Page 42 - Materials Chemistry, Second Edition

28 E. I. Wiloso and R. Heijungs

one company or farm. An example of this is that someone may pay to have their
residues picked up, while someone else must pay to receive it. Further, due to
technological developments, fluctuations in markets, and governmental interven-
tion, goods may rapidly turn into a waste or the other way around (Heijungs and
Wiloso 2012).

3.5 Data Variability

An LCA depends on a large number of input elements, and these elements are
often based on data of varying quality. The variability in input quality will, in turn,
influence the robustness of outcome estimates. This is an important issue that
deserves more attention in LCA. A strong challenge for LCA in addressing
uncertainty is to provide and track metrics of data quality with respect to how data
are acquired (measurements, assumptions, expert judgment), to what extent the
data have been validated (checked with respect to mass and energy balance), and
how well the data capture technological, spatial, and temporal variations. Some of
these uncertainties and variabilities cannot be reduced with the current knowledge
(through improvements in data collection or model formulation) because of their
spatial and temporal scale and complexity (McKone et al. 2011).
When developing LCIs, one needs quantitative data on the inflows and outflows
of the included processes such as resource use, emission data, energy use, and
waste production. The limited accuracy and availability of LCI data are generic
problems of LCAs. Uncertainty can be due to various reasons that may stem from
geographical, temporal, and technological differences. In the case of bioenergy
systems, common sources of uncertainty include variability in agriculture yield as
it depends on soil conditions, weather, and agricultural practices; variability in
biomass conversion technology at different development status; and regional
variability as the data are known only for certain countries (Heijungs and Wiloso
2012). Despite the above difficulties, doing LCA is now much easier than ten years
ago since there are now a number of online data repositories for different conti-
nents. Some of these databases are quite extensive, though mostly for the USA and
EU. The Ecoinvent database, for example, contains thousands of processes from
electricity production to transport by truck and from palm oil production to pes-
ticide production (Ecoinvent 2010).

3.5.1 Agricultural Process Variability

Data variability in the agricultural chain of bioenergy systems is an issue in LCI.
For example, there are a large number of potential biomass feedstocks with dif-
ferent characteristics. This presents substantial challenges for current LCA
approaches because of the vast scope of information needed to address so many
alternatives (McKone et al. 2011). The production of biomass feedstock is likely to