Chapter 11 -- National Handbook of Recommended Methods for Water Data Acquisition

11.B.1. Concepts

11.B.1.a. Offstream use

Water-use processes are discussed here as they relate to data collection and in determining the total quantity of water used for a specific offstream water-use category.

Withdrawal --

The quantity of water removed from a ground-water source or diverted from a surface-water source for use. Self-supply withdrawal is water withdrawn directly by commercial, domestic, industrial, and other users and is represented in figures 1 and 2 as "A2". Public water supply withdrawals are represented in figures 1 and 2 as "A1".

Delivery --

Deliveries are the quantity of water delivered at the point of use or treatment (B1, B2, B3, and B4).

Release--

Releases are the quantity released after use or treatment (C1, C2, C3, C4).



Figure 1. Water use flow diagram

---




Figure 2. Relations between water-use processes

---

Conveyance

--The systematic and intentional movement or transfer of water from one point to another, for example, from point of withdrawal to point of delivery (A1-B1, A2-B2); release from public water supplier to user (C1-B3); release from user to wastewater-treatment facility (C2-B4); after release to return flow (C3-D1, C4-D2). Withdrawals, deliveries, releases and return flows are the endpoints of conveyances. Conveyance loss may occur through evaporation from an open system or seepage. Conveyance gain may occur through infiltration and inflow (C2-B4). Conveyance of water may occur across political and hydrologic boundaries that range from small Minor Civil Divisions (MCD) to State borders, and from minor subbasins to Continental Drainage Divides and may have a major impact on water availability on local and regional scales.

Consumptive use

--That part of water withdrawn that evaporates, transpires, is incorporated into products or crops, is consumed by humans or livestock, or otherwise removed from the immediate water environment. In some cases, consumptive use will be the difference between the volume of water delivered and the volume released (B2+B3-C2-C3). The definition of consumptive use may be more restrictive if it focuses on water that is available for immediate reuse. For instance, water that has undergone degradation of water quality during use (irrigation in areas where the soil contains salts) may be considered consumptive use. Similarly, water that is returned to the soil to percolate downward to a deep water table that may be a thousand feet below the surface also is considered by the U.S. Bureau of Reclamation to be consumptive use.

Return flow

--The quantity of water that is discharged after use or wastewater treatment (D1, D2) to surface water or ground water so that it becomes available for reuse.

Reclaimed wastewater

--Wastewater-treatment plant effluent that has been diverted for beneficial use before it reaches a natural waterway or aquifer.

11.B.1.b. Instream use

Instream water use is water that is used, but not withdrawn, from a surface- or ground-water source. Instream uses can be broadly characterized as streamflows to meet human needs and streamflows to meet ecological or environmental needs. Human needs include recreation, hydroelectric power generation, transportation, waste assimilation, aesthetics, and cultural resource preservation. Ecological or environmental needs include fish and wildlife preservation, biodiversity, wetlands preservation, freshwater dilution of saline estuaries, and maintenance of the riparian zone.

Instream conveyance (fig. 1, E) occurs when water flows or is pumped from one reservoir to another without being used. Water may flow through a chain of reservoirs, from a river to a chain of lakes, from a river to a canal or aqueduct, or even from surface water to ground water. In some cases, the rate of water flowing or being pumped from one reservoir to another may be reported to a State regulatory agency and should be tagged as in instream use to avoid double or triple accounting of offstream withdrawals.

Quantitative estimates for most instream uses are difficult to compile. However, because such uses compete with offstream uses and affect the quality and quantity of water resources for all uses, effective water-resources management requires that methods, definitions, and procedures be devised to assess instream uses quantitatively. The only instream use described in this report is for hydroelectric power generation. Unlike other instream uses, hydroelectric power generation water use is a measurable quantity because the amount of water passed through a power plant can be documented.

Although the need to have a minimum quantity of water in the streams has long been recognized as an important component of the stream ecosystem, it is only recently that the importance of instream flows have been acknowledged legally to the point where increases in instream flows have been mandated in many court and compact decisions throughout the country.

Methodologies and terms concerning instream flows currently (1995) have not been standardized. Previously, the need for standardization of nomenclature has been noted but not acted upon. Finally, the West Division, American Fisheries Society responded to this need for standardization and developed a Glossary of Stream Habitat Terms.

The approach for determining instream flow use and requirements are significantly different from offstream use and may be addressed in a subsequent chapter.

11.B.1.c. Conservation activities

During the past two decades, public water suppliers, industries, and irrigators have become more involved in water conservation as a means for either expanding services (population served) with a finite supply or meeting current needs with a decreasing supply. Withdrawals by some public water suppliers have decreased through implementation of (1) leak detection programs, (2) domestic conservation programs featuring public education and retrofit low-flow fixtures, (3) conservation consulting services to industrial and commercial users, and (4) increased rates for water supply and wastewater treatment. Industries have decreased water withdrawals by redesigning their plant operations to maximize recycling water and using new technologies that require less water. Many irrigators are using low-pressure drip and trickle systems instead of high-pressure spray systems.

Areas with significant seasonal climatic variation may experience attenuated seasonal variation through restrictions or conservation programs that affect uses such as outdoor watering. Temporary restriction programs resulting from a drought can decrease water use markedly; however, when restrictions are not permanent, users tend to regress to previous water-use patterns. Some communities have permanent water-use restrictions, especially in arid States. Conservation programs, including retrofitting older buildings with low-flow fixtures, replacing grass with low water-use landscaping, and increasing prices for metered water users have a more permanent impact on water-use patterns. Restrictions on use, local ordinances, and conservation practices in an area need to be investigated before estimating use.

Water also can be reused as wastewater reclamation after release from a wastewater treatment plant. Decreasing availability of freshwater and increasing treatment costs tend to increase wastewater reclamation for irrigation or even commercial and industrial uses.

Choosing whether to use site-specific data or area estimates to estimate water use will depend on the objective of the water-use data collection; availability of statewide reported data; availability of time, manpower, and funds; and the area to be covered. The objective of the water-use data collection determines the required degree of accuracy and reliability and identifies individual data elements that are relevant. For example, if the objective of a water-use study is to compare the need for water by the textile industry with the need for water by the petroleum-refining industry in the United States, an estimate based on a coefficient per textile or petroleum-refining employee and the number of textile or petroleum-refining employees could be used. Similarly, if the objective is to determine withdrawals by people with their own wells for each county in a State, a coefficient of water use per person per day (per capita) could be used. However, if the objective is to determine industrial withdrawals within a small watershed, then either a complete inventory or a statistical sample of the industries would be required.

Area estimates are determined by using coefficients relating water use to another characteristic that is available for the area of interest, such as people per county or specific types of industry per state. Livestock water use almost always is estimated from published livestock surveys and established coefficients for water used by dairy cattle, beef cattle, sheep, pigs, and other livestock. Coefficients are most reliable when they are applied to uniform groups of users from which the coefficients were developed, such as livestock water use.

Applying coefficients to groups of users that are non-uniform, such as industries, must be approached cautiously. Water use by specific types of industries varies widely because of age and condition of the plants, processes used, the amount of recycled water used at each plant, and quality of the cooling water (James and others, 1980, p. 5). Any coefficients used for non-uniform users, however, can be improved locally through field verification of large facilities in the area of interest.

An inventory of all users, either by mail or site visit, is feasible if there are only a few users in the category of interest, such as thermoelectric facilities; available resources or time and manpower are relatively abundant; or the area to be covered is small. Even if an inventory of reported data is available, the accuracy of the data must be investigated.

The objective of statistical sampling is to collect information on a small but representative segment (sample) of a population and generalize the results to the total population. Statistical sampling methods are used to make an inference about groups of water users and their water-use characteristics on the basis of information obtained from a relatively small number of users. If done appropriately, the results of the sample can be used to develop a coefficient to apply to the population of users. Statistical sampling can decrease the cost of data collection and still maintain reliable estimates of water use. More accurate data collection from each site is feasible if fewer sites are sampled and time is spent collecting detailed water-use information on how the water is used rather than merely collecting data on the rate of withdrawal of use.

Stratified-random sampling is one form of sampling that offers considerable efficiency in water-use estimation and involves separating sites with similar water-use characteristics into groups or strata. Stratified-random sampling can increase the reliability of estimates on the basis of sample "means." The technique is used when there are known groups with particular water-use characteristics (for example, corn or tobacco growers, motels with swimming pools and restaurants, or golf courses). Stratification divides the population into internally similar groups. The greater the internal similarity of a group, the smaller the sample needed to adequately characterize water use for the group. Sampling a larger percentage of the major users than the minor users improves the reliability of the overall water-use estimates.

Selected statistical methods for estimating ground-water withdrawals were analyzed by Luckey. He determined that regression analysis and random sampling maintained an acceptable degree of accuracy and markedly decreased data collection. Statistical methods to estimate historical and 1980 irrigation requirements in the High Plains of Colorado, Kansas, Nebraska, New Mexico, Oklahoma, South Dakota, Texas, and Wyoming were successfully applied by Heimes and Luckey (1982, 1983). Shoemyen (1979) presented a detailed description of the statistical sampling techniques used in the Suwannee River Water Management District.

11.B.3. Methods

11.B.3.a. Primary data acquisition

Primary data can be obtained by direct methods such as using cumulative water meters; or indirectly by (a) measuring the pumping rate with instantaneous flow meters, and (b) measuring or estimating pumping duration. Some commonly used methods for measuring water use already have been discussed in previous chapters of this National Handbook of Recommended Methods for Water-Data Acquisition. Specifically, methods for measuring surface-water flow are described in Chapter 1, and methods for measuring and estimating ground-water pumpage are discussed in Chapter 2.

Primary data collection should be preceded by a public-relations bulletin to inform the public about the purpose of the field inventory and the legal authority for conducting the survey. A public-relations bulletin will help to insure the cooperation of the landowners and farmers. The bulletin must explain the ultimate use of the data, and more importantly, how the participants will benefit.

11.B.3.a.1. Direct methods

Cumulative meters record the volume of water that has passed through the meter in a manner similar to an odometer. The meter readings increase until the highest number on the meter is reached, whereupon the meter dial returns to zero and begins again. These meters are commonly used by public water suppliers to record the number of gallons or cubic feet the customer has received since the meter was last read. The accuracy and reliability of these meters varies considerably. When relying on these meters for information, the following factors need to be taken into account: range in volume for which the meter was designed; design accuracy of the meter; age of the meter; the date of the most recent calibration; and whether the meter will over-registration or under-register as the meter wears. The meter reader needs to become familiar with each type of meter to be read to ensure that the dial readings are not misinterpreted. Hurr and Litke (1989) illustrate several examples of meters and how to read them correctly. It is important when reading a meter to keep records of meter types and serial numbers to avoid incorrect readings when meters are repaired or replaced.

Table 1. Methods used in determining site-specific water use



11.B.3.a.2. Indirect methods

Indirect methods for measuring water use make use of instantaneous flow meters to measure the volume of water currently passing through a pipe. These meters can be permanently positioned or may be portable. Instantaneous meter readings need to be coupled with a running-time measurement in order to determine water used over a day, month, growing season, or year.

Sonic flowmeters may be the quickest and easiest method of determining instantaneous pipe flow. Portable sonic flowmeters can be attached to the outside of a pipe, calibrated, record the flow rate, and removed in usually less than an hour with no interference to the flow or the pumping operation. Several types and brands are available from various manufacturers. Flow rates may vary during the pumping period depending on drawdown in the well and possible system modifications; therefore, several measurements may be needed to determine an average flow rate over time.

The total volume of water use can be calculated by multiplying the instantaneous flow rate times the total time of operation for the period of interest. Fairly accurate values of running time may be available if an hour meter is located on the pumping equipment. An estimate of total flow can be calculated from the running time by recording the meter reading at the beginning and the end of the period of interest. If there are no timers on the existing equipment, a vibration time totalizer (VTT) can be installed on the pumping equipment. When placed on a vibrating pipe or pump, a VTT will sense the operation of the equipment and record the cumulative operating time (Shoemyen, 1979; Cordes, 1984). Electric pumps produce less vibration than nonelectric pumps, which may preclude the use of a VTT. However, an inductive time totalizer (ITT) can be installed on the electric wire supplying alternating current and provide the same information as the VTT (Cordes, 1984). Some users may keep an accurate log of operating times for the pump, which can be used to estimate total withdrawals.

Selection and use of meters are described by the U.S. Bureau of Reclamation, "Water Measurement Manual" (1974). In addition, the American Water Works Association (1986) describes the selection, installation, testing, and maintenance of water meters. However, the accuracy of the meter readings from older meters should be checked with another measurement method or a nonintrusive flowmeter, such as a sonic flowmeter (Arvin, 1992, Shoemyen, 1979; Cordes, 1984).

11.B.3.b Secondary data acquisition

Secondary data acquisition involves compilation evaluation and analysis of measured or estimated data sent by water users to State and Federal agencies. The data may be routinely reported to a State or Federal agency in response to a legal requirement or policy requirement and is referred to in this section as reported data. The State or Federal agency distributes data-collection forms on a routine basis to be filled in by the user. Frequently, reported data are required to adhere to a certain degree of accuracy. Surveyed data usually are collected in response to a specific need for the data by a Federal or State agency and response to the forms may be voluntary. In the initial stages of water-use data collection, check with other Federal and State agencies to determine the availability of reported data and completed or planned surveys. USEPA has published a guide to Federal water quality programs and information (U.S. Environmental Protection Agency, 1993) that provides a good list of sources of data and contact points.

Important sources of reported data may be available from State allocation programs, Health Departments, and Pollution Control Agencies, as well as Federal agencies like the USGS and USEPA. Approximately 85 percent of the States have a permitting or registration program for those who withdraw water that is accompanied by either a request or requirement reporting withdrawals (Don Arvin, U.S. Geological Survey, written commun., 1993). The USGS compiles water-use data in support of a Congressional mandate to compliment the USGS data on the availability and quality of the Nation's water resources. USEPA requires major industrial users and wastewater-treatment facilities to report the volume of discharge to surface-water bodies at least annually. Records of metered water-use data also may be available from public water suppliers and wastewater-treatment facilities.

Surveys, particularly mail surveys, may provide timely and relatively inexpensive estimates of water use. However, selection of populations, collection of data, and tabulation of results need to be properly monitored to ensure reliability. Respondent cooperation and overall response can be greatly enhanced by site visits and telephone follow-ups, but the cost also increases with such visits and follow-ups.

Reported and surveyed water-use data must be used with caution because the accuracy and reliability of individual withdrawal or discharge reports varies considerably. Reported or surveyed water-use data may be either metered or estimated. Metered data generally are reliable; however, metered water-use values may be less accurate if the design flow of the meter is not within the actual flow through the meter, if the meter has not been calibrated lately, or if it is inexpertly read. Some users mistakenly report the actual reading from a cumulative meter, which would indicate an increased use each month. Further errors may occur during transfer of readings from the meter to the field sheet and to the data base. Reported or surveyed water-use data, which are based on estimates, require evaluation for the estimation method. In some cases, the reported water use may conform more closely with withdrawal and discharge permits rather than with actual use. Estimates of total water use depend on accurate reports, and an adequate number of returned questionnaires. Reported or surveyed water-use data are rarely field verified.

11.B.3.c. Derived data

Derived data depend on a coefficient to estimate water use or a value calculated from models, including extrapolation, non-economic multiple-coefficient, econometric, and accounting models. The coefficients may be obtained from literature, refined using secondary data, or developed from primary data collection. For example, a per capita value of 100 gal/d (gallons/day) may be obtained from the literature and be used to estimate domestic use. Data from a local water supplier may indicate that a population of 1,000 people was supplied with 80,000 gal/d, for a locally refined per capita of 80 gal/d. Measurement of a statistically significant number of samples may yield an average per capita value of 77 gal/d.

Models that use extrapolation techniques especially are appropriate for estimating total water use in situations where extreme accuracy and detail are not required. Extrapolating water-use values through time is based on the assumption that current water use can be determined by water-use trends in the past. Time is the independent variable and water use is the dependent variable. Models based on extrapolation include simple regression and trend analysis.

Non-econometric multiple-coefficient models consider water use as a mathematical function of two or more explanatory variables, related to such factors as weather or demographics and do not include price or economic factors. Water use is considered a necessary requirement unaffected by price. The variables are incorporated in a model that fits historical data and the coefficients are estimated statistically, usually by multiple regression.

Econometric models are based on economic factors and the demand for water is estimated as a function of the price of water and other economic factors. Multivariate demand models determine water-use demand as a function of weather, economics, and demographics, and the function is statistically determined. Econmetric models are especially appropriate for estimating total public water-supply water withdrawals and deliveries. One of the more frequently used models, the Corp of Engineer's Institute for Water Resources-Municipal and Industrial Needs Model (IWR-MAIN model) is described in Davis, and others, 1991.

Accounting models use a general equation that relates various water-use activities. In general, the volume of use is equal to the total volume of water received by the user (fig. 2). The volume of use also is equal to the sum of consumptive use plus the total volume of water released. If measured or estimated values are placed into the following equation (letters and numbers in parenthesis refer to Figures 1 and 2).

withdrawals (B2) + public water supply deliveries (B3) = water use

water use - consumptive use = wastewater collection (C2) + return flow (C3),

Estimates can be developed for the remaining components. For example, if 100 Mgal/d (million gallons/day) was delivered to a factory, 10 Mgal/d was incorporated into the product, and 80 Mgal/d was discharged into the wastewater-collection system, then 10 Mgal/d could be assumed to have evaporated or have been return flow. Conveyance losses or gains and unmetered use complicate the water-use equation. For instance, conveyance losses need to be subtracted from withdrawal before it can equal use.

Quality assurance is important for all data collection, compilation, analysis, and use of water-use data. Each of the previous sections on water-use determination in this chapter discuss the weakness of the different determination methods. Metered data must be reviewed in the context of the meter accuracy and reliability. Coefficients need to be calibrated to local conditions. Calibration doesn't always require the collection of site-specific data, but may be accomplished by using corroborative data, as well as statistical analysis of the range and standard deviation to identify outliers or inconsistent data. Edit programs can be written to check the reliability of automated data and identify any discrepancies from one year to the next.

American Water Works Association, 1986, Water meters--Selection, installation, testing and maintenance: Manual M6.

American Water Works Association and Water Environment Federation, 1994, Proceedings of the Water Reuse Symposium, Feb 27-March 2, 1994, Dallas, Tex.:, 845 p.

Anderson, K.E., 1966, Water well handbook (4th ed.): Water Well Drillers Association, Rolla, Mo., 281 p.

Arvin, Don, 1992, Feasibility of using portable, noninvasive pipe flowmeters and time totalizers for determining water use: U.S. Geological Survey Water-Resources Investigations Report 91-4110, 65 p.

Bayha, Keith, 1978, Instream flow methodologies for regional and national assessments, Instream Flow Information Paper: No. 7, Fort Collins, Colorado: U.S. Fish and Wildlife Service, Office of Biological Services, Western Energy and Land Use Team, Cooperative Instream Flow Service Group; FWS/OBS-78/61; 97 p.

Brown, T.C., Taylor, J.G., and Shelby, Bo, 1991, Assessing the direct effects of streamflow on recreation: A literature review: Water Resources Bulletin, American Water Resources Association Paper no. 91122.

Canadian Government Publishing Centre, 1986, Canada water year book, water use edition: Ottawa, Canadian Government Publishing Centre Supply and Services, 98 p.

Cordes, E.H., 1984, Ground water instrumentation program of the U.S. Geological Survey: Ground-Water Monitoring Review, v.4, no. 4, p.103-114.

Davis, W.Y., Rodrigo, D.M., Opitz, E.M., Dziegielewski, Benedykt, Baumann, D.D., and Boland, J.J., 1991, IWR-MAIN water use forecasting system, version 5.1--users manual and system description, consultant report:, Carbondale, Ill., U.S. Army Corps of Engineers and Planning and Management Consultants, 307 p.

Gleick, P.H., 1993, Water in crisis, A guide to the world's fresh water resources: New York, Oxford University Press, 473 p.

Heimes, F. J., and Luckey, R.R., 1980, Evaluating methods for determining water use in High Plains in parts of Colorado, Kansas, Nebraska, New Mexico, Oklahoma, South Dakota, Texas, and Wyoming, 1979: U.S. Geological Survey Water-Resources Investigations 80-111, 118 p.

-----1982, Methods for estimating historical irrigation requirements from ground water in the High Plains aquifer in parts of Colorado, Kansas, Nebraska, New Mexico, Oklahoma, South Dakota, Texas, and Wyoming: U.S. Geological Survey Water-Resources Investigations Report 82-40, 64 p.

-----1983, Estimating 1980 ground-water pumpage for irrigation on the High Plains in parts of Colorado, Kansas, Nebraska, New Mexico, Oklahoma, South Dakota, Texas, and Wyoming: U.S. Geological Survey Water-Resources Investigations Report 83-4123, 36 p.

Helm, W.T., ed., 1985, Glossary of stream habitat terms: Western Division, American Fisheries Society, 34 p.

Holland, T.W., 1992, Water-use data collection techniques in the Southeastern United States, Puerto Rico, and the U.S. Virgin Islands: U.S. Geological Survey Water-Resources Investigations Report 92-4028, 186 p

Holland, T.W., and Baker, N.T., 1993, Evaluation of pumpage data furnished by selected public water suppliers in Arkansas, May 1990 through March 1991: U.S. Geological Survey Water-Resources Investigations Report 93-4104, 80 p.

Horn, M.A., Craft, P.A., and Bratton, Lisa, 1994, Estimation of water withdrawal and distribution, water use, and wastewater collection and return flow in Cumberland, Rhode Island, 1988: U.S. Geological Survey Water-Resources Investigations Report 93-4023, 54p.

Hurr, R.T., and Litke, D.W., 1989, Estimating pumping time and ground-water withdrawals using energy-consumption data: U.S. Geological Survey Water-Resources Investigations Report 89-4107, 27 p.

International Great Lakes Diversions and Consumptive Uses Study Board, 1981, Great Lakes diversions and consumptive uses: Windsor, Ontario, Canada, International Joint Commission, 56 p.

International Joint Commission, 1985, Great Lakes diversions and consumptive uses: Windsor, Ontario, Canada, International Joint Commission, 82 p.

James, I.C., II, Kammerer, J.C., and Murray, C.R., 1980, How much water in a 12-ounce can? A perspectqive on water-use information: Pamphlet reprint from U.S. Geological Survey Annual Report, fiscal year 1976, 18 p.

Kammerer, J.C., 1982, Estimated demand of water for different purposes, in Water for human consumption, man and his environment: International Water Resources Association, Dublin, Tycooly International Publishing Limited, 606 p.

LaTour, J.K., 1991, Determination of water use in Rockford and Kankakee areas, Illinois: U.S. Geological Survey Water-Resources Investigations 90-4166, 79 p.

Luckey, R.R., 1972, Analyses of selected statistical methods for estimating ground-water withdrawal: Water Resources Research, v. 8, no. 1, p. 105-210.

Mather, J.R., 1984, Water Resources: Distribution, use, and management, John Wiley and Sons and W.H. Winston and Sons, Silver Spring, Md., 439 p.

National Water Use Information Program, 1994, A supplementary guide to using the IWR-MAIN water use forecasting system with special case study. Emphasis on the MWD-MAIN version: University of Wyoming in cooperation with the U.S. Geological Survey.

Schefter, J.E., Moody, D.W., 1981, Water use data--who needs it: Water Resources Bulletin, v. 17, no 6, 978 p.

Shoemyen, J.L. (ed.), 1979, Procedure for assessing agricultural water use: Suwannee River Water Management District, Department of Planning and Operations, 114 p.

Solley, W.B., Pierce, R.R., and Perlman, H.A., 1993, Estimated use of water in the United States in 1990: U.S. Geological Survey Circular 1081, 76 p.

Sudman, Seymour, 1976: Applied sampling: New York, Academic Press, Incorporated, 249 p.

Sweat, M.J., Van Til, R.L., 1988, Water use and methods of data acquisition in Michigan, in Symposium on Water-Use Data for Water-Resources Management, Tucson, Ariz., 1988, Proceedings: American Water Resources Association, TPS-88-2, p. 872.

Tate, D.M., 1990, Water demand management in Canada--a state-of-the-art review: Environment Canada, Social Science Series no. 23, 52 p.

U.S. Bureau of Reclamation, 1974, Water measurement manual: Denver, Co., U.S. Bureau of Reclamation, 327 p.

U.S. Department of Agriculture, 1989, the Second RCA appraisal: Soil, water, and related resources on nonfederal land in the United States; Analysis of Condition and Trends: 280 p.

U.S. Environmental Protection Agency, 1993, Guide to Federal water quality programs and information: Washington D.C., EPA-230-B-93-001, 194 p.

U.S. Office of Management and Budget, 1987, Standard industrial classification manual, 1987: U.S. Government Printing Office, 705 p.

Van Der Leeden, Frits, Troise, F.L., and Todd, D.K. (eds.), 1990, The water encyclopedia (2d ed.): Chelsea, Mich., Lewis Publishers, 808 p.

Wilson, B.C., 1992, Water use by categories in New Mexico counties and river basins, and irrigated acreage in 1990: New Mexico State Engineer Office, Technical Report 47, 141 p.