The low-altitude economy has emerged as a strategic emerging industry in China, involving economic activities within airspace below 1,000 meters, such as manned/unmanned cargo/passenger transport. This three-dimensional economic form faces core challenges in ensuring flight safety, optimizing operational efficiency, and achieving large-scale commercialization, primarily due to the complexity of low-altitude environments characterized by dense urban structures, dynamic meteorological factors, and mixed aircraft operations. This demands innovative digital management solutions. A low-altitude three-dimensional data space, defined as a digitally integrated environment for organizing, managing, and applying multi-source heterogeneous data within this airspace, serves as a critical foundation for supporting safe, efficient, and intelligent development. This study proposes a comprehensive low-altitude three-dimensional data space framework grounded in China's National 3D Realistic Geospatial Landscape Model (3dRGLM). The framework establishes a spatiotemporally integrated digital environment supporting critical applications in low-altitude airspace planning, intelligent 3D navigation, and safety-guaranteed airspace management. It is structured around entity-based digital modeling of low-altitude elements, multi-source spatiotemporal data fusion via grid-discretized management, and trusted data circulation mechanisms. By leveraging 3dRGLM's capabilities in realistic 3D representation and dynamic geographic entity association, the proposed data space enables end-to-end digital twin modeling of low-altitude operational scenarios. Pilot programs in Wuhu and Deqing demonstrate the framework's effectiveness in enhancing decision-making precision for air route network design, 3D dynamic navigation, and risk-aware flight control, confirming its value as a replicable model for scalable low-altitude economy development.