The University of Arizona

The Energy-Water Nexus in Electricity Generation in the Southwest | CLIMAS


The Energy-Water Nexus in Electricity Generation in the Southwest

Wednesday, February 8, 2012

Freshwater is a precious resource, especially in the arid and drought-prone Southwest. But what you may not know is that the biggest user of freshwater in the U.S. is not our everyday needs, or even farms, but power plants. In 2005, the nation’s thermoelectric power plants—which account for about 90 percent of the electricity in the U.S.—withdrew more than four times as much water as all U.S. residents (Averyt et al.,2011)!

What’s more, although 99 percent of those withdrawals nationwide were from surface water, in the Southwest, surface water is relatively scarce and thermoelectric power plants have been forced to use groundwater, which then raises concerns over aquifer depletion. In 2008, power plants in the Southwest withdrew an average of 125 million to 190 million gallons of groundwater per day—the equivalent of a foot of water on about 500 football fields (Averyt et al.,2011)! However, it is important to keep in mind that water that is withdrawn is not the same as water that is consumed—evaporated or otherwise lost as liquid water. Most of the amount that is withdrawn (depending on the system) is returned to the stream, lake, ocean, or aquifer. In fact, in 1995, 132 billion gallons of freshwater was withdrawn for thermoelectric power plants per day, and of that amount all but about 3.3 billion gallons was returned to the source each day (DOE report to Congress, 2006).

Figure 1. Palo Verde Nuclear Generating Station located 50 miles West of Phoenix. Image from US Nuclear Regulatory Commission.

So exactly how does a thermoelectric power plant use water? Coal, natural gas, nuclear, and geothermal plants are all thermoelectric. They use heat, such as from burning coal, to create steam (from water) which then drives a turbine, creating electricity. After the steam passes through the turbine, it is cooled, using either withdrawn water or air, so that it condenses and can be reused. The cooling process is where almost all of the water usage takes place in the plants, as some of the water that is withdrawn is lost through evaporation.

There are three types of systems used to cool and condense steam in thermoelectric plants. “Once-through” systems withdraw the most water because they use the cooling water only once before discharging it to a river or pond, often at a higher temperature, which can have negative effects on ecosystems. However, such systems only “lose” (consume) a small fraction of the cooling water through evaporation, as most of the water is discharged. The majority of plants built since 1970 use “recirculating” systems which withdraw about 5% of the water withdrawn by once-through systems, but most of the water is lost to evaporation. And dry-cooled systems, which use air instead of water to cool the steam, use almost no water at all. However, they are less efficient in hot areas, where efficiency can drop by 25%—leading to increased cost and emissions. Thus, these regions are turning more and more to hybrid systems that use a dry-cooled system during the winter when the ambient temperature is cool and a wet-cooled system in the summer.

The above information is included in two recent reports, one released by the Pacific Institute and the other by the Union of Concerned Scientists (UCS). The report by UCS is the first of a series from the Energy and Water in a Warming World Initiative, a collaboration between UCS and numerous independent experts to synthesize policy-relevant research related to water use by power plants in a warming world. The initiative seeks to illustrate the water demands of these plants, especially in the Southwest and Southeast, to motivate low-carbon, low-water energy solutions.

In the Southwest, these types of reports are especially useful because they outline the type of cooling systems used by power plants in the region, where power plants get their water from (surface water, groundwater, wastewater, etc.), and the areas where power plants drive water-supply stress. According to the reports, most plants in the Southwest use recirculating systems. Several plants, mostly along the coast of California, use once-through cooling systems; however, since they are on the coast they use mostly ocean water, not freshwater. Some dry-cooled systems, which use almost no water, are located in northern California and near Las Vegas, most likely due to technological advancements of plants built after 1970.

Sources of water for thermoelectric power production vary throughout the Southwest based on location (Figure 2). For example, the Upper Colorado and Texas-Gulf regions use almost all surface water, whereas the Great Basin and Rio Grande areas use much more groundwater. What’s most interesting, however, is that the only region in the U.S. that gets a large portion of its cooling water from wastewater is the Lower Colorado River Basin. This reflects the usage of the Palo Verde Nuclear Generating Station—the largest nuclear generation facility in the U.S. and the only nuclear power plant in the world that uses treated wastewater—located outside Phoenix (Nuclear Power Industry News, 2010). Even though the plant uses wastewater instead of fresh water, it is still “consuming” water, because had that wastewater not been sent to Palo Verde it would have been put back in the ground where it would percolate through the soil and replenish aquifers. Is this the trade-off we’re asking people to make: use water to power our communities or put water back into aquifers?

Figure 2. Water consumption sources for each region. Figure from UCS report.


Climate change is expected to increase drought frequency and reduce runoff in the Southwest, where surface water is already scarce. And as population grows and increases the demand on water supplies, the needs of thermoelectric power plants will only amplify water stress. The high costs of retrofitting current plants to use dry-cooled or hybrid systems means that alternative energy sources need to be assessed for their low-water potential. Most low-carbon energy choices, such as wind and solar photovoltaics, are inherently low-water systems as well, and are thus the best option for new power plants in the Southwest.

The next report from the Energy and Water in a Warming World Initiative will evaluate the water needs of different electricity technologies, primarily focusing on the Southwest and Southeast—the two regions where competition for limited (and potentially shrinking) water resources is predicted to be high. To ensure that this precious resource remains for generations to come, decisions need to be made soon on how we will continue to produce electricity for the millions of people living in the region in both a water-efficient and energy-efficient way.

Special thanks to Ardeth Barnhart for her advice and editorial assistance


*Averyt, K., J. Fisher, A. Huber-Lee, A. Lewis, J. Macknick, N. Madden, J. Rogers, and S. Tellinghuisen, 2011. Freshwater use by U.S. power plants: Electricity’s thirst for a precious resource. A report of the Energy and Water in a Warming World initiative. Cambridge, MA: Union of Concerned Scientists.

*Cooley, H., J. Fulton, P. H. Gleick, 2011. Water for Energy: Future Water Needs for Electricity in the Intermountain West. Oakland, CA: Pacific Institute.

*U.S. Department of Energy, 2006. Energy Demands on Water Resources: Report to Congress on the Interdependency of Energy and Water.