Integrer la speciation des metaux en ecotoxicite terrestre pour l'analyse du cycle de vie: Le cas du zinc

The goal of this project is to generate new regionalized characterization factors (CFs) for zinc (Zn) that takes into account speciation for terrestrial ecotoxicity. Zinc has been chosen in this project notably because it is one of the main contributors to the ecological footprint of Canada and the world for terrestrial ecotoxicity and also because of the availability of field data on its speciation in soil.;1. Determine the bioavailable fraction of Zn at global scale with limited data on soil properties. The first specific goal of this project aimed at determining the bioavailability of Zn in soils at global scale using only the physicochemical properties of soil available in world soil databases. In LCA, metal emissions can occur everywhere in the world and it is necessary to use a coherent method for the world. This goal has led to the writing of the first manuscript entitled "Assessing the variability of the bioavailable fraction of zinc at the global scale using geochemical modeling and soil archetypes" and published by The International Journal of Life Cycle Assessment (chapter 3) [1]. The bioavailable fraction of Zn is determined for soils of the world by using the geochemical speciation model WHAM 6.0 (Windermere humic aqueous model) and the Harmonized world soil database (HWSD). The first step was to validate with field data the applicability of the WHAM 6.0 model for soils when using as input data only the following soil parameters available at global scale: soil pH, cationic exchange capacity (CEC), organic matter (OM) and carbonate contents, and soil texture. This method has been compared with the use of empirical regressions.;2. Calculate regionalized CFs for Zn that take speciation into account. Following this step, BFs were integrated in the determination of CFs for Zn for terrestrial ecotoxicity and the regionalization level required to adequately represent the spatial variability of CFs has been defined. This led to the writing of the second manuscript entitled "Characterization factors for zinc terrestrial ecotoxicity including speciation" submitted to The International Journal of Life Cycle Assessment (chapter 4). First of all, fate factors (FF) were determined using the USEtox model and specific partition coefficients (Kd) obtained with WHAM 6.0 model. Then effect factors (EF) were calculated using the AMI method (Assessment of mean impact) with terrestrial ecotoxicological data (EC50) available in ecotoxicological databases. To do so, various scenarios have been tested in order to be as coherent as possible with bioavailability factors (BFs) despite the lack of precision of ecotoxicological data, especially regarding the soils used for the ecotoxicological tests. The definition of soil archetypes established in the previous step has been refined in order to better represent the CF spatial variability and to obtain a spatial resolution more manageable for LCA. In order to consider also the cases in which the emission sites are not known, like it happens frequently in LCA, generic CF values were obtained using the population density as an emission proxy for Zn. Results are then presented in a map of the world in order to facilitate their use.;3. Determine the importance of including Zn speciation in terrestrial ecotoxicity in LCA. The third specific goal of this project was to determine the importance of integrating Zn speciation in the CF definition for terrestrial ecotoxicity by carrying out a case study where Zn is one of the major contributors to the terrestrial ecotoxicological impacts. This led to the writing of the third manuscript entitled "Case study: Does taking zinc speciation into account in terrestrial ecotoxicity make a difference for LCA results?" and submitted to The Journal of Cleaner Production (chapter 5). A case study on the ecoinvent "market for electricity, low voltage CA-QC" data has been carried out using the IMPACT 2002+ methodology. Different CFs for Zn were tested in this study (IMPACT 2002 CF and USEtox derived terrestrial CF without speciation; generic default CF values for the world including speciation based on either soluble or true solution Zn as well as minimum and maximum regionalized CF values including speciation). The importance of including speciation has been determined by comparing, for the different options, 1) the contribution of Zn emitted to soil to terrestrial ecotoxicity impact category and to ecosystem quality damage category impact scores (PDF.m2.yr) (PDF: potentially disappeared fraction of species), 2) the rank of Zn emitted to soil as a contributor to terrestrial ecotoxicity impact category and 3) the total terrestrial ecotoxicity and total ecosystem quality impact scores. (Abstract shortened by UMI.).