DFO MSc Defense: Kevin Bennett

This thesis describes an underwater video camera and a solar-powered single-board computer system designed and assembled with commercially available components to monitor living marine resources in their habitats. The system can directly observe and record migrating river herring, a collective name for Alewife (Alosa pseudoharengus) and Blueback Herring (Alosa aestivalis). River herring populations are depleted relative to historic levels due to decades of habitat loss and overexploitation; harvest is now widely prohibited. River herring in select river systems are monitored by human observers, and some have electronic counting systems. However, current fish-counting technologies are phasing out, and developing additional fish enumeration methodologies is a high priority. Video recording systems made with commercially available components are accessible, relatively low-cost, and offer standardization of equipment and data between users. Electronic video monitoring systems are also easily replicable and suitable for species such as river herring. To monitor and record the annual spring migration of spawning river herring in New England rivers, I created and deployed multiple solar-powered computers paired with underwater cameras. These systems are designed to operate in remote locations and record videos of migrating fish, which can then be reviewed to produce spawning population count estimates. Video monitoring systems are also useful for comparing fish stocks before and after restoration efforts and fish passage installations. This portable, solar-powered, single-board computer system is intended to be an economical, robust, and modular field computer system that can be easily customized to fit user needs. A video surveillance system provides observation coverage for the times of the day when human observers are not present; recorded fish migrations improve population estimates through more complete monitoring of the event. This system can also support upgrades and future expansion; a temperature probe can be installed alongside the camera. Recent developments in edge computing technology offer the potential for in-situ object recognition through on-board image processing. Natural resource managers can leverage a combination of consumer technologies to monitor ecosystems and make effective decisions.

Meeting ID: 915 4061 6096

Passcode: 068189