Effort continues to replace humans with cameras on fishing boats
Several years into the controversial effort to bolster Alaska’s fisheries observer program, a top federal fisheries official defended the work at a Seattle gathering of fishermen.
Eileen Sobeck, the NOAA Assistant Administrator for Fisheries, took the stage this past Nov. 18 to talk to fishermen gathered for the annual Fish Expo event to recap the program.
Observers are the eyes and ears on boats, collecting a range of data, she explained.
“We have been monitoring fisheries for decades, and we do it in a lot of different ways,” Sobeck said.
But the details of the program have been under fire over the past few years. Federal efforts to put a human on smaller boats was met with concerns about safety and efficiency, and fishermen’s requests to use cameras have had logistical difficulties.
Over the past few years, the effort to use cameras has increased nationwide, and the National Marine Fisheries Service has been tasked with sorting out how to make that work, both logistically and cost-wise.
Over 10 years, the National Marine Fisheries Service has helped fund more than 30 electronic monitoring, or EM, pilot programs. Expenses include the cost of cameras, the cost to install them, and the cost of going through the immense amount of data they can collect.
“We have, collectively, an interest in being as cost-effective as we can possibly be,” Sobeck said.
That effort has translated into regional electronic monitoring plans that were finished more than a year ago, and are now being implemented with plans for regular reviews, said George LaPointe, one of the point people on the project.
Although monitoring in some fisheries has developed successfully, like in the groundfish fisheries, LaPointe said, the agency is still working toward certain implementation, such as in Alaska’s small boat fixed gear and pot fisheries, where the target date is 2018.
That fleet includes about 630 vessels right now, with a much smaller number that have opted in for 2017.
The EM development effort has taken several years, from the 2013 decision to restructure the observer program, to 2016, when 51 vessels participated in a pre-implementation program.
This year, the agency is hoping that 120 of the smaller fixed gear and pot boats will be on board with the program, preparing for 2018 implementation. For 2018, the vessels that are required to have some monitoring, but not be covered full-time, will have the option of electronic or human observation.
As the agency has worked on implementation, several challenges have arisen, LaPointe said.
“We can put cameras on boats. And we can get the data out of those. But it’s expensive,” he said.
Now, work is underway to find a more efficient way to review the camera-collected data. Ideally, the agency wants accurate fish identification from computers, rather than requiring humans to review the data.
While the agency is helping fund EM for now, LaPointe said they eventually want to transition to funding the program on its own.
The agency is looking both at regular cameras, and stereo cameras, mostly testing those on longline boats. Those are machine vision systems, which ideally can process the imagery as fish come on board, limiting the time it takes to process images, as well as the cost. But it’s still in testing phase.
The agency has heard years of critiques on the program, from costs to the logistical difficulties boats face in carrying an observer or camera. But in the November discussion, the first question was about fishermen who want to help test EM.
“There are some remote places that would like to try this stuff,” Doug Rhodes, a longliner out of Prince of Wales, told the agency. He said in past years, he hadn’t been to a port where he could get a camera to try, but thought that many rural fishermen would give it a shot if there was a way to let them install it or get it installed at a closer port.
Although that’s not yet possible, Suzanne Romain from the Pacific States Marine Fisheries Commission said the ultimate goal is to make the technology open source.
Eventually, they want a system that fishermen can install themselves, she said.
That’s one way to keep the cost down, and ultimately, minimizing the cost of the program is a primary goal.
Fishermen in the North Pacific, unlike their Atlantic counterparts, have borne much of the cost of observation, although the agency has helped with costs of testing pilot electronic monitoring programs, and have said that it’s too much added expensive, particularly for small boat, lower-margin fisheries.
Addressing that is a goal, Sobeck said.
“We are trying to be innovative,” Sobeck said. “We are trying to find cost savings.”