Abstract: A number of shipwreck archaeological sites worldwide have underlined the importance of shipwreck localization and detection. Accidents that led to sinking are one of the possible causes of those shipwrecks. The shipwreck of MV Bahuga Jaya, which is located in the Sunda Strait, Indonesia could be such an example. A multibeam swath survey is a suitable technique to map the wreck location since it can produce high-resolution Digital Elevation Model (DEM) and backscatter imagery. Both the analysis of the bathymetry DEM and backscatter use visual examination. However, morphometric analysis of the DEM and texture analysis of the backscatter, subsequently combined with the machine learning classification, could give a preferable result in shipwreck detection and monitoring. In this paper, slope analysis of DEM bathymetry and texture analysis of multibeam backscatter imagery are presented. Those first-order textural features are used to carry out a Support Vector Machine (SVM) classification to separate between the wreck and non-wreck objects. A combination of SVM classification and slope analysis is investigated to detect the wreck location. Following that, K-means clustering is also performed to obtain the seabed characterization. Results indicate that the combination of machine learning and morphometric analysis can give a promising outcome in shipwreck detection. In addition, the result of K-means clustering reveals that soft seabed is more dominant than the hard seabed in the study area with 56.4% and 43.6% respectively. This study could play a role as a complementary tool to monitor and manage the shipwreck archaeological site location.