Environmental variation influences fish assemblage structure; however, fish assemblage composition shifts due to natural or anthropogenic stressors have been observed in marine ecosystems worldwide, altering ecosystem structure and function, and fisheries sustainability. Previous research in waters off North and South Carolina (USA) was limited in scope and repeatability or did not quantify fish assemblages at a scale suitable for monitoring composition shifts. Concurrent chevron traps, underwater video, and environmental data from an annual fisheryindependent survey were used to characterize the environment and enumerate demersal reef fishes caught in the traps, and priority fish species observed in video in depths similar to 15-110 m. Multivariate analyses detected assemblage patterns and environmental influences. An 8-variable (distance to shelf edge, depth, consolidated substrate size, latitude, percent biotic cover, temperature, undercut height, biotic class) model predicted assemblages, while 4 variables explained the greatest variation (distance to shelf edge [19%], depth [15%], consolidated substrate vertical relief [4%], size [4%]). The largest number of discriminator species occurred in mid- to outer shelf areas with greater substrate complexity (i.e. increasing substrate size and relief). Assemblages dominated by Centropristis striata at depths (similar to 15-40 m) with little substrate complexity transitioned towards assemblages dominated by Pagrus pagrus, higher prevalence of larger-bodied predators, and invasive Pterois spp. at depths (similar to 40-110 m) with greater substrate complexity. Understanding baseline assemblage characteristics and natural drivers of variability in fish assemblage structure is vital to conservation and management efforts that monitor changes in population abundance, the presence/absence of key species, and stressor-induced modifications of local assemblages, which are all measures of ecological health that underpin comprehensive assessments and management.