Life cycle assessment has been recognized as an important decision-making tool to improve the environmental performance of agricultural systems. Still, there are certain modelling issues related to the assessment of their impacts. The first is linked to the assessment of the metal terrestrial ecotoxicity impact, for which metal speciation in soil is disregarded. In fact, emissions of metals in agricultural systems contribute significantly to the ecotoxic impact, as do copper-based fungicides applied in viticulture to combat downy mildew. Another issue is linked to the ways in which the intrinsic geographical variability of agriculture resulting from the variation of management practices, soil properties, and climate is addressed. The aim of this study is to assess the spatial variability of the terrestrial ecotoxicity impact of copper-based fungicides applied in European vineyards, accounting for both geographical variability in terms of agricultural practice and copper speciation in soil. This first entails the development of regionalized characterization factors (CFs) for the copper used in viticulture and then the application of these CFs to a regionalized life-cycle inventory that considers different management practices, soil properties, and climates in different regions, namely Languedoc-Roussillon (France), Minho (Portugal), Tuscany (Italy), and Galicia (Spain). There are two modelling alternatives to determine metal speciation in terrestrial ecotoxicity: (a) empirical regression models; and (b) WHAM 6.0, the geochemical speciation model applied according to the soil properties of the Harmonized World Soil Database (HWSD). Both approaches were used to compute and compare regionalized CFs with each other and with current IMPACT 2002+ CF. The CFs were then aggregated at different spatial resolutions—global, Europe, country, and wine-growing region—to assess the uncertainty related to spatial variability at the different scales and applied in the regionalized case study. The global CF computed for copper terrestrial ecotoxicity is around 3.5 orders of magnitude lower than the one from IMPACT 2002+, demonstrating the impact of including metal speciation. For both methods, an increase in the spatial resolution of the CFs translated into a decrease in the spatial variability of the CFs. With the exception of the aggregated CF for Portugal (Minho) at the country level, all the aggregated CFs derived from empirical regression models are greater than the ones derived from the method based on WHAM 6.0 within a range of 0.2 to 1.2 orders of magnitude. Furthermore, CFs calculated with empirical regression models exhibited a greater spatial variability with respect to the CFs derived from WHAM 6.0. The ranking of the impact scores of the analyzed scenarios was mainly determined by the amount of copper applied in each wine-growing region. However, finer spatial resolutions led to an impact score with lower uncertainty.

Bioavailability of organic contaminants is a key parameter in estimating the fate and transport of contaminants in the environment. Its quantification is necessary for assessing the bioremediation potential of a contaminated site or for evaluating the hazard associated with the presence of the contaminant. However, there is a lack of methods that quantify biovailability under flow-through conditions. Therefore, a method was developed based on mass-balance analysis of contaminant breakthrough curves to quantify a bioavailability rate. The method was performed on an eight-column study where pentachlorophenol (PCP) was injected through saturated columns packed with a blend of sterile or inoculated sand and PCP-sorbing resin. Columns varied in sorption capacity, pore-water velocity and biodegradation activity. Breakthrough curves were modeled using a classic approach that includes advection-dispersion, equilibrium and first-order sorption, as well as biodegradation. Results showed that the bioavailability rates (non-dimensional numbers ranging from 0.02 for a sorptive column to 0.32 for a non-sorptive one) were directly related to the limiting kinetic mechanism. It was also demonstrated that the contaminant aqueous phase concentration does not adequately represent its bioavailability under flow-through conditions.