Winter Issue of The Bridge on Sustainable Water Resources

The Sustainability of Water Resources in the Colorado River Basin

Studies of Colorado River flows have called into question traditional assumptions about long-term mean flows and availability. 

The hydrologic limits of U.S. water supply systems, conflicts over shared water resources, and drought-induced water shortages are increasingly prominent topics of conversation, not only in the water science and engineering communities, but also in media and public policy discussions. Examples of recent books on these topics include Unquenchable: America’s Water Crisis and What To Do About It (Glennon, 2009) and Running Out of Water: The Looming Crisis and Solutions to Conserve Our Most Precious Resource (Rogers and Leal, 2010). Whether or not the nation is indeed facing a water “crisis” may be open to debate, but there is a growing sense that providing reliable water and related services to a full range of diverse users, especially during periods of drought, is becoming increasingly difficult.

Supply, demand, and drought are prominent issues throughout the Colorado River basin. In 2005, the National Research Council (NRC) convened a panel of experts to review and evaluate the scientific database on the climate and hydrology of the Colorado River basin and the long-term implications of hydro-climate variability for operating water projects and meeting obligations for water delivery. This article provides a summary of findings and recommendations from the NRC report (NRC, 2007) and some observations about long-term prospects for providing water supplies in the rapidly growing southwest region of the country.

The Colorado River and Basin

The headwaters of the Colorado River are in Rocky Mountain National Park northwest of Denver (Figure 1). The river flows westward through Glenwood Canyon toward Grand Junction, Colorado, where it is joined by the Gunnison River. Once it enters Utah, the Colorado is joined by the Green River, a major tributary that drains parts of Colorado, Utah, and Wyoming. Just before it flows into Lake Powell, the Colorado is joined by the San Juan River, which drains the San Juan Mountains and the Four Corners region. Fifteen miles below Glen Canyon Dam, the river passes Lees Ferry, Arizona—a river gauging station that marks the legal demarcation point between the upper and lower Colorado River basin.

Figure 1

From there, the river flows through Grand Canyon National Park and then is joined by the Virgin River just before it flows into Lake Mead. Below Hoover Dam, the center of the Colorado streambed marks the boundary between Arizona and California. The Colorado then enters Mexico on its way to the Gulf of California. In Mexico, however, flows are often fully consumed by irrigated agriculture and, in some years, the river does not reach the gulf.

The river basin drainage area covers portions of seven U.S. states—Colorado, New Mexico, Utah, and Wyoming in the upper basin and Arizona, California, and Nevada in the lower basin. The Colorado River is primarily a snowmelt-driven hydrologic system. Roughly 90 percent of the river’s flow is derived from snowmelt from precipitation in three upper basin states, Colorado, Utah, and Wyoming. However, most of the demand and use of the flows are in the lower basin states, Arizona, California, and Nevada (Hundley, 1975).

Based on measurements at Lees Ferry, the mean flow of the Colorado River is 15 million acre-feet (MAF) per year1—a much smaller volume of water than in other major U.S. river systems, such as the Columbia or Mississippi rivers. However, the Colorado flows through what author Wallace Stegner described as “the dry core” of the arid western United States and is the largest source of surface water in this very large region. Roughly 30 million people depend on the Colorado for drinking water, and its waters are essential to farmers, tribes, industries, anglers, power distributors, and rafters.

Drought, Stream-flow, and Storage in the Early 2000s

Large variability in flow, both seasonal and inter-annual, is a prominent feature of the Colorado River. As Figure 2 shows, annual flows often depart substantially from 15 MAF/yr. The figure also shows some markedly wet and markedly dry periods during the 20th century. For example, there was a very wet period at the beginning of the century, a drought in 1976–1977, and El Niño conditions (which generally entail heavy winter precipitation in much of the western United States) in the early 1980s.

Figure 2 also reflects pronounced drought conditions throughout the upper basin in the early 2000s, especially in the (water) years2 of 2000–2004 when inflows into Lake Powell were well below 15 MAF/yr; in water year 2002, for example, the flow was more than 50 percent below average (Fulp, 2005). These low-flow conditions had many hydrologic effects, including a sharp decrease in the amount of water stored in Lake Powell, which dropped to its lowest level since 1969 when the reservoir was initially filling. Water storage in Lake Mead also dropped to a level not experienced since the 1960s (Fulp, 2005).

Figure 2

Lake Powell and Lake Mead, which together represent roughly 90 percent of the surface water storage capacity in the Colorado River basin, are crucial to ensuring that legal water-delivery obligations are met during periods of drought. The most prominent of these obligations, which is enshrined in the Colorado River Compact of 1922, calls for the upper basin states to provide an aggregate flow of 75 MAF to the lower basin states over any 10-year period. Thus, a sharp drop in Lake Powell water storage is a legal, scientific, and public policy matter of the utmost concern and was an impetus for the NRC study and report.

Population Growth and Increasing Water Demands

A reliable water supply is a function of water-supply issues discussed above and water demand. Discussions during the NRC study with state water managers revealed that growing water demand was as important as drought and variability in the water supply. In the 1990s, the states in the Colorado River basin had the highest rates of population growth (by percentage) in the country. The four fastest growing states were Nevada, Arizona, Colorado, and Utah. In terms of absolute growth, California added more than 4 million people (a 13.8 percent increase). The fastest growing major U.S. metropolitan area was Las Vegas, which increased at the remarkable rate of 83.3 percent.

Discussions about population growth and the limits of western water supplies—or lack thereof, depending on one’s point of view—date back more than 100 years. Relationships between water supply and demand are complex, and trends can change in either or both, sometimes rapidly. Many water conservation programs in the Colorado River basin have successfully reduced per capita water demand and improved water efficiency. At the same time, as populations continue to grow, attendant increases in water demand eventually negate those savings.

Figure 3

Figure 3 shows the effects of population growth on water supplies in the southwestern United States. Population growth during the 20th century has steadily eroded the historic cushion between the region’s water supply and water demand. Thus, for the first time, aggregate water demand has exceeded available water supplies in some years.

Regional Climate Trends

Because Colorado River flows are derived primarily from snowmelt in high elevations in the upper basin, precipitation in the lower basin, which can have significant local impacts, does not affect flows in the large upstream tributaries or water storage in Lakes Powell and Mead. Thus, changes in winter precipitation and temperature patterns in the upper Colorado River basin are of great concern in terms of long-term water availability.

The 2007 NRC report includes reviews of 20th century precipitation and temperature trends in the Colorado River basin. Although precipitation records for the upper basin show great variability over that period, the committee concluded that “there is no significant trend in inter-annual variability of precipitation over the past 110 years.”

However, a review of temperature data for the entire river basin shows that “since the late 1970s, the Colorado River region has exhibited a steady upward trend in surface temperature. The most recent 11-year average exceeds any previous values in the over 100 years of instrumental records” (NRC, 2007). The report goes on to say that “the Colorado River basin has warmed more than any [other] region of the United States . . . This warming is well grounded in measured climatic data, corroborated by independent data sets, and widely recognized by climate scientists throughout the West.”

In addition to instrumental records of past climate data, the report includes reviews and summaries of model-based climate forecasts based on studies of precipitation and temperature futures across the basin. An early study in 1979 estimated that a temperature increase of 2°C would, by itself, result in a decrease in mean Colorado River flows of 29 percent (Stockton and Boggess, 1979). Since then, numerous other climate modeling studies have also concluded that modest temperature increases in the upper basin could result in marked reductions in stream-flow and inflows into Lake Powell (e.g., Christensen et al., 2004).

The NRC committee came to the following conclusions:

Collectively, the body of research on prospective future changes in Colorado River flows points to a future in which warmer conditions across the region are likely to contribute to reductions in snowpack, an earlier peak in spring snowmelt, higher rates of evapotranspiration, reduced late spring and summer flows, and reductions in annual runoff and stream-flow.

Reconstructions of Stream-flow Based on Tree Rings

The NRC committee also reviewed studies reconstructing long-term (on the order of several centuries) stream-flow based on the annual growth rings of coniferous trees (pines and firs) in the region. Coniferous trees growing at lower elevations on well-drained slopes with southern exposures are particularly well suited for these reconstructions (Woodhouse et al., 2006). Dendrochronologists correlated annual increments of tree-ring growth in pines and firs with hydroclimatic variability, including reconstructed historic river flows.

Figure 4

The first tree-ring based reconstruction of Colorado River flow was published in 1976 (Stockton and Jacoby, 1976), and several more followed. An example of one reconstruction by Woodhouse and colleagues (2006) is shown in Figure 4. Based on this study and other tree-ring based reconstructions of past Colorado River flows, several conclusions may be drawn (NRC, 2007; Woodhouse et al., 2006):

  1. Long-term Colorado River mean flow calculated over hundreds of years is significantly less than the 15 MAF/yr figure based on 20th century flows recorded at Lees Ferry.
  2. The early decades of the 20th century, one of the wettest periods in the entire reconstruction, was characterized by high-flow conditions.
  3. The reconstructed records reveal that droughts prior to the 20th century lasted much longer than droughts in the early 2000s.

Relatively wet conditions across the upper basin in the early 20th century turned out to have legal implications of great historical importance. The Colorado River Compact—the cornerstone legislation for water-management treaties, acts, allocations, and contracts—was signed in 1922. At that time, based on flow data collected during that wet period by the U.S. Bureau of Reclamation, it was assumed that the mean annual average flow of the Colorado River was 16.4 MAF/yr (Hundley, 1986), and the division of the Colorado’s flows between the upper basin states and lower basin states, with 7.5 MAF/yr for each, was based on this assumption.

Over time, and with additional data from both Lees Ferry and reconstructed flow data based on tree-ring analyses, it has become clear that the average annual flow is less than 16.4 MAF/yr. This hydrological reality has sobering implications for areas that plan to base future economic development on water rights, especially areas in the upper basin that do not have high priority rights under the Colorado River Compact and other water-use and sharing agreements. The effects on these areas could be exacerbated if future changes in climate further reduce Colorado River flows (Kenney, 2010).

Options for Augmenting Water Supplies

In considering limited water supplies in the Colorado River basin and possible short- or long-term reductions in water availability, it is natural to consider how water supplies might be augmented. The traditional approach was by constructing multi-purpose dams and storage reservoirs. However, for a number of reasons, including costs and potential environmental impacts, prospects for constructing large-scale water projects today are much less likely than in the past.

As a result, water managers are considering other strategies, some novel and some that have been the subject of experiments for decades. Alternatives include weather modification (i.e., seeding clouds with various agents, such as silver iodine or dry ice, to induce or enhance precipitation), desalination, removal of water-consuming plant species (e.g., tamarisk), agricultural and urban water conservation, changes in water pricing policies and rate structures, wider use of reclaimed wastewater, and off-stream water banking (i.e., storing water underground in aquifers for later use).

The latter option, off-stream water banking, is a promising technique for improving the efficiency of water management. The state of Arizona established the Arizona Water Banking Authority in 1996 to store Colorado River water by recharging groundwater. Arizona, California, and Nevada have also engaged in creative interstate agreements whereby one state banks groundwater in another state for withdrawal at a future date.

Although lessons and results tend to be site-specific, many conservation programs in the region, such as those that emphasize new landscaping techniques and technologies, have resulted in reductions in urban water demand. The U.S. Bureau of Reclamation, state and municipal water agencies, the private sector, and nongovernmental entities have all promoted and participated in these efforts, which will surely continue to be refined and improved.

Another means by which urban water supplies might be augmented is via the sale, transfer, or lease of water rights from agricultural users to growing urban areas. Historically, the majority of water diversions have been for the purpose of irrigation, and water diverted to irrigated agriculture in the western United States represents a considerable supply.

Today, agriculture-urban water transfers are taking place throughout the Colorado River basin, including in Denver, Las Vegas, and Phoenix. In strictly monetary terms, these transactions often represent “win-win” situations for buyers and sellers, as water typically shifts from lower value agricultural uses to higher value urban uses.

However, these transactions are not without costs and limitations. For example, the direct effects associated with water rights moving away from agriculture include the reduced capability of domestic food production. In addition, such transfers usually entail “third-party” effects beyond those that accrue to buyers and sellers. Examples include reduced agricultural return flows that support riparian ecosystems and reduced business and sales by merchants in agriculture-related sectors (NRC, 1992).

Third-party effects that harm rural communities and valuable ecosystems may well prevent some transfers of water to western cities.  Furthermore, even though the amount of water diverted to irrigated agriculture in and near the Colorado River basin is considerable, the volume of agricultural water is finite, so transfers may not be an option at some point in the future.

In short, none of these options resolves the fundamental tension between limited supplies and steadily growing demand, which will inevitably require costly and controversial trade-offs. In addition, the combination of increases in population and water demand also reduces the region’s capacity to cope with droughts and water shortages.

Findings and Conclusions

In the late 20th century, there was a strong trend of rising mean temperature in the region. The preponderance of evidence—both instrumental data and projections based on modeling—strongly suggests that warmer temperatures will reduce future Colorado River stream-flow and water supplies. In addition, tree-ring based reconstructions of Colorado River stream-flow have shown that extended droughts are likely to occur. These droughts could be even more severe than the drought of the early and mid-2000s, which resulted in sharp reductions in inflows into Lake Powell and prompted concerns about meeting water-delivery obligations. These studies of Colorado River flows have called into question traditional assumptions about long-term mean flows and availability.

Today, the Colorado River basin continues to be home to the fastest growing states in the nation adding to the strains on limited water supplies. Measures to extend and conserve water supplies, such as conservation programs, changes in landscaping practices and related technologies, aquifer storage, and desalination, have improved water use efficiencies, and agriculture-urban water transfers have increased water supplies available to urban areas. However, the benefits of all of these options are limited. Rapid population growth has already increased aggregate water demand to the point that it exceeds the available water supply in some years.

Future choices for water use will no doubt unfold in complex, perhaps unanticipated, ways, and future warming and droughts may reduce the availability of water resources even further. Current scientific understanding of the river’s historical flows and regional droughts, coupled with the potential for future reductions in flows, raises fundamental questions about the sustainability of current population growth and development. Moreover, some existing paradigms and principles that have governed Colorado River water use in the past will undoubtedly have to be adjusted to fit these realities.


Christensen, N., A. Wood, N. Viosin, D. Lettenmaier, and R. Palmer. 2004. The effects of climate change on the hydrology and water resources of the Colorado River Basin. Climatic Change 62(1): 337–363.

Fulp, T. 2005. Response of the System to Various Hydrological and Operational Assumptions: Reclamation Modeling Results. University of Colorado Natural Resources Law Center, Twenty-sixth Annual Conference.

Glennon, R. 2009. Unquenchable: America’s Water Crisis and What to do About it. Chicago, Ill.: Island Press.

Hundley, N. Jr. 1975. Water and the West: The Colorado River Compact and the Politics of Water in the American West. Berkeley, Calif.: University of California Press.

Hundley, N. Jr. 1986. The West Against Itself: The Colorado River—An Institutional History. In New Courses for the Colorado River: Major Issues for the Next Century, edited by G. Weatherford and F.L. Brown. Albuquerque, N.M.: University of New Mexico Press.

Jerla, C., K. Morino, R. Bark, and T. Fulp. 2011. The Role of Research and Development in Drought Adaptation on the Colorado River Basin. Pp. 423–438 in Water Resources Planning and Management, edited by R.Q. Grafton and K. Hussey. West Nyack, NY: Cambridge University Press.

Kenney, D. 2010. Rethinking the Future of the Colorado River. Draft Interim Report of the Colorado River Governance Initiative, Boulder, Colo.

NRC (National Research Council). 1992. Water Transfers in the West: Efficiency, Equity, and the Environment. Washington, D.C.: National Academy Press.

NRC. 2007. Colorado River Basin Water Management: Evaluating and Adjusting to Hydroclimatic Variability. Washington, D.C.: National Academies Press.

Rogers, P., and S. Leal. 2010. Running out of Water: The Looming Crisis and Solutions to Conserve our Most Precious Resource. New York: Palgrave Macmillan.

Stegner, W. 1954. Beyond the Hundredth Meridian: John Wesley Powell and the Second Opening of the West. Boston, Mass.: Houghton Mifflin.

Stockton, C., and G. Jacoby. 1976. Long-term surface-water supply and streamflow trends in the Upper Colorado River Basin based on tree-ring analyses. Lake Powell Research Project Bulletin 18: 1–70.

Stockton, C., and W. Boggess. 1979. Geohydrological Implications of Climate Change on Water Resources Development. U.S. Army Coastal Engineering Research Center, Fort Belvoir, Va.

U.S. Bureau of Reclamation. 2011. Colorado River Basin Water Supply and Demand Study: Interim Report No. 1. Available online at http://www.usbr.gov/lc/region/programs/crbstudy/report1.html.

Woodhouse, C., S. Gray, and D. Meko. 2006. Updated streamflow reconstructions for the Upper Colorado River Basin. Water Resources Research 42(W05415): 16 pp.



1 An acre-foot is the volume of water that covers an acre to a depth of one foot. It is roughly equivalent to 326,000 gallons.

2 A “water year” begins on October 1 and ends on September 30 of the following year.

About the Author: Jeffrey Jacobs is a scholar with the National Research Council Water Science and Technology Board.