In the middle of the fast-flowing Chilkoot River, an Alaska state employee sits on a small perch over a narrow, fence-like structure and stares down into the rush of water.

Eagles look on from the trees overhead as the river thunders around boulders nearby. The worker’s back is turned to a female grizzly bear creeping up just a few dozen feet away.

The bear cautiously wades upstream toward the man, then sidles close to a low metal railing stretching across the river, the only structure separating the two. Suddenly, the worker springs up, turns around and strides toward the bear, shouting and stomping on the metal platform. The bear stops and stares. After more yelling, the employee raises an air horn and hits the grizzly with a deep blast of noise. She turns, slips into the current and floats back downstream.

The employee returns to work.

This is a typical day on a salmon weir in southeast Alaska. Weirs, common in Alaska and the Pacific Northwest, are man-made fences that span a stream and force migrating salmon to pass through a single opening, where technicians stand watch for weeks or months and count the returning fish. Weirs are an essential tool for getting accurate counts of the state’s critical salmon runs.

They can also create prime fishing grounds for bears.

The technicians who operate them often count thousands of fish per day, while also fending off grizzlies and black bears who get too close.

“It takes a unique kind of person,” said Shelby Flemming, a Haines-based salmon research biologist with the Alaska Department of Fish and Game. “There’s a calmness and meditative aspect, and also the aspect of knowing bear behavior and bear hazing.”

The fish’s annual migrations from the ocean back to their freshwater spawning grounds, known as salmon runs, sustain a multibillion dollar commercial fishing industry in Alaska that employs thousands of workers in boats and processing centers. They’re also critical for the many residents who rely on wild foods for subsistence and for Alaska natives who retain cultural and spiritual connections to the salmon.

“We’re all dependent on salmon,” said Justin Priest, the wildlife agency’s Southeast Alaska salmon research lead. “Whether it’s economically, subsistence or culturally, we’re a salmon people.”

But salmon runs are unpredictable and threatened by climate change, ocean conditions and overfishing. Managing the fisheries sustainably depends on accurate, real-time data. When salmon counts are low, the state might shut down or limit fishing to allow enough salmon to reach their spawning grounds and reproduce. When returns are bountiful, they can allow for a larger harvest.

In many watersheds, those decisions depend on the seasonal weir technicians who spend long days in the middle of rivers and creeks. Wildlife officials sometimes use aerial surveys or sonar scans to get an idea of salmon returns, but nothing comes close to the precise data provided by human eyes watching from a weir.

Technicians count each fish that passes through, categorizing them by species. They also periodically scoop up individual salmon with a net, noting weight, length and sex, and taking a scale sample that can determine age.

“There’s so many [weir] crews across the state getting mobilized between April and July, it feels like an army at times,” Priest said. “We depend on the technicians. They’re incredible people—brand-new folks who are 18 years old to technicians who have made a lifetime out of working seasonal positions.”

Weirs have been around for thousands of years. Many Indigenous groups used systems of stakes or poles to guide fish into traps where they could be harvested.

“Weirs were a tool of governance, an assertion of that village to manage that river,” said Will Atlas, senior salmon watershed scientist with the Wild Salmon Center, a group that works to protect rivers in the North Pacific region. “They inform decision making in major ways around fishery management.”

In the present day, weirs are used for research, to collect fish for hatchery operations and for traditional harvests. They’re deployed by federal agencies, Native tribes and state wildlife managers ranging from California to Idaho to Michigan. Alaska, said Atlas, operates the largest weir program.

Alaska also uses counting towers, which are platforms that sit above clear, shallow streams and allow technicians to take periodic counts to extrapolate salmon totals. The state operates 43 weirs and towers, with most having two to four technicians to work them throughout the season.

As salmon runs fluctuate dramatically from year to year, especially as climate change alters ocean conditions, wildlife officials may need to rely more heavily on weirs to make decisions.

“Forecasting salmon returns has gotten harder and harder because the ocean and the climate have gotten more unpredictable,” Atlas said. “In-season management is really the paradigm of the future when it comes to decision making. You need to make sure enough fish reach the spawning grounds in every generation to regenerate the run.”

In July, Alaska wildlife officials shut down sockeye salmon fishing on the Chilkoot River because the numbers of returning fish were worryingly low. When an abundance of late-arriving sockeyes showed up later in the summer, the river was reopened for fishing.

“We all care deeply about making sure we have sustainable salmon populations for the future,” said Priest, the state researcher. “That starts with the work our weir technicians do.”

Priest said weir technician jobs have long been coveted positions—”the most fun you can have getting paid.” Many of the state’s wildlife officials got their start working on a weir. But it’s become more difficult to recruit as housing and food prices have soared in many rural communities. At the same time, fewer candidates seem eager to give up the creature comforts of modern life to work in remote locations.

But those who have worked on weirs say it’s a special job.

“You see the first fish come through the fence and you see the last fish come through,” said Atlas, speaking over the phone from a weir operation on the Koeye River in British Columbia, Canada. “Very rarely does our life revolve around the cycles and patterns of nature anymore. Those human insights are still incredibly valuable.”

Atlas is working with the Heiltsuk Nation, a First Nation in the region, on a project to restore traditional weir methods while incorporating modern technology. Foreseeing an increased need for weirs and the challenges facing their workforce, Atlas’s group has developed a new approach.

In place of technicians, the Wild Salmon Center is setting up underwater cameras at weirs that link to an artificial intelligence program. The “Salmon Vision” technology counts the fish and identifies their species and sex. They hope it can supplement existing salmon research, especially as more Indigenous nations seek to establish weirs and reassert their sovereignty over fisheries management. The program has largely partnered with First Nations in Canada.

Still, much of the crucial data that informs salmon management comes from seasonal workers, sitting in the middle of a river, clicking a hand counter over and over and over.

And keeping an eye out for grizzlies.

2024 States Newsroom. Distributed by Tribune Content Agency, LLC.

Ensuring that robots can smoothly collaborate with humans in real-world environments is a crucial step towards their large-scale deployment. While some robotic systems are already engaging daily with human agents, for instance at partially automated industrial and manufacturing facilities, human-robot collaboration on everyday tasks remains scarce.

Researchers at University of Padova and Mitsubishi Electric Research Laboratories (MERL) in Cambridge have developed a framework that helps to plan tasks that involve human-robot collaboration. This framework, introduced in a paper pre-published on the arXiv server, specifically focuses on tasks that entail the collaborative assembly of complex systems with various underlying components, such as pieces of furniture.

The researchers called their framework DECAF, which stands for Discrete-Event based Collaborative Human-Robot Assembly Framework for furniture. DECAF has various underlying components, including a discrete-event Markov decision process (DE-MDP) model, a HTM description of the assembly process and a Bayesian interference module.

“The human is characterized as an uncontrollable agent, implying, for example, that the agent is not bound by a pre-established sequence of actions and instead acts according to its own preferences,” Giulio Giacomuzzo, Matteo Terreran and their colleagues wrote in their paper. “Meanwhile, the task planner computes reactively the optimal actions for the collaborative robot to efficiently complete the entire assembly task in the least time possible.”

With the newly developed framework, the collaborative assembly process spans across various steps. Firstly, the robot observes the actions performed by the human agent, via a camera or other sensors.

Based on these observations, the DECAF model plans actions for the robot that would maximize the robot-human team’s efficiency in completing the assembly tasks at hand, while also adapting these actions following unpredictable events. The team modeled the assembly of furniture or other objects utilizing a mathematical framework often used to frame specific decision-making processes, known as DE-MDP.

“We formalize the problem as DE-MDP, a comprehensive framework that incorporates a variety of asynchronous behaviors, human change of mind and failure recovery as stochastic events,” wrote Giacomuzzo, Terreran and their colleagues.

“Although the problem could theoretically be addressed by constructing a graph of all possible actions, such an approach would be constrained by computational limitations. The proposed formulation offers an alternative solution utilizing Reinforcement Learning to derive an optimal policy for the robot.”

Essentially, the DE-MDP model is used to break down an assembly task and identify optimal actions that would allow the robot to efficiently tackle in collaboration with a human agent. The second component of the DECAF framework, namely the HTM model, encodes the interdependence between various sub-tasks, thus facilitating the planning process.

Finally, the team integrated a module based on a statistical approach known as Bayesian interference, which is typically used to continuously update the probability that a given hypothesis is true as more information becomes available. This module allows the framework to monitor the actions of the human agent and use them to predict the intentions of a human agent.

The researchers evaluated DECAF in a series of tests, and ran both in simulation and in a real-world setting. For the real experiment, 10 adult participants were asked to assemble a chair purchased from IKEA in collaboration with a 7-DoF robotic manipulator (i.e., the Franka Emika Panda arm).

The findings of initial tests run by the team were highly promising. In simulations, the DECAF framework was found to outperform standard planning policies, while during the real-world experiments it appeared to improve the efficiency and quality of the human-robot collaboration.

“In the future, we plan to include other optimal metrics beyond the execution time, such as human safety, action correlation and human ergonomics,” wrote the researchers.

© 2024 Science X Network

An orca has been born to the southern residents: L128, calf of a first-time, 31-year-old mom, L90.

The baby is tiny, with clear fetal folds, making it probably about three days old. It was seen for the first time on Sunday, said Michael Weiss, research director for the Center for Whale Research, which confirmed the birth on Monday.

Mom and baby appear to be doing fine, Weiss said. The baby is nursing and moving normally, and mom looks physically robust. The pair was seen traveling up and down the west side of San Juan Island all day Monday.

“Any birth is good news,” said Weiss, who helps track the J, K, and L pods that make up the southern residents, a distinct population of fish-eating orcas that frequent the Salish Sea. The orcas, listed as endangered under the Endangered Species Act, have struggled to rebuild their numbers after years of losses.

Everyone is hoping for a female, but the gender of the baby is not yet known. The July 1 census was 73 whales, which does not count the new baby.

Also seen recently was L25, confirming she is still leading the pods as the oldest southern resident—perhaps the oldest orca ever known. Possibly born in 1928, L25 carries the vast generational knowledge of countless miles, traveling and finding fish for her family. The older matriarchs lead the pods, particularly in times when fish are scarce.

New calves face tough odds, with about half of newborns not surviving their first year. One of the challenges is toxins in their mother’s milk, especially for an older, first-time mom like L90. Pollutants are carried to the calf in the fat of the mother’s milk.

The southern residents are battling extinction. They are facing a lack of adequate, regularly available food because of the decline of Chinook salmon throughout their range. Orcas preferentially target Chinook, the biggest, fattiest salmon. They will also eat chum and coho.

Orcas form lifelong family bonds, with the young never leaving their mothers. They are highly social, deeply intelligent top predators. They share language, greeting ceremonies and care of their young. Matriarchs such as L25 also share most of what they catch.

J pod is the pod most likely seen year-round in the waters of the San Juan Islands, British Columbia and even the urban waters of Seattle. According to the Center for Whale Research, J pod has 25 members. K pod has the fewest members, with 15 members. L pod is the largest of the three, with 33 members, not counting the newest baby.

2024 The Seattle Times. Distributed by Tribune Content Agency, LLC.