Page 107 - Materials Chemistry, Second Edition

LIFE CYCLE IMPACT ASSESSMENT 89

body." Although this is not a mandatory "shall" requirement, only a few char-
acterization models and factors currently satisfy this recommendation: the IPCC
(Intergovernmental Panel on Climate Change) model calculating the Global
Warming Potential (GWP) for the climate change midpoint impact category
(Forster et al 2007), the WGMO (World and Global Meteorological Organization)
model calculating the Ozone Depletion Potential (ODP) for the stratospheric
ozone depletion impact category. For other impact categories, there has been
only modest activity in international harmonization and scientific consen-
sus (Hauschild, Goedkoop et ah 2012). This, however, is recently changing as
researchers strive to find consensus in the models and underlying data.
A breakthrough in identifying and recommending state-of-the art character-
ization models and factors has been made by the European Platform on LCA.
An extensive evaluation of existing LCIA methods and characterization mod-
els was performed by model developers and scientific experts with the aim to
identify the best existing practice. Through a consultation process involving
listening to domain experts as well as stakeholders, the evaluation formed the
basis of recommendations of characterization models and factors for impact
categories at midpoint and at endpoint level (EC-JRC 2010b).
The evaluation process was conducted in three steps. In the first step, the
different characterization models used by each LCIA methodology in the
characterization of impact categories and areas of protection were identified.
This resulted in the identification of 156 characterization models stemming
from eleven LCIA methodologies. In addition there were a few models which
are not part of formal LCIA methodologies but showed interesting features.
Of these, 91 were pre-selected and included in the following analysis.
In the second step, criteria and procedures for the evaluation of character-
ization models addressing midpoint and endpoint levels were developed. Five
scientific criteria (completeness of scope, environmental relevance, scientific
robustness and certainty, documentation & transparency & reproducibility,
and applicability) and a stakeholder acceptance criterion were developed to
evaluate all impact categories at the midpoint level and at the endpoint level.
Each of these criteria was further detailed into a set of sub-criteria. Many
sub-criteria were general and applied to each impact category. But for the
scientific criteria on environmental relevance and scientific robustness and
certainty, the sub criteria were developed specifically for each impact category,
reflecting the central characteristics of the underlying impact pathway. An
analysis of the impact pathway of each category helped identify key processes
or aspects that should be considered in the characterization modeling, and
these were the basis of formulating the category-specific sub criteria (flow
sheets for each impact category can be found in EC-JRC 2010b).
In the third step, the 91 shortlisted characterization models were further
analyzed and compared to each impact category. The quality of the selected
characterization model was assessed along three levels of recommendation:

I - Recommended and satisfactory;
II - Recommended but in need of some improvement; and
III - Recommended, but to be applied with caution (See Table 4.10).