|
Prioritisation within The Integrated Product Policy 1 Environmental impact of product groups1.1 Environmental impact of Danish production and consumption 1.1 Environmental impact of Danish production and consumptionAs a decision basis for planning and selecting products for the future product-oriented activities, this chapter provides lists of the product groups and industries with the largest environmental impact potentials. The lists have been made with the method described in Chapter 2, combining environmental statistics and the national accounts. The assessment has been performed for the year 1999, since at the start of the project this was the most recent year for which comprehensive data were available. In the context of this report, environmental impact potentials are defined in terms of eight impact categories:
The methodology for assessing these impacts is described in Chapter 2.10. For ease of reading, we use in the remainder of this report the short-hand “environmental impact” instead of “environmental impact potential”, although it should be understood that all mentioning of impacts in this report relate to impact potentials, not actual impacts. The system boundaries for Danish production and consumption are drawn from a lifecycle perspective, i.e. including all upstream processes from the “cradle”, i.e. material extraction from nature, and downstream to the “grave”, i.e. waste treatment. To provide the most complete picture possible, this report applies several different perspectives on Danish production and consumption:
The lifecycles of each product group have generally been constructed by linking the upstream processes proportionally to the monetary value of the flows between the processes, as is traditionally done in economic input-output analysis and product life cycle assessment. This implies the assumption that a change in demand for a product will lead to a proportional change in production volume in the entire supply chain. To take into account that not all industries can change their production volume in response to a change in demand (for example, because of the quotas on milk production, a change in the output of milk from the dairies will not be able to influence the amount of milk produced in agriculture, and therefore not the environmental impacts from agriculture either), we analysed all industries systematically for long-term production constraints, i.e. constraints that influence investment decisions, like the one mentioned for dairy farms. For the most important constrained industries we have divided the industry in a constrained and a non-constrained part, transferred the constrained supplies to the alternative non-constrained industry and added the constrained outputs as separate products in new final consumption group, typically named “industry name (constrained)”. Since a constrained production is still relevant for non-market-based environmental measures, a constrained product takes part in the same way as any other product in the prioritisation in the supply and process perspectives. More detail on the treatment of constrained industries can be found in Chapter 2.9. A quantitative uncertainty assessment of the results has been performed and is reported in Chapter 1.5. Confidence intervals are provided on the prioritisation results in Chapter 1.4.1. Generally, the difference between the product groups are so large that their overall position in the prioritisation (among the 10 most important, among the 20 most important etc.) is very stable, even for product groups where the environmental impact is determined with relatively large uncertainty. In monetary terms, Danish consumption amounted in year 1999 to 840 GDKK (840*109 DKK) not including product taxes. Out of this, 90 GDKK was products directly imported for final consumption, while 750 GDKK was from domestic production. The domestic production also had an import, amounting to 250 GDKK (not including re-export), but also an export, with a total value of 380 GDKK. These product flows are illustrated in Figure 1.1. It is noteworthy that imports and exports practically outweigh each other, and amounts to less than half of the Danish consumption.
Figure 1.1. The flows of products related to Danish production and consumption, in monetary terms (Data based on the National Accounting matrices for year 1999 as modified in this project, see Chapter 2). A similar picture can be drawn for the environmental impacts related to these product flows, see Figure 1.2. The flows are shown as percentages of the total environmental impact from Danish production and consumption, expressed as an average of the eight environmental impact categories, i.e. where all environmental impact categories are weighted equally, see Chapter 2.10.4. For a more detailed picture, please refer to the similar figures for each impact category in Chapter 1.4. Seen from the supply side, the total environmental impacts (100%) can be split into those related to Danish activities (42% from Danish production and 6% from the final use stage) while the remaining 52% are environmental impacts abroad related to the products imported to Denmark (12% directly for final use and 40% for the products used by Danish industries).
Figure 1.2. The environmental impact potential related to Danish production and consumption, in percentage of the total. Seen from the consumption side, the same 100% can be split into that which is related to Danish consumption (12%+29%+6% = 47%) and the 53% related to the products exported. Comparing Figure 1.1 and Figure 1.2, it is clear that Danish foreign relations are proportionally much more environmentally important than their monetary flows indicate. In other words, both imported products and products produced for export in general cause more environmental impact than products produced in Denmark for the Danish market. Figures similar to Figure 1.1 and Figure 1.2 could be made for each single product group, thus providing information on how environmental impact is related to import and export of that commodity. This could be useful e.g. when discussing how emission quota can best be designed and administrated. 1.2 Product groups with largest environmental impactsThis sub-chapter provides the overall results of the developed prioritisation method, as applied to Danish production and consumption. For details on the methodology, please see Chapter 2, and for details for each impact category, please see Chapter 1.4. 1.2.1 Product groups within Danish productionFirst, we apply the supply or net production perspective, i.e. we look at the product groups supplied by Danish industries going either to final consumption or export. This is a “cradle to gate” perspective, where the gate is the point where the product leaves the Danish industry. For example for pork and pork products, it includes all processes (and their environmental impacts) upstream of and including the meat processing industry, but not the wholesale and retail sale and the final use in the households (or in industries abroad for exported products). Wholesale trade and retail sale are included as separate services. Within this perspective, the product groups with the largest environmental impacts are shown in the list below (the export percentage is shown in brackets, except for product groups where the production is not demand-driven, due to constraints on production volume or emissions, see also Chapter 2.9):
Together, these 10 products groups (out of a total of 138) account for 45% of the total environmental impact from Danish production and consumption. Pork and pork products rank high on all impact categories; see Chapter 1.4.1. This is partly due to the large share of pork production in the Danish economy (1.3% of the total production value), but also due to a high impact per monetary value for many of the impact categories; see Chapter 1.4.2. Dwellings, i.e. the management of residential buildings, also rank high on all impact categories; see Chapter 1.4.1. This is mainly due to the large share of this industry in the Danish economy (more than 5% of the total production value). Transport by ship ranks high on all impact categories, except nature occupation; see Chapter 1.4.1. This is partly due to the large share of shipping in the Danish economy (close to 3% of the total production value), but also due to a relatively high impact per monetary value (see Chapter 1.4.2), especially for the impact category ecotoxicity (due to the antifouling agent tributyltinoxide (TBTO)), but also for acidification (due to emissions of SO2 and NOx). The latter is in spite of a specific 75% reduction in the value attached to these emissions from shipping due to the lower expose expected from emissions at sea (see Chapter 2.10.1). As can be further seen in Chapter 1.4.1:
Industrial cooling equipment has been included on the list mainly because its net production alone accounts for 17% of the total ozone depletion potential related to Danish production and consumption. In chapter 1.7, we discuss the possible implications, in terms of improvement options for the above product groups. 1.2.2 Product groups within Danish consumptionNext, we apply the consumption perspective, i.e. we look at the product groups from either foreign or Danish industries going to final consumption in Denmark, both private and public. This is a complete “cradle to grave” perspective on these product groups. This implies that wholesale, retail sale and the use stage are included for each product, unless specifically excluded. For example, a product group is specifically called “meat purchase” to denote that cooking is not included (since it is reported separately as “cooking in household”), while “Dwellings and heating” include all use stage emissions from the dwellings. The following product groups within Danish consumption (both private and public) have been identified as the ones with the largest environmental impacts:
Together, these 10 products groups (out of a total of 98) account for 57% of the total environmental impact from Danish consumption, and 25% of the total impact from Danish production and consumption. Dwellings (with or without inclusion of heating), meat purchase, tourist expenditures, clothing and catering rank high on all impact categories; see Chapter 1.4.1. General public services, public order and safety affairs, Personal hygiene and Education and research rank high on all impact categories except nature occupation. Heating and electricity are important contributors to this, while sewage treatment is also important for personal hygiene and buildings play an important role for education and research. The public consumption group “General public services, public order and safety affairs” has its main input from the above-mentioned public industry “Defence, justice, public security and foreign affairs.” As can be further seen in Chapter 1.4.1:
In chapter 1.7, we discuss the possible implications, in terms of improvement options for the above product groups. 1.2.3 Inherent limitations of product group aggregationWhen identifying the most important product groups, as in the preceding sections, it is unavoidable that the result is influenced by how the product groups are defined, and especially their level of aggregation. A highly aggregated product group is more likely to show up among the top 10, and by disaggregating it into a number of smaller product groups, it can be made to disappear from the top 10. For example, education and research only reaches the top-10 of environmental impact because it is a very aggregated product group. In it self, education has very low environmental impact intensity (see chapter 1.4.2) and would not have reached the top 10 if it had been divided into primary, secondary and higher education, and adult education etc. To counter this inherent arbitrariness in the ranking, several complementary approaches can be applied: One option is to apply a functional approach, where the division between product groups is based on what needs the different products fulfil. Since this approach breaks down the entire consumption top-down, it becomes impossible to hide important product groups. A first step of this approach is applied in Chapter 1.2.4. The linking of products in the use stage, described in Chapter 2.7.2, is also a part of this approach. Another way of avoiding arbitrariness is to rank the product groups according to their environmental impact intensity, i.e. their impact per monetary value, as is done in Chapter 1.3. A product with a large impact per economic value will then appear on the top 10 also when disaggregated. In this approach, the only way an important product can disappear from the top 10 is if it is aggregated with another product with a low environmental impact. This means that it is still possible that very inhomogeneous product groups (in terms of impact intensity) can conceal products with large impact intensities. However, this problem can be solved by appropriate disaggregation. Thus, to limit the arbitrariness in the ranking, it is recommended first to apply a ranking according to impact intensity (as in Chapter 1.3), then to disaggregate the most inhomogeneous product groups (already done for the results presented in Chapter 1.3, see also Chapters 2.6 and 2.7), and finally to supplement this with a ranking that also takes into account the volume of the product groups, first in terms of need groups (see Chapter 1.2.4) and secondly in a disaggregated analysis (as in Chapters 1.2.1, 1.2.2). Another advantage of the ranking according to impact intensity is its ability to answer questions related to sustainable consumption, such as: “Given a specific level of consumption in monetary terms, which products should be chosen to reduce the overall impact the most, and what products should be deselected?” The first part of the question focuses on the product groups with the least impacts per monetary unit, i.e. the opposite focus compared to the top 10 ranking. By combining the impact intensities with consumption trends, it is also possible to calculate the environmental impact of the marginal consumer spending. These issues are treated in Chapters 1.4.2 and 1.4.3. 1.2.4 Danish consumption divided according to product functions (need groups)Product groups or groups of final consumption can be divided according to product functions, i.e. relating to the satisfaction of specific human needs. There are several suggestions on how to classify human needs. Within the field of psychology, Maslow (1954) and more recently Max-Neef (1992) have proposed sets of basic human needs. Sen (1998) and Nussbaum (1998) propose to characterize basic needs as necessary “capabilities to function.” Segal (1998) noted that this concept can be quite contestable across cultures, and how degree of satisfaction for many of the need categories would be difficult or impossible to assess in practice. He proposed instead a more physically-grounded, less psychologically descriptive need framework, focusing on a subset of the basic human needs, which he termed “core economic needs”. Advantages of this approach are that its applicability has been demonstrated in practical empirical work and that it provides a stronger linkage between consumption and affluence and its basis in products. We have applied a slight modification of Segal's set of core economic needs, in order to adequately cover all consumption groups in the NAMEA. We have expanded Segal's concept of child care to social care in general and the concept of economic security to security in general, added hygiene and leisure as need groups, and redefined the need for transportation into a need for communication, while splitting out part of car driving on food purchase and leisure. The resulting 10 need groups are (with share of total economic expenditure in brackets):
An additional group of “Other consumption not elsewhere classified” accounts for the remaining 7%. This group covers mainly “infrastructure” expenditures, such as interest etc. on financial investments, and economic affairs and services. It is interesting to note from Figure 1.3 that the environmental impact is concentrated on a few need groups and does not follow the economic expenditure. This means that the need groups differ significantly in impact intensity, as is also shown in Table 1.1.
Figure 1.3. Environmental impact per need group in Danish consumption. Table 1.1. Need groups in Danish consumption, ranked according to environmental impact intensity. Impacts are shown in person-equivalents (PE), i.e. the total environmental impact caused by the production and consumption of an average Dane in 1999 (see Chapter 2.10.3).
Note that the sum of the environmental impacts in Table 1.1 amounts to 44% of the impacts from the Danish production and consumption, compared to the 47% in Figure 1.2. The difference is due to constrained productions within the Danish industries, which are not included in the demand-driven environmental impact in Table 1.1 (see also Chapter 2.9). In chapter 1.7, we discuss the improvement options for the most important need groups. 1.2.5 Largest environmental impacts per DKKThe following product groups supplied by Danish industry (for domestic final consumption or for export) have been identified as the ones with the largest environmental impact per DKK:
Meat and meat products covers beef, pork and chicken meat and thus also represents the alternative supply for the constrained production of fish, seafood and fish products. Agricultural products in general covers the products directly bought on farms for final consumption (including export). These product groups rank high per DKK on all impact categories, except ozone depletion and human toxicity; see Chapter 1.4.2. As can further be seen in Chapter 1.4.2:
The following product groups within Danish consumption have been identified as the ones with the largest environmental impact per DKK:
Fireworks rank high per DKK on all impact categories, except nutrient enrichment and nature occupation; see Chapter 1.4.2. Car driving abroad ranks high per DKK on all impact categories, except ecotoxicity and nature occupation; see Chapter 1.4.2. Car driving in Denmark is not included in the list because it is more expensive than car driving abroad, which make it come out lower per DKK. Meat purchase rank high per DKK on all impact categories, except ozone depletion, photochemical ozone and human toxicity; see Chapter 1.4.2. Non-durable household goods n.e.c. (not elsewhere classified) ranks high on all impact categories except global warming, nutrient enrichment and nature occupation; see Chapter 1.4.2. The product group is very diverse, covering items such as labels, polishes, minor textile items, wrapping paper, brooms and brushes, carbondioxide cartridges and pesticides. It is one of the product groups that would be recommendable to subdivide for a more detailed analysis. As can further be seen in Chapter 1.4.2:
It is interesting to note that six of the ten listed product groups within Danish consumption relate exclusively to imported products. This reflects the relatively large environmental impact intensity of foreign production; see also Chapter 2.8. If we focus exclusively on domestically produced product groups within Danish consumption, the ones with the largest environmental impact per DKK are (besides fireworks, meat, non-durable household goods and potatoes):
As can be seen in Chapter 1.4.2:
Toilet flush ranks high per DKK on nutrient enrichment and ecotoxicity; see Chapter 1.4.2. The ecotoxicity can be traced back to emissions of copper, zinc and cadmium from corrosion of galvanised products. The share of this corrosion that is attributed to toilet flush is relatively small (15%), but because toilet flush is a relatively cheap activity, it still comes out high per DKK. In chapter 1.7, we discuss the possible implications, in terms of improvement options, for the above product groups. It is noteworthy that many of the product groups appearing in the lists in Chapters 1.2.1 and 1.2.2 (large overall improvement potentials) also appear here with large improvement potentials per DKK. This implies that these product groups are not only of interest due to their size, but also “in their own right”. At the other end of the scale, we find the products with low environmental impact intensity, which appear particularly to be services, e.g. bookkeeping and auditing, insurance, social security, financial and legal services, education and research, kindergartens and crèches, home and day care services and retirement homes; see Chapter 1.4.2. It is obvious that the products with high environmental impact intensities, such as food and transport, cannot be directly substituted by these low impact intensity services, since they do not fulfil the same needs. Likewise, even though transport by air has a lower environmental impact intensity than transport by ship, an item transported by air still involves more environmental impact than when transported by ship, simply because transport by air is more costly. However, the information on impact intensities can be used to point out the products for which it would be highly desirable to search for satisfactory substitutes, which may go beyond the mere substitution of products with identical functional properties. For example, the alternative to transport by ship is not necessarily another form of transport, but could also be a relocation of the production. Similarly, the general consumer welfare would not necessarily be affected by a non-compensated reduction in the amount of (high-impact-intensity) meat consumed. This could point to possible, desirable changes in the general consumption pattern. At a more general level, the information on impact intensities points out that it is an environmentally beneficial strategy to increase the service content of the products – provided the customers are willing to pay for this – since the value added by human labour adds no environmental impact. 1.3 Processes with largest environmental impactsIt is also possible to analyse the results across all product groups, to identify processes that have large contributions to the overall environmental impact without necessarily being suppliers of final consumption goods. For results per impact category, please see Chapter 1.4.4. The processes with the largest contributions to the environmental impacts from Danish production and consumption have been identified as:
It should be noted that while Danish processes (DK) are true gate-to-gate processes, the foreign (ROW – Rest-Of-World) processes are terminated cradle-to-gate supply chains. As can be seen in Chapter 1.4.4:
Further, Danish production of industrial cooling equipment accounts for 29% of the total ozone depletion potential related to Danish production and consumption. This may be compared to the 17% noted in Chapter 1.2.1, which is for the net production only, i.e. the industrial cooling equipment entering into final consumption, which in this case is entirely export. The difference (12%) is the amount used by Danish industry itself. 1.4 Results per impact category1.4.1 Environmental impact of Danish production and consumptionIn this sub-chapter, we look at each environmental impact category separately, providing both the overall picture (in the Figures) and ranked data tables showing the most important product groups. The tables include all product groups with a result of more than 10% of the top-ranking product group, or at least 15 product groups. 1.4.1.1 Global warming
Figure 1.4. The Global Warming Potential (GWP) related to Danish production and consumption, in percentage of the total, of which the GWP from Danish activities amount to 53%. The GWP related to Danish consumption is 11%+36%+8% = 55%, while 45% is related to Danish export. Table 1.2. Product groups within Danish net production with the largest Global Warming Potential (GWP), in person-equivalents (PE) and % of total GWP from Danish production and consumption.
1) The value shown represents the total impact from Danish electricity and heat minus the values shown for “Electricity (unconstrained)” and “District heat (unconstrained)” Table 1.3. Product groups within Danish consumption with the largest Global Warming Potential (GWP), in person-equivalents (PE) and % of total GWP from Danish production and consumption.
1.4.1.2 Ozone depletion
Figure 1.5. The Ozone Depletion Potential (ODP) related to Danish production and consumption, in percentage of the total, of which the ODP from Danish activities amount to 29%. The ODP related to Danish consumption is 16%+26% = 42%, while 58% is related to Danish export. Table 1.4. Product groups within Danish net production with the largest Ozone Depletion Potential (ODP), in person-equivalents (PE) and % of total ODP from Danish production and consumption.
Table 1.5. Product groups within Danish consumption with the largest Ozone Depletion Potential (ODP), in person-equivalents (PE) and % of total ODP from Danish production and consumption.
1.4.1.3 Acidification
Figure 1.6. The Acidification Potential (AP) related to Danish production and consumption, in percentage of the total, of which the AP from Danish activities amount to 55%. The AP related to Danish consumption is 10%+29%+4% = 43%, while 57% is related to Danish export. Table 1.6. Product groups within Danish net production with the largest Acidification Potential (AP), in person-equivalents (PE) and % of total AP from Danish production and consumption.
1) The value shown represents the total impact from Danish electricity and heat minus the values for “Electricity (unconstrained)” and “District heat (unconstrained)” Table 1.7. Product groups within Danish consumption with the largest Acidification Potential (AP), in person-equivalents (PE) and % of total AP from Danish production and consumption.
1.4.1.4 Nutrient enrichment
Figure 1.7. The Nutrient Enrichment Potential (NEP) related to Danish production and consumption, in percentage of the total, of which the NEP from Danish activities amount to 62%. The NEP related to Danish consumption is 13%+28%+3% = 44%, while 56% is related to Danish export. Table 1.8. Product groups within Danish net production with the largest Nutrient Enrichment Potential (NEP), in person-equivalents (PE) and % of total NEP from Danish production and consumption.
1) The value shown represents the total impact from Danish electricity and heat minus the values for “Electricity (unconstrained)” and “District heat (unconstrained)” Table 1.9. Product groups within Danish consumption with the largest Nutrient Enrichment Potential (NEP), in person-equivalents (PE) and % of total NEP from Danish production and consumption.
1.4.1.5 Photochemical ozone creation
Figure 1.8. The Photochemical Ozone Creation Potential (POCP) related to Danish production and consumption, in percentage of the total, of which the POCP from Danish activities amount to 38%. The POCP related to Danish consumption is 14%+31%+17% = 62%, while 38% is related to Danish export. Table 1.10. Product groups within Danish net production with the largest Photochemical Ozone Creation Potential (POCP), in person-equivalents (PE) and % of total POCP from Danish production and consumption.
1) The value shown represents the total impact from Danish electricity and heat minus the values for “Electricity (unconstrained)” and “District heat (unconstrained)” Table 1.11. Product groups within Danish consumption with the largest Photochemical Ozone Creation Potential (POCP), in person-equivalents (PE) and % of total POCP from Danish production and consumption.
1.4.1.6 Ecotoxicity
Figure 1.9. The Ecotoxicity Potential (ETP) related to Danish production and consumption, in percentage of the total, of which the ETP from Danish activities amount to 74%. The ETP related to Danish consumption is 5%+21%+2% = 28%, while 72% is related to Danish export. Table 1.12. Product groups within Danish net production with the largest Ecotoxicity Potential (ETP), in person-equivalents (PE) and % of total ETP from Danish production and consumption.
Table 1.13. Product groups within Danish consumption with the largest Ecotoxicity Potential (ETP), in person-equivalents (PE) and % of total ETP from Danish production and consumption.
1.4.1.7 Human toxicity
Figure 1.10. The Human Toxicity Potential (HTP) related to Danish production and consumption, in percentage of the total, of which the HTP from Danish activities amount to 14%. The HTP related to Danish consumption is 14%+36%+3% = 53%, while 47% is related to Danish export. Table 1.14. Product groups within Danish net production with the largest Human Toxicity Potential (HTP), in person-equivalents (PE) and % of total HTP from Danish production and consumption.
1) The value shown represents the total impact from Danish electricity and heat minus the values for “Electricity (unconstrained)” and “District heat (unconstrained)” Table 1.15. Product groups within Danish consumption with the largest Human Toxicity Potential (HTP), in person-equivalents (PE) and % of total HTP from Danish production and consumption.
1.4.1.8 Nature occupation
Figure 1.11. The Nature Occupation Potential (NOP) related to Danish production and consumption, in percentage of the total, of which the NOP from Danish activities amount to 58%. The NOP related to Danish consumption is 14%+26%+6% = 46%, while 54% is related to Danish export. Table 1.16. Product groups within Danish net production with the largest Nature Occupation Potential (NOP), in person-equivalents (PE) and % of total NOP from Danish production and consumption.
Table 1.17. Product groups within Danish consumption with the largest Nature Occupation Potential (NOP), in person-equivalents (PE) and % of total NOP from Danish production and consumption.
1.4.2 Environmental impact intensitiesIn this sub-chapter, we look at what product groups have the largest environmental impact intensity, i.e. environmental impact per DKK, still for each environmental impact category separately. We also look at the product groups with the smallest impact intensity, i.e. with the least environmental impact per DKK. This information is especially relevant when discussing “de-coupling”, i.e. how a reduction in environmental impact can be achieved without necessarily reducing the total level of consumption. All product groups with a result of more than 10% of the top-ranking and 10 times the bottom-ranking product group are included in the tables, except when closer than a factor two to the average product, and never less than 15 product groups. 1.4.2.1 Global warming intensities within Danish productionTable 1.18. Product groups within Danish production with the largest Global Warming intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product from Danish production.
Table 1.19. Product groups within Danish production with the smallest Global Warming intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product from Danish production.
1.4.2.2 Global warming intensities within Danish consumptionTable 1.20. Product groups within Danish consumption with the largest Global Warming intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product within Danish consumption.
Table 1.21. Product groups within Danish consumption with the smallest Global Warming intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product within Danish consumption.
1.4.2.3 Ozone depletion intensities within Danish productionTable 1.22. Product groups within Danish production with the largest Ozone Depletion intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product from Danish production.
Table 1.23. Product groups within Danish production with the smallest Ozone Depletion intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product from Danish production.
1.4.2.4 Ozone depletion intensities within Danish consumptionTable 1.24. Product groups within Danish consumption with the largest Ozone Depletion intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product within Danish consumption.
Table 1.25. Product groups within Danish consumption with the smallest Ozone Depletion intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product within Danish consumption.
1.4.2.5 Acidification intensities within Danish productionTable 1.26. Product groups within Danish production with the largest Acidification intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product from Danish production.
Table 1.27. Product groups within Danish production with the smallest Acidfication intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product from Danish production.
1.4.2.6 Acidification intensities within Danish consumptionTable 1.28. Product groups within Danish consumption with the largest Acidification intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product within Danish consumption.
Table 1.29. Product groups within Danish consumption with the smallest Acidification intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product within Danish consumption.
1.4.2.7 Nutrient enrichment intensities within Danish productionTable 1.30. Product groups within Danish production with the largest Nutrient Enrichment intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product from Danish production.
Table 1.31. Product groups within Danish production with the smallest Nutrient Enrichment intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product from Danish production.
1.4.2.8 Nutrient enrichment intensities within Danish consumptionTable 1.32. Product groups within Danish consumption with the largest Nutrient Enrichment intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product within Danish consumption.
Table 1.33. Product groups within Danish consumption with the smallest Nutrient Enrichment intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product within Danish consumption.
1.4.2.9 Photochemical ozone intensities within Danish productionTable 1.34. Product groups within Danish production with the largest Photochemical Ozone intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product from Danish production.
Table 1.35. Product groups within Danish production with the smallest Photochemical Ozone intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product from Danish production.
1.4.2.10 Photochemical ozone intensities within Danish consumptionTable 1.36. Product groups within Danish consumption with the largest Photochemical Ozone intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product within Danish consumption.
Table 1.37. Product groups within Danish consumption with the smallest Photochemical Ozone intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product within Danish consumption.
1.4.2.11 Ecotoxicity intensities within Danish productionTable 1.38. Product groups within Danish production with the largest Ecotoxicity intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product from Danish production.
Table 1.39. Product groups within Danish production with the smallest Ecotoxicity intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product from Danish production.
1.4.2.12 Ecotoxicity intensities within Danish consumptionTable 1.40. Product groups within Danish consumption with the largest Ecotoxicity intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product within Danish consumption.
Table 1.41. Product groups within Danish consumption with the smallest Ecotoxicity intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product within Danish consumption.
1.4.2.13 Human toxicity intensities within Danish productionTable 1.42. Product groups within Danish production with the largest Human Toxicity intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product from Danish production.
Table 1.43. Product groups within Danish production with the smallest Human Toxicity intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product from Danish production.
1.4.2.14 Human toxicity intensities within Danish consumptionTable 1.44. Product groups within Danish consumption with the largest Human Toxicity intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product within Danish consumption.
Table 1.45. Product groups within Danish consumption with the smallest Human Toxicity intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product within Danish consumption.
1.4.2.15 Nature occupation intensities within Danish productionTable 1.46. Product groups within Danish production with the largest Nature Occupation intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product from Danish production.
Table 1.47. Product groups within Danish production with the smallest Nature Occupation intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product from Danish production.
1.4.2.16 Nature occupation intensities within Danish consumptionTable 1.48. Product groups within Danish consumption with the largest Nature Occupation intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product within Danish consumption.
Table 1.49. Product groups within Danish consumption with the smallest Nature Occupation intensity, in person-equivalents per monetary unit (PE/kDKK) and relative to an average product within Danish consumption.
1.4.3 Impact of average and marginal consumptionIn the preceding sub-chapter on the environmental impact intensity of products, we already introduced, for the purpose of comparison, the concept of the environmental impact intensity of an average consumed product. This concept can be broken down into the impact intensity of the average consumer spending, and the impact intensity of the average public spending. These values can be found in Table 1.50. The values can be derived from the normalisation reference (see Chapter 2.10.3) as the share of each impact category related to Danish consumption (split out on private and public consumption) divided by the value of the total annual Danish consumption including product-related taxes (which is 644 GDKK for private consumption and 312 for GDKK public consumption). Note that share of each impact category related to Danish consumption in Table 1.50 does not necessarily add up to the percentage calculated in the figure text of each Figure in Chapter 1.4.1, since the values in Table 1.50 only relates to the part of the environmental impact that can be influenced by demand changes (see also Chapter 2.9). Table 1.50. The environmental impact of the average Danish consumption
An interesting comparison of impact intensities of private and public consumption can be made from Table 1.50. It can be seen that one DKK used by public authorities has an environmental impact between 13% and 64% of that of one DKK used by a private Dane, depending on the impact category. Popularly speaking, we may thus pay our tax with a smile, at least seen from an environmental perspective. When applying the data on impact intensity for comparisons, e.g. in the context of “de-coupling”, it is in fact not the environmental impact of the average spending which is of interest, but rather the environmental impact of the marginal spending, i.e. the impact of the last or an additional DKK spent. The composition of the marginal spending can be derived by studying the change in consumption either as the entire economy grows (as shown in Weidema 2003, Figure 3.2) or as the spending of individual consumers grow. The latter approach will give a better estimate, since the change in consumption over time for the entire economy may be influenced by many other trends in consumption patterns than that relating to increased spending. Data on the consumption patterns of consumers with different income levels are available from Statistics Denmark, which should allow determination of the change in spending with increasing income. It has not been possible within the limits of this project to determine the composition of the marginal spending in this preferred way, but once such a composition is specified, it is straightforward to calculate the environmental impact of the marginal spending by combining the composition of the spending with the emission intensities per industry provided by the database from this project (see Chapter 7). 1.4.4 Processes with large contributions to each impact categoryWe have also analysed the results for each environmental impact category across all product groups, to identify the processes that have large contributions without necessarily being suppliers of final consumption goods. These processes are shown in tables 1.51 to 1.58. The tables include all processes with a result of more than 10% of the top-ranking process, or at least 15 processes. Note that while Danish (DK) processes are true gate-to-gate processes, the foreign (ROW) processes are terminated cradle-to-gate supply chains. Table 1.51. Processes within Danish production and consumption with the largest Global Warming Potential (GWP), in person-equivalents (PE) and % of total GWP from Danish production and consumption.
1) The value shown represents the total impact from Danish electricity and heat minus the values shown for “Electricity (unconstrained)” and “District heat (unconstrained)” Table 1.52. Processes within Danish production and consumption with the largest Ozone Depletion Potential (ODP), in person-equivalents (PE) and % of total ODP from Danish production and consumption
Table 1.53. Processes within Danish production and consumption with the largest Acidification Potential (AP), in person-equivalents (PE) and % of total AP from Danish production and consumption.
1) The value shown represents the total impact from Danish electricity and heat minus the values shown for “Electricity (unconstrained)” and “District heat (unconstrained)” Table 1.54. Processes within Danish production and consumption with the largest Nutrient Enrichment Potential (NEP), in person-equivalents (PE) and % of total NEP from Danish production and consumption.
Table 1.55. Processes within Danish production and consumption with the largest Photochemical Ozone Creation Potential (POCP), in person-equivalents (PE) and % of total POCP from Danish production and consumption.
Table 1.56. Processes within Danish production and consumption with the largest Ecotoxicity Potential (ETP), in person-equivalents (PE) and % of total ETP from Danish production and consumption.
Table 1.57. Processes within Danish production and consumption with the largest Human toxicity Potential (HTP), in person-equivalents (PE) and % of total HTP from Danish production and consumption.
Table 1.58. Processes within Danish production and consumption with the largest Nature Occupation Potential (NOP), in person-equivalents (PE) and % of total NOP from Danish production and consumption.
1.5 Uncertainty of the resultsThis sub-chapter explains the results of the uncertainty analysis performed. More detail on the actual procedures applied for uncertainty analysis is provided in Chapter 2.11. 1.5.1 Confidence intervalsApproximate 95% confidence intervals are given with the ranking results in Tables 1.2 to 1.15 (95% confidence interval given by 2 divided by the mean value). The confidence intervals are expressed as a percentage so as to give an indication of the relative uncertainty of the product group totals. This relative uncertainty can best be explained graphically, as in Figure 1.12, which is a graphical representation of the data in Table 1.15. It shows the most likely value and the range in which 90% of the data sample falls. The higher the % uncertainty in Table 1.15, the greater the range spanned by the 90% confidence interval in Figure 1.12.
Figure 1.12. Top fourteen product groups within Danish consumption contributing to the human toxicity potential (all private consumption in Denmark, unless otherwise noted). The bars give the median value, while the whiskers show the 90% confidence intervals. The "whiskers" in Figure 1.12 show the degree of overlap between the product group totals, and is thus an indication of the reliability of the ranking. For example, there is a fair degree of overlap in the confidence intervals of "Dwellings and heating" and "Car purchase and driving", although the former is shown to contribute 30% more to human toxicity potential than the latter. The degree of overlap can be quantified by taking the normalised difference between the two product groups and plotting the resulting cumulative probability; see Figure 1.13. The point at which the normalised difference curve cuts the x = zero line gives the cumulative probability that "Car purchase and driving" always has a higher human toxicity potential than "Dwellings and heating" (approximately 10% in Figure 1.13). Conversely, this means that for 90% of the cases, the shown ranking will occur.
Figure 1.13. Cumulative probability curves of a pair-wise comparison of top four product groups shown in Figure 1.12 Looking further down the ranking list in Table 1.15, the difference between "Car purchase and driving" and "Tourist expenditure" is predicted with a high degree of confidence (no overlap in their 90% confidence intervals in Figure 1.12), which is confirmed in Figure 1.13 by the fact that the normalised difference between these two product groups lies almost completely above the zero line (99% confidence). However, the subsequent product groups shown in Table 1.15 do not differ substantially from each other in their median values. Even with fairly low uncertainty values, the small differences between the mean values of these product groups implies that it can not be stated with confidence whether one group can always be predicted to have a higher contribution to human toxicity than another. This is shown by the zero line having a y-intercept of 0.5 in Figure 1.13, i.e. about equal probability of one being higher than the other. It should be noted when interpreting the confidence intervals in the Tables 1.2-1.15 that these account only for empirical variability in the data on economic flows and emissions (see section 2.11). Uncertainty in the final use stage (both economic and emissions data) is not included in the analysis, and neither is the uncertainty of the impact assessment factors used to calculate the impact potentials. Including these later steps would be likely to reduce the uncertainty further, since the aggregation involved has the effect of "levelling out" the high uncertainties in some of the economic and emissions data, without adding to the overall uncertainty. The confidence intervals reported in the tables are calculated for the particular product group, normalised by an average product group. This normalisation step is carried out because it removes the distorting effect of uncertain elements common to all the product groups. However, the effect of this is limited in this analysis (there is very little difference between the actual confidence interval of the product group alone and the confidence interval of the ratio). If the uncertainty of impact assessment had been included, the normalisation step would be of larger importance, since it would introduce a large amount of uncertainty common to all product groups. Generally, the difference between the product groups are so large that their overall position in the prioritisation (among the 10 most important, among the 20 most important etc.) is very stable, even for product groups where the environmental impact is determined with relatively large uncertainty. 1.5.2 Causes of the highest variations in the resultsIn addition to giving an indication of the reliability of the observed ranking, the uncertainty analysis provides information as to where the highest variations are occurring, and thus where efforts can best be focussed to reduce the uncertainty. Unfortunately, the very large number of data inputs in the model and the consequent very high computer memory requirement, has made it impossible for us to apply a complete correlation analysis, as would typically be the case in an uncertainty importance analysis. Instead, we have done this manually, by identifying which product groups are calculated to have high variation and then moving backwards through the calculations to identify those data on emissions and economic flows contributing to that particular product group that have disproportionately high variation. The causes for these high input variations were then investigated. For example, in Figure 1.12, "Toys" stand out as having a particularly wide confidence interval. With a coefficient of variance of 0.53, "Toys, DK private consumption" has the highest variance of all product groups with respect to human toxicity potential. Looking back one level into the calculation, it can be seen that "Toys" is the most uncertain product group in both "human toxicity water" and "human toxicity soil". Looking further back at the individual emissions, mercury emissions to soil and water both show up with high variance for "Toys". Notably, the uncertainty in mercury emissions for "Toys" is higher than for most other mercury emissions, because the data on economic flows for “Toys” is also relatively uncertain due to the large number of diverse products included in this group. However, other emissions contributing to human toxicity are shown to have higher uncertainty than mercury (e.g. Cu to air at a coefficient of variance of 0.79), but these do not show up with high relative uncertainty for the product group "Toys". This indicates that a product group with high mercury emissions is causing the high uncertainty in this product group. On the side of economic data, a number of product groups with high variability contribute substantially to "Toys", notably "Industrial machinery", with a coefficient of variance of 0.86 for its supply to "Toys, gold and silver articles etc.". In general, what is dominating the overall uncertainty depends on the particular impact category and the particular product group. For certain impact categories one emission can be clearly seen to dominate the uncertainty. For example, the higher than expected uncertainty values for global warming are primarily a result of a very high uncertainty estimate for nitrous oxide emissions. This explains why various farming and food related product groups account for all but three of the top thirty product groups showing the highest uncertainty on global warming. However, identifying one emission with a far higher uncertainty than the other emissions is not always sufficient to explain the dominant source of uncertainty, as e.g. for acidification. Here, even though ammonia emissions are estimated with far higher uncertainty than the other emissions contributing to acidification, sulphur dioxide emissions are the most significant source of uncertainty for the product group with the highest acidification uncertainty (“Car driving for holidays abroad”). This is because in an uncertainty importance analysis, the magnitude of the emission is as important at its uncertainty. For photochemical ozone formation, the uncertainty estimates for non-methane VOC are somewhat higher than the other emissions contributing to this impact category, although this is less marked than for global warming and acidification. Nonetheless, it is sufficient for non-methane VOCs to account for "Hairdressing etc. in DK, private consumption" showing up as the product group with the highest photochemical ozone formation uncertainty. For ecotoxicity, human toxicity and nutrient enrichment, many highly uncertain emissions contribute. For these impacts the overall uncertainty is therefore more a function of which product groups have high emissions of a particular emission type, rather than the uncertainty of predicting that emission. It is also difficult to generalise the relative uncertainty importance of the economic data versus the emissions data. In certain cases, such as the high acidification uncertainty predicted for "Car driving for holidays abroad", it is clearly the economic data that dominates. This is because the emission type showing up with the highest uncertainty importance (in this case sulphur dioxide) has a relatively low emission data uncertainty. However, where the overall uncertainty of the product groups is dominated by a substance with high emission uncertainty estimates, the relative importance of the emission data uncertainty and the economic data uncertainty is more difficult to gauge without further analysis. 1.5.3 Uncertainty from looking at one single yearBesides the quantitative uncertainty analysis reported above, we have also investigated to what extent the results of the prioritisation is influenced by the fact that we have used data for one specific year only (year 1999). If one industry had an unusual low output in year 1999, this could mean that its impact would be underestimated, and vice versa an unusual high output could mean an overestimate of its impact in a life cycle perspective. This would be especially true for products with a lifetime beyond one year, where large variations in consumption could occur. We therefore analysed the production volume in fixed prices in the period 1990-1999, to see if 1999 was an untypical year for any particular industry. Here we define untypical as a deviation of more than 12% over the average of the last 3 years). Using this definition, we found that following industries with particularly high outputs in 1999 (with percentages above the average of the last 3 years in brackets):
In the few cases were these industries appear on the lists of highly prioritised industries in chapters 1.2 to 1.4 a decrease in output of 18-22% would only imply moving one step down on the lists. Even the highest variation in output (for office machinery and computers) would not move the position of this industry notably. This is due to the fairly large differences between the top-ranking industries, in terms of environmental impacts. We thus conclude that the prioritisation has not been notably influenced by the possible overestimation of the environmental impact from these industries due to an untypical high output in year 1999. Industries with particularly low outputs in 1999 were (with percentages below the average of the last 3 years in brackets):
As for the industries with untypical high outputs in 1999, these small percentage variations are not enough to influence the prioritisation notably. We similarly analysed the Danish consumption in fixed prices in the period 1990-1999, and found only few consumption groups that were atypically high in 1999, notably insurance, computers and renting of machinery and equipment. As above, the variations were not of a size that could influence the prioritisation. We did not find any product groups with a particularly low level of consumption in 1999. 1.6 Comparison with results of previous similar studiesOther previous studies with similar objectives, i.e. to identify the most important product groups from an environmental perspective, include Hansen (1995a) and Dall et al. (2002) for Denmark, Finnveden et al. (2001) for Sweden, Nijdam & Wilting (2003) for the Netherlands, Nemry et al. (2002) for Belgium, and Labouze et al. (2003) for EU. The Swedish and Dutch study use the same general methodology as our study (IO-analysis) while the remaining studies use a bottom-up process based analysis. Due to the environmental indicators used (energy consumption and resource loss) the product groups that are ranked high by Hansen (1995a) are those with either large energy consumption or which are destroyed or dissipated during use. This includes the main energy carriers, transport activities (represented by the vehicles including their use phases), fertilizers, animal feeds, meat and dairy products, and building materials. These items are also ranked high by our study (see Chapter 1.2.1), although under slightly different names, for example we regard electricity and heating as the products to be ranked, while Hansen (1995a) ranks the energy carriers including their use phase. The focus on resource loss implies that Hansen (1995a) ranks some products high for which a large part of the material volume is dissipated, such as detergents, newspaper, beer and furniture. Such products do not appear as high in our prioritisation. Dall et al. (2002) have a consumption perspective and include only private consumption. The study focuses mainly on energy consumption and concludes that food, car driving, and housing are the most important product groups, which confirms our findings. Also clothing and personal hygiene appear high in energy consumption. The aggregation of product groups, as well as the differences in methodology, makes it difficult to perform further comparisons at a more detailed level. The product groups that are ranked high by Finnveden et al. (2001) for the emissions of CO2, SO2 and NOx, are electricity and heat, food, dwellings, transport activities, and hotels and restaurants, which are also ranked high by our study. The fact that retail trade and public services, such as waste handling and recreational activities, also come out high in the Swedish study is probably due to the specific infrastructure of the Swedish economy. Finnveden et al. (2001) also rank the product groups according to emission intensity, and here we find transport by ship at the top of the list, similar to our results. Also construction materials, fish & seafood, metals and agricultural products are ranked high on impact intensity by both Finnveden et al. (2001) and our study (Chapter 1.2.5). When considering the ranking by CO2 and SO2, it is also not surprising to find electricity and heat among the important products (see our Tables 1.18 and 1.26). It is less expected that transport by air and road appear among the products with high impact intensity. However, air transport would be the next item to be included if the list in Table 1.18 had been expanded, and also freight transport by road is in the same order of magnitude. Swedish pulp and paper industry also appears to have a relatively high impact intensity. The corresponding Danish industry has a completely different product composition (more finished products), which explains its lower impact intensity. Tourist expenditures and car driving (private fuel use) does not appear in the Swedish ranking, since these product groups were not included in the Swedish data. Nijdam & Wilting (2003) use a number of environmental indicators, including global warming, acidification, nutrient enrichment and photochemical ozone. For global warming they find the most important consumption groups to be food (30%), followed by leisure (22%, mainly due to transport for holidays), and housing (17%; mainly for heating and electricity), confirming our main findings (see Chapter 1.2.2). The study applies the same general methodology as our study (IO-analysis) and their detailed reporting of consumption data and environmental impact intensities should therefore allow a more complete comparison between Dutch and Danish consumption, which is, however, not possible within the limitations of the current study. Nemry et al. (2002) and Labouze et al. (2003) find dwellings and transport to be the most important product areas, which confirms our finding in spite of a completely different methodological approach (bottom-up process analysis). This points to these two product areas as being of such size that they are likely to appear in any priority list, despite differences in methodology and data basis to derive these lists. Nemry et al. (2002) do not include food products in their ranking, while Labouze et al. find food products to be the largest source of eutrophication (due to fertilizer application) and a large source of global warming and photochemical oxidation (due to enteric fermentation and manure management). Nemry et al. (2002) furthermore point to packaging and electrical appliances as important products, while Labouze at al. (2003) find textiles among the largest sources of acidification and photochemical oxidation. This may be seen as corresponding to the importance assigned to wholesale trade (partly due to packaging use), electricity, and clothing in our study. Due to the differences in methodology, correspondence between the results would not be expected at a more detailed level. 1.7 Implications of the results for important product groups1.7.1 IntroductionIn this sub-chapter, we discuss the improvement options for the product groups identified in Chapters 1.2 and 1.3 as having high environmental impacts. This should not be seen as an exhaustive treatment of all current activities and improvement options, but rather as tentative indications on constructive ways to apply the results of the prioritisation. To be relevant for product-oriented environmental policy, a product group must have both high total impact and high impact intensity. Surprisingly, this is the case for most of the top-10 product groups in Chapters 1.2 and 1.3. Notable exceptions are “Education and research” which, as already mentioned, has a high level of aggregation that places it high in total impacts in spite of a low impact intensity (and thus with an inherently lower relevance for specific policy interventions) and tobacco products and fireworks that have high environmental impact intensity, but a low volume that make them less relevant for a policy intervention, although pointing to these two product groups being under-priced compared to their environmental externalities (which are not even completely covered by the impact categories applied in this study, which does not include such issues as noise and the health impacts from passive smoking). The top-10 product groups in Chapters 1.2 and 1.3 account for a surprisingly large share of the total environmental impacts from Danish production and consumption. In the supply perspective, ranked according to total impacts, the top-10 product groups (out of a total of 138) account for 45% of the total environmental impact from Danish production and consumption. In the consumption perspective, ranked according to total impacts, the top-10 products groups (out of a total of 98) account for 57% of the total environmental impact from Danish consumption, and 25% of the total impact from Danish production and consumption. This implies that the product-oriented environmental policy may reach large improvements by focussing on this rather small number of product groups. Generally, there are large improvement potentials for all the priority product groups, generally falling within the following categories:
In one of the more thorough, recent studies in cleaner technology options, Phylipsen et al. (2002) also concluded that there is still significant potential to reduce the environmental impact by “traditional” material technologies, such as more efficient material production, material-efficient product design and material recycling. 1.7.2 FoodBoth in absolute terms and in terms of impact intensity, food appears as the most important need group. Also from the production perspective, a major share of the total environmental impacts is related to food products. Both from the consumption and net production perspectives, pork meat is the most important food item. In the lifecycle of pork, pig farming is the most important process, with nature occupation (from fodder production) and nutrient enrichment (mainly from nitrogen compounds in the form of nitrate to water and ammonia to air) as the most important impact categories. To reduce the land (and thus nature) area appropriated for pig production, it is necessary to apply technologies for fodder production that demand less area, i.e. have a higher yield per land area. The main components of animal feed are protein and carbohydrates. Protein production today is largely based on soybeans and other legumes while grains are the most important sources of carbohydrates. Reductions in area use for both protein and carbohydrate crops can be achieved by selection of crops, crop varieties and growing conditions with high protein and/or carbohydrate yields per hectare. Given the large variation in these parameters, this appears the most straightforward improvement option. Options that utilise non-agricultural protein and carbohydrate sources may be relevant for specific purposes (such as fish fodder), and as a possible long-term relief. It is possible to produce protein independent of arable land, by microbial conversion of natural gas, a process normally occurring under deep-sea conditions. A Norwegian company, Norferm, has developed an industrial chain imitating this natural conversion process (www.norferm.no). The reaction also requires ammonia, oxygen and some minerals. The product has high protein content and a balanced amino acid composition, which makes it suitable in feed for fish, pigs and chicken, and in human food. Although carbohydrate production does not have similar opportunities to become completely independent of land use, it is possible to produce carbohydrate feeds from cellulose by a process involving low-temperature thermal treatment and enzymatic hydrolysis (a different process than the microbial degradation of cellulose in ruminants, but with the same result). Reduction or separation of lignin and tannin in the cellulose raw materials is necessary, since these substances reduce digestibility. With this process, straw and forest products can be used as raw materials for carbohydrate feeds, thus relieving some of the pressure on agricultural land. Currently, the cellulose-based process it too expensive, but research is ongoing to improve the process efficiency (Palonen et al. 2004). Another option in the same line of thought is the direct use of forestry by-products, such as leaf protein, as food and feed sources (Speedy & Pugliese 1991). With nitrogen often being the limiting factor for agricultural yields, there is a natural incentive to economize with this resource and thus to avoid emissions. Nevertheless, it has been pointed out that there are still significant improvement potentials and also many available technologies that are not yet fully implemented (Skov- og Naturstyrelsen 2003). Imported meat also appears as one of the most important food products in terms of environmental impacts, with the same impact categories being important as for domestic pig production, namely nature occupation and nutrient enrichment. This points to the same kind of improvement options, although it is obviously not possible to influence foreign producers of meat by direct incentives, leaving supplier requirements as the most relevant way to reduce these impacts. In Denmark, production of beef meat mainly takes place at the dairy farms, which also appear as important contributors to the total environmental impacts related to food products. Here as well, the most important impact categories are nature occupation and nutrient enrichment, thus pointing to the same kind of improvement options as for pig farms. Dairy farms are constrained by production quota, which implies that these farms can best be influenced through incentives directly targeted at the farms, or through labelling initiatives. Other important product groups within the need group “food” are:
For these product groups, it is still nature occupation and nutrient enrichment that are the most important impact categories, although both the energy related impacts and ecotoxicity (from pesticide use) are relatively more important for these products than for the pure meat products. Measures aimed at reducing food spillage throughout the value chain will obviously also reduce the demand for primary production. In the food industries, cleaner technologies for resource savings and waste minimisation traditionally focus on better housekeeping options e.g. more efficient use of raw materials and improved waste segregation. Especially efficient blood handling and recovery and minimised water use in eviscerating have been applied. Nevertheless, improvement potentials are still assessed as being high. Also in dairy industries, further possibilities for product recovery exist, while in brewing most options are assessed to be already implemented (EurEco et al. 1998). Car driving for shopping is also among the large contributors to the overall environmental impact from the need group “food”. Obvious improvement options are alternative distribution systems for groceries, e.g. direct delivery. For other improvement options for car driving, see Chapter 1.7.9. Imported vegetables and fruit, both fresh and processed, do not reach the top-10 lists of environmental impact, but are still among the more important items within the need group food. For these products, nature occupation plays a less important role, while nutrient enrichment is still important, and both the energy related impacts and ecotoxicity (from pesticide use) appear among the important impact categories. Several other agricultural products appear high on the lists of high environmental impact intensity (Chapters 1.2.5 and 1.4.2). For all of these products, the picture is the same as above: For meat products (including eggs), nutrient enrichment and nature occupation dominate, while for other food products, pesticides play a relatively more important role. In Denmark, pesticide use has been the topic of action plans aiming at reduced use. Potentials for further reduction in pesticide use are assessed in Christensen & Huusom (2003). 1.7.3 HousingAs a need group, housing appears as having the third largest environmental impact after food and leisure (see Figure 1.3 in Chapter 1.2.4). The main part of the impacts comes from the construction, repair and maintenance of the dwellings, except for the energy related impact categories where heating has a substantial contribution. The most direct way of reducing the environmental impacts from heating are savings in consumption, for which substantial potentials exist, both by improvements in construction and in user behaviour, as demonstrated by large variations in heat consumption per m² dwelling (Hans Bjerregård Rådgivning ApS 2001). Substitution of the heat source towards more renewable energy sources is also an obvious possibility. The actual construction, repair and maintenance processes also involve some energy use, but the main impacts relate to the materials and equipment used, notably wood products, basic non-ferrous metals, plastics, cement, bricks and tiles. A building is a very complex product, and improvement options will often require coordination between large numbers of actors. Realising this, buildings were one of the first areas where the Danish Environmental Protection Agency initiated a product panel (see www.byggepanel.dk). Both the panel and a recent status report (Øhlenschlæger 2003) point to the need for more knowledge dissemination, and stronger and more far-sighted regulatory incentives. Imported wood products and textiles receive a high ranking on ozone depletion (see Table 1.52 in Chapter 1.4.4), which is mainly due to fumigants (methyl bromide) and solvent use (methyl chloroform), respectively. These uses are generally being phased out as a result of the Montreal protocol. Basic non-ferrous metals are treated separately in Chapter 1.7.4. The most important emissions in relation to plastics are the VOC emissions occurring from liquid resin mixtures prior to their full polymerisation. Several VOC emission reduction options exist, both preventive (modifying process equipment and conditions) and end-of-pipe (combustion). The environmental impact from cement, bricks and tiles are mainly due to energy-related emissions of CO2 and SO2. Recycling is the most obvious option for reducing energy use for bricks and tiles. For cement, a low-energy alternative based on magnesite has been developed by an Australian company, Tec Eco, (www.tececo.com). Furniture and furnishing is also included in the need group “Housing.” The main environmental impacts of his product group are due to the components of imported wood, textiles and basic non-ferrous metals, all mentioned above. 1.7.4 Basic non-ferrous metalsThe basic non-ferrous metals industry, generally outside of Denmark, appears as having high impacts in the impact category “human toxicity.” This is mainly due to emissions of mercury to soil, mainly from primary extraction, smelting, casting and primary processing of aluminium, zinc, copper, gold and silver. Mercury has been the subject of an international study (UNEP 2002) and UNEP is presently initiating and implementing a mercury programme (www.chem.unep.ch/mercury/). The most obvious way of reducing the mercury releases from the basic metals industries is to increase the recycling of these metals, thus completely avoiding the primary processes. This would also reduce other emissions. The substitution of metals by e.g. new composites will have the same effect. The major non-industrial technique for gold extraction in South America (especially the Amazon), China, Southeast Asia and some African countries is mercury amalgamation, since it requires little start-up investment and very little technical know-how. Alternative extraction methods exist, but banning the use of mercury has not proven successful. Instead, UNEP (2002) recommends preventive measures such as educating the miners and their families about hazards, and putting in place facilities where the miners can take concentrated ores for the final refining process. We are not aware that anyone has considered product-oriented measures, such as a certification scheme allowing the processing technology to be traced, but the accounting systems applied in the processing of precious metals could be copied to the raw material extraction with major reductions in exposures and losses. 1.7.5 Transport by shipTransport by ship is an environmentally important part of the Danish production, both in absolute terms and in terms of impact intensity. The most important impact categories are ecotoxicity (mainly due to the antifouling agent tributyltinoxide (TBTO)) and acidification (due to emissions of SO2 and NOx), while the other impact categories, except nature occupation, are also of importance. The issue of TBTO is well recognized and in Denmark a ban was already enforced from 1991 for ships below 25 meters. The 2001 International convention on the control of harmful anti-fouling systems on ships (IMO 2001) requires that by January 2003, all ships shall not apply or re-apply organotin compounds in anti-fouling systems, and by January 2008, such compounds shall either be removed from the hulls or protected by a coating that prevents leaching. Although this convention does not come into force before a certain number of countries have ratified it, the EU has decided to put the convention into force already on their own territory (CEC 2002a). Alternatives to TBTO are currently being developed (for Danish research see Højenvang 2002, Madsen et al. 2003, Allermann et al. 2004) and the European paint manufacturers have created a web-site informing on TBT-free alternatives. The emissions of SO2 and NOx from marine fuels are larger than those from comparable amounts of fuels used in land transport. For SO2, the emissions from ships in European waters in 2010 will correspond to 75 % of the total emissions from the EU land-based sources (CEC 2002b). This is due to the lack of comparable regulation on marine fuels and engines. The European Commission has adopted a strategy to reduce emissions from seagoing ships (CEC 2002b). The main aim of the strategy is to reduce the impact of ship emissions on local air quality and acidification through the reduction of the sulphur contents of marine fuels used in the EU (2002c). For NOx and SO2, there are sufficient improvement options in the availability of cleaner fuels and advanced emission-control technologies already required and upcoming for heavy-duty diesel trucks and buses. In fact, engine manufacturers state “it is essential that engine emission standards be implemented to allow technologies to transfer in an orderly manner from on-highway applications, to nonroad applications, and then to marine applications” (French 2001). For all energy-related impact categories as well as ozone depletion, international requirements for pollution prevention are contained in Annex VI of the MARPOL convention (international convention for the prevention of pollution from ships). It contains requirements for NOx and VOC emissions, sulphur content of fuel, fuel oil quality, emissions of ozone depleting substances (although HCFCs will be allowed until 2002), ship incinerators and International Air Pollution Certificates. Annex VI is not yet in effect internationally, but the target date for implementation is 2004. In an extensive study on reduction options for greenhouse gas emissions, Skjølsvik et al. (2000) identified reduction of speed in general as the single measure that results in highest reduction of CO2 emissions. In general, improved fleet planning could yield more reduction than improved hull and propulsion designs, which also have a longer implementation time due to the long lifetime of ships. Emissions from ships in harbours have been investigated for Danish harbours by Oxbøl & Wismann (2003) and for the EU by Whall et al. (2002). The shares of total ship emissions are generally below 6%. Oxbøl & Wismann (2003) note that pumping of liquid goods cause 59% of the harbour emissions of SO2 and 36% of the harbour emissions of NOx, and recommend increasing the use of land-based power supply. The above considerations also apply to the operation of navy ships, classified under the public industry “Defence, justice, public security and foreign affairs,” which feeds into the public consumption group “General public services, public order and safety affairs.” Besides the production and maintenance of the transport equipment (ships and boats), refining of petroleum products is the most important upstream process feeding into transport by ship (and other forms of transport, see also Chapter 1.7.9). Variations in refinery emissions deduced from the data reported in Frischknecht (1996) point to significant improvement options. The degree of heat recovery and co-generation is especially important. The refinery BREF (EC 2001) list production techniques and has a special focus on abatement techniques for air emissions. 1.7.6 Wholesale tradeWholesale trade has a relatively large environmental impact, mainly due to a large consumption of transport and packaging, and to a lesser extent consumption of advertising and buildings. Improvements are thus mainly dependent on improvements in, and more efficient utilisation of, these upstream supplies, i.e. especially choice of transport and packaging solutions, and choice of suppliers with lower environmental impacts. Regarding transport, special attention should be given to reduce road transport, e.g. by shifting to rail transport, and to choose suppliers of ship transport with less environmental impact. Regarding packaging, special attention should be given to reduction in or substitution of metal and paper packaging. Both these material categories are also in focus of the EU directives on waste and packaging, and re-use and recycling targets are enforced. A recent study (Petersen and Jørgensen 2004) shows that collection of waste package at source e.g. factory, office or household, together with municipal waste, gives the best results. 1.7.7 ElectricityThe dominating environmental impacts from electricity production are global warming and acidification. Regulation of these environmental impacts needs to be directly targeted at the electricity producers, except where it is possible to create separate markets for environmentally preferable power, as e.g. Københavns Energi has done for photovoltaic-based electricity. There is already significant regulation of the electricity industry, including quotas on CO2, SO2 and NOx. Increased focus on CO2 emissions is likely to favour the competitiveness of wind power, which is perceived as an important part of a sustainable power supply (Eltra 2004, MVTU 2003). Photovoltaic cells are another development area with a promising environmental profile, although farther from being economically competitive (PA Energy 2004). 1.7.8 Industrial cooling equipmentThe main reason that industrial cooling equipment is singled out as a product group is that it is practically solely responsible for the contribution to ozone depletion from Danish activities, or for 29% of the total ozone depletion potential related to Danish production and consumption (the remaining 71% is due to foreign activities related to imports to Denmark). In our base year, 1999, the substances HCFC-141b and HCFC-22 were both used, while in 2002, the use of HCFC-22 has been discontinued, resulting in an approximate halving of the total consumption compared to the years before that (Poulsen 2004). We do not have any information to reveal which applications of industrial cooling equipment the HCFCs are actually used in, which means that it is distributed evenly per DKK over the buyers of industrial cooling equipment. For this reason, we are also unable to say whether it is correct that the Danish production of motor vehicle trailers appears as one of the more environmentally intensive product groups, which is for a large part due to the consumption of industrial cooling equipment in this industry. 1.7.9 AutomobilesBoth in absolute terms and in terms of impact intensity, car purchase and driving is an environmentally important part of the Danish private consumption. The emissions from fuel combustion during driving are dominating the energy-related impact categories, especially photochemical ozone formation due to the VOC emissions. The vehicle production itself also contributes significantly to the overall impacts, both due to its chemicals use (VOC emissions), energy use (especially electricity) and materials use (aluminium and steel). The most direct improvements option is to focus on reducing the need for car driving (Banister & Marshall 2000), through physical planning that reduce attractiveness of car driving (Paulley & Pedler 2000, Hartoft-Nielsen 2001a, 2001b, 2002, Christensen 2001), mobility management (Rambøll Nyvig A/S 2002, Grell & Kjerulf 1999, VTPI (2004); see also http://www.epommweb.org), and improving the distribution of retail goods. The emissions from car driving may be substantially reduced through improved fuel efficiency. The EU Commission has negotiated a voluntary agreement with the European, Japanese and Korean vehicle producers to achieve an average CO2 emission of 140 g/km by 2008 for all new cars sold in the EU, a reduction of 25% compared to 1995. A recent EU Directive (1999/94/EC) requires comprehensive labelling for all new cars providing information on carbon dioxide emissions and fuel economy. Weight reduction is one of the most practical ways to increase the fuel economy of vehicles, and is thus complementary to the efforts to reduce chemical, electricity and materials use in vehicle production (Fox & Cramer 1997). 1.7.10 LeisureLeisure is the second most important need group according to figure 1.3 (Chapter 1.2.4). A large share of the impact is due to car driving, both in Denmark and on holiday abroad. Leisure accounts for the largest share of car driving. Car driving has been dealt with separately in Chapter 1.7.9. As an important improvement option, mobility management could be applied by leisure service providers. For inspiration, it may be noted that the Swiss Federal Office for Spatial Development has initiated a leisure traffic programme (www.are.admin.ch/are/en/verkehr/freizeitverkehr/). Another large share of the impacts in the need group “leisure” is due to tourist expenditures by Danes travelling abroad. These expenditures relate to goods and services purchased while travelling abroad. Food, transport and accommodation account for the most important impacts. Significant coordinated improvements are only realistic for package tours that include one or more of these items. Current green tourism initiatives such as the Tour Operators Initiative (www.toinitiative.org) and the European VISIT initiative (www.yourvisit.info) focus mainly on the accommodation site and have no requirements in the food area and only minor focus on transport / mobility management. Separately from these, NETS, the Network for Soft Mobility in European Tourism (www.soft-mobility.com), focus specifically on the transport aspects. Thus, there is an apparent need for more coordinated efforts that integrate all environmental aspects of tourism, including the important role of food. Although less important in terms of overall size within the need area “leisure”, pet foods have high environmental impact intensity (see Chapter 1.2.5). In terms of environmental impacts and improvement options, pet foods are very similar to human foods (see Chapter 1.7.2). Also recreational items n.e.c. have a high impact intensity for ecotoxicity (mainly due to copper in lost fishing gear), ozone depletion and photochemical ozone (mainly from plastics production for Christmas decorations and similar items). A complete substitution of copper in fishing gear should be possible. Ozone depleting substances are generally being phased out due to the Montreal protocol. VOC emission reduction options for plastics were mentioned already under housing (Chapter 1.7.3). Fireworks also have high impact intensity, mainly due to copper and VOC emissions during use. The only realistic improvement option appears to be a reduction in consumption. 1.7.11 ClothingClothing is the fifth most important need group and has a relatively high impact intensity (see Chapter 1.2.4). The largest part of the impacts is caused by emissions from the foreign textile and clothing industries. The most important impacts are ozone depletion and photochemical ozone, mainly due to solvent use and the production of synthetic fibres. The ozone depleting methyl chloroform is being phased out as a result of the Montreal protocol. Improvement options have been extensively dealt with by Smith (1994), Eastern Research Group (1996), EurEco et al. (1998), and Laursen et al. (1997). A major obstacle appears to be the highly fragmented nature of the industry, which makes incentives for life cycle thinking hard to implement. Better information exchange systems appear to be an important part of any improvement programme. Detergents have high impact intensity (see Chapter 1.2.5), which is again mainly caused by solvent use and VOC's in surfactants. Several VOC emission reduction options exist, both preventive (substitution, modifying process equipment and conditions) and end-of-pipe (combustion). 1.7.12 HygieneHygiene is the sixth most important need group and has relatively high impact intensity (see Chapter 1.2.4). The most important contributions come from detergents and other chemical products (see Chapter 1.7.11), and energy use for hot water and sewage treatment (removal and disposal). Energy use for sewage treatment is also the main reason for toilet flush having high impact intensity (see Chapter 1.2.5), although the toilet paper and the nitrogen and phosphorous content in the flush is also of importance. For energy use there is large variation in impacts depending on the heat source, and thus much room for improvement through substitution of heat sources, especially the use of photovoltaic energy. For toilet paper, criteria for environmental labelling exist. 1.7.13 Education and researchEducation and research only reaches the top-10 of environmental impact in Danish consumption because it is a very aggregated product group. In itself, education has very low environmental impact intensity (see chapter 1.4.2) and would not have reached the top 10 if it had been divided into primary, secondary and higher education, and adult education etc. Nevertheless, since it is a relatively homogeneous product group, it may still be relevant to look for improvement potentials that may be implemented with small effort for the entire product group. However, a closer analysis reveals that the impacts come from many different sources, with no source particularly dominating, although buildings, heating, electricity and paper use are the largest contributors. Thus, improvements would require implementation of rather extensive environmental management systems, which may be out of proportion to the importance of the area. However, implementing such systems particularly in this field will have an additional educational effect, which may make it worthwhile in itself. Footnotes [1] Net production of Danish industries is the products supplied by Danish industry for domestic final consumption or for export, as opposed to the gross production that includes also the products supplied for internal use in Danish industry. [2] Products for re-export are not included in any of the three perspectives.
|
