Page 186 - Materials Chemistry, Second Edition

MODELING THE AGRI-FOOD INDUSTRY WITH LIFE CYCLE ASSESSMENT 171

(SOM) (Milä i Canals, et al. 2007) is considered a valid overall midpoint indica-
tor but it gives no indication on other impacts such as soil erosion, compaction
and salination. Müller-Wenk & Brandäo (Müller-Wenk & Brandäo, 2010) have
considered the carbon transfer between vegetation/soil and land as a means of
measuring the impact of land use.
"Endpoint Approach" indicators (Koellner & Scholz, 2007; Koellner &
Scholz, 2008) take into account the naturalness of the system analyzed (e.g. bio-
diversity in terms of: potential disappeared fraction of vascular plant species)
and are better related to other traditional impacts (e.g acidification and eutro-
phication). A recent approach to biodiversity assessment is found in (Koch,
et al. 2010); however, it has only been tested on two indicators (grassland flora
and grasshoppers) and needs to be broadened to include other indicators. In
general "Endpoint Approach" indicators do not consider aspects such as the
effects on human health or loss of landscapes (Mattsson, et al. 2000; Lindeijer,
2000). New methodological approaches that take into consideration these
impacts still need to be developed.
Since there are many possible land use indicators that can be considered,
it is advisable to choose a set of indicators that best represents the agri-food
environment being modeled. This will most likely force the LCA to follow
a determined impact assessment path that might not represent all the possible
impacts. On the other hand, the combination of all the indicators, considered
in LCA study, into a final aggregate overall impact value is often difficult to
achieve and at times unadvisable due to the very different nature of the col-
lected information. For example, Mattsson when considering the land use for
vegetable oil crops suggested that land use assessment is likely to be more
descriptive and a step closer to an Environmental Impact Assessment study
resulting in multiple impacts which should be associated with traditionally
aggregated LCA results.
Site specificity is an issue that can cause great variability of the land use
LCA results (Schryver, et al. 2010). Different regionalized datasets are already
available, e.g. ReCiPe and LIME (Goedkoop, et al. 2009; Itsubo, 2008), and
should be used when possible in order to obtain realistic results. Recently
Pfister et al. (Pfister, et al. 2010) used regionalized inventory and impact
assessment data based on ecosystem vulnerability and net primary produc-
tivity of energy crops in order to demonstrate the tradeoffs between land and
water use. This approach clearly demonstrates the variability of the regional-
ized results compared to the global averages; specifically although globally
water only causes 25% of the global land impact, the land use impact is least
in regions that have a high water use which, in this case, causes most of the
damage. The authors of (Geyer, et al. 2010; Geyer, et al. 2010; Nunez, et al.
2010) developed methods that make use of geographical information sys-
tems (GIS) to couple site specific information with indicators regarding land
use. Such bio-geographical differentiation approaches need to be further
refined in order to be made mainstream but should nonetheless be imple-
mented whenever possible in order to improve the quality of the agri-food
LCA results.