Page 202 - Materials Chemistry, Second Edition

188 R.K. Rosenbaum et al.

• Steady-state: Although exceptions exist, LCIA models are usually not dynamic
(i.e. representing the variation of an environmental system’s state over time and
for specific time steps), but represent the environment as a system in steady
state, i.e. all parameters which define its behaviour are not changing over time.
• Linearity: As life cycle inventory (LCI) data are typically not spatially and/or
temporally differentiated, integration of the impact over time and space is
required. In LCIA, this leads to the use of characterisation models assuming
steady-state conditions, which implies a linear relationship between the increase
in an elementary flow and the consequent increase in its potential environmental
impact. In other words, e.g. doubling the amount of an elementary flow doubles
its potential impact.
• Marginal versus average modelling: These terms are used in different ways and
meanings in the LCA context; here they describe two different impact modelling
principles or choices: a marginal impact modelling approach represents the
additional impact per additional unit emission/resource extraction caused by the
product system on top of the existing background impact (which is not caused
by the modelled product system). This allows, e.g. considering nonlinearity of
impacts depending on local conditions like high or low background concen-
trations to which the product systems adds an additional emission). An average
impact modelling approach is strictly linear and represents an average impact
independent from existing background impacts, which is similar to dividing the
overall impact by the overall emissions. This is further discussed by Huijbregts
et al. (2011). Note that marginal and average modelling are both suitable for
small-scale interventions such as those related to a product or service. However,
when medium-scale or large-scale interventions (or consequences) are to be
assessed, the characterisation factors should represent non-marginal potential
impacts and may also have to consider nonlinearity.
• Potential impacts: LCIA results are not actual or predicted impacts, nor
exceedance of thresholds or safety margins, or risk. They are relative expres-
sions of impacts associated with the life cycle of a reference unit of function
(=functional unit), based on inventory data which are integrated over space and
time, representing different locations and time horizons and based on impact
assessment data which lack information about the specific conditions of the
exposed environment.
• Conservation of mass/energy and mass/energy balance: Mass/energy cannot be
created or disappear, it can only be transferred. Following this principle, pro-
cesses of transport or transformation of mass or energy are (or at least should be)
modelled assuming that the mass/energy balance is conserved at all times.
• Parsimony: This refers to the basic modelling principle of “as simple as possible
and as complex as necessary”, an ideal balance that applies to LCIA charac-
terisation models as well as to the entire LCA approach.
• Relativity: LCA results are relative expressions of impacts that relate to a
functional unit and can be compared between different alternatives providing the
same function (e.g. option A is more environmentally friendly than option B).