In America, events can move from the impossible to the inevitable without ever stopping at the probable.
—Alexis de Tocqueville
In the spring of 2008, a hot wind blew out of the Sonoran Desert, blasting sand through the air and rolling tumbleweeds across the streets of Phoenix, Arizona. The temperature was nearing one hundred degrees, and the air was bone-dry. The few people caught outside shielded their eyes, hunched their shoulders, and ran through the flying grit from one air-conditioned sanctuary to another. On the third floor of a large beige office building downtown, Herb Guenther, the director of the Arizona Department of Water Resources, and one of the last of the classic Water Buffaloes of the Floyd Dominy school, was committing near-blasphemy. “Climate change is for real,” he said. “We can’t keep doing things the same old way.”
I was there to find out how a man of his experience was coping with increased water demands in a state that was already the hottest in the nation, and how it would adapt to even hotter, drier, more crowded conditions in the future.
For an old-school water manager such as Herb Guenther to acknowledge “climate change” and the need for “sustainability” was a major leap. He joined the Bureau of Reclamation as a wildlife biologist in 1971 and inherited Commissioner Dominy’s legendary harpoon-size electrical pointer from the bureau. Guenther has no use for what he calls “environmental extremists” who have saddled Americans with “a Chicken Little mentality,” in which “everything man does is considered bad” for the planet.
But in discussing climate change, the sometimes reactionary Guenther sounded almost as progressive as Peter Gleick: “We can’t just pray for more rain. We need to develop creative strategies that will allow us to use our limited water in a sustainable way.”
Guenther walks with a slight limp and tilts to one side as he sits. “It’s from flood fighting in the desert,” he explained. “It’s ironic, but that’s life in Arizona.” In 1993, fighting a flash flood on the Gila River, he blew out two of his cervical disks trying to save his airboat in a biblical downpour. “We’re either in drought or flood here. There’s no such thing as normal,” Guenther said. “They tell me that’s what it’s going to be like in climate change”—more violent shifts in weather, causing droughts one minute and floods the next. “Life is getting to be a whole lot more interesting.”
Although Arizona is defined by its deserts, it had “banked,” or stored underground, a surprising amount of water, some 3.1 million acre-feet, or about one trillion gallons. Guenther has ensured that Phoenix has a diverse portfolio of water supplies: the Central Arizona Project (CAP), a massive 336-mile aqueduct that annually siphons 1.5 million acre-feet of water from the Colorado River and jags across the desert like a dark scar; the Salt River Project, a series of canals and hydroelectric dams on the Salt River; and from aquifers. Begrudgingly, Arizonans have accepted what Guenther calls “the low-water-use lifestyle”—which includes xeriscaping (replacing grass lawns with cacti and rock gardens), strict limits on lawn watering, the restoration of rivers, interstate water deals, and using treated effluent for irrigation—which has allowed Arizona to use less water per capita every year.
But the problem in Phoenix is the same as in Las Vegas, Los Angeles, or Dallas: as the state’s climate warms and population climbs, so does its need for water. By 2009, Arizona had 6.6 million residents, and by 2030 that number is set to double. “I understand why people want to come here. It’s a nice place to live,” said Guenther, himself a transplant from Long Island, New York. But sometimes, he said, “people have to be reeducated” about water use in the desert.
In 1980 the USGS estimated that groundwater beneath Phoenix had dropped 220 feet in places, sometimes causing sinkholes. With the threat of a federal ultimatum looming, the state legislature restricted ground-water use in parts of the state and mandated that new homes have a hundred-year supply of renewable groundwater. But people accustomed to cheap, plentiful supplies have found ways around the regulations.
In 2008 Arizona used about 8 million acre-feet, or 2.3 trillion gallons, of water. Demand continued to increase, as weak zoning laws promoted growth. In over 80 percent of Arizona, developers can build subdivisions with hundreds of houses even if water supplies are known to be insufficient. Planned communities such as Verrado, on the western edge of Phoenix, would not have been built under the original water laws because they were too far from renewable supplies; but with help from the legislature, Verrado worked around the renewable-supply requirements. Rural wells—many of which are not required to use meters—are steadily drawing from the same aquifers that cities use for water banking. “That really scares us,” said Kathryn Sorensen, water resources director for Mesa, a city that has banked water for years. “We hope [the water] is there when we need it. But we don’t have control over the [rural] water pumping.”
Phoenicians, many of them from snowbelt states, insist on artificial lakes, green lawns and golf courses, shopping malls kept at arctic temperatures, and other resource-intensive methods of beautification. One of the most popular cooling methods for restaurants, shopping centers, and even homes is to use outdoor misting systems, which create a light fog on patios. (Misters are popular in other desert cities, most notably Palm Springs, California.) A typical home misting nozzle uses about one to fifteen gallons of water per hour, equivalent to running a large load of laundry. But a fifty-foot misting system on a restaurant patio might consume as much as fifteen hundred gallons of water per night, which is equivalent to a week’s worth of water for a household of two. It is so hot in Arizona that water misted during the day “flash evaporates” almost immediately. The state does not regulate misters, though there are “plans” to ban them in case of severe water shortage.
Arizona doesn’t know exactly how much groundwater it has, or how fast it is being used, and, according to the Bureau of Reclamation, it could face a water crisis by 2025.
With a glint in his eye, Guenther disputed that assessment and said that while he was concerned about climate change, he fully expected man to engineer himself out of trouble “as we always have in the past.” He believes that by refining existing technologies such as “weather modification” and desalination, we already have drought-proof water sources in hand. People have used these technologies for years, but results have been mixed, and skeptics question whether they are worth the effort.
I will give you rain in due season, and the land shall yield her increase, and the trees of the field shall yield their fruit.
—Leviticus
When rain falls from clouds, it does not always arrive when and where it is useful to man. But what if we could control the weather and produce rain whenever we like? This question has consumed people for years.
Meteorology was considered a black art until the nineteenth century, when James Pollard Espy, aka the Storm King, transformed it into an empirical science. Through careful study of weather, he developed a convection theory: air rises as it heats, expands, then cools; water vapor condenses in the clouds, then falls to earth as rain, snow, sleet, or hail. Espy believed that he could summon rain on command and tried to convince Congress to pay him to start forest fires so that he could demonstrate how his on-demand precipitation would snuff them out. Congress declined to fund Espy’s project.
In the late nineteenth and early twentieth centuries, “pluviculturists” promised they could draw rain from clouds by “puncturing” them with pointed balloons, blasting them with percussive artillery barrages, releasing steam from chimneys, or sprinkling clouds with “electrified sand” from a plane.
In 1946, two General Electric chemists in Schenectady, New York—Vincent Schaefer and Irving Langmuir—discovered that when dry ice was dropped into a cold cloud, it produced crystallized water vapor, better known as snow. Bernard Vonnegut, another GE chemist (and the older brother of novelist Kurt Vonnegut), found that silver iodide created snow in clouds. Their insight was that they didn’t have to create new clouds; rather, they could induce precipitation by adding crystallization to existing rain clouds. They termed their methods cloud seeding. Langmuir, who won the 1932 Nobel Prize in Chemistry, believed that this was his most important discovery and spoke of the potential to end drought and hail, control snowfalls, and turn the Southwest into a verdant garden.
The trick is to fly above already moisture-laden clouds in an airplane and “seed” them with a trail of silver iodide or dry ice (frozen CO2) particles, which causes the precipitation to fall. Some scientists estimate that under perfect conditions spreading silver iodide into a cloud can increase precipitation by 10 to 15 percent.
Cloud seeding has been used to produce rain around the world for over sixty years, and today numerous dry nations—Morocco, Saudi Arabia, countries of West Africa—are researching the technology as an answer to their chronic lack of water.
The Xinjiang region of China is home to the world’s largest cloud-seeding initiative. The China Meteorological Administration fires tens of thousands of rockets and cannon rounds loaded with silver iodide into the skies to promote rain. From 1999 through 2006, China claims to have produced 36 billion metric tons of artificial rain per year, reduced the size of hailstones (which destroy crops and houses), and suppressed forest fires caused by lightning strikes. China’s latest five-year plan calls for increasing artificial rainfall to 50 billion metric tons a year.
But tinkering with the weather can go badly wrong. In November 2009, clouds were seeded over China to alleviate a drought, but the temperature suddenly dropped, and the resulting blizzard closed highways and the Beijing airport, while heavy snows in other cities collapsed roofs, injured scores, and killed at least eight people.
Does weather modification really work? Proponents such as Herb Guenther believe it does. But critics say it is impossible to distinguish cloud-seeded rain from natural rain, that the amount of precipitation created by silver iodide cannot be measured, and that controlled experiments to answer these questions are impossible to construct.
In the 1970s, the federal government was spending $20 million a year on cloud-seeding studies, but its scientists concluded that most “moisture accelerators”—cloud-seeding entrepreneurs who charged farmers hefty fees to produce rain—were charlatans, and federal research money disappeared. In 2003, the National Research Council of the National Academies issued a report that stated, “There is still no convincing scientific proof of the efficacy of intentional weather modification efforts.”
Nevertheless, industry groups such as the Weather Modification Association have continued to lobby for federal support, and adherents persist. By 2006, sixty-six weather-mod research programs were under way in ten states. Ski areas, such as Vail, contract private firms to seed clouds for snow. A five-year $8.8 million experiment in Wyoming is under way to determine, once and for all, whether rain and snow can be squeezed from clouds on demand, and when and where people want it. At the conclusion of the study, lawmakers will have to decide if research should continue. “If this is shown to increase the [water] supply by ten percent, it would be very valuable,” said Mike Purcell, director of the Wyoming Water Development Commission.
The evidence suggests that cloud seeding works to a limited extent, when conditions are right, but that it is difficult to control. It could be a useful tool but will not be a major new source of water. Other, more proven technologies for creating new drinking supplies hold much greater promise.
Water, water, everywhere,
Nor any drop to drink.
—Samuel Taylor Coleridge,
The Rime of the Ancient Mariner
If we could ever competitively, at a cheap rate, get freshwater from salt water, that would be in the long-range interests of humanity which would really dwarf any other scientific accomplishments.
—President John F. Kennedy, 1962
Ninety-seven percent of the earth’s water is too salty for human consumption, yet dreams of turning salt water into a “new” supply of freshwater keep Herb Guenther up at night. “It is the municipal water supply of the future,” he said. The technology could provide a “limitless supply” of drinking water, on demand.
The oceans hold 321 million cubic miles of water. When salt is dissolved in water, it breaks the ionic bonds that bind salt crystals together; removing those resulting salt ions requires a lot of energy. The problem with distilling seawater is that it requires so much energy that it can be used only in small shipboard systems, or in places such as the Middle East where money and fuel are plentiful and freshwater is scarce.
A few “multi-effect” distillation plants, which use a cascading system of chambers, were built in the early nineteenth century, but mineral deposits on heat exchangers hampered the system. By the mid-twentieth century, a second method was developed that pushed salt water through a set of fine semipermeable membranes, which remove most of the salts. The best-known form of this process is called reverse osmosis, or RO.
In 1965, an experimental RO plant was built in Coalinga, California. And in 1980, the world’s first municipal seawater desal plant opened in Jidda, Saudi Arabia. These early plants required huge amounts of energy to run (the Jidda plant needed more than eight kilowatt hours [kwh] of electricity to produce one cubic meter of drinking water) and were considered too expensive and technically complex to be widely popular. Furthermore, the membranes were expensive to make and difficult to keep clean.
But the dream lived on. Cruise ships and aircraft carriers, arid seaside resorts in the Caribbean, and wealthy petro nations in the Middle East began to desalinate water regularly. This pushed down prices and led to the development of new technologies. Today, greater economies of scale, improved energy-recovery devices, and more sophisticated membranes have lowered the cost further, making desal on a larger scale more feasible.
During the extended drought of the late 1980s, several desal plants were built along the California coast. One was operating in Santa Barbara, but three months after the plant was completed, the drought abruptly ended with a series of drenching rainstorms referred to as the “March miracle.” The Santa Barbara plant was mothballed. (Much of the equipment has been sold off or is obsolete, and the city’s homeless have taken up residence around the plant’s perimeter.) Since then, refinements have brought down the cost of desalting ocean water from about $2,000 an acre-foot in 1990 to about $800 in 2007, which was slightly above the cost to coastal California cities of importing water from the Sacramento Delta and the Colorado River.
By 2005, more than two thousand desalination plants were operating in the United States, and today many more are on the drawing boards. The American Water Works Association, a trade group, predicts the desalination business will grow by $70 billion over the next two decades.
Not all desal plants are built on the ocean. About a quarter of them, such as the planned American Waters plant on the Hudson River, just north of New York City, process river water. Others, such as the Bureau of Reclamation’s Yuma Desalting Plant in Yuma, Arizona, process brackish inland water. Because inland water has less salt and other pollutants than ocean water, desalting it requires less power and fewer membranes than ocean desal and is therefore less expensive. The Yuma plant was built to reduce the salts in the Colorado River water that the United States is obliged by treaty to deliver to Mexico. It took twenty years and $250 million to build the plant, but thanks to design flaws and changing environmental circumstances, it has been idled since 1993. When I visited Yuma in the spring of 2007, the plant had just undergone a $30 million upgrade and was being run at 10 percent of capacity. It may be used for its original purpose in the future, but at present the Yuma desalter is run as the nation’s only full-scale RO research-and-testing facility, to help answer the question, is desal the magical tool that will solve the world’s quest for a drought-proof supply of water?
Saudi Arabia, where temperatures hover at 130 degrees in the desert, and the population is rising to an expected 36.4 million by 2020, once considered towing an iceberg from Antarctica or filling oil tankers with water from Norway to slake its people’s thirst. But as prices and technology have improved, the Saudis have become the world’s largest users of desalination, with twenty-eight plants supplying 70 percent of the kingdom’s water. In April 2009, King Abdullah pushed a button to start up the world’s largest desalter: the $3.4 billion Shoaiba Desalination Plant, in Jubail Industrial City, on the Persian Gulf, which produces eight hundred thousand cubic meters of water a year and generates electricity for 1.5 million people. But the nation’s desal plants consume 1.5 million barrels of oil a day, and the Saudi government plans to open the world’s largest solar-powered desal plant in 2012.
One lesson learned in Saudi Arabia is that the high cost of a desalter can be reduced when built next to a coastal power plant, which typically uses over a million gallons of seawater a day and can supply relatively cheap electricity. This was the theory behind choosing Tampa, Florida, as the site for the largest desalination plant outside the Middle East.
In 1998, twenty local governments in Florida, locked in a fierce competition over dwindling groundwater supplies, put their differences aside and established Tampa Bay Water (TBW). It is now the largest water wholesaler in the state, serving over 2.5 million people. In 1999, TBW approved a plan to construct a large desalination plant at Apollo Beach. The company driving this plan was a small firm called Poseidon Resources Corp., based in Stamford, Connecticut. Poseidon and its partner, the engineering giant Stone and Webster, estimated the desalination plant would cost $110 million to build and would produce water costing $677 an acre-foot.
Walter Winrow, Poseidon’s cofounder and president, was an engineer who had spent most of his career working on power projects around the world for GE. “I travel a bit,” he said blandly, by which he meant he was in near-constant motion, monitoring desal projects and drumming up new business for Poseidon (he declined to be specific about where he travels). “Freshwater,” he said, “is a growth opportunity.”
In 1994 Winrow saw an unfilled need: “On one side, the equipment and service providers were looking to sell power and their services; on the other side were public agencies; but there was nobody in between to mediate.” He quit GE to help form Poseidon Resources, where he put together small water deals that eventually led to bigger deals. He forged a partnership with Pemex, Mexico’s giant state-owned petroleum company, and Ionics, an American company experienced in membrane technology, to develop desal plants in Mexico.
In 1999, in Tampa, Winrow saw another opportunity: a bay with water of low salinity, a supply of cheap energy from Big Bend Power Plant, a growing population, and generous tax breaks. The desal plant he envisioned would be privately owned and operated. According to the plan, it would begin producing a steady supply of freshwater in 2002, at the unprecedented wholesale cost of $1.71 per thousand gallons. (At the time, big desalination plants could produce a thousand gallons of water for $4 to $6, under ideal conditions, and for as much as $10 in less ideal conditions.) Groundwater pumping, by contrast, cost about $1 per thousand gallons wholesale but was subject to the vagaries of supply, which in turn depended on the weather, pollution, and other people’s pumping. But after Winrow’s partners went bankrupt, Tampa Bay Water grew restive and bought out Poseidon in 2002. TBW hired other companies to finish and operate the plant, but there were numerous problems. Finally, in December 2007, five years behind schedule and $48 million over budget, the Tampa Bay desalination plant began to produce 25 million gallons of drinking water a day for $1,100 an acre-foot ($423 per acre-foot more than Poseidon had estimated it would cost).
The $158 million plant is the largest desalinator in the country. It aims to produce 25 million gallons a day (mgd) of freshwater, but during the drought of 2009 it was limited to 14 mgd by a cracked pipe, a blown transformer, and other issues. In 2010, the plant was producing 15 mgd, but with heavy rains and an 18.4 percent slump in demand since 2007, caused by Florida’s mortgage crisis, the plant was temporarily shut down to save money; this angered state regulators, who threatened to penalize Tampa Bay Water for failing to live up to its agreement.
An industry expert told the Wall Street Journal that everyone involved in the Tampa project was guilty of “sloppy work.” Ken Herd, TBW’s operations director, said that the plant is “mostly” running smoothly, and that while desalination “is not the cheapest source of supply, it is drought-proof.”
Winrow, meanwhile, was looking for the right combination of demand, location, and financing to cobble together his next big desal deal. He found it in Carlsbad, California.
In 2002, California passed Proposition 50, which provides grants for desalination, and in 2005 the state’s Department of Water Resources underwrote $25 million worth of desalting projects. About twenty of these have been proposed for the coast, but they are all waiting to see what happens to Poseidon’s latest gamble. If the Carlsbad desalination plant is built, it will supplant Tampa’s as the largest desalter in the western hemisphere.
Carlsbad lies between San Diego and Los Angeles. For the moment, Poseidon’s operation is not much to look at: a wide green lagoon, the old Encina Power Station, a shimmering black parking lot, a mobile home, and a long blue contraption that is mostly tanks, pipes, and valves—a scaled-down version of a reverse-osmosis desalinator.
Peter MacLaggan, the Poseidon senior vice president in charge of the project, looks like a surfer but speaks like an industrialist. He toured me along the scaled-down desalinator: it is about as long as a railway car, begins with a couple of tanks, proceeds to a panel full of gauges, then to two long white tubes, and ends with a water fountain. It is essentially a mechanical intestine. Seawater used for cooling the Encina Power Station is piped through a series of filters to remove impurities. Purifying chemicals are added, then the water is pumped through reverse-osmosis membranes (the white tubes), which remove salts and other microscopic impurities. (If the plant is built, it will use a slightly different process developed in Israel by IDE Technologies.) It takes two gallons of salt water to make one gallon of “ultra-high-quality freshwater.”
At the end of the tour, MacLaggan offered me a drink at the fountain. I leaned down and took a sip. What was ocean water about half an hour earlier now tasted like cool, clean tap water. Actually, it tasted more like the RO-cleansed bottled waters made by Aquafina and Dasani, which is to say it has been so thoroughly stripped of minerals that it doesn’t have much identifiable character at all. MacLaggan said that the total dissolved solids in this water are about half that of the existing water supply, which pleases local biotech and other high-tech industries that rely on superclean water for their manufacturing.
It will cost an estimated $450 million to build the full-scale plant in Carlsbad, which will be housed in a low, square one-story building across the lagoon. The plant will produce 50 million gallons of drinking water a day, enough for three hundred thousand residents, or about 8 percent of San Diego County’s water consumption in 2020.
In keeping with the new emphasis on “blue,” or water-smart, technology, Poseidon will offset its footprint by building or remediating 66 acres of wetlands, planting trees, using efficient pumps, and purchasing renewable energy credits. The company intends its Carlsbad desalinator to be the first major infrastructure project in California to be completely carbon-neutral.
MacLaggan had been pushing the plant through the permitting process for five and a half years, facing down protests, lawsuits, and negative editorials. The Carlsbad project is the test case for desal in Southern California, and many people were still following its stop-and-start progress with zealous interest.
Critics such as Food & Water Watch noted that due to spiraling costs, Poseidon might need to sell its water for as much as $2,000 to $3,000 an acre-foot, which is far more than the $950 an acre-foot the company said it would charge in 2008. (MacLaggan said his water would cost about $1,690 an acre-foot. By comparison, local water agencies charge $1,150 per acre-foot.) They also complained that desal requires finicky, expensive membranes to remove salts from seawater, and that desal requires tremendous amounts of power, which will be costly as fuel prices rise and adds greenhouse gases.
“Ocean desalination, quite frankly, is the SUV of water,” Mindy McIntyre, of the Planning and Conservation League, editorialized in the Los Angeles Times. “It requires more energy to desalinate a gallon of ocean water than it does to pump water from Northern California over a mountain range all the way to Southern California.”
In a counter-editorial, MacLaggan replied, “McIntyre’s Model T–era assertion [is] incredible and outdated…. Yes, energy is one of the cost variables associated with the production of desalinated water; however, the same is true for the transportation of imported water and the treatment of reclaimed water. In truth, the escalating energy costs … will affect all means of new drinking water.”
A broader environmental critique holds that desal plants use enormous amounts of seawater, which presents two problems: first, seawater provides habitat for plankton, fish, seaweeds, and other marine life, some of which are killed as the plant inhales water; second, seawater is considered a public resource, one that desalinators want to use for private profit.
To build the plant, Poseidon must get permits from numerous bodies, including the powerful California Coastal Commission, whose staff scientists rejected Poseidon’s proposal four times and recommended that the commissioners not approve the project because of the potential destruction of marine life. MacLaggan grew frustrated and in an unguarded moment snapped to local newspapers, “Our intake will kill about two pounds of fish a day. That’s less than the daily consumption of one pelican.” (A few months later, a scientist with the San Diego Regional Water Quality Board found that, due to mathematical errors, Poseidon had underestimated the number of fish it would kill by a factor of four.)
Perhaps the most contentious aspect of desal, however, is how to dispose of the highly concentrated salt brine left over from the plant’s water-cleansing. Every hundred gallons of desalinated seawater yields fifteen to fifty gallons of drinking water (depending on the process, and how salty the water is to begin with), and fifty to eighty-five gallons of brine. When the highly concentrated brine is flushed back to sea, it can destroy aquatic species, particularly those in the egg or immature phase. One environmentalist described the brine as “like the blood of the creature in Alien—it’ll eat through anything. The desalinators don’t know how to get rid of it, and they don’t want to talk about it.”
At the Carlsbad plant, the plan is to mix the brine with seawater discharging from the Encina Power Station and pump it offshore; Pacific wind, tide, and currents will mix and diffuse the brine, MacLaggan said. “A thousand feet offshore, we will get [the brine salts] down to thirty-six parts per thousand. By the time it reaches the kelp beds two thousand feet offshore, it’s down to thirty-four parts per thousand. It quickly dissipates. The impact on the marine environment will be de minimis.”
Surfers, fishermen, and environmentalists don’t buy the argument. The Surfrider Foundation has filed several lawsuits to stop Poseidon, claiming that the Carlsbad project would “kill everything that floats,” in the words of Surfrider’s Joe Geever—including the garibaldi, the state marine fish—and that it is not required to use the best water-intake technology available.
In 2009, the state’s coastal commissioners overrode their staff scientists and approved Poseidon’s permit, albeit with a list of twenty conditions attached. After a decade of contentious debate, which included fourteen public hearings and five revisions to the plan, the Carlsbad plant cleared its final regulatory hurdle when the San Diego Regional Water Quality Board unanimously approved permits for the Carlsbad desalter. Environmental groups have vowed to keep fighting the Carlsbad project and a sister plant in Huntington Beach, just south of Los Angeles. MacLaggan says that the fight over the Huntington plant has been even more vituperative than the one in Carlsbad. But as wildfires raged and the public worried about drought, Governor Schwarzenegger said of desalination, “We need it. It’s not a choice,” and in 2009 he green-lit both Poseidon projects.
By late 2010, the company had all its permits for Carlsbad in place when a deal for nine cities and agencies to buy its water fell apart over the issue of financial guarantees. The San Diego County Water Authority, a water wholesaler, stepped in. But the city of Carlsbad balked and demanded a guarantee that it wouldn’t lose money. As of this writing, Poseidon is negotiating with the county and is preparing to offer at least $530 million in tax-exempt bonds to private investors in 2011 to finance construction.
It remains to be seen whether the two plants will be built, and, if they are, how that will impact numerous other desalination projects proposed across the country.
Herb Guenther in Phoenix and Pat Mulroy in Las Vegas predict that a greater use of desal in America is “inevitable.” But with the enormous costs to build and operate a plant, the politics of desalination, and the energy and environmental hurdles, it won’t be easy. Some of the more ambitious programs require the use of nuclear power, which will entail further complications. As landlocked states, Arizona and Nevada intend to build a nuclear-powered facility on the Mexican coast, which would add diplomatic, trade-pact, and environmental considerations. (Freshwater created by the plant would be used in Mexico, in exchange for giving the states greater use of Colorado River water.)
“At the end of the day we will use desalters, but they will only be one tool in the toolbox,” Mulroy concedes, in what has become a common refrain.
The growing need for freshwater and the rising costs of procuring it from distant sources has raised the stakes for desal. In 2008, according to the Wall Street Journal, 13,080 desalination plants around the world produced some 12 billion gallons of water a day. Even some longtime critics have been won over. “Ten years ago desalination was the crazy aunt in the attic. That’s changed. It is now entering the mainstream and being taken seriously,” NRDC’s Barry Nelson said in 2003.
Desal has yet to gain wide acceptance, but the technology is being refined and costs are dropping. This has led some of the nation’s leading companies, such as the computer-chip-maker Intel, to aggressively pursue desal as a viable alternative to water imported from the overextended Colorado River and Sacramento Delta. To those who can afford it, then, desalination provides a relatively “drought-proof” (if not entirely green) source of freshwater.
For many years, Intel was so focused on creating the world’s smallest, fastest microprocessors and building its presence in the global marketplace, that the company didn’t pay close attention to the impact its chip manufacturing had on the environment. Intel is responsible for three Superfund sites in California, and the company has faced messy public opposition in New Mexico, where state regulators cited Intel for violations of wastewater and air-emissions-equipment rules, and residents worried that its chip-making plants were using millions of gallons of water while impacting air quality and the fragile desert ecosystem near Albuquerque. In the early 2000s, Intel built two new “fabs,” or semiconductor fabrication plants, Fab 12 and Fab 22, in a place even hotter and drier than New Mexico: Chandler, Arizona, next door to Phoenix. Then, in late 2007, the company opened a “mega-fab” in Chandler, Fab 32, which cost $3 billion to build and is a state-of-the-art facility: it recycles or stores about 75 percent of the water it uses and is among the most water-efficient plants in the company’s global operations.
Fab 32 is a behemoth in the desert: a long, low, wide building clad in gray metallic tiles. Big enough to fit seventeen football fields inside, the 1-million-square-foot factory employs over a thousand people, some of whom spend their days clad in full-body white suits, like Woody Allen in Sleeper, padding up and down white hallways and etching computer chips in 184,000 square feet of clean rooms. Fab 32 was Intel’s first high-volume producer of its forty-five-nanometer (45 billionths of a meter) transistors, which are so small that more than 2 million of them can fit on the period at the end of this sentence. Millions of these transistors are used in processors for computers and servers. They are produced on three-hundred-millimeter silicon wafers, which cost less, and use 40 percent less water, than the older two-hundred-millimeter wafers. Fab 32 produces some of the most advanced computer chips in the world. (Though, in a demonstration of Moore’s law, Intel was producing twenty-two-nanometer transistors at other plants by late 2010. To keep pace, Intel plans a $7 billion upgrade to its Arizona fabs to manufacture its latest chips there.)
To many observers, Chandler seemed an eccentric place for Intel to build such a large, costly, high-tech fab. For one thing, the city was not much to look at. Chandler began as a ranch, grew into a small town, and eventually expanded into a pleasant, bland satellite of Phoenix. Forbes ranked Chandler as “one of the most boring cities in America,” based on how rarely it was mentioned in the press (it is best known for its Peacock Day parade); it hardly seemed like a natural fit for a sophisticated, global, bleeding-edge technology firm such as Intel. More important, Chandler sits adjacent to the hottest city in the nation, in the shimmering Valley of the Sun, in the blazing Sonoran Desert, where water and the ecosystem are constant worries.
Computer-chip fabrication is water intensive. In Silicon Valley, California, the center of US microchip production, fabs from several companies account for a quarter of the water consumed and have faced complains about air and groundwater pollution. Of the twenty-nine Superfund sites in Silicon Valley (the most concentrated number in the United States), nineteen were contaminated by TCE, PCBs, and Freon from computer-chip manufacturers. Fabs are thirsty because as each of several dozen semiconductor layers is applied and etched to a silicon wafer, it must be rinsed by an atomized spray of water to keep it clean. Water has unique properties that remove molecular contaminants. Microprocessors are so sensitive that even minute particles in the water—traces of perfume or cologne, lotion, mold spores, or even smoke particles—can destroy a wafer. Computer-chip fabrication requires “ultrapure water,” which acts as a sponge for microcontaminants, such as colloidal solids, particles, total organic carbon, bacteria, pyrogens (fragments of bacteria), metal ions, and the like.
A few years ago, the three Intel fabs at the Ocotillo campus in Chandler would have required at least 7 million gallons of water per day to produce their transistors. But in the pointy-headed world of Intel, experience shapes ideas, ideas lead to actions, actions have consequences, and consequences provide further experience, which feeds back into the learning loop. Or, as Len Drago, Fab 32’s environmental health and safety director, put it, “We learned from our mistakes.”
As it was being criticized in Albuquerque in the late 1990s, Intel apparently underwent an epiphany and made green technology a corporate priority. The company invested heavily in conservation, pollution control, renewable energy, emission reductions, and recycling initiatives—all of which saved resources and emphasized efficiency, provided environmental credibility, and ultimately saved it money. “It’s about doing the right things right,” says Intel’s CEO, Paul Ottolini, who has linked employee compensation to achieving environmental goals.
“Water conservation was a big, big focus here from the get-go,” said Tom Cooper, a cheerful blond hydrologist who runs the company’s water programs worldwide, as he toured me around Fab 32.
Between 1998 and 2006, Intel invested about $100 million on water conservation projects, $20 million of that in Chandler alone, which saved over one hundred thousand acre-feet of water, equivalent to about 36 billion gallons. By using recycling and conservation technologies, and streamlining their manufacturing processes, the fabs in Chandler have reduced water use to about 2.5 million gallons a day.
Intel is still the largest water user in Chandler, by far, but these numbers are remarkable and set a standard for other manufacturers to emulate, or at least to aim for, as water demand rises and supplies grow stretched across the country.
Worried about industrial espionage, Intel bars outsiders such as me from touring the inside of its fabs, so Cooper showed me around the outside of Fab 32’s futuristic circulatory system. City water, which originates from the Colorado River, is squeezed through a series of membranes until its mineral content is a hundred-thousandth that of water in the river. The ultrapurified water goes into the fab to wash chips. Brine, left over from the water cleansing, is fed into a tall silver evaporating tower that looks like a toy rocket. It sends rinsed water back into the system and the salts to a series of evaporation ponds. The rinsed water is treated, and the resulting gray water is used in cooling towers and air scrubbers, and to irrigate the campus’s xeriscaped grounds. Finally, Intel sends 1.5 million gallons of water a day to Chandler’s $19 million RO desal plant, which the company paid for. After being cleaned to drinking standards, this water is injected six hundred feet underground into a sandstone aquifer beneath the city, which has enough stored to survive a major drought.
Intel hopes Fab 32 will become a LEED (Leadership in Energy and Environmental Design, a green building standard)-certified manufacturing facility. It has room to build two more fabs on the Ocotillo campus and has already worked their water budgets into its calculations. The company has invested some $9 billion in Chandler; it employs ten thousand people statewide, and it pays an average of four times the median salary. Its three fabs recycle tons of waste, and the company donates wooden packing boxes to local nurseries, used copper to sculpting classes, and tons of coffee grounds as mulch to the local botanical garden. In Chandler, Intel had extensive negotiations over water use with the city before building its fabs, built a desalination plant and brine-evaporation ponds for the city, and has been a model corporate citizen. But one of the lessons of the greening of Intel is that such innovation is not easy, or cheap.
Wired points out that the company’s environmental record doesn’t take into account the energy required to operate the pumping, recycling, and recirculating of water on the Ocotillo campus. The three fabs there use enough power to supply fifty-four thousand homes a year, much of which comes from the Palo Verde nuclear plant, which uses 20 billion gallons of water a year in its massive cooling towers. Nor do Intel’s numbers take into account the water footprint of its workers.
Nevertheless, the chips Intel builds at Fab 32 help power the computers used to keep American water supplies abundant and clean. And the company has produced a “tool kit” on water use developed by a “virtual team” (every Intel fab around the world contributes at least one employee to the virtual team) to use internally—and, in theory, with other companies willing to share information about costs, return on investment, problem solving, and the like.
Some Arizonans worry that Chandler is more beholden to Intel than the other way around, and that if an environmental problem arises the company could easily steamroll city leaders. (Though when an evaporation pond—built by Intel and run by the city—began to stink, upsetting neighbors, Intel bent over backward to fix the problem.) Traditionalists worry that Intel is changing the nature of the region and supplanting them, though this is a common reaction to the shifting demographics and rising urbanism of the Southwest.
In places such as Chandler, the past, present, and future of water come together. Here the New West and the Old West are learning to cohabitate and share, and water is the key resource that brings them together and could split them apart.
In Unquenchable, Robert Glennon writes approvingly of Intel’s economies of scale at the Ocotillo campus, but his words might send a chill down the spine of local agriculturists: “It takes roughly 135,000 gallons of water to produce one ton of alfalfa, but it takes fewer than 10 gallons to produce [an Intel] Core 2 Duo microprocessor…. Each acre-foot [of water] used to grow alfalfa generates at most $264. That same acre-foot used to manufacture Core 2 Duo chips generates $13 million.”
Fab 32’s futuristic circulatory system must have seemed like a nearly unattainable dream a few years ago, when its critics sniggered at Intel’s environmental missteps. But just as it transformed itself into the world’s largest semiconductor maker, Intel refashioned itself into a leading environmental steward and built one of the most water-efficient factories in the world.
I asked Tom Cooper if Intel could ever become “water neutral,” meaning that it would recycle or offset all of its water use so it had zero net impact on the environment, as Coca-Cola and others aspire to do. Cooper stared off into the distance, then changed the subject. Clearly, the question bugged him. A few days later he e-mailed me: “To be candid, we don’t know yet. What we can be sure of is that the solution will boil down to: collect and analyze lots of data, communicate internally and externally, have persistence, and use a lot of patience.”
Cooper doesn’t believe Intel will revolutionize the way Americans use water overnight. There is no systems theory of water, no silver bullet to solve every water problem. But the efficiencies built into Fab 32 indicate a “blue” path to follow in coming decades. As Cooper put it, “Lots of little, constant improvements eventually add up to big improvements.” And in this drying century, every drop counts.