Recent research has emphasized the need to adapt life cycle inventories and life cycle impact assessments to account for changes in future scenarios. This is mostly achieved by using Integrated Assessment Models (IAMs), which combine environmental and economic data to develop prospective life cycle inventories (LCIs). However, running complex IAMs to simulate scenarios is computationally expensive, and the resulting data is not always aligned with the geographical scope of background inventories. This study explores the potential of machine learning (ML) to create prospective water-scarcity characterisation factors by forecasting the AWARE factor for ~ 9700 watersheds globally. Historical time series of water-scarcity characterisation factors are generated using the global freshwater model WaterGAP v2.2d and the AWARE method. Several ML models are trained on these historical datasets and benchmarked using symmetric mean absolute percentage error (sMAPE). The best-performing model, N-Beats, is then used to forecast AWARE values through 2032. The results demonstrate that ML can produce spatially and temporally explicit forecasts with reasonable accuracy (median sMAPE ~ 28%). However, the models primarily capture seasonal patterns rather than long-term structural trends and the results are sensitive to the quality and representativeness of the training data. This study highlights both the potential and the limitations of time series forecasting for developing prospective characterisation factors in life cycle assessment. ... Read more

Purpose
This work aims to provide insights on the application of social life cycle assessment (S-LCA) in evaluating the social impacts associated with municipal wastewater treatment (WWT). The study assesses the social risks and social performance of two municipal WWT systems in Catalonia, Spain: a conventional wastewater treatment plant (WWTP) (Reference System) and a novel system that recovers water and other valuable resources (Novel System).

Methods
S-LCA was conducted at Generic and Site-Specific levels using 1 m3 of wastewater treatment as the functional unit (FU). The Generic assessment was conducted via the Product Social Impact Life Cycle Assessment (PSILCA) database, while the Site-Specific assessment employed the Subcategory Assessment Method (SAM) with four-level reference scales to assess the social performance of the WWTP operator and its first-tier suppliers. Furthermore, activity variables were calculated based on organizations’ shares in the total costs per FU, and the Novel System’s multifunctionality was solved through economic allocation. Results were aggregated by (i) assigning equal weights to organizations and (ii) factoring in organizations’ weights and the allocation factor, leading to results per FU.

Results and discussion
The Generic analysis results indicated that the Novel System entailed fewer social risks than the Reference System. Most social risks in both systems occurred in the subcategories “Access to material resources,” “Fair salary,” “Freedom of association and collective bargaining,” “Contribution to economic development,” and “Corruption.” In the Site-Specific assessment, the Novel System presented better social performance than the Reference System per 1 m3 of wastewater treatment. The latter’s performance per FU did not meet the basic requirement in four out of eleven subcategories, mainly due to the performance and weight of a chemical supplier. Allocation greatly benefitted the Novel System’s results per FU compared to the results obtained when equal weights were applied.

Conclusions
Activity variables were used to connect organizations’ conduct with particular WWT systems, and multifunctionality was solved. This approach allowed for obtaining results per FU at both assessment levels. However, social performance was also evaluated by calculating the average social performance of each system without considering activity variables and the FU, leading to different results. The social performance of the Novel System per FU was satisfactory across all subcategories but required improvement in four subcategories based on the average results. Given the limitations of using activity variables and allocation in S-LCA, further research is necessary to appropriately evaluate and compare the social effects of novel resource recovery systems. ... Read more