The role of atmospheric dispersion models and ecosystem sensitivity in the determination of characterisation factors for acidifying and eutrophying emissions in LCIA

The methodological choices and framework to assess environmental impacts in life cycle assessment are still under discussion. Despite intensive developments worldwide, few attempts have been made hitherto to systematically present the role of different factors of characterisation models in life cycle impact assessment (LCIA). The aim of this study is to show how European average and country-dependent characterisation factors for acidifying and eutrophying emissions differ when using (a) acidifying and eutrophying potentials alone, (b) depositions from an atmospheric dispersion model or (c) critical loads in conjunction with those depositions. Furthermore, in the latter case, the contributions of emissions, an atmospheric transport model and critical loads to changes in characterisation factors of NO₂ are studied. In addition, the new characterisation factors based on the accumulated exceedance (AE) method are presented using updated emissions, a new atmospheric transport model and the latest critical loads.

In this study, characterisation factors for acidifying and eutrophying emissions are calculated by three different methods. In the ‘no fate’ (NF) methods, acidifying and eutrophying potentials alone are considered as characterisation factors. In the ‘only above terrestrial environment’ (OT) approach, characterisation factors are based on the deposition of the acidifying or eutrophying substances to terrestrial land surfaces. The third method is the so-called AE method in which critical loads are used in conjunction with depositions. The results of the methods are compared both at the European and the country level using weighted mean, weighted standard deviation, minimum and maximum values. To illustrate the sensitivity of the AE method, changes in European emissions, employed atmospheric dispersion model and the critical loads database are conducted step-by-step, and the differences between the results are analysed.

For European average characterisation factors, the three characterisation methods of acidification produce results in which the contributions of NH₃, NO₂ and SO₂ to the acidification indicator do not differ much within each method when 1 kg of each acidifying substance is emitted. However, the NF methods cannot describe any spatial aspects of environmental problems. Both OT and AE methods show that the spatial aspects play an important role in the characterisation factors. The AE method results in greater differentiations between country-dependent characterisation factors than does the OT method. In addition, the results of the AE and OT methods differ from each other for individual countries. A major shortcoming of the OT approach is that it does not consider the sensitivity of the ecosystems onto which the pollutants are deposited, whereas the AE approach does. In the case of the AE method, a new atmospheric dispersion model, new information on emissions and critical loads have a different influence on the characterisation factors, depending on the country. The results of statistics show that the change in the atmospheric dispersion model has a greatest influence on the results, since ecosystem-specific depositions are taken into account for the first time.

The simple NF methods can be used in a first approximation to assess the impacts of acidification and terrestrial eutrophication in cases where we do not know where the emissions occur. The OT approach is a more advanced method compared with the NF method, but its capability to describe spatial aspects is limited. The AE factors are truly impact-oriented characterisation factors and the information used here represents the current best knowledge about the assessment practice of acidification and terrestrial eutrophication in Europe. The key message of this study is that there is no shortcut to achieving advanced characterisation of acidification and terrestrial eutrophication: an advanced methodology cannot develop without atmospheric dispersion models and information on ecosystem sensitivity.