China, the world's largest developing country, faces a severe water shortage. As the government has set a goal of limiting water use to 7000 × 10⁸ m³ by 2035, how to control the increase in water use will be a thorny issue for China. Unbalanced and uncoordinated regional socio-economic development is an important feature of China. Research on the interaction between provincial water use will help to optimize the rational allocation of water resources and control of water use. In this paper, SNA (social network analysis) method is first used to explore the characteristics of social network relationship between inter-provincial water use, construct a two-stage model of SNA–LMDI, and decompose the driving factors of inter-provincial water use evolution. We found the following points. (1) From 2000 to 2018, the spatial correlation network structure of water use is tending to be stable, and the stability and risk resistance of the whole network are enhanced. (2) From different angles to quantify the centricity analysis, can be found that eastern provinces located right in the heart of water network, obviously larger impact on water resources utilization in other provinces, Shanghai and Beijing is located in the former two, and central and western provinces in the edge position. (3) The national water use spatial correlation network can be divided into four blocks, net beneficial block, bidirectional spillover block, brokers block, and net spillover block. (4) Technological progress and industrial structure adjustment were the primary and secondary factors inhibiting the increase of total water use, while income increase was the main factor promoting the increase of total water use, population scale expansion had a weak role in promoting the increase of total water use. Some policy implications are put forward related to our research conclusions.

China experienced a decline of water use intensity in the 11th Five Year Plan, but the water use intensity actually increased in 2009. To the best of our knowledge, the index decomposition analysis method was rarely used to analyze changes in water use, and no decomposition analysis has investigated the role of regional economy in the decline of water use intensity. In this paper, we use logarithmic mean Divisia index (LMDI) techniques to decompose the change of water use intensity in the period 2006–2010. We find that the change of industrial water use intensity is confirmed as the dominant contributor to the decline in the overall water use intensity; the regional structure effect and the industrial structure effect is positive to the decline of overall water use intensity; the decline of China’s water use intensity is mainly attributed to the effect of developed eastern provinces; meanwhile, the effect of central and undeveloped western is also positive to the decline of overall water use intensity; at least one out of three effects is positive to the decline of water use intensity in the different provinces; the intensity effect is positive and the industrial structure effect is positive to the declines of China’s water use intensity based on chaining approach except the period 2008–2009, individually; and the deviation of regional structure effect and industrial structure effect between with regional economy and without regional economy in LMDI is 0.9 and 2.3 m³/10⁴ RMB, respectively.