Exploring the Water–Energy Nexus in Italian Wineries: An Integrated Analysis of Resource-Use Associations

The wine sector is increasingly required to improve its environmental performance by reducing the consumption of key resources such as water and energy. However, these two dimensions are often analyzed separately, even though winery operations may generate interdependent demands for both resources. This study explores the water–energy nexus in the wine sector by investigating the relationships among annual water consumption, electrical energy use, outsourcing intensity and production configuration in a sample of 21 Italian wineries. A dedicated dataset was developed from company-level operational data, including winery reference models, outsourcing and packaging indices, production volumes, electricity consumption, water consumption and specific performance indicators. The analysis combined descriptive statistics on winery-average data with Pearson correlation analysis, exploratory pooled multiple regression, Principal Component Analysis and a log-linear mixed-effects model applied to 74 winery-year observations nested within 21 wineries. Pearson correlation analysis showed strong positive associations between electrical energy consumption and water consumption (r = 0.891), between wine outsourcing index (WOI) and electrical energy consumption (r = 0.796), and between WOI and water consumption (r = 0.780). The complete within–between mixed-effects model showed that a 1% higher-than-usual electricity use within a winery was associated with 0.595% higher annual water use (95% CI: 0.418–0.772; p < 0.001) after accounting for within-winery variation in annual production. In the conventional mixed-effects model, the corresponding electricity coefficient was 0.571 (95% CI: 0.406–0.735; p < 0.001), conditional on production and the other covariates. Given the strong collinearity between electricity and production (r = 0.955; VIFs = 11.6 and 13.2), the 0.571 estimate should not be interpreted as an independent electricity effect. These findings indicate that aggregate water and electricity consumption covary strongly across the sampled wineries and that this relationship persists after accounting for repeated winery observations and annual production volume. WOI and production configuration contribute to the structural interpretation of resource demand, although their conditional associations are less stable. The results support the joint assessment of absolute resource demand and specific performance indicators for more context-sensitive winery benchmarking and managerial decision making.