With population increases and the socio-economic development of Chinese society, the demand for and use of water in China is increasing rapidly. Water scarcity is one of the main problems for sustainable development, especially in arid areas in the northwest of China. Water saving is thus one of the main concerns of local and national policy makers. Analysis of the linkage effects of water resource use between economic sectors, and the movement of commodities and services using water is helpful in water conservation management, water use rationalization, and in strengthening water-saving practices. The Heihe River is the second largest inland river basin in the northwest of China, and water scarcity is the key problem for its sustainable development. The Zhangye city area, which is in the middle Heihe River Basin, is a pilot area for water saving. For this reason, it was chosen as the research site for this study. To produce commodities and to supply services an area needs water. Water-movements associated with the complex linkages among sectors are usually ignored, however, in favor of measures of direct consumption. Studying the movement of commodities and services can reveal the movement of water. In this study, an input-output resource analysis is integrated with an hypothetical extraction method, to uncover the in-depth characteristics of the inter-sectoral patterns of water movement. Commodity input-output tables and water input-output tables were developed for Zhangye for the years 2000, 2002, 2005, 2007, 2010 and 2012. Based on a mixed type input-output model of Zhangye, water resources for the period 2000-2012, a modified hypothetical extraction method was applied to identify indicators and analyze the linkage effects among six economic sectors of Zhangye in relation to water resource use. These sectors were the planting industry, livestock farming, other agriculture, industry, construction, and service sectors. The results showed that: (1) by comparing direct water consumption with vertically integrated sectoral water consumption, the direct water consumption of all sectors was not equal to the water required to meet final demand in any of them. However, in view of the state of the economy, total direct water consumption in all sectors was deemed equivalent to the total of water required to meet internal demand. With respect to the water resources transferred among productive sectors, the planting industry consumes the largest amount of direct water, and of vertically integrated water consumption. However, because vertically integrated consumption is less than direct water consumption, the planting industry is a real net output sector with respect to water resources in Zhangye; (2) by measuring the internal effects, mixed effects, net forward linkages and net backward linkages in the six sectors, compared with other sectors the internal effects and mixed effects in the planting industry were the largest of all. In contrast, the net backward linkages in the service industry were the largest determinant of inter-sectoral water consumption, which shows that this sector has the highest dependency on other sectors; (3) in terms of the water resources transferred among sectors, the largest water resource provider in Zhangye was the planting industry, and transferred a large amount of water to the service industry by intermediate inputs (the largest recipient of water). Conversely, the construction industry does not make net water transfers to other sectors, but receives a certain amount of net water inputs from them. Thus, the construction industry is a "pure" input sector; and (4) the internal effects, mixed effects, net forward linkages, and net backward linkages of the six sectors were analyzed in terms of the Zhangye economic system between 2000 and 2012. These data show the dynamic linkages relating to water use among industrial sectors, and indicates clearly that the key sector in terms of water use in the Zhangye economic system is the planting industry. This paper argues that the linkage effects of water resource movements among different sectors actually stem from the demand that transfers from one sector to another. Using a modified hypothetical extraction method, and analyzing direct consumption and decomposing the components of water resources, this study could more directly and clearly quantify the linkage relationships found between and within each industrial sector in relation to water resources. This in turn revealed the regularity of industrial water consumption patterns, and therefore may provide a new perspective on structural adjustment, and a more scientific basis for policy-making.