Monday, May 1, 2017

Brink of Disaster?

At least one dam fatally collapsed during the 1913 flood, and others threatened to. Today, the U.S. has 15,000+ “high hazard potential” dams whose failure would likely kill people. Most dams are nearing the end of their design life and their condition is “deficient.” What would happen if 1913-scale flooding swept the nation today?

[Happy May! This month includes National Dam Safety Awareness Day, Flood Safety Awareness Week, and National Drinking Water Week (see “Days of Warning”).]

                “After watching the rapid ride of water in our business section and lowlands” of Fremont in northeast Ohio, Cornelia A. Gast and two unnamed companions took “a long, muddy tramp” to Ballville, two miles upstream along the Sandusky River (and about 18 river miles from Lake Erie). It was Tuesday, March 25, 1913, two days into the heavy rains that had been falling since Easter Sunday; they wanted to see what
Two men survey the raging waters of the Sandusky River from the top of the incomplete Ballville Dam before its collapse. Credit: Charles E. Frohman Collection, Rutherford B. Hayes Presidential Center

was happening at the concrete dam (about 400 feet across and three stories high) and retaining wall under construction there to build a hydroelectric power plant. From the river road, the three friends were astounded to behold “water at a depth of five feet or more pouring over the dam and the top of the wall.” 

Wanting a closer view, they were about the cross the road when the northeast end of the retaining wall suddenly caved in and “with a rush and roar its pent-up waters swept all in their path. In a moment’s time the great derrick was toppling over, the buildings torn from their foundations and swept away, the great steel flume 14 feet in diameter, was lifted up and carried a distance of half a mile or more.” Several houses were rocking to and fro from the force of the surging waters, two finally carried down the river. 

Terrified, the three scrambled through wire fences and up a hillside to the cemetery. As they looked back, they saw the horse and carriage of the local doctor “carried away by the on-rushing waters, and the doctor, who so bravely rushed to the rescue of his horse, was seen clinging to the iron hitching post in water shoulder deep.” Around them, men were running and warning everyone to flee. “Signal shots were heard in every direction, and in the distance could be plainly heard the fire bell warning the people at Fremont of the coming danger.”1
This is what a turbulent flow of 65,000 cfs looked like in February 2017 over the damaged spillway of the Oroville Dam—tallest dam in the nation—after record winter rains in California. Photo is the seventh image of 22 compiled from several news sources into a dramatic slide show. Credit: Reuters

At least three men died during the Fremont flood. Moreover, according to stats kept by the U.S. Geological Survey, the 1913 flood is still the flood of record in Ballville, with “an estimated peak discharge of 63,500 cfs” or cubic feet per second.2

           Now, the magnitude of a peak discharge of 63,500 cfs that approximates the normal flow of the ferocious Class 5 and 6 rapids of the Colorado River. It was also the catastrophic flow over the spillway of the Oroville Dam in California this past February (2017) that eroded away huge blocks of concrete and compelled the evacuation of some 190,000 people downstream.

Panics about dam failures
                As far as I have been able to determine, the Ballville Dam may have been the only dam to have collapsed during the 1913 flood—or at least the only one in Ohio (although many levees collapsed in various states). I’m not completely convinced of that as I have not verified this for every state affected by the 1913 flood, but in 1913 there were not all that many dams in Ohio or the rest of the country. However, the nation’s worst dam failure, and the resulting loss of 2000+ lives in Johnstown, Pennsylvania in 1889, was not yet 24 years in the past and its memory was still fresh (see “An Unnecessary Tragedy: The Johnstown Flood”). 

Columbus Dispatch,
March 27, 1913, p. 8
So in 1913, with all that water falling out of the sky, people were understandably hypervigilant about the possibility of other dams collapsing. And with the powerful weather system’s decimation of telegraph and telephone communications (see “The First Punch”, “Crisis Communications in a Communications Crisis”, and “Heroism of the ‘Hello Girls’”), and no way of obtaining reliable information, rumors flew and panic broke out in several cities around Ohio. 

The most famous incident was in Columbus, when on Wednesday, March 26, people fearful of the collapse of Griggs Dam (a water storage dam north of the city) started running—only to discover they had panicked on the basis of an unfounded rumor. Two decades later, that panic inspired a famous short story by humorist James Thurber (see “The Day the Dam Broke?”). 

But Columbus was not the only city during flood week to be gripped by such panic. In northern Ohio the next day (Thursday, March 27), residents of Kent just south of Cleveland were similarly spooked by rapidly spread rumors of the failure of Kent Dam across the Cuyahoga River. The flood actually did damage the dam but the structure did not collapse.

Cleveland Plain Dealer, March 28, 1913, p. 13
Similar panics were sparked by rumors of the collapse of at least three other dams: the Lewiston dam across the Miami River in Logan County north of Dayton, the Loramie dam across the Loramie River in Shelby County, and the Celina dam in Mercer County near the headwaters of the St. Mary’s River. The Celina dam impounded the Grand Reservoir, what was then the largest artificial lake in the state (15,748 acres) and the dam was overtopped and the banks considerably washed. 

“While the immediate danger was small and there was little need for the excitement which prevailed, such mental disturbances cannot help but have a decidedly detrimental effect upon the people,” wrote E. F. McCampbell, secretary of the Ohio State Board of Health, in a special report published May 1913. Indeed, he noted, “The reports circulated concerning the breaking of the reservoirs had very grave results and probably caused the loss of several lives that might have been otherwise saved.” 3  

Creeping hazards, deficient condition4
After several major fatal dam collapses in the 1970s, the U.S. Army Corps of Engineers (USACE) set up the National Inventory of Dams (NID): a publicly accessible database of dams of any size that potentially threaten at least one human life, plus dams bigger than a certain size (25 feet high storing more than 15 acre-feet, or 6 feet high if storing more than 50 acre-feet; an acre-foot, a volume unit for measuring reservoirs, is 325,851 gallons) regardless of hazard to human life. As of the 2016 update, the NID included 90,580 dams, half of which (45,402) are under 25 feet high. 
Akron Beacon-Journal, March 27, 1913, p. 1
Dams in the NID are categorized by hazard potential. Hazard potential is not condition; it is purely an assessment of the likely severity of downstream damage should the dam fail, given its location, size, reservoir capacity, downstream assets and population, etc. A high hazard potential dam—also called Class 1 (or I) by some states—is one whose failure would probably result in loss of human life regardless of property damage. 

Some 17 percent—15,498—of the dams in the NID are categorized as high hazard potential. A dam does not have to be big to threaten lives. According to a 2011 statistical analysis by the U.S. Department of Homeland Security, at least two-thirds of fatal dam failures since 1900 resulted from the collapse of dams considered “small” or “intermediate” between 20 and 49 feet high, with half a dozen impounding less than 100 acre-feet.5

The number of high hazard potential dams is inexorably growing—a phenomenon called “hazard creep:” One cause is increasing population around the country, which encroaches into previously uninhabited downstream hazard zones of dams originally built in comparatively isolated areas. Another cause is the decades-long increasing trend toward exceptionally heavy rainfall events.6 Both factors are forcing regulators to reclassify dams. Indeed, across the country, the number of dams now classified as high hazard has grown by nearly 60 percent in less than two decades, up from 9,281 in 1998.7 

About 14 percent (2,171) of the 15,498 high hazard potential dams in the NID are rated poor (red) or unsatisfactory (black). Moreover, the condition of another 23 percent (3,442) is not rated, either because it is unknown or undisclosed; however, if the same proportion (14 percent) of the unrated dams also is in poor or unsatisfactory condition, then another 500 or so of the unrated high hazard potential dams nationwide likely are also deficient.

Dams in the NID are also rated by condition or structural soundness. The USACE and the Association of State Dam Safety Officials (ASDSO) use a five-point scale: satisfactory, fair, poor, unsatisfactory, or not rated.8 As of the 2016 NID update, 14 percent of high hazard potential dams are considered deficient—having not only potential to take lives, but also being at significant risk of failure9—and another 23 percent are not rated.  

Independently, the American Society of Civil Engineers (ASCE) in its quadrennial Infrastructure Report Card reports, awards letter grades (A, B, C, D, F, plus ?) for condition.10 For the nation as a whole, in 2017 the ASCE assigned dams a grade point average of D, meaning “Poor: At Risk.” That national G.P.A. for dams has bounced around the basement since 1998. 

But wait, there’s more:

Age and neglect
By 2020—just three years hence—nearly two-thirds of U.S. dams will be older than half a century. Since many dams, like much other heavy infrastructure (such as interstate highways and power plants), are built with a nominal design life of 50 years, that means a “staggering percentage of our dams… have either just surpassed the extent of their intended design lives or will do so in very near future,” noted the Center for American Progress (a public policy think tank) in 2012.11

Heavy infrastructure can long outlast its design life if it is well-maintained. But therein lies the rub. Age exacts a heavy toll with neglect. 
Of the 76,361 dams in the NID (of all hazard potentials) giving completion dates, 63 percent (48,516) were built before 1970, so are now at least 47 years old. Moreover, well over 15 percent are 75+ years old (built before 1942), including some in areas afflicted by the 1913 flood. Credit: USACE NID

According to the Federal Emergency Management Agency (FEMA), an analysis of 656 U.S. dam failures from 1975 to 2011 reveals that the single greatest cause of dam breaches (70.9 percent, 465 incidents) is flooding and overtopping from heavy rainfall. The second biggest cause of failure (14.3 percent, 94 incidents) is piping or seepage through or under a dam.12

Overtopping can happen to a dam of any age, but how well the dam withstands it depends also on the soundness of its internal structure and maintenance of its exterior. Risk of overtopping increases if a reservoir’s capacity is diminished due to the deposition of sediments, or if runoff increases from pavement and other impermeable surfaces from subdivisions and businesses later built upstream, stressing the dam’s original design.13 Dams can also be weakened by the burrowing of rodents, the penetration of tree roots, or the rusting of metal pipes or structural components. Such risk factors can be managed only through thorough inspections and timely maintenance.

Today’s dams in a 1913 redux?
In 2009—eight years ago—the ASCE awarded Ohio’s dams a C, and in 2010 gave Indiana’s dams a D-. Neither state received a grade in 2017, but it is hard to believe that either state’s dams would be in better shape now than they were nearly a decade ago, given that both expend significantly less than the national average per dam on dam safety inspections and maintenance. According to ASDSO stats, in 2015 Ohio’s annual budget of $1.3 million for dam safety amounted to $876 per dam per year, and Indiana’s budget of $505,000 came to only $551 per year (the national average was $1,267 per dam per year).

A whopping majority of U.S. dams—nearly two-thirds—are privately owned, whether by corporations, homeowner associations, or individual property owners. Many cannot afford to maintain the dams.

One big problem in Ohio and Indiana as well as nationwide is that nearly two-thirds of dams are privately owned, and neither individuals nor homeowner associations have the resources to keep up with maintenance needs that get larger each year with continued neglect.

Another big problem, especially in Ohio, is that increased earthquakes and other seismic activity due to hydraulic fracturing for natural gas and high-pressure disposal of the highly saline and toxic wastewater in underground injection wells is endangering dams. For example, accelerated slippage, cracks, and separation of concrete in Mineral Ridge Dam in 2014 and 2016 led the board president of the Mahoning Valley Sanitary District—which owns the dam—to seek help from state lawmakers and oversight agencies. Slippage is instability in the slopes of the dam; it accounts for about 30 percent of dam failures.

Mineral Ridge Dam is a large earthen dam built in 1932. At its crest, the Class I dam is 3,480 feet long  and 60 feet high. It impounds the seven-mile-long Meander Reservoir in Mahoning County, one of the 15 biggest artificial lakes in Ohio (and bigger than any natural inland lake in the state). Credit: Google Earth; WKSU 

 Today’s dams in a 1913 redux?
As with much of the nation’s other infrastructure today, dams—and specifically high hazard potential dams—are concentrated in the eastern half of the nation, right under the footprint of the 1913 storm system whose high winds, tornadoes, and monumental flooding devastated significant parts of 15 states (see “Benchmarking ‘Extreme’” ).
Repeatedly throughout federal agency reports in FEMA’s most recent (fiscal years 2014–15) biennial report to Congress on the National Dam Safety Program (NDSP), the theme was lack of resources. The most recent NDSP report was blunt: “FEMA…strongly believes…that many Americans are living below structurally deficient high hazard dams; Americans are unaware of the risk; there is no plan in place to evacuate them to safety in the event of a failure; or there is a plan in place but they are not aware of it.”16 As daunting as the magnitude of investment to rehabilitate U.S. dams may be, CAP warned: “…the costs of inaction are exponentially higher and will likely not be measured only in dollars spent but, more importantly, in lives lost." 
In this context, McCampbell’s words in 1913 stand relevant and prophetic today for Ohio, Indiana, and other states afflicted by the record 1913 flood:

While [reservoirs and canals] were not responsible to a considerable degree for the recent flood they came so near to causing considerable additional damage that those who are interested in the protection of the life of the citizens of our commonwealth are much concerned and immediate steps should be taken in order to have an absolute guarantee against the loss of life and property from this source. …[I]n order to guard against serious catastrophies [sic] in the future a very careful examination of the reservoirs should be made and all defects remedied as soon as possible. [emphasis added]
©2017 Trudy E. Bell

Next time: Desperate Medicine

Selected references
1Gast, Cornelia A. “Breaking of the Dam,” in Historical Souvenir of the Fremont Flood, March 25–28, 1913, published by The Finch Studio,  Fremont, Ohio, 1913, pp. 35–37.
2Ballville Dam Project Final Environmental Impact Statement: Appendix A3. Ballville Dam Flood Storage Capacity Memorandum.  From Scott Peyton to Brian Elkington, July 24, 2012. Page 22 of the unnumbered 407-page PDF.
3McCampbell, E.  F. “Special Report on the Flood of March, 1913,” Monthly Bulletin, Ohio State Board of Health, May 1913, p.
4This section and the next one are largely drawn from my article “Brink of Failure? High Hazard Potential Dams,” The Bent 108(2): 8–13, Spring 2017.
5 U.S. Department of Homeland Security (DHS), Dams Sector: Estimating Loss of Life for Dam Failure Scenarios, Sept. 2011.
6 NCAR/UCAR , “Extreme Downpours Could Increase Fivefold Across Parts of the U.S.” Dec. 5, 2016.
7 ASDSO, “State and Federal Oversight of Dam Safety Must Be Improved,” Oct. 2016.
8 USACE, ASDSO, and U.S. Army Topographic Engineering Center, National Inventory of Dams Methodology: State and Federal Agency Manual, Version 4.0, April 2008, p. 13.
9 Gratitude is expressed to Rebecca M. Ragon of the USACE for providing statistics from the 2016 NID update.
 10 Amekudzi, Adjo A. , Rebecca Shelton, and Tim R. Bricker, “Infrastructure Rating Tool: Using Decision Support Tools to Enhance ASCE Infrastructure Report Card Process,” 2013 .
 11 Center for American Progress (CAP), Ensuring Public Safety by Investing in Our Nation’s Critical Dams and Levees, Sept. 2012.
 12 FEMA, Federal Guidelines for Inundation Mapping of Flood Risks Associated with Dam Incidents and Failures, P-946, July 2013.
 13 FEMA, Living With Dams: Know Your Risks, P-956, 2013. See also a different document with the same title by ASDSO, Living With Dams: Know Your Risks, April 2012.
14 FEMA, The National Dam Safety Program, Biennial Report to the United States Congress, Fiscal Years 2014–2015, P–1067, Aug. 3, 2016.
15 Axam, Shahid, and Qiren Li, “Tailings Dam Failures: A Review of the Last One Hundred Years,” Geotechnical News, 28:50–53, Dec.  2010.

Bell, Trudy E., The Great Dayton Flood of 1913, Arcadia Publishing, 2008. Picture book of nearly 200 images of the flood in Dayton, rescue efforts, recovery, and the construction of the Miami Conservancy District dry dams for flood control, including several pictures of Cox. (Author’s shameless marketing plug: Copies are available directly from me for the cover price of $21.99 plus $4.00 shipping, complete with inscription of your choice; for details, e-mail me), or order from the publisher.