Keywords

1 Introduction

This chapter deals with the category ‘abiotic resource use’. Resources may be defined as those elements that are extractable for human use and that have a functional value for society (Udo de Haes et al. 1999). This general definition will be considered in this chapter. Resources may be classified according to different typologies:

  • Renewable and non-renewable: According to the EPA (2014), renewable resources are natural resources that can be replenished at approximately the same rate at which they are used (e.g. wind and solar energy). On the other hand, nonrenewable resources cannot be produced (or re-grown) at the same rate at which they are consumed (e.g., coal and natural gas).

  • Biotic and abiotic: Biotic resources are materials derived from presently living organisms. In addition to the resource value, they typically have an important role in maintaining ecosystem services and also intrinsic value (examples are tropical hardwood and ivory), while abiotic resources are the product of past biological processes (coal, oil and gas) or physical/chemical processes (deposits of metal ores) (Müller-Wenk 1998; Guinée 1995).

  • Funds, flows and stocks: In the case of stocks extraction inevitably leads to the depletion of the resource, i.e. reduction of the available amounts in nature, whereas funds may be depleted but also have a renewal rate which is high enough to allow the resource to recover. Usually biotic resources are categorized as funds, but also groundwater can be considered as fund resource. Flow resources though cannot be depleted. Their availability per unit time however is limited, and thus their extraction is marked by competition (e.g. wind energy) (Heijungs et al. 1997; Lindeijer et al. 2002).

In contrast to many other impact categories discussed in the other chapters of this volume on life cycle impact assessment, ‘abiotic resource use’ is actually not an environmental impact category that is directly related to the Area of Protections (AoPs) ‘natural environment’ and ‘human health’, as climate change or human toxicity. The title of this chapter suggests that it deals with the environmental concerns due to the use of abiotic resources. The use of resources may cause environmental impacts to several AoPs, as shown in Fig. 13.1 in Sect. 2, and therefore can be considered an important issue when performing a Life Cycle Assessment (LCA). Some of these environmental impacts at the AoP Human Health and Natural Environment (ISO 2006) were already considered in previous chapters, and this discussion will not be repeated here. However, there are still damages to the AoP Resources to be discussed. According to Jolliet et al. (2003b), damages to this AoP consist in the reduced availability of the corresponding type of resource in the future, which is mainly known as ‘resource depletion’ (or ‘resource scarcity’). On the other hand, we can also find several methods that are able to give answers to questions of environmental sustainability, by considering the abiotic resource use in a life-cycle perspective (Stewart and Weidema 2005). These methods do not evaluate ‘resource depletion’, but they are able to consistently account for the resource use of a product.

Fig. 13.1
figure 1

Simplified impact pathway for the category abiotic resource use

Full size image

2 Impact Pathways

Figure 13.1 shows the impact pathways for the category ‘abiotic resource use’. At the left side there are types of resources, grouped as biomass, water, land, flow energy resources (e.g. wind and hydropower from dammed water), atmospheric resources (e.g. argon), metals and minerals, nuclear energy, and fossil energy. The first step in the impact pathway can be evaluated by the resource accounting method (RAM), which can affect several AoP (category #1). The next step in the impact pathway is a group of approaches that evaluate the scarcity of resources at a midpoint level (category #2), and affect solely the AoP Resources. Finally, the last step in the impact pathway is a group of approaches that evaluate the scarcity of resources at endpoint level (category #3), also affecting solely the AoP Resources. For reasons of completeness Fig. 13.1 also includes biomass (i.e. biotic resources), water and land (the two latter are dealt in dedicated chapters of this book, see the Contents page).

3 Scale, Spatial Variability, Temporal Variability

The scale of the impacts on the AoP Resources depends somewhat on the resource in question. For metals, nuclear and fossil energy carriers and atmospheric resources impacts can be considered to be on the global scale. Metals, fossil energy carrier and nuclear energy carrier are typically traded globally (except natural gas, which can be traded regionally as well) and atmospheric resources are in essence distributed equally over the planet (Hoekstra and Hung 2002). Some low value minerals like gravel are usually supplied from local sources. For this type of resources impacts can differ based on location. For flow resources location can be relevant, i.e., they are not available everywhere to the same extent and the resources cannot be moved as such. The energy content needs to be converted into a suitable form first, before it can be traded.

Short term temporal variability (diurnal, seasonal) can be neglected when determining impacts on the AoP Resources. Longer term temporal variability, however, can be deemed an important source of uncertainty for the life cycle impact assessment (LCIA) approaches discussed below. As humans influence the availability of resources, e.g. by their consumption pattern, but also by discovering new resources, and developing new technologies for resource extraction, e.g. fracking, impacts on the AoP Resources can be considered to be changing with time.

4 Methods for Abiotic Resource Use

LCIA methods for ‘abiotic resource use’ can be classified in different categories, considering some common characteristics. Finnveden (1996), Lindeijer et al. (2002) and Steen (2006) classified the approaches in four categories: (1) Approaches based on energy or mass; (2) Approaches based on ratio of use to deposits; (3) Approaches based on future consequences of current resource extractions; and (4) Approaches based on exergy consumption or entropy production. The ILCD Handbook (European Commission 2011) classified the methods for abiotic resource use in four other categories: (1) Category 1 includes methods that use an inherent property of the material as basis for the characterisation; (2) Category 2 includes methods that address the scarcity of resource; (3) Category 3 includes methods focused on water depletion; and (4) Category 4 includes methods that evaluate the depletion of resources at an endpoint level.

The available amount of material can be evaluated using different definitions, e.g. ‘(economic) reserve’ and ‘reserve base’ for metals and minerals from the United States Geological Survey (2010). The former are the resources that can currently be economically extracted and the latter includes also additional resources which meet certain criteria that are relevant for mining and production practice (e.g. depth of deposits). Another important definition for resource depletion used in LCIA methods is the ‘ore grade’ of a substance, which is the concentration of the targeted substance in the ore, and is typically represented in mass percentages.

4.1 Resource Accounting Methods (RAMs)

From the impact pathway in Fig. 13.1, we can see that the ‘abiotic resource use’ can be analysed through RAMs before reaching the environmental damages at specific areas of protection or even the environmental impacts at specific midpoint categories. These methods are far from giving a direct quantitative value for environmental damages, but they are still able to provide results on the environmental sustainability of a product due to the philosophy of ‘less is better’.

These RAMs generally sum up all the resources consumed/used in the life cycle of a product. In order to provide results in single indicators, the resources are usually represented in common units (e.g. energy), otherwise the same information as given by the Life Cycle Inventory (LCI) would be obtained.

4.1.1 Mass

In Material Flow Analysis (MFA) resources are also typically aggregated based on mass. There are different MFA approaches. One of them is the Material Intensity Per Unit Service (MIPS) method (Ritthoff et al. 2002; Spangenberg et al. 1999). Though it is not usually classified as an LCIA method, it is similar to LCA in that it models at the system level (Finnveden and Moberg 2005). The MIPS method, pioneered by Schmidt-Bleek (Schmidt-Bleek et al. 1998), distinguishes between five resource categories: abiotic raw materials, biotic raw materials, movement of soil (agriculture and forestry, incl. soil erosion), water and air. These categories can be further divided into subcategories. A general guide to MFA was published by the OECD (2008a, b, c).

4.1.2 Energy

Accounting for energy use is a concept that was introduced in the 1970s (Boustead and Hancock 1979; Pimentel et al. 1973), and standardised by VDI (1997). Energy-based RAMs account for the energy extracted from the natural environment (i.e. the cradle) to support the technosphere system. They account not only for energy flows but also for material flows, by quantifying their energy content.

These methods have been operationalised for LCA, for instance as the Cumulative Energy Demand (CED) for the ecoinvent database (ecoinvent 2010; Hischier et al. 2009) and as the Primary Energy Demand (PED) for the GaBi database (PE International 2012). In principle, CED and PED are the same, only differing in names and compatibility to databases and/or software. The results are generated in a unit easily comprehended by stakeholders (e.g. MJ). However, since some materials have low energy value (e.g. water, metals and minerals), energy-based RAMs do not have a desirable completeness for abiotic resource use accounting. Also, biotic resources are accounted through their energy content and, to avoid double-counting, the use of land is neglected. This methodological approach results in two weaknesses: (1) agricultural systems with higher yields do not show better results although they have lower land use; (2) because these methods account for the energy content of the biomass harvested, the system boundaries may be inconsistent with the difference between natural environment and technosphere. The biomass harvested is basically an output from agricultural production, rather than a natural resource that should be accounted for. In this sense, the biomass harvested may be considered as still at the supply chain level (see Liao et al. (2012a)). In spite of these limitations, energy-based RAMs are applicable for LCA, but they should be complemented by methods accounting for the use of other resources (e.g. water, metals and minerals), for reasons of completeness.

Fossil energy consumption (one category of the CED and the PED) can be a useful screening tool (Huijbregts et al. 2010; Huijbregts et al. 2006) and is able to provide consistent results when LCA studies are interested in information solely regarding the consumption of fossil fuels during the product’s life cycle. It is also common to find energy-based RAMs in some other traditional midpoint LCIA methods, i.e., the energy content of fossil fuels is used as characterisation factor (CF). For instance in the category ‘Fossil depletion’ of the method ReCiPe Midpoint, the mass of oil equivalent (kg oil eq.) is used as unit; this type of method will be discussed further later on.

4.1.3 Exergy

By definition, the exergy of a resource or a system is the maximum amount of useful work that can be obtained from it (Dewulf et al. 2008). Exergy analysis is usually used in industry to assess the efficiencies of processes. For natural resources that are exploited to convert their energy content into work (or heat) the idea that amounts of those energy carriers can be expressed in exergy terms is quite straightforward. Even though most metal and mineral resources are not extracted from nature with the aim to directly produce work from them, they still contain exergy. This is because these resources differ from the reference environment with respect to their chemical composition and their concentration. For example, the copper in a copper deposit is much more concentrated and occurs in another chemical form (e.g. CuFeS2) than copper dissolved in seawater (the reference species for copper). These differences can in principle be used to produce work.

The cumulative exergy consumption (CExC), introduced by Szargut et al. (1988), is the exergy of the overall natural resources consumed in the life cycle of a product. Exergy-based RAMs have been operationalised to LCA through different LCI modelling approaches. For the process-based ecoinvent database, the Cumulative Exergy Demand (CExD) is operationalised in Bösch et al. (2007) and the Cumulative Exergy Extraction from the Natural Environment (CEENE) is operationalised in Dewulf et al. (2007). The two operational methods have some differences, including the approach to account for metals and minerals, but also the approach to account for biotic resources: While the exergy of the biomass is accounted in the CExD, in the CEENE the exergy deprived from nature due to land use is accounted. For the economic input-output U.S. 1997 database, the Industrial Cumulative Exergy Consumption (ICEC) is operationalised in Zhang et al. (2010).

The CExD and the CEENE are able to account for several resources, even though the results are generated through units not easily comprehended (e.g. MJex). The system boundary of the CExD is similar to the CED, leading to similar limitations, i.e., land use is neglected and its system boundary does not correspond to the interface between natural environment and technosphere for biomass. The CEENE method accounts for land in exergy terms through the accounting for the quantity of photosynthetic solar exergy deprived from nature due to land use. This procedure allows accounting not only for land use for biotic resources, but also for other purposes (e.g. built-up land). However, by choosing to account for land, biotic resources from natural systems which had no land occupation during its production (e.g. wood from natural forests) are not accounted through CEENE. Nevertheless, the CEENE method was recommended as the most appropriate thermodynamic indicator for resource use accounting, in Liao et al. (2012b). Trying to tackle some limitations of CEENE and CExD, Alvarenga et al. (2013) proposed a new approach to account for land resources (i.e., land and biotic resources) by classifying the system studied as natural or human-made beforehand. On top of that, they provided spatially differentiated CFs for land occupation, through the potential natural net primary production.

4.1.4 Emergy and Similar Methods

Introduced by Odum (1996), emergy accounts for the total available energy that has been used to make a product (including human labor). It is sometimes referred to as energy memory, since it is supposed to be a record of the previously used-up available energy to make a product. In contrast to other RAMs (e.g. exergy-based), which usually set the natural environment as ‘cradle’, emergy has a different system boundary. The natural environment is part of the system, and the ‘cradle’ is considered to be the energy forces outside of the geobiosphere, e.g. the sun (Liao et al. 2012a) (Fig. 13.2). Emergy considers tidal, geothermal and solar energies as main energy sources that rule life on Earth, and the latter is taken as reference for its unit (Joules of solar energy – Jse). Emergy has received several criticisms from the scientific community, as combining disparate time scales, allocation problems, and the lack of uncertainty quantification on the numbers used to calculate transformities (Hau and Bakshi 2004a, b). Some propositions to overcome them, together with challenges to implement emergy into LCA, have been suggested in literature (Rugani and Benetto 2012; Ingwersen 2011).

Fig. 13.2
figure 2

Simplified scheme representing different system boundaries considered in RAM

Full size image

Due to the limited acceptance in the scientific community, Hau and Bakshi (2004a) developed the ecological cumulative exergy consumption (ECEC). According to the authors, it overcomes some weaknesses of emergy, and if identical system boundaries, allocation and quantification methods are used, emergy and ECEC should produce equivalent results. The ECEC has been operationalised to LCA for the economic input-output U.S. 1997 database in Zhang et al. (2010). It accounts for several ecosystem services as well and is commonly used in complementation to the industrial ICEC (Zhang et al. 2010; Urban and Bakshi 2009; Baral and Bakshi 2010; Baral et al. 2012). However, it is questioned if certain emergy-based RAMs, as the ECEC, may be applicable to ‘abiotic resource use’ in LCA, since some regulating and supporting services considered in this methodology (e.g. air quality regulation) appear not to be resources according to the definition from Udo de Haes et al. (1999).

Using emergy as starting point, the Solar Energy Demand (SED) accounts for the amount of solar energy needed to produce a certain product. It has been operationalised to the ecoinvent database (Rugani et al. 2011), and according to these authors, it shares the same conceptual rationale as emergy, although it does not use the same approach for allocation. Unlike emergy, the SED does not account for human labor and most of the ecosystem services: it accounts for provisioning services only.

4.1.5 Ecological Footprint

Developed by Wackernagel and Rees (1996) and further enhanced by the Global Footprint Network (2009) and Ewing et al. (2010), the Ecological Footprint is defined as the ecological surface area needed to sustain a certain system. When applied to products, the requirement of area to produce the raw materials and to absorb CO2 emissions is calculated (m2). It has been operationalised to LCA through the ecoinvent database (Huijbregts et al. 2008), where nuclear energy is also (indirectly) accounted for. In this methodology, solely land use, nuclear energy, and fossil energy (indirectly through the fossil CO2 emissions) are accounted. Water, metals, minerals and other resources are not accounted; therefore the method does not provide satisfactory completeness for ‘abiotic resource use’. Nevertheless, it has a strong appeal to society, since it can directly be compared with the Earth’s carrying capacity, and as a consequence has a strong communication ability for dialogue with stakeholders (by its immediately understandable unit).

The Ecological Footprint is able to provide other information (e.g. the ecological deficit of nations) than solely resource accounting (see Global Footprint Network – www.footprintnetwork.org), but since this chapter is focused on ‘abiotic resource use’ in LCA, only its potential use as RAM is discussed.

4.1.6 Ecological Scarcity

Dating back to the 1990s, the ecological scarcity method (Ahbe et al. 1990; Brand et al. 1998) is a distance-to-target methodology developed based on the Swiss context and encompassing both resource use and emission-related environmental impacts. Its most recent implementation is described in Frischknecht et al. (2009). The resource types treated in the method and relevant for this chapter are flow energy resources, metals and minerals, fossil energy and nuclear energy.

The so-called ecofactors are derived on the basis of political targets (critical flow) and actual resource flows (current flow, normalisation flow) and expressed in units of eco-points. The targets are set for the year 2030. For the resources discussed in this chapter the derivation of the ecofactors does not involve an environmental impact pathway. The factors published in Frischknecht et al. (2009) are specific for Switzerland, but in principle the methodology can also be adopted to other regions, e.g. Baumann and Rydberg (1994) and Miyazaki (2006). Rather than a characterisation method, the ecological scarcity method can be seen as incorporating both normalisation and weighting resulting in Ecofactors, which are to be multiplied with the appropriate inventory flow to arrive at final scores which can be aggregated. Ecofactors are calculated according to (Eq. 13.1):

$$ \mathrm{Ecofactor}=K\times \frac{1\times EP}{F_n}\times {\left(\frac{F}{F_k}\right)}^2\times c $$
(13.1)

With

  • K the CF (kg · kg−1 or MJ-equivalent/MJ), EP the unit eco-point, F the current flow (kg · year−1 or MJ · year−1), F n the normalisation flow (kg · year−1 or MJ-equivalent · year−1), F k the critical flow (kg · year−1 or MJ · year−1) and c the constant (1012 · year−1). The resulting unit is then EP · MJ−1 or EP · kg−1.

For energy resources flows are accounted in units of MJ. The only distinction made is between renewable and nonrenewable energy resources. Whereas the CF for nonrenewable energy is one MJ-equivalent/MJ, it is one third MJ-equivalent/MJ for renewable energy resources, due to political targets. For energy resources no environmental mechanism is involved.

The only other abiotic resource covered by the ecological scarcity method and relevant for this chapter is natural gravel. Due to its low market value on a mass basis resulting in low economic viability of extensive transport, gravel availability can be considered as a regional issue. In the calculations the critical flow is assumed to be equal to the current flow, because the life time of economic gravel reserves in Switzerland has been stable for quite some time, though eventually the resource is considered to be finite.

The methodology is interesting where the LCA target is an assessment with respect to country specific policy targets. As policy targets are the benchmark the environmental relevance of the method is limited to the relevance of these policy targets.

4.2 Resource Depletion at Midpoint Level

As illustrated in Fig. 13.1, the second category of resource impact assessment methods evaluates the scarcity of resources, but still at a midpoint level, i.e., they do not evaluate the actual damages to the AoP Resources.

4.2.1 EDIP 97 and EDIP 2003

The EDIP methodology has its baseline literature in Hauschild and Wenzel (1998). The approach for ‘abiotic resource use’ in EDIP 2003 and EDIP 97 is the same, except that the values used in the calculations in EDIP 2003 are updated for the year 2004, while in EDIP 97 the data is from 1990 to 1991. Based on the economic reserves, the ‘abiotic resource use’ is evaluated by the scarcity of resources naturally available, which means that even though metals may not disappear after their use (unlike fossil energy), they will no longer be available in their natural deposits, but in other places (e.g. landfill).

To calculate the CF, the authors divided the procedure in two steps. In the first, called normalisation, the global production of a substance (i) for a specific year (2004 in EDIP 2003) is considered, and this value is divided by the world population from that year. In the second step, called weighting, the economic reserve of the substance (i) is divided by the global production from the same substance (i) in a particular year (2004 in EDIP 2003), providing the supply horizon of the substance, in years. Finally, the CF for the substance (i) is calculated by the reciprocal of the product between the normalisation and the weighting factors (Eq. 13.2). Taking a look at the equation, we can notice that the ‘global production’ can be erased from the equation, and effectively the CF are based solely on the economic reserves, but normalized for the World population from the year 2004, expressing the resource use in so-called person-reserves – the available economic reserve per person in the world in 2004. We can conclude that this approach goes in accordance with ‘option 2a’ from Lindeijer et al. (2002).

$$ {\mathrm{CF}}_{\mathrm{i}}=\left[\frac{1}{\left(\frac{\mathrm{Global}\ \mathrm{prod}{.}_{\mathrm{i}.2004}}{\mathrm{World}\ \mathrm{pop}{.}_{2004}}\right)\times \left(\frac{{\mathrm{Economic}\ \mathrm{reserves}}_{\mathrm{i}}}{\mathrm{Global}\ \mathrm{prod}{.}_{\mathrm{i}.2004}}\right)}\right]=\left[\frac{1}{\left(\frac{{\mathrm{Economic}\ \mathrm{reserves}}_{\mathrm{i}}}{\mathrm{World}\ \mathrm{pop}{.}_{2004}}\right)}\right] $$
(13.2)

This LCIA method calculated CFs for several nonrenewable resources, including fossil energy, nuclear energy (uranium), metals and some minerals. In Hauschild and Wenzel (1998) the method is also elaborated for renewable resources for which a distinction is made according to whether the annual extraction exceeds the regeneration capacity. If this is not the case, the CF is zero (the resource use is sustainable) but if it is the case, the CF is calculated using Eq. 13.2, replacing the annual production by the annual exceedance of the regeneration capacity. CFs for water and wood on a global scale are presented, but they are often not considered due to the low significance of non-spatially differentiated CFs for these renewable resources.

4.2.2 Abiotic Depletion Potential

Guinée (1995) developed the abiotic depletion potential (ADP) as an approach applicable for ‘abiotic resource use’. Later on the approach was implemented in the CML (Institute of Environmental Sciences Leiden) LCIA method by Guinée et al. (2002) and further updated by van Oers et al. (2002). These updates were implemented in the CML method in 2009 and 2010. The latest implementation of the CML method can be found on the CML website (http://www.cml.leiden.edu/software/data-cmlia.html). The CML method implements the ADP for metals and minerals, fossil energy, atmospheric resources and, since version 4.1, also for nuclear energy. In van Oers et al. (2002) CFs can also be found for some mineral configurations (e.g. fluorspar and bauxite). Due to relatively good data availability (for the reference years) a large number of substances are covered.

From a conceptual point of view the approach is similar to the approach used for resources in the EDIP methodology, as it is based on use-to-resource ratios. In contrast to the EDIP methodology the amount of remaining resource is squared in order to take into account that extracting 1 kg from a larger resource is not equivalent to extracting 1 kg from a small resource, even if the use-to-resource ratio is the same. In this sense, we can conclude that this approach goes in accordance with ‘option 2c’ from Lindeijer et al. (2002). Equation 13.3 represents the generic calculation of the ADP of substance i expressed in kg of reference substance (Guinée et al. 2002):

$$ AD{P}_i=\frac{D{R}_i}{R_i^2}\times \frac{R_{ref}^2}{D{R}_{ref}} $$
(13.3)

With ADPi the abiotic depletion potential of substance i (kg reference substance · kg−1), Ri the ultimate reserve of substance i (kg), DRi the extraction rate of substance i (kg · year−1), Rref the ultimate reserve of the reference substance (kg reference substance) and DRi the extraction rate of the reference substance (kg reference substance · year−1).

The ultimate reserve is the total amount (mass for elements and minerals, energy for fossil fuels) of the substance available on Earth, be it in the Earth’s crust, in the oceans or the atmosphere. The reference substance is antimony for elemental species, and originally also for fossil energy. By now fossil energy is treated separately from elemental species in the CML implementation, with the total of fossil energy as reference. While Guinée (1995) had calculated ADPs for fossil energy based on their individual annual production and ultimate reserves, full substitutability based on energy content was assumed in Guinée et al. (2002), i.e. one MJ of oil is equal to one MJ of coal. As a consequence of these changes the ADPs of fossil energy are equal to their lower heating values when applied to elementary flows (ISO 2006) in units of kg, respectively m3 for natural gas. Therefore for fossil energy the current CML implementation of the ADP is comparable to methods accounting only for the energy content of resources, as in the CED method discussed in Sect. 4.1.2. In the CML methodology the ADP is used as CF, while normalisation is done per impact category.

The ADP approach was criticised by Müller-Wenk (1998), because the amount of ultimate reserves could satisfy human consumption for ‘millions of years’, which would imply that there is no scarcity issue. Moreover, the approach lacks consideration of quality aspects of the resource. Already earlier Guinée (1995) had remarked that what was relevant were the reserves which can eventually be extracted, called ‘ultimately extractable reserves’, likely to be very different from the ultimate reserves. Guinée (1995) implicitly assumed ‘the ratio between the ultimately extractable and ultimate reserve to be equal for all resource types.’ With the updates to the ADP approach, alternative CFs are available which use economic reserves or the reserve base (United States Geological Survey 2010) in the reference year 1999 instead of ultimate reserves. For resource depletion the International Reference Life Cycle Data System (ILCD) (European Commission 2011) recommends the use of the CML method at midpoint, in particular the alternative CFs using reserve base. In addition, it is advised to perform a sensitivity analysis using economic reserves and ultimate reserves.

4.3 Resource Depletion at Endpoint Level

The approaches that evaluate the damages to the AoP Resources, for instance extra economic costs, are considered to be resource depletion methods at endpoint level (see Fig. 13.1), and are discussed below.

4.3.1 Eco-Indicator 99

The approach for ‘abiotic resource use’ in Eco-indicator 99 (EI99) (Goedkoop and Spriensma 2000) focused on the evaluation of the depletion of resources (minerals and fossil energy), which is based mainly on Müller-Wenk (1998). The decrease in resource concentration due to extraction is modelled and evaluated by the concept of surplus energy, i.e. the difference between the energy needed to extract a resource now and at some point in the future (‘option 3’ from the set of depletion approaches from Lindeijer et al. (2002)). The depletion of nuclear energy is not assessed in this methodology.

For metals and minerals, the authors considered geostatistical models in order to evaluate the relationship between availability and quality. As an approximation at higher ore grades, it was assumed that the logarithm of cumulative amount of minerals mined increases linearly with the decline in the logarithm of the ore grade. The slope of the curves between the cumulative amount of minerals mined and ore grade are of key interest. In Fig. 13.3 we can see that a metal with a high slope (e.g. metal A) has a small change in ore grade when a certain amount is mined, while another metal with a lower slope (e.g. metal B) has a higher change in ore grade when this same amount is mined. Since the authors considered that the energy requirement needed to extract, grind and purify an ore goes up as the grade goes down, the CFs were calculated by considering the slope. The calculation of the additional energy requirement is based on energy requirements per kg of ore treated that are fixed for each metal. In the EI99 methodology report (Goedkoop and Spriensma 2000) they argue that in line with the modelling of other impacts technological development, which might lead to efficiency increases, is also not considered for the surplus energy method. This argumentation might be debatable as technology and grade are not fully independent, because mining has to be economic.

Fig. 13.3
figure 3

Slope of the availability against the grade (Based on Chapman and Roberts (1983))

Full size image

For fossil energy, the resource analysis had to be done differently from metals and minerals, mainly for two reasons: (1) The geological processes involved in the generation of fossils in nature are quite different from the processes that have caused the lognormal distribution of mineral ore grades in the Earth’s crust; and (2) the effort to exploit a resource does not increase gradually when the resource is extracted (as for minerals), but rather more abruptly when the marginal production of fossil fuels changes from conventional to unconventional sources (e.g. natural gas from tight reservoirs). In this sense, the surplus energy (i.e., the CF) is calculated by the difference between energy requirements for current and future sources.

In the EI99 methodology, different cultural perspectives are used according to the cultural theory of risk (Individualist, Hierarchist and Egalitarian), and they play an important role in the resource analysis for fossil fuels (while for minerals the evaluation is the same independent of the cultural perspective). For the Individualist perspective, fossil fuels are considered not to be a problem, and since the long term perspective is not relevant with this cultural perspective, fossil fuels would not switch to unconventional sources, leading to null values of surplus energy. In the Hierarchist perspective, it is assumed that a certain fossil fuel (e.g. oil) would not be substituted by another fossil fuel (e.g. coal), but by unconventional sources of the same fossil fuel (e.g. oil shale). Finally, in the Egalitarian perspective, substitution between fossil fuels is assumed, for instance coal-shale mix is considered to substitute conventional natural gas in the future.

4.3.2 Impact 2002+

The LCIA methodology Impact 2002+ (Jolliet et al. 2003a) considers three types of resources in their approach for ‘abiotic resource use’: Metals and minerals, fossil energy and nuclear energy. In the former, the authors use the same approach as the methodology EI99 (Goedkoop and Spriensma 2000), i.e., the surplus energy concept. However, they assume an infinite time horizon for fossil and nuclear energy instead of assessing the surplus energy required for future unconventional technologies (as in EI99). This implies that the total energy content of the fossil energy and nuclear energy are lost due to their consumption. Therefore, CFs for fossil energy and nuclear energy are based on their energy content, making the approach to be classified as an energy-based RAM.

4.3.3 ReCiPe

ReCiPe is a Dutch LCIA method created in 2008, which combines the scientific efforts of several institutes. The main information can be found in their report (Goedkoop et al. 2009) and the updated CFs can be found in http://www.lcia-recipe.net/. This method provides indicators at two levels: midpoint and endpoint. The midpoint indicators for ‘abiotic resource use’ from ReCiPe will be discussed together with the endpoint indicators in this subsection.

The approach of ReCiPe for metals and minerals uses focuses on the depletion of deposits, instead of individual commodities. According to the authors, it allowed them to take into account the actual geological distribution of metals and to cover more commodities, especially those that are always mined as co-products (but the CFs are provided for the elements, e.g. ‘copper, in ground’). ReCiPe provides midpoint and endpoint indicators for minerals. The damage to the AoP Resources due to the extraction of a certain mineral (resource depletion at endpoint level) is evaluated by the additional costs society has to pay due to this extraction, and it is expressed in US dollars, $ (present value in 2000). The CFs for the midpoint indicators are calculated by an equation similar to the equation from the endpoint indicator (except by the exclusion of some factors) and then normalized for the value obtained for iron, with iron equivalents as unit. Cultural perspectives are not considered for minerals, therefore the CF are the same for the Individualist, Egalitarian and Hierarchist versions. Nuclear energy (uranium) is considered together with other metals; therefore, the indicator ‘Metal depletion’ evaluates the depletion of metals and minerals and nuclear energy.

The approach for evaluating damage to AoP Resources from fossil energy use is also based on the marginal cost increase. However, as in the method EI99, the marginal increase is not related to the decrease of the grade of the metal, but to the shift from conventional to unconventional sources (Fig. 13.4). The method also provides CFs for midpoint and endpoint indicators. However, the characterisation model for the midpoint indicator is basically energy-based RAM, since the CFs are based on the energy content of the fossil fuels. The authors divided the fuel energy content by the energy content of a specific type of crude oil (42 MJ/kg), generating values in mass of oil equivalents (kg oil-eq.). For the endpoint indicator, the authors calculated the marginal cost increase for oil when changing to unconventional sources (e.g. oil from tar/bituminous sands). Different cultural perspectives were considered, and the marginal cost increase for oil was considered to be lower in the Individualist perspective, making the CF of the Hierarchist and Egalitarian perspectives (which were the same) to be approximately three times higher. After the CF for oil was calculated for the three cultural perspectives, the CF of fossil fueli (e.g. natural gas) was calculated by multiplying the CF of oil by the ratio between the energy content of fossil fuel i and the energy content of oil (42 MJ/kg).

Fig. 13.4
figure 4

Oil production costs for various resource categories (Resources to Reserves 2013© OECD/IEA (2013), fig. 8.3, p. 228)

Full size image

4.3.4 Sustainable Process Approach (EPS2000)

The sustainable process approach is an endpoint method implemented in the Environmental Priority Strategies for product development (EPS2000) (Steen 1999a, b). Information on this method is also available on the website of the CPM (Swedish Life Cycle Center) database (http://cpmdatabase.cpm.chalmers.se/). In the EPS2000 methodology the approach is implemented for metals and minerals, fossil energy, nuclear energy and atmospheric resources. Alternative factors for some metals are available in Steen and Borg (2002).

As for the other impact categories in EPS2000, the idea is to quantify the willingness to pay (WTP) for restoring damage done to the safe-guard subject. In the case of abiotic stock resources, not only the present generation but also future generations are included; therefore the WTP is calculated based on hypothetical sustainable processes which could produce resources like those extracted today once these are depleted. Thus, the method can be classified as one based on future consequences of current abiotic resource extraction (Steen 2006; Lindeijer et al. 2002). The calculated costs include direct production costs and external costs due to emissions and resource use.

The sustainable processes are assumed to produce resources from average rock (most elements and gravel), from seawater or air. Fossil fuel alternatives are vegetable oil, charcoal and biogas, respectively. In this context it should be noted that it is acknowledged that the biomass products cannot fully substitute fossil use for energy production. The sustainable processes are further optimised, by using electricity from solar energy and wood as an energy source, among others.

The sustainable process approach has been criticised for its rather long time horizon and the many assumptions associated with it (Müller-Wenk 1998; European Commission 2011).

4.4 Summary of the Methods

A summary of the methods presented in this chapter can be found in Table 13.1, where they are classified into the three ‘abiotic resource use’ categories mentioned in the beginning of the chapter, their main literature reference are given, and the types of resource they evaluate are marked. We also included some well-known methods for water and land (dealt with in dedicated chapters of this volumeFootnote 1) for completeness.

Table 13.1 Summary of the methods for abiotic resource use, showing the type of resources they consider and to which category they belong (e.g. resource depletion at midpoint level)
Full size table

5 New Developments and Research Needs

There are still many gaps in the way ‘abiotic resource use’ is evaluated in LCA. Especially for midpoint methods that are based on use-to-resource ratios, it is debatable how relevant the different resource definitions are for resource availability for future generations. Moreover, the dynamic nature of the concept of resources, which is affected by technology and economics to a varying extents depending on which resource definition is employed, and the estimate of the actual amount and quality of available stocks remain pending issues. At the same time, the needs of the future generations are not easy to define. For the future generations, needs may move to resources that are not in frequent use today. Overall pressure on the resource base is likely to increase as a consequence of increases in population numbers and material standard of living, and urban mining could become much more important. Overall, data availability for quantifying impacts at the AoP Resources is a considerable concern.

Recent research is being done to tackle some of these issues, for instance in the project LC-Impact (http://www.lc-impact.eu/). One example is the method from Vieira et al. (2012), which evaluates the resource depletion of metals based on the global ore grade information. Therefore, it would belong to the category ‘resource depletion – midpoint’, but considering future consequences. It provides CFs for three different types of copper deposits.

Other methods try to include the stocks of metals which have accumulated in the technosphere, e.g. Schneider et al. (2011), reflecting the idea of urban mining. Furthermore, the inclusion of competition and regional availability could be aspects to be included when assessing abiotic resource use in LCA (Yellishetty et al. 2011). By some it is even suggested to model process changes due to current resource dissipation, like increased energy needs, in the inventory phase of an LCA, rather than in the LCIA phase (Finnveden 2005; Weidema et al. 2005).

Abiotic resource availability is an important issue for society, yet there is still no clear consensus method for LCIA (European Commission 2011). The existing methods are quite diverse. Furthermore, uncertainties and data availability remain a challenge in the modelling of impacts of abiotic resource use as considerable parts of the planet are not fully explored and well documented.