*2.3. Natural and Artificial Structures*

On each of the 8 reef sites used for this study, 1 soft coral sea rod, and 2 healthy, 2 diseased, and 2 recently dead hard coral colonies were selected as the natural structures for further monitoring (Figure 2). Thus, a total of 8 soft coral sea rods, 16 healthy, 16 diseased, and 16 dead hard coral colonies were tracked across our 8 reef sites. The hard coral colonies included five common boulder coral species (*Colpophyllia natans, Montastraea cavernosa, Orbicella faveolata, Porites astreoides,* and *Siderastrea siderea*), whereas the soft coral was a species from the family Plexauridae. "Sea rod", "dead", "diseased", and "healthy" were used as terms to refer to these different types of natural structure. In 2018, each of these corals was tagged, photographed, and analyzed using ImageJ software to estimate the

percent of live, diseased, and dead coral tissue. Estimated percent cover of dead, diseased, or live tissue was converted into surface area using the surface area formula of a half-dome (2 πr2), with the radius estimated as half the mean of the height, length, and width of the coral. In 2019, the resulting 56 corals were re-photographed and re-analyzed to evaluate any changes in tissue cover and surface area. Of the 16 diseased hard corals infected with Stony Coral Tissue Loss Disease (SCTLD) in 2018, 15 had survived and were classified as "healed coral colonies" for the purpose of the 2019 survey.

**Figure 2.** Reef fishes' use of structures was monitored on natural and artificial structures. Natural structures (**top**) included sea rods, and dead, actively diseased, and healthy boulder corals. Artificial structures (**bottom**) included a control and three structures that were built to mimic historically abundant natural structures including soft coral, boulder coral, and branching coral (elkhorn coral, *Acropora palmata*).

Separately from the survey of the existing corals, at each reef site, 4 artificial reef structures (ARs) were deployed, 1 of each of 4 different types, variously created out of PVC piping, concrete, rope, tomato cages, and Vexar mesh (plastic coated wire grid) (Figure 2). The 4 structures were designed to imitate different types of natural habitats. The first type consisted of just 1 single length of PVC pipe standing up from a concrete block; this was a structure that lacked physical complexity. The other artificial structures represented a soft coral (created with PVC arms and frayed rope), a boulder coral (created with Vexar mesh around a tomato cage), and a branching elkhorn coral (created with arms wrapped in Vexar extending out from the center) (Figure 2). These different types of artificial structure were referred to as "control", "soft", "boulder", and "elkhorn" respectively. These 4 artificial structures were designed to assess which characteristic (holes, surface area, edge space), if any, influenced use by the reef fish community. These artificial structures were similar in height (100 cm) and diameter (20–100 cm) to the natural boulder hard corals (height 25–200 cm and diameter 30–150 cm) and natural sea rod soft corals (height 30–125 cm and diameter 30–100 cm) present on the reef. In 2018, the artificial structures were deployed, left to acclimate for 2 weeks, surveyed, then removed from the site. In 2019, the artificial

structures were redeployed for 2 weeks and surveyed again before being removed. The 2-week deployment and removing schedule was set to minimize the amount of biotic build up on the artificial structures as well as to avoid the potential for reef damage from breakage during the hurricane season.
