Faculty Miao Yu
, Nikhil Chopra
, Yiannis Aloimonos
, Yang Tao
, Matthew Gray, University of Maryland Center for Environmental Science, Chengchu Lin, UMD Extension Service, Matt Parker, UMD Senior Agent Associate AGNR-UME-Sea Grant Extension, Donald Webster, UMD Senior Agent AGNR-UME-Wye REC, Yuanwei Jin, University of Maryland Eastern Shore, Brian Callam, Louisiana State University, Bobbi Hudson, Pacific Shellfish Institute, Jonathan van Senten, Virginia Tech
USDA National Institute of Food and Agriculture
This is a five-year, $10M project to transform shellfish farming with smart technology and management practices for sustainable production.
The smart sustainable shellfish aquaculture management (S3AM) framework entails an integrated, system-based approach to address the industry needs and tackle the challenge of establishing sustainable shellfish farming with significantly enhanced productivity and profitability. An interdisciplinary team consisting of investigators with diverse expertise and engaged stakeholders from three coastal regions (East, West, Gulf) is brought together to pursue integrated research, education, and extension activities.
The S3AM framework employs two novel technology tools to revolutionize farm management practices: S3AM monitoring and S3AM harvesting. In S3AM monitoring, the team will develop novel environmental sensing and imaging tools and AI-based mapping algorithms and implement them on an underwater drone to perform lease environmental monitoring and crop inventory monitoring. Before planting seeds, S3AM environmental monitoring will be used to map the water quality and bottom substrate conditions. This will provide farmers with accurate bottom lease conditions to enable precision planting, which will help reduce losses due to seed mortality and increase farm productivity. During growing seasons, S3AM crop inventory monitoring will be conducted multiple times to create high-precision crop inventory maps to help farmers improve inventory records, as well as make predictions on future farm productivity and profitability. By using the precise crop inventory maps created by S3AM monitoring, S3AM harvesting will be developed to create an optimized path for the dredging vessel to perform high-efficiency precision harvesting, which maximizes coverage, minimizes the dredging path, and reduces labor and energy used during harvest. In addition, the team will develop user-friendly farm management, data analysis, and visualization software based on the mapping data obtained with S3AM monitoring and existing farm management models. This software will help farmers better maintain their data assets, manage and predict the conditions of their crops, as well as develop and evaluate business plans for improving the economic viability for farming operations.
In Year 1, the team will collect water quality data from sources including buoys and NOAA's System Wide Monitoring Program (SWMP) for 12 U.S. coastal states. The data will be analyzed to ensure wide-range applicability of S3AM monitoring across the Nation's bays and estuaries. In addition, the team will work with growers from three costal regions to obtain benthic habitats images to understand habitat complexity. In Year 2, the team will use the collected water quality and benthic images to recreate a range of environmental conditions and test the S3AM monitoring at Shellfish Aquaculture Innovation Laboratory, located in the Chesapeake Bay. In Years 3 and 4, field tests will take place in the Horn Point Laboratory Demonstration Farm, a 2.25 acres licensed lease used to evaluate shellfish gear and management practices. S3AM will be used to evaluate substrates within the farm as well as evaluate oyster densities in Sandy Hill Oyster Sanctuary. The results will be validated using side-scan sonar and coring sediment samples. In Years 3, 4 and 5, S3AM inventory monitoring and harvesting will be evaluated in active bottom leases of the three coast regions. During a single season, multiple scans of the same beds will be performed to track growth of individuals, confirm live and dead animals, and determine how growth rates vary within a single farm. An animal size and location map will be created to obtain a smart harvesting dredge route that enables growers to more efficiently target harvest of market-size animals and limit disturbance of young, smaller animals. The size distribution and catch per unit effort of smart harvest dredge routes will be statistically compared to traditional harvesting (dredging indiscriminately back and forth within the lease).
The team will access the economic feasibility of S3AM, and identify economic barriers and opportunities through modeling. First, the team will engage with national stakeholders to develop economic cost models to reflect farming practices on the East, West, and Gulf Coasts. In addition to the annual cost models for each region/scale cost model, an investment analysis will be developed. The economic cost models and the associated performance metrics will constitute the base against which the economic performance of S3AM will be measured. Second, the team will input the performance data from S3AM lab test and field trials into each of the cost models developed. A number of metrics will be compared between models with and without S3AM. Effects of region and production scale on whether S3AM results in improved economic outcomes for farmers will be examined. In addition, analysis will be performed to evaluate the potential of providing S3AM as a service through a business instead of requiring each separate farm to purchase the equipment.