Projects in the Field is a series of independently produced articles profiling work supported by NFWF’s Electronic Monitoring & Reporting Grant Program, and is meant to raise awareness and support for these important initiatives. As always, your feedback is welcome.
The Gulf of Maine Research institute (GMRI) is committed to developing Electronic Monitoring (EM) into a powerful and affordable fisheries monitoring tool. To date, many electronic monitoring efforts have failed to scale and have not proven cost effective due to technology constraints. Through their Maximized Retention EM Program, GMRI is addressing these constraints by utilizing advanced EM technology in the New England Multispecies Groundfish Fishery.
The Maximized Retention EM Program is designed to improve the timeliness and accuracy of monitoring data, while decreasing the cost and catch-handling burdens of EM. Participating vessels retain all allocated groundfish (excluding Halibut) regardless of size, and EM is used as a tool to ensure all groundfish are landed. Rather than recording discards at sea, allocated fish are accounted for on shore by a dockside monitor. For fishing year 2018, the program includes only three vessels, but the new technology and processes they are testing could make EM accessible to a wider array of fishermen. Funded by the National Fish and Wildlife Foundation, program partners — GMRI, Integrated Monitoring, Cvision AI and the Sustainable Harvest Groundfish Sector — are developing solutions to three major shortcomings of traditional EM systems: data transmission, video analysis, and data storage.
Retrieving EM data from vessels has been a bottleneck for EM programs around the globe, resulting in slow data turnaround, delayed feedback to vessels, and logistics challenges (e.g. meeting a vessel to retrieve data or ensuring a hard drive is shipped on time). To mitigate these challenges, GMRI’s Maximized Retention EM program is testing wireless transmission of video data.
Currently, all video data from the program is transmitted via cellular service. Advanced servers aboard participating vessels are cellular-compatible, and novel data compression and transmission processes enable delivery of EM footage to video analysis stations as early as the day after landing. To date, the program has transmitted over 120 sea days of video footage using this technology. Not only does this eliminate complicated data shipping and retrieval logistics, it significantly decreases the time between data collection, review, and submission, resulting in faster assimilation of EM data into management processes and rapid feedback to vessel participants.
The process of generating data from EM video is a major cost to most EM programs. Artificial intelligence is an oft-touted (but rarely-applied) method to reduce these costs. As part of the Maximized Retention EM program, GMRI and partners have developed algorithms that are currently detecting specific activities on deck, including: hauling (setting out and hauling back), catch sorting, fish-hold loading, and offloading. These events are critical to identifying high priority portions of video that require detailed review. Human review is still being used for quality assurance and to help further refine the software.
The project team is currently working towards using the systems powerful server to run this software onboard the vessels. Bringing this technology onto the vessel and closer to the action will make data transmission and video recording more efficient. For example, on board activity recognition software has the potential to identify and select specific video segments for transmission, making cellular transmission faster and more affordable while simultaneously decreasing required storage capacity and cost.
EM practitioners and advocates have long worried about data storage issues such as cost, security, access, and capacity. The Maximized Retention EM program uses Amazon Web Services’ cloud-based storage to alleviate many of these concerns. The program utilizes Amazon S3 services for video that needs to available at an instant and Amazon Glacier to cost effectively archive older video. Cloud storage puts the responsibility of costly IT services on the third-party providers, alleviates infrastructure (server) costs, and requires less maintenance and fewer upgrades over the long-term. A reputable cloud provider also provides excellent data security, remote access, and practically unlimited capacity to scale.
By piloting new technologies across these three important areas of need (data transmission, video analysis, and data storage), GMRI is creating a vastly more efficient EM program. Furthermore, by deliberately building complementary technologies into the EM system design, like in GMRI’s MREM program, we can begin to revolutionize EM systems. As these efforts increase the efficiency and capabilities of EM programs for fishermen and regulators we hope to realize a true win-win opportunity in our regional fishery.
Editor’s Note: We hope articles like these can spur learning and discussion across our entire network of EM/ER/fishtech stakeholders, as well as the public. Please leave any comments, questions or suggestions below. You can also email the author at firstname.lastname@example.org.
Photos courtesy GMRI — Reprint inquiries: email@example.com