Spring 2012

Thirsty for Power | Catalyst Spring 2012

The U.S. electricity sector is addicted to water. Research coordinated by UCS explains the dimensions and implications of this addiction, and how utilities can kick the habit.


By John Rogers and Erika Spanger-Siegfried

Take the average amount of water flowing over Niagara Falls in a minute. Now triple it. That’s almost how much water U.S. power plants take in for cooling purposes every minute, on average. At the same time, water demand is increasing and heat waves and drought are compounding the strain placed on vital freshwater supplies—a problem that global warming is projected to worsen. Understanding this energy-water nexus is critical to making smart policy and planning decisions, but until recently even basic information on power plant water use was difficult to obtain.

To bridge this gap, UCS and a team of independent experts undertook the first-ever systematic assessment of both the effects of power plant cooling on U.S. water resources and the quality of information available to public- and private-sector decision makers. Our analysis provides a strong initial basis for making water-smart energy choices—the only choices suitable to a warmer, water-constrained future.

The Scope of the Problem

Our analysis used new analytical approaches to calculate the water use of virtually every power plant in the United States, looking across the range of fuels, technologies, and cooling systems. As it turns out, thermoelectric power plants—those that boil water to create steam, which in turn spins the turbines that produce electricity—have a massive thirst. On an average day in 2008 (the latest year for which data were available), these plants withdrew roughly 100 billion gallons of freshwater from rivers, lakes, streams, and aquifers, to condense the steam from the turbines. They consumed (i.e., evaporated) billions of gallons of that water rather than returning it to waterways. Our nation’s coal fleet alone, which generates less than half of our electricity, was responsible for two-thirds of both power plant withdrawals and consumption.

We also found tremendous variation in water use, not just by technology but also by location. For example, plants in Michigan, Missouri, North Carolina, and Virginia withdrew 40 to 55 times as much freshwater as plants in California, Nevada, and Utah in 2008 (see the map). U.S. nuclear plants withdrew nearly eight times more freshwater than natural gas plants on average. And some renewable energy facilities used essentially no water, but others consumed more than the average coal or nuclear plant. (All comparisons are per unit of electricity generated.)

Freshwater Use for Electricity Generation

Water use intensity can vary greatly from state to state based on electricity mix (including water-free sources such as wind) as well as power plant efficiency and cooling. States that make water-smart energy choices may set an example for others to follow.



Water Supplies Feel the Stress

Using these data, we then assessed the stress that power plants place on water systems across the country. According to our analysis, 400 out of 2,106 watersheds across the country in 2008 experienced water supply stress—the point at which demand for water by all users (including power plants, agriculture, and municipalities) exceeds a critical threshold of the available supply, typically surface and groundwater. Power plants, by tapping this overstretched resource for cooling purposes, contributed to water supply stress in nearly 80 of those watersheds.

The Carbon-Water Balancing Act

Decision makers must consider the impacts that global warming, electricity generation, and water all have on each other.

As scientific evidence shows, heat-trapping carbon emissions from human activities are driving up global average temperatures. Because warmer air can hold more water, precipitation patterns that affect water resources are changing, contributing to flooding in some areas and drought conditions in others (which has a direct impact on the operation of water-cooled power plants). Water temperatures are also rising in many streams, lakes, and rivers, causing water-cooled power plants to run less efficiently, which in turn increases the cost of electricity and the amount of water these plants need.

Compounding the problem is the fact that the power sector is a major source of carbon emissions, accounting for one-third of the country’s total in 2009. As fossil-fueled power plants are forced to run longer and harder to provide consumers with relief from rising temperatures, they generate even more emissions.

No Easy Answers
Power plant fuels and cooling technologies vary greatly in both water requirements and carbon emissions: traditional coal-fired plants are carbon- and water-intensive; nuclear is virtually carbon-free but water-intensive; wind power is both low-carbon and low-water. Solutions to one problem do not necessarily address the other; adding cooling towers to coal plants, for example, reduces water use but not carbon emissions.

And in some cases, solutions create problems elsewhere. Consider carbon capture and storage, a set of technologies that can prevent power plants’ carbon emissions from escaping into the atmosphere by injecting them underground instead. Though this strategy would curb heat-trapping emissions, it could increase water consumption between 35 and 95 percent.

Water quality can be similarly stressed by high temperatures. Even with the sparse temperature data available, we found more than 350 power plants across the country had discharged water that exceeded 90°F, the limit set by many states to control harm to fish and other wildlife. Warmer water stresses power plants, too, decreasing their efficiency (see the sidebar) and forcing operators to reduce output, or even temporarily shut down, to avoid the risks—to the environment or the plants themselves—posed by higher water temperatures.

Our analysis also revealed a number of gaps and inaccuracies in even the most comprehensive water-use database, with details reported by plant operators and compiled by the federal government, in 2008. About 25 to 30 percent of power plant water use went unreported, and discrepancies were widespread across the country. As a result, analyses based on federal data would have overlooked regions facing water stress.

A Water-Smart Energy Future

Getting better information is a critical step, but only a first step. Decision makers must then put that information to work in curbing electricity’s thirst, especially in water-stressed regions. We offer the following recommendations for making water-smart energy choices:

Get it right the first time. Particularly in regions of current and projected high water stress, utilities and other power plant developers planning new generating facilities could prioritize low-water options including dry cooling (which uses fans instead of water) and technologies that do not require cooling at all, such as wind and solar photovoltaics. Some developers are already deploying dry cooling in projects located in the desert.

Retool existing plants. Owners and operators of existing water-hungry power plants in water-stressed areas could consider retrofitting to low-water cooling. Some plants have already made the switch; for example, the 1,250-megawatt (MW) Plant Yates near Newnan, GA, added cooling towers in 2007, cutting water withdrawals by 93 percent (while also eliminating the large fish kills caused in part by hot water discharges from the plant). Xcel Energy slashed freshwater consumption at its 1,080 MW Harrington Station in Amarillo, TX, by switching to treated wastewater for its cooling needs in 2006.

Set strong guidelines for power plant water use. Public officials including legislators and public utility commissioners can play an active role in averting energy-water collisions. For example, the Colorado legislature’s 2010 decision to retire more than 900 MW of coal plants in favor of natural gas, energy efficiency, and renewable energy will annually save a volume of water roughly equivalent to that used by 50,000 people.

Renewable energy development should be approached with an eye toward minimizing water use. The Ivanpah solar thermal plant in California (left) uses dry cooling to save water, while wind turbines require no water at all.


Engage diverse stakeholders. Informed and empowered stakeholders—mayors securing water supplies for their cities, anglers concerned with sport and commercial fishing, water resource managers at all levels, and many others—can help ensure strong decisions are made and implemented.

Reduce power plant carbon emissions. Because human-caused climate change is worsening water stress across much of the United States, water-smart energy choices should also be low-carbon. However, not all water-saving technologies reduce heat-trapping carbon emissions, nor do all low-carbon options save water (see the sidebar).

Averting energy-water collisions means taking a long view. Power plants are designed to last for decades, and much of our existing infrastructure will continue functioning for years as well. Yet over the next several decades, our nation’s precious freshwater resources will face ever more stress from growing populations, a changing climate, and other trends. Making smart decisions today about which power plants to build, which to retire, and which technologies to develop and deploy will help ensure that both our electricity system and our water supplies remain viable for future generations.

John Rogers and Erika Spanger-Siegfried are co-managers of the UCS energy-water initiative and senior analysts in the Climate and Energy Program.


Learn more about the energy-water nexus in Freshwater Use by U.S. Power Plants: Electricity’s Thirst for a Precious Resource