Can robotic ‘sensor fish’ save salmon from hydroelectric dams?

The McNary Dam spans the Columbia River, stretching between Washington and Oregon. (Andrea Starr / Pacific Northwest National Laboratory Photo)

Some 71 percent of the world’s renewable energy comes from hydropower and more dams are being built all of the time. But while energy wrung from rivers doesn’t release greenhouse gases and contribute to global warming, there are still environmental costs.

And in the Northwest, the main concern is the harm caused to salmon traveling up river to spawn and downriver as juveniles returning to the ocean. The dams injure and kill fish in a variety of ways as they navigate fish ladders and bypasses, plunge through turbines and swim through unnaturally warm reservoirs.

Initial Sensor Fish prototype. (Photo via PNNL)

Scientists at the Pacific Northwest National Laboratory (PNNL) in the Eastern Washington city of Richland hope to make that journey less life-threatening with new technology to measure the physical impacts of hydropower infrastructure on fish and through improvements to tools that track salmon migration.

The Department of Energy’s PNNL has invented something called the Sensor Fish, a device that’s about 3 1/2 inches long, or roughly the size of juvenile salmon called a smolt. The sensor is dropped into the water at the top of a dam and travels through it, collecting information on pressure, acceleration, rotational velocity and orientation. The next-generation device takes about 2,000 measurements per second and is retrieved below the dam.

The reusable sensors will be available beginning this spring for about $3,500 a piece. They’re being manufactured by Minnesota-based Advanced Telemetry Systems (ATS).

“There is nothing else like this out there,” said ATS sales manager Joe Allen. “This is a first.”

PNNL also created the first and possibly only injectable tracking device that can be inserted into young salmon to monitor their migration for up to three months. The minute acoustic transmitter is roughly twice the size of a grain of basmati rice. These are also licensed for manufacture by ATS and cost $250. They’re used only once.

The hope is that data from the two devices can help dam engineers and operators design and manage hydropower plants in a more fish friendly manner.

“We want to identify what kind of dam operations are affecting fish behavior,” said Daniel Deng, a laboratory fellow at PNNL who helped lead the research. Using the two technologies together, “we can correlate the fish behavior with the physical conditions.”

Debate over the effects of Northwest dams on salmon has been on a steady simmer for years. Salmon and orca scientists and advocates have sought for more water to be spilled at the dams and in some cases called for their removal in order to save endangered marine life.

Granny the orca
The orca known as J2, or Granny, was a member of the Southern Residents that frequent Puget Sound. She died in  2016. (Center for Whale Research / 1998)

But the fight is heating up over the breaching of four dams in the lower Snake River, which flows through Southeastern Washington and into the Columbia River.

The Puget Sound orca population known as the Southern Residents have seen numerous deaths in recent years, and their population is hovering at 74 — an alarming number not seen since their capture for aquariums and aquatic parks ended decades ago.

A key cause of their decline is a shortage in chinook salmon, their preferred food, including fish that spawn in the Snake and Columbia river systems.

In his two-year budget, Washington Gov. Jay Inslee proposed spending $1.1 billion on orca recovery. That includes $750,000 for a stakeholder process to consider the environmental, economic and social impacts of removing the lower Snake River dams.

Those concerned about salmon and orcas say in this case that data from new tech innovations won’t help the problem. The government has already spent vast sums making these dams less harmful to fish, but salmon survival still suffers.

The Sensor Fish devices collect data that reveals the pressure and other physical conditions that a young salmon experiences traveling through a hydropower dam. (Pacific Northwest National Laboratory Photo)

“These new gizmos might provide new information, but I’m not convinced that it’s a good use of money and that it’s going to help restore salmon in any regard,” said Joseph Bogaard, executive director of the nonprofit Save Our Wild Salmon, which supports the dams’ removal.

“People are well meaning and trying to find ways to maintain the hydro system and protect and restore salmon,” he said. “But we know that some major changes are needed.”

There are hydropower projects elsewhere that could potentially benefit from improved data collection. China, Brazil and Canada each produce more hydropower than is generated in the U.S. The technology created at PNNL could help people strike a better balance between energy production and wildlife in the design and operation of the power facilities.

The PNNL researchers are working on ever better devices. A mini Sensor Fish is in development, and work is underway to make acoustic tags that are smaller, more powerful and longer lasting.

“We are just happy to see this out in the market,” Deng said, “so people can use these tools and people can gather information that hasn’t been gathered before.”