Other Scholarly Content


Emerging Technologies in Fisheries Science: A Transdisciplinary Report Public Deposited

Downloadable Content

Download PDF


Attribute NameValues
  • The Pacific Coast Groundfish Fishery harvests a diverse and large grouping of fishes, but it did not become heavily fished until around WWII. This makes the groundfish fishery a comparatively young fishery. Despite its youth, it is one of the largest and most lucrative fisheries in Oregon—with a current harvest value of approximately $48 million per year, which is exceeded only by the Dungeness crab fishery. Northeastern Pacific Coast Groundfish species are also important for recreational and tribal purposes, although it is difficult to compare these to the commercial industry. With over 90 different species to consider, this commercial fishery is complex, and there are many different stakeholder groups involved, each with their own goals, values, and perspectives. Fishing regulations greatly impact local stakeholders, some of whom rely on the fishery for their livelihoods. These local stakeholders are dependent on accurate stock assessment surveys and models so that the fishing regulations are appropriate. Some stakeholders feel that regulations tend to be overly cautious to compensate for the large amount of uncertainty involved with managing a fishery and estimating a fish population. To reduce this uncertainty and the need to err so heavily on the side of caution, stock assessment surveys could include innovative technologies and novel datasets. For example, these stock assessments do not currently use automated video surveillance on their bottom trawl surveys, an emerging form of machine learning. As understood by the NSF-funded National Research Traineeship (NRT) training, there are three interwoven core concepts: 1) Big Data (BD), 2) Coupled Natural-Human (CNH) systems, and 3) Risk and Uncertainty (R&U) analysis and communication. Big Data refers to any high volume of data with high throughput. Coupled Natural-Human systems are the biological and human worlds, as well as their overlap and interaction. Risk is the potential and likelihood of an unfavorable event, and uncertainty refers to the unknowns of a likelihood, process, or analysis. This project chose to investigate these three concepts within the framework of emerging technologies and fisheries science. Emerging technologies are those dealing with BD, since this is a relatively new area of study, and this project specifically focused on computer vision within machine learning. This technology was applied to the realm of fisheries science and ultimately management, which is the study of a coupled natural-human system. Changing oceans conditions mean that Northeastern Pacific groundfish are at risk and their future is uncertain. Therefore, this project set out to determine how the influence of big data, machine learning, ecological inference, and environmental decision making overlap. The story of the life and study of these fishes in a newly Americanized sea is ready for a closer examination. It is for these reasons combined that Pacific coast groundfish fishery science provides a robust platform in which to explore the autonomous capacity of technology and data production at the intersection of environmental science and decision making. More specifically, to what extent are large, ecological datasets informing the production and application of emerging technologies in fisheries science, and how are these new technologies and sampling methods being integrated into fisheries management frameworks? A case study in which to explore this concept can be found in the testimony of a flatfish, or rather, the complex, ecologically and economically important assemblages of numerous groundfish species in the northeastern Pacific Ocean where flatfish are found.
Resource Type
Date Issued
Degree Level
Rights Statement
Peer Reviewed
Additional Information
  • This report was completed as a requirement of an NSF funded graduate training program and OSU minor. It has undergone several revisions and edits since it's submittal to the ScholarsArchive@OSU. To follow up email samm.newton@gmail.com



This work has no parents.