Water-Use Data-Gap Analysis

This geonarrative explores water-use data availability in the U.S. and identifies gaps in those data that limit their use in water-use models and water-availability and water-management models.

A water-use data-gap analysis is a process that determines the differences between the current state of the Nation's water-use data and the data needed to achieve the long-term goals of the USGS Water Availability and Use Science Program. The data-gap analysis includes examining the important regional uses of water, how those uses are spatially distributed across the region, the availability of and level of uncertainty in historical data, the triggers behind water-use decisions, and the suitability of existing data to meet modeling and water-management needs.

The USGS is transitioning to the use of models for estimating historical and forecasting future water use in order to improve the temporal and spatial scales for which water-use estimates are available and to provide estimates in “near real-time.” An analysis of water-use data was conducted to identify data that are needed for current and future modeling work.

Several methods were used to identify water-use data gaps, including an analysis of the county- and State- level water-use data from 1985 to 2015 from the USGS Aggregated Water-Use Data System (AWUDS), inventory and evaluation of the USGS Site Specific Water-Use Data System (SWUDS), inventory of the USGS Water-Use Data and Research Program (WUDR) data from State agencies, a literature review primarily focused on socioeconomic drivers or triggers that affect the timing and volume of water use and water availability, and discussions with selected water-use project teams. As data gaps were identified, additional sources of data to fill those gaps were investigated. Table 1 describes sources of information used in this analysis.

This assessment was initially done by State for the three largest water-use categories, or priority categories: public supply, irrigation, and thermoelectric power. The priority categories comprise 90 percent of the total water withdrawals in the U.S. The assessment was then repeated by county for all water-use categories:

The water-use category with the largest amount of water withdrawals, referred to as the top category, was determined by State and county for selected water sources: groundwater, surface water, and total (groundwater and surface water combined).

Historical water-use estimates can be mapped to identify the predominant water-use categories at various spatial scales, as well as the source of the water (groundwater or surface water).

A map of total water use by State for 2015, shows that thermoelectric power is the top priority category for most States in the eastern part, and irrigation is the top priority category for most States in the western part of the U.S. (fig. 1). Results are similar when all categories are evaluated, however a few States do have a different top category when all categories are used rather than the priority categories.

The Water-Use Maps section contains several interactive maps that explore historical water-use data at the county level. A total of 3,236 counties were evaluated. Some county boundaries change over time. A county may have been added, removed, or merged with another county. This assessment includes the set of counties that were active from 1985 to 2015 (  U.S. Census Bureau, 2022  ). Similar to the State assessment, first an evaluation of the priority categories was done followed by an evaluation of all categories.

The importance of irrigation water use, primarily by groundwater but also by surface water, is clearly visible in the Mississippi Alluvial Plain in Arkansas, Louisiana, and Mississippi. Mining water use, which is often related to oil and gas production, emerges as the top category of groundwater use in the Permian Basin of western Texas, The Williston Basin of western North Dakota, and in the Alaska North Slope. 

Historical water-use data, which are obtained primarily from State agencies, are received in various formats and differ in quality. Reported water-use withdrawal data, for example, can be metered but are typically self-reported by a facility or system to a State agency on an annual basis. To detail the differences in the data obtained, the USGS began documenting the method used for computing the withdrawal volumes by county, category, water source, and element (for example, fresh surface water for mining or saline surface water for mining are reported separately).

Withdrawal volumes obtained from State agencies are characterized as reported (metered or measured), estimated, or computed from an aggregate area. Withdrawal volumes can be estimated by multiple methods including coefficients, permitted volumes, data from prior years, or disaggregation from a larger geographic water-use accounting unit. Data that are estimated typically have the most uncertainty.

The map titled, “Estimation Method for the Top Category of Groundwater and Surface-Water Use by County, 2015”, located in the Maps tab shows the primary estimation method for the top category of water use by annual volume for groundwater and surface water withdrawal data in 2015. A category, element, or county may have multiple reported estimation methods, but only the primary estimation method is shown.  

In order to optimize the predicting capability of the water-use models, it is important to identify the socioeconomic drivers that may impact water use and availability at various spatial and temporal scales. Socioeconomic drivers used to predict public-supply water use, for example, include population, economic, housing, and educational drivers (fig. 2). Specific drivers in literature include water pricing, housing tenure, persons per household, outdoor housing features, age of population, indigenous communities, native language, race/ethnicity, interpersonal/institutional trust, conservation efforts, and political drivers.

The figure below lists socioeconomic drivers that may improve the prediction capability of the public supply water-use model.

Likewise, socioeconomic drivers that may help improve the predicting capability of the irrigation water-use model include economic drivers, demographics, water-use conservation education and incentives, and water-use regulations.

Published literature indicates that the relationships between socioeconomics and water use may vary based on unique regional characteristics such as differences in climate or household income.

For example, though literature suggests that urban areas use less water, the opposite effect may be seen for arid areas due to increasing water necessities during high temperatures exacerbated by the urban-heat-island effect (  Guhathakurta and Gober, 2007  ). Nieswiadomy (  1992  ) suggests that increases in average water prices may decrease domestic water usage. However, Agthe and Billings (  1987  ) suggest that water use in higher income households was less sensitive to changes in water prices.

These publications illustrate the importance of studying interacting socioeconomic drivers that may influence water use.

The groundwater and surface water-use maps below show the top category by county. The data compiled for these maps are available in a companion data release (  Houston and others, 2022  ). The color of the county shows the top category (public supply, irrigation, thermoelectric power, industrial, mining, livestock, domestic, or aquaculture), as represented in the explanation in the bottom left corner.

Clicking on a county will trigger a pop-up window that shows detailed water-use information. Select the specific year of data that is displayed by using the layer list to toggle layers on and off. Pan (click and drag) or zoom (using the plus and minus buttons) to explore areas of the map that are not yet in view.  Adjusting the extent of zoom on your web browser may provide a better fit for the display of the web maps.

Use the time slider below to view changes in the top total water-use category from 1985 to 2015. Total water-use is the sum of groundwater and surface-water use.

 Many water-use data gaps were identified during discussions and meetings with the national water-use modeling teams for public supply, irrigation, and thermoelectric power. Some of these data gaps were identified at the beginning of the modeling projects, and others were identified over time as the models were being developed.

Measured or reported withdrawal information are a spatial and temporal data gap for the public-supply, irrigation, and thermoelectric-power water-use models that are already in progress. At the State level, more site-specific or facility-level data are available for the public-supply category than the irrigation category. States that do not have reported withdrawals for irrigation often estimate annual or monthly water withdrawals or consumptive use at the county level. Location of the irrigated field, which may be different than the location of irrigation withdrawal, and source of the irrigation withdrawal are often unreported or known.

Modeling provides a more complete picture of water use, enables the U.S. Geological Survey to estimate water use at a finer spatial and temporal scale, and provides the ability to forecast and predict water use. The end goal of the modeling approach is to estimate water use at the hydrologic unit code (HUC) 12 level and at a daily time step and report the modeled water-use estimates annually.  Without site-specific or facility-level data for model training, calibration, and verification, daily estimates will be of limited accuracy.  

Agthe, D.E., and Billings, R.B., 1987, Equity, price elasticity, and household income under increasing block rates for water: The American Journal of Economics and Sociology, v. 46, p. 273–286, accessed August 22, 2022, at   https://doi.org/10.1111/j.1536-7150.1987.tb01966.x  . 

Dieter, C.A., Maupin, M.A., Caldwell, R.R., Harris, M.A., Ivahnenko, T.I., Lovelace, J.K., Barber, N.L., and Linsey, K.S., 2018, Estimated use of water in the United States in 2015: U.S. Geological Survey Circular 1441, 65 p., accessed August 12, 2022, at   https://doi.org/10.3133/cir1441  . [Supersedes USGS Open-File Report 2017–1131.] 

Guhathakurta, S., and Gober, P., 2007, The Impact of the Phoenix urban heat island on residential water use: Journal of the American Planning Association, v. 73, p. 317–329, accessed August 25, 2022, at   https://doi.org/10.1080/01944360708977980  . 

Houston, N.A., Merriman, K.R., Dieter, C.A., and York, B.C., 2022, Data tables associated with an analysis of the U.S. Geological Survey's historical water-use data, 1985–2015: U.S. Geological Survey data release, accessed September 30, 2022, at   https://doi.org/10.5066/P94Y93BW  . 

National Research Council, 2002, Estimating water use in the United States—A new paradigm for the National Water-Use Information Program: The National Academies Press, Washington, DC, accessed August 22, 2022, at   https://doi.org/10.17226/10484  . 

Nieswiadomy, M.L., 1992, Estimating urban residential water demand—Effects of price structure, conservation, and education: Water Resources Research, v. 28, p. 609–615, accessed August 22, 2022, at  https://doi.org/10.1029/91WR02852  . 

U.S. Census Bureau, 2022, Substantial changes to counties and county equivalent entities—1970-Present, accessed August 22, 2022, at   https://www.census.gov/programs-surveys/geography/technical-documentation/county-changes.2010.html#list-tab-6BLKW6MNG8V0IX4GF0.   

Disclaimer: Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government.