The input–output data were collected by Agris Sardegna under the framework of the “CANOPAES” project at two experimental sites located in Sardinia . The first experimental site contains low levels of HM-contaminated soil, while the second experimental site contains high-level HM-contaminated soils . At both experimental sites, the “Futura 75” industrial hemp cultivar was sown at a rate of 40 kg ha− 1 and fertilized with 60 kg ha− 1 nitrogen , while the irrigation water was distributed at a rate of approximately 4,500 m3 ha− 1 based on the thermopluviometric trend and soil moisture content. The input–output data of the processes after the field gate were gathered from the most recent scientific literature and from the Ecoinvent database . As recommended by Siregar et al., to meet the data quality requirements, data verification must be conducted. The data adopted in this study were checked and verified according to the data quality parameters proposed by ISO 14044. Specifically, the data quality included the time coverage, geographical coverage, precision, completeness, representativeness, consistency, reproducibility, data sources, and information uncertainty. In addition, an uncertainty analysis was performed to address any parameter uncertainties and to test the robustness of the Life Cycle Impact Assessment . The Monte Carlo simulation method was conducted by using 1,000 Monte Carlo analysis runs. This analysis was carried out by collecting a set of uncertainties for the parameters that affected the variation of each scenario. Specifically, the uncertainties for industrial hemp cultivation, seed processing plants, anaerobic digestion plants, biomass-fired power plants and transportation were accounted for. Where available, uncertainty was explained by the results of this study , while for the other subsystems, greenhouse benches the mean and standard deviation were provided from the referenced sources or recalculated based on available published data.

The selected functional unit of this study consisted of 1 kg of dry matter industrial hemp product and 1 ha of phytoremediated area. An assessment of the cumulative energy demand and environmental impacts of each production phase, from raw material extraction, the manufacturing process, cultivation, transportation and biomass utilization, to the end of life, was analyzed in this study, and 4 different industrial hemp supply chain scenarios were designed . The system boundaries of the 4 scenarios were defined from cradle-to-grave. Scenario 1 – Industrial hemp supply chain scenario 1 includes, as shown in Fig. 1, industrial hemp cultivation and harvesting of the crop at the full ripening stage of the seeds , with two main products obtained, namely, industrial hemp seeds and residual straw. These two products share the same on field activities; thus, the energy and environmental burdens were distributed between these products by following the economic allocation method. After transportation, the industrial hemp seeds were treated in a processing plant to obtain several final products , while the industrial hemp straw was pressed and delivered to an anaerobic digestion plant for biogas production. The biogas produced was used to generate electricity and heat, where a portion of the heat was used to dewater the digestate. Next, the dry digestate was transported to a biomass power plant to produce electricity for the grid. Scenario 2 – Fig. 2 shows industrial hemp supply chain scenario 2 , which consists of, as specified in HSC1, industrial hemp cultivation, management and harvesting of the seeds and straw, and the treatment of industrial hemp seeds in a processing plant. The industrial hemp seeds and residual straw share the same on-field activities; thus, the energy and environmental burdens were distributed between these products by following the economic allocation method. In this scenario, the straw bales from the field were transported directly to the biomass power plant for electricity generation. Scenario 3 – Industrial hemp supply chain scenario 3 includes industrial hemp cultivation and then mowing the dried plants to obtain a dried whole plant product . Industrial hemp straw bales, after transportation, were used as solid fuel in a biomass power plant to produce electricity for the grid. Scenario 4 – Industrial hemp supply chain scenario 4 includes, as shown in Fig. 4, industrial hemp cultivation and harvesting of the crop after the complete flowering stage by using a self-propelled forage harvester to obtain a chopped whole plant with a short cutting length .

The industrial hemp biomass was transported to an anaerobic processing plant where after an ensiling process, the product was used as a substrate for biogas production. The successive steps of this industrial hemp supply chain scenario were the same as those presented for HSC1. An LCI includes an extended set of inputs and outputs that are associated with the studied system. Analyses of the inventoried data were conducted by splitting the overall processes into 5 subsystems, as described below. 1- Industrial hemp field cultivation for the phytoremediation of HMcontaminated soil. A field-level dataset was obtained from on-site measurements that included the overall inputs , outputs and detailed descriptions of each cultivation activity for each designed scenario . Specifically, the information consisted of the field task conducted, the type and power of the agricultural machinery used and the time spent per activity, which were collected for industrial hemp cultivation. The equations applied to assess the requirements for the amount of diesel fuel are available in the associated dataset .Moreover, as suggested in other energy and environmental studies, the indirect energy requirements of capital goods for on-field activities were included to obtain a broader overview of the examined system. For these reasons, the embodied energy of the machinery and equipment that was used for the field operations was considered in this study. 2- Seed processing plants transform seeds into refined products. After the harvesting operations, the industrial hemp seeds were delivered to a processing plant for further treatment in scenarios HSC1 and HSC2. The industrial hemp seeds were immediately dried with air ventilation because improper moisture levels of the seeds may cause considerable problems in their conservation and in subsequent processing phases. The preprocessing activities consisted of sieving and ventilation by rotary sieves. Then, the industrial hemp seeds were processed in an expeller screw press to extract the oil and to obtain the cake byproduct. The cumulative energy demand and environmental burdens of industrial hemp seed processing were offset by the savings that were obtained from the replaced products. Industrial hemp oil and its byproducts are widely used in the nonfood sector to produce biofuels and other industrial products. The seed processing dataset for this work comes from the processing trials that were carried out and are still in progress at the experimental stations of Agris. 3- Anaerobic digestion plants for biogas production and energy conversion. The industrial hemp byproducts for scenario HSC1 and fresh industrial hemp biomass for scenario HSC4 were used in an anaerobic digestion plant for biogas production. Based on the available literature, it was assumed that the fresh industrial hemp biomass was ensiled at the ADP with a dry matter loss of 12%, while the industrial hemp straw bales were stored at the ADP.

Both the industrial hemp silage and industrial hemp straw bales were mechanically pretreated with a cross-flow grinder before feeding the anaerobic reactor, and an energy expenditure of 12 kWh t− 1 of feed stock was adopted. The quantity and quality of the biogas may vary based on the input material compositions and operational parameter settings . A specific methane yield of 92 m3 t− 1 of volatile solids was adopted for the industrial hemp straw residues under mesophilic conditions of VS for the industrial hemp silage was used. Moreover, it was assumed that 1.73% of the methane produced was lost as diffuse methane emissions in the combined heat and power unit. The produced biogas was converted into a CHP unit for electricity and heat generation; electricity was conveyed to the national grid, while heat was used for dewatering the digestate. According to the national directive, the agronomic use of digestate, which is produced with the addition of biomass derived from contaminated areas, is not allowed, while the energy valorization of the digestate is encouraged after a rigorous dewatering process. Other studies have indicated that dewatering digestates improves their disposal , but the water content of the digestate should be reduced to 50–60%. Several technologies are available to reduce the water content of digestates obtained from anaerobic digestion , such as a belt conveyor dryer. In this study, the heat from the CHP unit is used to dewater the digestate, since only a minor proportion of the heat produced from the CHP unit is used to maintain the optimal thermal conditions of the anaerobic reactor, while the remaining heat proportion is generally not exploited in commercial ADPs due to their locations in rural areas and because there are only a few potential heat consumers for bulk consumption. For these reasons, in this study, the surplus heat that was produced from the CHP unit was released as waste heat into the atmosphere. The production of capital goods was excluded from the inventory assessment of this study since the capital equipment of ADPs is commonly not considered due to their long life spans and low environmental impacts. 4- Biomass-fired power plant for industrial hemp biomass incineration and energy generation. The dewatered digestates from scenarios HSC1 and HSC4, the industrial hemp straw bales from scenario HSC2 and the dried whole plant bales from scenario HSC3 were incinerated in a biomass-fired power plant for electricity generation. The exploitation of the waste heat that was derived from the BFPP was not considered in this study since reliable energy and emissions factors were not available.

Additionally, growers equipment the limited availability of national district heating structures causes the exploitation of waste heat to be difficult to pursue. Accordingly, the BFPP dataset was compiled from the available scientific literature. The BFPP was assumed to have a life span of 20 years and a power generation efficiency of 20%. In addition, due to the nonnegligible capital goods burdens, the cumulative energy demand and carbon emission impacts of BFPP upstream manufacturing were 0.35 MJ kWh− 1 and 0.036 kg CO2e kWh− 1 , respectively. The dewatered digestate and industrial hemp straw bales were stocked at the BFPP. The industrial hemp straw bales were mechanically pretreated with a cross-flow grinder before incineration, with an energy expenditure of 12 kWh t− 1 of biomass. The specific lower heating values adopted in the hemp supply chain scenarios are available in the associated dataset . In this study, the BFPP employed a mixture of biomass in the furnaces; thus, the suitability of ash as a mineral soil amendment is unknown and requires further investigation. In fact, Cavalcanti et al. and Ferreira et al. found that the contribution of the final disposal of biomass ash was not significant or was excluded from the study boundaries. For these reasons, the disposal of biomass ash was considered to be outside the system boundaries. This aspect represents a limitation of the study. 5- Transportation phases. The transportation of the products among the different processing sites was included in the system boundaries of this study. The transportation data were obtained from the Ecoinvent 3 dataset and from onsite information. Specifically, the industrial hemp seeds and industrial hemp straw bales were transported from the field gate to the seed processing plant and to the ADP , respectively, and 16–32-t Euro 5 trucks for road transportation were used. The fresh industrial hemp biomass that was harvested in scenario HSC4 was transported to the ADP by tractors . The dewatered digestates were delivered from the ADP to the BFPP in scenarios HSC1 and HSC4 with 16–32-t Euro 5 trucks for road transportation. These trucks were also used in scenarios HSC2 and HSC3 for industrial hemp bale transportation from the field gate to the BFPP . When addressing the cumulative energy demand and environmental impacts of the field subsystem inputs, the overall scores were reported based on the amount of product that was harvested; thus, lower production yields had greater impacts. Moreover, the working times that were related to on-field activities were likely influenced by the biomass yields. In fact, the harvesting operations in S1 were slightly greater than the activities carried out in the S2 experimental field.