We can't stop rivers from flooding. But we can stop making the floods worse.
by Bob Schildgen
The Deluge is ancient, universal, inevitable. Stories of a catastrophic flood
were told 4,000 years ago in Mesopotamia, and have been recited from desert to
rainforest, from the Natives of Australia to the Maya in Central America. In many
of these myths, the flood is a punishment for the sins of humanity.
What sins are we suffering for now? Floods are so frequent and intense that it
seems we've returned to the drenched mythic dreamtime. The trouble begins with a
deep snow cover in the mountains, a quick spring thaw on the prairie, a sudden
downpour, a relentless gray month of steady rain. As waters rise, people in the
floodplain anxiously listen to weather reports and upstream flood measurements,
and watch the muddy torrents on the news. Then comes the desperate sandbagging,
pumping, bulldozing, emergency levee building. Families flee to high ground before roads
and bridges wash out. Homes, farms, and businesses are ruined, submerged or
ripped away in a rush of water and mud.
The human cost can be read on the faces of those huddled in makeshift shelters
and school gyms, and heard in their dazed accounts of loved ones who disappeared
under the waves. Our sense of what is important changes: in Grand Forks, North
Dakota, a police officer boats to his ruined house, and steps out of the water
into his second-story window to save his wife's wedding dress. There is heroism
and sacrifice and community solidarity. One after another, survivors tell how the
sundering flood made them closer than before.
Then the politicians arrive by boat and helicopter, looking grave and rugged in
their flannel shirts and hunting caps and outdoor gear. The president declares a
disaster, and another relief effort begins.
We call these disasters "natural" and even "acts of God." True, rivers always
have and always will overflow their banks. But there is increasing evidence that
human hands are roiling the already angry waters; we have forgotten the ancient
lesson that floods are the price we pay for our own actions.
In an attempt to save ourselves, we build not arks, but massive dams and levees
that enable us to live and farm on the lands where the water belongs. This
reflexive reliance on technical fixes is notorious for ruining vital natural
ecosystems. Dams obliterate river valleys, turning them into artificial lakes,
while levees cut off rivers from riparian habitat. Moreover, these remedies can
defeat their purpose. A growing number of flood watchers warn that excessive
dependence on structures actually aggravates floods, as does our destruction of
water-storing wetlands and reckless development of floodplains. We spend
billions, first to prevent floods and then to recover from them, but much of that
money is merely subsidizing disaster.
THE U.S. GOVERNMENT DOES more to promote floods than any other entity. More than
40 separate federal programs and agencies, governing everything from highway
construction to farm export policy, encourage building and farming on floodplains
and wetlands. In 1996 alone, according to an analysis by Sierra Club Midwest
representative Brett Hulsey and the National Wildlife Federation's David Conrad,
over $7 billion was poured into ten programs that aggravate flooding. "So much
subsidy goes into the development of floodplains that there's no
incentive to stay out," says Nancy Philippi, vice president of the Wetlands
Initiative in Chicago.
Between 1960 and 1985, the federal government spent $38 billion on flood control,
yet average annual flood damage-adjusted for inflation-continued to increase,
more than doubling. Since 1990, damages have averaged more than $5 billion a
year. When rains pounded the Upper Mississippi watershed for days on end in the
spring of 1993, the cost was $6.5 billion. When a "Pineapple Express" from the
subtropical Pacific dumped heavy rains on California and brought on the New
Year's Flood of 1997, rivers swelled and broke or overtopped many stretches of
California's thousands of miles of levees-just as they had in the Midwest-at an
estimated cost of $1.7 billion. When the Red River, which flows up through the
prairie between North Dakota and Minnesota, flooded later that year, another $3
billion in damages was added.
The human toll is also staggering. More than 500 people have been killed since
1993 in the Great Midwest Flood and the many floods that followed, a loss that
would have been far higher without modern weather forecasting and communications
to spread the word to sandbag or flee.
THE TRADITIONAL DEFENSE against floods is to treat them as a plumbing problem.
Dams are built to contain the water, and levees-mounds of earth, riprap, or
concrete along the banks of the river-seek to confine it. When another 100-year
flood comes ahead of schedule and washes these structures away, they are rebuilt
bigger and stronger.
The bulk of the $4 billion annual budget of the U.S. Army Corps of Engineers, the
federal agency primarily responsible for flood control, goes to building and
maintaining waterworks. The Corps boasts that its dams and 8,500 miles of levees
have saved some $387 billion in damages since 1928. No similar assessment exists
of the damages wrought in areas where Corps projects encouraged development that
was later inundated.
The barriers to change may be as much a matter of political culture as of the
Corps' long tradition of engineering. Congress is a significant obstacle, says
Larry Larson, executive director of the Association of Floodplain Managers. "They
still believe all their constituents want structural solutions, because the
structural cadre has been lobbying for a hundred years. Other people don't get
heard from." In addition, Larson notes, there's the congressional ego factor:
"Dams are 'plaqueable.' The plaque says, 'that's the senator's dam.' " Less
intrusive approaches tend not to be monumental, and thus not so politically
Flood management is complicated by the fact that federal agencies often work at
cross-purposes. For example, while the Department of Agriculture's Wetland
Reserve Program is buying up marginal cropland and restoring it as wetlands, the
Corps is issuing permits that allow drainage and destruction of wetlands at the
rate of 70,000 acres a year. When government-built dams and levees fail,
emergency relief and federal flood insurance encourage rebuilding in the same
locations. "The federal government tries to adjust nature to us, rather than
letting us adjust to nature," says Larson. "It does too good a job bailing people
out. If a community knows it can get a hundred percent aid to rebuild, there's no
incentive for moving out." Inappropriate development in flood-prone areas occurs,
he adds, because "too many city councils say, 'if we don't let them build,
they'll go someplace else.' I don't blame the developers as much as city
An indication of confusion as murky as the Missouri River itself comes from House
Majority Leader Dick Armey. "Don't move away-rebuild," he exhorted the
waterlogged folks of Grand Forks. "If I were sitting here today making that
decision I'd come back." Armey's advice runs directly counter to the policies of
the Federal Emergency Management Agency (FEMA), which has been trying to break
the traditional flood-and-rebuild cycle by moving property out of harm's way and
discouraging reckless development.
In recent years, the Corps has begun, ever so slowly, to seriously consider
nonstructural approaches to flood control. It is even attempting to undo some of
its past environmental damage in the Everglades and around the upper Snake River
in Wyoming, where it is removing levees to let water flow into old channels.
Nevertheless, the engineers still rule. In California, failed levees are being
rebuilt, raised, and hardened-just as they are being built along the
Mississippi-while other strategies are largely neglected. "Two years ago the
water started to recede in California," says Jeffrey Mount, chair of the geology
department at the University of California at Davis. "The reality is that despite
thousands of hours of work, we're no further along than two years ago, and in
many ways worse off. We remain addicted to levees as the first line of defense."
Mount advocates getting rid of some levees and moving others farther away from
riverbanks, giving the river room to slow down and flow through natural paths, to
loaf in sloughs and swamps and spill over into farmlands that can tolerate a
certain amount of flooding. This approach would not only curb further development
in these areas but also help revive much-abused riparian areas.
Even when they stand firm, levees can actually cause or exacerbate floods.
Constricting and intensifying the river's flow, they can channel a flood
downstream, making one community's salvation another's trauma. Conversely, a
breach in upstream levees can be a blessing for those downstream. During the New
Year's Flood in California, says Mount, "if there had not been breaks at
Olivehurst on the Feather River, it may well have flooded Sacramento."
Like California's capital, St. Louis may have been spared enormous damage in
1993, when the Mississippi and its tributaries topped or broke more than 1,000
upstream levees, causing vast inundations. Even so, the flood still crested in
St. Louis at over 47 feet, six feet above its previous record. No one knows how
much greater the destruction might have been had the levees held.
Neither the Mississippi and its tributaries' thousands of miles of levees and
dozens of huge dams nor California's 6,000 miles of levees and more than 1,400
dams can reasonably be expected to do the entire job of flood control. "We cannot
prevent floods," Mount warned in congressional testimony after the California
flood. "The hard lesson learned is that despite our seemingly Herculean
engineering efforts, floods are going to happen."
NOTHING DEMONSTRATES MORE vividly than a river the degree to which one's own
backyard is linked to the ecosystem. The Mississippi watershed, for example,
drains more than one-third of the United States. A raindrop falling in Montana
can end up in the Gulf of Mexico. Its destiny is determined not just by weather
and the lay of the land, but by myriad human actions: agricultural tilling and
drainage, suburban development, deforestation, and the decisions of hundreds of
local, state, and federal agencies as well as thousands of private landowners.
Runoff from a flooded cornfield in Minnesota can end up killing fish off the
coast of Louisiana, because fertilizers washed downriver promote the growth of
algae that deprives the water of oxygen. (After the 1993 Mississippi flood, the
zone of oxygen-depleted waters off the Louisiana coast doubled to almost 7,000
square miles.) That cornfield in the floodplain might have been planted because
of agricultural subsidies, a drought in Russia, or a growing demand for
bacon-burgers. A case can be made that Ronald McDonald, not nature or an angry
God, is the true Lord of the Floods.
Even the amount of rain falling on the field may be tied to human activity. The
recent El Niņo weather system, with its increased precipitation and more intense
storms, has been linked to global warming (see "The Invisible Hand" in the
May/June 1998 issue). Thus the sport utility vehicle that gets washed away in the
flash flood may have helped bring about its own destruction.
A turning point in flood control came in the wake of the 1993 flood when Corps
General Gerald Galloway bucked tradition and called for more emphasis on
nonstructural methods, including the acquisition and restoration of wetlands and
riparian habitat, stricter limits on development in floodplains, and even a farm
policy that discourages the conversion of wetlands to cropland.
Wetlands can store excess water like a sponge, and in some cases may be more
efficient than manmade reservoirs. Depending on the soil type, they can contain 1
million to 1.5 million gallons of water per acre, and can alleviate flooding,
though no one really knows to what extent.
Wetland destruction, however, clearly aggravates flooding, even on a local scale.
Doris Wilson, an elementary school teacher from Louisville, explained in
congressional testimony how her home flooded on March 1, 1997, because a nearby
wetland had been drained by a developer. "I realize we had a lot of rain that
day," she said. "However, the developer created the situation that made the flood
What happened in Wilson's yard may be a miniature version of what occurs in an
entire watershed. Almost 120 million acres of U.S. wetlands have been destroyed
for agriculture and development, more than half of what existed prior to European
conquest. California, Missouri, Iowa, and Illinois have allowed over 85 percent
of their wetlands to be destroyed. Motorists rolling along midwestern highways
through the vast stretches of corn and soybean fields don't realize that they're
driving over the grave of a wetland, sometimes with miles of underground pipes
draining off the water to prevent it from returning.
Some geologists and soil scientists consider it more than coincidence that the
Red River flood occurred in an area where there has been large-scale drainage of
wetlands. "Water retention is significantly less than ten, twenty, or thirty
years ago," says Dexter Perkins, a geology professor at the University of North
Dakota in Grand Forks. "Seventy-five percent of the wetlands in the Red River
basin have been drained. In several counties it's ninety-nine percent."
While scientists agree that wetlands can reduce flooding in upland areas, there
is disagreement on whether they can prevent the flood peaks that are the major
cause of damage. Donald Hey, a hydrologist with the Wetlands Initiative, contends
that the excess water that poured through St. Louis in 1993 would have covered
slightly more than 13 million acres-half the amount of wetlands lost in the Upper
Mississippi region since 1780, but only 3 percent of the land in that area. Hey
argues that "by strategically placing at least thirteen million acres of wetlands
on hydric soils in the basin, we can solve the basin's flooding problems in an
ecologically sound manner." The Corps' hydraulic engineers sharply dispute this
analysis, saying that it would have been physically impossible for the wetlands
to have contained the 1993 flood. Even if this is so, argues Perkins, the prudent
course is to regard wetlands as insurance against floods, protecting and
restoring what we can.
The idea that swamps and bottomlands can tame a torrent is hardly new. In 1849,
Louisiana Senator Solomon Downs testified on the Swamp Act, which transferred
federally owned wetlands to the states and opened them up to development. "It is
reasonable to suppose that the whole country is now more rapidly and thoroughly
drained into the Mississippi than when in a state of nature," Downs said. "Then,
no doubt, a great quantity of water was collected in pools and swamps, and there
remained until carried off by gradual evaporation." The issue of flood control on
the river was hotly debated for years after the Swamp Act began to bring massive
new settlement on the floodplains, with one camp in favor of levees and the other
for backing off from the river. The levee advocates won out, but the most massive
federal projects on the Mississippi were not undertaken until after the flood of
1927. Gilbert F. White, generally acknowledged as the nation's foremost authority
on floods, was already making a case for nonstructural methods in the 1930s,
when he served as an advisor to President Franklin D. Roosevelt's administration.
ONCE THE WATERS RECEDE, the debris has been hauled away (100 tons a day in Grand
Forks in 1997), and rebuilding begins, people tend to leave flood control to community leaders
and the federal government. Making public policy is tedious enough in good times,
let alone when a family is confronted with a houseful of mud, molding wallboard,
and wrecked appliances. "Our flood-memory half-life is remarkably short," says
Galloway. One victim of the Red River flood told Minnesota Public Radio that her
home was in the 500-year floodplain, "so I figure I won't see another flood like
this 'cause I won't live that long."
More and more Americans are living in harm's way: FEMA has identified 10 million
households and businesses with property valued at a trillion dollars on some
150,000 square miles in flood-prone areas. Cities and towns grew up on rivers
because of the need for inland water transport, and once established they cannot
very well pack up and move. But ceasing to build on the most vulnerable areas
would clearly reduce the damage and loss of life. And while the rich alluvial
soil of floodplains makes fine farmland, not every acre has to be cultivated
right to the edge of a river.
Increasingly, communities are opting out of the flood-and-rebuild cycle. After a
1972 flood killed 238 people and caused $500 million in damages, Rapid City,
South Dakota, used federal funds to buy 1,400 pieces of property and create a
greenway. After the town of Valmeyer, Illinois, was flooded in 1993, 600
residents used $35 million in state and federal aid to move to higher ground. In
St. Charles County, Missouri, a similar relocation after the 1993 flood meant
that when the river flooded again in 1995, damage and the cost of disaster relief
was a fraction of the previous total.
Soldiers Grove, Wisconsin, has avoided
disaster by moving its business district away from the Kickapoo River and
creating a new town center. And in Napa, California, where flooding costs have
averaged $15 million a year since 1960, citizens rejected a plan by the Army
Corps to dredge the Napa River and build more levees. Instead, a broad coalition
campaigning on the slogan "a living river" mustered the necessary two-thirds vote
last year to restore 600 acres of marshland, move levees back from the river, and
relocate more than 60 structures off the floodplain.
Tulsa, Oklahoma, has gone from almost total reliance on structural control to a
more integrated approach. In 1964, after the Corps of Engineers completed the
Keystone Dam 15 miles upstream, Tulsa believed it was finally secure from the
constant flooding of the Arkansas River. Lowland meadows were paved over and
developed, and the population grew by 25 percent in the 1960s. When flooding did
occur, the response was the typical return and rebuild-until the 1976 Memorial
Day Flood that took three lives and wrought $40 million in damages.
citizens recognized that it might have been worse if a large local park had not
been preserved as open space. So after the deluge, Tulsans elected new city
commissioners who declared a moratorium on floodplain construction. After a
second Memorial Day Flood killed 14 people and caused $180 million in damage in
1984, Tulsans pushed the nonstructural solution further. Working with FEMA, the
city relocated over 500 houses and mobile homes and ultimately moved more than
900 buildings out of the most critical areas.
In most of these situations, local citizens broke the pattern of leaving things
to the experts and became deeply involved in the planning themselves. Larson of
the Association of Floodplain Managers emphasizes the importance of citizen
participation in developing a comprehensive plan that takes into account the
entire local economy. Local citizens are also needed to change attitudes in
Congress, and with more people involved, it is easier for municipalities to
thread through the various bureaucracies to find and secure state and federal
In the states affected by the Midwest floods, FEMA has helped buy out or relocate
more than 20,000 properties at a cost of only $480 million-a bargain in that
every dollar spent saves two on future disaster-relief costs-and the agency would
like to acquire more. "We have identified 35,000 repetitive flood-loss properties
across the country that have had two or more flood-loss claims in the past ten
years," says FEMA director James Lee Witt. The agency hoped to relocate 7,300 of
these properties, which would cost $300 million over a three-year period, but
ultimately save an estimated $1 billion in damages. Under the Clinton
administration's new budget, however, FEMA expects an $88 million shortfall for
Unfortunately, the disaster policy mandated by Congress remains backward,
treating symptoms rather than causes and shelling out more money for disaster
relief than for prevention. (For example, FEMA is allowed to use only 15 percent
of the total spent on disaster relief for mitigation.) Thus Witt is proposing
other measures to discourage building and living in the danger zone, such as
refusing to issue flood insurance to property owners who have filed two or more
Another important effort is the Agriculture Department's Wetlands Reserve
Program, which has set aside 665,000 acres of wetlands and is slated to acquire
and restore 310,000 more by 2001 at an average cost of $850 an acre. The cost of
restoration overall averages about $200 per acre; in some places the wetland can
return simply by being left alone, while in others considerable work is required
to recontour and revegetate the landscape. The program usually purchases a
perpetual easement to the land at prices based on its value as agricultural land.
"This totally changed the land use on marginal lands," says Wetlands Reserve
Program Director Bob Misso. "It's a hell of a deal for taxpayers, for landowners,
and for the environment." Since the acquisition cost is often capped at $800 an
acre, however, and the price is pegged to agricultural value, the program doesn't
work when speculative developers have driven up land prices. And some farmers
consider the compensation inadequate. In North Dakota, only three square miles
has been enrolled in the program.
Although there are many mysteries in the soul of a river and the heart of a
flood, it is becoming clear that we all would benefit from a remedy that
environmentalists have long pleaded for: reducing human impact on the earth by
protecting and restoring natural places, building compactly and halting sprawl,
and curbing population growth. The other thing we have learned about flood
prevention is the oldest truism of democracy, that active engagement by a broad,
well-informed citizenry is key to improving the way things are done.
No matter what we do, catastrophic rains will fall and implacable torrents will
flow, and we will never control them completely. But with a combination of
respect for nature and restraint in our own actions, we stand a better chance of
riding out the storm.
Protecting Our Families From Floods
Floods are inevitable, but they don't have to be disasters. Across the country, a
number of initiatives can move beyond reliance on dams and dikes.
Floodplain sprawl. Much of the damage from floods comes from homes, farms, and
businesses built in the river's natural channel or floodplain. Congress will soon
be reauthorizing the Water Resources Development Act, and conservationists want
it to include a 100-year moratorium on floodplain development.
Wetlands and Conservation programs. The Wetlands Reserve Program pays landholders
to restore wetlands, and the Conservation Reserve Program compensates farmers for
leaving critical parts of their land undeveloped, thus reducing agricultural
runoff and soaking up flood waters. The programs also help improve the farm
economy, which is suffering from overproduction. Both of these voluntary
programs, however, face annual battles for funding in Congress.
Relocation. The Federal Emergency Management Agency's hazard-mitigation program
moves people out of harm's way. The program is a bargain: for every $1 spent on
relocation, $2 is saved in disaster relief. The agency is seeking to increase its
budget for these buyouts to $50 million a year.
Wetlands protection. Under the Clean Water Act, it's up to the Army Corps of
Engineers to issue permits to destroy, fill, or drain wetlands-and the Corps
grants 90 percent of all requests. Development interests like the National
Association of Homebuilders are lobbying the Corps to further weaken its
requirements. In addition, a recent court victory by the American Mining Congress
means that developers can now drain and destroy wetlands without having to get a
The Corps still has the administrative power to protect wetlands, and the Sierra
Club is now lobbying President Clinton and the Corps to do so.