Cliff Notes - Where the Water Goes 💧.

In “Where the Water Goes” author David Owen follows the Colorado River from its high-altitude headwaters to its eventual, trickling desert end - every drop over-promised, underdelivered, or disputed along the way. The system now has an average “structural deficit” of 1.2 million acre-feet a year, “which is equivalent to covering almost four Phoenixes with a foot of water” (source).

What to do? There’s no shortage of attempted solutions. The thing that struck me while reading this book though, was how often some unforeseen downstream circumstance counteracts a seemingly intuitive water management practice. Solving the water deficiencies presentsa complex problem, made more difficult by the Colorado’s age-old “first in, first out, forever” water rights policies (as opposed to needs-based “riparian” policies on other rivers), and the fact that so vastly many people in such increasingly dry regions depend on the Colorado, which, by many measures, is not a sizeable river.

The Colorado provides an exceptionally useful introduction to water issues because we literally use it up. The following is a quick trip downstream through the salient bits of Owen’s book, starting in Colorado.

How Water Rights Work

“Water law in most states in the West is based on the doctrine of ‘prior appropriation.’ That doctrine holds that the first person to make “beneficial use” of water gains the right to use that quantity for that purpose forever. The claim takes precedence over every claim made later and is unrelated to the user’s distance from the stream.

Eighty percent of Colorado’s precipitation, for example, falls on the western half of the state, yet eighty-five percent of the population lives to the east, in the mountains’ “rain shadow.” If a small community on the west has a prior beneficial claim, they might get the water, regardless of whether the city of Denver goes dry or not.

An even more complicated use case is California. Even though the Colorado does not flow through it, California has a priority claim on a vast amount of its flow, dating back to the early days of the gold rush. This and most other allocations were established in the 1920s in the Colorado River Compact, a doctrine endlessly disputed by its member states. Under this doctrine, Native and Mexican lands are almost entirely overlooked, and non-human animals are not considered at all.

“The fact that a natural resource might have value for species other than our own, or even that it might have aesthetic, spiritual, or recreational value for us, is not a consideration.”

Changes in supply (e.g., drought) are also not well considered. The years on which the original estimates in the Colorado River Compact were based were the wettest since the 1400s. While drought periods have recently been lasting only a few years at a time, looking back as far as the medieval era, we can see droughts lasting 70 to 140 years.

These rules are slowly changing through things like instream-flow water rights. In Colorado, “only the Colorado Water Conservation Board can appropriate them, and because the idea of preserving the environment hasn’t been around for as long as mining, farming, and ranching have, instream flows usually have very junior priority dates. During dry periods, when exercising them would be the most useful, they inevitably get called out by something senior.”

Rights are also occasionally renegotiated, such as between hydroelectric power companies and water bureaus, to address times of genuine need, illustrating that “even century-old legal precedents are not necessarily immutable—a good thing to keep in mind when people who live west of the Continental Divide talk about cutting off water to people who live east of the Continental Divide, or when people in Los Angeles talk about allowing Phoenix and Tucson to disappear. In a genuine water crisis, a small power plant in a canyon in a sparsely populated area would not be allowed to make the state capital uninhabitable.”

So, in a system like this, how do you know if the water is yours to use or whether it has some prior downstream claimant? David Owen discusses this with water rights lawyer Kent Holsinger, who recounts growing up on a small Western Colorado farm. “If the senior wasn’t getting all his water, he would call the water rights commissioner,” Holsinger said. “The commissioner would go down the list to the next junior, and if that was us he would make the dreaded call, and say we had to turn off our headgate…”

“Upper-basin states still take less than their theoretical entitlement” (~60% in 2012).“Owners of water rights cannot sell them or modify their decreed use without the court’s approval, and, if they seek that approval, potentially affected users are given an opportunity to object.”

You can lose your rights, however. One way this happens is abandonment. “When Shoshone Dam suffered a mechanical failure a few years ago its owners rushed to make repairs. The plant is especially beloved on the West Slope because its ‘return flow’ is nearly a hundred percent. Even though the plant has the theoretical power to call out the largest city on the East Slope, essentially all the water it uses goes right back into the river.” But not all water consumers are so beloved.

Hazardous effects of water use and mismanagement

Subsurface mines have typically been a major source of this consumption and abandonment pattern. Groundwater seeps in and has to be pumped out (dewatered), but it inevitably settles below the water table, leaching out all manner of toxic chemicals in a phenomenon called “acid mine drainage.” Open pit mines require massive dewatering, often lowering the water table by hundreds of feet.

Often, these mines source materials for electronics and batteries. “A cell phone alone can contain aluminum, antimony, beryllium, bismuth, bromine, cadmium, copper, gold, iron, lead, mercury, nickel, palladium, silver, tantalum, and zinc, and most of the energy that the phone and its mobile network depend on comes from the ground as well.

“One consequence of our growing appetite for technological wonders and our unease about the environmental and human costs of producing them has been a shift in the worst of the direct damage to places we can’t see—to parts of the world where labor and land are cheap, and where regulatory oversight is minimal.”

Then there’s fracking, where high-pressure water jets are shot into drilled shafts of rock to release trapped gas and oil. “Modern fracking technology is so effective and efficient that it has transformed global energy markets by creating gas and oil gluts that have pushed down prices by mind-boggling amounts” (increasing our dependency on fossil fuels over renewable energy).

The original fracking experiment took place in 1969 as part of “Project Rulison” (on a patch of land bordering my aunt’s Horse Ranch in Parachute, Colorado, coincidentally). Instead of water, the government decided it would be a good idea to set off a small scale nuclear bomb. The gas they freed was far less than anticipated, and most of it was rendered unusable by high levels of radioactive tritium.

Irrigation, and the problem with efficiency

In examining irrigation, we start to see some of the complexity of water rights, particularly on the Colorado, where Patricia Mulroy, senior fellow for environmental policy at UNLV describes the maze:

“We take water in Wyoming—outside the river’s watershed—and move it to Cheyenne. Come down to Colorado: we move it across the Continental Divide, from the West Slope to the Front Range, into the Kansas-Nebraska basin—outside the watershed of the Colorado. We move it across the Utah desert to the Wasatch Front, to Salt Lake, Provo, Orem, and all those agricultural districts—not in the Colorado watershed. In New Mexico, we move it to Albuquerque, which straddles the Rio Grande. In Arizona, we move it across 360 miles of desert, to Phoenix and Tucson and still more agricultural districts. And in California, we move it over hundreds of miles of aqueduct, from Lake Havasu to the coastal cities—not in the Colorado watershed.

“We may be citizens of a community, state, or country, but we are also citizens of a basin. It’s a web, and if you cut one strand the whole thing unravels. Either we all win, or we all lose.”

Simply becoming more efficient with water use doesn’t always translate to net positive environmental outcomes, as evidenced by David’s next stop on the Colorado - the wine-growing region of the Grand Valley, which is heavily dependent on irrigation.

Water for irrigation is counted as either consumed or non-consumed, which sounds straightforward, but the actual accounting for this is far more complex. For example, “not all of the farmer’s irrigation water goes into the grapes or evaporates. Some of it runs off the end of the field and channels back into the ditch system, from which it can be diverted again, by other farmers—and that excess is deemed “non-consumed.” Some non-consumed water makes it all the way back to the river, earning a “return-flow credit” for the irrigation district.” On average, river water in Colorado is reused 6 times before it leaves the state. Even irrigation water that soaks into the ground can be considered non-consumed because it maintains the water table and replenishes aquifers.

“Colorado high-mountain ranchers say that overwatering in the spring and summer isn’t wasteful because it leads to delayed subsurface return flows, through groundwater, that keep our rivers higher in the fall and winter than they would be otherwise. And in many cases they are right.” Requiring such ranchers to adopt more efficient irrigation techniques can have the perverse effect of increasing the proportion of water they consume, by enabling them to irrigate additional acres with water that used to be non-consumed.”

“Irrigation improvements can actually increase net water use by crops, by allowing either more intensive use of irrigation water on a given field (which raises both yields per acre and net water use per acre), or through extensive use of ‘saved’ water on nearby fields that were previously less irrigated.” Irrigation improvements can harm entire ecosystems, far beyond the cultivated area.”

Water that flows away from irrigated farmland sometimes sustains neighboring wetlands, and if farmers become more efficient the wetlands disappear. Increasing efficiency also does nothing to address over-allocation. “Typically, irrigators implement efficiency improvements because someone is paying them to do so.”

Non-consumption vs. consumption also becomes important in accounting for needs-based water use (e.g. flushing a toilet) vs. nonessential water use (washing a car). In a modern sewer system, flushing a toilet is mostly non-consumptive, because the water is captured, treated, and used again downstream, whereas washing a car, as far as a municipality is concerned, is almost entirely consumptive, the water disappearing the minute it hits your ride. The author proposes a potential solution: “Imagine a municipality with tiered water rates which make ordinary household use affordable, and then rise dramatically, to discourage people from casually doing things like washing cars.”

But what happens when are successful at cutting water use? Here again are unforeseen adverse circumstances.

“In 2015, Californians were remarkably successful at cutting domestic water use, but an unanticipated consequence was clogging, corrosion, intrusion, and other damage within waste systems which were not designed to function without big flows to keep things moving.” Similarly, reducing waste can create problems for downstream users, especially farms, who depend on the waste of municipal users. And the biggest problem, possibly:

“With water as with energy, we almost always reinvest our savings in additional consumption.”

Tribal Lands

The next stop along the tour is the dry, desert regions of Arizona, where groundwater lies deep, and access to it requires a lot of work (Lifting water out of the ground takes energy, and an acre-foot of water weighs almost 1400 tons). As such, residents often have to order their water from a commercial supplier, which is expensive. It is perhaps no coincidence that this problem is felt most acutely on tribal lands, a problem foreseen by John Wesley Powell in 1893 when he addressed Congress saying, “You are piling up a heritage of conflict and litigation over water rights, for there is not sufficient water to supply these lands.”

The irony for the tribes here, though is that

“tribes, collectively, could conceivably make priority claims to all the freshwater in America, since Indians were using streams long before the invention of the sluice box.

…the rationale lying in the only mention of Native Americans in the Colorado River Compact: “Nothing shall be construed as affecting the obligations of the United States of America to Indian Tribes.” It’s likely, for example, that nearly all the agricultural irrigation water in Arizona will end up under tribal control, since the tribes’ rights are senior even to California’s (sounds like a rife sequel to Killers of the Flower Moon.


Further downstream on the Colorado lie some of the U.S.’s largest dams, foremost of which is Hoover Dam: 700 feet tall, 660 feet thick at the base, and made of 7 million tons of concrete, on which lies the heavily touristed Lake Mead. Roughly sixty percent of the output goes to metropolitan L.A., 270 miles to the west, and one of its functions there is to power the pumps that move water from places that have it to places that don’t.”

The Hoover Dam is essential to California, but the problem is it’s drying up (you may have read about the bodies turning up in the drought-stricken Lake Mead). This decrease in water levels would not only cause downstream deficiencies, but would eventually lead to cavitation and destruction of the dam itself.

You’ll no doubt have heard the endless controversy over dams, essential for storing water for human consumption, but detrimental to countless other natural processes. An undammed river allows the uninhibited passage for wildlife, of course, but also, for example, allows free-flowing water to purge the river corridor and tear out the non-native, noxious weeds.

A key to protecting rivers from the construction of dams may be assigning legal rights to non-human entities, through rights such as “recreational in-channel diversions (RICDs)” which can essentially be thought of as water rights that belong to the water itself. But what to do with existing dams, especially those on a scale of the Hoover Dam? The problem lies not only in the number of people who depend on it but also in large-scale energy-generating processes.

“Taking [the Hoover Dam] or any other large hydroelectric facility entirely offline would represent a significant setback for non-carbon energy production in the United States—like winding back the clock on solar and wind—both because of the direct loss and because most of the replacement plants would inevitably be powered by fossil fuels…”

“Visitors to Hoover Dam often assume that the electricity produced by its power plant goes entirely to Las Vegas, but in fact, Nevada receives only about a quarter of the output (Nevada’s full entitlement to the Colorado equates to only about 45 inches, while California, Arizona, and Mexico use 54, 35, and 18 feet respectively).

Water Wars and Water Banking

California in particular, saw the water shortage coming in its early days. William Mulholland, responsible for much of Los Angeles’ early water infrastructure (and Parker Dam, and Lake Havasu), began arranging the purchase of land in the Owens Valley to construct an aqueduct to divert water, which met with revolt (and was popularized/fictionalized in the classic film Chinatown). Given the meteoric population growth in LA, he approved nearly the full cost of constructing Parker Dam (Lake Havasu) and built the Colorado River Aqueduct.

In response to seeing its water increasingly being spoken for, Arizona recognized it needed to develop its own diversions, forming a program called the Central Arizona Project (CAP). Many legal disputes between Arizona and Colorado followed, as California tried to kill CAP out of economic protectionism more than actual need - and while Congress ruled Arizona could proceed, California won a major concession - a provision stating that Arizona’s entire Colorado River water right was junior to California’s - “which theoretically means California could divert all of its 4.4 million acre-feet before Arizona takes any.”

Arizona wasn’t able to use all the water it was entitled to, but to keep California from taking it, Arizona began diverting it anyway, storing it underground by flooding it onto “spreading basins” in the desert - a practice known as “water banking” - advantageous in that, unlike in reservoirs, the water does not evaporate. Nevada and Arizona subsequently began paying Arizona for access to this banked water.

But banked water needs treatment to be rendered potable. A key issue in these desert regions is salt - which is not removed through wastewater treatment or water recycling, thus clogging air and water channels in the soil, reducing the ability of roots to absorb nutrients and promoting previously unfamiliar plant diseases. Because the land is dry, rain alone cannot leach the salt. Irrigation is needed, but the more one irrigates, the more the water table rises, leading to surface evaporation of water, which in turn leaves more salts - an inescapable cycle. “You know the salt flats in Utah? They’re devoid of plants for a reason.”

In examining desert states’ water use, people will point to frivolous uses such as maintaining verdant green Vegas Golf Courses and lawns, but here again are lie conundrums. Grass is not a huge consumer of water (one mature oak tree requires as much water as sixteen hundred square feet of Bermuda grass, for example). And “people will sometimes remove turfgrass but leave fifty-foot-tall trees behind, without realizing that the trees were totally dependent on the irrigation the turfgrass was receiving.” And the grass ameliorates a critical problem in Vegas: airborne dust, and the related issue of urban heat islands through ground absorption and re-radiation of sunlight (a better target for animosity might be Vegas Fountains and water features, which, because of evaporation, can require more water than fairways, and whose water, with nowhere to go, requires underground pumps to “dewater”, or recycle).

Clearly, resource management is needed - and here again, a paradox. “Successful conservation programs often depend on population growth — a force that pulls in the opposite direction.”

The mortgage crisis had reduced Vegas’s consumer demand for water—“a beneficial outcome, you would think—but it also halted new construction throughout the valley, and without new construction, the revenues from connection fees fell by ninety percent. Rapid population growth was both the cause of and the solution to Las Vegas’s water challenges.”

The Nation’s Breadbasket (and Driest Desert)

“The Imperial Valley in California is the largest single user of Colorado River water. It’s also one of the most productive agricultural areas in the US. If you eat fresh fruits and vegetables during the winter, you eat produce from the Imperial Valley,” despite an average annual rainfall of only ~3 inches.

“Growing food in a desert may seem nutty, but there are many advantages.” Frost, hail, and damaging rainstorms are far less common. Jobs are year-round, and precise crop planning is possible.

Of course, there are drawbacks - summers so hot the rain can scald plants as it falls, in part because the heat drives oxygen from the falling water. Salt,again, is a problem - to keep it from accumulating, farmers have to water it enough to push salt below the root zone of plants and into the subterranean drainage, flushing it out to the Salton Sea. But as irrigation practices become more efficient, that becomes harder to do.

The Salton Sea itself is a fascinating debacle - a man-made lake (California’s largest) that once saw countless tourists and has become a crucial stop for migratory birds on the Pacific Flyway (60% of birds that breed in North America stop there) but is now dwindling to a giant, steaming, algae-infested hazard, brought on by 125-degree temperatures, industrial fertilizer disposal, and flash flooding. Its poisonous waters have been the cause of countless natural casualties, including the death in 1991 of 150,000 grebes wintering in the area, in which were found “worrisome quantities of selenium, arsenic, cadmium, chromium, zinc, and DDT.”

The area poses risks to humans as well, including the 650,000 people who live in the Salton Sea’s ‘air shed’ where, as the lake recedes, the substances left behind evaporate. The area has seen record numbers of respiratory illnesses, and the towns closest to the lake have 3 times the rate of childhood asthma than California as a whole. But where would the agricultural runoff go if not for the Salton Sea?

Michael Cohen, of the Pacific Institute’s 2014 report on the Salton Sea, explains the issue: “One of the challenges with the Salton Sea—which I think is emblematic of a lot of the water problems in the West—is that any effort to protect it is an effort to stop natural processes, and that requires a project with inputs in perpetuity.”

“Environmentally problematic measures that sound green (favoring locally grown food, switching to electric cars) are vastly easier to sell than environmentally useful measures that sound like trouble (dumping agricultural runoff into a reeking man-made desert lake).”

For example, the least significant (and most expensive) salt-reduction measure was to construct a desalination facility, which, after many years of planning and construction, was operated only twice and then damaged by flooding shortly after.

The current proposal for the Salton Sea is to capture some of the water in shallow dikes and spread it out. “It’s not sexy, it doesn’t generate a lot of recreational revenue or economic development, but as long as there’s agriculture in the valley, the farmers will need some way to leach their soils and discharge the water. If you capture that water and spread it out, you can reduce dust emissions and generate a lot of ecosystem benefits.”.


The Bureau of Reclamation was established in 1902 at the behest of President Roosevelt, who “wanted the government to make rivers useful by regulating, capturing, and diverting their flow.” Under this definition of reclamation, there was a nationwide fervor to take up lands along which water could be diverted. “There was a quasi-theological element in the notion of reclaiming land, of taking it back, as though the western deserts were in a fallen state and could be returned to their rightful place in Creation through determined intervention.” It’s perhaps no surprise then that the rights of Mexico, the final stop along the Colorado River, weren’t even considered until 1944, when a treaty was signed that set Mexico’s allotment at 1.5 million acre-feet of excess flows or, barring that, equal contributions from upper and lower basin states.

To prevent this, the Bureau of Reclamation created Senator Wash Reservoir, an “off-stream retention reservoir” two miles upstream from Imperial Dam, which could keep canceled water orders and unexpected inflows from Mexico by pumping them uphill from Imperial Reservoir. This practice is considered “saving” water by US agencies (as opposed to water that is “lost” by Mexico).

The more literal definition of water saving the bureau encounters is in fixing leaks, such as the ones occurring in the “All-American Canal”, a 30-mile trench in the desert which, when originally constructed, used to leech tens of thousands of acre-feet of water into the ground each year, causing floods in Mexico until it was lined in 2010. However farmers came to rely on the influx, which was useful for irrigation and reduction of salt concentration. “Eliminating leakage didn’t reduce water use, it transferred an existing resource from water users and ecosystems in Mexico to city dwellers in Southern California.

What’s to be done?

The multitude of issues uncovered in examining the Colorado as it moves from source to mouth makes the conversation around the river’s “Highest and Best Use” incredibly complex. So what to do? The author identifies several key areas to examine.


Finding flaws in how the western states have handled water isn’t hard. “No system whose central purposes include allocating nonexistent water can truly be said to work.” Effective long-term programs must address over-allocation, perhaps through concepts like “water footprint”, the idea that, for example, when you sell a crop, you’re also selling the water used to grow it.

Robert Glennon, in ‘Water Follies’, writes.

Such an allocation system creates tremendous inefficiencies. It ignores the economic value of the activity, treating higher and lower-value uses alike. It creates incentives to hoard resources, promotes financial speculation, and thus turns a shared common resource into private property.

“Over-allocation can only be addressed from the top—beginning with a scientifically defensible determination of how much water is likely to be available for human exploitation.” According to Denver Water’s CEO, “You can only do two things: augment the supply of water from an outside source or reduce demand.”

And of course, demand is spiking and time is ticking. “Current estimates of climate-related reductions in the Colorado’s annual flow range from 10-30% by mid-century, making existing allocation systems useless. The federal government is inextricable here - “almost all major water infrastructure in the West was financed by the federal government, and in times of shortage, a variety of federally mandated emergency reductions kick in. As such, states are incentivized to work together to forestall any shortage declarations that would force water reduction.

Water Diversion

The Colorado alone cannot continue to provide enough water to meet the growing needs of the western states that currently rely on it. One idea is to divert water from other river systems—a sort of ‘water wheel’ from state to state, starting perhaps in the Great Lakes region, down through the Missouri, into Kansas, Colorado, and so on. But such a system would require an enormous amount of cooperation from entities with their own water problems.


Desalination places saline water under enormous amounts of pressure to separate the dissolved salt. But the process is expensive and consumes a lot of energy, and the equipment is complex. The biggest obstacle is that you must ensure the water is clean before removing the salt. It’s certainly possible - cruise ships desalinate the water used aboard all the time - the Royal Caribbean’s Oasis of the Seas, for example, produces 800,000 liters of fresh water a day through its two desal units for its ~6,000 passengers. But to extend this sort of solution to an increasing global workload would put enormous strain on our already dire energy capacity. Another potential result of desalination might be increasing development in regions where that might otherwise never occur, such as Dubai.

Cloud Seeding

Cloud seeding, a process of shooting silver iodide into the clouds to stimulate the formation of ice crystals that fall as snow, is already a common practice in many western states. It is a bit of a shell game, in that rain that falls in one place can’t fall in another. But, “since one of the effects of climate change is a catastrophic increase in precipitation in places that already have more than they need - maybe everything might balance out.”


There are countless ways to make agriculture more efficient, from the simple but laborious (level the ground), to the traditional but economically disruptive (fallowing).

“How far to go in protecting farmers, and which ones, is among the more complicated western water issues.”

Water Footprint

People argue irrigation shouldn’t be wasted on “low-value” crops. But “high-value” crops are often thirsty, like the decorative plants people put in their irrigated yards. One of the unintended consequences of cutting back on irrigation is that it opens up former agricultural land for urban sprawl and other “high-value” uses. “A new subdivision on the outskirts of Phoenix will almost always use less water than any irrigated farm it replaces. But water shouldn’t be the only consideration.”

Growth Management

The American cities with the most significant increases in population in recent years have been concentrated in some of the most water-threatened parts of the country. 3 of the top 10 rely on the Colorado: LA, San Diego, and Phoenix. But from a broader environmental point of view, these aren’t necessarily terrible places to live. They bring the efficiencies of population density, year-round solar exposure (cooling requires much less energy than heating, and solar power is much more readily available) - “water scarcity can even be a useful tool for containing the heedless sprawl of human habitation.”

Outdoor Recreation

A long-standing tenet (and one I believe in), is that humans must experience nature in order to value and protect it. But wild landscapes, argues Owen, “are less often ruined by people who despise them, than by people who love them. The rise of national parks closely parallels the rise of the automobiles that made them accessible.” Today people recreate outdoors not just in vehicles, but in massive RVs running generators and pulling trailers loaded with gas-powered recreational equipment.

How do we balance our impact on natural areas with the advocacy and support that results in its protection and betterment? The decisions we make in managing the Colorado will become a model that much of the world will soon be forced to examine when dealing with their own dwindling supply.

Water connects us all. There can be no winners and losers if we want to preserve it for future generations.