AI Northern Shrimp Abundance Estimation Project
The Canadian Centre for Fisheries Innovation (CCFI) has approved a project partnership with three fishing industry associations to explore the use of Artificial Intelligence (AI) in developing a population assessment model for the Northern Shrimp resource off Eastern Canada. Currently, interconnected shrimp resources from Baffin Island to the Grand Banks off Newfoundland are scientifically assessed independently from each other. In addition, environmental and ecosystem dynamics that are important drivers of shrimp populations are not well understood, with a paucity of this data in more northern areas. Especially in the face of climate change, fisheries scientists are seriously challenged to employ traditional methods/models to assess this stock complex, and to understand its population dynamics. This project will explore the use of AI in developing a population assessment model for the entire Northern Shrimp stock complex. The use of AI to solve such complex challenges is becoming more possible as machine learning evolves to its next level - Deep Learning (DL). DL is finding a larger role in human healthcare, manufacturing and banking, where optimization, speed, and accuracy are required. In collaboration with DFO NL Region scientists, experts from Canadian AI firm Intelleo AI Inc., will employ the first known use of this methodology for assessing and projecting fish stock abundance. “To this point, no population model has been adopted for assessing the Northern Shrimp stock complex, or its individual components. While correlations between shrimp abundance and some environmental and ecosystem factors are able to be observed in some areas, there is little understanding of underlying population productivity and dynamics, and virtually none relative to the stock complex as a whole. This Project has the potential to contribute significantly to sustainable resource management,” stated Keith Hutchings, Managing Director of Canadian Centre for Fisheries Innovation (CCFI). The three fishing industry partners in this Project are the Canadian Association of Prawn Producers (CAPP), the Northern Coalition (NC) and the Nunavut Fisheries Association (NFA). Bruce Chapman, CAPP’s Executive Director added: “The application of this exciting technology, with its capacity to identify and ‘interpret’ relationships within large volumes of data, may be able to unlock some of the mysteries of shrimp stock dynamics across this very broad geographic area. Over one thousand people, and hundreds of millions of dollars in existing investments are dependent on achieving a more robust understanding of how our fishing activity impacts the future of this shrimp stock complex. Alastair O’Rielly, Executive Director of Northern Coalition Corporation noted “Traditional approaches to stock assessment in the North are costly and inadequate. We need to explore the application of new technologies and methodologies to improve sustainable resource use. We are excited to work with the three industry associations, DFO, and Intelleo AI in bringing this innovative stock assessment methodology to Northern Shrimp populations, and the role that it could play in projection of fish stocks, Brian Burke, Executive Director with the Nunavut Fisheries Association stated. “NFA is supportive of new approaches to stock assessment, including the utilization of AI and Machine Learning, to build and improve on the longstanding traditional approaches.