7
A Vision for the NWUIP

The National Water-Use Information Program (NWUIP) currently focuses on the collection, estimation, and management of county-level water use data. The committee found significant opportunities to advance the techniques and technology used to estimate and manage national water use data. There is a compelling need for unbiased, science-based national water use information—information that will only become more vital for management and policy decisions in the future. Together, these findings led the committee to frame a broad vision for the NWUIP.

Long-term planning and management decisions need water use information that shows how water is an essential economic commodity and a vital natural resource. To use water use information in investigations of water resources or ecosystems requires placing water use prominently within the hydrologic cycle. The NWUIP should, therefore, be viewed as much more than a data collection and database management program focused on county-level categorical water use. Rather, it should become more integrated with the other efforts of the U.S. Geological Survey (USGS) to provide unbiased water resources information for assessing impacts and sustainability of current and future water use.

REQUIREMENTS FOR ANY NATIONAL PROGRAM OF WATER USE ESTIMATION

First and foremost, the NWUIP needs to obtain estimates of water use that meet unambiguous and meaningful objectives. Setting clear objectives is essential to selecting the estimation methods at the proper resolution and scale. Although many possible objectives can be chosen, most of these should be at the national level.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 115
Estimating Water Use in the United States: A New Paradigm for the National Water-Use Information Program 7 A Vision for the NWUIP The National Water-Use Information Program (NWUIP) currently focuses on the collection, estimation, and management of county-level water use data. The committee found significant opportunities to advance the techniques and technology used to estimate and manage national water use data. There is a compelling need for unbiased, science-based national water use information—information that will only become more vital for management and policy decisions in the future. Together, these findings led the committee to frame a broad vision for the NWUIP. Long-term planning and management decisions need water use information that shows how water is an essential economic commodity and a vital natural resource. To use water use information in investigations of water resources or ecosystems requires placing water use prominently within the hydrologic cycle. The NWUIP should, therefore, be viewed as much more than a data collection and database management program focused on county-level categorical water use. Rather, it should become more integrated with the other efforts of the U.S. Geological Survey (USGS) to provide unbiased water resources information for assessing impacts and sustainability of current and future water use. REQUIREMENTS FOR ANY NATIONAL PROGRAM OF WATER USE ESTIMATION First and foremost, the NWUIP needs to obtain estimates of water use that meet unambiguous and meaningful objectives. Setting clear objectives is essential to selecting the estimation methods at the proper resolution and scale. Although many possible objectives can be chosen, most of these should be at the national level.

OCR for page 115
Estimating Water Use in the United States: A New Paradigm for the National Water-Use Information Program The following are proposed as minimum requirements for any national program for accurate water use estimation: Water use estimates should be consistent across the country and should support meaningful comparisons of patterns and trends across geographic, climatic, and political boundaries. Estimates should be linked to the surface and groundwater resources affected by withdrawals. Linking water use to the resource emphasizes the importance of such water use categories as consumptive use, recharge, and return flows in maintaining water resources. Estimates should be consistent with the categories and spatial structure of previous national estimates, enabling the analysis of regional and national water use trends. Fundamental changes in the behavior and technologies that determine water use patterns should be detected and quantified. These changes (in, for example, water use efficiency, application rates, sources, consumptive use, etc.) may be caused by institutional changes (such as the legal need to exercise water rights), macroeconomic changes, or hydroclimatic changes. This suggests that before each national assessment is done, methods have to be developed to screen or “pre-sample” in order to identify major changes in water use and to refine sampling and estimation methods. Statistical sampling and inference should be done within the context of an unambiguous and statistically meaningful definition of the populations of water users to be estimated. Only by this kind of analysis can a hypothesis-based framework for estimation, error analysis, and the analytical determination of sample size requirements be meaningfully done. Estimation and data collection should incorporate error analysis designed to give relative standard errors (and thus confidence limits) on all quantities, including estimated changes from prior estimates. Error analysis should distinguish different types and sources of error (e.g., sample vs. nonsample error, nonresponse bias, etc.). Estimates should account for “intermittent” changes in water use (e.g., irrigation in the humid East only in dry years; the exercise of junior water rights in the West only in wet years). Estimates should account for interbasin transfers (e.g., Colorado River diversions to southern California) and for transfers between surface and groundwater sources (e.g., aquifer storage and recovery) over multiple scales. GOALS OF THE NWUIP By meeting these requirements, the NWUIP would address two complementary goals: (1) understanding spatial water use, and (2) understanding hydrologic water use. Spatial water use refers to estimates of the total use of surface and

OCR for page 115
Estimating Water Use in the United States: A New Paradigm for the National Water-Use Information Program groundwater, by category, within any designated area. The area may be defined by political or hydrologic boundaries. Understanding spatial water use requires knowledge of human behavior and consumption patterns and decisions. It directly supports information generation associated with water use as a material flow in regional and national economic activities. Hydrologic water use refers to water use as the human component of the hydrologic cycle and emphasizes the impact of humans on the natural resource. Understanding hydrologic water use requires determining the magnitude, location, and timing of water withdrawals, and determining consumptive use and return flows affecting a designated resource. Although consumptive use and return flows are essential for understanding the water budget, these water uses are often calculated as the residual in a hydrologic budget and are not routinely obtained from site-specific water withdrawal records. The NWUIP can and should characterize and quantify the importance of water as an essential commodity in the economy and the use of water as a critical hydrologic stress affecting the sustainability of the nation’s water resources. This requires applied research and techniques development in both estimation and sampling techniques as well as science-based research on the determinants and impacts of water use behavior. Such a vision for the NWUIP leads to the USGS fully integrating water use with process-based science in its hydrologic and water resources investigations. CONCEPTUAL FRAMEWORK FOR THE NWUIP A conceptual framework for the NWUIP that addresses the goals identified above is illustrated in Figure 7.1. As suggested in the figure and in this report, a future NWUIP would be supported by a foundation of water use data, water use estimation, and water use science. Water use data refers to measurements or estimates of the amount of water use at a site or for a region. These data would include direct measurement, as well as estimates obtained from surrogate measurements of activities involving water use. The term also refers to national geospatial databases and data inventories on water resources and water facilities, which support water use estimation. Water use estimation refers to statistical sampling, inference, and estimation techniques for estimating spatial water use. An array of estimation techniques will be needed for the NWUIP, since water use estimation is inherently tied to the water use data available for individual states and the nation, and the goals of the NWUIP. With the growing water use data that are available, there are significant opportunities for advancing water use estimation. Water use science refers to the hypothesis-driven investigation of the behavior and phenomena that determine spatial and temporal patterns of water use. This science will directly contribute to the development of techniques that support improved water use estimation. Water use science, discussed further in Chapter 8, also includes scientific assessment of the sustainability and impacts of

OCR for page 115
Estimating Water Use in the United States: A New Paradigm for the National Water-Use Information Program FIGURE 7.1 Conceptual framework for the National Water-Use Information Program. water use on aquatic ecosystems, on the hydrologic cycle, and on the reliability and vulnerability of the nation’s water resources. The goals of understanding spatial water use and hydrologic water use require a view of water use from the perspective of both the infrastructure water system and the natural water system. The infrastructure water system is described by locations of water withdrawals and the movement of water through the landscape in constructed water systems to users and eventually to points of discharge. The natural water system is described by streams, rivers, lakes, aquifers, and watersheds, which coexist with the infrastructure water system. Exchanges of water occur between the two systems, primarily at the points of water withdrawal and discharge. The conceptual framework envisages two major applications of information and data provided by the NWUIP:

OCR for page 115
Estimating Water Use in the United States: A New Paradigm for the National Water-Use Information Program Water use reports, incorporating the results of spatial water use estimation, currently being done by the USGS in its five-year national summaries of water use, and also incorporating interpretive summaries of water use data prepared at the state and regional levels. Water resource assessments, involving hydrologic water use analysis, where the availability of water from surface and groundwater resources is balanced against how these waters are being used. The sections that follow elucidate some of the major features on the conceptual framework for the NWUIP. THE NATURAL AND INFRASTRUCTURE WATER SYSTEMS Site-specific water use data (represented as points and/or polygons) can be viewed as data points in a spatial network, wherein each water use is associated with a node, and connections to other water use nodes are described by a set of directed links. Such a link-node representation gives a sense of direction to the flow of water use, just as the National Hydrography Dataset defines upstream-downstream relationships in the stream network. A link-node data structure can be used in a spatial database, defining all of the upstream uses that affect each water use node, as well as all of the downstream uses that may be affected by every water use. A network data structure for water use provides additional information, linking the impacts of water use in both the natural and the built environments. Links to the natural water system directly connect water use to surface and groundwater sources by withdrawals, discharges, recharge, instream flow, and ecosystem impacts. Links to the human environment connect water use to the infrastructure water system of pipes, canals, treatment plants, and other man-made structures that extract, convey, distribute, and collect water. The natural water system, at least with respect to surface water, is convergent; natural stream and river systems form dendritic networks draining many sources to a few sinks. The infrastructure water system can be divergent; a single source such as a reservoir may supply many points of water use distributed over the landscape. Conceptually, a link-node data structure for water use integrates spatially referenced water use in the natural water system and the infrastructure water system. A region in central New Hampshire near Lake Winnipesaukee is shown in Figure 7.2, for which the link-node water use data were obtained from the New England Water-Use Data System (NEWUDS) (Horne, 2001). Figure 7.2(a) shows the natural water system for this region, composed of Lake Winnipesaukee and its neighboring lakes, the streams and watersheds of this region, and groundwater aquifers (not shown). Superimposed on the natural water system are the points of withdrawal of water from groundwater and surface water, and a point of discharge of water back to the natural water system.

OCR for page 115
Estimating Water Use in the United States: A New Paradigm for the National Water-Use Information Program FIGURE 7.2 (a) Natural water system near Lake Winnipesaukee in central New Hampshire, (b) infrastructure water system for the towns in this region, and (c) combination of the natural and infrastructure water systems.

OCR for page 115
Estimating Water Use in the United States: A New Paradigm for the National Water-Use Information Program Figure 7.2(b) shows the infrastructure water system, composed of areas defining the legal boundaries of the towns, nodes at the centroids of these areas, and links symbolizing water movement within the infrastructure water system. The linkages are shown by solid lines for the water supply and distribution systems and by dashed lines for the wastewater collection and disposal systems. Although there are many points of water withdrawal, the local concern about wastewater discharge to lakes led to the construction of a regional wastewater treatment plant near the most downstream town, Franklin, from which the treated wastewater is discharged to the Merrimack River. Figure 7.2(c) shows the overlay of the natural and infrastructure water systems, from which the complexity of the interaction of these systems is readily apparent. Link-node data structures retain all the attributes of a site-specific database, and they incorporate additional information showing how water use connects to the natural and infrastructure water systems. This information might include estimates of the leakage from the distribution system to the resource, infiltration from the resource into the wastewater collection system, consumptive use, and other water fluxes between the natural and infrastructure water systems. A network presentation of water use with natural hydrologic water movement enables mass balance calculations to be done for both systems. Indeed, an approximate mass balance inherently supports the hypothesis that the quality of data used to estimate water use (and other hydrologic fluxes) is high.

OCR for page 115
Estimating Water Use in the United States: A New Paradigm for the National Water-Use Information Program Linking each water use node to upstream influences and downstream effects provides a direct framework for examining policy and management questions, such as the impacts of water use on instream flow or the identification of source areas and vulnerable uses affected by a contaminant spill. Moreover, the network representation powerfully supports modeling and assessment of policy options for complex water-resource systems. Network flow models are robust tools used to plan and analyze extremely complex regional water-resource systems. For example, the U.S. Bureau of Reclamation (2000) used a network flow model to analyze the effects of diverting up to 1.4 million acre-feet from irrigation water use to instream flow needs in the Snake, Boise, and Payette River basins. Similarly, the CALVIN model (Howitt et al., 1999) is an extremely detailed network flow optimization model of the water resource system for the entire state of California. The model includes the statewide water system, surface and groundwater resources, storage and conveyance facilities, and agricultural, environmental, and urban water uses. It was developed to support operational planning and policy analysis of the economics and sustainability of the state’s future water uses and water supplies (Newlin et al., 2001). A link-node structure for water use data is a direct extension of a site-specific water use database. This conceptual framework for water use data, currently used in NEWUDS, can support the representation of water use in both the natural and infrastructure water systems. A link-node structure for water use data integrates the goals of spatial and hydrologic water use estimation conceived for the NWUIP, as suggested in Figure 7.1. The complementary characteristics of spatial and hydrologic water use are shown in Table 7.1. TABLE 7.1 Characteristics of Spatial and Hydrologic Water Use Spatial Water Use Hydrologic Water Use Estimates the quantity and type of water use in an arbitrarily defined area Estimates spatial and temporal impacts of water use in the hydrologic cycle Reflects water use as a material flow or as a commodity input to human activities Reflects water use as a hydrologic flux resulting from human activities Takes place principally within the infrastructure water system, but is connected to the hydrologic system through withdrawal, discharge, recharge, and return flows Takes place principally within the natural water system, but is connected to the infrastructure water system through withdrawal, discharge, recharge, and return flows NWUIP products describe quantity, trends, and the spatial and temporal distribution of water use, by water use category NWUIP products describe impacts on vulnerability and sustainability of the resource, aquatic ecosystems, and riparian vegetation, by hydroclimatic category

OCR for page 115
Estimating Water Use in the United States: A New Paradigm for the National Water-Use Information Program A HIERARCHICAL WATER USE DATA STRUCTURE The link-node data structure we suggest can also be visualized in the context of a data hierarchy depicted conceptually in Figure 7.3. At the lowest level, the focus is simply on water withdrawal site locations. Information from the federal databases described in Chapter 3 can be used to help develop an initial baseline water-withdrawal database for a state, if that information does not already exist. When combined with appropriately selected estimation methods, these water withdrawal data can support spatial estimation of total water use, such as that reported in the NWUIP’s current five-year national summary reports. At the second level in the hierarchy shown in Figure 7.3, water discharge points may be added to the water withdrawal points and linked by their location to the streams, rivers, lakes, and aquifers of the natural water system. If the amount of water withdrawn or discharged at each of these water use sites is known, this information may be sufficient to support regional studies of water availability. At the third level in the hierarchy, information is added describing the water infrastructure: the jurisdictional boundaries of the institutions that manage water use, and the locations of their water and wastewater treatment plants and water distribution and collection systems. The quantity and quality of water flowing FIGURE 7.3 A hierarchy of water use representations. The symbols used are the same as those used in Figure 7.2.

OCR for page 115
Estimating Water Use in the United States: A New Paradigm for the National Water-Use Information Program through these facilities may also be defined. A water use representation at this level can be used to create a comprehensive understanding of the movement of water through the landscape in both natural and infrastructure water systems, and to show how human use of water impacts the quantity, quality, and sustainability of water resource systems. This level of complexity may be especially appropriate for jurisdictions with complex water management and accounting concerns, such as aquifer storage and recovery, artificial recharge, water reuse, desalination, total maximum daily load (TMDL) issues, and/or interbasin transfers. Water use data, water use estimation, and water use science are the foundation for achieving the goals of understanding spatial and hydrologic water use within the natural and infrastructure water systems. The following sections outline these three components of a future NWUIP. Water Use Data The Arkansas water use program has an excellent database of permitted water withdrawal points maintained to support the state’s regulatory responsibilities. To varying degrees, similar management and regulatory needs have resulted in comparable programs in many of the states. Concurrently, the maturation of geographic information system (GIS) technology as a standard data-management tool has made it common for states to organize and manage their water use data as GIS databases. The status and data availability of state water use programs have been summarized by the USGS district water use specialists (see Chapter 2). The convergence of states’ regulatory needs and the use of GIS technology has resulted in the emergence of site-specific water use databases. Many of these databases, including the Arkansas database with nearly 45,000 permitted water withdrawal points used throughout this report, are easily accessed and manipulated with over-the-counter GIS programs. Thus, it is now practical and efficient for the USGS to manage site-specific water use data (which would consist of literally millions of water withdrawal points and their associated attributes) for every state. The increasingly common availability of site-specific water use databases and GIS technology among the states is a key opportunity for the NWUIP. The last few decades have also seen the emergence of major national databases on water-using facilities and activities, including many with site-specific data. Examples include the Safe Drinking Water Information System (SDWIS) database on public water supplies from the U.S. Environmental Protection Agency, databases on energy generation facilities and electricity production time series from the Energy Information Administration, and the Census of Agriculture information on irrigation water use from the U.S. Department of Agriculture (see Chapter 3). These databases are a rich source of basic information for new water use estimation techniques.

OCR for page 115
Estimating Water Use in the United States: A New Paradigm for the National Water-Use Information Program Although site-specific water-use databases are very useful, they have limitations. Total water use by category is easily calculated by summing water use for all withdrawal points within any geographically defined region (e.g., Arkansas). If the only goal for the NWUIP were to accurately estimate total withdrawals by county and state, a site-specific database would suffice. However, even detailed site-specific databases of permitted withdrawals alone cannot fully satisfy the requirements for national water use estimation proposed in this chapter. For example, site-specific data developed to support regulatory activities may omit water use that falls outside of the state’s regulatory authority. This error may be negligible if the data are used to estimate total water use in a state, but may be unacceptably large for assessing drought vulnerability, sourcewater protection, or ecosystem impacts (see Box 7.1). BOX 7.1 Limitations of Site-Specific Databases—Arkansas Example Arkansas’s site-specific water use database (see Chapter 3) is distinguished by its high-quality and complete coverage of permitted withdrawals. Global Positioning System (GPS) verification and georeferencing of water withdrawal points, and quality assurance/quality control activities that discovered and corrected errors in about 80% of the records, make this database a model for state regulatory programs. However, alone, this comprehensive database is insufficient to satisfy the requirements for the NWUIP envisioned by the committee. Consider the site-specific data for domestic water use (Table 5.1). Only four of the nearly 45,000 water withdrawal points represent domestic water use. This shows not the absence of domestic water use in Arkansas, but rather the sparse number of domestic users large enough to be covered by the state’s permit program. The USGS estimates domestic water use in Arkansas based on the difference between the state population and the population served by public supplies. Annual domestic water use not included in the state permit system can be estimated as: 450,000 persons × 89 gal./person/day × 365 days/year = 14.6 MG/year. This represents only 0.1 percent of the total annual use reported in the state database. The error introduced by neglecting these domestic users is therefore negligible for estimating total water use for the state. However, this represents the primary water supply for about 20 percent of the population of Arkansas. Neglecting these domestic users would be unacceptable for assessment of drought vulnerability and statewide drought planning. Even the most accurate and complete site-specific water use database must be supplemented with indirect estimates for uses that are not captured by a regulatory permit program or, like instream flow, are not naturally associated with a single point of use.

OCR for page 115
Estimating Water Use in the United States: A New Paradigm for the National Water-Use Information Program Also, at a regional scale, spatial water use can be estimated without any site-specific data, and these estimates may be sufficient to meet many of the data-assessment requirements of the NWUIP. For example, Vörösmarty et al. (2000) estimated the global-scale vulnerability of the world’s freshwater resources to climate change and increases in water use. They used publicly available spatial data to estimate global water use for 0.5-degree grid cells, and linked these estimates to the natural hydrologic drainage network extracted from digital elevation model data. To estimate urban water use without site-specific data, urban population was distributed between 1-km-resolution city polygons and 1-km grid cells classified as “city lights” from nighttime remote sensing images. The Vörösmarty et al. (2000) comparison of spatial water use estimates and regional water balances foreshadows the committee’s vision of NWUIP integrating water use information in water resource assessments. Site-specific permit data are typically associated with water withdrawal points, whereas water use is often more appropriately associated with an area (e.g., an irrigated field, a polygon describing the limits of a municipal water distribution system). Spatially distributed water use can be represented as a single point, such as the centroid of the population served or of a service area, the location of a treatment plant, or even the withdrawal location. However, this introduces estimation error whenever the arbitrary polygons used in estimation (e.g., census tracts, cities, counties, hydrologic units) do not correspond to the true spatial limits of each water use (see Box 7.2). One of the strengths of GIS technology is that GIS data structures can combine both point and spatial data. Even so, no single spatial structure (e.g., polygons representing census tracts, county boundaries, or irrigated acreage) can fully support the goal of water use estimation for any arbitrarily defined geographic area. This goal also requires techniques to interpolate and apportion both water use data and supporting data such as population, employment, and irrigated acreage that are uniformly available from consistent national databases. These data are typically compiled by state, county, or census tract boundaries and therefore must be extrapolated over different spatial domains—such as a watershed. The extrapolation error from, for example, translating county-level data to a smaller hydrologic unit is inescapably part of national assessment programs such as the NWUIP. Identifying the best methods to quantify and minimize these approximation errors is a fruitful and widely applicable research area for the NWUIP. Water Use Estimation The nature and availability of data resources for water use estimation impose many constraints. For instance, the quantity and quality of water use data vary across the country, as each state tailors water use data collection to meet individual regulatory or resource assessment needs. As the discussion in subsequent sections will show, even in states with extensive site-specific water use and

OCR for page 115
Estimating Water Use in the United States: A New Paradigm for the National Water-Use Information Program BOX 7.2 Limitations of Site-Specific Databases—Maryland Example To manage water use in a site-specific database, spatially distributed water uses can be represented by a single point such as the point of withdrawal or the location of a treatment plant. Consider public supply water use for the western Maryland city of Cumberland, in Allegany County. The city draws its raw water supply from surface water reservoirs in Bedford County, Pennsylvania. Treated disinfected potable water is distributed to the city’s retail customers as well as to wholesale customers that operate their own distribution systems—including Allegany County. One of the city’s largest single customers (accounting for about 25 percent of total demand) is a co-generation plant that uses potable “public supply” water for evaporative cooling in thermoelectric power generation. The city also sells treated disinfected drinking water to wholesale customers across the Potomac River in West Virginia. Depending on the convention adopted, representing this spatially distributed public supply system by a single point could variously shift the associated water use between three different counties in three different states and at least two separate categories of use. Despite this complexity, the city of Cumberland’s water use is less than 1 percent of the total public supply for the state of Maryland, which is dominated by public systems serving the Baltimore-Washington corridor. Errors introduced by treating Cumberland as a single water use point would have a small impact on the accuracy of estimated public supply water use for the state of Maryland. However, as with the Arkansas example in Box 7.1, these errors would have a significant impact in assessments of drought vulnerability, stresses on aquatic habitats, or the sustainability of water use in the affected watersheds. spatial databases, there are limitations on the use of these databases in water use estimation. As a result, a “one size fits all” approach to water use estimation is impractical and undesirable for the NWUIP. Instead, a variety of estimation techniques, tailored to the data resources of individual states, are needed. As the survey of estimation techniques in Chapter 4 showed, there are many promising approaches for water use estimation. Chapter 5 illustrated the advantages of statistical sampling of site-specific water use for estimating spatial water use and its uncertainty, while Chapter 6 demonstrated how statistical inference based on statewide categorical water use data can help reveal the structure of water use on a national scale. As part of the NWUIP, there is a strong need for the USGS to systematically compare these and other water use estimation methods in order to identify the techniques (or combination of techniques) best suited for the individual states or regions. Although data resources supporting water use estimation are not consistent across the country, consistency is still important in the products from the NWUIP. For all states, water use estimates are needed for consistently defined sets of

OCR for page 115
Estimating Water Use in the United States: A New Paradigm for the National Water-Use Information Program water use categories and spatial regions (e.g., counties). This situation makes the reporting of uncertainty in water use estimates critically important. Users of water use estimates need to be aware that uncertainties of estimates within these similar products are quite variable from one state to the next. Water Use Science Despite the importance of the conceptual framework outlined in this chapter, water use is more than a matter of databases, statistics, site locations, and estimates. One of the most fascinating aspects of water use is that human activity alters the quantity and quality of water and the patterns of water flow through the landscape. Human impacts on water systems have ecological consequences. Chapter 8 of the report looks at water use in this larger context of integrative water use science. Still, with the growing availability of site-specific water use data and ancillary datasets, water use science also has an important role in improving water use estimation. There are many opportunities for research on technique development. These include the testing and evaluation of statistical sampling strategies, the transfer of information from data-rich areas for estimation in data-sparse regions, and the exploration of indirect techniques for water use estimation using ancillary datasets. CONCLUSIONS AND RECOMMENDATIONS There is a compelling national need for unbiased, science-based water use information for current and future management and policy decisions. To meet this need, the NWUIP must obtain estimates of water use that meet unambiguous and meaningful objectives. Specifically, the national program should at a minimum (1) produce consistent estimates across the country, (2) be linked to the surface and groundwater resources, (3) detect fundamental changes in the behavior and technologies that determine water use patterns, (4) utilize statistical sampling and inference within the context of an unambiguous and statistically meaningful definition of the populations of water users to be estimated, (5) incorporate error analysis designed to give relative standard errors, (6) account for “intermittent” changes in water use, and (7) account for interbasin transfers and transfers between surface and groundwater sources over multiple scales. The committee sees the NWUIP addressing two complementary goals, namely, the understanding of (1) spatial water use, i.e., the use of surface and groundwater, by category, within any designated area, and (2) hydrologic water use, i.e., water use as the human component of the hydrologic cycle. Spatial water use is linked to human behavior and consumption patterns and decisions. Hydrologic water use emphasizes the impact of humans on the natural resource. The NWUIP can and should characterize and quantify the importance of water

OCR for page 115
Estimating Water Use in the United States: A New Paradigm for the National Water-Use Information Program use as both an essential commodity in the economy and water use as a critical hydrologic stress affecting the sustainability of the nation’s water resources. Achieving these goals requires applied research and techniques development in both estimation and sampling techniques as well as science-based research on the determinants and impacts of water use behavior. These considerations suggest a conceptual framework for the NWUIP that links the infrastructure water system, described by locations of water withdrawals, water discharges, and the principal water facilities, with the natural water system of streams, rivers, lakes, aquifers, and watersheds. The increasingly common availability of site-specific water use databases and GIS technology among the states is a key enabling factor for this approach. Likewise, “link-node data structures” associate each water use with a node and connect the node to other water use nodes by a set of directed links. Such a link-node representation gives a sense of direction to the flow of water use. Such a framework may not be appropriate at all scales (e.g., regional) and adequate for all purposes (e.g., drought management) and will need to be supplemented with other methods and frameworks. Regardless, any framework chosen should be supported by water use data, water use estimation, and water use science.