Fish community data were collected at the location of the oyster castles and at a control area nearby. Samples were collected using a 5.5-meter (18 ft) bag-less seine net (dimensions: 5.5 m long, 1.2 m deep x 3.2 mm sq. mesh). Monitoring was performed around either a low- ebbing or low flooding tide during daylight hours from July-October in 2015, 2017-2019, 2021 . The net was actively towed across the bottom in 0.3-1 meters of water for a distance of about 25 meters. Non-reef reference site samples were taken by either pulling the net parallel to shore for approximately 20 meters then turned and pulled perpendicularly toward the shore or towed perpendicular to shore for 25 meters. To sample the treatment areas, the net was towed around the footprint of a reef structure or along one side of several structures. The seine net pole closest to the reef was kept in contact with the structure while the net was in motion. Both site types were along a contiguous shoreline with similar vegetation, however, sediment composition and other parameters were not identical. Hence, we use the term ‘non-reef reference site’ rather than ‘control site’ (Bailey et al., 2004).

Each animal was identified to species or lowest possible taxon and up to 30 individuals of each species were measured (TL or FL for fish, carapace length for crabs, and length from rostrum to tail for shrimp). Four tows each were made on reef structures (treatment areas) and non-reef reference areas for a total of eight per sampling event. A few deviations to this sampling plan were made during unexpected tidal conditions where less than eight tows were made. The exact seine sample locations varied slightly throughout the duration of this multi-year project. Though all tows were made within the intertidal zone, exact distance from shore was based on site location and tidal stage.