Challenges and solutions for scaling electronic monitoring: highlights from first session

Electronic monitoring, or EM, technologies are playing a key role in advancing effective and equitable fisheries management. EM has been around for close to three decades — and yet widespread uptake has remained elusive. Is it the cost, the reluctance of fishermen to put a camera onboard, or the lack of broader policy standards for EM programs and data sharing? Maybe it’s all these factors (and more) that have held EM back, even as efforts have increased to scale the technology to more and more fisheries around the world.

But the tide is finally rising. Within the last five years, we have witnessed a digital revolution in fisheries-focused technology and a rapidly expanding EM market, now with over 20 companies offering products and services for both industrial and small-scale fisheries alike. As the technology becomes cheaper, more functional and more widely available, it looks like EM is set to go from its current use level (~2,500 vessels) to over 50,000 vessels in the next 10 years (CEA 2020). But what are the technological advances that are enabling this growth?

That’s the question that motivated over 140 participants from around the globe to drop in recently to the first session of the 20/21 Seafood and Fisheries Emerging Technologies (SAFET) conference. Over the course of two and a half packed hours — including a keynote by moderator Kate Wing, diverse speaker panel, breakout sessions and tech spotlights — we learned about the latest and greatest in EM technologies, and discussed the challenges holding it back.

Below is a curated list of key insights from our speakers and from breakout conversations. It seems the only things missing were the name badges and coffee breaks!

The potential benefits of automating EM review are huge, and so are the challenges with getting artificial intelligence (AI) up and running

Kydd Pollock and Matt Merrifield of the Nature Conservancy described their efforts to install EM systems and automate human review of longline tuna fisheries in the South Pacific. Working with the 8 member-nations of the Parties to the Nauru Agreement, they’ve collected EM data on over 500 trips, and are using the imagery to develop algorithms that will assist and eventually automate review of the EM data. This dramatically speeds up the review process by showing the reviewer only the priority events (e.g., when a fish is coming on or off the boat), creating greater cost efficiency, not to mention happier reviewers spending less time in front of a screen.

AI-enabled EM systems for activity recognition and species identification show significant promise for overcoming some of the largest barriers to greater EM uptake, namely human review and data storage costs, and in some contexts reduced privacy concerns, especially where the AI works on the “edge,” meaning analysis is done on board the vessel and only summary information rather than images is sent to fishery managers.

Development of standards and protocols so that governments and institutions are comfortable sharing data is critical

For the high seas fisheries that are managed by regional fishery management organizations, or RFMOs, standards of use and privacy for data collected using EM is critical. From the Pacific Community, Malo Hosken presented on that organization’s efforts to integrate EM with existing monitoring practices for the tuna fleets of Pacific Island countries and territories — no small feat! Malo and his team have helped regional leaders establish data analysis standards and have developed an online reporting system (Dorado) to integrate data from various sources such as logbooks, observers and EM. He believes that data analysis standards are necessary for effective, self-sustaining EM, and ultimately for long-term tuna stock health and abundance in the region. This is an important consideration for other EM programs around the world.

The perceived cost of EM technology can be a barrier in small-scale fisheries, when in fact, inexpensive and easy-to-use hardware exists that can be matched to the needs of small-scale fisheries

Most EM programs have been implemented in larger, industrial-scale fisheries, leaving those in smaller fisheries wondering if EM is too expensive or difficult for the value of small-scale fisheries, or SSF. This is particularly unfortunate as small-scale fisheries are critical for livelihood and food security of hundreds of millions of globally, and are particularly vulnerable to the impacts of climate change.

Dr. Todd Gedamke (MER Consultants), who has spent many years working in the smaller islands of the Caribbean, sees EM technology as a way to solve an age-old fisheries management problem: lack of basic life history data. Without robust quantitative data on size classes in the fishery and their mortality, it’s hard to know stock health and how hard to fish without depletion. Keeping low cost and simplicity in mind, Dr. Gedamke and his colleague Bill Hartford (Nature Analytics) have engineered a camera-based length measurement system for SSF that costs less than US$1,000 per system. They’ve also developed annotation software and have logged over 20,000 images for training algorithms. As Dr. Gedamke puts it: “Doing nothing is not an option … We have technology, we have cheap solutions, we should be getting this out to small-scale fisheries all around [the world].” We couldn’t agree more, and plan to cover the opportunities to apply technological solutions to SSF challenges in more detail in the December 2020 SAFET session.

Radical transparency can bring benefits to fishermen

Another innovative use of an EM camera is to earn the complete trust and confidence of your buyers. New Zealand fisherman Karl Warr is pioneering a whole new level of transparency using a deck camera and streaming live 24/7 to his website so anyone can view his fishing operation. His aim is simple: Karl is betting on attracting sustainability-focused customers who care where their seafood comes from and how it is caught. Of course, it takes more than just the video to get the fish where it needs to go, so Karl is partnering with Cathie Gould of Daily Catch, a new e-commerce platform to educate the seafood consumer, enhance the product and build relationships. This flips the driving force for monitoring on its head, from government regulators (i.e. fishery managers) to the progressive fisherman who wants to demonstrate accountability for business reasons, including a growing consumer base that wants to know how and where their fish is caught.

EM is not a panacea to solve all oversight issues, and should be used in concert with human observers and other tools

It is easy to get excited by all the digital potential that EM/AI tech has to offer. However, throughout the session we were reminded that human intelligence, or HI, is the most valuable asset we have in managing our fisheries well. By working together, with the current observer programs, fishery participants, managers and regulatory entities, we can ensure the technology we implement is practical and effective. By sharing what we know through papers, workshops and conferences like SAFET, we can overcome the longstanding challenges with EM to take us beyond a myriad of pilots to actual large-scale, fully implemented programs. That’s how we will get to 50,000 vessels by 2030, resulting in thriving fisheries and fishing communities even in the face of climate change.

THANK YOU to all speakers, and all of those who attended. The full plenary and speaker presentations can be found here.

Don't miss the next two sessions!

Oct. 29th 10-11am PST, Emerging Technologies and their Application in Fisheries Management (Special Session of the UK Space Agency, white paper release) REGISTER

Nov. 12th 8-10am PST, What has your Data done for you lately? Getting the Most from Sustainable Fishing Technologies REGISTER

- SAFET Connections -

Subscribe to Updates from the SAFET Team

@2020 SAFET Conference