Cities are hotspots of commodity consumption, with implications for both local and systemic water resources. Water flows “virtually” into and out of cities through the extensive cross-boundary exchange of goods and services. Both virtual and real water flows are affected by water supply investments and urban planning decisions, which influence residential, commercial, and industrial development. This form of water “teleconnection” is being increasingly recognized as an important aspect of water decision-making. The role of trade and virtual water flows as an alternative to expanding a city’s “real” water supply is rarely acknowledged, with an emphasis placed instead on monotonic expansion of engineering potable water supplies. We perform a literature review of water footprint studies to evaluate the potential and importance of taking virtual flows into account in urban planning and policy. We compare and contrast current methods to assess virtual water flows. We also identify and discuss priorities for future research in urban water footprint analysis.

Research has yet to operationalize water footprint information for urban water policy and planning to reduce vulnerability and increase resilience to water scarcity. Using a county‐level database of the U.S. hydro‐economy, NWED, we spatially mapped and analyzed the Water Footprint of Flagstaff, Arizona, a small city. Virtual water inflow and outflow networks were developed using the flow of commodities into and out of the city. The power law distribution of virtual water trade volume between Flagstaff and its county trading partners broke at a spatial distance of roughly 2000 km. Most large trading partners are within this geographical distance, and this distance is an objective definition for Flagstaff's zone of indirect hydro‐economic influence—that is, its water resource hinterland. Metrics were developed to measure Flagstaff's reliance on virtual water resources, versus direct use of local physical water resources. Flagstaff's reliance on external water supplies via virtual water trade increases both its hydro‐economic resilience and vulnerability to water scarcity. These methods empower city managers to operationalize the city's Water Footprint information to reduce vulnerability, increase resilience, and optimally balance the allocation of local physical water supplies with the outsourcing of some water uses via the virtual water supply chain.

In complex coupled natural‐human systems (CNH), multitype networks link social, environmental, and economic systems with flows of matter, energy, information, and value. Embedded Resource Accounting (ERA) is a systems analysis framework that includes the indirect connections of a multitype CNH network. ERA is conditioned on perceived system boundaries, which may vary according to the accountant's point of view. Both direct and indirect impacts are implicit whenever two subnetworks interact in such a system; the ratio of two subnetworks' impacts is the embedded intensity. For trade in the services of water, this is understood as the indirect component of a water footprint, and as “virtual water” trade. ERA is a generalization of input‐output, footprint, and substance flow methods, and is a type of life cycle analysis. This paper presents results for the water and electrical energy system in the western U.S. This system is dominated by California, which outsources the majority of its water footprint of electrical energy. Electricity trade increases total water consumption for electricity production in the western U.S. by 15% and shifts water use to water‐stressed Colorado River Basin States. A systemic underaccounting for water footprints occurs because state‐level processes discount a portion of the water footprint occurring outside of the state boundary.

This paper quantifies and maps a spatially detailed and economically complete blue water footprint for the United States, utilizing the National Water Economy Database version 1.1 (NWED). NWED utilizes multiple mesoscale (county-level) federal data resources from the United States Geological Survey (USGS), the United States Department of Agriculture (USDA), the US Energy Information Administration (EIA), the US Department of Transportation (USDOT), the US Department of Energy (USDOE), and the US Bureau of Labor Statistics (BLS) to quantify water use, economic trade, and commodity flows to construct this water footprint. Results corroborate previous studies in both the magnitude of the US water footprint ( F ) and in the observed pattern of virtual water flows. Four virtual water accounting scenarios were developed with minimum (Min), median (Med), and maximum (Max) consumptive use scenarios and a withdrawal-based scenario. The median water footprint ( F _CUMed ) of the US is 181 966 Mm ³ ( F _Withdrawal : 400 844 Mm ³ ; F _CUMax : 222 144 Mm ³ ; F _CUMin : 61 117 Mm ³ ) and the median per capita water footprint (FCUMed′) of the US is 589 m ³ per capita (FWithdrawal′: 1298 m ³ per capita; FCUMax′: 720 m ³ per capita; FCUMin′: 198 m ³ per capita). The US hydroeconomic network is centered on cities. Approximately 58 % of US water consumption is for direct and indirect use by cities. Further, the water footprint of agriculture and livestock is 93 % of the total US blue water footprint, and is dominated by irrigated agriculture in the western US. The water footprint of the industrial, domestic, and power economic sectors is centered on population centers, while the water footprint of the mining sector is highly dependent on the location of mineral resources. Owing to uncertainty in consumptive use coefficients alone, the mesoscale blue water footprint uncertainty ranges from 63 to over 99 % depending on location. Harmonized region-specific, economic-sector-specific consumption coefficients are necessary to reduce water footprint uncertainties and to better understand the human economy's water use impact on the hydrosphere.

Through the trade of products and services, cities indirectly depend on distant water sources to function, prosper, and grow. To fully account for indirect (virtual) water dependencies, virtual water flows need to be known along complex supply chains. To this purpose, we build a new environmental multiregional input–output model for U.S. regions. The model is used to quantify the domestic, blue virtual water flows and analyze the water footprints of 69 major U.S. cities. Our results show a large inequality in the urban water consumed for economic production: just 7 out of the 69 cities included in this study account for 35% of the U.S. national water footprint of production. This is due to the production of water‐intensive agricultural products in the metropolitan areas of western cities. The inequality reduces for the urban water footprint of consumption because, through the supply chains of industrialized food sectors, western virtual water is partially transferred to eastern cities as final demand. The water embodied in industrial products and services tends to be higher in western cities than in eastern cities; that is, the water embodied in food services could be several times higher in Los Angeles than in New York City. Trade hub cities attract large inflows of products which are mostly transformed for consumption elsewhere. Thus, the omission of product interdependencies within trade hub cities can increase by several times their water footprints of consumption. Overall, the proposed model is able to enhance subnational estimates of U.S. virtual water flows.