the 1920 kg CO2-eq/person/year of the NND in scenario 3, or by 14 % and 19 % respectively (Fig. 2) compared to the 6 % reduction of the NND in scenario 3.

Scenario 6 shows that when the content of organic produce in the NND was reduced to 6.6 %, the GHG emissions were only 1510 kg CO2-eq/person/year (Table 2), or 25 % lower than the ADD of scenario 3 (Fig. 2).

Scenario 7 shows that the GHG emissions of the NND were reduced to 1480 kg CO2-eq/person/year (Table 2), or 27 % lower than the ADD of scenario 3, when only the ‘positive’ organic products were included in the NND instead of similar conventional products.

Scenario 8 shows that an ovo-lacto-vegetarian version of the NND was not a more efficient strategy for reducing the GWP than scenario 7. Both resulted in a 550 kg CO2-eq/person/year reduction (Table 2) which equals a 27 % reduction in potential GHG emissions relative to the ADD of scenario 3 (Fig. 2) when all three diets had the same energy and protein content.

4 Discussion

4.1 Comparison with other studies

When comparing the results from this study with others, the chosen energy and protein normalisation gave lower GWP reductions for the healthy diets compared with the standard diet, than if the original recipes had been used. Only a few studies have a simultaneous focus on the environmental consequences of diets and the nutritional aspects. The two healthy Nordic diets, the NNR and the NND, potentially improved the GWP of eating and drinking relative to the average Danish diet (ADD) by 6–7 %. This is in the same range as found for the WHO/EFSA-recommended healthy diets where a reduced content of red meat and dairy reduced the overall environmental impact by 4–8 %, relative to the average EU-27 diet (Tukker et al. 2009). The Livewell study by Macdiarmid et al. (2011) is another example of a recent study with a dual focus on environment and health. The study concluded that it is possible to reduce the GWP of the UK diet by 25 % by making small changes to a familiar, normal and varied diet. However, the Livewell study achieved this reduction by lowering the meat and dairy content, while accepting lower protein content than the present UK diet, as long as it remained within the minimum requirements. The authors argue for lower protein content because it contributes to a lower energy intake (Macdiarmid et al. 2011). Audsley et al. (2009) modelled the reductions in GWP of the UK average diet with consumption measures comparable to those in this study. They confirm a striking alignment between healthy eating and GWP-reducing consumption measures such as reducing the intake of livestock products. Taking the higher meat consumption in Denmark into account, the results of Audsley et al. (2009). are in line with the results of this study.

Worldwide, the livestock sector is responsible for 18 % of total greenhouse gas emissions (Garnett 2008). Feeding wheat, barley and soy to domestic animals, rather than using these plant products directly for human consumption, increases the GWP of our protein intake. Domestic animals eat many times their final meat or milk weight in feed. Typically, however, the animal protein is more concentrated and digestible, and it is increasingly the preferred protein of the majority, though meat production and consumption are having an increasingly negative environmental impact, including global warming.

4.2 Six lessons for developing healthy and climate-friendly diets

This study has identified six lessons for developing healthy and climate friendly diets.

Lesson 1 (refers to scenario 1) is that reducing the intake of meat and dairy products, particularly beef, significantly reduces the GWP of diets in agreement with findings by Audsley et al. (2009), Cederberg et al. (2011), Foster et al. (2006), Macdiarmid et al. (2011), Tukker et al. (2009), and Vieux et al. (2012). It is easy to design a climate friendly diet if it is low in energy and protein. However, the OPUS project seeks to reduce the GWP while maintaining a high level of protein in the NND, both quantitatively and qualitatively to prevent obesity and chronic diseases (Larsen et al. 2011).

Lesson 2 (refers to scenario 2) is that the inclusion of local produce may reduce the GWP due to reduced transportation and cooling/freezing en route. For the NND, this reduction was 5.4 %. However, in the present study, the actual production conditions in the country that a particular food item is imported from are not taken into account, and in some cases the GHG emissions from production abroad will be lower and in some cases higher than the GHG emissions from local production (Saunders et al. 2006). Especially for animal products with a high GHG emission per kg product, the significance of the GHG contribution from transportation might be less important as the relative GHG contribution from transportation is much smaller than the GHG contribution from production, whereas for vegetable products it might be the opposite.

Lesson 3 (refers to scenario 3 compared with scenario 2) is that substituting conventionally produced products with organically produced analogues can have an overall negative effect on the GWP of a diet. The overall result depends both on the absolute content of organics included in a diet (scenario 2, 3 and 6) and on the ratio of ‘negative’ and ‘positive’ organics (scenario 7). Furthermore, the result is dependent on the assumptions, as some studies find higher GHG emissions of, for example, organic milk compared with conventional milk, whereas in other studies it is opposite (Thomassen et al. 2008). These findings are probably due to the large variation in GHG emissions per kg milk between farms in both the organic and the conventional system (Kristensen et al. 2011). In agreement with our finding, Niggli et al. (2008) showed that the product type (plant or meat) has a much greater influence on GHG emissions than the method of production whether it is organic or conventional. In a larger perspective, section 4.3 (see below) cautions against a simple interpretation of organic products being negative for the environmental impact of diets.

Lesson 4 (refers to scenarios 4 and 5) is that the composition of meat in a diet has a significant impact on its GWP. Eating less red meat, particularly beef, and more white meat, such as chicken and pork, whilst keeping the weight, energy and protein content constant, reduces the GWP of the NND by 9 to 14 % (NND in scenario 4 and 5 respectively relative to NND in scenario 3). This is in agreement with Audsley et al. (2009).

Lesson 5 (as deducted from the basic scenarios in Table 1) is that beer, wine and alcohol contribute 9 %, sweets and candy 7 %, and coffee, tea and cocoa 6 % to the total GWP of the average Danish diet (ADD). The aggregated GWP of these products for the ADD is 22 %, which is more than half of the meat’s 37 % share (Fig. 1). For health reasons, the content of alcoholic drinks, hot drinks and sweets and candy in the basic NND (using the preliminary OPUS-definitions from 2010) was reduced to 50 % of that of the ADD. This is just as efficient as reducing the meat intake by 30 %.

Lesson 6 (refers to scenarios 7 and 8) is that a vegetarian diet does not necessarily reduce the GWP of a diet more than an optimised omnivorous diet with a high degree of meat-to-meat substitution and where only the ‘positive’ organic products are included—even with energy and protein normalisation. However, if LUC (see section 4.4 below) had been included, the vegetarian version of the NND (scenario 8) would have had a lower GWP than the omnivorous version of the NND (scenario 7) (Audsley et al. 2009).

4.3 Caution: Functional unit, limited scope and missing data

The result that substituting conventional with organic produce overall has a negative effect on the GWP of a diet must be treated with caution if the results are to be implemented as diet recommendations. First of all, the GWP was calculated using the functional unit ‘kg’ food, as is common practice in this type of LCA, to mirror the need to feed the global population. But if the priority is on long-term sustainability in feeding the World, a more appropriate functional unit would be ‘ha crop land’. But the lower food production per ha in organic production compared to conventional production is an immediate problem in a world with an increasing demand for land for food and bioenergy, and is the reason behind the negative effect of substituting conventional products with organics on the GWP when using the functional unit ‘kg’.

Secondly, if the priority of the LCA of diets is to investigate their long-term sustainability, more impact categories than the GWP must be included. Measured by other impact categories, organically produced food typically performs better than conventionally produced food, e.g. retention of soil carbon by applying organic fertilisers, improved ecotoxicity and human toxicity by limiting the use of pesticides, hormones and antibiotics, improved fertility, animal welfare, and biodiversity (Mäder et al. 2002).

Thirdly, the available data on damage caused by pesticides only include a limited part of the complete life cycle of pesticides and associated consequences.

4.4 Land-use change (LUC)

The UN has recently predicted that the world population will grow to 9.3 billion by 2050. The FAO (2009) has calculated that world food production will consequently have to increase by 70 %, which will require the expansion of arable land in developing countries, mainly sub-Saharan Africa and Latin America, by around 120 million hectares. Furthermore, the increasing demand for organically grown foods, associated with lower land use efficiency, and a growing demand for biofuels will put further pressure on agricultural commodities and land use in the future (Edwards et al. 2010). Even today, the increasing supply of feed and food, particularly to the industrialised countries, is indirectly responsible for significant LUC (Cederberg et al. 2011).

Since the GHG emissions from LUC were not included in this study, the total impact of food and beverages on climate change is underestimated, especially for beef (Cederberg et al. 2011) imported from South America where it causes deforestation, the most harmful type of LUC (Dalgaard et al. 2008).

4.5 Implementing healthy diets

The development of diets that are both healthy and sustainable (e.g. Audsley et al. 2009; Macdiarmid et al. 2011, Tukker et al. 2009, this study) is important. The crucial step, however, is to get people to prefer these diets. Despite regular campaigning for healthier eating, the populations of the Nordic countries have not adopted healthier dietary habits. Differential taxes on the ‘good’ vs. the ‘bad’ components of diets may direct more people towards healthy and sustainable diets, but new taxes are so far met with political opposition. In the OPUS project, we test whether consumers can be tempted by the high palatability of the New Nordic Diet. Additionally, branding by famous chefs and the publication of several hundred NND recipes will hopefully encourage a reasonable proportion of the Nordic population to make the transition to a healthier and more sustainable, locally grown diet. The NND was developed with the aim that ingredients are easily available and do not induce additional cost in terms of buying or in terms of time spent on preparing the meals. The vision is that the OPUS's New Nordic Diet will have a significant, positive impact on health and become a realistic strategy for the mitigation of climate change.

5 Conclusion

This study evaluated the GWP of two healthy Nordic diets relative to the Average Danish Diet and found that a change towards the Nordic diets supports climate change mitigation due to their content of less animal produce and more fruit and vegetables. The fact that the produce is local in the New Nordic Diet increased mitigation, while the substitution of large amounts of conventionally produced foods with organically produced foods decreased mitigation.

This study found ways to further reduce the GWP of the basic NND. The GWP can be cut by nearly 30 %. The main driver was to change the composition of meat to include less red meat, particularly beef, and more white meat, such as chicken and pork. The impact of the product type (plant vs. meat products), even without including the effects on land use change, is much greater regarding GHG emissions than the effect of buying either imported or local produce, or of choosing either conventional or organic produce.

This study summarises the analyses into six specific recommendations for designing healthy and climate friendly diets. A well-designed diet simultaneously contributes to a better environment in terms of GWP and to improved health.

Notes

Acknowledgements

The study is part of the OPUS project 'Optimal well-being, development and health for Danish children through a healthy New Nordic Diet. Supported by a grant from the Nordea Foundation. We thank the Danish National Food Institute and 2.-0 LCA Consultants for providing the data on the volumes of food and beverages produced to make up the three diets

References

  1. Anderson K, Ohlsson T, Ohlsson P (1998) Screening LCA of tomato ketchup. J Clean Prod 6:277–288. doi:10.1016/S0959-6526(98)00027-4 CrossRefGoogle Scholar
  2. Ardente F, Beccali G, Cellura M, Marvuglia A (2006) Poems: A case study of an Italian wine-producing firm. Envir Man 38:350–364. http://www.isa.utl.pt/der/ASAmb/DocumentosAulas/ParaAlunos/ACV_vinho.pdf. Accessed 12 April 2012.
  3. Astrup A, Andersen NL, Stender S, Trolle E (2005) Kostrådene 2005. http://www.meraadet.dk/gfx/uploads/rapporter_pdf/Kostraadene%202005_www.pdf. Accessed 12 April 2012.
  4. Audsley E, Brander M, Chatterton J, Murphy-Bokern D, Webster C, Williams A (2009) How low can we go? An assessment of greenhouse gas emissions from the UK food system and the scope to reduce them by 2050. Cranfield University, UK http://assets.wwf.org.uk/downloads/how_low_can_we_go.pdf. Accessed 12 April 2012.
  5. Berlin J (2002) Environmental life cycle assessment (LCA) of Swedish semi-hard cheese. Int Dairy J 12:939–953. doi:10.1016/S0958-6946(02)00112-7 CrossRefGoogle Scholar
  6. Carlsson-Kanyama A (1998) Climate change and dietary choices—how can emissions of greenhouse gases from food consumption be reduced? Food Policy 23:277–293. doi:10.1016/S0306-9192(98)00037-2 CrossRefGoogle Scholar
  7. Cederberg C (2003) Life cycle assessment of animal products. In: Mattsson B, Sonesson U (eds.) Environmental Friendly Food Processing. CRC Press. ISBN 1 85573 677 2.Google Scholar
  8. Cederberg C, Persson UM, Neovius K, Molander S, Clift R (2011) Including carbon emissions from deforestation in the Carbon Footprint of Brazilian Beef. Envir Sci & Tech 45:1773–1779. doi:10.1021/es103240z. http://pubs.acs.org/doi/pdfplus/10.1021/es103240z. Accessed 12 April 2012Google Scholar
  9. Dalgaard R, Schmidt J, Halberg N, Christensen P, Thrane M, Pengue WA (2008) LCA of Soybean Meal. Int J LCA 13:240–254. doi:10.1065/lca2007.06.342 CrossRefGoogle Scholar
  10. Dalgaard R, Halberg N, Hermansen JE (2007) Danish pork production. An environmental assessment. DJF Animal Science 82, 34pp. http://www.lcafood.dk/djfhus82ny.pdf. Accessed 12 April 2012
  11. Ecoinvent (2010) Swiss Centre for Life Cycle Inventories. Ecoinvent database version 2.1. Process: Diesel, burned in diesel-electric generating se/GLO U'. http://www.ecoinvent.ch/. Accessed 12 April 2012
  12. Edwards R, Mulligan D, Marelli L (2010) Indirect Land Use Change from increased biofuels demand. JRC Scientific and Technical Reports. http://ec.europa.eu/energy/renewables/studies/doc/land_use_change/study_4_iluc_modelling_comparison.pdf. Accessed 12 April 2012
  13. FAO (2009) How to feed the World in 2050. Expert paper. http://www.fao.org/fileadmin/templates/wsfs/docs/expert_paper/How_to_Feed_the_World_in_2050.pdf . Accessed 12 April 2012
  14. Foster C, Green K, Bleda M, Dewick P, Evans B, Flynn A, Mylan J (2006) Environmental impacts of food production and consumption. A report produced for the Department for Environment, Food and Rural Affairs. http://www.ifr.ac.uk/waste/Reports/DEFRA-Environmental%20Impacts%20of%20Food%20Production%20%20Consumption.pdf. Accessed 12 April 2012
  15. Garnett T (2008) Cooking up a Storm. Food, greenhouse emissions, and our changing climate. Food Climate Research Network. Centre for Environmental Strategy, University of Surrey. http://www.fcrn.org.uk/fcrn. Accessed 12 April 2012
  16. Halberg N, Dalgaard R, Rasmussen MD (2006) Miljøvurdering af konventionel og økologisk avl af grøntsager. Danish Environmental Protection Agency. Working Report No. 5. http://www2.mst.dk/udgiv/publikationer/2006/87-7614-960-9/pdf/87-7614-961-7.pdf. Accessed 12 April 2012 (Danish)
  17. Halberg N, Hermansen JE, Kristensen IS, Eriksen J, Tvedegaard N (2008) Comparative environmental assessment of three systems for organic pig production in Denmark. Proceedings of ISOFAR Conference: Organic Agriculture in Asia, Korea 13–14 March. Pp. 249–261. http://orgprints.org/15585/1/15585.pdf. Accessed 12 April 2012
  18. Halberg N, Hermansen JE, Kristensen IS, Eriksen J, Tvedegaard N, Petersen BM (2010) Impact of organic pig production systems on CO2 emission, C sequestration and nitrate pollution. Agron Sustain Dev. 30:721–731. http://www.agronomy-journal.org/index.php?option=article&access=doi&doi=10.1051/agro/2010006. Accessed 12 April 2012
  19. Hanssen OJ, Rukke E-L, Saugen B, Kolstad J, Hafrom P, Krogh L von, Raadal HL, Rønning A, Wigum KS (2007) The environmental effectiveness of the beverage sector in Norway in a factor 10 perspective. Int J Life Cycle Assess 12:257–265. http://www.springerlink.com/content/g6550442q5j957w3/. Accessed 11 May 2012.Google Scholar
  20. IPCC (2007) IPCC Fourth Assessment Report: Climate Change 2007 (AR4). http://www.ipcc.ch/publications_and_data/publications_and_data_reports.shtml. Accessed 12 April 2012
  21. Kilic A, Midilli A, Dincer I (2010) A strategic program to reduce greenhouse gas emissions from food industry. In: Dincer et al. (eds.), Global Warming, Green Energy and Technology, Springer Science+Business Media, LLC2010. doi:10.1007/978-1-4419-1017-2_10. Accessed 12 April 2012
  22. Kristensen T, Mogensen, L. Knudsen MT, John E Hermansen (2011) Effect of production system and farming strategy on greenhouse gas emissions from commercial dairy farms in a life cycle approach. Livestock Sci. 140:136–148. doi:10.1016/j.livsci.2011.03.002. Accessed 12 April 2012
  23. Larsen TM, Dalskov S-M, Baak M van, Jebb SA, Papadaki A, Pfeiffer AFH, Martinez JA, Handjieva-Darlenska T, Kunesová, Pihlsgård M, Stender S, Holst C, Saris WHM, Astrup A (2011) Diets with high or low protein content and glycemic index for weight-loss maintenance. New England J Med 363:2102–2113. http://www.nejm.org/doi/pdf/10.1056/NEJMoa1007137. Accessed 12 April 2012.
  24. LCA Food Database (2004) http://www.lcafood.dk. Accessed 11 May 2012.
  25. Macdiarmid J, Kyle J, Horgan G, Loe J, Fyfe C, Johnstone A (2011) Livewell: a balance of healthy and sustainable food choices. WWF. Rowett Institute of Nutrition and Health, University of Aberdeen. http://assets.wwf.org.uk/downloads/livewell_report_jan11.pdf. Accessed 12 April 2012
  26. Mäder P, Fliessbach A, Dubois D, Gunst L, Fried P, Niggli U (2002) Soil Fertility and Biodiversity in Organic Farming. Science 296:1694–1697. doi:10.1126/science.1071148 CrossRefGoogle Scholar
  27. Mithril C, Dragsted LO, Meyer C, Blauert E, Holt MK, Astrup A (2012) Guidelines for the New Nordic Diet. Public Health Nutrition. doi:10.1017/S136898001100351X
  28. Niggli U, Schmid H, Fliessbach A (2008) Organic farming and climate Change. FiBL. International Trade Centre (ITC), Geneva. Technical Paper, 30pp. http://orgprints.org/13414/3/niggli-etal-2008-itc-climate-change.pdf. Accessed 12 April 2012
  29. Nguyen TLT, Hermansen JE, Mogensen L (2010) Environmental consequences of different beef production systems in the EU. J Clean Prod 18:756–766. doi:10.1016/j.jclepro.2009.12.023 CrossRefGoogle Scholar
  30. Nordic Council (2004) Nordic Nutrition Recommendations 2004—integrating nutrition and physical activity. http://www.norden.org/da/publikationer/publikationer/2004-013. Accessed 12 April 2012
  31. Renouf MM, Wegener MK, Nielsen LK (2008) An environmental life cycle assessment comparing Australian sugarcane with US corn and UK sugar beet as producers of sugars for fermentation. Biomass Bioenergy 32:1144–1155. doi:10.1016/j.biombioe.2008.02.012 CrossRefGoogle Scholar
  32. Salomone R (2003) Life cycle assessment applied to coffee production: Investigating environmental impacts to aid decision making for improvements at company level. Food, Agric Envir 1:295–300. http://isites.harvard.edu/fs/docs/icb.topic267876.files/Coffee%20Life%20Cycle.pdf. Accessed 12 April 2012
  33. Saunders C, Barber A, Taylor G (2006) Food miles—comparative energy/emissions performance of New Zealand’s agriculture Industry. Research Report No. 205. http://www.lincoln.ac.nz/documents/2328_rr285_s13389.pdf. Accessed 12 April 2012
  34. Saxe H (2011) Diet as a healthy and cost-effective instrument in environmental protection. In: Nriagu JO (ed.) Encyclopedia of Environmental Health, volume 2, pp. 70–82 Burlington: Elsevier. http://www.waterstones.com/waterstonesweb/products/jerome+o-+nriagu/encyclopedia+of+environmental+health/7968186/. Accessed 12 April 2012
  35. Saxe H, Jensen RB, Petersen ML (2006) Fødevarers Miljøeffekter. Det politiske ansvar og det personlige valg. Institute of Environmental Assessment. English abstract. http://www.dors.dk/graphics/Synkron-Library/Publikationer/IMV/2006/fdevare.pdf. Accessed 12 April 2012 (in Danish)
  36. Schmidt J (2010) Comparative life cycle assessment of rapeseed oil and palm oil. Int J Life Cycle Assess 15:183–197. doi:10.1007/s11367-009-0142-0 CrossRefGoogle Scholar
  37. Statistics Denmark (2011) Value of import and export after SITC main categories, country and import and export. http://www.statistikbanken.dk/statbank5a/default.asp?w=1280
  38. Stehfest E, Bouwman L, van Vuuren DP (2009) Climate benefits of changing diet. Clim Chang 95:83–102. doi:10.1007/s10584-008-9534-6 CrossRefGoogle Scholar
  39. Thomassen MA, van Calker KJ, Smits MCJ, Iepema GL, de Boer IJM (2008) Life cycle assessment of conventional and organic milk production in the Netherlands. Agric Syst 96:95–107. doi:10.1016/j.agsy.2007.06.001 CrossRefGoogle Scholar
  40. Tukker A, Bausch-Goldbohm S, Verheijden M, de Koning A, Kleijn R, Wolf O, Domínguez IP (2009) Environmental impacts of diet changes in the EU. JRC European Comission. European Joint Research Centre. Institute of Prospective Technological Studies. Report no. EUR 23783 EN. ftp://ftp.jrc.es/pub/EURdoc/JRC50544.pdf. Accessed 12 April 2012
  41. Vieux F, Darmon N, Touazi D, Soler LG (2012) Greenhouse gas emissions of self-selected individual diets in France: Changing the diet structure or consuming less? Ecological Economics 75:91–101. http://dx.doi.org/10.1016/j.ecolecon.2012.01.003. Accessed 12 April 2012
  42. Wallén A, Brandt N, Wennersten R (2004) Does the Swedish consumer’s choice of food influence greenhouse emissions? Environ Sci & Pol 7:525–535. doi:10.1016/j.envsci.2004.08.004 CrossRefGoogle Scholar
  43. Weidema BP, Wesnæs M, Hermansen J, Kristensen T, Halberg N (2008) Environmental Improvement Potentials of Meat and Dairy Products. JRC Scientific and Technical Report. EUR 23491 EN. ftp://ftp.jrc.es/pub/library/Impro_Meet.pdf. Accessed 12 April 2012
  44. Weidema BP (2009) Using the budget constraint to monetarise impact assessment results. Ecol Econ 68:1591–1598. doi:10.1016/j.ecolecon.2008.01.019 CrossRefGoogle Scholar
  45. Weiske A, Vabitsch A, Olesen JE, Schelde K, Michel J, Friedrich R, Kaltschmitt M (2006) Mitigation of greenhouse gas emissions in European conventional and organic dairy farming. Agric Acosys & Envir 112:221–232. doi:10.1016/j.agee.2005.08.023 CrossRefGoogle Scholar
  46. Williams AG, Audsley E, Sandars DL (2006) Determining the environmental burdens and resource use in the production of agricultural and horticultural commodities. Defra project report ISO205. Natural Resource Management Institute, Cranfield University. Silsoe Research Institute.Google Scholar
  47. Økologisk Landsforening (2009) Økologisk markedsnotat. http://www.okologi.dk/media/229957/markedsnotat2009.pdf. Accessed 12 April 2012 (in Danish)

Copyright information

© Springer Science+Business Media B.V. 2012

Authors and Affiliations

  • Henrik Saxe
    • 1
  • Thomas Meinert Larsen
    • 2
  • Lisbeth Mogensen
    • 3
  1. 1.Institute of Food and Resource Economics, Faculty of ScienceCopenhagen UniversityCopenhagenDenmark
  2. 2.Department of Human Nutrition, Faculty of ScienceCopenhagen UniversityCopenhagenDenmark
  3. 3.Department of AgroecologyAarhus UniversityAarhusDenmark

Personalised recommendations