Water is for Fighting Over

by John Fleck

On paper, the idea won early diplomatic support. In 2000, the two nations signed a formal agreement known as “Minute 306” to begin a formal binational conversation about putting water back in the delta. (Called a “minute” after the arcane diplomatic process that generated it—the agreement reads like the “minutes” of a meeting between the diplomats of the two nations—the deal exploited a much-used process that allows negotiated tweaks to the 1944 treaty between the two nations. They are technically not “amendments” to the treaty, but have the same effect.)

  The agreement talked of “joint studies” and the need to “establish a framework for cooperation.”15 But it was toothless—just an agreement to talk, not to act. Crucially, the arena of action seemed to be all wrong. The deal had been signed by diplomats, but the paper document seemed far removed from the real world of moving water around the Colorado Basin. In practice, among water managers on both sides of the border the suggestion of somehow finding enough water in the system to send a pulse of water down into the delta was a hard problem.16 Moving beyond the accident of filled reservoirs and excess water to intentional restoration seemed in the years that followed to be beyond the Colorado River management system’s capabilities. Human demands upstream were just too great. How could water users possibly give up a share for the beavers’ sake?

  By the mid-’00s, people working on delta restoration began to realize that any agreement to put water back in the Colorado River’s channel had to be about far more than just putting water back in the Colorado River’s channel. “It started becoming clear that we needed to look at what we were doing in the delta in the context of bigger water-management decisions,” explained Jennifer Pitt of the Environmental Defense Fund.17

  Pitt, whom High Country News’s Matt Jenkins described as “a woman with a seemingly pathological drive to decipher the river’s byzantine operational mechanics,”18 had worked for years to learn the details of river management. Instead of disdaining the world of the “water buffalos,” she embraced it, learning the complex computer simulations the US Bureau of Reclamation and other water agencies used to track the Colorado’s stocks and flows.19 Any solution to the environmental part of the problem, Pitt and her colleagues came to realize, would have to be linked to the costs and benefits of problem-solving efforts across the Lower Colorado River Basin.

  The discussions seemed stalled until the ground shook, literally, on Easter Sunday in the spring of 2010. Two people died and at least 233 people were injured in the magnitude 7.2 quake, centered in the farming region of the Mexicali Valley on the western flank of the Colorado River Delta. The earthquake damaged the irrigation system used to deliver water to some 80,000 acres of farmland.20 It was a striking demonstration of the vulnerability of Mexico’s aging and fragile regional water-distribution system, according to an analysis by Vicente Sanchez and Alfonso A. Cortez-Lara, researchers at El Colegio de la Frontera Norte.21

  The system’s seismic vulnerability was certainly a problem, but also an opportunity, and the water-management institutions quickly kicked into gear. With no official deal in place, US and Mexican water managers immediately agreed at an informal level to reduce Mexican deliveries, storing the unused water in Lake Mead. For the first time in history, Mexican water was being quietly stored behind a US dam.

  Then, in what passes for haste in the normally ponderous world of international diplomacy, representatives of the two nations negotiated an agreement that implemented on a temporary basis an idea that had been kicking around for several years without a tool for its implementation. A new accounting category was created on the US water ledgers for what came to be called “earthquake water”—unused Mexican water now stored in Lake Mead.

  Interior Secretary Ken Salazar wasn’t shy about sketching out the importance of a deal. It was, he said in a Mexico City announcement, “a remarkable achievement from a humanitarian perspective, but it also lays important groundwork for a much-needed comprehensive water agreement with Mexico on how we manage the Colorado River.”22 The earthquake had been turned from problem to opportunity, Mexican water was sitting behind Hoover Dam, and the diplomatic logjam was broken.

  The Network in Action

  Much of the negotiations that followed hinged on the development of a shared understanding of a seemingly straightforward problem: how to account for the water moving through the Colorado River System from one nation to the other. But to deal with the issue required doing something that had previously been beyond reach: extending “the network” of US water managers into Mexico.

  One of the network’s key tools is a sophisticated computer model of the basin known as the Colorado River Simulation System. The Bureau of Reclamation supported an entire research team based at the University of Colorado in Boulder to develop and maintain it, and its calculations are used to make decisions about how much, where, and when to release water.

  Mexico’s water agency, CONAGUA, operated an entirely different computer model of its own design, running on different software platforms, based on different assumptions about system operations, using different underlying data sets, and therefore yielding different results. The disconnect illustrated one of the crucial insights from Elinor Ostrom’s early research on California groundwater-basin management a half century before: a shared understanding of the resource, including the detailed math needed to measure the water’s flow, is critical to any agreement. So one of the most important steps was entirely technical in nature—a series of exchanges that ultimately allowed the Mexican technicians to understand how the CRSS model works and to operate it for themselves. It let negotiations proceed, a group of participants in the talks wrote, “on a common factual footing.”23

  That shared understanding helped, but language still posed difficulties. In the end, a small negotiating group from both countries holed up in a Tijuana hotel with side-by-side English and Spanish versions of the bilingual agreement in a marathon session in late 2012.24

  Consensus on the numbers only got them so far; there was still fundamental disagreement over the final terms of the deal. It involved the water itself that would be used for the pulse flow. US officials insisted that that no American water would be sent to the delta. Instead, they offered to pay for water-system improvements in Mexico to free up water, with a part of the conserved water dedicated to the environment. But this made Mexican officials uneasy because of the perception back home that the rich gringos north of the border were unwilling to contribute the one resource that really mattered.

  In the end, a reading of the final deal, dubbed “Minute 319,” shows the two sides papering over the problem with an uncomfortable agreement that left the tough stuff for later. In the short run, US water agencies would make modest contributions ($21 million) for water-saving system improvements in Mexico. Mexico retained title to nearly all the saved water, and Mexico would in return use some of its conserved “earthquake water” for the pulse flow. The US municipal water agencies, contributing a significant share of the money for Mexican water-conservation measures, got a share of the saved water as part of a complex water-accounting swap. US environmental groups would contribute money to buy up additional water rights in order to provide water to maintain environmental restoration sites in Mexico. That allowed the modest first experiment in Colorado River Delta restoration to succeed.

  Along with the environmental benefits, two other major pieces formed the deal’s three-legged stool. The second leg was clarification about how shortages and surpluses on the river would affect Mexico’s deliveries. Under the deal, Mexico shares modestly in shortage if Lake Mead drops below elevation 1,075, the level at which Arizona and Nevada begin to take shortages. It also allows Mexico to share in surpluses if and when there is extra water in the river. The third leg was operational—the ability for Mexico to continue to store unused apportionment in Lake Mead.

  But the big stuff, where the real needs and opportunities lie, was left for later. The deal talks about $1.7 billion worth of water-saving improvements t
hat are possible in Mexico, to be jointly funded by water agencies on both sides of the border, “that could generate sufficient volumes of water to benefit both countries.” They range from adding reservoir capacity in Mexico similar to what was done with Drop 2 on the US side, to building a big new plant to desalinate ocean water on the Mexican coast in exchange for a reduced delivery obligation, freeing up that water for use in the United States. But unable to come to terms on the actual projects, the agreement embodies the promise that discussions will continue.25

  Bringing Water Back to the Delta

  In one sweeping agreement, Minute 319 formally brought two traditionally excluded groups, environmentalists and Mexicans, into the Colorado River’s formal management framework. It also felled two myths. The first was that environmentalists and water managers could not compromise to achieve common goals. The second myth was deeply embedded in our western narrative—the idea that the delta was dead, that environmental rejuvenation was impossible in a world of expanding populations and growing water demands.

  The Colorado River returns to San Luis Río Colorado (© John Fleck).

  Minute 319’s crowning bauble, the pulse of water released from Morelos Dam into the dry channel of the Colorado, briefly brought the river back to life. On March 23, 2014, the day the first water of the pulse flow was to be released, I pulled my wife’s Subaru into the dirt parking area across Baja Highway 2 from Morelos Dam. A fast-food breakfast lay on the passenger seat; the sun wasn’t yet up and I was the first to arrive. Journalist Matt Jenkins showed up next, followed shortly by environmentalist Jennifer Pitt, University of Arizona scientist Karl Flessa, Yamilette Carillo-Guerrero of the Colorado River Water Trust, and many more of the people who had worked so long to make the pulse flow happen. The formal ceremonies, with dignitaries and speeches, were not scheduled until days later, but none of us wanted to miss the first pulse as the dam’s gates were opened.

  Beyond the success of a mutually beneficial deal, putting water back into the delta just plain felt good. John Entsminger, the Colorado-born water lawyer who now heads the Las Vegas metro area’s main water utility, remembers the day as a young college student that a friend explained to him that the Colorado River didn’t reach the sea. “It was appalling,” he recalled.26

  Bob Snow, a US government lawyer who helped negotiate the agreement, recalled an uncle from Ireland telling him, “You know, rivers really should reach the sea.”27

  Minute 319 has rightly been hailed as a milestone and a significant success story. But it also demonstrates the limitations of water-management deal making in a complex, evolving environment. The deal is short-term in nature, an experiment. While it was enthusiastically received in Mexico, support was not universal, with some agricultural interests viewing the water diverted to environmental purposes as “wasted.”

  The water had not even finished flowing before the water managers who arranged it began talking about a follow-on deal—the next agreement that they know will be needed in the ever-evolving management of the Colorado River.

  CHAPTER 13

  Conclusion

  ARIZONA SENATOR JEFF FLAKE had sharp questions for Deputy Interior Secretary Mike Connor during a hearing in the fall of 2015. Arizona water managers had been working hard all year to reduce the state’s water use, hoping to leave water in Lake Mead as a buffer against shortage.

  “The number-one priority in Arizona is to make sure that when Arizona, or any other state, voluntarily contributes their water to the health of the Colorado system, the contributed water actually stays in the system and doesn’t disappear along somebody else’s canals,” Flake said, looking up from his notes with a smirk. Flake asked Connor to pledge that the federal government would not simply reassign the water to some other user. “Without these assurances,” Flake said, “obviously such preventative measures don’t make sense.” So far so good—the senator was trying to guard against the tragedy of the commons. But then Flake’s argument jumped the rails: “It’d be like having a savings account and just being able to see your neighbor reach and grab money from it.”1

  Flake was in part playing to the hometown audience. The references to “somebody else’s canals” and “some other user” were dog whistles to Arizonans and to water-management insiders across the basin, signaling his vigilance against the ever-present threat (real or imagined) that California and the federal government might be in cahoots to steal Arizona’s water.

  But beyond Arizona’s historic political paranoia, Flake’s question highlights the core problem with current management of Colorado River water. As we have seen again and again—in farm districts like Yuma and Imperial, and in urban areas like Albuquerque, Phoenix, Las Vegas, and Southern California—the communities that depend on the Colorado are learning to thrive on less water. But how to manage “saved” water remains the central dilemma.

  The problem is the clumsy metaphor itself—the idea that Arizona, in conserving water in 2015, had been contributing to its own personal Colorado River water savings account. A much more apt metaphor would be that, in leaving water in Lake Mead, Arizona is simply trying to prevent the account from being overdrawn. Given that, overall, water users are demanding more water than nature is providing, sooner or later the water that is conserved will have to be water that no one will ever get back.

  Albuquerque

  My home town of Albuquerque, New Mexico, is a perfect case study of the problem. It is no exaggeration to say that we have become very, very good at conserving water. The question of what to do with the water we’ve saved illustrates, in microcosm, the problems facing the Colorado River going forward.

  In the early 1970s, the US Bureau of Reclamation opened the gates on the San Juan–Chama Project, which diverts Colorado River water outside the river’s hydrologic basin to be used in central New Mexico, with Albuquerque getting the largest share. The project was intended to shore up supplies on the Rio Grande, a relatively small river even by the standards of the arid West, which by the 1950s was over-allocated.2 It also marked the culmination of an effort to bring more water to a rapidly growing and thirsty city. A burst of Cold War growth, fueled by one of the United States’ major nuclear-weapons research centers, made the acquisition of water essential, the state’s leadership told Congress as it was considering legislation to fund the San Juan–Chama Project. Even with the new water, the region was believed to be at risk of outstripping its supplies.3

  In the 1980s and 1990s, it looked as though that risk was becoming real. Despite the imported Colorado River water, the growing metropolitan area continued to lean heavily on its aquifer, and groundwater levels beneath some parts of the city had dropped more than a hundred feet.4

  In response, the community in the mid-1990s launched a massive water-conservation effort that over two decades cut per capita water use nearly in half. While population grew modestly in the years that followed, water use kept dropping. The community spent some $500 million on a new water-distribution system that allowed more efficient use of the imported San Juan–Chama water, and by 2015 the aquifer was rising across the big groundwater basin that underlies the Albuquerque metro area.

  It took me a long time to grasp that this was actually happening. As a newspaper reporter covering the region’s water supply, I remained steeped in old narratives of crisis, of communities at risk of running out of water. Monthly, I asked for copies of the water utility’s pumping reports, and I made a habit of routinely checking the US Geological Survey’s groundwater monitoring well data, ready to pounce at the first sign of failure. But our water use keeps dropping and our aquifer keeps rising. Grudgingly at first, but then with increasing fascination, I began writing stories about what success looked like.

  This has created an entirely different problem for the community’s water managers. By the end of 2015, the Albuquerque Bernalillo County Water Utility Authority estimated that the aquifer beneath the city held nearly a million acre-feet more water than it would have held absent the conserv
ation efforts. While the size of that number is based on generous assumptions made by the agency, there is no question about which direction water management and policy in the New Mexico city are now headed. Between its imported San Juan–Chama water and its rights to local groundwater and surface water, Albuquerque has rights on paper to 32 percent more water than it is currently using. Instead of scrambling to find new water, the agency has shifted to finding ways of managing the surplus—expanding storage space in a big northern New Mexico reservoir and ramping up efforts to inject water into the aquifer for later use.5

  But nowhere in the community’s water-management dialogue is there any mention of what might seem like an obvious response to the current situation: maybe Albuquerque shouldn’t take so much water from the over-allocated Colorado River Basin every year? Albuquerque’s water managers could offer a rational response, of course. If they left San Juan–Chama Project water unused, to paraphrase Flake’s question to Connor, wouldn’t it just end up in someone else’s canals, watering someone else’s golf courses and lawns?

  What to Do with Saved Water

  Albuquerque is not unique.

  Las Vegas, as we have seen, has demonstrated remarkable water-conservation success, using just 75 percent of its Colorado River allocation in 2014. But it still claims the full amount, stashing the surplus in groundwater storage projects scattered around the West. By the end of 2014, southern Nevada’s municipal water managers had more than a million acre-feet of water in its water banks. The result is that Lake Mead drops just as much as if Las Vegas had not conserved a drop.

  Consider Phoenix, which has important similarities to Las Vegas and Albuquerque, as well as important differences. In 2014, Phoenix used just 70 percent of its 186,557 acre-foot Colorado River water-rights entitlement. But because of the “use it or lose it” rules under which the Arizona water rights are administered, Phoenix had no way to simply leave that water in Lake Mead for the good of the overall Colorado River system. Instead, any of Phoenix’s allocation that went unused simply reverted to other users in Arizona. Where Albuquerque and Las Vegas have an incentive to conserve and hoard the water for the cities’ own growth and future use, Phoenix has little incentive to conserve water at all.6