A novel integrated framework for water-energy-food system efficiency evaluation and multi-scenario simulation incorporating water footprint accounting

Traditional water–energy–food (WEF) efficiency assessments predominantly rely on water withdrawal statistics, which do not accurately represent the water consumption and pollution characteristics of production activities, potentially leading to a systematic underestimation of water scarcity risks and misdirection of resource management policies. This study introduces the water footprint method, utilizing blue water and green water as inputs and gray water as an undesirable output, to establish a refined assessment system. Combining GDIM decomposition and Monte Carlo simulation, this study develops three scenarios are developed to forecast the evolution of WEF efficiency in nine provinces of the Yellow River Basin from 2024 to 2035. The results show that: (1) From 2005 to Li et al., 2023, comprehensive efficiency increased from 0.61 to 0.95, but inter-provincial differentiation is significant, with Sichuan being the only province displaying “high efficiency-low pressure-high coordination.” (2) The blue water stress index is decoupled from efficiency; major grain-producing areas experience blue water overload without exhibiting low efficiency, indicating that the efficiency indicator cannot accurately reflect the relationship between absolute consumption and carrying capacity. (3) Scenario simulations reveal three response modes: constant efficiency type (Inner Mongolia, Sichuan), synchronous improvement type (Shanxi, Henan), and lagging type (Shaanxi). (4) Policy marginal benefits diminish; the general control scenario already achieves most water-saving benefits (e.g., a 29% reduction in agricultural water footprint in Shanxi), while the additional reduction under the enhanced control scenario narrows to 6%–12%. This study provides a scientific basis for refined WEF management in the Yellow River Basin.