by John Fleck
Thus each basin had its own great savings bank—Lake Powell for Wyoming, Utah, Colorado, and New Mexico, and Lake Mead for Nevada, Arizona, and California. Lake Powell’s role was to store water needed to pay the Upper Basin’s 7.5 million acre-feet per year bill to the Lower Basin. Lake Mead was a savings bank to store water needed for water users in Nevada, Arizona, California, and Mexico. Manage the savings accounts right, and there would be enough water to meet everyone’s need when it was time to dig their metaphorical shovels into the ground and turn water into their ditches. But the rules were never entirely clear, and by the early 2000s, Upper Basin water users feared a downstream decree, in the form of legal action to enforce the 7.5 million acre-foot obligation, would trump their upstream shovels and ditches—so, as Lake Powell dropped, they would have to curtail their use to keep the full amount flowing past Lee’s Ferry.
The Mexican Obligation
January 2000, as Fulp and his colleagues were finishing up their analysis of Colorado River Basin operations, would be the last time for a while that Lake Mead lapped at Hoover Dam’s spillways.3 The decline of the great reservoir began in the summer of that year. By 2015, Lake Mead had reached the lowest level it had been since the Bureau finished Hoover Dam and began filling it in the 1930s.
The problem, as Fulp saw so clearly, is simple: downstream water use is greater than the supply that nature and the upstream water users provide. But after a big policy push in 2002 and ’03 to try to curtail California’s overuse of Colorado River water, the years that followed settled back into a predictable debate: not over how to use less water, but rather over the rules that decide who takes the hit when shortage sets in.
While the California experiment, cutting that state back to its 4.4 million acre-feet per year allotment, seemed successful, the next steps proved substantially harder. No one else wanted to take on the burden of using less water, which in the end is the only way to keep from draining the basin’s reservoir supplies. Everyone was trying to find a way of reading the rules so that, as scarcity set in, they wouldn’t have to take a water-supply hit.
To grasp the legal and political debates among basin water users in the years that followed requires a deeper understanding of the rules that lay behind that seemingly sacrosanct Upper Basin 8.23 million acre-foot delivery requirement at the heart of Fulp’s calculations. That number depended on a particular interpretation of the river’s operating rules, and as Lake Mead and Lake Powell continued to drop, that interpretation was called into question.
You need to invoke two separate rules to get to a total of 8.23 million acre-feet. The first is the Upper Basin’s legal requirement under the Colorado River Compact to deliver a minimum average of 7.5 million acre-feet per year of water past Lee’s Ferry. But to get up to 8.23 million acre-feet, you also needed to assume that the Upper Basin is required to deliver extra water above and beyond that minimum obligation to provide a share of the water delivered to Mexico under a 1944 treaty between the new nations. That number was buried in four pages of turgid prose that only a water lawyer could love: the 1970 federal regulation known as the LROC—“Criteria for Long-Range Operation of Colorado River Reservoirs.”
In 2004, the states of the Upper Basin did not like the way the LROC (they pronounce it “el-rock”) was being used. They balked at the idea that they were required to deliver a share of the Mexico water. In the current state of the river’s management, the representatives of the four Upper Basin states wrote, “the Upper Basin has no obligation in this regard.”4
If they were right, it would make the math problem identified by Fulp four years earlier even worse. With 7.5 million acre-feet delivered instead of 8.23 million acre-feet, Lake Mead would go down that much faster. To the states of the Lower Basin, it was as if a business partner with a $1.5 million annual partnership declared that he had no obligation to pay his share of the debt, a trio of Arizona water lawyers wrote.5
Any attempt to decide what’s really “fair” in distributing the basin’s water is a doomed exercise, but there is one crude, easy-to-see metric: how much water is in Lake Powell? How much in Lake Mead? The two reservoirs, book-ending the Grand Canyon, act like savings accounts for the two basins, and if one is empty while the other is full, that just looks unfair. By 2004, the balances clearly looked unfair.
Bad drought years—2002 was the worst on record—had left the upstream reservoir, Lake Powell, so enfeebled that the National Park Service abandoned the marina at Hite, in the red-rock canyon country at Powell’s upstream end. What once had been lake at the site of the marina was turning rapidly into narrow river channel as Lake Powell dropped away. Workers towed the floating gas station and marina into deeper water, leaving behind a forlorn concrete ramp stretching into what used to be a recreational boaters’ paradise but by the early 2000s was little more than a muddy former lake bottom.6
Lake Mead was dropping, too—its surface elevation down 70 feet in the first five years of the twenty-first century. Together, the combined water in storage in the two reservoirs had been cut nearly in half.
Meanwhile downstream, water users watched demand skyrocket. The population of Las Vegas grew 23 percent from 2000 to 2005, and Phoenix grew 18 percent. Water managers were nervous, but as they watched the two reservoirs drop in tandem, they had the feeling that they were all in this together. There were feuds, but the shortage was primarily treated as a shared problem.
One of the first obvious signs that it might not be such a shared problem came in 2002, the year Hite was abandoned. The lowest snowpack in history left the Upper Basin parched. Farmers in Utah and Wyoming estimated that they suffered a 337,000 acre-foot shortfall.7 With less water in the snowpack in the mountains above them, they had no choice but to use less water. Meanwhile downstream, there was so much water in Lake Mead that the US Bureau of Reclamation declared a surplus. California got 875,000 extra acre-feet of extra water in 2002, 20 percent above the state’s normal-year legal entitlement. Lake Mead was dropping, too, but Lake Powell was dropping more.8
To Upper Basin water users, this seemed unfair, and they thought they understood the cause. The states of the Upper Basin argued that their reservoir was being lowered so quickly because they were being overcharged—that they were being made to contribute too much water to meet US delivery obligations to Mexico.
Under existing interpretations of the Colorado River Compact, it had been generally agreed that each basin was responsible for half of the United States’ treaty obligation to deliver 1.5 million acre-feet per year past the US-Mexico border at Morelos Dam. But water managers in the Upper Basin had always chafed at the agreement, harboring their own legal interpretation that, when there was “surplus” in the system, the entire 1.5 million acre-feet should come from that surplus. Instead, surplus water had for years been sent to California while the Upper Basin’s dwindling bank account in Lake Powell was being charged to meet the US obligation to Mexico. Instead of declaring a “surplus” and sending the extra water to California, the Upper Basin argued, the extra water should have been sent to Mexico, allowing more water to be held back in Lake Powell as a savings account for the future.
This is one of those problems that did not come up for most of the history of the management of the Colorado River because there had long been enough extra water for everyone. But much in the same way that the basin states were forced to grapple with California’s excess use as demand finally rose to meet supply, or deal with the loss of water because of the salty water flowing from Wellton-Mohawk in Arizona, they were now forced to grapple with the issue of who bore what responsibility for the Mexican allocation.
“They’ve never had to face a shortage of this consequence,” Las Vegas water chief Pat Mulroy said at the time. “When you’re right up against it and facing the possibility of inadequate supplies to municipalities or farmers or jeopardizing recreation values, these are very tough choices.”9
With Lake Powell dropping, the Upper Basin saw a need to protect “their�
�� water. The ability of Glen Canyon Dam to generate power for some 200 mostly rural communities was at risk. If the drought continued to sap the basin’s snowpack at the rate it had been, Glen Canyon Dam would be unable to generate electricity at all by February 2006.10 Upper Basin agriculture also was being hit hard by the drought.11 With demand across the basin growing, “it is apparent these issues will not go away, even if we are blessed with a few years of favorable runoff,” the Upper Basin representatives wrote.12 The states were at an impasse, and there was a very real chance that the whole thing would end up in court.
Most of the water managers involved were too young to have lived through the cloud of uncertainty that hovered over the basin the last time such a dispute had ended up in court, when the US Supreme Court worked from 1954 to 1963 to sort out the Arizona v. California litigation. But they had internalized the lesson from that era, that courts are a lousy way to deal with questions like this. “That route takes the decision out of the water managers’ hands,” explained John Entsminger, one of Nevada’s representatives in the negotiations that followed. “Are we going to let guys and gals in black robes start making these decisions for us? Or are we going to come together and maybe not have a perfect solution from everybody’s perspective, but a solution that works for everybody?”13
The Network
The group of people who gathered to work on this issue lacks a formal name. John Entsminger, the Colorado-born lawyer who rose to head Las Vegas’s largest municipal water agency, is one of the people who calls it simply “the network.” The existence and functioning of the network illustrates the reality of problem solving in a river basin where water crosses borders, where it must be shared, but where no one is in charge.
Though there are some formal structures through which it operates, the network is not a formal thing. It includes representatives of the seven US Colorado River Basin states and the major water-using agencies within those states. A team of federal lawyers, hydrologists, and water managers are central players. It also includes a handful of outsiders—attorneys and representatives of environmental groups who have learned the lingo and earned the trust to participate in the discussions.
They are people, explained Anne Castle, a water attorney, former assistant secretary of the interior, and one of the network’s key participants, who are “laser-focused” on the reality that in order to meet their own states’ and water agencies’ needs, the basin’s broader problems need collective action toward collective solutions.14
You have to have formal institutions—federal and state agencies, formally established working groups reaching across boundaries, and officially designated negotiating teams, all operating under written rules. But for those institutions to function well, you also need informal relationships, across institutional boundaries, among people who represent different communities and interests, yet understand one another’s needs and share common values.
These people meet regularly. It has not been uncommon, when working on this book, for me to bump into more than one of them in the halls of a hotel in Las Vegas, Nevada, or Santa Fe, New Mexico. When the network gathers in Yuma, Arizona, I learned, it’s a good bet that you can find the people you want to talk to in the lobby or out by the pool at the Hilton Garden Inn on the banks of the Colorado River.
Researchers who study these formal and informal institutions talk about “social capital”—“the shared knowledge, understandings, norms, rules, and expectations about patterns of interactions that groups of individuals bring to a recurrent activity,” in the words of Elinor Ostrom.15 The word “capital” is chosen carefully, suggesting that its role is every bit as important as the physical capital: the dams, pumps, and canals through which the water moves.
It was “the network” that ultimately put together the 2002 deal to reduce California’s water use, and by 2005 it was being put to a test. Somehow, the network’s members now had to come up with a new set of rules that could both balance reservoir levels in Mead and Powell, as well as provide some certainty for how shortages would be handled among the Lower Basin states if Lake Mead kept dropping.
It meant understanding one another’s positions, but it also meant honoring a shared goal—keeping their dispute out of court. “We knew where the disagreements were,” said Entsminger, “and the choice was litigate those disagreements or find a working solution.”16
The Negotiations
In the two years that followed, there were times when Terry Fulp thought the whole thing would fall apart. Fulp had risen to be deputy director of the Bureau of Reclamation’s Boulder City office by that point, and he was given the unenviable task of trying to shepherd the negotiations to come up with new operating rules for the basin. “I had many nights that I went home late at night and sat there and thought, ‘We’re not going to make it,’” he said.17
Just as in the close-but-not-quite-there moments in 2001 and ’02 over the California allotment, a deal seemed within reach. The biggest hurdle was the need for an agreement on reduced allocations to downstream water users as Lake Mead dropped. California wanted deep cuts in the amount of water released each year to Lower Basin water users, knowing that its priority status would mean the burden of those cuts would fall entirely on Arizona and Nevada. Arizona pushed back, and the states agreed to more-modest cuts that were not deep enough to completely halt the drop in Mead, but were a start.18
But the details—how much to release from Lake Powell each year, and under what circumstances—increasingly seemed insurmountable. Rules that left enough water in Lake Powell to make Upper Basin states comfortable left Lower Basin water users, especially in Arizona, nervous that Lake Mead would drop to dangerously low levels. By the summer of 2007, the network was staring failure in the face. They had gathered in Phoenix (they picked the site to coincide with the wedding of Colorado negotiator Scott Balcomb’s daughter) when they awoke to a newspaper headline announcing that Arizona wanted to pull out of the negotiations. “We worked hard to try to put together this agreement, and the only thing Arizona asked is that it did not harm Arizona water users,” said Herb Guenther, director of the state department of water resources, to the Arizona Republic.19
Said Entsminger, “Everybody was blindsided.”20
With the possibility for agreement in tatters, a meeting the following month at Bishop’s Lodge outside Santa Fe offers one of the clearest examples of how the network operates.
Bishop’s Lodge is a historic site in Colorado River Basin history, the place where the original version of the network had gathered seventy-five years earlier to finalize the Colorado River Compact. In recent decades, a nonprofit called the Water Education Foundation has organized an invitation-only return to Bishop’s Lodge every two years to bring together senior basin water managers. It has become one of the most important gatherings of the network. There are seminars and talks, but more important are the full breakfasts before the day begins, the hallway conversations during the day, and the happy hours after. It is the place where network governance happens.
It was there in 2007, in a bar after a day of meetings, that Nevada hydrologist David Donnelly hatched the scheme crucial to saving the agreement. Donnelly’s plan offered a way, under certain circumstances, to release a bit of extra water from Lake Powell to bolster Lake Mead.
Donnelly’s breakthrough required two things that are critical to the success or failure of the process. The first was Donnelly’s deep understanding of how the system itself works, the arithmetic of water stored in one dam moving downstream to be captured by the next. But more importantly, the idea evinced a deep understanding by Donnelly, a Nevada representative, of the other states’ needs. If they failed to come up with a deal, Nevada arguably had the most to lose. If they couldn’t come up with rules to slow the decline of Lake Mead, the Las Vegas water-intake pipes were vulnerable, posing the risk of a city going without water. So for Nevada, the most important thing was a deal that could get the other six states to sign on. “Nevada had a lot
of skin in the game,” Donnelly told me. “Our main goal was to try to get everyone else on board.”21
They succeeded.
The Deal
When it was finally signed in December, the resulting agreement was historic. For the first time since the Colorado River Compact was signed in 1922, the states on the US side of the border had agreed that there might not be enough water to cash the check written in the compact—that there might be times when users would simply have to do with less.
Under the deal Arizona and Nevada agreed that when Lake Mead dropped to a surface elevation of 1,075 feet above sea level, each would take a cut in their guaranteed annual delivery. When Mead hit 1,050, the cut would be bigger. At 1,025, bigger yet.
Never before had Colorado River water users formally agreed that shortage was a reality—that full deliveries year in and year out were not an immutable right. The agreement also provided new flexibility for managing the system, allowing major water users to conserve water without losing it, banking the savings in Lake Mead for use in later years. No longer was “use it or lose it” the official operating policy of the Colorado River.
But you can see the deal’s poorly stitched seams—choices that were necessary to keep it together, but that weakened its ability to truly address the problems it was meant to address.
Under the deal, when Lake Mead drops to a surface elevation of 1,075 feet above sea level, Arizona and Nevada agreed to take less water from the reservoir. The purpose was to slow the decline, but the shortages are modest—an 11 percent cut to Arizona’s allotment, and a 4 percent cut to Nevada’s. California, the victor in the battles for priority rights in the 1960s, still gets its full allocation.
When the negotiations began, some of the participants, led by California, argued that far deeper cuts in usage would be needed to preserve the water supply over the long run. But California’s proposal came with a poison pill that made bigger cuts a nonstarter: the reductions would fall entirely on Arizona and Nevada. California thought that water conservation was critical, but that it was its neighbors’ responsibility to do it.