*2.3. Life Cycle Analysis*

There is not one strictly defined methodology for conducting LCA analyses for agricultural production. However, according to the principles set out in ISO PN-EN 14040 [22], the full analysis should include the four phases of PN-EN ISO 14040 (goal and scope definition), PN-EN ISO 14041 (LCI-life cycle inventory), PN-EN ISO 14042 (LCIA-life cycle impact assessment), and PN-EN ISO 14043 (life cycle interpretation). The LCIA is the phase in which the environmental impact assessment (e.g., land use) of the products applies. Data obtained in the previous LCI phase were transformed into impact category indicators. This was done by selecting the impact category and impact indicators, assigning LCI results, and calculating the category indicator values. According to the methodology developed by the Society of Environmental and Chemical Sciences (SETAC), 14 environmental impact categories were taken into account in LCA, of which three were considered in this study, namely, global warming potential (GWP), acidification, and eutrophication.

LCA was done by applying SimaPro 8.4 software [23]. To calculate the emissions of inputs production for the Italian agroforestry systems, the Ecoinvent database 3.3 was used. Dinitrogen monoxide (N2O), ammonia (NH3), and nitric oxide (NO) were modelled based on methodology described in [24]. N2O follows IPPC guidelines [25] Tier 1 for animal production and Tier 1 for crop production. NH3 calculations were based on emission factors for NH3, based on application of mineral N fertilizer and as a function of soil pH. NO is relatively of low importance compared to other sources, for that reason, simple emission factors were used [26]. Carbon dioxide (CO2) emissions after urea or lime application were calculated based on the factor (1.57 kg CO2/kg Urea-N<sup>10</sup> for urea and 44 kg CO2/kg limestone or 48 kg CO2/kg dolomite [26]. Irrigation quantity was calculated based on Methodological Guidelines for the Life Cycle Inventory of Agricultural Product [26] as consumed water for yield production (m<sup>3</sup> <sup>t</sup>−1). The emissions related to pesticide use were not included due to low influence on calculated environmental impacts. The impacts of GWP, acidification, and eutrophication were calculated using the CML method [27]. The system boundary was cradle to olive farm gate, i.e., from the extraction of raw materials to the farm gate until the olives were harvested. For the Danish farm, the direct NO2-N and indirect NO2 emissions were calculated based on IPCC 2006 methodology [25], while the ReCiPe method [28] was applied to calculate the potential of GWP, acidification, and eutrophication. Due to the agroforestry systems producing different crops, the yields are not directly comparable. Hence, the yields were converted to monetary values based on the prices indicated: 0.49 \$ kg−<sup>1</sup> wheat, 0.45 \$ kg−<sup>1</sup> barley, 0.16 \$ kg−<sup>1</sup> clover, and 0.14 \$ kg−<sup>1</sup> woodchips [19].
