*2.2. Surveys and Sampling*

Surveys were conducted using the standard Reef Check survey method (see http: //reefcheck.org/ecoaction/Monitoring\_Instruction.hp, accessed on 30 August 2018) [39]. A 100 m transect was divided into four 20 m long segments, each separated by a 5 m interval from the next one, so that the four segments could be considered as replicates. On each segment, substrate cover was recorded beneath the transect line at 0.5 m intervals, so that 40 data points were recorded per segment (160 per 100 m transect). The transect lines were laid down parallel to the shore, along the long axis of the reef. The work was undertaken using SCUBA at 2 to 6 m depth, and snorkel at 1 to 2 m depth. The substrate was recorded using the following categories (and codes): hard coral (HC), soft coral (SC), dead coral with macroalgae or nutrient indicating algae (NIA), recently killed coral (RKC), sponge (SP), coral rubble (RB), hard substrate of dead coral more than one year old covered by turf algae or encrusting coralline algae (RC), sand (SD), silt (SI), and others (including sea anemones, tunicates, gorgonians, or non-living substrate) (OT) [39]. Digital photography and videography were employed (using a GoPro Digital Hero 5 Black, GoPro Inc., (San Mateo, CA, USA) to provide a visual record of the reef habitats and coral assemblages along the transects. In addition, images of live coral specimens were recorded with a dedicated underwater camera (Canon PowerShot A620, Canon, USA).
