Global Biodiversity Loss by Freshwater ... - ACS Publications

Subscriber access provided by UNIV OF NEBRASKA - LINCOLN

Article

Global biodiversity loss by freshwater consumption and eutrophication from Swiss food consumption Laura Scherer, and Stephan Pfister Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.6b00740 • Publication Date (Web): 02 Jun 2016 Downloaded from http://pubs.acs.org on June 8, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 37

Environmental Science & Technology

Global biodiversity loss by freshwater

1

2

consumption and eutrophication from Swiss food

3

consumption

4

Laura Scherer* and Stephan Pfister

5

Institute of Environmental Engineering, ETH Zurich, 8093 Zurich, Switzerland

6 7

*

Institute of Environmental Engineering, ETH Zurich, John-von-Neumann-Weg 9, 8093 Zurich, Switzerland, phone: +41 44 632 31 72, e-mail: [email protected]

8 9

ABSTRACT

10

We investigated water-related resource use, emissions and ecosystem impacts of food

11

consumed in Switzerland. To do so, we coupled LCA methodologies on freshwater

12

consumption, freshwater eutrophication and the consequent local and global biodiversity

13

impacts with Swiss customs data and multi-regional input-output analysis. Most of the

14

resource use, emissions and impacts occur outside the national boundaries which illustrates

15

the extent of environmental outsourcing facilitated by international trade. Countries that are

16

severely affected by Swiss food consumption include Spain, the United States and Ecuador.

17

Cocoa, coffee and almonds stood out as products with high impacts. By identifying spatial 1

ACS Paragon Plus Environment


18

hotspots and impactful products, awareness of policy-makers as well as individual consumers

19

can be raised and efforts of detailed assessments can be streamlined. However, political and

20

economic constraints and the resistance by individual consumers limit the high potential of

21

changes in diets and trade relations to decrease the environmental impacts of food.

22 23

TOC ART

24 25 26 27 28

KEYWORDS Biodiversity loss; Freshwater consumption; Freshwater eutrophication; Life cycle assessment (LCA); Multi-regional input-output analysis (MRIO)

29 30

INTRODUCTION

31

The world is facing the twin challenges of satisfying a growing food demand while

32

reducing its impact on the environment. The global population is growing and with a

33

simultaneous increase in affluence there is also a rising calorie intake per capita as well as

34

diets changing towards a higher consumption of animal products. While agriculture is 2


Page 2 of 37

Page 3 of 37


35

degrading land and water resources, an increase in food production is constrained by the same

36

resources and has to compete with biofuel production. Also biodiversity is severely threatened

37

by our current food production practices.1,2 Biodiversity not only contributes to the provision

38

of ecosystem services to humans such as water purification, flow regulation and carbon

39

storage,3 but also has an intrinsic value irrespective of its societal benefits.4 Altogether, there

40

are two urgent needs: an increasing food supply and the preservation of biodiversity and

41

ecosystem services.

42

With advancing globalization, trade gains in importance and leads to increasing spatial

43

disconnection between production and consumption.5,6 Switzerland, as many other developed

44

countries, is running an ecological deficit and requires external resources from an area about

45

six times larger than their own ecologically productive land to produce food, wood products

46

and energy.7 More recent studies confirm that more land is occupied abroad to satisfy Swiss

47

demand for crops than within its own borders.8,9 Lenzen et al.10 analysed biodiversity loss

48

driven by international trade and identified Switzerland as a net importer of biodiversity

49

threats and as having a biodiversity footprint that is higher abroad (132 species threats) than

50

domestically (56 species threats). The biodiversity in developing countries is especially

51

threatened by consumption in developed countries.

52

A useful tool to link local consumption to global impacts is the coupling of life cycle

53

assessment (LCA) with multi-regional input-output analysis (MRIO).6 LCA is a methodology

54

to quantify the environmental impacts of products from “cradle to grave”.11 For some impact

55

categories spatially explicit methods were developed, which includes water consumption and

56

eutrophication as well as their impacts on biodiversity.12,13 However, the link between the

57

production at one location and the consumption elsewhere is missing. MRIO can fill this gap, 3



58

as it links various regions by their trade patterns.14 The aspect of water was missing in

59

previous studies on Swiss trade with agricultural products.8 In contrast to most studies on

60

virtual water trade that only analyse water inventories,15 this study assesses the water-related

61

impacts of traded food products. Additionally, direct trade was combined with MRIO and

62

FAO data to improve the results of the trade analysis.

63

The study aims at estimating the biodiversity impacts caused by Swiss food consumption

64

and thereby makes use of the model described above. It tests the hypothesis that Switzerland

65

is heavily outsourcing biodiversity impacts9,10 and considers the impact categories water

66

consumption and eutrophication due to their high significance in agricultural production. The

67

products and countries with the biggest effect on biodiversity loss will be identified.

68 69

METHODS

70

Linking food products to crops and countries of origin

71

A list of food items produced in, imported to and exported from Switzerland is obtained

72

from the Swiss Farmers’ Union (Obrist, personal communication). In addition, the Swiss

73

Customs Administration provided a more detailed list of imports into Switzerland including

74

the exporting countries (Obrist, personal communication). The food items include derived

75

crop products and livestock products. Life cycle inventories are only given for crops so that

76

processed products have to be converted to the required crop inputs. Likewise, only direct

77

trading partners are given even if they are re-exporters. Therefore some trade origins have to

78

be traced back. Product conversions and trade analysis are described below.

4


Page 4 of 37

Page 5 of 37


79

Derived crop products. Crop derived products such as chocolate were assigned to their

80

main ingredient which is cocoa beans in this case. The major ingredient and its mass fraction

81

(fmass) for all products are obtained from FAO16 and Chapagain and Hoekstra.17 In order to

82

avoid double counting where the same crop can produce multiple derived products, economic

83

allocation was applied by multiplying the mass of the root product (mrp) needed to produce

84

the derived product with its monetary value fraction (fvalue) provided by Chapagain and

85

Hoekstra:17

86

mrp = mdp · fmass · fvalue (1)

87

where mdp is the mass of the derived product.

88

Where the above mentioned sources did not provide major ingredients, mass or value

89

fractions, own assumptions were established mostly based on product derivations of similar

90

products (Supporting information S1).

91

Livestock products. For livestock products, the cultivation of animal feed is taken into

92

account. The FAO delivers the production volume of different animal products for three

93

farming systems (extensive, intensive or mixed) at the global level18 and of meat in general

94

for six world regions.19 The two scales were merged to get product- and region-specific shares

95

(Supporting Information S1):

f′, = f , ∙ f , / f, f, = f′, /

f′,

96

where fsystem,i indicates the fraction of the farming system i (extensive, intensive or mixed)

97

for each product in the region, fproduct,i is the product-specific share of the farming system at

98

the global scale, fregion,i is the region-specific share of the farming system for meat and fglobal,i 5



99 100

is the global share of the farming system for meat. The animal feeding operation is not distinguished beyond the three farming systems.

101

Specifically reported proportions were assumed for Swiss livestock production which

102

involves more extensive farming systems than the global average.20 The fraction of

103

concentrate feed (fconcentrate) for different animal products, farming systems and 10 world

104

regions is obtained from Mekonnen and Hoekstra.21 The remaining feed is roughage and is

105

assumed to consist of either mixed grass, alfalfa, clover or grass (not else specified; nes) from

106

the country where the animal product was produced. Mekonnen and Hoekstra21 also provide

107

feed conversion efficiencies (FCE) for the same categories which enables one to translate the

108

feed mass into weight of carcass, milk and eggs. The concentrate feed is composed of nine

109

crops (neglecting fish meal and peas) which differ in their fractions (ffeed) according to three

110

animal categories (dairy and beef cattle, pigs and poultry) and six world regions22 although in

111

reality the composition might differ from country to country or even farm to farm. The mass

112

of feed (mfeed) for one of the concentrate feed crops is calculated as:

113

m = m ∙ ∑ f, ∙ FCE ∙ f , ∙ f (2)

114

and for roughage as:

115

m = m ∙ ∑ f, ∙ FCE ∙ #1 − f , & (3)

116

where mfeed is the mass of one of the feed crops (e.g. maize), mrp is the mass of the root

117

product (e.g. milk) calculated using equation 1.

118

Aquatic animal products such as fish are assumed to only cause impacts on freshwater

119

resources when they originate from aquaculture in contrast to wild fisheries. The respective

120

global fraction is provided by the FAO23 for different products. Fishes are assumed to be fed

121

by the same concentrate feed as poultry, but with 65% of the feed conversion efficiency.24 6


Page 6 of 37

Page 7 of 37


122

m = m ∙ f' ∙ FCE ∙ f (4)

123

Trade analysis. Exports were proportionally subtracted from Swiss production and

124

imports. With regards to imported food, the exporting country is not always the producer

125

(trade origin A, Figure 1). For instance, Switzerland imports cocoa from Germany, but

126

Germany does not cultivate cocoa. Therefore the concerned products are traced back to the

127

trade origin (B, Figure 1) using EXIOBASE14 for multi-regional input-output analysis. In

128

order to conform to the EXIOBASE data, 343 food items from 194 countries are aggregated

129

to 14 food cultivation and 11 food processing industries (Supporting information S1) and 47

130

regions (excluding Switzerland). The structural matrix (A) of the economy as available in

131

EXIOBASE was inverted25 and multiplied with the Swiss food imports (FI) to result in the

132

crop production (CP) required to satisfy the demand for imports:

133

CP = (I – A)-1 ·FI (5)

134

where A is a square matrix with 1175 rows and columns representing industry-region

135

combinations, I is an identity matrix with the same dimensions as A and FI is a vector of

136

length 1175. The demand was assigned to the food processing industries, while the crop

137

production was taken from the food cultivation industries. The concentrate feed of animal

138

products was traced back in the same way, but for each country separately, whereas roughage

139

was always assumed to be from the country where the product is produced. Information on

140

Swiss feed production is not contained in our databases and was estimated as the difference

141

between feed demand and feed imports.

142

The food industries and regions were disaggregated to food items and countries. For food

143

items the same proportions were assumed as for the corresponding industries; however, some

144

of the industries are very broad such as “vegetables, fruit and nuts”. For countries the export 7



145

shares26 of producing countries of the food items within a region were considered (trade

146

origin C, Figure 1) and where these data were missing production shares26 were used instead

147

(trade origin D, Figure 1). For ~19% of the crops where some crop-country combinations are

148

invalid, production data had to be used.

149

Life cycle assessment

150

Freshwater consumption. Crops consume blue water (irrigation water from surface or

151

groundwater resources) and green water (precipitation or soil moisture). Since green water is

152

only available to plants occupying the land and cannot be transferred to other locations or

153

used for non-agricultural purposes, it can be considered a land rather than water use impact.27

154

Therefore we only consider impacts of blue water consumption in this study. Blue water

155

consumption is given for 160 crops from 160 countries.28,29 We distinguish expected water

156

consumption, which represents the geometric mean of full and deficit water consumption

157

(variant V1), from deficit water consumption (V2), which represents a minimum estimate.

158

Especially for trees in tropical regions such as cocoa, the minimum estimate might be more

159

representative.29 If not otherwise specified, expected values (V1) were taken.

160

For roughage, we assumed either deficit irrigation (V1) or no irrigation (V2), assuming that

161

the grass is often from non-irrigated pastures and expected irrigation would therefore

162

overestimate blue water consumption. If neither of the roughage alternatives was produced as

163

a crop in the corresponding country, blue water consumption of roughage was also set to zero.

164

Water consumption is translated to impacts on biodiversity by characterisation factors

165

relating it to precipitation, net primary production and its fraction that is limited by water

166

availability. It is expressed as potentially disappeared fraction of species (unit: PDF m2 a / 8


Page 8 of 37

Page 9 of 37


167

m3).12 Water consumption and the characterisation factors are both weighted by national crop

168

production. Global biodiversity loss additionally considers endemic species richness30 and

169

total land area. It is expressed in global PDF (unit: gPDF a / m3) and the data was provided by

170

Verones et al.31

171 172

Besides impacts on biodiversity, water scarcity footprints as the product of water consumption and water scarcity indices32 were also investigated (Supporting information S3).

173

Freshwater eutrophication. Eutrophication describes the phenomenon of increased faunal

174

mortality as a consequence of an over-enrichment of nutrients in the aquatic environment and

175

the subsequent oxygen depletion. Freshwater eutrophication is mainly driven by excessive

176

emissions of phosphorus. Crop- and country-specific phosphorus emissions are obtained from

177

Scherer and Pfister.13 The emissions arise from mobilisation of soil phosphorus by erosion

178

and of phosphorus contained in fertilizers by surface runoff, drainage and groundwater

179

leaching.

180

The emissions are translated to biodiversity impacts by its residence time in freshwater

181

bodies (fate factor)33 and a typical effect factor of 200 PDF m3 a / kg P.33,34 Methods to assess

182

global biodiversity loss from phosphorus emissions are not yet available. The unit of local

183

biodiversity impacts from phosphorus were converted to match that for water consumption

184

and enables aggregating the impacts from both categories. We used a conversion factor of

185

6.25 PDF m3 a / PDF m2 a, which was derived from the average depth of freshwater

186

ecosystems and the ratio of global freshwater volume to global land area.35

187

Missing inventories. Some products have been disregarded because i) they could not be

188

assigned to any crop for which inventories are given and the inventories are assumed to be

189

negligible (popcorn, mushrooms, honey, yeast, vinegar, salt) or ii) their major ingredient was 9



190

unknown (infant food, food preparations). These products amount to ~1% of the consumed

191

food by mass.

192

For some exporting countries listed in the Swiss customs data, water consumption data is

193

not provided, as they are not cultivating the crops. It is assumed to equal nearby or associated

194

countries. For example, Solomon Islands are assigned to Papua New Guinea and Taiwan to

195

China.

196

Inventories are also not provided for all crops that are cultivated in Switzerland. For those

197

products, the average of the inventories in Germany, Austria, France and Italy was assumed

198

and water consumption was scaled with the aridity index36 to take into account that a more

199

humid climate generally requires less irrigation. Some products could still not be quantified.

200

Therefore, the water consumption of root (nes) was assumed to equal that of potatoes and for

201

eggplants, garlic and raspberries values of other nearby countries were taken.

202

Sensitivity analysis

203

We investigated the effect of not tracing back crop origins, but instead always assuming

204

that the crop is grown where the product is imported from (trade origin E, Figure 1). Since

205

inventories are not given for countries which do not produce a specific crop, the values of the

206

nearest country were assumed. If the nearest producing country was more than 1000 km away

207

from the exporting country, the average of the three nearest countries was assumed. Water

208

consumption was again scaled with the aridity index.36 For the biodiversity characterization

209

factors, country averages were taken. As an alternative sensitivity analysis, the major

210

producing exporter (trade origin F, Figure 1), or if unavailable the major producer, was

10


Page 10 of 37

Page 11 of 37


211

assumed as trade origin (G, Figure 1). As mentioned above, for ~19% of the crops where

212

some crop-country combinations are invalid production data had to be used.

213 214

RESULTS

215

Trade analysis

216

Switzerland produced 4.8 Mt of food, imported 5.0 Mt and exported 2.8 Mt in 2012. The

217

country is a net food importer, although about half of the food consumed is produced

218

domestically (49%). Almost half of the mass of imported food comes from its four

219

neighbouring countries Italy, Germany, France and Austria, but Brazil also exports many

220

products to Switzerland (Supporting information S2). Switzerland imports a variety of

221

products, but above all beverages and wheat (Supporting information S2).

222

Over 40% of the product-country combinations of imported food were traced back because

223

the exporting country does not produce the food. For instance, Switzerland imports cocoa

224

from Germany which, in turn, imports “crops nec” (not else classified, equivalent to not else

225

specified), to which cocoa but also coffee and pulses (excluding soya beans) belong, mainly

226

from Brazil, Rest of the World (RoW) America and RoW Africa based on EXIOBASE,

227

which, based on FAO data, results in Brazil, Ecuador and Ivory Coast as being the biggest

228

cocoa exporters to Germany. The implications of the biodiversity impacts due to different

229

trade origins are discussed in the following sections.

230

Freshwater consumption and eutrophication

231

The production of food consumed in Switzerland required almost 1400 million m3 of blue

232

water (

Global Biodiversity Loss by Freshwater ... - ACS Publications

Recommend Documents