Object-based image analysis supporting site selection in Posidonia oceanica restoration efforts

Posidonia oceanica meadows are a valuable yet vulnerable coastal ecosystem endemic to the Mediterranean Sea, providing key ecosystem services. However, these habitats have declined markedly over the last decades due to multiple anthropogenic stressors. Given their low natural recovery rates, active restoration through transplantation has become a widely adopted strategy to mitigate this loss. Nevertheless, restoration outcomes remain variable and are influenced by multiple factors, with site selection playing a critical role, as it is rarely addressed through structured and systematic approaches, particularly those accounting for the geomorphological context of the meadows. This study contributes to addressing this methodological gap by proposing a geospatial framework that integrates high-resolution acoustic remote sensing, geomorphometric analysis, and Object-Based Image Analysis (OBIA). MultiBeam EchoSounder (MBES) bathymetry and Side Scan Sonar (SSS) backscattering data were used to characterize the spatial extent and morphometric features of P. oceanica meadows. The research was conducted within the RENOVATE project offshore Civitavecchia (central Tyrrhenian Sea, Italy), and resulting habitat mapping was validated through Single Beam EchoSounder-equipped USV surveys and ROV inspections. A semi–automated OBIA workflow was developed to identify patches of bare sediment within the meadow characterized by geomorphological conditions potentially favourable to restoration, based on attributes related to stability, morphology, erosional signatures, and spatial context. The proposed approach introduces a geospatially and quantitatively informed framework to support site selection in P. oceanica restoration. By incorporating measurable geomorphological attributes derived from acoustic remote sensing data, this method provides an objective, reproducible and scalable tool to guide decision making and prioritize areas for further site-specific assessment.