Soil landscape models at different scales portrayed in virtual reality modeling language

Most state-of-the-art manipulations and visualizations of soil data use geographic information systems to portray soil landscapes in two dimensions (2-D). Nevertheless, soil attributes are distributed continuously in three dimensions (3-D) across landscapes. The objective of this study was to investigate the use of Virtual Reality Modeling Language (VRML), a 3-D graphics language suitable for stand-alone or browser-based interactive viewing, to create 3-D soil landscape models at different scales. Four different locations in southern Wisconsin were selected to represent pedon, catena, catchment, and soil region scales. Soil data, including texture, cone index, and depth of soil layers, were used in conjunction with topographic attributes to create 3-D soil landscape models. Spatial modeling techniques comprised 2-D and 3-D ordinary kriging. We used Environmental Visualization Software (EVS) to export the geometry of 3-D objects, which were enhanced to include: (i) viewpoints, (ii) Munsell colors, (iii) texture maps, (iv) 3-D cross-section animation, (v) animations such as zooming and rotation, and (vi) primitive shapes to highlight areas of interest. Virtual reality modeling language is capable of describing and visualizing extremely complex shapes, such as complex soil layers or terrain. Visualization of Munsell soil colors was difficult to implement because there is no hardware or software independent color-management system available in VRML. Animation techniques were valuable to high-light specific characteristics of each model. The accessibility of interactive VRML models via the World Wide Web and the portability of these models across platforms facilitate the entry of soil science into the virtual world of cyberspace.