Skip Navigation

Environment Magazine September/October 2008


January-February 2012

ResizeResize Text: Original Large XLarge Untitled Document Subscribe

U.S. Energy Security and Water: The Challenges We Face

Caption: Cooling tower and cooling water discharge at a nuclear power plant.

Caption: Cargo ship under attack in Tanker War during Iran-Iraq War.

Caption: Cargo ship under attack in Tanker War during Iran-Iraq War.

U.S. national security interests can cover a variety of issues, but one that has been enduring over the last 40 years has been energy security, which means that there are sufficient supplies of energy at prices that do not disrupt ordinary economic and social activity.1 The nature of the energy security challenge has morphed over those decades to include carbon dioxide emissions from fossil fuels and, more recently, the security of electricity supplies.

Now, as the nature of interdependencies between energy and water is becoming clearer, so too the connections between water and national security are also becoming clearer and adding new challenges. Leading options to respond to the need to increase energy security by increasing the domestic production of oil and oil substitutes and reducing greenhouse gas (GHG) emissions through carbon capture and sequestration (CCS) may be limited because of water quantity or quality constraints. And in the electric power sector, expansion of conventional supplies to meet demand may be constrained because in significant parts of the country, water supplies are expected to be insufficient.2

There are energy options though, that can reduce energy dependence, cut emissions, and conserve water. This article explores these issues.

Energy Security in the United States

Energy security can be susceptible to market manipulation, political instability in producing countries, damage to infrastructure from natural disasters, accidents or attacks, and competition for energy resources.3 Any of these can disrupt supplies, or create a fear of such disruption, leading to price surges.

The United States became vulnerable to oil shocks because consumption grew after World War II and domestic production failed to keep up. In the early 1950s, production matched demand. The pace of consumption accelerated to support growing standards of living, while domestic production rates slowed, then peaked in the late 1960s and early 1970s. This opened a gap that was filled by imports, which the country became dependent upon.4

Energy security gained broad public attention during the oil embargo of 1973–1974, with substantially higher gas prices and lines at filling stations. The Organization of Arab Petroleum Exporting Countries cut oil exports in response to the United States' support of Israel during the October Yom Kippur War,5 though this also served to increase its oil revenues. At that time, net oil imports by the United States were about 35 percent of consumption.6

The Carter Doctrine, put forward by Jimmy Carter in 1980, proclaimed that the United States would use “any means necessary, including military force,” to keep oil flowing through the Persian Gulf. This was in response to the Soviet Union's invasion of Afghanistan and the perceived danger to Iran, which the President identified as “a grave threat to the free movement of Middle East oil.”7 Since that time, reliance on imported oil peaked in 2005 at 59 percent. It has since dropped back to about 49 percent due to the recession, enhanced efficiency, gains in domestic production, and the substitution of domestically produced ethanol and biodiesel.8

There are various options to address energy security, including increasing the reliability of supplies, whether domestic or imported, increasing the availability of new domestic supplies, fuel substitution, and reducing demand through conservation and efficiency. The focus on energy security has most often been on external supplies of oil, although no matter how much the United States imported, any global disruption would cause economic pain.9 Fortunately, the nation has become much less dependent on energy to drive economic output. Between 1990 and 2009, overall energy needed to produce $1,000 of gross domestic product (GDP) went down by 30 percent.10

In addition to minimizing the risk of economic disruptions, another reason for the United States to reduce reliance on oil, particularly from the Middle East, is to minimize the need to maintain a military presence at maritime chokepoints for oceangoing oil transport, which carries two-thirds of the oil that is traded.11 The probability of disruption equal to or greater than the one that occurred when Iraq invaded Kuwait was estimated to be 25 percent between 2005 and 2014.12 Disruptions have occurred in the past, notably the “Tanker War” of 1983–1988, when Iran attacked 554 oil tankers and killed 400 crew members.13 If there were no concerns about securing oil supplies from the Persian Gulf, the U.S. military might avoid expenses equal to 12–15 percent of the 2008 defense budget,14 which puts the cost between $55 and $69 billion.15

Energy security has become more complicated since the 1970s, in part because of the security concerns tied to climate change caused primarily by the consumption of fossil fuels, including oil. The probable unintended consequences of the United States' fossilfuel- based economy include a litany of effects including higher average global temperatures, changes in precipitation amounts and patterns, sea-level rise, additional risks to human health, major impacts on agricultural production, more powerful hurricanes, accelerated extinction of species,16 and even the prospect for more conflict in weak nations (see Water and Conflict, p.7).17

The national security considerations of these effects include significant deleterious impacts to the U.S. economy, damage to coastal infrastructure including Navy, Marine, and Coast Guard bases, heightened stress on water resources, higher global food prices, increased demands on the military and aid agencies for humanitarian responses, and greater instability in high-risk regions of the world.

Options to Replace Oil and Their Implications

To make things still more difficult, a variety of options to respond to the need to increase energy security by increasing the domestic production of oil and oil substitutes and reducing GHG emissions may be limited because of water constraints. For example, for the foreseeable future, ethanol in the United States will be derived from corn, which, depending upon how it is grown and processed, can use 130 to 6,200 gallons of water per 100 miles driven, while a gasoline-powered vehicle only requires 7 to 14 gallons18—a multiple of 19 to 443.

Water and Conflict

The word rival is derived from the Latin rivalis, meaning “one using the same stream as another.”61 Numerous articles speak to the issues of future water scarcity and offer the possibility of rivalry or conflict. In “Water Wars,” the author reported that in the mid- 1980s, U.S. intelligence identified 10 places on the globe where war could break out over water scarcity and that “sewage could eventually become a catalyst for armed conflict in the Middle East” if allowed to continue to flow untreated and uncontrolled.62

However, a study of the history of international water conflicts shows little tendency for nations that share transboundary rivers to fight wars over water access. In fact, it seems that water is more often a cause for cooperation, perhaps because there is so much at stake.

During the last 50 years of the 20th century 1,831 events have been identified—military actions, hostility, support, or agreements—where issues such as water quantity, infrastructure, control, or quality were central. Two-thirds were cooperative, about a quarter were conflictive, and the rest were neutral. Of the 37 cases of acute conflicts, 30 were between Israel and one of its neighbors, with the final one in 1970.63

Since World War II, internal wars have been far more common and have had a much higher human cost than interstate conflicts64; following this pattern, internal civil unrest appears much more likely than water wars. Drought events big enough to cause a 5 percent drop in gross domestic product (GDP) in a poor country can increase the likelihood of a civil war. A study of 41 African countries from 1981 to 1999 found that there was a 50 percent increase in the probability of civil conflict in the year after an economic shock caused by a drought.65 Between 1980 and 2005, there were 21 conflicts in Africa, 16 of which involved water. In five, water was the sole contributor.66

The El Niño/Southern Oscillation (ENSO) has been connected to internal conflict. Between 1950 and 2004 the likelihood of new civil conflict in the tropics doubled from 3 to 6 percent in El Niño years, when the climate is warmer and dryer, versus La Niña years. The El Niño conflicts account for 21 percent of all civil conflicts in the period.67 The four factors tied to water that can lead to civil conflicts include water degradation, food insecurity caused by drought, storms and floods, and environmental migration caused by the first three.68

Not only is conflict in weak and fragile nations tied to water, but so are large-scale natural disasters, which may be increasing. Research on the frequency of such weather-related natural disasters as hurricanes, floods, drought, and windstorms shows that they have quadrupled since the mid 1970s, “surging from 428 in 1974–78, to 817 in 1984–88, to 1,707 in 1999–2003.” The number of disasters that require international assistance because they have overwhelmed local capacity has roughly tripled between 1950–1959 and 1996–2005, from 21 in the first period to 57 in the second.69 It's possible that the number and magnitude of these emergencies may increase if the earth's climate becomes more variable over time, with more extreme events that lead to food shortages, economic distress, more displaced persons, and perhaps even more conflict.

Caption: A typical ethanol plant in West Burlington, Iowa.

Caption: A typical ethanol plant in West Burlington, Iowa

Caption: Figure 1. Corn for ethanol—percent of total production.

Caption: Figure 1. Corn for ethanol—percent of total production.
Caption: The extensive and growing use of corn as an energy feedstock means that its price is becoming tied to the price of oil.

Caption: The extensive and growing use of corn as an energy feedstock means that its price is becoming tied to the price of oil.

Caption: Figure 2. Violent protests are linked to high food prices. The death toll is shown in parentheses. The red lines indicate the beginning date of unrest.

Caption: Figure 2. Violent protests are linked to high food prices. The death toll is shown in parentheses. The red lines indicate the beginning date of unrest.

The higher numbers in this range are due to irrigated corn production. From 2001 to 2006, irrigated corn-planted acreage held fairly steady between 8.6 to 8.9 million acres. Beginning in 2007, however, irrigated corn acres varied between 9.1 and 10.1 million, an increase between 6 and 18 percent from the prior average.19,20 To put this in some context, this means that in 2007, when 20 percent of corn production went for ethanol, water consumption for irrigated corn for ethanol amounted to approximately 3 percent of total water consumption in the United States. In 2010, 40 percent of the crop went to ethanol, bumping water consumption to roughly 6 percent of the U.S. total.21

Increases in corn production have been induced by new federal mandates. Congress called for less dependence on foreign oil in the Energy Independence Act of 2007. This act establishes a target, called the renewable fuel standard (RFS), which requires a minimum amount of renewable fuels in the transportation fuel mix. The new standard mandated 9.0 billion gallons per year in 2008, increasing to 22 billion gallons in 2022. Corn-based ethanol cannot make up more than 15 billion gallons per year, although there are no cost-effective alternatives on the horizon with the exception of ethanol from sugar. Although there is considerable discussion of second- generation fuels derived from such cellulosic materials as grasses, trees, and wastes, these production methods are more expensive than corn-based ethanol and require breakthroughs to be viable.22

An increasing amount of corn production in the United States is going to make ethanol. In 2001, 707 million bushels of grain went for ethanol. In 2010, 5,020 million bushels out of 12,447 bushels produced, or 40 percent, went to produce ethanol, the highest share yet (see Figure 1). In that year, ethanol contributed 12.8 billion gallons, or about 9 percent of the 138 billion gallons of finished gasoline use.23

In addition, corn production for ethanol is taking up a growing share of the incremental corn production in the United States. Average national corn production for 2001 to 2003 was about 9.5 million bushels. The percent of the incremental production in 2004 that went to ethanol was 58 percent, while in 2010 it was 171 percent—corn production went up, but corn use for ethanol went up much faster.24

World grain prices have increased dramatically in the last five years, in part due to diversion of corn to ethanol with implications for instability in some countries. The Food and Agriculture Organization (FAO) World Grain Price Index was relatively stable during the period from 1990 to 2007, generally between 90 and 130. After that though, prices rose dramatically, peaking at 265 in April 2011.25 About two-thirds of the increase in prices was due to increased food demand and production issues in some places, while about one-third has been attributed to increased production for ethanol.26 The extensive and growing use of corn as an energy feedstock means that its price is becoming tied to the price of oil.27

There are international security considerations tied in here. The recent run up of food prices has been identified in one study as the trigger for unrest in North Africa and the Middle East in 201128 popularly known as the “Arab Spring.” Figure 2 shows the overlap between FAO's composite Food Price Index and the start date of unrest in various countries in the region. The probability is low that these violent protests would have occurred by chance at the same time that food prices were high.29 There are certainly other underlying factors that create conditions amenable to instability. And, while these widespread episodes have had near-term international security implications, in the long-term there may be benefits. The protests movements are living history and remain to be played out.

In addition to using ethanol as an oil substitute, new technologies may enhance domestic production of oil from shale in fields that were very recently considered unrecoverable. Hydraulic fracturing, or fracking, has opened up a significant new source in Texas, the Eagle Ford. A report by the National Petroleum Council says that this area could produce 2–3 million barrels per day by 2035,30 although another study suggests that level could be reached by 2020. This would be 1.5 to 2 percent of current finished gasoline production from this single area,31 and there are other formations that are or could become recoverable with the technology including North Dakota, Pennsylvania, Wyoming, and Montana.32 Oil companies have invested $25 billion in the Eagle Ford during 2011.33

There is a catch, though—the process requires large volumes of water in a semi-arid area where water is already scarce and groundwater tables are declining. Local water officials say “there's definitely going to be a problem,” and oil company staff say that water is the process's “Achilles' heel.”34 In addition, there is growing interest in regulating fracking to address concerns that it contaminates water.

The water use for any single well would not likely be a problem; it's the accumulative effect that is a concern. A recent study in Texas estimated that water demand for fracking in the Eagle Ford formation will increase 10-fold by 2020, and double again by 2030. There would also be additional water demand from the workers who will come into the area. A single company that currently has 10 wells in the area is planning to drill another 200 in 201235; that's enough water for about 50,000 people for one year.36 There will likely be thousands of wells drilled in the area in the coming years.

Plug-in electric cars may also prove challenging because of their additional water demands. The argument for electric cars is that they can be charged at night when power plants are operating well below capacity. Though there may be unused capacity, the plants would still require cooling water to operate, perhaps significantly more. Cooling for thermoelectric power already withdraws more water than any other water demand; in some areas of the country, more than 75 percent of the surface water is used to cool power plants37. And if carbon capture and sequestration (CCS) technology is used to eliminate the greenhouse gases, water for cooling could double because of there is an energy penalty associated with capturing and compressing the carbon dioxide.38

Electricity and Water

There is an additional concern for the security of energy supplies, in this case electricity. While the United States has plenty of coal and growing reserves of natural gas due to fracking, water competition may constrain increased production. In the United States, freshwater withdrawals for irrigated agriculture and thermoelectric energy production show an almost equal percentage of the 345 billion gallons per day (gpd) of water withdrawn (Figure 3). Most of the water withdrawn for thermoelectric production is returned and can be re-used, though its quality may be lower due to thermal or chemical pollution. Seventy percent of the water used for thermoelectric cooling is fresh water from rivers and lakes; the rest is saline. Ninetynine percent of it is surface water.39

Where water for irrigation makes up 40 percent of total withdrawals, it comprises more than 80 percent of the 100 billion gallons of water consumed daily in the U.S.; these losses occur due to evapotranspiration.40 About 40 percent of water withdrawn for irrigation is returned for re-use, whereas almost 98 percent of withdrawals for thermoelectric cooling are returned.

While most conversations about the security of energy supply focus on oil, the security of the electricity supply is also in question. According to analysis by the Electric Power Research Institute (EPRI), it is likely that expansion of conventional thermoelectric power production will be constrained in 14 percent of U.S. counties by 2025 because water supplies will be insufficient. Nearly every state is expected to be affected.42

Nuclear power generators use about 2.5 times the water per unit of electricity than gas and 25 percent more than coal. For this reason, nuclear power plants are more susceptible to drought than other types of thermoelectric power plants. During an extended drought in 2007 in the southeastern United States, 24 of the nation's 104 nuclear power generators faced shutdowns because of water limitations, including lack of supply but also because low flows and higher ambient water temperatures limited how much heat could be expelled.43

During an extreme drought in Texas in 2011 (Figure 4), the Electric Reliability Council of Texas (ERCOT) warned the Public Utility Commission that “ERCOT anticipates higher outage rates associated with cooling water issues.”44 ERCOT had previously cut supplies to industrial users and warned of rolling blackouts due to record demand and complications from the extreme drought.45

The impact of climate change on water availability could lead to even greater constraints in some areas. Roy et al., analyzed the prospects for water sustainability out to 2050, including increases in demand from projected population growth and power demands, as well as possible changes in precipitation and evaporation as given by the median values of global circulation models. Their work shows that water use in more than half of the counties in the United States could be unsustainable in the future due to the combination of growth in demand and climate change. The areas most impacted include the sothwest, the southern plains, and Florida.46

Hydroelectric power production may also be constrained, particularly in the Southwest. The trends for water levels needed to produce electricity from Hoover and Glen Canyon dams look particularly troubling. Electricity from the dam powers large parts of southern California, Nevada, and Arizona. Water levels must be maintained above minimum power pool thresholds to turn the turbines, but there is an even chance that by 2017 water levels will be too low to enable the dams to generate electricity. There is also a 50 percent chance that by 2021 the live storage will be gone, meaning the dams will be unable to release any water at all. These calculations assume modest climate change, natural variability, and no change from current allocations or management regime.47 These conclusions are supported by trends in the flow of the Colorado River. Flow data for most of the 20th century show an asymptotic decline in water reaching the river's delta, with numbers centered around 25,000 million m3 in the early 1900s to zero for most years since the early 1960s.48

Both in terms of absolute degrees and in terms of annual standard deviation, the Colorado River basin has warmed more than any region of the United States … Key manifestations of warmer temperatures in western North America are a shift in the peak seasonal runoff (driven by snowmelt) to earlier in the year, increased evaporation, and correspondingly less runoff.49

Two things are likely going on here. First, persistent droughts have occurred in the past, and the period when much of the river was developed and allocated was “exceptionally wet”—it may be that the river has reverted to its norm. Second, climate change is likely a significant factor.

Caption: Figure 3. Fresh water withdrawals and consumption in the United States.41

Caption: Figure 3. Fresh water withdrawals and consumption in the United States.41

Caption: Figure 4: The 2011 drought in Texas impacted electricity production due to water constraints.

Caption: Figure 4: The 2011 drought in Texas impacted electricity production due to water constraints.

Power Production Comparisons

Caption: Photograph of the Hoover Dam from across the Colorado River; from the series Ansel Adams Photographs of National Parks and Monuments, compiled 1941–1942.

Caption: Photograph of the Hoover Dam from across the Colorado River; from the series Ansel Adams Photographs of National Parks and Monuments, compiled 1941–1942.

Caption: Lake Mead, upriver of the Hoover Dam, in July 2009, showing a depleted water level.

Caption: Lake Mead, upriver of the Hoover Dam, in July 2009, showing a depleted water level.

Projections of stationary power supply for the United States made by the Energy Information Agency (EIA) suggest that electricity generation will grow by about 30 percent between 2009 and 2035.50 An increase of this sort raises the possibility of a large increase in water withdrawals for thermoelectric cooling, which already makes up the largest share of water withdrawals in the United States. This is far from a fait accompli, however, as the withdrawal and consumption of water for thermoelectric cooling are highly dependent on the fuel, combustion technology, and cooling method. These in turn are predicated on relative costs and regulatory choices.

Coal produces about 44 percent of the United States' electric power, followed by natural gas at 22 percent and nuclear power at 19 percent. Conventional hydropower comprises about 9 percent, while wind is 3 percent. Petroleum is less than 1 percent, and biomass, solar thermal, and solar photovoltaic (PV) together are about 3 percent.51

Power production technologies vary widely in cost, water use, and environmental characteristics. Table 1 shows various energy technologies in use today. The first column shows levelized costs, which are the amortized capital and operation and maintenance costs. Because the factors that determine costs are varied, a range is provided. The levelized costs are taken from or derived from the Energy Information Administration, which uses these costs for energy and policy modeling.52 Subsidies are not included.

Table 1. Cost and environmental factors for various electricity generating technologies

Table 1. Cost and environmental factors for various electricity generating technologies

Not surprisingly, the three most dominant fuels in the United States—coal, natural gas, and nuclear—are the cheapest, at least for existing generation at the lower end of the ranges. The lower costs represent older facilities that have been fully amortized so there are no capital costs, just operation, maintenance, and fuel.

The fleet of existing coal plants was built between the 1950s and 1990, with a bulge between the late 1960s and the mid 1980s. The oldest ones are likely to have the lowest costs. Nuclear plants also tend to be older, with the majority between 21 and 40 years old. In contrast, gas plants are relatively new, with the largest age category less than 10 years old. With the exception of a very few coal and petroleum plants, almost all the capacity built since the latter 1990s was natural gas and since the mid 2000s, wind.53

The reason why becomes apparent under the new generation costs. Natural gas has the lowest range and average levelized cost by quite a bit, even comparable with the average cost for existing coal. Hydroelectric also has an inexpensive low end and average, but most of the potential sites for hydro are already in use. The next cheapest options are new conventional coal and onshore wind. Wind currently has the fastest growth rate in capacity of any of the options, 36 percent per year.54 Carbon capture and sequestration (CCS) options add significantly to the cost of gas and coal.

The cheapest of all is energy efficiency, which is shown as a point of comparison though it does not supply energy except in the “negawatts” sense.55

Also shown are the average costs with carbon taxes of $25 and $50 per ton of carbon dioxide. There is no impact on the costs for nuclear or renewable technologies. Natural gas costs increase modestly, while there is a significant additional cost on the coal options. New coal becomes uncompetitive with gas and wind, even at the lower of the carbon tax rates.

The table also shows the range of estimates for water withdrawal and consumption for each power production technology. The ranges are so large because the water use characteristics of the four main cooling options, oncethrough, pond, recirculating, and tower, vary quite a bit56 (see Water Requirements for Thermoelectric Generation, p.16). The low end of the ranges for withdrawal typically show values for cooling towers, and the larger values are for once-through cooling. The highend extremes for water withdrawals are lower for new capacity, because oncethrough cooling is not used in new construction due to its water intensity. The new high-end values are for recirculating cooling.

The high-end values for withdrawals show a significant difference between the fuel types and their use of oncethrough cooling. Natural gas is the most efficient, so more of the energy goes into power and less into waste heat that requires cooling. Nuclear is the thirstiest of all the options because it uses lower temperature steam, which requires more steam and cooling water.57 These numbers show why nuclear is the most vulnerable to drought and lowflow conditions.

Coal options use a bit less than nuclear, with integrated coal gasification combined cycle (IGCC) using less than pulverized coal because it is a bit more efficient. Natural gas uses considerably less water than coal because it is much more efficient. Where the efficiency of a coal plant may be 35–40 percent, natural gas may be 65–70 percent. As there is less waste heat, there is a lower demand for cooling water. In some parts of the country, particularly in the Midwest, more than 75 percent of river flows are used for thermal cooling,58 largely for power production from coal.

The lower consumption values for cooling water are generally for cooling towers, though they also have higher values for consumption. When CCS technology is added to conventional fossil fuel generators, there is a considerable energy penalty encountered as carbon dioxide must be stripped from the flue gas, compressed, and disposed. For coal, this increases water needed for cooling by 90 percent, and for gas by 80 percent.59 These numbers assume that only the most efficient cooling options would be used with CCS.

Caption: A bank of photovoltaic cells.

Caption: A bank of photovoltaic cells.

Caption: The Brazos Wind Farm, also known as the Green Mountain Energy Wind Farm, near Fluvanna, Texas.

Caption: The Brazos Wind Farm, also known as the Green Mountain Energy Wind Farm, near Fluvanna, Texas.

Water Requirements for Thermoelectric Generation70

Thermoelectric power production is the conversion of thermal energy into electrical energy. Thermoelectric generation relies on a fuel source (fossil, nuclear, biomass, geothermal, or sun) to heat the fluid to drive a turbine. If natural gas is the fuel, a combined cycle is used and it is called a natural gas combined cycle (NGCC); this is the most efficient way to generate electricity. If coal gasification is used to produce synthetic gas, it is called an integrated gasification combined cycle (IGCC). This is the second most efficient generation method. If coal is burned directly, a boiler is needed to transfer heat from the combustion of the coal to water. There are many boiler designs, but for electricity generation, the coal is pulverized into a fine powder and fed as slurry into the boiler; this is called pulverized coal (PC) combustion.

The higher the temperature and pressure of the steam, the more efficient is the conversion process. For nuclear reactors, the nuclear fuel is the heat source and the steam remains below subcritical pressures and is less efficient than PC plants. Other sources of heat are geothermal and concentrating solar power plants.

Thermoelectric power generation requires large volumes of water, mostly used to cool and condense the steam after it exits the turbine. There are three general types of cooling system designs used for thermoelectric power plants: once-through, wet recirculating, and dry. In once-through systems, the cooling water is withdrawn from a local body of water such as a lake, river, or ocean, and the warmer cooling water is subsequently discharged back to the same water body after passing through the surface condenser. As a result, plants equipped with once-through cooling water systems have relatively high water withdrawal, but low water consumption.

The most common type of recirculating system uses wet cooling towers to dissipate the heat from the cooling water to the atmosphere. In the process, a portion of the warm water evaporates from the cooling tower and forms a water-vapor plume. These systems have low withdrawals but higher consumption than once-through cooling.

Though hydroelectric and solar thermal can have relatively high water consumption numbers, other renewable options use little or no water, and even solar thermal can use a dry cooling option, as could other thermal options. Wind uses no water and PV only consumes water for washing. And of course, efficiency also avoids water use.

The last comparison point for the power options looks at greenhouse gas emissions and air quality indicators. Coal has the highest emissions across the board, with the exception of NOx, where the natural gas combined cycle (NGCC) is somewhat higher. However, natural gas emits no sulfur dioxide (SO2), mercury (Hg), or particulate matter (PM), and its GHG emissions are only 43 percent of that of coal without CCS. CCS can cut CO2 emissions from coal and gas by about 90 percent. Nuclear energy, renewables energy, and energy efficiency have neither associated greenhouse gas emissions nor other air pollutants.

The withdrawal numbers given in Table 1 are for water use on-site for generator operation. There is also water use “upstream” during mining or extraction, fuel preparation, transportation, and construction or manufacturing processes. The water withdrawals upstream can be comparable to the lower end numbers in the table. Upstream water withdrawals for coal can vary from 36 gallons per MWh (g/MWh) to 975. The largest single use factor for all fuel cycles is transport by slurry pipeline, which withdraws about 815 g/MWh.60

The two other thermoelectric fuels fall into a similar range. For natural gas extracted onshore, upstream water withdrawals are 323 g/MWh, which includes water used for fracking (see Water and Fracking, p.17). Water used for environmental controls is the largest use factor. Nuclear power's upstream water withdrawals range from 42 to 339 g/MWh depending upon the enrichment process; centrifuge enrichment uses less than 10 percent of that needed for diffusion.

Wind and photovoltaic use water mainly upstream in the manufacturing process. For wind, this amounts to 66 g/MWh, mostly for steel, iron and glass fibers. The spread for PV is 151 to 403, with thin-film at the low end.

Where Are We Headed?

One commentator I recently spoke with called the energy–water–climate linkages “the big gnarly”—touching on everything and hard to get a handle on. If approached thoughtfully though, recognition and consideration of these very real connections can improve security, help mitigate climate change, and save water, avoiding major disruptions and providing water supplies for future economic development.

The United States has operated as if these linkages do not exist. For example, though thermoelectric power is regulated by the U.S. Environmental Protection Agency (EPA), and the rules it promulgates inevitably impact water use, the agency does not measure or manage those impacts because it has no authority over water quantity. And if it tried to do that, the states would almost certainly voice strong objections.

Even in terms of policy assessments, there is no energy model at the U.S. EPA or Department of Energy (DOE) that includes water supply feedbacks to limit thirsty energy choices if water use was constrained. For this reason, it is possible that energy policy evaluations for the future could be wrong.

Other energy policy decisions are similarly made without considering the broader impacts on water or security, some positive, some not. Corn-based ethanol has a negative impact on water quality and use, contributes little to greenhouse gas reduction, and is costly, while efficiency standards for cars, buildings, and appliances reduce energy consumption, reduce water use, and can pay for themselves.

Some recent developments, such as emphasis on natural gas and wind for new electric power capacity, are very good news. Fracking to open up new sources of domestic gas and oil could go either way if not managed. If the potential for water quality problems is tightly controlled, and supply issues are managed through recycled fresh water or saline water, then this could enhance energy security while minimizing water impacts. Greater use of natural gas from fracking would also reduce greenhouse gas emissions.

Energy and water policy is disjointed, with many federal, state, and local decision makers but few mechanisms to coordinate action. Yet there are technologies and policy approaches that could be adopted that would improve the country's position with regard to energy, water, and climate security, if only means of coordination were in place. There is also an opening for productive policy conversations, particularly in the West, where energy and water problems are generally acknowledged and under discussion.

Synergies exist, but the question is whether or not policymakers will recognize and act on them. If so, then the country could be headed for a manageable future. If not, then the United States will remain on course to muddle through or worse, with policies continuing at cross purposes and irregular crises that could have been mitigated.

In the current political environment in Washington, the prospect of any agreement on energy, water or climate is slim to none. However, for the states, energy and water issues are here and now and can't be put aside. Texas, for example, is facing its worst recorded single-year drought with enormous problems across the state. The impacts would be much worse if not for the fact that Texas has the largest installed capacity of wind power in the U.S. And so, though there's little sympathy for climate change policy there, they are not likely to back off wind power anytime soon. Water use for fracking is also coming under more scrutiny there and may constrain oil and gas development. It's likely that the Western states, hardest pressed on water issues, will be where the policy action is for some time to come.

Water and Fracking

Hydraulic fracturing, or “fracking,” is a technology used to extract gas and oil from rock formations that have low permeability, making the resource inaccessible using normal approaches. Fractures and fissures are created in the rock formation using horizontal drilling and high levels of hydraulic pressure and are propped open by “proppants” such as sand or ceramic beads, allowing the gas or oil to move to the production well for extraction.71

Fracking has expanded rapidly and opened up major new domestic resources, of which the biggest is the Marcellus shale basin, which extends from northern Georgia into New York. There are various other formations as well, though, in Texas, Oklahoma, Indiana, Michigan, Arkansas, Louisiana, and the Rockies.72 The Barnett Shale formation in Texas is the most developed in the United States. The number of producing wells there rose from 400 in 2004, to more than 10,000 in 2010. The EIA has a dramatic graphic that shows this development.73 According to the EIA, shale gas will make up 46 percent of the total supply in 2035.74

There are concerns about water use and fracking, which include water volumes as well as contamination. Shale oil production requires about 3–4 million gallons of water for an average well, which could produce 800 million cubic feet of gas over a 7.5-year lifetime.75 The fluid that is used in fracking has numerous chemicals in it, and the various formulations are secret. A Congressional Committee asked 14 oil and gas companies to disclose the types of products used in their fluids. The companies reported 750 different products, from harmless one such as salt, citric acid, coffee, and walnut hulls, to toxic ones such as lead, benzene, and methanol. Twenty-nine were known or possible carcinogens.76 Fracking is specifically excluded from regulation under the Safe Drinking Water Act.77

1. D. A. Deese, “Energy: Economics, Politics, and Security,” International Security, 4, no. 3 (1979–1980), 140–153.

2. R. Goldstein, EPRI's Water/Energy Sustainability Initiative (2004),

3. M. Wesley, Power plays: Energy and Australia's security, Australian Strategic Policy Institute, 2007.

4. Energy Information Administration (EIA), “How Dependent Are We on Foreign Oil?” (2011),

5. D. A. Rustow, “Who Won the Yom Kippur and Oil Wars?,” Foreign Policy, no. 17 (1974–1975), 166–175.

6. EIA, see note 4.

7. J. E. Carter Jr. “The Defense of the Gulf.” January 23rd, 1980.

8. EIA, see note 4.

9. RAND, “Does Imported Oil Threaten U.S. National Security?” (2009),

10. Millennium Development Goals Indicators, Energy Use (kg oil equivalent) per $1,000 GDP (Constant 2005 PPP $),

11. U.S. Energy Information Administration, “World Oil Transit Chokepoints” (January 2008),

12. P. C. Beccue, “An Assessment of Oil Market Disruption Risks,” Stanford Energy Modeling Forum, 3 October (2005),

13. W. Komiss and L. Huntzinger, The Economic Implications of Disruptions to Maritime Oil Chokepoints, CNA Corp. CRM D00234669.A1/Final (March 2011).

14. RAND, see note 9.

15. The 2008 budget for the Department of Defense was $460 billion.

16. U.S. Global Change Research Program Global Climate Change Impacts in the United States (2009),

17. P. Faeth, Water Vulnerabilities, Climate Change and Conflict in Poor Countries (CNA Corp., 2011).

18. M.E.Webber, “Catch 22: Water vs. Energy.” Scientific American. Volume 18, Number 4, 2008

19. Derived by CNA from data compiled by the National Agricultural Statistics Service, U.S. Department of Agriculture,

20. It is impossible to tell how total irrigated acreage changed in the United States over the period, as NASS does not report irrigated acreage for wheat in 2009 and 2010.

21. Irrigation uses 80.6 percent of all water consumption; in 2007, the last year for which data are available, there were 10.1 million acres of irrigated corn out of a total of 56.6 million acres of all irrigated land; 20 percent of corn production went to ethanol in 2007. (0.806*(10.1/56.6)*0.20)*100 = 2.8 percent.

22. B. D. Yaccobucci and R. Schnepf. CRS Report for Congress: Selected Issues Related to an Expansion of the Renewable Fuel Standard (RFS). Congressional Research Service, Order Code RL34265. 2007.

23. American Fuels, 2010 Gasoline Consumption,

24. CNA analysis, derived from USDA production statistics accessed at

25. FAO Food Price Index,

26. B.A. Babcock and J. F. Fabiosa. The Impact of Ethanol and Ethanol Subsidies on Corn Prices: Revisiting History. Center for Agricultural and Rural Development, Iowa State University. CARD Policy Brief 11-PB 5 April 2011.

27. L. Brown, “The New Geopolitics of Food,” Foreign Policy, May/June (2011).

28. M. Lagi, Marco, K. Z. Bertrand, and Y. Bar-Yam, The Food Crises and Political Instability in North Africa and the Middle East (New England Complex Systems Institute August 10, 2011).

29. Lagi et al., see note 28.

30., Report: U.S. Has “Surprisingly” Large Amount of Oil, September 15 (2011),

31. Derived from American Fuels, 2010 Gasoline Consumption,

32., “South Texas Enjoys Major Boom From Oil Fracking,” June 12 (2011).

33. New York Times. Shale Boom in Texas Could Increase U.S. Oil Output. May 27 (2011).

34. New York Times, see note 33..

35. Bloomberg, “Worst Drought in More Than a Century Strikes Texas Oil Boom,” June 13 (2011),

36. Based on Bloomberg, see note 35.

37. E. Jenicek et al., Army Installations Water Sustainability Assessment: An Evaluation of Vulnerability to Water Supply, U.S. Army Corps of Engineers, Engineer Research and Development Center, ERDC/CERL TR-09-38, September (2009), map D3.

38. NETL, Bituminous Performance Tool, (accessed September 21, 2011).

39. USGS, Total Water Use in the United States, 2005, (accessed September 15, 2011).

40. Water use is reported in two ways: withdrawals and consumption. Not all sources differentiate between the two, but the difference is crucial. Water withdrawals measure how much water is removed from the source for a particular use. Some may be returned for reuse, while some may evaporate or otherwise become unavailable for reuse. Water that becomes unavailable is said to be consumed. The water consumed plus the water returned equals the total water withdrawal.

41. U.S. Department of Energy, Energy Demands on Water Resources: Report to Congress on the Interdependency of Energy and Water, December (2006).

42. B. Goldstein and M. Hightower. Energy and Water: Partnerships for Energy-Water Research. Powerpoint presentation accessed at N.d.

43. MSNBC, January 23 (2008),

44. Letter from Trip Doggett to Donna L. Nelson, Chairman, Public Utility Commission of Texas, August 18, 2011,

45. Dallas Morning News, ERCOT Warns of ‘High Probability’ of Rolling Blackouts as Heat Wave Strains Power Grid, August 4 (2011),

46. S. B. Roy, L. Chen, E. Girvetz, E. P. Maurer, W. B. Mills, and T. M. Grieb, Evaluating Sustainability of Projected Water Demands Under Future Climate Change Scenarios (Tetra Tech, Inc., July 2010),

47. T. P. Barnett and D.W. Pierce, “When will Lake Mead go dry?” Water Resources Research, Vol. 44, 2008.

48. P.H. Gleick, “Global Freshwater Resources: Soft-Path Solutions for the 21st Century.” Science Volume 302: November 28, 2003.

49. National Research Council of the National Academies.. Colorado River Basin Water Management: Evaluating and Adjusting to Hydroclimatic Variability. Committee on the Scientific Bases of Colorado River Basin Water Management, Water Science and Technology Board, Division on Earth and Life Studies. National Academis Press, Washington, D.C. 2007.

50. Annual Energy Outlook with Projections to 2035, DOE/EIA-0383(2011), April (2011), Table A8,

51. EIA, Electric Power Monthly, September 15 (2011),

52. EIA, Levelized Cost of New Generation Resources in the Annual Energy Outlook 2011 (n.d.),

53. EIA, How Old Are U.S. Power Plants?, August 8 (2011),

54. Ibid.

55. Consumer Reports, Buzzword:Negawatts,

56. J. Macknick, R. Newmark, G. Heath, and K. C. Hallett, A Review of Operational Water Consumption and Withdrawal Factors for Electricity Generating Technologies, National Renewable Energy Laboratory, NREL/TP-6A20-50900, March (2011).

57. DOE National Energy Technology Lab, Water Requirements for Existing and Emerging Thermoelectric Plant Technologies, Revised 2009, DOE/NETL-402/080108,

58. Jenicek et al., 2009.. See note 37.

59. NETL, Bituminous Performance Tool.

60. The numbers in this section were derived by CNA from V. Fthenakis and H. C. Kim, “Life-Cycle Uses of Water in U.S. Electricity Generation,” Renewable and Sustainable Energy Reviews 14 (2010): 2039–2048.

61. Merriam-Webster, 1981. Webster's New Collegiate Dictionary.

62. Starr, Joyce R. “Water Wars.” Foreign Policy 82 (Spring 1991): 17–36.

63. A. T. Wolf, S. B. Yoffe and M. Giordano. “International waters: identifying basins at risk.” Water Policy 5 (2003) 29–60.

64. D. Michel, “A River Runs Through It: Climate Change, Security Challenges, and Shared Water Resources.” In, Troubled Waters: Climate Change, Hydropolitics, and Transboundary Resources. David Michel and Amit Pandya, Editors. The Henry L. Stimson Center. 2009.

65. E. Miguel, S. Satyanath and E. Sergenti. “Economic Shocks and Civil Conflict: An Instrumental Approach.” Journal of Political Economy, 2004, vol. 112, no. 4, pp 725–753.

66. R. Schubert et al., Climate Change as a Security Risk (German Council on Global Change, 2007),

67. S.M. Hsiang, K. C. Meng and M. A. Cane. “Civil conflicts are associated with the global climate.” Nature Vol. 476, 2011.

68. Schubert, et al., 2007. See note 66.

69. Michel, 2009. See note 64.

70. Adapted from B. Carney, T. Feeley, and A. McNemar. Water Requirements for Thermoelectric Generation. (U.S. Department of Energy/National Energy Technology Laboratory, n.d.),

71. Earthworks, Hydraulic Fracturing 101,

72. U.S. Department of Energy, Office of Fossil Energy and the National Energy Technology Laboratory, “Modern Shale Gas Development in the United States: A Primer,” April (2009),

73. EIA, Technology Drives Natural Gas Production Growth From Shale Gas Formations (2011),

74. EIA, What Is Shale Gas and Why Is It Important? (2011),

75. EIA, see note 72.

76. U.S. House of Representatives Committee on Energy and Commerce, Minority Staff, Chemicals Used in Hydraulic Fracturing, April (2011),

77. U.S. EPA, Regulation of Hydraulic Fracturing by the Office of Water (n.d.),

Paul Faeth is a Senior Fellow at CNA, a not-for-profit research and analysis organization. He was previously President at Global Water Challenge and Executive Vice-President at the World Resources Institute. 

In this Issue

On this Topic

Taylor & Francis

© 2018 Taylor & Francis Group · 530 Walnut Street, Suite 850, Philadelphia, PA · 19106