When water chokes you, what are you to drink to wash it down?
—Aristotle
Throughout history, people have built communities near reliable drinking supplies and created vast hydraulic systems—soaring dams, astonishingly long pipelines, and massive canals—to collect, purify, and distribute drinking water. So successful were these waterworks that by the twentieth century humans began to take water for granted and forgot about its destructive nature. People built in areas previously considered out-of-bounds—deserts, floodplains, filled-in wetlands, and low-lying coastal zones. But conditions are changing in the twenty-first century, and now communities around the world are increasingly vulnerable to rising seas and “extreme weather events,” such as flash floods.
While many experts are racing to find ways to avoid our running out of water in coming years, others are deeply concerned about the opposite problem: too much water, of the wrong kind, at the wrong time. Most countries have systems to collect, purify, and distribute water, but few, including the United States, have sufficient infrastructure to keep water at bay.
Floods are a natural occurrence that can be highly beneficial to the ecosystem—by providing nutrients that fertilize soil, cleaning silt from channels, removing invasive species, maintaining biodiversity, recharging ground-water supplies, and providing hydration to arid regions. But floods can also be devastating, tearing boulders and trees from embankments, scouring out old channels and creating new ones, and setting off mudslides. In many cases, human interventions—straightened rivers, destroyed marshes, new canals, old levees, cities built in floodplains—set the stage for catastrophic flooding.
As the world urbanizes, naturally absorbent wetlands, fields, and woods are converted to “hard-scape” paved roads and concrete buildings, which amplify the effects of storm-water runoff. According to the USGS, hard surfaces in urban areas increase storm-water runoff from two to six times over what occurs on unpaved ground. Ill-conceived or poorly built flood-control infrastructure worsens the impact of overflowing rivers. Floodwalls designed to restrain water in one section of a river can push dangerously high water downstream. Poorly built levees, such as those that ringed New Orleans during Hurricane Katrina, can breach and allow the gushing floodwaters to sweep away people and their homes, destroy roads and bridges, and submerge cities.
After high waters recede, flooded communities suffer from contaminated drinking supplies, waterborne disease, ruined crops, washed-out roads, and long-term economic hardship.
In the United States, seven of the nation’s ten costliest disasters have been caused by floods. Although death by flooding has declined since the 1950s, thanks to improved weather forecasting and early-warning systems, the USGS estimates that flooding kills 140 Americans a year, on average, and causes $6 billion worth of property damage.
In Britain, where floods have been labeled “a twenty-first-century catastrophe,” unusually hard, prolonged rainstorms between 2000 and 2007 caused over 2 billion pounds’ worth of damage and left more than a third of a million people without drinking water, and thousands homeless. A single violent stormburst, in July 2007, produced more than a month’s worth of rain in one hour; towns were swamped, public transportation was paralyzed, and Prime Minister Gordon Brown declared an emergency. The 2007 flood, according to the nation’s Environment Agency, eclipsed the UK’s previous benchmark, the floods of 1947, which inundated seven hundred thousand acres and were the worst in two hundred years. “We have not seen flooding of this magnitude before,” an agency statement read. While experts were leery of attributing the disaster to global warming, the “extreme rainfall event” of July 2007 was consistent with forecasts of what climate change will bring.
Similar statements were made in Australia three years later. After a decade of drought, torrential rains arrived in Queensland in late 2010, creating record-breaking floods that killed dozens and caused millions of dollars’ worth of damage. Scientists blamed La Niña weather patterns rather than climate change, but said it was “indisputable” that global warming was leading to more severe storms. “People have to accept that the game’s changed,” said Chris Cocklin, an environmental scientist at James Cook University.
But as weather patterns change, communities with poor flood defenses will find themselves increasingly vulnerable. This has already been hinted at by storms in usually temperate regions of the US, such as the Northeast and Southeast.
In March 2010, two immense storms soaked Rhode Island and Massachusetts with record-breaking precipitation. The second storm dumped almost nine inches of rain on Providence, Rhode Island, and the Pawtuxet River crested at 20.79 feet, nearly 12 feet above flood stage, causing rampant flooding. More than four thousand homes lost power; over a hundred people were evacuated to shelters; parts of I-95, the major East Coast artery, were flooded; Amtrak service was suspended; sewage treatment plants failed; and President Obama declared a state of emergency. New Englanders were bewildered. “Nobody was prepared,” said a college student. “When it comes to man versus nature, nature wins every time.”
As in Australia, years of water scarcity in Atlanta, Georgia, masked the threat of suddenly having too much.
With over 5.4 million residents, the Atlanta metropolitan area was the fastest-growing, most populous region in the Southeast in 2007, and the city promoted itself as being “the economic engine of the South.” But as Atlanta grew rapidly—starting in the mid-seventies, with the city’s rise accelerated by its hosting of the 1996 Olympic Games, which led to a construction boom—city and state leaders failed to create comprehensive water policies or invest in water infrastructure.
The spring and summer of 2007 were virtually rainless, and Atlanta’s main reservoir, Lake Sidney Lanier, dropped a record fifteen feet. Frontpage photographs across the nation showed docks high and dry and boats stranded on the lake’s gravel ledges. In April, Georgia was placed under statewide restrictions that limited outdoor watering to three days a week. In May, Atlanta allowed watering only on the weekends. In August, temperatures reached 104 degrees, one degree below Atlanta’s record, set in 1980. In September, officials banned all outdoor watering in the northern half of the state for the first time in history. In October, Atlanta officials asserted that Lake Lanier was less than three months from turning empty, while smaller reservoirs were dropping even faster. In November, Georgia governor Sonny Perdue declared a state of emergency for the northern third of his state, asked President George W. Bush to label it a major disaster area, and cut public utilities’ water withdrawals by 10 percent. Then Perdue joined hands with supporters on the statehouse steps to pray for rain.
To some, the calamity was no surprise. Years of pro-growth policies and lax zoning had led to poor water management and urban sprawl; hydrologists had warned Georgia for nearly two decades that such a drought was possible, but legislators had never developed a coherent response. In the 1990s, plans to build a network of state reservoirs were defeated, largely by developers who were angered that they would not be allowed to build homes around the new lakes. A 2003 plan to sell water permits, which would limit water use, was derailed by Georgians who feared that neighboring states would be able to outbid them. A 2004 initiative to build a state-funded regional reservoir was defeated. At the same time, local farmers planted thousands of acres of water-intensive sod to embellish the growing supply of new housing developments, while golf courses and car washes faced no restrictions on water use during the three-year drought.
“There’s no question this situation could have been avoided,” said former governor Roy Barnes. “We’ve known this for a long time. We have a state approaching nine million people … [and] we have no plan for water.”
The Southeastern drought began in late 2005 and lasted through the summer of 2007. Many commentators blamed global warming, which seemed to make sense. But after carefully reviewing historical climate data, experts concluded that global warming was not the culprit. In 2009, a team of climate researchers led by Columbia University’s Dr. Richard Seagar (who argues that the Southwest is facing a permanent drying out) undertook a dispassionate appraisal of the Southeastern drought and discovered that the three-plus-year dry spell was “quite typical” for the region and will be repeated.
What Atlantans didn’t focus on was the second major finding of Seagar’s study: “In the near future, precipitation will increase year around in the Southeast.” This prediction was borne out almost immediately.
In June 2009, Governor Perdue’s theatrical prayer for rain was finally answered with light precipitation, and Atlanta was able to lit water-use restrictions for the first time in three years. Over the summer, the weather seemed to normalize. Then, on Tuesday, September 15, a low-pressure system crossed Georgia, collided with a high-pressure system over the East Coast, and stalled. It began to rain. As the week wore on, the rain fell harder and then harder still.
On Saturday, September 19, some 3.7 inches of rain fell on the city, which was more than double the record for that date, while over 5 inches fell on the suburbs. By Monday, creeks had overtopped their banks. Forty homes were flooded, power was knocked out across the Atlanta metropolitan area, trees heavy with water crashed to the ground, and the Red Cross began to evacuate people. It rained for eight days straight. In one seventy-two-hour stretch, 20 inches of rain fell on parts of Atlanta.
In what seemed like the blink of an eye, fear of drought turned into fear of drowning.
As Atlanta expanded rapidly in the 1980s and 1990s, its failure to upgrade its water system—by building new reservoirs and other water supplies, limiting growth, and erecting flood defenses—left it vulnerable, first to drought, and then to flood. The drought baked soils hard, making them unabsorbent; when the deluge of 2009 hit, the runoff streamed over the hard soil and the acres of concrete and tarmac that had been laid down, sending storm water crashing through sewers, ditches, and rivers; after eight days of ceaseless drenching, the soils became so saturated that ponds, creeks, and rivers overflowed.
US Geological Survey crews monitored three hundred streamgages—devices that measure the flow of streams in real time—across Georgia. The numbers were stunning. On September 22, the USGS measured water flowing down Sweetwater Creek at twenty-eight thousand cubic feet per second and thirteen feet above flood stage, the greatest flow ever recorded there. The Chattahoochee River rose to heights not seen since 1919. Studying historical patterns, USGS discovered that the chance of a flood of this magnitude hitting this region was 1 in 10,000.
By September 23, the rains eased and the clouds over Georgia began to dissipate. Eleven people were dead, sixteen thousand were homeless, and some thirty thousand were let without electricity. Seventeen bridges across the state were closed, as were stretches of interstate highways. A large sewage treatment plant north of the city was flooded, which caused millions of dollars in damage and released untreated sewage into residential neighborhoods, leading to fears of mass contamination. (The rains also added more than three feet of water to Lake Lanier and returned the reservoir to “full pool” in October. Allatoona Lake, nearby, rose more than thirteen feet over its full pool.) The flood caused at least $250 million worth of damage across seventeen counties, mostly to homeowners who did not have flood insurance. Governor Perdue, who had prayed for rain two years earlier, declared a state of emergency and requested $16.35 million in federal assistance.
“The flooding in Atlanta is certainly near the top of the list of the worst floods in the United States during the last hundred years,” said Robert Holmes, the USGS National Flood Program coordinator.
Meteorologists concluded that the eight days of steady rain that caused the 2009 Georgia flood was an unusual storm pattern for the Southeast. It raised the question of whether the region will become more prone to extreme shifts in weather—years of drought followed by prolonged drenching and floods—as the climate warms in coming decades. This is a question people around the world are starting to ask: its answer, as yet unknown, will have major implications for how runoff is managed and how flood controls are built.
In a sign of the new hydrologic reality, as some places grapple with devastating drought, others are facing unprecedented floods, and some—such as Georgia, Texas, or Australia—are facing both.
Floods are acts of God; flood damages result from acts of men.
—from a congressional report on flood control, 1966
A flood can be local, or it can affect entire river basins and cover several states. It can arrive slowly and gently, or it can appear with sudden, terrifying violence. There are riverine floods, in which heavy precipitation causes rivers to overrun their channels; estuarine floods, caused by storms or tidal surges; coastal floods, which result from hurricanes or tsunamis; catastrophic floods, which are caused by significant events, such as the breach of a levee; and muddy floods, which are generated by agricultural runoff. The most destructive are flash floods, in which water rises to dangerous levels within hours, and these are typically caused by dam failure, collapsing ice jams, or an intense downpour.
Many factors contribute to flooding, but two key elements are the intensity and duration of rainstorms. The less ground cover or wetland available to absorb rising waters, the more likely a river is to flood destructively.
Ground cover—grass, bushes, trees—can help sop up rising water, as can spongy wetlands, the “kidney of the environment.” Wetlands absorb rainfall and waves and help to mitigate the impact of flooding (they also absorb carbon, a greenhouse gas).
In the United States, some wetlands are regulated by states and some by the federal government, under the Clean Water Act. But wetlands are transitional zones that lie between wet or swampy areas and dry upland areas; delineating the boundary between regulated wetland and nonregulated lands has proven contentious and led to numerous court cases (a subject I will return to presently).
An estimated 60 percent of the world’s wetlands—and 90 percent of Europe’s wetlands—have been destroyed over the past century, according to Science Daily. When wetlands are filled in and built upon, the displaced water will try to reassert itself. Often the result is flooding, erosion, and wet basements filled with mold.
Floods hold a special, dark grip on the human imagination. People from many societies and religions have envisaged the end of the world arriving in the form of a “great flood,” or “deluge,” as told in the story of Noah’s Ark in the book of Genesis, or in Ovid’s Metamorphoses, or the Babylonian Epic of Gilgamesh. The single deadliest natural disaster ever recorded was a series of terrible flash floods in central China in 1931. After a prolonged drought, heavy snowstorms arrived, along with rains, causing the Yellow, Yangtze, and Huai Rivers to overflow, killing an estimated 2.5 to 3.5 million people.
Between 1991 and 1999, floods in the United States killed 850 people and caused over $89 billion in property damage, according to the Sierra Club. The deadliest natural disaster in American history was the flooding of Galveston, Texas, by a hurricane in September 1900, which killed about eight thousand people. The Okeechobee hurricane of 1928 killed about four thousand people in Florida, the Bahamas, and Puerto Rico. Hurricane Katrina, which devastated the coastal regions of Louisiana, Alabama, and Mississippi in August 2005, killed 1,836 people and caused at least $150 billion worth of damage—a figure that is still being tabulated and does not include the billions of dollars in insurance payouts and court-awarded damages still being litigated.
Nearly half the world’s population—some 3 billion people—live in coastal regions vulnerable to the rising oceans, intensifying storm surges, and less obvious problems such as saltwater pollution of aquifers. Rising waters are already a problem for low-lying nations such as the Maldives, the Seychelles, Bangladesh, and Holland. The World Bank estimates it will take $75 billion to $100 billion to build adequate flood defenses in developing countries. Rich countries—the ones largely responsible for pumping climate-warming gases into the atmosphere—have been reluctant to fund flood protection in poorer nations.
By 2025, when the world’s population will be around 8.5 billion, some 6 billion people are expected to live in coastal zones. Predicting future sea-level rise is a murky and inaccurate science, but the National Academy of Sciences (NAS) estimates seas will rise by five to twenty-one inches by 2025, and by 2100, the NAS expects seas to rise from two to eleven feet above current levels. At that point, rising waters will become a far greater danger to humans than they are now.
The United States has a flawed flood-protection system. It relies on a mishmash of federal, state, and local bureaucracies—a dysfunctional system notorious for political turf wars, chronic underfinancing, and a lack of leadership. Moreover, certain policies—such as the federal government’s “hundred-year levee” design and its flood insurance program—have unintentionally provided a false sense of security and encouraged people to build on risky sites.
In erecting floodwalls, federal agencies, such as the US Army Corps of Engineers, use a “hundred-year” design criterion to determine how big a levee should be. This does not mean a flood will occur once every hundred years; it means there is a 1 percent chance of the levee’s being overtopped each year. (A levee is overtopped when high waves wash over it and swamp the “protected” area behind the levee.) The risk of overtopping increases with time, as levees weaken and subside. Yet, people who build behind a hundred-year levee are not required to elevate their buildings, floodproof their structures, or purchase flood insurance—and as a result, their property is susceptible to flooding.
The levees ringing New Orleans were supposedly built to withstand a hundred-year flood, but the wind and waves of Hurricane Katrina plowed right over, under, and through them. Climate change will exacerbate the problem: a one-foot rise in ocean water levels, scientists estimate, will increase the frequency of a hundred-year storm event to once every ten years.
In certain regions, the federal government requires people to buy policies from its National Flood Insurance Program in order to get a mortgage. Most private insurers don’t offer insurance in floodprone areas because the damage from high waters is expensive to clean up. But the government’s insurance program keeps premiums artificially low (to make them affordable) and often relies on outdated flood maps (because updating them is expensive and is not a priority), which, perversely, encourages building in risky areas.
In 2006, a year after Katrina, only 20 percent of American homes at risk of flooding were properly insured. Despite years of brave talk about fixing the insurance program, Congress has failed to do so. When major floods occur, federal agencies and the National Guard provide substantial aid—meaning that taxpayers ultimately end up shouldering the homeowners’ risks of building in flood zones.
One place that experts agree has a high probability of being hit by a severe hurricane in coming years is the Northeast, including New York City. By 2050, climatologists estimate, the city is likely to experience a three-to-five-degree rise in temperature and a 5 percent increase in rain and snow. As the climate warms, ocean waters will expand and cause the seawater around the city to rise about ten inches by 2050 and two feet by 2080, according to a study by Columbia University. When stirred by a hurricane, high waters could swamp the city.
According to a NASA climate study, if a Category 3 hurricane, like Katrina, were to hit New York, it could create a storm surge of up to twenty-five feet high at Kennedy Airport, twenty-four feet at the Battery, twenty-one feet at the entrance to the Lincoln Tunnel, and sixteen feet at La Guardia Airport. Such a deluge would destroy billions of dollars’ worth of property and could shut the city down. While the sandhogs and pencils at the city’s Department of Environmental Protection (DEP) are racing to finish Tunnel No. 3 to avoid the city’s running dry, other experts at the agency are concerned about the opposite problem: the prospect of Manhattan turning into a modern Atlantis.
In 2006, Max Mayfield, then director of the National Hurricane Center, told Congress, “It is not a question of if a major hurricane will strike the New York area, but when.” That same year, just a few months after Hurricane Katrina, an insurance-industry assessment ranked New York City as “the second worst place for a hurricane to hit,” after Miami. Another survey that year found that New York residents had taken only a few of eight basic steps—such as getting flood insurance or putting together a disaster evacuation plan—to protect themselves.
In the spring of 2007, New York was deluged in a series of vicious stormbursts; one, in April, saw seven inches of rain fall in hours, the biggest precipitation measured there since 1882. If coupled with a nor’easter storm or a hurricane, extreme flooding could hit the city once every forty-three years by the 2020s, once every nineteen years by the 2050s, and once every four years by the 2080s, the DEP predicts.
To help absorb more precipitation, New York has begun to restore wetlands, promote green roofs and bigger tree pits, and build porous “green-streets,” which allow storm-water runoff to filter into the ground. But these are incremental steps. The city relies on an old, already overburdened sewer system and a few catch basins and inflatable flood barriers. Its flood maps, which show areas prone to hazardous flooding, are outdated, and the city has no major flood defenses, large catchments, or storm-surge barriers.
“We’ve been talking about this lack of preparation since the early 1990s, but before Hurricane Katrina no one would listen to a bunch of academics,” said Dr. Malcolm Bowman, a professor of physical oceanography at the State University of New York at Stony Brook. “Now a few people in the city government are returning my calls.” Hurricane Katrina, he said, was “a warning to other cities of what kind of disasters could be in store. A temperature shift will change our world, and we’d better be prepared for it.” With higher sea levels, he said, weak storms of the future will be as destructive as severe storms are today.
To defend against the catastrophic flooding of New York City, Bowman and his colleagues at Stony Brook’s Storm Surge Research Group have proposed building three enormous storm barriers, thirty-five feet high and up to a mile long, to provide “a circle of protection” from surging ocean waves.
In England, the Thames River is outfitted with giant swinging floodgates; in Italy, the Venice lagoon is defended by inflatable gates that rise from the seafloor; in Holland, massive swinging barriers protect Rotterdam from the North Sea.
The enormous gates that Bowman suggests for New York have not been designed yet but are likely to be swinging metal structures, with navigation locks to allow boats and water to flow through. They would be placed at the Verrazano Narrows (the gateway to New York Harbor), the upper reaches of the East River (where it joins Long Island Sound), and across the Arthur Kill (a stretch of water between Staten Island and New Jersey).
A similar scheme, designed by Halcrow Group Ltd., a British infrastructure consulting firm, calls for a single five-mile-long barrier to be built between Sandy Hook, New Jersey, and the Rockaway Peninsula, in Queens. A third floodgate plan, from New York–based engineers Parsons Brinckerhoff, envisions a wall that lies flat on the bed of the East River and would crank upward to block storm surges as high as twenty-five feet. Each of these projects would take years to build, might impact water salinity and aquatic life, and would cost an estimated $1 billion to $6.5 billion.
City officials have called these ideas “intriguing” but “theoretical,” and say floodgates will not be needed for several decades. In the event of a major flood, they plan to evacuate some 3 million city residents via overland routes.
“That seems a bit hopeful,” said Dr. Douglas Hill, a colleague of Bowman’s at Stony Brook, who has been exasperated by the city’s lack of interest in their warnings. “We are expecting to be flooded. It happens here.” He recalled Hurricane Floyd in 1999, when “the data says that was the worst flooding here in the last half century. It also says that we will see more hurricanes like it. [Floyd’s 155-mile-per-hour winds hit five East Coast states, causing some $4.5 billion in damages and killing sixty people.] After the disaster in New Orleans, we can’t say we haven’t been warned.”
Hill and Bowman estimate that their three barriers would cost $9 billion, a figure that sounds scary. But Hill believes it is a relative bargain. (By comparison, a proposed railroad tunnel from New Jersey to Penn Station could cost $8.7 billion, and a plan to raise railroad infrastructure above flood levels might cost $5.6 billion.) “New York City is the financial center of the world. It is uniquely valuable, and vulnerable,” said Hill. “If New York floods, it will cost the nation at least 1.9 trillion dollars. If you look at it that way, 9 billion dollars for floodgates isn’t so bad.”