Seagrass meadows derived from field to spaceborne earth observation at Green Island (Wunyami), a reef habitat in the Cairns section of the Great Barrier Reef, November 2020

Seagrass meadow extent and meadow-scape was mapped using two alternative approaches at Green Island, a reef clear water habitat, in the Cairns section of the Great Barrier Reef, in November 2020. Approach 1 included mapping seagrass meadow-scape (including patches and scars) using imagery captured during low spring tides with a DJI Mavic 2 Pro UAV at an altitude of 100 m (85% sidelap and frontlap) on the 25 November 2020. The orthomosaic of the captured images was created in PIX4D and the resolution was 2.45cm/pixel. Approach 2 used PlanetScope Dove imagery captured on 05 November 2020 coinciding as close as possible to the field-surveys from 25 to 27 November 2020, with 3.7 m x 3.7 m pixels (nadir viewing) acquired from the PlanetScope archive. For Approach 1, spatially explicit seagrass maps were created, including seagrass cover, from the UAV nadir imagery using deep-learning techniques. The abundance classes used were: (1) absence of seagrass (0%), (2) low seagrass cover (1-15%), (3) medium seagrass cover (>15-50%), and (4) high seagrass cover (>50%). For Approach 2, we created spatially explicit seagrass maps from PlanetScope Dove imagery, and conducted the classification using a machine-learning model (Random Forest) coupled with a Boot-strapping process (100 iterations). The final model predictions were then gathered into separate rasters, based on Bootstrap Probability thresholds of 60% and 100%. The final rasters were cleaned using a majority filter algorithm, to eliminate stray pixel predictions using a moving window between 3 and 9 pixels depending on the size of the imagery.