Comprehensive empirical data to inform benthic species distribution models for marine hard-substrate-dominated environments, are pivotal. However, such data are difficult to obtain. These data are crucial to the definition and demarcation of protected areas and for assessment of the ecological status and function of hard-substrate habitats. In this study, underwater video-observations of hard-substrate habitats within four target areas in the sand-dominated German Bight (SE North Sea) were investigated to obtain comprehensive information on hard-substrate distribution patterns, on the amount and sizes of stones and on the presence of sessile organisms. Based on three size classes (cobbles, boulders, large boulders) three hard-substrate distribution classes were identified: (1) widely scattered stones, (2) accumulations of stones and (3) dense stone fields. The ratios between cobbles, boulders and large boulders differed significantly between the investigated areas. Boulders and large boulders were largely colonized by sessile organisms, whereas cobbles in coastal areas were least frequently colonized. Physical disturbances of epibenthos resulting from abrasion and coverage by mobile sediments are discussed as a possible explanation for the proportional differences in the colonization of stones. Hard substrates in shallower, coastal areas appeared to be strongly influenced by sand abrasion because of higher current velocities and storm-induced waves. In deeper areas, located further offshore, disturbances caused by migrating sandy ripples mobilized by storm-events seemed to be more relevant. Habitat modelling of hard substrates and resultant ecological studies require sound information on the probability of epifaunal colonization for different substrate sizes, hard-substrate distribution patterns combined with hydrodynamic and physicochemical properties of the marine environment to produce valid results. We used a structured approach for the video-based analysis of hard-substrate habitats and present estimates of the colonization probability of differently-sized stones. Our study shows that the analysis of drift videos provides basic data at a suitable resolution to contribute to the monitoring and modelling of marine ecological processes.