Page 43 - Materials Chemistry, Second Edition

24 LIFE CYCLE ASSESSMENT HANDBOOK

for climate change. We will illustrate their usage below. For now, it suffices
to say that GWPs provide one example of a set of characterization factors,
and that the IPCC-model from which they are derived is an example of a
characterization model. Note, by the way, that IPCC has not developed this
model as a characterization model for LCIA, but that the LCA-community
has adopted this model as such and its derived GWPs as characterization
factors. Also note that the characterization model itself is not used by LCA
practitioners; only the characterization factors that have been derived from
it as a one-time exercise are used. Characterization factors are often tabu-
lated in LCA guidebooks and are implemented in many LCA software pack-
ages, while the characterization models often require supercomputers and
expert knowledge.
In fact, one element is needed before one can select a category indicator
and a characterization model with associated characterization factors. It is the
selection of impact categories to be addressed. Some LCA studies concentrate
on just one impact category. For instance, the carbon footprint (of a product,
not of a company or country) is considered a form of LCA that addresses just
climate change at the midpoint level through GWPs. At the other extreme,
some LCA studies incorporate fifteen or more impact categories. For con-
sistency reasons, the choice of impact categories is often made on the basis
of a recommended impact assessment guidebook or its implementation in
software. Thus, in practice one often sees LCA-studies reporting the use of
"IMPACT2002+," "TRACI," "CML-IA," "ReCiPe," "ILCD," etc. All these
methods comprise a recommended set of impact categories with a category
indicator and set of characterization factors. ISO does not specify any choice in
these matters. Table 2.3 gives an overview of some often-used impact catego-
ries and category indicators. We see that the column with endpoint indicators
contains many times the same term (e.g., "loss of life years"). This suggests
that impact categories can be aggregated into fewer endpoint indicators than
midpoint indicators.
As a concrete example of how characterization works, let us study a fragment
of a hypothetical inventory table, containing the following information: emis-
sion of C0 2 100 kg, emission of CH 4 1 kg, emission of 0 2 1 kg. Characterizing
S
greenhouse gases with GWPs requires a table with GWPs. In such a table, one
can find that the GWP of C0 2 is 1 (by definition) and that the GWP of CH 4 is 25
(kg C0 2 -equivalent/kg CH 4). S0 2 has no GWP; it is assumed not to contribute
to climate change. Characterization now proceeds in the case of climate change
by calculating

k
1 x 100 + 25 x 10 = 350 g C 0 2 - equivalent

For the more general case, this can be written as

GW = ^GWP sxm s