Page 292 - Materials Chemistry, Second Edition

A Comparison of Life Cycle Assessment Studies 283

Finnveden (2000) noted the slightly different impact category lists that have
been proposed by different organizations. The lack of standardization of some
impact categories is demonstrated in the recent debate as to whether certain impact
categories such as soil salinity, desiccation, and erosion should be their own
category or part of another category such as land-use impact and freshwater
depletion (Jolliet et al. 2004). McKone et al. (2011) pointed out a key challenge for
applying LCA to a broadly distributed system (e.g., biofuels) is to rationally select
appropriate spatial and temporal scales for different impact categories without
adding unnecessary complexity and data management challenges as significant
geographical and temporal variability among locations over time could influence
not only the health impacts of air pollutant emissions, but also soil carbon impacts
and water demand consequences, among other factors. McKone and co-worker
suggested that accurate assessments must not only capture spatial and variation at
appropriate scales (from global to farm-level), but also provide a process to
aggregate spatial variability into impact metrics that can be applied at all geo-
graphical scales. The selection of midpoint or end point (damage) impact cate-
gories is another potential result affecting criteria for both the level of confidence
or relevance for decision making on the basis of LCA study results (Reap et al.
2008b). End point categories are less comprehensive and have much higher levels
of uncertainty than the better defined midpoint categories (UNEP 2003), and
midpoint categories, on the other hand, are harder to interpret because they do not
deal directly with an end point associated with an area of protection (Udo de Haes
et al. 2002) that may be more relevant for decision making (UNEP 2003).
The International Program on Chemical Safety (WHO 2006) proposed four
tiers, ranging from the use of default assumptions to sophisticated probabilistic
assessment to address uncertainty in risk assessment:

Tier 0: Default assumptions; single value of result
Tier 1: Qualitative but systematic identification and characterization of
uncertainties
Tier 2: Quantitative evaluation of uncertainty making use of bounding values,
interval analysis, and sensitivity analysis
Tier 3: Probabilistic assessments with single or multiple outcome distributions
reflecting uncertainty and variability.

Cherubini and Strømman (2011) reviewed several biofuel LCA studies and
found that very few studies (about 9 %) included land-use category in their impact
assessment. This is an important indicator particularly for bioenergy systems based
on dedicated crops or forest resources, since land use may lead to substantial
impacts, especially on biodiversity and on soil quality. This is mainly due to the
fact that there is no widely accepted methodology for including land-use impacts
in LCA, despite some recent efforts (Dubreuil et al. 2007; Koellner and Scholz
2008; Scholz 2007). Cherubini and Strømman (2011) also stated that for the same
reason, none of the reviewed studies included in the assessment the potential