LAUSR

Abstract The water footprint theory has provided an effective approach for evaluating the utilization of freshwater resources in agricultural production. However, there are few studies have explored the dynamic coupling relationship between water footprint and socioeconomic factors, especially from the perspective of regional ecological features. Therefore, the water footprint method was used to investigate the dynamic evolution of agricultural water consumption from 2005 to 2015 in Zhangjiakou, an extremely water-scarce city which is divided into six ecological zones (I, II, III, IV, V, and VI). Then mathematical models such as the Gini coefficient were first employed to evaluate the spatiotemporal matching characteristics of agricultural water footprint and socioeconomic factors. The results show that: (1) The agricultural water footprint increased by 1.69 × 109 m3 in Zhangjiakou, of which the animal products water footprint increased by 1.59 × 109 m3, accounting for 94%. (2) Cereals had always been the major contributors to crop water footprint, with an increasing contribution rate from 49% to 54%. Milk and egg products were the main drivers for the increasing water footprint of animal products, with a total contribution rate increased from 46% to 55%. Meanwhile, the spatial differentiation was significant. The contribution rate of the cereal water footprint was less than 50% in counties of high-altitude ecological zones I, II and IV in 2015, while it was higher than 50% in counties of low-altitude ecological zones III, V and VI. (3) The per unit area water footprint in counties of the ecological zone III was much higher than the average level, while per capita water footprint and per unit output value water footprint were far below it, indicating agricultural structure optimization requires a comprehensive consideration of multiple socioeconomic factors. This study is expected to provide policy-makers with scientific guidance that is conducive to agricultural water conservation and ecological zone planning.