W10 AM Life Cycle Approaches to Water Resources
Wednesday, 16 November 2005: 8:00 AM - 11:40 AM in 341-342

(MAR-1118-677907) World regional characterization factors for toxic emissions: does it make a difference?

Margni, M.^{1, 2}, Gigante, F.¹, Jolliet, O.¹, ¹ EPFL, Lausanne, Switzerland² CIRAIG, Montreal, Canada

ABSTRACT- One of the common criticisms to life cycle impact assessment methods is linked to the fact that characterization factors are usually calculated for a given geographical context, such as Western Europe or North America. The straightforward adoption of such factors to assess emission occurring in other continental regions could lead to misleading results, especially for impacts categories with a strong regional variation. This paper presents a fist attempt extending a spatial fate and exposure modeling approach at a global scale analyzing the variation of characterization factors developed for 13 different geographical regions, incl. Europe, Africa (2x), Asia (2x) , Oceania (2x), North America, Central America, South America (2x), as well as 8 oceanic zones. The multimedia fate and exposure model IMPACT 2002 is used. The model accounts for relationships between the location of food production and drinking water extraction, as well as where population cohorts live relative to where chemical emissions occur. Global databases provide reliable and consistent data among all continents, which allowed adapting the Western European IMPACT 2002 model to all other continents. The model considers chemical transport between the different spatial zones accounting for the potential impacts of an emission that leaves the systems. Results of a test set of organic chemicals are then cross-compared between the different emission locations. Despite important variations in characterization factors relative to which geographical location the pollutant is emitted (up to a factor of 100) this remains restricted compared the variations of the entire set (up to 9 orders of magnitude) and the ranking tends to remain constant. Variations linked to the geographical location are mainly due to population density, and also other specific continental parameters such as agricultural production.

Key words: Characterization factors, toxic chemicals assessment, spatial modeling