Oceanic islands can be relatively isolated from overfishing and pollution sources, but they are often extremely vulnerable to climate and anthropogenic stress due to their small size and unique assemblages that may rely on a limited larval supply for replenishment. Vulnerability may be especially high when these islands bear permanent human populations or are subjected to regular or intermittent fishing. Since the late 1970's, Brazil has been establishing marine protected areas (MPAs) around its four oceanic island groups, which concentrate high endemism levels and are considered peripheral outposts of the Brazilian Biogeographic Province. In 2018, the Brazilian legally marine protected area increased >10-fold, but most of the ~1,000,000 km2 of MPAs around Brazil's oceanic islands are still unknown and unprotected. Here, we provide the first detailed quantitative baseline of benthic reef assemblages, including shallow and mesophotic zones, of the Fernando de Noronha Archipelago (FNA). The archipelago is partially protected as a no-take MPA and recognized by the UNESCO as a World Heritage Site, but also represents the only Brazilian oceanic island with a large permanent human population (3,000 people), mass tourism (up to 90,000 people per year) and a permanent small-scale fishing community. The influence of depth, wave exposure, and distance from the island and shelf edge on the structure of benthic assemblages was assessed from benthic photoquadrats obtained in 12 sites distributed in the lee and windward shores of the archipelago. Unique assemblages and discriminating species were identified using Multivariate Regression Trees, and environmental drivers of dominant assemblages' components were evaluated using Boosted Regression Trees. A total of 128 benthic taxa were recorded and 5 distinct assemblages were identified. Distance to the insular slope, depth and exposure were the main drivers of assemblages' differentiation. Our results represent an important baseline for evaluating changes in benthic assemblages due to increased local and global stressors.