With the increasing demand for marine resource surveys, traditional marine data management models face challenges such as unscientific classification, incomplete systems, and information silos. To address these issues, this study develops a marine 3D spatiotemporal database and information platform, achieving an upgrade from single-element, static management to comprehensive, dynamic management. The platform integrates multi-source heterogeneous data and designs a four-layer database architecture (marine environmental layer, natural resource layer, management layer, and basic geographic information layer). It also constructs 2D and 3D spatial entity representation models along with a spatiotemporal evolution model to support full lifecycle tracking of resources. Leveraging microservices technology and knowledge graphs, the platform enables semantic association, data provenance, and systematic management of marine natural resource survey and monitoring data. By establishing a full-process data management system, it effectively resolves issues like data fragmentation and sharing difficulties, enhancing the digital and intelligent management capabilities of marine resource data. This provides scientific support for marine resource development, ecological conservation, and rights protection.