Chapter 4—
Developing New Water Supplies

by Harvey O. Banks, Jean O. Williams, and Joe B. Harris

Abstract

The inadequacy or maldistribution of water supplies for agricultural and other water users throughout most of the western U.S. has historically been a focus of attention for the local citizenry, for water planners and developers, engineers, conservationists, economists, promoters and others concerned with the future of the region. This paper looks at the potential for developing alternative water supplies for the region, with special emphasis on meeting the future water needs of the agricultural sector. Prospects for importing excess surface waters into the region, either from international or domestic sources, are examined along with several local water enhancement or augmentation potentials. Weather modification, water harvesting/water banking techniques, desalination and/or use of saline water supplies,  water Reclamation and reuse, surface/groundwater management, improvements in operation of existing projects to increase yields, and other prospects for local water supply enhancement are discussed. General conclusions are that the probabilities for large-scale new water supplies or developments for the region in the foreseeable future are not great. The potentials for significant breakthroughs in local water supply enhancement or any large-scale water importation for the semiarid West are limited.

Development of irrigated agriculture in the semiarid to arid regions of the Great Plains, the Southwest, and the far western states was an inevitable step in the westward growth in the United States. Given the physical conditions of the availability of land, deep, productive soils, advantageous terrain, and climate that appeared to be suited to large-scale dryland farming, the agriculturally oriented culture of the western expansion quickly established an economy of boom-and-bust dryland farming.

Early reclamation programs demonstrated the potential for irrigation in these vast western lands. When, in the 1930s, deep

wells became more feasible and economic, extensive drilling tapped what were apparently unlimited groundwater resources, and irrigation spread rapidly throughout the Plains states, the Southwest, and wherever the combination of good land, favorable climate and cheap, plentiful water supplies could be found. Irrigation expanded even more rapidly after World War II.

By the 1960s, it had become apparent that groundwater resources were being depleted. Commitments of surface water to irrigation through reclamation projects came under increasingly heavy fire from existing or potential competing water users. More sensitive environmental concerns were also expressed. Throughout the West, a critical water supply crisis was developing.

Thus, water planners and developers, economists, politicians, and citizens concerned with the future have turned their attention to the potential for developing new water supplies. Simultaneously, the options for improving the efficiency of use of existing supplies have been subjected to intense study, as is reflected in other chapters of this volume.

This chapter deals with the prospects for developing new or augmented water supplies. There are relatively few new water resources in the West remaining to be developed. The prospects of developing those are slight and the costs would be very high. There are, however, opportunities for better allocation of supplies from existing projects, for improvement in management of such projects, for intervention in the hydrologic cycle to modify precipitation events on a basis more nearly related to man's needs than nature provides, and for application of some of the newer technologies. With that understanding, we will look briefly at the potential for new or augmented water supplies for irrigation in the West.

Imports from Outside the United States

Both Canada and Mexico share borders with the United States, and across those borders occur common problems of matching water needs with available supply. Yet within the continent of North America vast quantities of surface water occur, and the developable yield of all of those resources could meet the needs of all three nations for the foreseeable future, if it were possible to develop, allocate, manage, and use the water in the common interest. Realistically, the difficulty of this coordination, while

not insurmountable, is awesome, particularly as regards the political/legal/institutional/financial aspects. Within our own United States, discussions of interbasin transfers of water within a state or between and among states are generally conducted with a great deal more heat than light, and often with extraordinarily slow results.

While international development and allocation of available water resources may be difficult and very long-term in prospect, they are not impossible, and in most cases could be shown to be mutually advantageous for each nation and regional (or basin) participant. One of the axioms of any water transfer proposal is that there must be demonstrable benefits for all parties—both importers and exporters.

But let us look at some of the international potentials that have been examined. These international water transfer schemes are highly conceptual and typically have been investigated at only reconnaissance levels of feasibility. Water development sites and facilities have not been identified specifically, nor have meaningful cost projections been prepared. Projected future costs in excess of $100 billion for complete system installation could be anticipated. Looking first to the case of Canada, we here give brief discussions of four proposals for moving water from Canada to the United States.

The Rocky Mountain Plan[en1]The Rocky Mountain Plan^[1]

The Rocky Mountain Plan, conceived by William G. Dunn, Consulting Engineer, is a potential massive, international water and power development project that would distribute water and power throughout the West from Canada to the Mexican border.

Principal sources of water are the Peace, Athabasca, and Smoky rivers in northern Alberta (Canada), and upper tributaries of the Mackenzie River in northern British Columbia, which flows into the Arctic Ocean. Additional sources of water are the Kootenai and Flathead rivers and Clark Fork in western Montana, which are upper tributaries of the Columbia River. Water would be diverted for use within the Yellowstone, Missouri, and the Snake rivers in the northwestern United States, and upper tributaries of the North and South Saskatchewan rivers in Alberta.

The water distribution system would include several large reservoirs with a total storage capacity of nearly 100 million acre-feet. Project yield would range from 12 to 25 million acrefeet per year, depending on aqueduct and reservoir sizing. This water would be distributed through more than 5,850 miles of

aqueduct for use in southern Alberta, Montana, Idaho, Wyoming, all of the western states on both sides of the Rocky Mountains including west Texas and California, and northern New Mexico in the Colorado River and Rio Grande valleys.

New energy developed under the Rocky Mountain Plan would come from a huge hydroelectric project called the Whitehorse-Skagway Division, collecting water from the upper tributaries of the Yukon River and releasing it through a 2,200-foot power drop into an interior inlet of the Pacific Ocean near Skagway, Alaska. The 33 billion kilowatt hours of power produced by this system would be conveyed in a 2,000-mile transmission line to Alberta, British Columbia, and the Pacific Northwest for general use in the power market, and for project purposes. Three large storage reservoirs with a total storage potential of 60 million acre-feet are proposed within the Columbia River Basin. These reservoirs would include large pumped storage facilities that would reregulate the power developed in the Columbia River plants and in the project power plants, and that also would produce some new power.

The entire Rocky Mountain Plan, including power facilities, was estimated to cost between $40 and $50 billion in 1977 dollars. One of the significant advantages of the Rocky Mountain Plan is that it could be staged to provide significant water and power benefits during early development.

Canadian Proposal

Three Canadians (Knut Magnusson, Edward Kuiper, and Roy E. Tinney) have proposed concepts for diverting waters from the Athabasca, Peace, and Laird rivers to be conveyed across the plains of northern Alberta, Saskatchewan, and Manitoba to the United States border in North Dakota. A full report on this conceptual plan was not available to the authors, but it is of interest, showing as it does the international concern with such possibilities.

North American Water and Power Alliance (NAWAPA)[en2]North American Water and Power Alliance (NAWAPA)^[2]

NAWAPA is a master plan concept that proposes taking advantage of the geographical and climatological factors of the North American continent in contrast to the single river basin plan. It would utilize the excess water of Alaska, the Northwest Territories, and the Rocky Mountain regions of Canada, and distribute it to the water-deficient areas of Canada, the United States, and northern Mexico.

NAWAPA, conceived by the Ralph M.Parsons Company, includes a possible Transcontinental Canal for Canada—a navigable waterway from Alberta to Lake Superior. The canal system, plus the development of rivers in north central Canada, would distribute irrigation water across the plains and increase the flow through the Great Lakes-St. Lawrence System to alleviate water pollution and lowering water levels of that area. The NAWAPA plan is projected to generate 60 to 180 million kilowatts of electric power (net after meeting its own needs), and to supply more than 75 million acre-feet of water annually. The cost to implement the entire NAWAPA concept is estimated at several hundred billion in current (1982) dollars, with spending spread over a 30 to 50-year period.

Western States Water Augmentation Concept[en3]Western States Water Augmentation Concept^[3]

This plan, proposed by Lewis Gordy Smith, is for a new water system that would permit any available surplus waters from the Fraser River near Hope, British Columbia, and from the Coastal Range in British Columbia to be passed into the Columbia River, and from there conducted within a distribution system both east and west of the Continental Divide. This system would supply water to the Upper Snake, the Humboldt River system of Nevada, the Salt Lake area, the Missouri River system, the Green River and the lower Colorado, the Rio Grande below Albuquerque, and the entire High Plains extending from Nebraska to western Texas. The plan would also look to the ultimate possibility of extending to the far north, to sources of British Columbia, Yukon, and Northwest Territories of Canada, and of later placing this water in the initial water conveyance system within the United States.

For the entire collection system from the Dean River to the Columbia, Smith projected a total of 26 dams, ranging in height from 200 to 1,200 feet, costing some $6.3 billion (1967 dollars). Approximately 26 power and pump plants would be required, with total pumping load and power generation potential about balanced. Almost 55 miles of open canal and five main tunnels totaling 56.5 miles, with capacities ranging from 5,500 to 49,000 cubic feet per second would be needed. The above features, along with transmission system, some railway relocation, and miscellaneous structures, would call for a total expenditure of about $11.5 billion in 1967 dollars.

In addition to these proposals for moving Canadian water across international boundaries, some planners have discussed the potential of moving water from northern Mexico into the

United States. These proposals have included capture and transfer of flood waters from the Rio Conchos in northwest Mexico and other potentials along the Mexican eastern coastal area. None of the discussions have been formally considered or presented.

Interstate Diversions of Water

Interbasin diversions of water have been in place in many parts of the United States for many years. Rarely have they been carried out without controversy. Where such diversions are contemplated across state lines, the opportunities for conflict and the complexities of law and equity, increase exponentially.

Certainly the major western intrastate, interbasin transfer project is the State Water Project in California, planned to have an ultimate firm yield of 4.3 million acre-feet per year and moving water from the north to the south through 715 miles of aqueduct serving municipal, industrial, and agricultural users en route.^[4] A careful study of the history of that project—its conception, design, and probably most importantly the ongoing intrastate controversies, conflicts, and regional bitterness generated in its implementation process—should be required reading for water planners and decision makers concerned with acquiring new water supplies. The Federal Central Valley Project in California, also wholly intrastate, has been and still is subject to many of the same problems. Diversions from the Lower Colorado River to California and in the near future to Arizona through the Central Arizona Project are other examples.

Major potential sources of water for interbasin diversion to arid western lands include the following.

The Columbia River Basin

This potential source of new water supply in the West has been considered from a conceptual standpoint, but federal legislation, sponsored by Senator Henry Jackson of Washington, has since 1968 precluded detailed studies by federal agencies of the potential for diversion into the Colorado River System or into northern California. Legislation has been introduced in the current session of Congress to extend a similar prohibition to all interstate waters.

The Missouri River Basin

Central to much of the development of the Midwest, the Missouri remains the object of interests in adjacent states as a potential export basin. Intrabasin states understandably object to any out-of-basin commitments of water from the Missouri in the absence of any institutional mechanism protecting long-term in-basin water needs.

The Missouri River main stem is already extensively controlled under the Pick-Sloan Plan (Flood Control Act of 1944) by the six main stem dams and reservoirs—Fort Peck, Sakakawea, Oahe, Sharpe, Francis Case and Lewis and Clark—for navigation, flood control, hydropower, and in-basin irrigation, municipal, and industrial uses. Any large exportation would involve trade-offs with these presently authorized commitments for in-basin uses.

In the recently (March 1982) completed report on the Six-State High Plains Ogallala Aquifer Regional Resources Study, conducted under the auspices of the Department of Commerce Economic Development Administration, the U.S. Army Corps of Engineers (Corps) examined the potential of the Missouri as a new water source for irrigation in the six states of Nebraska, Colorado, Kansas, New Mexico, Oklahoma, and Texas.^[5]

The legislation authorizing the Corps study explicitly limited the source basins for analysis to areas "adjacent" to the six-state region to be served. This eliminated the Columbia River from possible consideration. The Mississippi was also ruled out, and thus the Missouri was selected by the Corps as a potential source basin for its work.

Several diversion points and transfer routes were studied. Reconnaissance level design and cost estimates were made for ranges of transfer quantities. The Corps did not make a determination of the amounts of water that might be "surplus" to in-basin needs and thus available for diversion.

Transfer quantities of less than two million acre-feet annually to 3.4 million acre-feet were investigated for alternative routings and sizes of facilities. Resulting cost estimates (total investment costs in 1977 dollars) ranged from $2.9 billion to $7.4 billion for a 10-year construction period, and from $4.4 billion to $11.2 billion for a 20-year period for delivery to terminal reservoirs.

Unit costs (per acre-foot) of water delivered to terminal storage reservoirs under the alternative routes projected by the Corps from Missouri River sources ranged from $227 to $335. These costs are significantly in excess of the ability to pay for imported water by irrigation agriculture, in the time frame of

the High Plains Study to year 2020. They also do not include the additional water distribution costs from terminal storage to farm headgates.

Western Arkansas Basins

In addition to the potential diversions to the High Plains region from the Missouri River basin, the Corps also assessed the feasibility of interstate, interbasin transfers into the region from several streams in western Arkansas and northeastern Texas. The Arkansas, Ouachita, Red, and White rivers of western Arkansas, and the Sulfur and Sabine rivers in Texas were considered as possible sources.

Water transfer quantities for the southern alternative routes ranged from 1.26 maf per year to almost 8.7 maf annual diversion. Unit costs per acre-foot of water delivered to terminal storage sites by the two alternative southern routes of importation to the High Plains region ranged from $430 to $569, considerably more costly than the projected northern routes.

The Mississippi River System

The large flood flows of the Mississippi have long been studied as a potential source for water export. Key questions raised by in-basin interests are the long-term needs of in-basin users, high minimum flow requirements to repel intrusion of salt water up-river from the Gulf and for sediment transport, and the need for maintenance of fresh water inflows into coastal bays and estuaries of Louisiana. The extremely limited availability of storage for intermittent diversions of flood flows when they occur is a major roadblock to export from the Mississippi. A feasibility study by the Mississippi River Commission in 1973 of diverting water from the Lower Mississippi River Basin to West Texas and eastern New Mexico exemplifies previous studies of the Mississippi as a potential source basin.^[6] The study indicates a technical feasibility for such diversions, but a high cost of delivered water, at about $330 per acre-foot. Total capital costs for the system were estimated at $19.5 billion in 1972 dollars.

Weather Modification

Weather modification projects throughout the West have shown variable results to date. While some statistically significant precipitation enhancement results can be documented, they are not consistent and dependable, particularly for the

convective (summertime) cloud systems of the Great Plains area. Wintertime and high altitude (orographic) cloud seeding programs have been relatively more effective than the summertime experiments, particularly for increasing snowpacks, but significant operational and institutional problems confront all weather modification projects.

A related area of water supply augmentation is found in the treatment and management of snow accumulations in those regions or altitudes where significant snowpacks occur. Ongoing research and trials of methods and materials for improving water yields, decreasing evaporative losses, and managing the rate and timing of runoff from snow fields show promise of more dependable water supply management from this source.

By the late 70s, the scientific community (cf. U.S. Interdepartmental Committee on Atmospheric Sciences or ICAS)^[7] generally accepted the operational capability for seeding wintertime (orographic) clouds to increase precipitation by a factor of 10 to 20 percent. On the basis of a 15 percent increase in snowpack due to seeding, it has been projected that an additional 2+ million acre-feet of water per year, average, could be produced in the Colorado River Basin, at a (1977) cost of about $1.50 per acre-foot. In agricultural use, irrigation benefits are estimated at about $50 per acre-foot of available water. Most other uses have higher per acre-foot values than agriculture. Such large-scale precipitation enhancement would require much larger federal/state cooperative projects than have been attempted to date. A largely unresolved question is, who owns the additional water produced?

Water Harvesting—Water Banking

A local water supply enhancement method that has seen extensive development and use in the Mid-East, Africa, and other parts of the world, but limited application in the U.S., is the so-called "water harvesting" technique. This consists essentially of intensive watershed and vegetative management on nearby non-cultivated lands, in order to capture or "harvest" the water for use on cultivated areas. There are extensive areas throughout the West where this technique could be applied. "Water banking" is a technique for capturing available surface water in excess of immediate needs and overwatering areas with favorable infiltration rates. Excess waters are "banked" in groundwater storage through deep percolation for later recapture.

Such projects would necessarily be extensive in nature and involve many landholders. Where state laws direct the acquisition of groundwater rights, many questions of law as well as equity arise with respect to ownership of the banked water supply.

Conjunctive Use

The coordinated management of groundwater and usable underground storage capacity with surface water resources and surface storage as an integrated system can often increase available water supplies and reduce costs. The techniques for achieving conjunctive use vary with the specific situation involved. For example, where surface storage is limited or there is none, surface runoff that would otherwise be lost can be stored underground by artificial recharge for later extraction and use. Available surface storage can be used to regulate variable runoff to increase artificial recharge capability. Groundwater can be used to meet peak demands with resultant savings in transmission costs in some cases. Water storage underground minimizes evaporation losses. A degree of natural treatment results from passage of surface water through the soil column in transit to the water table. This is particularly important where polluted surface water or treatment and reclamation of wastewaters are involved. It is emphasized, however, that to achieve full benefits of the conjunctive use potential, the management plan must be based upon thorough considerations of hydrology, geology, and man-induced influences.

A carefully planned program of groundwater extractions with respect to areal pattern, amounts, and timing is required in order to maximize the potential for use of underground storage. The possibility of interference with vested groundwater rights must be recognized and any necessary arrangements made for compensation, either in-kind or monetary.

Conjunctive use has been extensively practiced in parts of Southern California in a variety of ways for many years with a high degree of success. Here, runoff is highly variable, available surface storage is very limited and costly, and groundwater basins are extensive, although the availability of land for artificial recharge operations is now limited. Artificial recharge and underground storage are used for conservation of local

runoff, for storage and distribution of imported water, and for treatment and storage of reclaimed water. Groundwater rights in several basins have been adjudicated. In at least one other basin, adjudication has not been necessary through acquiescence of the water users who have been more interested in assurance of an adequate water supply of good quality than in legal protection of water rights. Equitable physical solutions have been provided in all cases.

The State of California and local agencies are now developing plans to conjunctively use underground storage capacity in Southern California for long-term carryover storage of surplus water from Northern California imported by the State Water Project.

Desalting/Use of Brackish Water

There are modest success stories to relate in agricultural water supply enhancement for the semiarid West. The U.S. Salinity Laboratory and brackish water use programs in Arizona, New Mexico, Texas and other western states have shown significant progress in water management, crop adaptations, soil treatments, and other agricultural techniques for the use of brackish and moderately saline waters. Most western states have sources of largely unused brackish water, both ground and surface, that could be developed for agricultural purposes.

The State of New Mexico is estimated to have about 15 billion acre-feet of saline groundwaters (salinity ranges of 1,500 to 15,000 mg/L TDS). The economic and operational feasibility of using typical saline waters representative of New Mexico groundwaters has been investigated for several years. A variety of crops and cropping systems have been demonstrated to have suitable tolerance for such saline irrigation. Many of the more common field crops grown in the West—small grains, cotton, alfalfa, grain sorghums and others—demonstrate this adaptation.

The processes of desalting have yet to be established as a large-scale solution to the problem of providing new agricultural water supplies. The increasing costs of the very large amounts of energy required for desalting have made this potential less and less attractive. Continued advances in geothermal or solar energy generation processes may provide in the future a way to treat the available brackish to saline waters on a large scale.

Water Reclamation and Reuse

The potential for reusing water, and the requirements for reclaiming it, restoring it to a quality suitable for reuse, and redistributing it among users, is a cycle of legal, engineering, esthetic, and environmental complexity. Yet, since water is not destroyed by use, it is a cycle nature has always provided. The problems are twofold: separation of use and reuse over time and geography, and the persistent pollutants which our civilization manages to insert into the cycle.

The technology for treating wastewater to the point of making it suitable for reuse for irrigation is available, although public health questions about direct reuse for human consumption remain. Certainly the reallocation of reclaimed water to industrial and agricultural users is well within existing technology. However, irrigated areas where significant volumes of reclaimed water could be used are generally at considerable distances, often with ranges of intervening hills or mountains, from the urban areas where large amounts of wastewater are generated, thus adding significantly to the cost. An example is irrigation in the San Joaquin Valley of California, many miles from the metropolitan areas of the San Francisco Bay region and Southern California.

In irrigated agriculture, the increased efficiencies of present practices generally result in full use of applied water, with tailwater recovery and reuse a common practice. Opportunities for improved reuse of agricultural waters still exist on a limited basis, but do not represent significant potentials. Continued research into reuse, and its systematic inclusion in the water resource allocation planning process, are necessary steps in achieving the full potential of this measure.

Improving Existing Project Operations

Many projects, in fact most existing projects for surface water development, were planned and authorized under planning concepts, standards and criteria, economic conditions, projected downstream needs, projected upstream depletions, operational criteria, contractual requirements, and political attitudes that differed significantly from those prevailing today. This is true of the main stem developments on the Missouri River, the Federal

Central Valley Project in California, and the California State Water Project, to mention but three examples.

At the times these projects were originally planned and authorized, little if any thought was given to the potential for increased yield through conjunctive use with groundwater resources, to the potential benefits which could result from integrated operation with other projects on a "systems" basis, to requiring efficient use of water by water service contractors, or to operating the projects on benefit/risk basis, among other potentials for increasing yields.

As stated above, the Missouri River was studied by the U.S. Army Corps of Engineers as a potential source of water for exportation for irrigation in the High Plains area of Colorado, Kansas, Nebraska, New Mexico, Oklahoma, and Texas, under the recently completed federally funded High Plains-Ogallala Aquifer Regional Resources Study. The Missouri is now controlled by six main stem dams and reservoirs—Fort Peck, completed in 1935, and the other five authorized under the Pick-Sloan Plan by the Flood Control Act of 1944. By the authorizing legislation, these projects are committed to navigation, flood control, hydropower, and irrigation, municipal, and industrial uses in the basin states.

The Corps studies indicated that, under the present authorizations and commitments, little if any surplus water would be available for exportation without encroachment on navigation, hydropower generation, and future in-basin uses. However, more recent projections of in-basin uses and depletions are significantly lower, and questions have been raised as to the justification under present conditions for the present allocation of storage and water for the limited navigational use of the Missouri River to Sioux City. More water might be available for both in-basin use and exportation were the allocation and the operational criteria to be changed to accord with today's projected conditions and needs.

Were the Federal Central Valley Project and the California State Water Project, both of which divert from the Sacramento River and tributaries and from the Sacramento-San Joaquin Delta, to be operated as an integrated system with proper regard for hydrologic diversity, there could be a potential increase in yield of 500,000 to 1,000,000 acre-feet per year. Hydroenergy production might also be increased. Conjunctive use with groundwater resources would provide additional benefits.

Any proposal to improve the efficiency of operation of existing projects would require Congressional and state approval. No

doubt there would be strong opposition from some present project beneficiaries. The potential for increased water supply and other benefits seems to warrant the attempt.

Better Allocation of Resources

The discussion above of improved management and operation of existing projects implies reallocations of current supplies and allocation of augmented supplies in accordance with today's needs and conditions. There are also situations where undeveloped resources could be allocated and developed to sustain current uses. Only one example will be discussed here, that of the undeveloped groundwater resources of the High Plains-Ogallala Aquifer region of Colorado, Kansas, Nebraska, New Mexico, Oklahoma, and Texas.

In 1980, of the total of 140.8 million acres in the High Plains area, over 15 million acres were irrigated with groundwater extracted from the Ogallala and associated aquifers. It is projected that 5.4 million acres will revert to dryland farming or be abandoned by 2020 because of physical or economic exhaustion of the underlying groundwater resources if no new remedial actions are taken. The rate of reversion will accelerate thereafter.

Of the more than 125 million acres of nonirrigated land in the region, another 17 million acres were in dry cropland. Of the total area, almost 30 million acres are classified as marginal for irrigation, but are suitable for some types of dryland production such as grazing. These nonirrigated areas are underlain with groundwater resources in amounts that vary with location.

It is suggested that the groundwater resources underlying the marginal lands, and some of the other lands not especially well suited for irrigation, might be developed and conveyed over time to the presently irrigated lands and the nonirrigated prime lands that may go under irrigation. Of course, it would be necessary to obtain, by direct purchase or by condemnation, the water rights or surface development rights of those lands and owners from which the underlying waters would be purchased. Some changes in state laws would be necessary. There would be some legal/institutional difficulties to be overcome.

This concept would have several advantages:

· The lands from which the water would be taken could remain in their present uses or other dryland uses.

· There would be a number of relatively small projects which could be implemented at the appropriate times with respect to the declining availability of underlying water for the recipient lands, as contrasted to one or two very large importation projects. Implementation could be accomplished in stages.

· The investments required at any one time would be relatively small.

· Implementation could be accomplished by local public agencies rather than the master regional agency required for a large importation project.

· The elevation differences are relatively small and pumping costs would be much less than for importation.

· Surface storage reservoirs would not be required, thus minimizing evaporation losses.

· Total costs should be significantly lower.

· At least some of the owners no doubt would welcome the money derived from sale of their groundwater rights, which they might not ever exercise on their own behalf.

Groundwater Management/Recharge

Artificial recharge is frequently touted as a principal technique to be applied for alleviation of the present water supply deficiencies in the West. Artificial recharge is already being widely applied throughout the West. However, application of this technique to new situations will depend on the availability of water for increased recharge that would otherwise be lost for beneficial use. There is now little surface water wasted in the West on a regional or river basin basis; most surface water is already being used for one or more beneficial purposes. The feasibility of artificial recharge also depends on the availability of sufficient usable underground storage and transmission

capacities. Effective artificial recharge requires land and physical facilities. The investment and operation and maintenance costs may be substantial. In some states, there are legal questions as to the ownership of and control over the recharged water, i.e., the right to recapture.

There are opportunities for further augmentation of usable supplies in the West through artificial recharge, but they are not widespread. Artificial recharge will be particularly valuable for the underground storage of imported and reclaimed waters.

Other Possibilities and Research Needs

One very new technology may have potential application in areas throughout the West that have already seriously depleted available groundwaters. Methods are undergoing testing in the south area of the Texas High Plains for the secondary recovery of additional groundwater supplies from the unsaturated zones of an aquifer. The concept is that in many aquifers, as much or more water remains in the formations after depletion by usual extraction methods, due to molecular or capillary attraction, as was removed by normal gravitational (pumping) forces. This could represent a very significant supplemental water supply if this ongoing research demonstrates both a technical and cost-effective capability.

A variety of techniques for reducing nonproductive losses of water by evaporation, transpiration, and/or by losses to runoff or deep percolation to nonrecoverable areas such as saline sinks or aquifers are being investigated. Methods to improve infiltration and deep percolation on-site, and to reduce losses to nonproductive deep rooted vegetation like phreatophytes and noxious brush species, hold out strong prospects for local water enhancement in the West, but more research is needed.

Conclusions

Certainly there are no "quick fixes" for new water supplies for the semiarid West. Nor, to continue in the vernacular, are there any "free lunches". With complete rational planning and management, and no objections from the source areas, interbasin diversions of water could be achieved to the probable benefit of all concerned. The reality is that even on a small scale, the

probabilities are not great. Fear of exploitation on the part of export areas, and an appearance of greedy provincialism in some areas seeking imports, combine to create almost impenetrable barriers to the successful implementation of any diversion scheme. Lack of available funding now precludes large-scale structural solutions to problems of water supplies for irrigation.

Other processes show promise, but the potential for a significant breakthrough in local water supply or large-scale water supply augmentation for the West in the foreseeable future is limited.

A state-of-the-art evaluation of a large set of emerging technologies for enhancing local water supplies, while not consistently discouraging or pessimistic, nevertheless offers small relief for the major irrigated agricultural production areas of the West, which are presently dependent on seriously overdrafted groundwater sources or very limited surface water supplies.

Significant projects are still underway for agricultural water supply enhancement. Examples are the extensive weather modification and precipitation management research programs; the adaptation and use of brackish and saline waters for certain crops; and the secondary recovery of additional waters from aquifers where gravitational waters have already been depleted due to overdraft. These and other methods may provide some temporary and partial water supply sources for western agriculture, but none provide long-term solutions to the water crisis. These are nonetheless the only solutions, limited as they are, that appear to be implementable for many years.

Discussion:
Herman Bouwer

New water supplies or water importation schemes usually mean transferring water from places where the supply exceeds the demand to places where the demand exceeds the supply, or from places where the economic returns from the water use are low to where they are high, all in accordance with the second fundamental law in hydraulics: water runs uphill—to money! This does not bode well for agriculture, which traditionally is accustomed to inexpensive water for irrigation. Rather than for irrigated agriculture to acquire additional water supplies, current trends seem to be more in the opposite direction, i.e., sales of irrigation water rights for municipal and industrial uses.

The authors have done an excellent job in summarizing the various large water-transfer schemes that have been proposed over the years, and other possibilities for augmenting local water supplies. One source not mentioned is icebergs that would be towed from the Antarctic. Is this no longer a viable concept? Accurate cost figures for large water transfer schemes are difficult to obtain. Preliminary estimates all indicate, however, that costs are high: capital costs of several thousand dollars per acre-foot per year capacity, and total costs of several hundred dollars per acre-foot at the aqueduct or reservoir. To this must be added the cost of further distribution of the water to the points of use. For a simple project like the Central Arizona Project where water will be pumped from the Colorado River and transported a few hundred miles into south central Arizona, construction costs are already about 2.4 billion dollars for a capacity of 1.2 million acre-feet per year (or about $2,000 per acre-foot per year), and this does not include the cost of getting the water from the main aqueducts and reservoirs to the points of use. The cost of the water to consumers is projected at $52 per acre-foot for agricultural users and $82.50 for municipal and industrial users, again at the main aqueduct. These figures will soon be revised, probably upward. The cost of water in southern California from the California Aqueduct is about $100 per acre-foot. This figure could double in 1983, as new contracts for electric

power will be negotiated. New projects can be expected to be a lot more expensive.

Man-made obstacles (legal, social, environmental, etc.) to large transfer projects seem more difficult to overcome than the technical problems, which can be solved by good engineering. It should be possible, however, to develop long-range projections of water needs for selected basins, to identify basins of water surplus and water deficit, to design water transfer projects, and, if economically and environmentally attractive, to build them. The Sporhase decision (Sporhase v. Nebraska, U.S. Supreme Court, 2 July 1982), which declared groundwater an article of interstate commerce subject to congressional regulation, may help overcome political opposition from water surplus states against export of water to deficient areas. Long-term economic and social values for the life of the project should be considered rather than payout period economic aspects which, because of present cost levels for water, are almost always unfavorable. To translate from the Dutch, "A nation that lives builds for its future"!

If we leave out the element of moving water over great distances, development of new water supplies simply means transferring water from a use with a low economic return to one with a higher economic return. This, of course, includes water conservation, where losses and wastes of water are reduced and put to more beneficial use. In view of the costs and the many difficulties of water transfer schemes, water conservation is increasingly considered as the best and most immediate solution to problems of water shortage. The authors allude to water conservation and increased irrigation efficiency, as do other chapters in this volume. However, further discussion of some opportunities for water conservation seems warranted.

One such opportunity is to reduce water use by agriculturally nonbeneficial vegetation such as phreatophytes in floodplains. Phreatophytes have been estimated to cover about 15 million acres in the western states and to consume about 25 million acre-feet of water per year. This is the equivalent of 20 Central Arizona Projects! Complete eradication of the phreatophytes, as advocated a few decades ago, is not compatible with wildlife and scenic considerations. Thus, selective removal will be more acceptable. Proper control can be achieved by keeping the phreatophytes away from the water (by selective cutting and floodplain management), or by keeping the water away from the phreatophytes (by lowering groundwater levels or reducing seepage from stream channels). Care should be taken that

replacement vegetation does not use appreciable amounts of water. With the high cost of imported water, saving water by phreatophyte control may become attractive.

Runoff farming offers great potential for the millions of acres of marginal lands with insufficient rainfall or irrigation water for normal crop production. Crops can then be grown in widely-spaced rows at the base of contour strips that have been treated chemically or mechanically to increase runoff from rainfall, thus concentrating the rain on the crops. The systems can be designed to yield more runoff than can be used by the crops for evapotranspiration, thus increasing deep percolation from the crop rows and producing more groundwater recharge. Runoff farming and replenishment irrigation have great potential for the management of abandoned irrigated land, which otherwise could develop problems of dust and tumbleweeds. The crops should be deep-rooted or drought-tolerant to survive long periods of no rain. Supplemental irrigation may be desirable.

Reuse of wastewater, particularly municipal wastewater, requires considerable advanced planning to ensure that the treatment plants and the irrigated fields are not too far apart, and that land treatment or groundwater recharge opportunities can be utilized. If partially treated wastewater can be put underground with infiltration basins and pumped from wells after it has moved through the vadose zone and aquifer to become "renovated water", the cost of treating the wastewater to meet the public health, agronomic, and aesthetic requirements for unrestricted irrigation can be greatly reduced. There are also increasing trends toward local or on-site reuse of municipal wastewater for landscape irrigation, golf courses, cemeteries, etc.

Last but not least, there is irrigation efficiency, which often is the center of attention because irrigation uses so much water (almost 90 percent of all water in Arizona, 85 percent in California) and field irrigation efficiencies are low. Many people have the misconception that a field irrigation efficiency of 60 percent means that only 60 percent of the irrigation water is effectively used, and 40 percent is wasted. Of course this is not true. The forty percent of the water not used by the crop in this case is in the form of runoff at the lower end of the field and/or of deep percolation from the root zone. Both types of water can be recovered and reused again. For this reason, the irrigation efficiency of entire irrigation districts or irrigated valleys is much higher than the efficiencies of individual fields. As the saying goes, "the upper basin's inefficiency is the lower basin's water resource."

The real loss of water is the consumptive use or evapotranspiration, and that does not change much with irrigation efficiency. However, if the irrigation efficiency is increased, for example from 65 to 85 percent, less energy is needed for pumping, and higher yields are generally obtained, because of better water management and reduced leaching of fertilizer. This is really the main purpose of increasing irrigation efficiency: to increase crop yield per unit of water consumptively used.

It is, of course, also possible to reduce evapotranspiration by not growing crops in the hottest part of the year (late season cotton, winter vegetables instead of summer crops, etc.) and increase water use efficiency that way. However, the main prospects for water saving in irrigation lie in increasing crop yields. Average crop yields typically are only about 20 percent of record values, so there is still room for improvement in crop management. Also, research should be greatly stepped up to create new, high-yielding varieties, using new developments in genetic engineering. New approaches such as the use of growth hormones and biostimulators should be investigated with vigor. If we can double the yield per acre, the same crop can be produced with half the land, essentially half the water, and essentially half the salt load on the underlying groundwater due to deep percolation. Thus, growing the proverbial two blades of grass where only one would grow before is still the name of the game.

Discussion:
Marion Marts

The chapter by Banks, Williams, and Harris is a realistic and comprehensive assessment of the prospects for developing large-scale new agricultural water supplies in the semiarid West. The authors conclude that the prospects are poor in the foreseeable future. This discussant shares this conclusion, and indeed with respect to large-scale importation would argue that the prospects approach zero. Let me elaborate on this latter point a bit, and then proceed to speculation on some broader issues.

Large interregional transfer of water is an idea that won't go away, but whose time refuses to come. As the authors point out, the inhibiting factors have always been strong; my thesis is that they are growing stronger over time, so that whatever prospects once existed are fading. The Columbia River illustrates the point nicely.

Then Assistant Secretary of the Interior William Warne's famous "climb the ladder of rivers to the north" speech stimulated the first major review of large-scale interstate water transfer: the Bureau of Reclamation's United Western Investigation, which in 1950 and 1951 reported on a reconnaissance of a variety of possibilities for transferring "surplus" Pacific Northwest water to the Southwest, but concluded that tapping anything north of the Klamath River was economically infeasible. It is interesting that even in that heyday of water project development, economic feasibility was a constraining criterion. Also interesting was the fact that certain northwest waters—the Rogue River and Flathead, Pend Oreille, and Coeur d'Alene Lakes—were declared sacred and off-limits. The investigation was quickly and quietly terminated when Northwest congressmen discovered what was going on.

A blizzard of proposals followed the 1963 Arizona v. California decision. Anyone with a roadmap and pencil could play the game. By 1969, Bingham inventoried 14 interregional and 10 international proposals, and there were many variants of these. Banks and colleagues describe four of the major proposals as illustrations. Senator Jackson of Washington saw fit to take the Columbia out of the game by imposing a congressional moratorium, which effectively stopped federal agencies from planning to rearrange the Columbia.

While high cost can be cured by massive subsidy, and political clout can erode over time, a new and very fundamental element has been added. This element is, surprisingly, shortage. Competition within the Columbia River Basin for water for hydropower, for anadromous fish, and for additional irrigation has become fierce—reinforced by two drought years in the 1970s. The Northwest Regional Power Council, for example, has accepted a calculation that provision of adequate spring flows for the juvenile salmon migrating to sea will cost in the order of 500 to 550 megawatts of firm power. In the same vein, irrigation expansion will impose annual power costs on the region amounting to more than $100 per acre irrigated—in some cases more than $200—from a combination of lost generation and consumption of electricity for pumping. Hydropower operation is now a claimant for water that once was surplus during the high flow season (late spring and early summer) as a result of upstream storage, the ability to sell electricity to the Southwest via the Intertie, and the gradual transition to peaking mode for the hydropower plants. Indian claims to water and fish are far from quieted, and will complicate the planning picture for years.

All of these factors mean that there are substantial opportunity costs of diverting Columbia River water which must be included in any responsible benefit-cost analysis. The days when outflow to the seas could be considered surplus are gone. By extension, the principle of opportunity cost applies to all available rivers, although the details and values will vary. This is a powerful inhibition on any river pooling scheme. The authors illustrate this principle in the case of the Missouri River by pointing to the trade-off between navigation and out-of-basin diversion.

Banks and his colleagues proceed from the impracticality of large water transfers to an excellent review of the possibilities of improving the efficiency with which available water supplies are used, and point out institutional factors—water laws, property rights, custom, etc.—which inhibit increased efficiency. I applaud this discussion and can only add, let's get on with it.

But all of this suggests some interesting speculation. If my thesis that large water transfers have been priced out of the market is correct, does this mean that federal intervention and federal subsidy have lost their efficacy? What potential irrigation projects remain if we can't afford to augment? Does southern Idaho claim more Snake River water for irrigation regardless of the downstream costs in lost hydropower and fish? And into what maze does enhanced efficiency lead us? Will Wyoming surrender its rights to irrigate pastures in high mountain valleys to expand the intensive agriculture of the Imperial Valley? The social cost of interregional efficiency, like the cost of augmentation, may be excessive. The sobering fact is that state boundaries and state water codes constrain both efficiency and interstate transfer.

If big augmentation and big efficiency are beyond us, what then? Perhaps the time has come to accept the proposition that there is an inevitable equilibrium between availability and use of western water resources. There will no doubt be some modest additions here and there using locally available supplies, and some abandonment here and there as water tables drop or salinity rises, but these are just perturbations on the way to equilibrium. We may have seen the last of the large new irrigation projects in the West. The next ones may be in the Mississippi Valley, as we proceed to emulate the Italian experience in the Po river basin.

A final comment about long distance transfer. We are habituated to think of interstate and international water transfers only in terms of large scale, and chiefly for irrigation. This creates the twin problems of large cost and low repayment capacity, compounded by the perception of unfair interregional competition. If we have long distance transfers in the foreseeable future, they will more likely be of modest volumes for specific and high-value uses, such as coal gasification or transport, or municipal use, analogous to oil and gas pipelines and perhaps even sharing the latters' rights-of-way. Such systems would have much greater social and economic acceptability than would canals built to move the equivalent of the Colorado River, and might help retard the transfer of irrigation water to industrial and urban areas in water-short areas.

Macro-think has brought no large interstate water transfers in the three decades since the United Western Investigation, and may actually have impeded the search for more efficient allocation and use of available water. It may be time to try micro-think.