Accurate mapping of dominant seagrass species is an important aspect of supporting ecosystem monitoring and sustainable coastal zone management. This study aimed to integrate high-resolution PlanetScope SuperDove imagery with field observation data to map the spatial distribution of dominant seagrass species around Barrang Lompo Island, Makassar, Indonesia. Three machine learning algorithms, Random Forest (RF), Support Vector Machine (SVM), and k-Nearest Neighbor (kNN), were applied and compared to evaluate classification performance. The results showed that the total seagrass coverage reached 52.32 ha, with Thalassia hemprichii identified as the dominant species by the RF and SVM models (covering 20.85 ha and 31.95 ha, respectively), whereas the kNN model identified Cymodocea rotundata as the dominant species. Accuracy evaluation indicated that RF provided the best performance with an overall accuracy of 73% and a kappa coefficient of 0.57, compared to SVM (56%; 0.25) and kNN (60%; 0.36). These findings demonstrate that utilizing high-resolution PlanetScope imagery combined with machine learning approaches is effective for seagrass mapping at the species level in complex shallow waters. The generated distribution maps quantify the extent of dominant seagrass species, providing baseline data for blue carbon estimation, biodiversity assessment, and coastal habitat management.