Greg Thoma, Ph.D. | University of Arkansas
Figure 1. Impact World+ Methodology Framework (adapted from www.impactworldplus.org).
Table 1. Eco-profile data sources (Battagliese et al., 2013)
Eco-Profile |
Source, Year |
Comments |
Cardboard, recycled |
Ecoinvent 2.2, 2010 |
Ecoinvent profile: corrugated board, recycling fiber, double wall, at plant/RER U |
Paper |
|
Ecoinvent profile: Paper, wood free, uncoated at non-integrated mill /RER U |
Polypropylene |
BASF, 1996 |
|
Wood pallets |
|
Ecoinvent profile: wood container and pallet manufacturing (USA Input Output Database) |
The most important characteristic of a high-quality LCA is transparency of the data and data sources. Transparency allows users and reviewers to evaluate, in detail, the foundational information which has been used to support the study conclusions. However, situations exist where complete transparency is not possible due to aggregation and use of confidential data or trade secrets. In these cases, an explanation of aggregation and reasons for nontransparent data should be provided. Table 1 provides a sample of data sourcing with an appropriate note regarding confidential data coupled with a third-party review.
Table 2. Results of 1,000 Monte Carlo runs for uncertainty analysis of dry whey from cradle-to-customer per ton of dry whey solids (Kim et al., 2013).
Impact Category |
Unit |
Mean |
CV (%) |
95% CI |
|
Climate change |
Kg CO2e |
1.21E+ 04 |
15.3 |
9.11E+03 |
1.61E+04 |
Cumulative energy demand |
MJ |
5.81E+ 04 |
28.5 |
4.09E+04 |
8.93E+04 |
Freshwater depletion |
m3 |
1.45E+ 03 |
16.2 |
1.05E+03 |
2.00E+03 |
Marine eutrophication |
kg N eq. |
3.73E+ 01 |
12.2 |
2.92E+01 |
4.77E+01 |
Photochemical oxidant formation |
kg NMVOC |
4.40E+ 01 |
12.9 |
3.33E+01 |
5.60E+01 |
Freshwater eutrophication |
kg P eq. |
7.52E+ 00 |
15.6 |
5.53E+00 |
1.01E+01 |
Ecosystems |
Species/year |
3.51E− 04 |
13.4 |
2.70E-04 |
4.54E-04 |
Human toxicity |
CTUh |
2.27E− 04 |
116 |
7.78E-05 |
7.29E-04 |
Ecotoxicity |
CTUe |
7.57E+ 04 |
14.9 |
5.69E+04 |
1.01E+05 |
Another characteristic is an analysis of the data quality which was achieved in the inventory phase as it relates to the ability of the authors to achieve the goal of the study. As mentioned previously, high-quality data, the absence of gaps in data for unit processes and the utilization of primary data rather than secondary data are all characteristics of higher-quality studies. The paper should provide a discussion of whether data gaps or the use of proxy or surrogate datasets may have impacted the study conclusions. The influence of modeling assumptions on the study results – such as choice of allocation procedures and decisions to include or exclude some aspects of the supply chain – should be evaluated through scenario analysis. High-quality studies will also include uncertainty analysis. Typically, Monte Carlo simulation is used to demonstrate the effects of data input uncertainty on LCA results, as shown in the example in Table 2. Finally, a section in the paper which discusses the study limitations with respect to conclusions also demonstrates quality in the results.
Bottom Line: A comprehensive, high-quality life cycle assessment will be a cradle-to-grave assessment with multiple impact categories, spanning major areas of production and in compliance with ISO standards. In addition to these characteristics, the data used in the assessment should be transparent and a thorough analysis of the quality of the data as it pertains to the results should be performed. Studies which are focused on a single (i. e., footprints) or relatively few impact metrics are less comprehensive, because the ability to identify trade-offs is limited.