3.3.1 Impact category indicator and general characterization equation for three different impact categories of resource use
Three different impact categories of resource use
To assess the impact of the present use of elements, four different compromising actions were defined: exploration, dissipation environment, hibernation technosphere, and occupation in use. The next framework challenge was to merge these compromising actions into one overall characterization model assessing the impacts of these compromising actions on the future accessibility of resources.
In principle, a characterization model builds upon a cause-effect mechanism linking the use of resources to an impact on the accessibility of resources. Compromising actions with similar mechanisms or starting assumptions can be part of the same model and thus belong to the same impact category, while essentially different actions with different mechanisms or starting assumptions should have their own impact category. In the latter case, the impact categories are essentially different and any aggregation into a single impact score for the use of resources would require additional weighting.
We first argue that exploration is different from environmental dissipation, technosphere hibernation, and occupation in use. Exploration mostly adds to the stock (
R) from which humans can extract/use a resource, whereas dissipation, hibernation, and occupation determine the fate of the resource used. In other words, exploration adds to accessibility through the increase of the stock, while the other compromising actions add to some sort of inaccessibility. Thus, in terms of Eq.
1, exploration is not part of the
Ci but contributes to
Si (where the R is included). The other compromising actions (occupation in use, hibernation, and environmental dissipation) are part of the parameter that describes the fraction of the used elements that is made inaccessible (
Ci).
We next argue that due to the different character of the impacts and mechanisms of the compromising actions, a hierarchy of three different levels of reversibility of the inaccessibility of elements can be distinguished. Consequently, three different impact categories for assessing the impacts of abiotic resource use were distinguished:
1.
(Assumed) irreversible inaccessibility of a resource within the time horizon considered: environmental dissipation (see Section
3.1.2)
2.
Potentially reversible but temporary inaccessibility of a resource within the time horizon considered: technosphere hibernation (including also dissipation in the technosphere) (see Section
3.1.2)
3.
Reversible but temporary inaccessibility of a resource within the time horizon considered: occupation in use (see Section
3.1.2)
Depending on the exact goal and scope of an LCA case study, either environmental dissipation alone, or together with hibernation in technosphere, or together with hibernation in technosphere and occupation in use should be included in LCA studies. We recognize that occupation in use actually corresponds to the desired “role of resources” (Section
3.1.1) and delivers benefits now and for the next generation, but it nevertheless prevents benefits from a second application of the same unit of resource at the same time. Therefore, even for the short term, we argue that it would be more correct not to include occupation in use in the impact assessment at all, since LCA cannot capture such short-term dynamics.
In the case of three impact categories proposed in this article, optional weights could be applied to the different category indicator results to derive one overall indicator score for inaccessibility. When one considers reversible changes less problematic than irreversible changes, a lower weight can be given to the impact category that covers the reversible changes to inaccessibility, i.e., “occupation in use” and, to a less extent, “hibernation in technosphere.”
In the Electronic supplementary material (ESM
2), general equations are derived for the three impact categories based on Eq.
1. The generic equations are translated into specific characterization models for all three impact categories and the long and short time horizon.
Note that the way that exploration, environmental dissipation, technosphere hibernation, and occupation in use are handled in the further elaboration of practical methods for type B perspectives depends on the time horizon adopted. For example, for short time horizon, all may be relevant, while for the very-long-term time horizon, only environmental dissipation might be considered relevant, as will be discussed below.
From ESM
1, it becomes clear that particularly for the short time horizon, the elaboration of the characterization model into an operational set of characterization factors is challenging (see discussion in Section
4.3). For practical reasons of data availability, the generic equations provided in ESM
2 will only be elaborated for the environmental dissipation impact category for the very-long-term time horizon.
3.3.2 Equation for characterization factor for environmental dissipation for the very-long-term time horizon
Equation
1 provides the generic equation for characterization factors and is assumed to be valid for all three impact categories and both time horizons, very long term and short term. To calculate the characterization factor for environmental dissipation, the
Ct, T, i and
St, T, i are determined for the environmental dissipation potential (
CFt, T, i is
EDPt, Ti). For environmental dissipation, the two components are elaborated, namely,
Ct, T, i for the change of inaccessibility due to the compromising action and
St, T, i for its severity.
Note that Eq.
1 represents a top-down approach regarding the modeling of compromising actions. It models the compromising actions for the change in accessibility of a resource as a fraction of the global use of resources. A bottom-up approach, on the other hand, would rather model the compromising actions on the basis of individual applications (product) in which a resource is applied and then aggregate these resource applications to the global use of that resource. In this article, a top-down approach is assumed to be still challenging but more feasible than a bottom-up approach.
Table
2 summarizes the terminology and symbols for the equation of the characterization factor of environmental dissipation. Lowercase symbols
mi,
ei,
hi,
oi, and
sri refer to product system-related quantities (as a result from an LCA study), while the capital symbols
Ei,
Hi, and
Oi refer to the global total amounts of the same flows, which are used in one or more characterization models for the three impact categories.
Table 2
Overview of main terms and symbols adopted for describing the equation for the characterization factor of environmental dissipation
(Number of years of) the time horizon adopted for assessing the potential decrease of resource accessibility | T | Year (yr) | |
Time for which the characterization factors are assumed to be representative | t | Year (yr) | |
Amount of primary resource i consumed by a product system | mi | kg/FU | Calculated as part of the LCI of an LCA study |
Amount of resource i emitted by a product system | ei | kg/FU | Calculated as part of the LCI of an LCA study |
Amount of secondary resource i consumed by a product system | sri | kg/FU | Not reported by LCA studies yet |
Global amount of primary resource i consumed by all products in year t, equalling the world annual extraction of resource i as reported by USGS or BGS | Mt, i | kg/year | (British Geological Survey 2018; US Geological Survey 2018) |
Global amount of resource i emitted by all products in year t | Et, i | kg/year | |
Global amount of secondary resource i consumed by all products in a specific year | SRt, i | kg/year | |
Global amount of primary and secondary resource i consumed (represented by total primary and secondary production) by all products in year t | Pt, i | kg/year | =Mi + SRi |
Cumulative global emissions of resource i over time horizon T (25 years or VLT) | \( {E}_{t,T,i}=\sum \limits_{t=1}^T{E}_{t,i} \) | kg | |
Characterization factor for resource i | CFi | kg ref/kg i (e.g., kg Sb-eq./kg i) | |
Fraction of the present (t) global primary extraction and secondary use of resource i over time horizon T (25 years or VLT), made inaccessible | Ct, T, i | - | |
Severity of making 1 kg of resource i inaccessible for time horizon T (25 years or VLT) | St, T, i | Depending on method (e.g., 1/(year.kg i)) | |
Environmental dissipation potential (CFi for environmental dissipation of resources), based on the cumulative global emission of resource i that goes into hibernation within the time horizon adopted due to the use of resource i on time t | EDPt, T, i | kg-Cu-eq/kg i | |
Category indicator result for time horizon T (25 years or VLT) for “environmental dissipation” for a product system | EDt, T | kg-Cu-eq. | |
Global ultimate stock of resource i in the environment based on the crustal content | Rult, i | kg | |
Global accessible stock (e.g., economic reserve) of resource i in the environment as projected for year t | Renv, t, i | kg | (US Geological Survey 2018) |
Global accessible stock of resource i in technosphere as projected for year t | Rtech, t, i | kg | |
Total global accessible stock of resource i in the environment and technosphere | Rtot, t, i | kg | Renv, t, i + Rtech, t, i |
In the context of dissipation to the environment, the fraction
Ct, T, i in Eq.
1 refers to that part of the present use of resource
i that is made inaccessible due to emissions from successive applications within the time horizon
T considered. In this case
Ct, T, i for the impact category, the environmental dissipation is:
$$ {C}_{ED,t,T,i}=\frac{E_{t,T,i}}{P_{t,i}} $$
(2)
with
Et, T, i representing the cumulative global emissions of resource
i starting at time
t within the time horizon adopted due to the present use of resource
i and
Pt, i representing the annual global total primary extraction and secondary provision of resource
i in year
t.
To express the severity of making 1 kg of resource
i inaccessible for time horizon
T, we propose to use the total accessible stock (
R) in the environment and technosphere, complemented by the annual total production (
P) of that resource (as presently also defined in the ADP (Guinée and Heijungs
1995) (van Oers et al.
2019))
5. So, adopting the
\( \frac{P_i}{R_i^2} \) ratio from the original ADP equation for the
St, T, i, we get:
$$ {S}_{t,T,i}=\frac{P_{t,i}}{R_{tot,t+T,i}^2} $$
(3)
with
Rtot, t + T, i representing the total accessible stock in the environment and technosphere in year
t +
T, which represents the final year of the time horizon adopted. Based on the general equation for characterization factors (Eq.
1), we would get the following general equation for the environmental dissipation potential (EDP) as the characterization factor for the environmental dissipation of resources:
$$ {EDP}_{t,T,i}=\raisebox{1ex}{$\left(\frac{E_{t,T,i}}{P_{t,i}}\right)\times \frac{P_{t,i}}{R_{tot,t+T,i}^2}$}\!\left/ \!\raisebox{-1ex}{$\left(\frac{E_{t,T, ref}}{P_{t, ref}}\right)\times \frac{P_{t, ref}}{R_{tot,t+T, ref}^2}$}\right.=\frac{\raisebox{1ex}{${E}_{t,T,i}$}\!\left/ \!\raisebox{-1ex}{${R}_{tot,t+T,i}^2$}\right.}{\raisebox{1ex}{${E}_{t,T, ref}$}\!\left/ \!\raisebox{-1ex}{${R}_{tot,t+T, ref}^2$}\right.} $$
(4)
In order to be able to derive an operational set of characterization factors, additional assumptions are necessary. Equation
4 gives the characterization equation for the impact category Environmental Dissipation. Ideally the emission parameter
Et, T, i is based on the cumulative global emission over
T =
VLT due to the present (
t) and successive future applications of the present demand (use) of resource
i, e.g., the cumulative emissions from 2020 to VLT of all applications that presently (2020) use resource
i.
However, successive emissions over the future time horizon are difficult to estimate. Therefore, we here pose, as a rough assumption, that primary extraction at present (e.g., in the year 2020, so
M2020, i) equals the very-long-term emission to the environment (i.e
E2020, VLT, i):
$$ {E}_{2020, VLT,i}\approx {M}_{2020,i} $$
(5)
This assumption is true for the very-long-term time horizon (
T → ∞) but certainly will not be true for the shorter term (see discussion in Section
4.2). So, implicitly we assume that the relative differences between the amounts of various elements for “relative cumulative emissions over the time horizon” and “relative present extractions” will become more and more negligible for the very long term (VLT), starting with 100 years and more (note: it is a consequence of this assumption that any proxy greater than 100 years could be chosen without changing the result of the calculation, e.g., 500 years, 5000 years, 50,000 years, etc.).
Secondly, in the very long term, the total accessible stock
Rtot, t + VLT, i can be approximated by the crustal content stock
Rult,
i:
$$ {R}_{tot,t+ VLT,i}\approx {R}_{ult,i} $$
(6)
Referring to the general characterization equation of environmental dissipation (Eq.
4, adopting
T =
VLT, and assuming that emissions in the very long term equal present (e.g., 2020) production, the environmental dissipation potential (EDP) is then constructed as follows:
$$ {EDP}_{2020, VLT,i}=\frac{\raisebox{1ex}{${M}_{2020,i}$}\!\left/ \!\raisebox{-1ex}{${R}_{ult,i}^2$}\right.}{\raisebox{1ex}{${M}_{2020, ref}$}\!\left/ \!\raisebox{-1ex}{${R}_{ult, ref}^2$}\right.} $$
(7)
Here, M2020, i is now the world’s annual primary production (kg/yr) in 2020 of resource i, and Rult, i (kg) represents continental crustal content of resource i, which is taken to represent the total accessible stock in the environment and technosphere over the very long term.
Equation
7 resembles the old ADP except that we don’t apply it to the extraction of
i from the environment, but to the total emission of
i to the environment (see Section
3.4). In fact, when the reference substance is the same for EDP and ADP, we can write:
$$ ED{P}_{2020, VLT,i}(ref)\approx AD{P}_{2020,i}(ref) $$
(8)
When the reference substances are not the same (see below), they are merely proportional:
$$ ED{P}_{2020, VLT,i}\left( re{f}_1\right)\propto AD{P}_{2020,i}\left( re{f}_2\right) $$
(9)
3.3.3 Data for characterization factor
For the calculation of the environmental dissipation potential (EDP) of the different elements over the very long term, the following input data are needed:
a)
Rult, i, the global stock in the environment and technosphere of resource
i, represented by the continental crustal content data of resource
i by Rudnick and Gao (
2014)
b)
M2020, i, the world’s annual primary production of elements for a specific year (here 2020) which can be directly taken from van Oers et al. (
2019)
In Oers et al. (
2019), accessible stock estimates for the environment are derived from available stocks, consisting of the continental crust, ocean, and atmosphere. For these calculations, the average continental crust concentrations are taken from Rudnick and Gao (Rudnick and Gao
2014).
If the
ADP2020, i is taken as a second
6 best or currently best and most feasible estimation of the
EDP2020, VLT, i, it can be directly taken from van Oers et al. (
2019) who updated production data based on the USGS and BGS reports (British Geological Survey
2018; US Geological Survey
2018) for about 80 elements over the period 1970–2015. The
ADPi, 2020 would only be available in the course of 2021 but could be approached for the time being by the
ADP2015, i.
Abiotic depletion potentials (ADPs) are expressed in kg antimony (Sb) equivalents extracted per kg extraction. It is proposed to express environmental dissipation potentials (EDPs) in kg copper (Cu) equivalents emitted per kg emission, for the following reasons:
-
Using another reference substance emphasizes that we deal with a different impact category using a different model (depletion versus dissipation) based on a different problem definition.
-
A different unit will help to avoid that indicator scores of abiotic depletion and environmental dissipation are mistakenly aggregated.
-
Finally, copper is generally perceived as more illustrative to express the problem of decreasing resource accessibility than antimony.
With these choices for the reference substances, the proportionality factor between the two potentials is
$$ \frac{ED{P}_{2020, VLT,i}(Cu)}{AD{P}_{2020,i}(Sb)}\approx 36.41 $$
(10)