A four-minute video called “How Wolves Change Rivers” has been watched more than 46 million times. It tells a tidy story: bring back the wolves, and the rivers themselves move. That is a lot of people absorbing a claim that the scientists who study Yellowstone still cannot agree on. 

A quick note before going further. I am not an ecologist, wildlife biologist, or hydrologist, and nothing here is a verdict on the science. This is a piece of reading and reflection on a live scientific disagreement. The studies at its center are observational, drawn from field sites in one specific park, and even the researchers involved describe their findings in careful, hedged terms.

The viral story, told cleanly

Gray wolves had been largely absent from Yellowstone, mostly wiped out by a predator-control program, between 1914 and 1926. Then, in early 1995, they came back.

The first wolves arrived by truck through the Roosevelt Arch on January 12, 1995. In total, fourteen wolves went into three Lamar Valley acclimation pens that first year and were released by late March. A second group followed, bringing the total to 31 wolves by the end of 1996.

The narrative that grew from this is what ecologists call a trophic cascade. Wolves thin and spook the elk, elk stop overgrazing the young willow and aspen along the streams, the recovering plants stabilize the banks, beaver return, and the water runs differently. As put it in the video, “We all know that wolves kill various species of animals, but perhaps we’re slightly less aware that they give life to many others.” It is a nice line, and also the part scientists keep flagging as too clean.

Where the science actually stands

The case for a strong cascade has been built mostly by William Ripple and Robert Beschta of Oregon State University, who have argued the idea for two decades. Their earlier work reported that young aspen and willow along the streams recovered as elk browsing eased after wolves returned.

More recently, a 2025 paper from Ripple and colleagues put a striking number on it: roughly a 1,500 percent increase in willow crown volume.

That figure is where things get interesting, because it rests on a method: the team used willow height, converted into an estimate of crown volume, as a stand-in for how much plant there actually is. That conversion is exactly what the critics went after.

The reanalysis that complicated everything

In October 2025, a team led by Daniel MacNulty of Utah State University published a formal comment arguing that the 1,500 percent figure does not hold up. Their core objection is mathematical.

As MacNulty describes it, “Because height was used both to compute and to predict volume, the relationship is circular—mathematically guaranteed to look strong even if no biological change occurred.”

Co-author David Cooper of Colorado State University put the reanalysis’s conclusion bluntly, saying that once the problems are corrected, “there is no evidence that predator recovery caused a large or system-wide increase in willow growth.” That is a hard line, and it is the critics’ reading of the same dataset, not the field’s settled position. 

Ripple’s group did not concede the point. In a 2026 reply covering the closely related dispute over aspen, they corrected an error in their effect-size calculation and reported a smaller figure, but still described the result as a strong, ecologically significant cascade, while acknowledging it was never a complete restoration of every aspen stand. The corrected figure was smaller, but the disagreement itself was not resolved.

What ecologists do and don’t agree on

The scientists arguing about this are not split into wolf-lovers and wolf-doubters.

MacNulty is careful to say the critique is not an attack on predators at all: “Our goal is to clarify the evidence, not downplay the role of predators,” he said. His stated principle is the sort of thing most researchers would sign: “Predator effects in Yellowstone are real but context-dependent—and strong claims require strong evidence.”

Read across the two camps and a rough middle emerges. Almost everyone agrees some cascade happened. What they contest is its size and its cause, whether the willow recovery is large and system-wide or modest and scattered, and how much of it is wolves versus hydrology, climate, beaver, bison, and the specific luck of a given streambank. 

Why the cleaner story keeps winning

Arguably the reason the river-changing version travels so far is that it is simply a better story. Wolves as landscape architects, one decision in 1995 rippling outward into the shape of the water. It has a shape our brains like. “Willow crown volume responds modestly and variably to a mix of browsing pressure, water tables, and local conditions” does not get 43 million views.

The messier version is, I think, more interesting for being so. The wolves are real, their return is real, and Yellowstone looks different for it. Whether they bent the course of the rivers is the question still sitting on the workbench, and the truthful answer right now is something like: probably, at least partly, in some places, and the researchers are still arguing over how much. That is not the ending the video offers, but it is the one the evidence, so far, will support.