The goal of this PhD is to create a new method for creating realistic artistdirected terrain. Terrain generation has been developed for over 40 years with fractals and erosion simulation being among the most popular methods. With the ever-increasing demand for more realism in film and games, current terrain generation algorithms are no longer suitable and are being overlooked in favour of hand modelling. I am proposing to investigate new algorithms to create large scale realistic terrain that can be artist-directed by creating a general terrain system. I will also address the problems concerning iterative terrain editing and converting heightmaps into highly detailed polygon meshes.