This thesis develops algorithms and mathematical concepts for the visualization of curved spacetimes. Special attention is given to methods suitable for analysis of data given purely numerically in the guise of a tensor field originating from large-scale supercomputer simulations. Existing visualization approaches are reviewed for applicability upon relativistic data sets and new ideas for the full graphical representation of tensor fields are explored. A mathematically founded model unifying the treatment of diverse scientific data is presented to support high reusability of implemented algorithms. All methods have a wide range of applicability even beyond general relativity. The technology has been worked out and used in a cooperation project among the department for visualization and data analysis in the Zuse-Institute Berlin (ZIB), the origin of the reknowned amira visualization software, and the Max-Planck Institute for Gravitational Physics (the Albert-Einstein-Institute AEI), which facilitate the world's largest research group in general relativity. Results from this project have appeared in scientific journals as well as in many general public magazines and TV broadcasts such as the Discovery Channel.weiterlesen