Environmental News, 78 – Impact categories, normalisation and weighting in LCA

Impact categories, normalisation and weighting in LCA

Summary and conclusions

The present Guidelines present an extension of the original EDIP methodology in the form of normalization references and weighting factors for EU-15 and the world. At the same time, the normalization references and weighting factors for Denmark have been updated. The results are presented in the table below:

Click here to see the table.

The establishing of the normalization references and weighting factors is documented in a technical report (Stranddorf et al., 2005). In short, the new normalization references have been established in two ways, i.e. by using actual emissions in the calculations, or by extrapolating the emissions from one region (e.g. the EU) to a global scale by using the gross domestic product (GDP). Obviously, the first option provides the most precise results and has been the preferred choice, but emission inventories of sufficient quality are only available for a few impact categories and therefore the developed extrapolation method was applied for the local and regional impact categories to yield figures on the Global scale.

The weighting factors have been calculated according to the methodology described in the original EDIP reports. Consistent information on reduction targets for local and regional impacts was difficult to find on the Global scale and were therefore not calculated; the Guidelines provide recommendations on what to do instead.

The general recommendations are fairly simple, and can in short be expressed in the following way:

For global impacts (global warming and stratospheric ozone depletion) always use the worldwide normalisation reference and weighting factor in the base case analyses
For regional impacts (acidification, photochemical ozone formation and nutrient enrichment) and local impacts (ecotoxicity, human toxicity) use the EU-15 normalisation reference and weighting factor as the base reference
Where relevant, use normalisation references and weighting factors for other geographical regions as an element in the sensitivity analyses, acknowledging the inherent uncertainties.

The recommendations above reflect to some extent the inherent uncertainties in normalisation, especially if the step has a broad scope. Obviously, the more is known about the product (system) investigated as regards the geographical extent of its potential impacts, the more precisely the normalisation step will mirror the relative importance of different impacts.

The recommendations are a modification of earlier recommendations for EDIP₉₇, where Danish normalisation references were recommended for regional and local impacts. The suggested shift to the EU normalisation reference is justified by the better scope for many industrial products, combined with the fact the EU-normalisation references are of an acceptable quality.

Although the general recommendations are simple, they still leave the LCA practitioner with the final choice. The reason for this is that most LCAs using the original EDIP methodology will include impact potentials that occur in a wide selection of countries and regions and it is not possible to discriminate between these in the final results. The practitioner must therefore decide which approach gives the most suitable results for the user of the LCA.

The Guidelines include an example, where the recommended values are highlighted and the alternative results presented alongside, together with the results obtained by using the original EDIP method. The example thus provides an interesting overview of both the importance of the development in emission inventories since the original EDIP and the differences that are observed when using different approaches.

It is outside the scope of this report to discuss the findings in the example in detail, but it is obvious that the increased number of possibilities for choosing can provide a more balanced view of the life cycle impacts in the life cycle of a product. At the same time, however, the increased amount of information can also cause additional questions to be asked.

Exactly which questions that will emerge cannot be determined at this moment, but it is a general recommendation that they should concentrate on those issues where significant changes are observed.

The improved possibilities for normalisation and weighting have not been utilised in "real" LCA so far. LCA-practitioners will most probably quickly find a way to use the possibilities, e.g. as a part of the sensitivity analysis. The efforts needed in doing so will decrease significantly when the first experiences have been gained.

Finally, it is mentioned that in Subproject 5 of the Danish LCA Methodology Development and Consensus Project, an alternative to the original EDIP methodology has been developed and documented (Hauschild og Potting, 2005). This alternative focuses on using the site-specificity of the contributions to different environmental impacts in the assessment and thereby reducing the inherent uncertainties associated with the normalisation and weighting procedure used in the original EDIP method.