Quantifying biodiversity footprints of Dutch economic sectors: A global supply-chain analysis

Journal of Cleaner Production 156 (2017) 194e202

Contents lists available at ScienceDirect

Journal of Cleaner Production journal homepage: www.elsevier.com/locate/jclepro

Quantifying biodiversity footprints of Dutch economic sectors: A global supply-chain analysis Harry C. Wilting*, Mark M.P. van Oorschot PBL Netherlands Environmental Assessment Agency, The Hague, The Netherlands

a r t i c l e i n f o

a b s t r a c t

Article history: Received 9 November 2016 Received in revised form 11 April 2017 Accepted 11 April 2017 Available online 12 April 2017

Economic sectors contribute to biodiversity loss via environmental pressures, such as land use and greenhouse gas (GHG) emissions, directly and via their supply chains. This study is the first that systematically quantifies supply-chain-related environmental pressures and terrestrial biodiversity losses in relation to sectoral production by presenting so-called biodiversity footprints for 47 sectors in the Dutch economy. The supply chains of the food and chemical sectors were investigated in more detail, by applying a structural path analysis. Mean Species Abundance (MSA) was used as a biodiversity indicator, representing the degree of ecosystem naturalness. Our results revealed that (i) the largest supply-chainrelated biodiversity losses occur in land-intensive and energy-intensive sectors; (ii) sectors that produce primary resources, such as crops and livestock, showed the largest biodiversity footprint per EUR of output; (iii) for most sectors in the Dutch economy, more than 50% of the biodiversity losses related to their supply chains were being caused abroad; and (iv) more than 45% of the supply-chain-related losses caused by the food and chemical sectors occurred upstream of the direct suppliers. Our results imply that mitigation of GHG emissions as well as land-use-related options should be considered in sectoral strategies to protect global biodiversity. The results create a clear rational for not only improving sectoral production efficiency, but also for taking supply-chain responsibility. Supply-chain-related biodiversity losses often cannot be directly influenced by the sector, or occur in other countries. Additional strategies may be needed then to reduce global biodiversity losses. © 2017 Elsevier Ltd. All rights reserved.

Keywords: Biodiversity loss Sector footprints Netherlands Supply chains Input-output analysis

1. Introduction Current demand for food, wood, energy and water has negative consequences for global biodiversity, in general (Butchart et al., 2010), and animal species, in particular (Dirzo et al., 2014). Without further policy action biodiversity loss will continue under continuation of current trends in population and income growth (Tittensor et al., 2014). Newbold et al. (2015) predict further declines in global terrestrial biodiversity, in a business-as-usual landuse scenario, with losses concentrated in biodiverse but economically poor countries. Policy strategies averting biodiversity loss commonly focus on primary sectors, such as agriculture, mining and forestry, which directly drive biodiversity loss through habitat conversion (CBD, 2016). However, actors more downstream in the supply chains, such as companies in secondary and tertiary sectors, and consumers, can also play a role in mitigating biodiversity losses, by influencing primary sectors. By including biodiversity concerns in their decision-making processes and by taking * Corresponding author. E-mail address: [email protected] (H.C. Wilting). http://dx.doi.org/10.1016/j.jclepro.2017.04.066 0959-6526/© 2017 Elsevier Ltd. All rights reserved.

responsibility for supply-chain impacts, they could move primary producers in a more biodiversity-friendly direction (Kok et al., 2014). Several studies on the supply-chain-related impacts of consumption on biodiversity have been published in recent years. Lenzen et al. (2012) calculate consumption-based biodiversity losses, measured as the number of Red List species threatened, for 187 countries. Moran et al. (2016) apply a similar approach in investigating the biodiversity losses related to the supply chains of certain materials and products. Chaudhary et al. (2016) calculate the impacts of Swiss consumption on global biodiversity by including the relationship between impacts and land-related pressures. However, studies on the supply-chain-related consequences of economic sectors did not consider impacts on biodiversity yet. For instance, Foran et al. (2005) calculate the social, economic and environmental consequences of the supply chains of 135 sectors in the Australian economy, and Acquaye et al. (2017) analyse the environmental pressures in the supply chains of two heavy polluting industries, namely electricity and chemical industries, in several large economies. Where these studies focus on supply-chain

H.C. Wilting, M.M.P. van Oorschot / Journal of Cleaner Production 156 (2017) 194e202

195

pressures of sectors, such as greenhouse gas emissions and land use, this paper goes a step further by presenting the combined impact these pressures have on global biodiversity loss, which can be interpreted as a biodiversity footprint by sector. A growing interest in the impacts of Dutch economic activities on global biodiversity loss in combination with the lack of information on these impacts caused by sectors, led to the main research questions for this paper. How large are the supply-chain-related impacts of individual economic sectors on biodiversity? Where in the world and in which sectors do these impacts take place? And what environmental pressures are the most important? We answered these questions in two steps: (i) we created an overview of the biodiversity footprints of 47 Dutch economic sectors, including primary, secondary and tertiary sectors, covering all stages of production in the Dutch economy; (ii) we investigated the biodiversity footprints of two sectors that showed large footprints (the food, beverages and tobacco sector and the chemicals and chemical products sector), in more detail. The answers to the research questions revealed which sectors were responsible for most of the biodiversity loss, and supports Dutch policymakers and sector organisations on what mitigating measures should be taken, and what national and international policy tools could be used to stimulate such measures. Furthermore the answers support companies that feel a growing need to investigate and report on their supply chains impacts (see also section 1.1). We used an environmentally extended multiregional inputoutput (EEMRIO) model to calculate the biodiversity footprints of Dutch economic sectors. EEMRIO models trace the flows of goods and services between sectors all over the world, and are appropriate for analysing the supply-chain-related environmental pressures of consumption activities (Wiedmann, 2009). Acquaye et al. (2017) applies an EEMRIO model to analyse the environmental pressures in the supply chains of two heavy polluting industries, but, to our knowledge, this study is the first to focus on the supplychain impacts on biodiversity of all sectors in a country. We included multiple environmental pressures, such as land use, infrastructure and greenhouse gas (GHG) emissions, in our model, providing a more explicit relationship between local production processes and impacts on biodiversity, so that our model results would be more policy-relevant, as is argued by Spangenberg (2007). Hertwich (2012) also suggests to explicitly model the cause-effect relationships between environmental pressures and biodiversity loss in calculating biodiversity footprints. We used Mean Species Abundance (MSA) as an indicator for biodiversity, since our focus was on connecting local biodiversity losses to sectoral activities. MSA is one of the various indices that were found relevant for measuring human impacts on biodiversity 
r et al., 2012). The MSA indicator expresses the mean (Va
cka abundance of original (i.e. naturally occurring) species in a disturbed situation, relative to their abundance in undisturbed ecosystems, as a reflection of the degree to which an ecosystem is intact (Alkemade et al., 2009). In this way, the MSA indicator is providing an aggregate and sensitive measure of local biodiversity loss, based on multiple pressure factors. It is not a measure of extinction, as that only occurs on larger scales of space and time, nor can it easily be attributed to short-term activities of individual companies. It is more effective to base mitigation measures on early warning signals, to prevent the extinction of species, in the long term.

priorities and selecting relevant mitigation measures to reduce biodiversity losses, both within sectors and in their supply-chains. Especially processing and retail sectors have a key role in influencing both primary producers upstream and consumers downstream, as they have a central and strong position in supply chains. In the market transformation strategy of the World Wild Fund for Nature, these actors are the primary target for engagement, and this may leverage the effect of their interventions on priorities for conservation (WWF Market Transformation Initiative, 2012). Mainstreaming biodiversity concerns in sectors and reducing direct pressures are also explicit targets of the Convention on Biological Diversity (CBD, 2014). The Dutch Government can use this information to prioritise and target specific sectors for engaging in, for example, sector covenants on mitigation. Such a strategy can contribute to national or regional biodiversity targets for reducing the impacts of production and consumption, both within and outside the country. Furthermore, sector footprints supply information on the average sector impacts that can be used as a general benchmark for individual company calculations. Wiedmann et al. (2009) apply such an approach by comparing the ecological footprint of a specific company in the recreational services sector to the average ecological footprint of that sector. When making corporate sustainability and impact reports, companies can apply data from sectoral studies to calculate their biodiversity footprint; for instance, by coupling the intensity of their impact on biodiversity (biodiversity loss per output in EUR) to their expenditures. Berners-Lee et al. (2011) implement such an approach for greenhouse gas (GHG) emissions in a tool to calculate the carbon footprint of small and mediumsized enterprises. Companies often calculate their supply-chain impacts with Life-Cycle Analysis (LCA) focusing on direct (on-site) pressures and pressures from direct suppliers, not including parts of the supply chain that are not directly visible (Lenzen, 2000). Data on sector footprints supply additional information for these calculations, providing the substantial part of the impact that is usually missing from LCAs of companies. Finally, insights into the contribution, in the form of environmental pressures, by regions and sectors to supply-chain-related biodiversity losses can be used for policy prioritisation, and help sectors with their procurement strategies, and with taking responsibility for the supply-chain impacts. Three well-known types of mitigation strategies to decrease supply-chain impacts are (van Oorschot et al., 2012): (i) reduction in direct pressures from local production processes; (ii) reduction in resource use by any sector, through improved resource efficiency, and (iii) changes in the procurement patterns of the supply-chain sectors, or changes in the locations from which resources are obtained. The choice for one of these strategies depends on the types of pressures and the contributions of regions and sectors to the sector footprint. Sectors with large direct impact on biodiversity losses should focus on their own activities. The two other strategies help to reduce the losses further upstream in supply chains. The third strategy may also influence direct suppliers and suppliers further upstream to improve their processes.

1.1. Use of sector footprints

We based our analysis on an EEMRIO model that had been used, previously, in research on consumption-based biodiversity footprints (Wilting et al., 2017). We modified the model by excluding final consumers from the model equation, in order to focus on sector footprints. The model relates production of a specific sector via monetary flows of goods and services to production in sectors, all

Biodiversity footprint studies help to raise awareness among different stakeholders and may help to define mitigation strategies. Insights in biodiversity footprints of sectors, as calculated in this study, can support sector organisations with information for setting

2. Methodology and data 2.1. General approach

196

H.C. Wilting, M.M.P. van Oorschot / Journal of Cleaner Production 156 (2017) 194e202

over the world. Environmental intensities per sector are used to couple this production to environmental pressures, which are used to calculate the impact on biodiversity. The environmental pressures considered are the pressures with the largest aggregated MSA losses, as distinguished in the global biodiversity model GLOBIO (Alkemade et al., 2009), including land use, infrastructure and GHG emissions. The GLOBIO is one of the models in our global climate change and biodiversity loss modelling framework (Stehfest et al., 2014). The impacts on biodiversity were attributed to habitat conversion into cropland, pasture and forestry areas, fragmentation as a result of existing cropland and infrastructure, disturbance caused by infrastructure use, and climate change. In calculating biodiversity footprints, we were interested in the impacts of all sector pressures in a specific year. Current biodiversity losses related to land use and infrastructure were combined with estimated future losses resulting from current GHG emissions in one composed biodiversity footprint indicator. Future biodiversity losses related to GHG emissions were based on the time-integrated impacts of climate change (see 2.3). All biodiversity losses were expressed in MSA-loss$km2$y, depicting the time integration aspect of the biodiversity losses due to GHG emissions. For land-related pressures (land use and infrastructure), the time period was one year, meaning that all land use in a specific year was considered responsible for the total MSA loss due to land conversion. In our approach, we allocated supply-chain biodiversity losses due to climate change to the countries and regions from which the GHG emissions originated. This is a logical approach when looking for strategies to reduce the impact on biodiversity by sectors. Insight into the sectors with high GHG emission levels is relevant for designing reduction options. However, those GHG emissions contribute to global climate change, which implies that pressures and their impacts do not occur at the same location. Biodiversity losses due to climate change vary per biome and especially occur in regions with high biodiversity values, such as the tropics (Alkemade et al., 2011; Arets et al., 2014). We explored the supply-chain impact structure of the two sectors, in more detail, by applying a structural path analysis (SPA) d a technique introduced in the 1980s (Defourny and Thorbecke, 1984). Using SPA, we decomposed the sector footprints into an infinite number of paths, characterised by the sectors involved in the paths, their length (indicating the number of sectors in the path) and the contribution of sectors to the total footprint (Lenzen, 2003). The SPA further unravelled and separated the impacts on biodiversity that can be attributed to direct suppliers and the suppliers further upstream in the supply chain. All paths with the same path length belong to the same production layer in the supply chain. The first production layer consists of the production in the sector itself (path length one), the second production layer represents all direct suppliers of the sector under consideration (path lengths two), the third production layer represents all direct suppliers of the direct suppliers of the sector under consideration, and so on (Lenzen, 2003). The contributions of the production layers give direction to the strategies to use in reducing the supply-chainrelated impacts of a sector as discussed in section 1.1.

the number of environmental pressures): Operation + is the element-wise multiplication of two matrices, named the Hadamard product (Styan, 1973). 2.3. Data parametrisation The model we used, as a direct follow up to our previous research on biodiversity footprints of nations, was parametrised with data on 2007 from several sources, as described by Wilting et al. (2017). The economic data were based on the WIOD database (Timmer et al., 2015), extended with data from GTAP version 8 (Narayanan et al., 2012), for disaggregating the agricultural sector and Rest-of-the-World region. We distinguished 47 sectors and 45 countries and regions in our analysis (see Table 1 for the sector classification and Table A1 for the classification of regions). The 47 sectors cover all production activities in the Dutch economy, from the primary production of agricultural products and the extraction of raw materials, to business and public services. Compared to the original data that we used, we omitted the sector paddy rice, since this sector has no economic significance for the Netherlands, in this respect. The 45 countries and regions cover the entire world. We included various environmental pressures, including GHG emissions and land use, in our biodiversity footprint calculations. The greenhouse gases included CO2, CH4 and N2O emissions from energy sources and agricultural and industrial processes. Emissions from land use, land-use change, and forestry (LULUCF) were not included. The types of land use included in our study were cropland, pasture, forested areas and infrastructure. Urban areas, including residential, factory and office locations, were not included in the sector footprints; a lack of data meant that production sectors could not be adequately related to those areas. We used factors on MSA loss in order to relate biodiversity losses to environmental pressures (matrix M in Equation (1)). The factors on MSA loss from cropland, pasture and forests depict the impact of habitat conversion on biodiversity. When deriving these factors per region, we distinguished three types of agricultural management for crop production and four types of forest management for wood production. We also included fragmentation of natural habitats as an indirect impact of the use of cropland. Smaller sized remaining natural patches of land lead to smaller remaining biodiversity values of natural areas (Alkemade et al., 2009). Biodiversity losses  pez due to infrastructure were related to disturbance (Benítez-Lo et al., 2010) and fragmentation (Verboom et al., 2014). We used one average worldwide factor for future MSA losses due to GHG emissions that was based on the time-integrated global temperature potential of greenhouse gases (Joos et al., 2013) and the relationship between a global mean temperature increase and MSA losses per biome (Arets et al., 2014). We considered a time horizon of 100 years in the calculations, consistent with the Intergovernmental Panel on Climate Change (IPCC, 2013). All sources, adjustments and derivations of pressure and impact data, and allocations to sectors are described, in detail, in Wilting et al. (2017). 3. Results

2.2. Environmentally extended MRIO model

3.1. Overall results

The starting point for our calculation of the biodiversity footprints of Dutch economic sectors was a modified version of the EEMRIO model, described by Wilting et al. (2017). Our model for calculating the biodiversity footprint of sector j, Bj, was:

Table 1 shows the calculated values for biodiversity footprint, measured as MSA losses, for 47 economic sectors in the Netherlands. The food, beverages and tobacco sector showed the largest impacts on biodiversity. The biodiversity loss caused was more than twice that by the sector with the second-largest impact, the electricity, natural gas and water supply sector. Other large contributors to biodiversity loss were found to be the chemicals and chemical products sector and the construction sector. That the food,

Bj ¼ i (M + D) (I e A)1 xj

(1)

With (for an r-region economy with s sectors per region and t

H.C. Wilting, M.M.P. van Oorschot / Journal of Cleaner Production 156 (2017) 194e202

i

2

m1;1 M¼4 « mt;1 2 1;1 d D¼4 « dt;1 I 2 11 A A¼4 « Ar1 xj

/ 1 / / 1 / / 1 /

3 m1;r « 5 mt;r 3 d1;r « 5 dt;r 1r

3

A « 5 Arr

197

i is a (1  t) vector of ones used for summing up the biodiversity losses caused by individual environmental pressures; M is the (t  r·s) matrix of biodiversity-loss factors; mi,j is a (1  s) row vector of biodiversity loss factors of direct environmental pressure i in region j (depicting the biodiversity losses per unit of environmental pressure); D is the (t  r·s) matrix of direct environmental pressures; di,j is a (1  s) row vector of direct environmental pressure intensities of pressure i in region j (depicting the direct environmental pressures of one unit of production, for all sectors); Matrix I is the identity matrix; (I e A)¡1 is the Leontief inverse matrix. A is the (r·s x r·s) matrix of input coefficients; Aij, i¼j is the (s x s) matrix of domestic input coefficients of region i, Aij, i≠j is the (s x s) matrix of import coefficients of region j importing from region i. The domestic and import coefficients depict the intermediate input requirements per unit of production (output), for each sector. xj is the (r·s x 1) adjusted vector of production with zeros for all sectors except for sector j. Note that this is different from consumption-based footprints that have a focus on final demand (often characterised by y).

beverages and tobacco sector topped the list is not surprising, as it is a relatively large sector, in terms of production value, with intensive land use in the primary supply sectors. Not all large sectors, in terms of production value, showed a large biodiversity footprint. For instance, Table 1 shows several sectors with production values similar to that of the food, beverages and tobacco sector, but with significantly lower biodiversity footprints. These sectors include the public sector and financial sector, which both mainly provide services. Forestry is a relative small sector in the Netherlands and, therefore, total biodiversity loss caused by this sector was relatively small, as well. Since the size of the footprint of a sector only partly depends on its economic size, we also calculated the losses per unit of production value (EUR) for all 47 sectors (Table 1). The Dutch primary land-related crop and livestock sectors, such as wheat, plant-based fibres and wool, showed the largest supply-chain-related biodiversity losses per EUR output. This is mostly due to the large effects that crop and livestock production have on biodiversity, as natural habitats are completely converted for agricultural production. Most secondary and tertiary sectors showed a relation to biodiversity losses per EUR production value below 1.0 MSA-loss$m2$y. Exceptions were the sectors of food, beverages and tobacco, electricity, gas and water supply, and air transport. For 35 of the 47 sectors (representing 89% of total Dutch production value), more than 50% of the biodiversity footprint was caused by pressures outside the Netherlands, for production, extraction and processing of raw materials (Table 1). For the sectors of leather and footwear and wood and products of wood and cork, more than 90% of the caused biodiversity loss occurred both indirectly and abroad. These sectors are characterised by large imports of raw and processed materials. Sectors with a large direct impact (shares of direct losses higher than 50%) were the utility sector (electricity, natural gas and water), the transport sectors and most of the primary crop sectors. The sector bovines, sheep and goats, horses showed a relatively large impact abroad, caused by its import of feed. Land use was the primary driver of biodiversity loss for most agricultural sectors (column 9 in Table 1). The Dutch wool sector showed a substantial contribution in GHG emissions, in the form of non-CO2 GHG emissions from sheep. The food, beverages and tobacco sector obtains its resources from the land-using agricultural sectors; more than three quarters of the biodiversity loss induced by this sector was due to indirect land use in the supply chain. Supply-chain GHG emissions contributed for slightly more than 20% to the biodiversity loss caused by this sector. For non-foodrelated sectors with high biodiversity losses per production value (in EUR), more than three quarters of the supply-chain-related biodiversity losses were caused by GHG emissions. These belong to the energy sector (electricity, natural gas and water supply and coke, refined petroleum and nuclear fuel) and the transport sector (air transport and water transport) with large amounts of fossil fuel

being used by the sector itself. 3.2. Detailed results: food and chemical sector To provide a more detailed analysis of sector footprints, we selected two sectors with relatively large supply-chain-related biodiversity losses, namely those of food, beverages and tobacco (‘food sector’) and chemicals and chemical products (‘chemical sector’). The food sector was chosen because it was found to have the largest impact via its supply chains, and the chemical sector because it has an especially high impact through its contribution to climate change. 3.2.1. Sector contributions The food sector showed the largest impacts on biodiversity measured as MSA losses (see 3.1). More than 98% of the losses were caused by indirect pressures from the supply chain, and 79% of these losses were related to land use. A relatively small share of the losses (<2%) was caused by the direct activities in the food sector that were mostly related to GHG emissions from operational energy use (lowest bar in Fig. 1a). Both domestic and foreign sectors contributed in large or small extent to the indirect biodiversity losses caused by the food sector. Fig. 1 shows these contributions by adding domestic and foreign contributions per sector. Main contributors were land use in arable sectors producing food and feed biomass, such as oil seeds, other crops and other cereals, and livestock sectors, such as bovines, sheep and goats, horses and raw milk. Main contributors via the pressure of GHG emissions were livestock farming (especially non-CO2 GHG emissions) and the utility sector (CO2 emissions from electricity generation), which is included in Other sectors (the aggregated contribution of 39 sectors with relatively small contributions). The production value in the chemical sector was 5% less than the production value in the food sector, but the total biodiversity losses were about one-third of those related to the food sector (Table 1). Biodiversity losses were mainly caused by GHG emissions (77%) from the sector itself or upstream in the energy and raw material sectors (Fig. 1b). Contributions from land use were less than onequarter of the total biodiversity footprint of the chemical sector. 3.2.2. Regional contributions The EEMRIO framework also provides information on the regions where the supply-chain-related environmental pressures from the Dutch sectors occur (Fig. 2a). These patterns were quite different for the Dutch food and chemical sectors. For the Dutch food sector, almost 36% of the biodiversity losses were caused by pressures occurring in European countries (10% in the Netherlands and 26% in other European countries). Outside Europe, land use in South America and Africa contributed the most to the supplychain-related biodiversity losses caused by the food sector. These regions include the main production locations of imported

198

H.C. Wilting, M.M.P. van Oorschot / Journal of Cleaner Production 156 (2017) 194e202

Table 1 Production value, total biodiversity loss, and biodiversity loss per EUR production value, for 47 sectors in the Netherlands, in 2007; biodiversity loss is distributed over direct and indirect causes (directly by the sector and indirectly in the supply chain, both domestic and via imports) and over environmental pressures. Land-related causes include land use and infrastructure. Primary sectors: 1e15; secondary sectors: 15e30; and tertiary sectors: 31e47. No.

Sector

Production (million V)

Total MSA loss (MSA loss$km2$y)

MSA.loss per V (MSA loss$m2$y/V)

Directly by sector itself (%)

Indirectly via domestic suppliers (%)

Indirectly via imports (%)

Landrelated (%)

GHG emissions (%)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Wheat Other cereals Vegetables, fruits, nuts Oil seeds Sugar cane, sugar beet Plant-based fibres Other crops Bovines, sheep and goats, horses Other animal products Raw milk Wool, silk-worm cocoons Forestry Fisheries Mining and quarrying Food, beverages and tobacco Textiles and textile products Leather and footwear Wood and products of wood and cork Pulp, paper, printing and publishing Coke, refined petroleum and nuclear fuel Chemicals and chemical products Rubber and plastics Other non-metallic minerals Basic metals and fabricated metal Other machinery Electrical and optical equipment Transport equipment Other manufacturing; recycling Electricity, natural gas and water supply Construction Sale, maintenance and repair of motor vehicles and motorcycles; retail sale of fuel Wholesale trade and commission trade, except of motor vehicles and motorcycles Retail trade, except of motor vehicles and motorcycles; repair of household goods Hotels and restaurants Inland transport Water transport Air transport Other supporting and auxiliary transport activities; activities of travel agencies Post and telecommunications Financial intermediation Real estate activities Renting of machinery and equipment, and other business activities Public administration and defence; compulsory social security Education Health and social work Other community, social and personal services Private households with employed persons

220 399 4520 439 249 8 8708 2140 4206 4892 34 322 943 20,744 55,739 3699 373 3189

1241 1558 9703 1315 763 32 15,213 6028 10,000 13,060 127 816 487 5755 103,142 1615 201 2825

5.6 3.9 2.1 3.0 3.1 4.2 1.7 2.8 2.4 2.7 3.8 2.5 0.5 0.3 1.9 0.4 0.5 0.9

89 44 38 15 88 91 64 43 15 47 91 81 32 26 2 6 0 3

1 1 5 1 1 1 6 6 13 4 2 3 10 20 8 9 7 5

11 55 57 84 11 8 29 52 72 50 7 16 58 54 90 84 92 92

90 83 68 80 85 86 60 58 69 65 5 84 20 16 79 39 62 77

10 17 32 20 15 14 40 42 31 35 95 16 80 84 21 61 38 23

19,928

6620

0.3

10

13

77

42

58

30,292

29,467

1.0

19

3

78

19

81

52,595 7387 6956 28,481 22,353 19,985 17,204 9905 34,875

33,896 3017 3497 15,060 6618 6725 5499 3254 46,141

0.6 0.4 0.5 0.5 0.3 0.3 0.3 0.3 1.3

28 4 30 23 2 3 1 6 50

7 11 12 10 13 13 11 10 17

65 85 57 68 85 85 87 84 33

23 35 18 18 28 33 29 47 10

77 65 82 82 72 67 71 53 90

78,210 16,049

23,336 3259

0.3 0.2

5 12

13 18

82 70

38 28

62 72

72,015

14,210

0.2

5

18

77

31

69

29,694

4316

0.1

9

25

66

33

67

18,507 20,397 6173 9145 16,966

12,415 8051 5695 10,222 5126

0.7 0.4 0.9 1.1 0.3

5 50 56 59 3

11 10 10 6 24

84 40 34 35 72

67 24 10 12 29

33 76 90 88 71

25,310 61,648 59,493 121,578

3278 6554 5950 17,035

0.1 0.1 0.1 0.1

4 6 2 11

18 19 15 18

78 75 83 71

36 37 46 35

64 63 54 65

61,638

12,029

0.2

8

23

69

39

61

29,737 59,794 35,044

2845 9640 20,859

0.1 0.2 0.6

15 8 38

24 16 14

61 76 48

30 46 32

70 54 68

2089

0

0.0

0

0

100

68

32

19 20 21 22 23 24 25 26 27 28 29 30 31

32

33

34 35 36 37 38

39 40 41 42

43

44 45 46 47

resources through the supply chain of the food sector. For the chemical sector, 65% of the biodiversity footprint was caused in Europe, mainly by GHG emissions from European sectors (Fig. 2b).

Biodiversity losses outside Europe were mainly caused by GHG emissions in Asia (China) due to energy production and manufacturing facilities in that region.

H.C. Wilting, M.M.P. van Oorschot / Journal of Cleaner Production 156 (2017) 194e202

199

b) Chemicals and chemical products

a) Food, beverages and tobacco Other sectors

Other sectors

Oil seeds

Wheat

Other crops

Vegetables, fruit, nuts

Other Community, Social and Personal Services

Raw milk

Bovines, sheep and goats, horses

Bovines, sheep and goats, horses

Coke, Refined Petroleum and Nuclear Fuel Forestry

Other cereals

Inland Transport

Other crops

Mining and Quarrying

Oil seeds

Electricity, Gas and Water Supply

Food, Beverages and Tobacco

Chemicals and Chemical Products 0

5000

10000

15000

20000

25000

0

5000

MSA-loss·km2 ·yr GHG

10000

15000

MSA-loss·km2 ·yr

Land use

GHG

Land use

Fig. 1. a-b. Supply-chain biodiversity losses measured as MSA losses related to the Dutch food sector (a) and chemical sector (b), caused by direct land use and direct GHG emissions from the sector itself (lowest bar) and caused by pressures in upstream sectors (indirect losses). The contributions from upstream domestic an foreign sectors were added per sector.

3.2.3. Production-layer contributions By using an SPA, we identified the contributions of the production layers in the supply-chain-related biodiversity losses of the two sectors. We already noticed that the direct pressures (production layer 1) from the Dutch food sector itself were very small, compared to the supply-chain-related losses caused by all suppliers within the chain (Fig. 3a). The direct suppliers in the food sector (at production layer 2) were found to be responsible for 53% of the biodiversity losses caused by this sector. Pressures further upstream in the chain caused the remaining 45%. This information is relevant when supply-chain responsibility issues are addressed. 28% of the biodiversity footprint of the chemical sector resulted from direct (production layer 1) pressures and 21% from production layer-2 pressures that mostly consisted of GHG emissions by direct suppliers (Fig. 3b). This implies that more than 50% of the biodiversity footprint of the chemical sector was caused upstream, at production layer 3 and higher, which means that supply-chain responsibility goes beyond the direct suppliers. Table 2 shows the ten paths with the largest contributions to the biodiversity footprint of the food sector. The direct pressures of the sector (with path length 1) were ranked 10th. There were nine paths with larger contributions, seven with path length 2 and two with path length 3. The paths with length 2 comprise the direct suppliers of various crops to the food sector, including oil seeds, cereals, vegetables, and raw milk. The largest paths with three sectors involved (path length 3) supply animal products to foreign food sectors (aggregated over regions), which in turn supply the

Dutch food sector with processed resources. The ten paths selected together contributed 55% of the total biodiversity footprint of the Dutch food sector. The food sector might focus on these paths first, when mitigating its supply-chain impacts. Table 3 shows the outcome of an SPA for the chemical sector by showing the top 10 paths with the largest contribution to the sector’s biodiversity footprint. The direct pressures of the chemical sector itself showed with almost 28% the largest contribution to the footprint (rank 1; see also Fig. 3b). The supply of chemicals and energy resources from the foreign mining and refineries sectors and domestic electricity sector, together, contributed more than 7% of the total impact of the Dutch chemical sector on biodiversity. Substantial contributions from sectors upstream the direct suppliers (path length 3 and higher) originated from the foreign chemical sector. 4. Discussion and conclusions 4.1. Modelling aspects We calculated biodiversity footprints measured in MSA losses of Dutch economic sectors by applying an EEMRIO model that included the consequences of specific environmental pressures for terrestrial biodiversity. The model included the environmental pressures that have the largest impacts on biodiversity, including land use, infrastructure and GHG emissions. Since not all pressures driving biodiversity loss were included, total loss per sector is

a) Food, beverages and tobacco

b) Chemicals and chemical products

Oceania

Oceania

Europe

Europe

Asia

Asia

South America

South America

North America

North America

Africa

Africa 0%

10%

20%

30% GHG

Land use

40%

50%

60%

70%

0%

10%

20%

30% GHG

40%

50%

60%

70%

Land use

Fig. 2. a-b. Distribution of sector biodiversity footprints due to environmental pressures (land use and GHG emissions) over contributing regions. Food sector (a) and chemical sector (b).

200

H.C. Wilting, M.M.P. van Oorschot / Journal of Cleaner Production 156 (2017) 194e202

a) Food, beverages and tobacco

b) Chemicals and chemical products

60%

60%

50%

50%

40%

40%

30%

30%

20%

20%

10%

10%

0%

0% 1

2

3 GHG

4

5

6 and above

1

2

3

Land use

4

GHG

5

6 and above

Land use

Fig. 3. a-b. Shares of biodiversity loss per production layer in the supply chains of the food sector (a) and chemical sector (b). Contributions for land use and GHG emissions. The first production layer are the impacts of the sector itself, the second production layer represents the impacts of the direct suppliers of the sector under consideration, and so on.

Table 2 Total biodiversity loss due to the Dutch food sector, broken down into supply chain contributions from different sectors. Contributions per sector are divided into domestic sectors (Dom) and foreign sectors (Imp). Absolute values in MSA-loss.km2.y. Rank

Path description

Path length

Path value

Share (%)

1 2 3 4 5 6 7 8 9 10

Oil seeds (Imp) - > Food sector Other crops (Imp) - > Food sector Other cereals (Imp) - > Food sector Raw milk (Dom) - > Food sector Vegetables, fruits, nuts (Imp) - > Food sector Bovines, sheep and goats, horses (Imp) - > Food, Beverages and Tobacco (Imp) - > Food sector Bovines, sheep and goats, horses (Imp) - > Food sector Raw milk (Imp) - > Food, Beverages and Tobacco (Imp) - > Food sector Wheat (Imp) - > Food sector Food sector All other paths Total

2 2 2 2 2 3 2 3 2 1

16.9 14.1 6.9 4.4 4.4 3.5 2.4 1.9 1.9 1.9 44.8 103.2

16.4 13.6 6.7 4.3 4.3 3.4 2.3 1.9 1.9 1.8 43.4 100.0

slightly underestimated (PBL, 2012). Some of the pressures that were not accounted for yet are mining impacts and acidification. To our knowledge, the impacts of these pressures on biodiversity are small on a global level, but might be substantial on local levels for specific sectors, such as the mining sector. Furthermore, Alkemade et al. (2009) show that the impact of nitrogen deposition on biodiversity is relatively small in most regions. We did not account for impacts on inland or marine aquatic biodiversity yet, due to methodological constraints in integrating terrestrial and aquatic indicators. The inclusion of aquatic biodiversity would have large consequences for specific sectors, such as the fisheries sector. We chose for using a model that was parametrised with data from the WIOD. The agricultural sector was already disaggregated to improve the allocation of biodiversity losses to the agricultural

subsectors. Compared to other MRIO databases, the number of regions in the WIOD is relatively low and the regional classification of WIOD is biased towards developed rich countries and large economies (Table A1). This implies that WIOD is less representative for developing countries with relatively large biodiversity hotspots. Employing MRIO databases with more spatial detail, such as Eora (Lenzen et al., 2013) and GTAP (Aguiar et al., 2016), would help to test the robustness of our conclusions. Further disaggregation at the sub-regional level in regions with large spatial variability in production and impacts also might improve the accounting of sector biodiversity footprints. Godar et al. (2015), for instance, showed the importance of fine-scale spatial resolution in data on soy production in Brazil for improving their land footprint calculations.

Table 3 Total biodiversity loss due to the Dutch chemical sector, broken down into supply chain contributions from different sectors. Contributions per sector are divided into domestic sectors (Dom) and foreign sectors (Imp). Absolute values in MSA-loss.km2.y. Rank

Path description

Path length

Path value

Share (%)

1 2 3 4 5 6 7 8 9 10

Chemical sector Chemicals and Chemical Products (Imp) - > Chemical sector Mining and Quarrying (Imp) - > Chemical sector Coke, Refined Petroleum and Nuclear Fuel (Imp) - > Chemical sector Electricity, Natural gas and Water Supply (Dom) - > Chemical sector Inland Transport (Imp) - > Chemical sector Electricity, Natural gas and Water Supply (Imp) - > Chemicals and Chemical Products (Imp) - > Chemical sector Chemicals and Chemical Products (Imp) - > Chemicals and Chemical Products (Imp) - > Chemical sector Mining and Quarrying (Imp) - > Coke, Refined Petroleum and Nuclear Fuel (Imp) - > Chemical sector Other Community, Social and Personal Services (Imp) - > Chemical sector All other paths Total

1 2 2 2 2 2 3 3 3 2

9.4 2.1 1.0 0.7 0.7 0.7 0.5 0.4 0.4 0.4 17.6 33.9

27.8 6.3 3.0 2.2 2.1 2.0 1.4 1.1 1.1 1.1 51.8 100.0

H.C. Wilting, M.M.P. van Oorschot / Journal of Cleaner Production 156 (2017) 194e202

4.2. Interpretation of the results The sectors that showed a large footprint are a mix of landintensive and energy-intensive sectors, weighted according to their impact on biodiversity in terms of MSA losses. High losses were found for sectors that depend on land-related resources, such as agriculture and the food sector, as well as for those with relatively high energy use and GHG emissions, such as electricity production and transport. This is not really surprising since, at the global level, food production, energy use and transport were identified as the main contributors to biodiversity loss (Kok et al., 2014). To our knowledge, there are no peer-reviewed studies available on the impacts on biodiversity of individual sectors including their supply-chain effects, to use for comparison. There are long lists of complementary indicators for measuring other aspects of the biodiversity concept (Curran et al., 2011). It would be interesting to see if other biodiversity measures lead to similar results and support our outcomes. The structural path analysis of the supply chains of the two sectors considered, the food and the chemical sectors, showed that 45%e50% of the impact on biodiversity was caused upstream of the direct suppliers of the sectors. These findings are in line with a study by Lenzen (2000) which shows that process-based life-cycle inventories do not include all upstream pressures or impacts; they suffer from truncation errors by up to 50%, compared with IO analysis. Since LCAs of the supply-chain impacts of companies mainly focus on direct (on-site) pressures and those from their most relevant suppliers, a substantial part of the impact is missing from these studies. Pelletier et al. (2014) suggest that results from IO analyses of sector footprints, which consider complete supply chains, can deliver additional insights into Organisation Environmental Footprinting (OEF) which is usually calculated using LCAs (see also section 1.1). Dutch companies can apply the MSA losses per EUR production value as calculated in this study in the analyses of their supply-chain-related impacts on terrestrial biodiversity. Sectors are active in all areas of the economy, from primary resource extraction to service and government sectors. Since supply chains of sectors partly overlap, an economy footprint based on the sum of the individual sector footprints would lead to double counting. When raw materials or intermediate goods produced in the Netherlands, subsequently, are used in sectors further downstream, for instance in the production of consumer goods, the related supply-chain losses will partly overlap. For instance, the biodiversity footprint of cattle farming, the food sector and hotels and restaurants all include biodiversity loss due to feed production in arable farming in the Netherlands and abroad. A sector-based footprint for the whole economy can be compiled by accounting for double counting, but that is beyond the scope of this paper. 4.3. Implications and recommendations In section 1.1 we mentioned three strategies for reducing supply-chain biodiversity losses, namely (i) reducing direct pressures; (ii) reducing the use of resources, and (iii) changing procurement patterns and locations. The choice of the mitigation strategy depends on the types of pressures and the contributions of regions and sectors in the supply chains to the footprint. Sectors with a high share of direct impacts on biodiversity should focus on their own activities and pressures. For instance, crop sectors with high impacts from land use should aim to produce their crops with lower impacts on biodiversity. By implementing energy-efficiency measures, energy-intensive sectors, such as electricity generation and several transport sectors, might reduce their GHG-related impacts on terrestrial biodiversity. For the chemical sector, with 28%

201

of its footprint resulting from direct pressures, a reduction of the direct pressures is one of the strategies that can be chosen. The two other strategies help to reduce the losses upstream in supply chains. Most secondary and tertiary sectors might use such a strategy by assessing their sourcing practices and influencing their direct suppliers and suppliers further upstream to improve their processes. Since the direct (domestic and foreign) suppliers are companies that are in direct contact with the sector, it is easier to influence them, for instance through supplier screening and selection. For the food and chemical sectors we investigated in more detail, however, more than 45% of the impacts were caused upstream of the direct suppliers. Influencing these suppliers requires transparency in the supply chains with insights in the sourcing practices of the direct suppliers. Further research is needed to obtain insights in the potential effects of the three strategies for reducing biodiversity losses in the supply chains of individual sectors. Relevant and effective policies that implement these strategies can then be identified. The Netherlands has no well-defined targets for reducing sectoral biodiversity losses. Where policies on mainstreaming biodiversity concerns in sectoral initiatives and involving sectors in conservation of biodiversity are still under construction, policies focusing on the underlying pressures on biodiversity already contribute to mitigation of biodiversity loss. For instance, policies on reducing GHG emissions, as a consequence of the Paris agreement (UNFCCC, 2016), that specify targets for individual sectors contribute to reducing the biodiversity losses from energy intensive sectors, such as refineries, electricity production and air transport. For cross-border linkages, adequate international policy is lacking. So, for reducing land-related impacts in supply-chains other approaches are necessary, for instance via certification by companies with specific voluntary sustainability standards initiatives that also take biodiversity loss into account (UNEP-WCMC, 2011). However, all policy strategies, both current and those still being developed, first need insight into the supply-chain contributions to global biodiversity loss as presented in this paper. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Acknowledgements We are grateful to Annemieke Righart at PBL Netherlands Environmental Assessment Agency for editing the manuscript. Furthermore, we thank our PBL colleagues Michel Bakkenes and Johan Meijer for providing data from the GLOBIO model. Appendix A. Region classification Table A1 Region classification in the EEMRIO model. Australia Austria Belgium Bulgaria Brazil Canada China Cyprus Czech Republic Germany Denmark Spain

Estonia Finland France United Kingdom Greece Hungary Indonesia India Ireland Italy Japan

Korea Lithuania Luxembourg Latvia Mexico Malta Netherlands Poland Portugal Romania Russia

Slovak Republic Slovenia Sweden Turkey Taiwan United States Rest of Oceania Rest of Asia Rest of Americas Rest of Europe Africa

202

H.C. Wilting, M.M.P. van Oorschot / Journal of Cleaner Production 156 (2017) 194e202

References Acquaye, A., Feng, K., Oppon, E., Salhi, S., Ibn-Mohammed, T., Genovese, A., Hubacek, K., 2017. Measuring the environmental sustainability performance of global supply chains: a multi-regional input-output analysis for carbon, sulphur oxide and water footprints. J. Environ. Manag. 187, 571e585. Aguiar, A., Narayanan, B., McDougall, R., 2016. An overview of the GTAP 9 data base. J. Glob. Econ. Anal. 1, 181e208. Alkemade, R., Bakkenes, M., Eickhout, B., 2011. Towards a general relationship between climate change and biodiversity: an example for plant species in Europe. Reg. Environ. Change 11, 143e150. Alkemade, R., van Oorschot, M., Miles, L., Nellemann, C., Bakkenes, M., ten Brink, B., 2009. GLOBIO3: a framework to investigate options for reducing global terrestrial biodiversity loss. Ecosystems 12, 374e390. Arets, E.J.M.M., Verwer, C., Alkemade, R., 2014. Meta-analysis of the Effect of Global Warming on Local Species Richness. WOt Paper 34. Statutory Research Task Unit for Nature and the Environment. Wageningen University, Wageningen, The Netherlands. pez, A., Alkemade, R., Verweij, P.A., 2010. The impacts of roads and other Benítez-Lo infrastructure on mammal and bird populations: a meta-analysis. Biol. Conserv. 143, 1307e1316. Berners-Lee, M., Howard, D.C., Moss, J., Kaivanto, K., Scott, W.A., 2011. Greenhouse gas footprinting for small businesses d the use of inputeoutput data. Sci. Total Environ. 409, 883e891. Butchart, S.H.M., Walpole, M., Collen, B., van Strien, A., Scharlemann, J.P.W., Almond, R.E.A., Baillie, J.E.M., Bomhard, B., Brown, C., Bruno, J., Carpenter, K.E., Carr, G.M., Chanson, J., Chenery, A.M., Csirke, J., Davidson, N.C., Dentener, F., Foster, M., Galli, A., Galloway, J.N., Genovesi, P., Gregory, R.D., Hockings, M., Kapos, V., Lamarque, J.-F., Leverington, F., Loh, J., McGeoch, M.A., McRae, L., Minasyan, A., Morcillo, M.H., Oldfield, T.E.E., Pauly, D., Quader, S., Revenga, C., Sauer, J.R., Skolnik, B., Spear, D., Stanwell-Smith, D., Stuart, S.N., Symes, A., , J.-C., Watson, R., 2010. Global biodiversity: inTierney, M., Tyrrell, T.D., Vie dicators of recent declines. Science 328, 1164e1168. CBD, 2014. Global Biodiversity Outlook 4. Secretariat of the Convention on Bioal, Canada. logical Diversity, Montre CBD, 2016. National Biodiversity Strategies and Action Plans (NBSAPs). Convention on Biological Diversity (Accessed 26 January 2017). www.cbd.int/nbsap/default. shtml. Chaudhary, A., Pfister, S., Hellweg, S., 2016. Spatially explicit analysis of biodiversity loss due to global agriculture, pasture and forest land use from a producer and consumer perspective. Environ. Sci. Technol. 50, 3928e3936. Curran, M., de Baan, L., de Schryver, A.M., van Zelm, R., Hellweg, S., Koellner, T., Sonnemann, G., Huijbregts, M.A.J., 2011. Toward meaningful end points of biodiversity in life cycle assessment. Environ. Sci. Technol. 45, 70e79. Defourny, J., Thorbecke, E., 1984. Structural path analysis and multiplier decomposition within a social accounting matrix framework. Econ. J. 94, 111e136. Dirzo, R., Young, H.S., Galetti, M., Ceballos, G., Isaac, N.J.B., Collen, B., 2014. Defaunation in the anthropocene. Science 345, 401e406. Foran, B., Lenzen, M., Dey, C., Bilek, M., 2005. Integrating sustainable chain management with triple bottom line accounting. Ecol. Econ. 52, 143e157. Godar, J., Persson, U.M., Tizado, E.J., Meyfroidt, P., 2015. Towards more accurate and policy relevant footprint analyses: tracing fine-scale socio-environmental impacts of production to consumption. Ecol. Econ. 112, 25e35. Hertwich, E., 2012. Biodiversity: remote responsibility. Nature 486, 36e37. IPCC, 2013. Climate Change 2013: the Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. Joos, F., Roth, R., Fuglestvedt, J.S., Peters, G.P., Enting, I.G., von Bloh, W., Brovkin, V., €licher, T.L., Halloran, P.R., Burke, E.J., Eby, M., Edwards, N.R., Friedrich, T., Fro Holden, P.B., Jones, C., Kleinen, T., Mackenzie, F.T., Matsumoto, K., Meinshausen, M., Plattner, G.K., Reisinger, A., Segschneider, J., Shaffer, G., Steinacher, M., Strassmann, K., Tanaka, K., Timmermann, A., Weaver, A.J., 2013. Carbon dioxide and climate impulse response functions for the computation of greenhouse gas metrics: a multi-model analysis. Atmos. Chem. Phys. 13, 2793e2825. Kok, M., Alkemade, R., Bakkenes, M., Boelee, E., Christensen, V., van Eerdt, M., van der Esch, S., Janse, J., Karlsson-Vinkhuyzen, S., Kram, T., Lazarova, T., Linderhof, V., Lucas, P., Mandryk, M., Meijer, J., van Oorschot, M., Teh, L., van Hoof, L., Westhoek, H., Zagt, R., 2014. How Sectors Can Contribute to Sustainable Use and Conservation of Biodiversity. CBD Technical Series No 79, PBL Report Number 01448. PBL Netherlands Environmental Assessment Agency, The Hague, The Netherlands. Lenzen, M., 2000. Errors in conventional and input-outputdbased lifedcycle inventories. J. Ind. Ecol. 4, 127e148. Lenzen, M., 2003. Environmentally important paths, linkages and key sectors in the Australian economy. Struct. Change Econ. Dyn. 14, 1e34.

Lenzen, M., Moran, D., Kanemoto, K., Foran, B., Lobefaro, L., Geschke, A., 2012. International trade drives biodiversity threats in developing nations. Nature 486, 109e112. Lenzen, M., Moran, D., Kanemoto, K., Geschke, A., 2013. Building Eora: a global multi-region inputeoutput database at high country and sector resolution. Econ. Syst. Res. 25, 20e49. Moran, D., Petersone, M., Verones, F., 2016. On the suitability of inputeoutput analysis for calculating product-specific biodiversity footprints. Ecol. Indic. 60, 192e201. Narayanan, G., Badri, A.A., McDougall, R., 2012. Global Trade, Assistance, and Production: the GTAP 8 Data Base. Center for Global Trade Analysis, Purdue University, West Lafayette, Indiana, USA. Newbold, T., Hudson, L.N., Hill, S.L.L., Contu, S., Lysenko, I., Senior, R.A., Borger, L., Bennett, D.J., Choimes, A., Collen, B., Day, J., De Palma, A., Diaz, S., EcheverriaLondono, S., Edgar, M.J., Feldman, A., Garon, M., Harrison, M.L.K., Alhusseini, T., Ingram, D.J., Itescu, Y., Kattge, J., Kemp, V., Kirkpatrick, L., Kleyer, M., Correia, D.L.P., Martin, C.D., Meiri, S., Novosolov, M., Pan, Y., Phillips, H.R.P., Purves, D.W., Robinson, A., Simpson, J., Tuck, S.L., Weiher, E., White, H.J., Ewers, R.M., Mace, G.M., Scharlemann, J.P.W., Purvis, A., 2015. Global effects of land use on local terrestrial biodiversity. Nature 520, 45e50. PBL, 2012. Roads from Rioþ20. Pathways to Achieve Global Sustainability Goals by 2050. PBL Netherlands Environmental Assessment Agency, The Hague, The Netherlands. Pelletier, N., Allacker, K., Pant, R., Manfredi, S., 2014. The European Commission Organisation Environmental Footprint method: comparison with other methods, and rationales for key requirements. Int. J. Life Cycle Assess. 19, 387e404. Spangenberg, J.H., 2007. Biodiversity pressure and the driving forces behind. Ecol. Econ. 61, 146e158. Stehfest, E.E., van Vuuren, D.P., Bouwman, A.F., Kram, T., 2014. IMAGE 3.0 Integrated Model to Assess the Global Environment, Version 3.90. Complete Model Overview and Recent Scenario Results. PBL Netherlands Environmental Assessment Agency, The Hague, The Netherlands. Styan, G.P.H., 1973. Hadamard products and multivariate statistical analysis. Linear Algebra its Appl. 6, 217e240. Timmer, M.P., Dietzenbacher, E., Los, B., Stehrer, R., de Vries, G.J., 2015. An illustrated user guide to the world inputeoutput database: the case of global automotive production. Rev. Int. Econ. 23, 575e605. Tittensor, D.P., Walpole, M., Hill, S.L.L., Boyce, D.G., Britten, G.L., Burgess, N.D., Butchart, S.H.M., Leadley, P.W., Regan, E.C., Alkemade, R., Baumung, R., Bellard, C., Bouwman, L., Bowles-Newark, N.J., Chenery, A.M., Cheung, W.W.L., Christensen, V., Cooper, H.D., Crowther, A.R., Dixon, M.J.R., Galli, A., Gaveau, V., €ft, R., Januchowski-Hartley, S.R., Gregory, R.D., Gutierrez, N.L., Hirsch, T.L., Ho Karmann, M., Krug, C.B., Leverington, F.J., Loh, J., Lojenga, R.K., Malsch, K., Marques, A., Morgan, D.H.W., Mumby, P.J., Newbold, T., Noonan-Mooney, K., Pagad, S.N., Parks, B.C., Pereira, H.M., Robertson, T., Rondinini, C., Santini, L., Scharlemann, J.P.W., Schindler, S., Sumaila, U.R., Teh, L.S.L., van Kolck, J., Visconti, P., Ye, Y., 2014. A mid-term analysis of progress toward international biodiversity targets. Science 346, 241e244. UNEP-WCMC, 2011. Review of the Biodiversity Requirements of Standards and Certification Schemes: a Snapshot of Current Practices. Technical Series No. 63. al, Canada. Secretariat of the Convention on Biological Diversity, Montre UNFCCC, 2016. The Paris Agreement. United Nations Framework Convention on Climate Change. unfccc.int/paris_agreement/items/9485.Php (Accessed 24 February 2017). 
r, D., ten Brink, B., Loh, J., Baillie, J.E.M., Reyers, B., 2012. Review of multiVa
cka species indices for monitoring human impacts on biodiversity. Ecol. Indic. 17, 58e67. van Oorschot, M., Rood, T., Vixseboxse, E., Wilting, H., van der Esch, S., 2012. De Nederlandse voetafdruk op de wereld: hoe groot en hoe diep? PBL Netherlands Environmental Assessment Agency, The Hague, The Netherlands. Verboom, J., Snep, R.P.H., Stouten, J., Pouwels, R., Pe’er, G., Goedhart, P.W., van Adrichem, M.H.C., Alkemade, J.R.M., Jones-Walters, L.M., 2014. Using Minimum Area Requirements (MAR) for assemblages of mammal and bird species in global biodiversity assessments. WOt Paper 33. Statutory Research Task Unit for Nature and the Environment. Wageningen University, Wageningen, The Netherlands. Wiedmann, T., 2009. A review of recent multi-region input-output models used for consumption-based emission and resource accounting. Ecol. Econ. 69, 211e222. Wiedmann, T.O., Lenzen, M., Barrett, J.R., 2009. Companies on the scale. J. Ind. Ecol. 13, 361e383. Wilting, H.C., Schipper, A.M., Bakkenes, M., Meijer, J.R., Huijbregts, M.A.J., 2017. Quantifying biodiversity losses due to human consumption: a global-scale footprint analysis. Environ. Sci. Technol. http://dx.doi.org/10.1021/ acs.est.1026b05296. WWF Market Transformation Initiative, 2012. Better Production for a Living Planet. WWF e World Wide Fund For Nature Gland, Switzerland.