Page 250 - Materials Chemistry, Second Edition

L1644_C06.fm Page 223 Monday, October 20, 2003 12:06 PM

of interest. Of course he must do this while taking into account budget restrictions.
In the goal definition it must be decided which situations or scenarios will be assessed
and compared. Here situations refer to existing process chains while scenarios mean
process chain options for the future (Pesonen et al., 2000).
In the scope definition, the decision-maker must select the functional unit (e.g.,
1 TJ electricity or 1 kg treated waste), the initial system boundaries (e.g., 1% con-
tribution to functional unit), the LCIA characterization potentials (e.g., GWP, AP,
NP, etc.) and/or single index methods or endpoint-orientated methods (e.g., eco-
indicator 95 or 99, EPS, etc.). These requirements correspond to those for LCA
according to the ISO 14040 series. However, in the methodology of environmental
damage estimations for industrial process chains, more information is necessary to
outline the study. These decision points are obligatorily the initial cut-off criteria
(e.g., site-specific 5% and literature values 1%) and the weighting and aggregation
scheme. Deciding if an uncertainty analysis should be included, if accidents should
be considered and if the eco-efficiency of the process chain should be calculated is
optional.
Initial cut-off criteria must be defined for the dominance analysis. These cut-off
criteria serve to determine which media, processes and pollutants must be further
studied in the fate and exposure and consequence analysis and in which way, e.g.,
site specific or by literature values.
The methodology is based on the principle of transparency in the way the results
are obtained. The format in which the results are desired, such as monetary values
or physical impacts (e.g., cases of cancer), determines this. Therefore, the subjective
elements, in particular the different parts of the weighting step, are assembled in the
goal and scope definition before carrying out any analysis. For the weighting,
decision-makers can follow the general decision tree presented in Figure 6.7. There
are different options to evaluate the ELs, whose choice depends on the worldview
of the decision-maker. Thus, the methodology avoids implicit decision-making com-
mon in endpoint-orientated LCIA methods.
For environmental loads that cause a GWP with global impact, ODP and other
global indicators can be calculated. In such a case these potentials have environ-
mental relevance in the form of life-support functions and depletion of natural
resources. First, decision-makers must select the environmental impacts they con-
sider relevant for the environmental management problem under study; then they
must decide, according to available knowledge, if the damages related to these
potentials may be estimated. If they think that these damages cannot be estimated
with acceptable reliability, then, for each global indicator, they must decide if they
prefer to monetize the potential impact using abatement costs or to express it directly
as a physical impact potential, e.g., CO equivalent. These can be assessed in
2
conjunction with the other environmental loads if they believe the damages to be
estimable (see Figure 6.7).
Other environmental loads may cause local and regional damages, which can
be divided into the AoPs published by Udo de Haes et al. (1999): manmade envi-
ronment, natural environment and human health. In the cases of manmade environ-
ment and natural environment, the questions on which to decide are the same. Due

© 2004 CRC Press LLC