Geo-visualization for geo-science education
Abstract (summary)
This thesis provides a study of terrain-rendering techniques and technology for the purpose of conveying geo-scientific information in the form of an educational, interactive, and scientifically representative geo-visualization. Specific techniques investigated include Level-Of-Detail (LOD), fractal, and voxel-based terrain rendering. The case study involves the use of high resolution LIDAR data of the McMurdo Dry Valleys of Antarctica to test the techniques used by specific software packages. These include Virtual Terrain Project, with LOD capabilities, NGRAIN, with voxel-based rendering, and the Virtual Reality Modelling Language (VRML), a programming language for creating three-dimensional objects and environments using LOD.
Through this research it has been found that three steps are common to all practices: (1) Data conversion and processing; (2) Terrain population with information and objects; and (3) Terrain Run-Time Viewing. The common limitation is the apparent trade-off between interactive quality and visual quality.
Conclusions are that LOD and Fractal based rendering are most suitable to maintain a balance between interactive and visual quality. In addition it was found that the proper use of landmarks and metaphor provide the best elements of design for Virtual Learning Environments (VLEs). VRML offers the best means to place landmarks, incorporate multi-media elements, and links to other sources of information while encouraging learning through exploration. VRML was therefore used in creation of the McMurdo Learning Module; a VLE teaching geomorphology using the high resolution LIDAR data.*
*This dissertation is a compound document (contains both a paper copy and a CD as part of the dissertation).
Indexing (details)
Science education;
Physical geography
0714: Science education
0366: Geography