In the fall of 2015, a smallmouth bass swam past a stone bridge in the suburbs of Pittsburgh and entered a stretch of Little Sewickley Creek that no bass had reached since the 1920s. The fish was not released there and was not chased into the area. It moved upstream on its own into a shallow section of the creek where an 87-year-old concrete barrier had stood just four months earlier.
Little Sewickley Creek is a high-quality coldwater fishery and the only stream in Allegheny County with a naturally reproducing brown trout population. The creek runs approximately 18 miles through some of the most developed areas of western Pennsylvania before joining the Ohio River. By most state standards, the stream was already considered healthy. Yet for most of the 20th century, the upper half of one of the county’s cleanest creeks contained barely a quarter of the fish species found downstream.
The barrier in question was a low-head dam built in 1928. A private estate on the hill above the creek wanted water and something pleasant to look at, so a concrete wall a few feet high was placed across the stream near an old stone bridge. According to Eric Chapman of the Western Pennsylvania Conservancy, who later spent years studying the dam’s impact, its original purpose was simple: it was built as a swimming pool. It was not intended for flood control, power generation, or municipal water supply. A wealthy family wanted a place to swim and a small waterfall to view from the bridge. The course of one of the region’s healthiest streams was altered to serve that wish for the next 87 years.
Structures like this are called low-head dams, and they make up a large share of Pennsylvania’s dams. Lisa Hollingsworth-Segedy, director of river restoration at American Rivers, estimates they account for 75 to 80 percent of the state’s total. They are short, inexpensive, and were built for reasons that seemed sensible at the time before those reasons faded and the concrete remained.
By the 1990s, no one was swimming behind the Woodland Dam. The estate that built it no longer needed it. The structure was cracked, its foundations were eroding, and it no longer held anything anyone wanted to retain. It had become, functionally, a piece of abandoned construction in the middle of a trout stream.
Pennsylvania’s Department of Environmental Protection classifies Little Sewickley Creek as high quality, one of the state’s strongest ratings for water chemistry and ecological value. This rating is unusual in any location, and it is especially notable here because the creek runs through Pittsburgh suburbs in a region whose waterways spent a century absorbing the effects of steel and coal production. The reason the creek remained healthy is its forested watershed, which keeps the shallow water cool enough in summer for brown trout to reproduce without hatchery support. No other stream in Allegheny County does this on its own.
Between 2009 and 2012, Chapman conducted fish surveys on the creek with students from nearby Quaker Valley High School. Over three years of wading the same waters with nets and identification keys above and below the dam, the results were clear. Downstream of the dam, survey teams counted more than 29 species. Upstream, they found only seven. The two numbers were separated by just a few dozen feet on the same stream on the same day, in water with the same temperature and chemistry. There was no pollution upstream, and the water was not too warm. The habitat was there, the food was there, and the fish that should have used it were blocked by a concrete wall roughly four feet high that they had no way to climb.
The seven species that persisted above the dam were those that could complete their entire life cycle within that isolated stretch. Everything else in the downstream count belonged to the Ohio River system, a watershed that had once flowed continuously from the creek’s headwaters through Ohio and toward the Mississippi. The missing species included spotfin shiners, rainbow darters, greenside darters, longnose dace, bluntnose minnows, and spotted bass — species that move seasonally, follow temperature and spawning gravel, and colonize new water when it becomes available. Little Sewickley Creek had been cut off from that system since 1928. Below the dam, the creek was part of a continental river network. Biologically, the section above it was closer to a pond with a current.
Chapman now had a warning sign in his data. He also had a problem: the Woodland Dam was private property, structurally complicated, and located in a neighborhood where residents had long considered the small waterfall it created one of the more attractive features of their daily commute.
Jessie Thomas-Blate, director of river restoration at American Rivers, compares a dam in a river to a blood clot in an artery. The body needs blood to flow freely, and the same applies to rivers and the water and nutrients that move through them. A dam affects a waterway in four ways at once. The obvious one is blockage: fish cannot pass, ranges collapse, populations fragment, and genetic isolation begins upstream. The second issue is temperature: water pooled behind a low wall becomes still and absorbs sunlight, a direct threat to a stream whose ecological identity depends on staying cool. The third is sediment: a dam traps the gravel and sand that must keep moving downstream. Upstream, retained material silts over spawning areas. Downstream, the river arrives short of sediment and begins cutting into its banks in search of what it has lost. The fourth is nutrient transport, the slow conveyor of organic matter that feeds insects, which feed fish, which feed everything above them. Cutting that conveyor weakens the entire food chain.
All four factors were present in a single stream that the state rated as high quality, and the result was exactly what Chapman’s students had measured: perfectly clean water with almost nothing living in it. This combination exposes a blind spot in how river health is measured. Water quality monitoring answers a chemical question. It samples temperature, dissolved oxygen, nutrients, and pollutants, and Little Sewickley Creek passed all those tests equally above and below the dam. What no chemical test panel measures is connectivity — whether an organism living in the water can physically reach the rest of the system it evolved within. A stream can receive the highest rating on a state scale and still be functionally severed. For 87 years, that was precisely the situation on Little Sewickley Creek, and the rating never once indicated it.
There is a second reason Pennsylvania has spent the past two decades removing these structures, and it has nothing to do with fish. Low-head dams have a particular nickname in the state. Ryan Walt of the Pennsylvania Fish and Boat Commission explained its origin. They are called drowning machines with good reason. Water flows over a low wall and curls back on itself at the base, creating a circular current. A person who falls over the wall is pulled into that loop and circulates inside it. The air-filled water at the bottom provides almost no buoyancy, and even skilled swimmers cannot escape. From the bank, none of this is visible. What a person sees is a smooth layer of water, a pretty white curtain, and a drop that looks ankle-high.
Approximately one in four people who die at low-head dams entered the water to try to rescue someone else. This statistic explains the pattern of failure. The first person falls in and cannot get out, and the water gives the second person no visible warning that the same fate awaits. Pennsylvania learned the cost of dam neglect earlier and more harshly than almost any other state. In 1889, the South Fork Dam on the Little Conemaugh River failed after years of poor maintenance, sending a 36-foot wall of water toward Johnstown at 40 miles per hour. More than 2,200 people died.
Rich Riegner, who runs the dam safety division at the state Department of Environmental Protection, describes the prevailing mindset: the dam has been there for a hundred years, and people tell him it will be there for another hundred. The state currently tracks about 3,380 dams. Of those, 744 are classified as high-hazard, meaning their failure could cause significant property damage and loss of life. Most are privately owned, more than half are over 50 years old, and funds for their repair are not available.
Pennsylvania has roughly 86,000 miles of rivers and streams, one of the densest freshwater networks in the eastern United States, with about 3,400 dams inside it. That averages out to roughly one barrier every 25 miles of waterway across the entire state. The vast majority are small, privately owned, and invisible to anyone not standing directly beside them. The Woodland Dam was not an exceptional case; it was typical of the general condition.
The character of this inventory has changed over the past decade. Among removals completed nationwide in 2025, the environment was the stated reason in 59 cases. Dilapidated or failing structures came second at 37, and safety came third at 20. Communities have largely stopped waiting for a dam to begin deteriorating before removing it. They are now removing structurally sound dams because of their impact on the river, a different standard from what prevailed over the past hundred years. The Woodland Dam sat on the wrong side of that standard twice over: it was environmentally damaging and it was deteriorating.
While Chapman was compiling fish statistics on a small creek in Allegheny County, a much larger inventory was underway across Pennsylvania. American Rivers, state agencies, regional conservancies, and local watershed groups were cataloguing obsolete structures, ranking them by hazard, removal cost, and environmental return. The Woodland Dam was small enough to disappear inside such a list. It was four feet high on a creek almost no one outside the town had heard of. Chapman contacted Hollingsworth-Segedy, telling her he had found a truly complicated, terrible dam and asking if she would like to work on it. She agreed.
Hollingsworth-Segedy has been involved in this field for years. Since joining American Rivers in 2008, she has worked on more than 100 dam removal projects in Pennsylvania and West Virginia and has helped reconnect more than 5,000 miles of free-flowing river. She has also created what may be the world’s only puppet show about dam removal, which she uses to explain to skeptical neighborhoods why an attractive little waterfall is a hazard rather than an amenity.
When the Woodland Dam was removed, many residents were angry, and their objection was aesthetic. They were losing the waterfall. The neighborhood organized a defense of it. They were defending the mechanism causing the damage, according to Hollingsworth-Segedy, who said people get genuinely angry when this happens, and then they realize things are not as bad as feared.
The removal itself was almost anticlimactic. Work began on July 28, 2015, and was completed by July 31 — within four days. The total cost was $20,000, funded by a grant from the Allegheny Conservation District along with money from American Rivers. The numbers explain why this is happening across the country: 87 years of ecological blockage removed in four days at the price of a used car.
The American Society of Civil Engineers rates US dam infrastructure with a grade of D, and the Association of State Dam Safety Officials has estimated the national repair bill for deficient dams at $66 billion. Removal is usually three to five times cheaper than repairing a structure and maintaining it indefinitely, and unlike repair, removal never needs to be done a second time.
The stream responded quickly. Within days of removal, invertebrates moved in. Caddisfly and mayfly larvae recolonized gravel that had been buried under retained silt for decades. Once this layer of the food web was reestablished, the animals that feed on them followed within weeks. The sequence repeats consistently enough across projects that restoration staff now treat it as a marker of progress.
Two months after the concrete was removed, in September 2015, spotfin shiners and rainbow darters were recorded half a mile upstream of the former dam site. Neither species had been documented in those waters in living memory. There was no relocation program, no hatchery truck, no stocking permit. There was simply an opening where a wall had been. Serena McClain, who runs the national dam removal program at American Rivers, said it is remarkable how quickly a river recovers once it returns to its natural free-flowing state. In some cases, fish return upstream within moments.
Then came the smallmouth bass. In a river the size of the Ohio, smallmouth bass is an ordinary medium-sized fish. In a shallow creek like Little Sewickley, it is the top predator. Its arrival was not just another name on a species list. Apex predators restructure everything below them, stabilizing smaller fish populations and large invertebrates, redistributing pressure across the food web, and pushing the system toward a complexity that makes it resilient to the next disturbance. The creek had been without its top predator since 1928. It took about four months to restore one.
Fifteen months after the removal, surveys were conducted again, across a full mile upstream. Nine new species had established themselves. The count rose from seven to sixteen, more than doubling the entire upper stream’s diversity. Alongside the shiners and darters, the returning species included greenside darters, longnose dace, bluntnose minnows, bluegill, and spotted bass. These species do not merely fill in survey forms. A more integrated fish community speeds up nutrient cycling and improves the stream’s ability to process organic matter, meaning gains in biodiversity and water quality reinforce each other rather than compete.
This timeline matters. In 2025, a Cornell University study concluded that dam removal rehabilitates aquatic ecosystems within about three years. Jeremiah Dietrich, senior aquatic ecologist at the New York State Water Resources Institute, said the research shows nature’s ability to recover from degraded conditions. Little Sewickley Creek accomplished most of its recovery within 15 months. The recovery was neither slow nor fragile. It had simply been waiting.
Pennsylvania has since turned this effort into a program of action. The state has removed 433 dams since 1912, more than any other state in the country. In 2024, the state removed 27 dams, nearly three times the total of Michigan, which ranked second with 10. In 2025, it removed 14 more, leading the country again.
The state has not limited itself to small four-foot walls. Between 2018 and 2019, crews removed the 83-foot-tall, 550-foot-long Gunter Valley Dam on Trout Run, the tallest dam ever removed in Pennsylvania. On September 18, 2023, the Oakland Dam was removed in what became the largest dam removal in state history, reconnecting more than 250 miles of habitat. In August 2025, two low-head dams were removed from the Juniata River, reopening more than 156 miles of connected waterways in a single project.
The state does not carry this burden alone. The Wildlands Conservancy handled eight of the 27 removals in 2024 by itself. The US Fish and Wildlife Service, regional watershed associations, and the Department of Environmental Protection all contribute to the same effort. There is now a regional goal attached to this work: the 2025 Chesapeake Bay Watershed Agreement requires signatory states to open 150 miles of rivers annually for fish migration through dam and barrier removal. By virtue of its location on the Susquehanna River, Pennsylvania carries a disproportionate share of this commitment.
Nationally, 2025 was the largest year on record. One hundred dams were removed across 30 states, reconnecting more than 4,800 miles of rivers — the most distance restored in a single year in the United States. More than 2,300 dam removals have been documented nationwide since 1912. American Rivers has set a goal of removing 30,000 dams by 2050, out of the estimated hundreds of thousands nationwide. Most of those structures will look exactly like the Woodland Dam: small, private, forgotten, and inexpensive to remove. Most will be defended by people who love the waterfall.
That is the part of this story worth holding onto. The residents of Sewickley Heights were not wrong about what they were losing. There had been a waterfall, and it was genuinely attractive, and for 87 years it was the most prominent feature of that part of the creek. What they could not see — because the water was clean, the trees were healthy, and nothing looked broken — was that the waterfall was the wound. Its removal took four days of work. Two months later, a predator that had been blocked since 1928 swam through the gap where the dam had been. Fifteen months after that, the upper portion of the creek contained twice as much life as it had for nearly a century. The world did not end. It came back to life.
