AQUA 2024

August 26 - 30, 2024

Copenhagen, Denmark

SMART, SUSTAINABLE, SHELLFISH AQUACULTURE MANAGEMENT: ADVANCING TECHNOLOGICAL DEVELOPMENT OF OYSTER AQUACULTURE IN THE USA

Matthew W. Gray*, Miao Yu, Michael  Xu,  Allen Pattillo, Yang Tao, Kaustub h Joshi,  Nikhil Chopra, Don Webster, Matt Parker, Cathy Liu, Bobbi Hudson, Yuanwei Jin ,  Chiao-Yi Wang, Yiannis Aloimonos , Alan Williams, Gudjon Magnusson.

 

University of Maryland Center for Environmental Science, Horn Point Laboratory, Cambridge Maryland USA 21613

 



 The oyster aquaculture industry has steadily grown in the United States in recent decades. Alterations in regulations and steady demand from consumers have fueled industry growth, which is principally driven by bottom-culture of crops, where oysters are planted on a bed of shells and typically harvested 2-3 years later. Although the oyster aquaculture industry has existed for centuries, the technology used to manage and harvest crops is antiquated and inefficient. Specifically, growers have used dredges to track the growth of their crops and harvest them. Dredges are inefficient at harvesting crops and may damage those left behind. To advance  the technological development of the industry, we have developed technology that enables off-the-shelf robotic underwater vehicles (Fig. 1a) to help growers manage their crops and monitor lease conditions. This presentation will provide a brief overview of the Smart, Sustainable, Shellfish Aquaculture Management (S3AM) robotic system and its products that extend from a highly interdisciplinary team consisting of mechanical engineers, roboticists, computer scientists, resource economists, ecologists, and aquaculture industry extension agents.

Tracking shellfish production begins with identifying crops within the lease. Distinguishing oysters from rocks or empty shells using optical sensors in real-time poses several challenges. We used machine-learning techniques to train computers to recognize and enumerate oysters on active leases (Fig. 1b). In highly productive estuaries, such as the Chesapeake Bay, optical sensing of oysters is challenging under low visibility conditions. Therefore, we also used sonar sensors and machine-learning approaches to distinguish crops, empty shells, and bare substrates. Underwater localization of S3AM was also developed so crop information would be spatially explicit. Over time, monitoring with S3AM will provide production rates (crop growth and health) and lease water quality information , which may be useful for detecting acute environmental changes that may harm crops. To package this information, we developed an application so growers could access their crop information for greater management control. This program  has  leadership throughout the US and  it is  intended to serve t he  shellfish industry along all coastlines.