This is the second part of a multi-part series about the impact of hydroelectric dams. Part one can be found here.
Imagine this idyllic scene: a sprawling lake nestles amid rolling forest-topped hills, projecting beauty and tranquillity. But sometimes the smothered reality rises to the surface. Disquieting the view, the top of a tree may protrude where pleasure boats are gliding. In a drought, a church steeple holds a finger above the water to register its objection. For a moment at least, we are forced to remember what we have been burying beneath our supposed improvements. Across the world, where large dams have been built, we can count the cost not only in what happens downriver of the obstruction (see “Voyage of the Dammed”). We can also follow the consequences behind the dam and see how the disruption that is nurtured there spells yet further devastation beyond it.
Every way in which dams damage rivers and their valleys is deadly for humankind because we rely on rivers, more than we realize, to sustain our own lives. It is a harmfully false dichotomy to imagine that what denudes and poisons watersheds, aquifers, riverbanks, riverbeds, estuaries, floodplains, and deltas harms “nature” but not us. We rely for our own continued life on what we are eradicating. However, there are certain kinds of consequences which immediately and sharply make this fact plain. Giant hydroelectric dams demonstrate their perverting and destructive essence as they visit cataclysmic disasters on people, first by drowning and displacing historic communities, and then by exacerbating flooding.
The Flood Every Dam Makes
With each great dam comes a deluge. Behind the towering edifice that awes disciples of gigantic infrastructure lies the proportionately vast reservoir it has necessitated. Every piece of river engineering can only redirect, not eliminate, the water. And, even if we were to believe all the hype of a dam’s benefits, the land flooded by the restrained river is a heavy price rarely considered. “Construction of the Tucurui hydropower station in the rain forest of the Amazon River basin,” Paul Josephson relates in Industrialized Nature, “inundated nearly 3,000 square kilometres (more than 740,000 acres) of land, [and] destroyed fragile ecosystems.” The rightly infamous Three Gorges Dam on China’s Yangtze River has, in the end, required a 410-mile-long reservoir. The dams on Russia’s Volga River—one of the most engineered rivers in the world—have cumulatively buried over 7,500 square miles of land.
It is typical of the flippancy with which dams have been promoted that such inundations have often been sold to the wider public as mere recreation areas. But rapidly adding millions of tons of water to a landscape has consequences beyond the scope of dam boosters’ intentions and interests. First, landslides often widen the devastation of a new dam reservoir, sometimes even crumbling areas where displaced people have been moved due to the dam. As with many of a dam reservoir’s affects, it is both the volume of new water and the subsequent swift and dramatic rise and fall of the water level in accordance with the dam’s operation that causes this structural instability. As Mara Hvistendahl describes in Scientific American, after the reservoir of the Three Gorges Dam was first filled up to the level of 445 feet (130 feet below its intended depth), “early warnings came true just a month later, when around 700 million cubic feet (20 million cubic meters) of rock slid into the Qinggan River, just two miles (three kilometres) from where it flows into the Yangtze, spawning 65-foot (20-meter) waves that claimed the lives of 14 people.”
The Chinese government had long denied baleful predictions during the decades of work constructing the dam, but in 2007, “Chinese officials staged a sudden about-face,” Hvistendahl continues, “acknowledging for the first time that the massive hydroelectric dam…may be triggering landslides, altering entire ecosystems and causing other serious environmental problems—and, by extension, endangering the millions who live in its shadow.” Yet it should not have taken the beginning of landslides to convince anyone of their likelihood. The deleterious effects of dam reservoirs have been observed for decades. In 1963, the worst landslide to date occurred behind the huge Vajont Dam in the Italian Dolomites. When the reservoir had been initially created and filled three years before, instability had been noted on Monte Toc, which towered above the new lake. Then one October evening, after a period of heavy rain, over nine billion cubic feet of rock broke from the mountain, crashing all at once into the reservoir, and producing a massive wave which surged 460 feet over the dam, drowning the town of Longarone below it and killing nearly two thousand of its residents.
Part of the deadly geological disturbance caused by the Vajont dam reservoir was an increase in seismic activity, and the evidence over many years of a correlation between large dam reservoirs and seismic activity is hard to deny. Indian seismologists Harsh Gupta and BK Rastogi meticulously compiled and analyzed this evidence in their 1976 book Dams and Earthquakes. Regarding the Vajont case, they noted that between the opening of the reservoir and the catastrophe, 250 tremors were recorded. “The three significant uprises in the Vajont reservoir [were] followed by three bursts of seismic activity…. Following every peak, the decrease in water level [was] followed by decreased seismic activity.” Furthermore, they added, “the maximum activity [in] September 1963 was followed by the disastrous landslide of October 9.” Similarly, at the Monteynard Dam on France’s Drac River, tremors were recorded a few days after the reservoir was first filled on April 15, 1963, culminating in 4.9 magnitude earthquake ten days later. As with Vajont, seismic activity rose and fell along with the depth of the reservoir.
The correlation between the filling of dam reservoirs and seismic activity is both present and urgent because large dams are frequently built upon fault lines, including the Three Gorges Dam. As Sichuan-based geologist, Fan Xiao explains in Hvistendahl’s article, “When you alter the fault line’s mechanical state,” with the addition of millions of tons of water which seeps into the rock, “it can cause fault activity to intensify and induce earthquakes.” In an article for the journal Scientific Reports, geologists Marco Scuderi and Cristiano Collettini further elucidate. “Fluid overpressure is one of the primary mechanisms for tectonic fault slip, because fluids lubricate the fault and fluid pressure reduces the effective normal stress that holds the fault in place.” In other words, there is an inverse relationship between fluid pressure and stabilizing geological friction—as the former rises, the latter dangerously decreases. In this way, even dormant fault lines have been reactivated, including the one beneath the Hoover Dam’s Lake Mead.
But there is more than rock beneath the reservoirs created by large dams. Entire habitats and regions are lost to these inundations, and with hydroelectric dams’ notoriously ballooning costs, little to no thought is given for what happens to the life that is drowned. Yet we live in a creation where all actions have consequences, whether they fit our priorities or not. The Balbina dam in northwest Brazil created a huge but shallow reservoir when it inundated between 1,200 and 1,700 square miles of forest in the late 80s. As one might expect, one result was the devastation of the animals living in the forest. Researcher Carlos Peres of the University of East Anglia’s School of Environmental Sciences, who documented this destruction in a 2015 study, told National Geographic “We’re watching extinction unfold right in front of us.”
Meanwhile, the reservoir also utterly fails to produce the promised “pollution-free electricity for the capital of the Amazon,” as Fred Pearce records in When the Rivers Run Dry, referring to the city of Manaus. As the vegetation of the forest rots in the new expanse of water, the greenhouse effect of the methane released is equal to a coal plant providing the same energy as the dam—times eight. Pearce also relates the conclusion of Vincent St. Louis of the University of Alberta that “reservoirs produce a fifth of all manmade methane in the atmosphere.” Surely, you might think, all vegetation decomposes eventually. This is true, but, outside of a reservoir, “the decomposition would most likely occur in a well-oxygenated river, producing carbon dioxide—whereas tropical reservoirs usually contain little oxygen, and as a result they generate methane instead,” Pearce further explains. “Methane is twenty times more potent as a greenhouse gas than carbon dioxide. Reservoirs thus change the way significant amounts of the earth’s vegetation rots, and with it dramatically raise the greenhouse effect of the rotting.” Furthermore, John Harrison of Washington State University argues, “there is strong and growing evidence that temperate reservoirs can produce methane at rates comparable to those reported from tropical reservoirs.”
Homes beneath the Reservoir
Beneath numerous dam reservoirs lie the remains of extinguished human societies. The 2000 World Commission on Dams report estimated that, in the previous fifty years, 40 to 80 million people had been displaced by dams. Millions more have suffered a similar fate since. There is ultimately no adequate restitution for the destruction of one’s homeland, but many have received no recompense at all. In 2007, China’s Prime Minister Wen Jiabao admitted that at least 23 million people had lost their homes to dams in that nation alone.
The people sacrificed to dams are ignored in official presentations of dam construction, as if they never lived in the places from which they have been removed; once condemned to a transient life in temporary settlements, they are often outside the reach of statistics or government assistance. “The millions of displaced people don’t exist anymore,” wrote Booker Prize-winning author, Arundhati Roy, regarding the many displaced by India’s hydroelectric dams. Even as former property owners are inadequately assisted, many of the uprooted are legally landless and are left to fend for themselves without the farms, folkways, and communities on which they relied. “Once they start rolling, there’s no resting place. The great majority is eventually absorbed into slums on the periphery of our great cities, where it coalesces into an immense pool of cheap construction labour (that builds more projects that displace more people).” For example, thousands of Indians living on fertile agricultural lands in Jabalpur were evicted in the 1980s so that their villages and fields could be covered by the reservoir of the new Bargi dam. When the reservoir was first filled, it even inundated areas where some had been resettled. “People were flushed out like rats from the land they had lived on for centuries,” Roy describes. “They salvaged what they could, and watched their houses being washed away.” Incredibly, some of those displaced are due to be forcibly removed again to make way for new nuclear power plants, which will also voraciously feed on the burdened waters of the Narmada River. Meanwhile, the whole sordid process starts over again upriver, with the massive newly inaugurated Sardar Sarovar Dam and the 200,000 forced from their homes for its benefit.
According to International Rivers, the Three Gorges Dam forced out 1.2 million people, as 13 cities, 140 towns, and 1,350 villages were engulfed by the waters of the dam’s reservoir. Hundreds of thousands more have to move due to the instability of the land around the dam. The toll goes on and on, as Pearce notes: “The Akosombo on the Volta in Ghana expelled 80,000 people, the Aswan High Dam in Egypt 120,000…the Kariba in southern Africa almost 60,000, the Tarbela in Pakistan more than 90,000.” Bountiful naturally irrigated farmland is often snatched away in exchange for less and worse territory and the promise of irrigation from the dam that either never arrives or brings salinization and water-logging. But while China, India, and Brazil have become centers of large-dam-building fever as it has spread across the planet, it was the United States and Soviet Union who pioneered the hubristic dreams and destructive reality at its core.
Once Stalin’s Soviet regime began choking the Volga with hydroelectric dams, over 500,000 Russians had to make way for the reservoirs, including the 130,000 who lived in and around the towns of Mologa and Kalyazin, buried under the Uglich reservoir in 1941. Today, bearing witness to the people who not only lost their homes, but were unable to speak of it for half a century, the belfry of Kalyazin’s former St. Nicholas monastery stands incongruously amid the man-made lake, like a tree growing from a smokestack. At the same time, in the United States, dam-building focused on the West, invariably displacing native tribes.
In the 1950s and 1960s, “Indian lands became sacrifice areas in massive flood-control and hydroelectric projects,” asserts Laurence Hauptman, one of the foremost experts on American Indian tribal history and dispossession. The US Army Corps of Engineers and the Department of the Interior’s Bureau of Reclamation worked at a fearsome pace and great expense to re-engineer the vast region, picking winners and losers as they went. The “Big Bend, Bonneville, Boyson, Cochiti, Dalles, Fort Peck, Fort Randall, Garrison, Glen Canyon, Grand Coulee, Oahe, Roosevelt, and Yellowtail Dams,” Hauptman explains, were all “devastating to Native peoples.” For example, the Garrison dam flooded 150,000 acres of the Three Affiliated Tribes’ (Arikara, Hidatsa, and Mandan) reservation in North Dakota, and “permanently shattered the lives” of this community, while the Big Bend, Fort Randall, and Oahe dams inundated 200,000 acres of Sioux territory in the Dakotas. One of the starkest examples, however, occurred in the northeast.
Many Americans are used to the idea that their government frequently ignored treaties made with native tribes, but may assume it to be a 19th century phenomenon. Big dams continued this sad practice into the 20th. One of the oldest treaties made by the young republic was the 1794 Canandaigua Treaty with the Six Nations of the Iroquois. Article III of the treaty delineated the lands of the Seneca within the western portions of New York State and Pennsylvania, acknowledging that “all the land within the aforementioned boundaries, to be the property of the Seneca nation; and the United States will never claim the same, nor disturb that Seneca nation…in the free use and enjoyment thereof: but it shall remain theirs, until they choose to sell the same.”
The Snake River Valley floods after the collapse of the Teton Dam near Rexburg, Idaho in June, 1976. Courtesy of WaterArchives.org, CC BY-SA 2.0.
This promise was tested when the Flood Control Act of 1936 authorized, among other things, the construction of a series of dams on the Allegheny River north of Pittsburgh after a serious flood. Eventually the chain of dams reached its most northerly extent with the Kinzua Dam, the plans for which included the flooding of 10,000 acres of Seneca land. In 1957, the first funds for the project were congressionally appropriated, which, the Supreme Court would later rule, in and of itself severed the treaty and left the government legally free to proceed in expropriating one-fifth of Seneca territory. Even though Article VI of the US Constitution states that treaties “shall be the supreme Law of the Land,” a minor initial appropriation, not a specific abrogation, was deemed sufficient to break the nation’s bond. No purchase was made; the land was just seized (although $15 million compensation was later paid), more than 800 people were evicted from their ancestral land (according to the Seneca), and the dam was built—land, homes, school, and forty-two cemeteries disappeared beneath the reservoir. “While a private corporation and the federal government profited financially by their partnership,” Hauptman concludes, “the Allegany Senecas’ old way of life—their fishery, their areas for hunting and gathering medicinal plants, their holy lands and sacred sites—were all lost forever.”
The Floods Dams Make Worse
Even if we open our eyes to the travesties visited upon peoples and places by dam reservoirs, can we at least rest assured that a positive service is, overall, being rendered? Most hydroelectric dams are built with the promise of flood control. But even as dams keep the rivers below from the regular flooding that sustained the life of its floodplain, they also worsen the severity, frequency, and destruction of irregular floods. Large dams make rivers narrower, deeper, and straighter than they would naturally be and denude them of life (see “Voyage of the Dammed”). Exacerbating this consequence is that all of the other various forms of river engineering—particularly dredging, channelization, and levees—produce the same effect.
Worse still, the practices and logic that have driven severe river engineering since the 19th century tend to make further interventions more likely and encourage patterns that increase the human cost. “Over the past century, we have straightened, channelized, dammed, and rip-rapped the Mississippi, all in an effort to keep it in its channel and away from its natural floodplain. Ever higher levees and stronger floodwalls—supplemented by federal insurance and disaster relief—lure more and more people into the floodplain, which in turn requires more extreme measures to protect those people,” summarize Christine Klein and Sandra Zellmer in their study, Mississippi River Tragedies: A Century of Unnatural Disaster. “Meanwhile, we have cut off the river’s natural pressure-relief valve—its floodplain—leaving it no good place to go when the rains come. Through these efforts, humans have demonstrated an uncanny ability to exacerbate the damage caused by natural hazards.”
The problem is not only that we view rivers as instruments in our hands, but that we assume we can, in Klein and Zellmer’s words, make “make the river predictable and unchanging,” in accordance with our new uses and intentions. On the Mississippi, the upper section from its Minnesota source to Cairo, Illinois, was dotted with locks, while the lower section from Cairo to the Gulf was heavily leveed, along with dredging along the whole stretch. While commercial shipping would be greatly enabled, hubs of industry would sprout and be made safe for growth. Such was the effect of this 19th century domestication that when Mark Twain returned to the lower Mississippi in 1882 after a long absence, he reflected, “The river is so thoroughly changed that I can’t bring it to mind even when the changes have been pointed out to me. It is like a man pointing out to me a place in the sky where a cloud had been.” This was more than the tricks or laments of nostalgia, Klein and Zellmer add: “He could see that the Mississippi was rapidly becoming one of the most heavily modified river systems in the world.”
The craze for giant dams beginning in the 1930s fit easily within the prevailing attitude and assumptions established in the 19th century—instrumentally modified nature would be compliant. But nothing could be further from the truth, and, as is consistently the case in our technological age, we tampered with what we had decided to no longer understand. “The main flooding problem,” asserts hydro-geologist Bob Criss of Washington University in St. Louis in an article for Scientific American, “is that we don’t get along properly with our rivers.” It ought to be elemental—what happens when you constrain and constrict something, the force of which you did not create and cannot extinguish? You delay and ultimately increase the expression of that force when it eventually finds release. So it is that the narrowing of rivers produced by dams, dredging, and levees ensures that the natural flooding which would occur over hundreds of miles—as rivers gradually slop over their channels and widen—is stored up and directed toward a point where it can, quickened and heightened, escape. “What might have been slow-spreading floodwaters when they were unconstrained turn into neighborhood-destroying mini-tsunamis when they burst all at once from behind failing levees,” comments Adam Rogers in his Wired piece, “Too Much Engineering Has Made Mississippi River Floods Rise.” This was a consequence that was foreseen. In 1852, engineer Charles Ellet Jr. warned the federal government that narrowing the Mississippi would, inevitably, cause the river to “rise higher and flow faster.”
Year after year, this continues to occur, but the reaction has consistently been to compound the “flood-control” measures that are at fault. No mistakes are admitted, no courses reversed. Writing of levees in India, Himanshu Thakkar—coordinator of the South Asia Network on Dams, Rivers and People—explains that “embankments are basically flood transfer mechanisms: they quickly transfer the floods from a given area to downstream areas. The floods resulting when embankments are breached are very different than a natural flood. Embankment floods are sudden, have greater destructive power….” In the United States, the same interests that built the dams and levees that narrowed the rivers, build more dams and levees to meet the crisis they created. “This is the problem with the Army Corps [of Engineers],” comments Criss. “It’s like a protection racket. They just squeeze the river, make more floods, and then say, ‘Oh, let us help you, you need more help, the floods are worse.’”
That the flood-control of dams and other river engineering has heightened floods is not mere conjecture. For example, in 2018, geoscientist Samuel Muñoz and his colleagues released a study on the Mississippi, concluding that “the artificial channelization (confinement to a straightened channel) has greatly amplified flood magnitudes over the past century.” As Emma Marris summarizes the research in Nature, “Now when the river overtops its banks, the flood is faster, bigger and more powerful than it would be without human intervention, the researchers found. In essence, the same engineering works that prevent small floods make big floods worse.” The findings about causation are important because we have come to lazily attribute all disastrous changes to climatic warming. While this has had some impact on the flooding Muñoz analyzed, Rogers reports in Wired that “75 percent of that [increased] risk comes from human engineering of the Mississippi for navigation and flood control.” Likewise, Criss, who published research in 2015 indicating that the Corps of Engineers was now significantly underestimating the extent of future flooding in the American Midwest, insists that climate change “is not our hugest problem,” in relation to such greater floods, “and that does not explain this. This river constriction is the No. 1 cause.” Across the world, where many billions of dollars have been spent on flood control, flooding is ever more severe.
A Sham of Stability
Despite all this downriver disruption, surely the dam itself, towering and impressive, is a model of control and stability? Instead, it is in the very nature of a hydroelectric dam to store up its own negation and expose its surroundings to even more cataclysmic risk than we have already considered. The height of a dam is partly based on projections of how high a once-in-a-century flood would be. But this reasoning is counteracted by the other main purpose of dams—efficient power generation—which requires a dam reservoir to be as full as possible. This means that dam operators will often not lower the level of reservoirs until it seems unavoidable. Consequently, by the time dams eventually release water from the reservoir, they do so in ways that create sudden serious floods. Thakkar describes what happened at the Ukai dam in India’s Gujarat region, when the dam filled up near the conclusion of monsoon season. “By the evening of August 8 [2006], the dam was releasing over twice the amount of water that the river downstream could carry.” As the New York Times reported, “With water brimming well past the permitted levels at the 350-foot Ukai Dam…the engineers apparently threw open the reservoir’s 21 sluice gates. Water then did what water does. It surged downriver, swallowing this city of three million people like a hungry beast.” Mercifully, there were “only” 120 deaths, but the damage was extraordinary.
This, of course, is not even the result of a dam failure. But dams do fail. In China alone, this has happened hundreds of times, most notably with the 1975 collapse of the Banqiao dam in Henan province. Its operators, Pearce relates, thought “that the dam could survive a once-in-a-thousand-years flood on the river,” and, therefore remained confident during a day that brought a torrent of rain. But when a dam upstream burst, the massive Banqiao followed suit, unleashing at least 21 billion cubic feet of water in a seven-mile-wide wave traveling 30 miles an hour into the valley. In a cascade of dam failures, sixty-two more dams crumbled downriver. It was a mind-boggling disaster in which at least 170,000 people died. The cascade of dam failures that night was not unique, since dams are often concentrated in particular valleys and regions. Pearce also tells of the situation at the turn of the millennium in Mozambique, when “dams upstream in South Africa, Botswana, and Zimbabwe…kept their reservoirs almost full at the start of the rainy season. When the floodwaters came down the rivers, dams that promised to provide flood protection became a lethal liability. Dozens of dams were washed away.”
The solution is not to simply build better dams, because a large dam is inherently endangered by its own functioning. As noted in “Voyage of the Dammed,” dams trap within their reservoirs all the silt and sediment that would otherwise travel on with the river. With nowhere to go, sediment accumulates, gradually reducing the capacity, effectiveness, and lifespan of a dam. Indeed, siltation was a major impediment for both the Ukai and Banqiao dams discussed above. But this is also a slowly multiplying problem that contradicts the assumptions about dam longevity, and will bring increasing instability, malfunctioning, and failure for the dams that have throttled our rivers. The Reservoir Sedimentation Handbook warned that while the capacity of China’s dam reservoirs was being reduced by 2.3% per year due to siltation. According to the paper “Accumulation of Sediment in Reservoirs” by T. Sumi and T. Hirose, “sedimentation problems are growing as today’s inventory of reservoirs ages, and severe sediment problems are starting to be experienced at sites worldwide, including major projects of national importance.”
The Myth of Efficiency
That which is destroyed by large dams is more valuable than quantification can capture. But, even if we were to accept another key rationale behind hydroelectric dams at face value—that they efficiently produce electricity—the reality falls far short of the boasts. “Those who promoted dam projects were not honest about costs and benefits,” admitted Daniel Beard, former commissioner of the Bureau of Reclamation, one of the primary forces behind US dam construction. Thayer Scudder is an American anthropologist who had long believed in the benefits of dams, consulting with the World Bank as they funded giant projects around the world. But, as he has just revealed in a new book, Large Dams: Long Term Impacts on Riverine Communities and Free Flowing Rivers, he can no longer maintain that position. “What I learned was that important short- and medium-term benefits of large dams tend to be followed by major and unacceptable longer-term economic, environmental, and social costs.” A 2014 study of 245 large dams by Oxford University academics concludes that, “even before accounting for negative impacts on human society and environment, the actual construction costs of large dams are too high to yield a positive return.” In an analysis of the study for Yale’s environmental magazine, author Jacques Leslie reports that “on average dams’ actual costs were 96 percent higher than their estimated costs, and the average project took 44 percent longer to build than predicted.”
As dams prove increasingly more fragile than expected, the long-term costs mount further. In 2017, the Oroville Dam in California came close to failure, sending 180,000 from their homes, and leading to $1.1 billion of restoration on the faulty spillways. Hydroelectric dams also routinely and significantly under-perform estimates in the actual production of electricity. As sharply diminishing power production from dams in India has illustrated, performance is likely to decrease over time, thanks to siltation. Hydroelectric dams simply do not bring the productivity and economic regeneration that supposedly induced the World Bank to invest $75 billion in hydroelectric dam-building during the peak of its popularity. Incredibly, however, gigantic new dams are still being planned and built worldwide. Despite a brief hiatus following the criticisms of the World Commission on Dams report, the World Bank is back in the game, while China has become the world’s primary source of loans to governments for great dams—in 2017 alone, Chinese banks provided $9.2 billion in such capital.
While long-term problems build up, the short-term cycle of vain promises and effusive spending continues. The modern economics of destruction—cutting across political systems, continents, and creeds—remains as trendy and hard to halt as it was when the Cold War’s opponents strove to match each other in the volume of concrete clogging up their waterways. As ever, vast monuments to artificial might cannot propel us away from the need to live within the realities of creation.