All notes and picture credits are presented in the final part of this series.
Demise of the Western Reservoir: Replacement, Breach, and Abandonment
The Western Reservoir and its dam were already outdated even as the crews finished construction. US civil engineers had, in the 1830s, begun dealing with unreliable canals by eliminating them using improved railroad technology. Work crews were laying track across both New York and Maryland by the 1840s, and the Pennsylvania General Assembly once again followed their lead. In 1846, the Assembly chartered the Pennsylvania Railroad to link Philadelphia to Pittsburgh via a trans-Appalachian route. The Railroad soon became known as the PRR or “Pennsy” and began continuous cross-state rail service late in 1852 after six years of surveying, design, and construction (Burgess and Kennedy 1949).
Initially, the PRR used the Main Line’s Portage Railroad to cross the Allegheny Mountains. Early in 1854, however, the Pennsy opened a new, permanent route through the Alleghenies which included both a massive switchback – quickly named the “Horseshoe Curve” – up the eastern side of the mountains and two tunnels below the crest of the mountains. The route remains in daily use almost 170 years after entering service. Traffic through the Appalachians, and especially over the Alleghenies, rapidly switched to the PRR. The days of the Main Line were over (Burgess and Kennedy 1949).
The Main Line went almost overnight from one of Pennsylvania’s largest assets to, perhaps, its largest liability. However, the PRR recognized that the real estate of the Main Line Johnstown Flood could serve as rights-of-way for future tracks. In 1857, the Pennsy purchased the Main Line from the Commonwealth for $7.5M. Almost immediately, the PRR slashed oversight of the Western Reservoir and dam to periodic check-ins by local employees. Some residents living near the now-unused reservoir used it for fishing and boating. Others stole lead from the seals of the discharge pipes and sold it for scrap (Coleman 2018, McCullough 1968, Shappee 1940).
Years of neglect and lead theft eventually caused leaks to form in the dam at the old Western Reservoir. In mid-July of 1862, the discharge culvert partially collapsed due to the leaks. A week later, PRR watchman Joseph Leckey was doing a routine check of the dam when he noticed muddy water flowing out of the remnants of the culvert. 21st-century geotechnical engineers would have recognized this muddy water as a sign of ongoing piping within the dam, although Leckey may not have. He certainly knew, though, that the turbid leakage indicated the potential for a major breach. Leckey rowed to the control tower and opened the discharge pipes to avert catastrophe, then rode to the nearest PRR telegraph station, where he and the operator warned Johnstown of the danger. Per Darcy’s Law, first quantified in France a few years earlier, the relatively low – less than 50 feet – water depth within the reservoir slowed the leaks. Yet they worsened, perhaps due in part to Leckey’s actions, and the dam finally failed about 12 hours after his timely observation (Coleman 2018).
The waters of the old Western Reservoir roared through the breach and down the South Fork of the Little Conemaugh River. As they did, they demolished a house, a sawmill, and hundreds of feet of PRR track. Yet Leckey’s timely warning about the pending breach, along with its slow formation, kept damage fairly minimal, and prevented fatalities. The media and the public, fixated on the ongoing US Civil War, paid little attention to the breach. In mid-1862, many young men from Cambria County were serving in the Union forces which fought violently and often with Confederate troops that summer in engagements such as the Peninsula Campaign, the Second Battle of Bull Run, and the Battle of Antietam (Coleman 2018, McCullough 1968).
The PRR had already been planning to abandon the old Western Reservoir. The Pennsy thus also cared little about the July 1862 breach, and it abandoned the former Main Line reservoirs and canals in 1864. Yet the PRR kept the old reservoir property for another decade. A 200-foot-long breach shaped like an inverted wedge, with its point above the ruined drainage culvert, now separated the remnants of the former dam. Once more, the remnants were exposed to the elements. Locals fished in the few deep pools remaining on the old reservoir bed and let their cattle graze there. In 1875, the Pennsy sold the site to John Reilly, a PRR employee who had just won a seat in the US House of Representatives. The Pennsy surely valued being represented favorably in Washington, DC, and Reilly bought the land fairly cheaply. It was during his ownership that the old control tower burned down (Coleman 2018, Hanna 2021, Unrau 1979).
Benjamin Ruff and The South Fork Fishing and Hunting Club
Congressman Reilly lost his re-election bid in 1878 and began looking to sell the Western Reservoir property. He soon found a willing buyer. Pittsburgh mogul Benjamin Ruff and some rich peers in the city incorporated the South Fork Fishing and Hunting Club early in 1879 and sought to establish a country resort for Pittsburgh elites. Ruff, now Club President, saw that the old reservoir site would work well. That fall, the Club agreed to buy the property from Reilly. The members planned to re-impound the reservoir, to be renamed Lake Conemaugh, as a leisure lake by rebuilding the dam, to be named the South Fork Dam. Ruff then hired a contractor for the reconstruction of the dam. Yet Reilly still owned the property, and Ruff’s crew most likely began work only with Reilly’s permission (Coleman 2018, Hanna 2021, McCullough 1968).
Qualified civil engineers had been involved throughout the design and construction of the Western Reservoir and its dam. The rebuilding of the South Fork Dam and re-impoundment of Lake Conemaugh, by contrast, involved little if any such input. Instead, Benjamin Ruff directed the work himself, although he had no experience or training as a civil engineer. Ruff had worked as a contractor for railroad embankments and tunnels, and perhaps he believed that this experience qualified him to supervise the rebuilding of the dam. However, civil engineers – especially those with field experience – appreciate that contractors and civil engineers often have different priorities and distinct types of expertise. Ruff also probably let money influence his decision to direct the project himself. Costs have always been a concern in construction and were likely even more worrisome for the Club, which still mostly existed on paper (Coleman 2018, Hanna 2021, McCullough 1968).
Rebuilding the South Fork Dam and Re-Impounding Lake Conemaugh
The 50-man crew began rebuilding the South Fork Dam late in 1879. As they did, Ruff made seemingly every decision based on finances. It remains unclear whose finances he was mainly considering, since Reilly still owned the property. What is clearer is that no qualified civil engineer in 1879 would have recommended the repairs which Ruff’s laborers began performing at the dam that fall. The crew first removed the discharge pipes, which were sold for scrap. Now, no one could control the lake levels either to repair the dam or to prepare for incoming storms. The laborers also lowered the crest of the dam by about 3 feet, most likely to reuse material to begin filling the 1862 breach. Whatever the reason behind the decision, no one appears to have considered the disastrous consequences of lowering the dam. Reducing the available freeboard by 3 feet had also significantly lowered the peak discharge of the main spillway and had reduced the auxiliary spillway to a mere drainage swale. A storm on Christmas Day washed out everything the crew had dumped into the 1862 breach, including the material presumably salvaged from the crest (Coleman 2018, Hanna 2021, McCullough 1968).
After the Christmas 1879 storm, the Club paused its dam reconstruction until the summer of 1880 to avoid further washouts. The Club had formally purchased the site by then, which had put Ruff fully in command of the dam rebuild. As the contractors resumed the project, their handiwork at the dam indicated little to no experience in dam reconstruction. The crew restarted work by driving hemlock sheet piles at the upstream end of what remained of the drainage culvert in an effort to block off the remnants. The double layer of sheet piles was less than 6 inches thick and was apparently poorly-driven. The laborers then dumped boulders, cobbles, and shale into the 1862 breach, along with clay which may actually have been plastic soil taken from the spoils of local coal mines. They neither puddled nor properly compacted these materials, which suggests that their work failed to meet the contemporary standard of care for placing material during dam construction (Kaktins et al. 2013, Wellington and Burt 1889 B).
The crew kept haphazardly filling the old breach, and Lake Conemaugh slowly rose behind the dam. However, the South Fork of the Little Conemaugh was not diverted during the reconstruction and flowed through the breach, saturating the fill material and reducing its shear strength. The laborers did cover the fill at the upstream face of the breach with brush and hay to curtail the leakage, although it remains unknown how much this technique improved the situation. The crew also blocked the remaining spillway, across which a wooden bridge had long stood, by installing a fish screen of iron bars between the bents of the bridge. The screen reduced the maximum discharge of the spillway and also created a potential clogging hazard for drainage during storms. Notably, the laborers never removed the trees and vegetation which had grown on the dam remnants over the decades since the dam had first been built. This made fully assessing the condition of the dam nearly impossible (Kaktins et al. 2013, NPS 2022 D, Wellington and Burt 1889 B).
Word spread as Benjamin Ruff’s laborers reconstructed the South Fork Dam that they were doing low-quality work. One civil engineer who visited the site in 1880 remarked that he had seen leaks near the old culvert before the breach had been filled even halfway. Yet he also noted that Ruff’s foreman had brushed off his concerns. News of the shoddy rebuild eventually reached Daniel Morrell, head of the Cambria Iron Company in Johnstown, whose concerns could not be brushed aside so easily. Morrell had built Cambria Iron into a leading US steel production firm that, by 1880, had about $1.45B (2023 USD) invested in its facilities in and around Johnstown. Some of his incentives were also civic. Cambria Iron was, by Gilded Age standards, a fairly benevolent employer that ran a hospital, an affordably priced store, a night school, and a public library within Johnstown. Lastly, Cambria Iron employed many trained engineers. One was Morrell’s top lieutenant, John Fulton, who also had experience in geology. Thus, Morrell, unlike Ruff, appreciated the value of competent engineering. Morrell soon decided that Cambria Iron should check in on the rebuilding of the South Fork Dam. He sent Fulton to meet with several Club officers, observe everything he could, and report back on his findings (McCullough 1968, Unrau 1979, Webster 2023).
In November 1880, Fulton visited the South Fork Dam site and toured the project with some Club officers. Benjamin Ruff was not on the tour, although civil engineer N.M. McDowell was. (It is unclear for how long the Club had retained McDowell prior to the tour.) Fulton did not observe the lowering of the dam, which might have escaped any first-time visitor. Yet he did see the haphazard dumping of fill into the 1862 breach and wrote to Morrell, “It did not appear to me that this work was being done in a careful and substantial manner, or with the care demanded in a large structure of this kind.” Fulton then noted the lack of discharge pipes in the rebuilt dam and made a startling prediction. “When the full head of sixty feet [in Lake Conemaugh] is reached,” Fulton stated, “it appears to me to be only a question of time until the former cutting [breach] is repeated. Should this break be made during a season of flood, it is evident that considerable damage would ensue along the line of the [Little] Conemaugh.” Fulton concluded that the dam should be rebuilt properly with both a new discharge pipe and large, heavy riprap on its upstream face. Fulton’s report so alarmed Daniel Morrell that Morrell forwarded it straight to Ruff (McCullough 1968, Unrau 1979).
Benjamin Ruff wrote back to Morrell quickly and rather curtly. He began by noting some minor mistakes Fulton had made, such as having the wrong name for the Club, then denied that the rebuilt dam was leaking. Next, Ruff imperiously dismissed Fulton’s suggested repairs by writing, “We consider his conclusions as to our only safe course [of action] of no more value than his other assertions.” Ruff then referred Morrell to an attached report by an unnamed civil engineer working for the Club, most likely McDowell, which apparently does not survive. Ruff signed off by snapping, “You and your people are in no danger from our enterprise” (McCullough 1968, Unrau 1979).
Daniel Morrell replied to Ruff a few weeks later. Morrell acknowledged Fulton’s minor errors, but also stated, “[I] think that his conclusions in the main were correct.” Furthermore, Morrell – now speaking for Cambria Iron – wrote, “We must protest against the erection of a dam ... that will be a perpetual menace to the lives and the property of those residing in the upper valley of the [Little] Conemaugh, from its insecure construction.” Morrell also fired back at Ruff and engineer McDowell, pointedly noting, “I think you will find it necessary to provide an outlet pipe or gate before any engineer could pronounce the job a safe one.” Yet Morrell closed his letter magnanimously. Cambria Iron, Morrell stated, “would contribute liberally toward making the dam absolutely safe” if Ruff rebuilt it properly. Morrell was now speaking the language of business in which Ruff was fluent and was suggesting that repairing the dam correctly would be both safe and economically sound. Ruff apparently never replied to Morrell’s second outreach. However, John Fulton kept a copy of his report and of all of the correspondence between Morrell and Ruff (McCullough 1968, Unrau 1979).
As 1881 began, Ruff’s crew finished reconstructing the South Fork Dam. The leaks persisted, though, and a winter storm caused another minor breach. The Club then quietly validated Fulton’s concerns by asking engineer McDowell how to patch the leaks. McDowell, per the contemporary standard of care, recommended capping each leak using hay, manure, and brush and then covering each cap using clay. Per his recollection, this solution stopped the leaks. Finally, Ruff’s crew finished the rebuild by installing a V-shaped log boom pointing into Lake Conemaugh to keep large debris from blocking the screens. This action suggests that the Club may have harbored concerns over the spillway screens after all. The South Fork Dam was at last finished in June of 1881. The Club then impounded Lake Conemaugh and officially opened for business. One of its members, surprisingly, was Daniel Morrell, who had joined partially to monitor the dam. The Club had spent about $495,000 (2023 USD) to rebuild the dam (Coleman 2018, McCullough 1968, Unrau 1979, Webster 2023).
The rebuild of the South Fork Dam and the re-impoundment of Lake Conemaugh by Ruff’s crew had clearly not met the contemporary standard of care for dam construction, or even the standard from when the dam had first been built. Ruff’s decisions to lower the crest of the dam and to remove the discharge outlet may have been his most egregious mistakes. The storage-elevation curve for the reservoir and lake constructed by the UPJ researchers makes the tragic ramifications of those decisions crystal clear. The team used the curve to estimate the capacity of the Western Reservoir and the peak discharges of the spillways and discharge outlet at its dam, then performed identical calculations for the rebuilt dam and re-impounded lake (Coleman 2018).
ake Conemaugh and the South Fork Dam, 1880s. Note the poorly rebuilt breach (1,6), the main spillway with fish screens and log boom (2), the lack of a control tower (3), the abandoned discharge culvert (4), and the lowered crest (5). Source: NPS (2022 D)</em>" data-entity-type="file" data-entity-uuid="4c3d06c9-0a64-43aa-8bb0-fc3eb3d3e8e9" src="/sites/default/files/inline-images/Figure%2015%20South%20Fork%20Dam%20as%20Completed.jpg" />
Ruff’s crew had lowered the dam by 3 feet, less than 5 percent of its original height of 72 feet. Yet, per the UPJ team, this decision gave Lake Conemaugh a capacity of about 514M gallons, a decrease of around 10 percent compared to the 574M-gallon capacity of the Western Reservoir. The difference in peak discharges between the two dams was even greater, given the removal of the auxiliary spillway, the reduction in peak discharge of the remaining (main) spillway, and the removal of the discharge outlet. Ruff’s crew had reconstructed the South Fork Dam to have a peak discharge of 3,050 CFS, a 56 percent reduction from the 6,950-CFS peak discharge of the original dam. Thus, the lake had 10 percent less capacity than the reservoir with which to handle floods, and its dam could not discharge its contents even half as quickly as the old dam could discharge those of the old reservoir. Nor could the lake, unlike the reservoir, be lowered before severe weather. Ruff’s decisions during the rebuild to lower the dam and to remove its discharge outlet had clearly been disastrous (Coleman 2018).
Operation of the South Fork Fishing and Hunting Club
The South Fork Fishing and Hunting Club filled Lake Conemaugh to its maximum safe depth immediately after completing the South Fork Dam, unlike the original engineers of the Western Reservoir. Citizens along the Little Conemaugh River became concerned about the dam almost right away. A rainstorm in June of 1881 just weeks after the Club’s opening caused extensive flooding in the area and terrified the citizens of Johnstown, who feared that the dam might breach. The Cambria Iron Company sent two engineers to the lake to evaluate the situation, and their report that all was well helped calm down the valley residents. Notably, however, the engineers seem to have overlooked how the storm had filled the lake to the point where the dam had just two feet of freeboard (Coleman 2018, McCullough 1968, Unrau 1979).
The June 1881 crisis at Lake Conemaugh also attracted the attention of Club member Robert Pitcairn, a PRR executive. He knew Benjamin Ruff well and would later describe Ruff as “better than any engineer”. Yet Pitcairn chose on this occasion to take several civil engineers from the Pennsy on an inspection visit to the South Fork Dam. The group members were all concerned by several apparent leaks near the abutments of the dam. Ruff, however, was accompanying the group and reassured Pitcairn and his men that these were actually springs. 21st-century geotechnical engineers might, given how full the lake was, find Ruff’s explanation unconvincing. Yet Pitcairn would later claim that Ruff’s explanation had addressed his concerns. Pitcairn remained wary enough, though, that he asked a local business connection of his to alert him if trouble ever arose at the South Fork Dam (Coleman 2018, McCullough 1968, Unrau 1979).
Ultimately, though, the storms subsided, the level of Lake Conemaugh gradually returned to normal, and local fears about the South Fork Dam disappeared. Throughout the decade, some Johnstown residents would briefly panic after nearly every major regional storm over whether the dam might breach. Inevitably, though, the flooding would subside, and so would the fears. As the dam survived each storm, locals slowly grew numbed by the false alarms. So did area engineers, who seem to hardly have visited the dam after 1881. The prospect of the South Fork Dam breaching had, by the late 1880s, almost become a running joke among those who lived and worked along the Little Conemaugh River. Eyewitness Victor Heiser, then a teenager in Johnstown, would long afterwards summarize their cavalier sentiment: “Sometime, that dam will give way, but it won’t ever happen to us” (McCullough 1968, Unrau 1979).
The opinions and conclusions expressed in this article are those of the author only and do not necessarily represent the views of ASCE or the Geo-Institute.