A juvenile white sturgeon drops into rushing water above a spinning metal wheel.
It has no choice but to go through.
On the other side, a tank waits to show whether it made it.
Safer fish passage at dams is usually chased with screens, bypasses and spill.
A research team in California went at it from the other end instead, and changed the blades.
What decided whether a fish lived or died was hiding in the edge of a single blade.
What makes one blade edge deadlier than another
Conventional turbine runners carry thin, straight leading edges. That profile is efficient at pulling energy from moving water, but when it meets a soft body roughly the scale of the gap it is passing through, the force of contact concentrates along a narrow line rather than spreading across a surface. The result is a blade that behaves more like a blunt knife than a paddle.
The alternative tested alongside it carried thick, slanted leading edges, angled back from the hub instead of cutting straight across. A blunter, raked edge is built to push a fish along the blade face and let the flow carry it clear before contact turns into a strike.
So the two runners went into the same machine, as close to identical as possible apart from that one shape. Same housing, matched head, matched shaft speeds. The only variable researchers allowed was the geometry of the edge the fish actually meets.
Inside a lab built to watch fish survive, or not
The trials ran at a model scale turbine, a scaled down but fully working machine of the kind used to represent the units at dams and small hydropower plants. Head and runner rotation were matched between the two runners to produce blade tip speeds from about 49 to 91 feet per second, roughly 34 to 62 miles an hour.
Fish were carried into the flow, drawn toward the runner, and passed straight through it. They were recaptured at the outlet.
High speed cameras recorded every passage, so researchers could see exactly how each animal was struck, deflected or missed entirely. Every pass was logged by speed, runner type and outcome, and that structure is what let two very different survival stories emerge from what otherwise looked like the same machine running the same water.
The numbers behind the two blades
The work, by a university fish conservation lab and a turbine maker, was published in the Journal of Ecohydraulics and ran juvenile white sturgeon through both runners under matched conditions. Fish went through in groups of five and were anesthetized for passage, with 112 passing the novel runner and 115 the conventional one, plus control groups handled identically but never sent through a turbine.
The gap was not subtle. Conventional runner passage produced 42 to 78 percent survival at 48 hours depending on speed, 64.5 percent across all speeds combined, with roughly a third of tested fish injured by blade contact.
With the thick, slanted runner in place, survival was 100 percent at every speed but the fastest, where it was 95.6 percent at 48 hours. A hydropower part often treated as fixed geometry turned out to be a powerful lever.
Where the effect gets smaller
Even the better runner was not flawless. Immediate survival through it was 100 percent, and the shortfall at the top speed traced to a single fish that did not make it through the 48 hour holding period, rather than a pattern spread across many animals.
The video told the same story from another angle. Of the fish visible on camera, 21.7 percent were seen being struck by a blade of the new runner, against 57.9 percent for the conventional one, and a further group entered on a collision path but were deflected clear.
White sturgeon make a useful test case. They move between rivers and the Pacific, with a native range running from Alaskan waters south to Monterey, California, so a turbine can sit between a young fish and decades at sea. Hardware on those rivers has had to be rethought before, much as the same industry keeps revisiting old assumptions.
What comes next for the blade that worked
None of this means every dam on the Columbia or the Sacramento swaps runners overnight. Model scale testing is a necessary first step, but full sized machines face more variables, different fish sizes and years of engineering before a design like this reaches a working river.
Fish and wildlife fixes often move in fits and starts, the way painted turbine blades have been trialled as a cheap way to cut bird collisions at wind farms.
Simple changes tend to get tested last.
Even so, the distance between roughly a third of fish being injured and almost all of them coming out intact is hard to argue with. It suggests the fix for one of hydropower’s oldest problems may be less a rethink of the whole machine than a blunter, slanted edge on the part the fish meets first.
Read the whole thing?
Get the week's signal, not the noise
Our sharpest reporting on energy, climate and nature — free, once a week.
