Designing visualizations of spatial data in 3D terrain is challenging because various heterogeneous data aspects need to be considered, including the terrain itself, multiple data attributes, and data uncertainty. It is hardly possible to visualize these data at full detail in a single image. Therefore, this thesis devises a scalable visualization approach that focuses on relevant information to be emphasized, while less-relevant information can be attenuated. In this context, a noval concept of visualizing spatial data in 3D terrain and different soft- and hardware solutions are proposed.