On July 1, 1999, a backhoe punched through the earthen shoulder of the Edwards Dam in Augusta, Maine, and the Kennebec River — held back for 162 years by a 917-foot barrier of granite and timber — began to drain around it. Within hours, water that had been pooled since Martin Van Buren was president was moving again. Within weeks, striped bass were hunting bait in stretches of river they had been locked out of since 1837. Within a decade, alewife runs on the Kennebec had exploded by more than 200 percent, and Atlantic sturgeon were leaping clear of the water at Augusta each spring, a spectacle no living person had ever seen there.
The dam came down because the Federal Energy Regulatory Commission ruled, for the first time in its history, that the ecological cost of a hydroelectric dam outweighed its 3.5 megawatts of power. The result was the loudest starting gun for the modern American dam-removal movement, and a live experiment in how quickly a New England river can remember what it used to be.
The dam that held back a river for 162 years
The Edwards Dam was built in 1837 to power sawmills and, later, textile and paper mills in Augusta. It stretched nearly the width of the Kennebec, roughly 25 feet high, and it did what dams do: it created a warm, slow, silty impoundment where a cold, fast, gravel-bottomed river had been. It also blocked eleven species of sea-run fish from reaching about 17 miles of the lower Kennebec and the tributaries that fed into it.
Atlantic salmon, alewives, American shad, blueback herring, striped bass, rainbow smelt, American eel, sea lamprey, Atlantic and shortnose sturgeon, and tomcod had all used the river before the dam. After 1837, most of them could no longer complete their life cycles above Augusta. Salmon that had spawned in Maine’s interior for more than 10,000 years — since the retreat of the Laurentide ice sheet — were reduced to a hatchery-supported remnant.
By the 1990s, the dam was old, structurally tired, and producing a trivial amount of electricity — enough for perhaps a few thousand homes. When its federal license came up for renewal in 1993, a coalition of state agencies, tribes, and conservation groups argued that the fish were worth more than the wattage. In November 1997, FERC agreed.
What came back, and how fast
The response was immediate and, to biologists watching in real time, faintly disorienting. Striped bass moved past the old dam site within days. Alewives — small, silvery river herring that spawn in freshwater ponds and feed almost everything larger than themselves — surged upstream the first spring. Sturgeon followed.
The numbers eventually became the argument. As The Maine Monitor reported, river herring runs on the Kennebec increased by 228 percent after the Edwards removal — a figure NOAA Fisheries also cites — and by another 1,425 percent after the Fort Halifax Dam came down on the Sebasticook tributary in 2008. Sean Ledwin of Maine’s Department of Marine Resources has said the lower Kennebec now hosts the largest river herring run in North America and the highest eel abundance the state has ever measured.
Atlantic and shortnose sturgeon — armored, prehistoric-looking fish that can top eight feet — began spawning again above Augusta. Each spring, adult sturgeon breach clear of the surface near the city in a behavior nobody fully understands, a sight the river had not offered for six generations.
The salmon story is harder
The alewives and shad came back because the barrier was gone and their spawning habitat was directly upstream. Atlantic salmon are a different problem. They need to travel much farther — up the Kennebec, past three more dams, and into the cold, spring-fed Sandy River — to find the gravel beds and 60-degree water where their eggs can survive.
Today, roughly 1,000 to 1,200 wild Atlantic salmon return each year to the entire Gulf of Maine, according to NOAA figures cited in Maine reporting. Fewer than 100 adults, on average, are captured at the Lockwood Dam near Waterville each year and trucked around the three upstream dams to the Sandy. The species is federally listed as endangered.
The Edwards removal opened the door. It did not open the whole hallway. That is why, more than two decades later, the fight over the Kennebec has moved 40 miles upstream.
A $168 million sequel
In 2025, The Nature Conservancy announced a landmark agreement to buy the four remaining lower-Kennebec hydroelectric dams from Brookfield Renewable for a total investment of $168 million, with the eventual goal of removing them. The dams — Lockwood, Hydro Kennebec, Shawmut, and Weston — sit between Waterville and Skowhegan and together generate 46 megawatts, about 6 percent of Maine’s hydroelectric capacity and roughly 0.43 percent of the state’s electricity.
Alex Mas, associate state director for The Nature Conservancy in Maine, has described the goal as creating a free-flowing lower Kennebec that would restore ecology and strengthen the economy. The plan is to keep the dams operating for five to ten years while a new nonprofit works through federal decommissioning and negotiates a technical solution with Sappi North America, whose Somerset paper mill in Skowhegan relies on water impounded by the Shawmut Dam.
John Burrows of the Atlantic Salmon Federation has said that opening the river could enable annual adult salmon returns to jump from fewer than 100 to several hundred in a short timeframe, and that combined with two additional tributary projects, dam removal could open 825 miles of stream habitat back to the Gulf of Maine.
Why old dams are coming down across New England
The Kennebec is the flagship, but it is not alone. New England has thousands of small dams left over from the 19th-century mill economy — grist, saw, textile, and paper mills that stopped running decades ago but left their impoundments in place. Many are unsafe, unproductive, and blocking rivers for no economic reason at all.
As The Christian Science Monitor documented in July 2025, New England towns have been methodically removing barriers on smaller waterways — among them the Bridge Street and East Elm Street dams on the Royal River in Yarmouth, taken up by the town alongside the national group American Rivers — to restore alewife runs and reconnect spawning ponds to the sea. In the decades since the Edwards project, dam removals have accelerated nationwide: more than 2,200 have come down across the United States since record-keeping began in 1912, with a record pace in recent years and New England among the most active regions.
The pattern is consistent. Take out a low-value dam. Watch a river herring run rebuild within three or four spawning cycles. Watch eagles and osprey and seals show up to eat them. Watch the whole food web thicken.
The Penobscot precedent
The other river worth watching runs east of the Kennebec. On the Penobscot, a 16-year, $60 million project removed two dams outright, built a stream-like bypass channel around a third, and installed a new fish lift at a fourth. The result was renewed access to roughly 2,000 miles of river habitat — the largest river-restoration project ever attempted in the Northeast.
A 2023 paper in Frontiers in Ecology and Evolution laid out seven interconnected ways dams still burden migratory fish on the Penobscot even after the restoration — from injury and mortality at turbines to blocked habitat access, migration delays, and the warm, slack water of impoundments — and concluded that fish passage remains among the hardest problems to solve. Every dam a salmon must clear costs the run a fraction of its adults, and the losses compound. Only about half of salmon even find the fish lift at the lowest Kennebec dam, state biologists estimate; apply a survival rate like that across four barriers and barely one fish in sixteen reaches the top.
What a river remembers
The Edwards Dam removal became a national reference point partly because of how quickly the Kennebec responded. Ecologists sometimes talk about “ecological memory” — the idea that a system carries the imprint of its former structure, and that if you remove the disturbance, it can, within limits, reassemble. Terra Daily has explored this pattern before in South Korea’s post-war reforestation, where a country stripped nearly bare in 1953 grew back to roughly two-thirds forest cover within a single lifetime.
Rivers, if anything, respond faster than forests. Alewives home to natal ponds using olfactory cues. If the pond is still there and the route is clear, the run rebuilds. Sturgeon return to gravel bars they haven’t touched in a century. Sea lampreys — often reviled but ecologically essential — dig nests that Atlantic salmon later use as their own spawning sites, a co-evolved relationship that took millennia to develop and years to reactivate.
The economics that finally shifted
What broke the stalemate on the Kennebec was not sentiment. It was arithmetic. As Mainebiz documented in a 20-year retrospective, the Edwards removal transformed Augusta’s waterfront from an industrial back-lot into a recreation corridor. Property values along the restored river rose. Kayak outfitters, restaurants, and trail systems moved in. The city gained more in tax base than it lost in hydropower.
Kristina Cannon of Main Street Skowhegan has said that a planned whitewater park in downtown Skowhegan alone could generate $625,000 in annual tax benefits — more than the roughly $500,000 in combined property tax revenue the four upstream dams currently pay to Waterville, Skowhegan, Winslow, and Fairfield.
Meanwhile, the 46 megawatts the four dams produce are being outpaced by other sources. Maine is projected to add 1,200 megawatts of solar generation over five years — roughly 26 times the capacity of the four Kennebec dams combined.
What still has to happen
The Nature Conservancy deal does not remove the dams tomorrow. Federal decommissioning takes years. Sappi’s mill needs a water-supply engineering solution. Two Wabanaki tribes with deep cultural ties to the Kennebec Valley — descendants of the community attacked at Norridgewock by English soldiers in 1724 — are advising the process. A Penobscot Nation citizen and University of Maine anthropologist on The Nature Conservancy’s board has described how opening the river creates deep connectivity to ancestral places.
The four upstream dams still stand. Salmon still ride in trucks. Fewer than a hundred wild adults still find their way to the Sandy River each summer, where volunteers plant hatchery eggs in the gravel and hope that some fraction of the fry that emerge will survive the trip downstream, past four sets of turbines, out to sea, up to Greenland, and back.
But the Edwards precedent is doing quiet work in every negotiation. It proved something concrete: a New England river held closed for 162 years can be reopened, and the fish — the ones that survived the long wait in the ocean and in tiny remnant runs — will find the door.
On a spring morning in Augusta now, a shortnose sturgeon can breach into the air above a stretch of Kennebec where, in 1837, Martin Van Buren’s America was busy sealing the river shut. The dam is gone. The fish, some of them, are older than the removal itself. They have been waiting.
