In the summer of 2010, funding was secured for a LiDAR survey of the CU-Boulder campus as part of a solar energy potential and photovoltaic planning study. Following data acquisition, 20 square kilometers were modeled using only open source tools. While Martin Isenburgâ€™s NSF-funded work on LiDAR processing (LAStools) was used to extract and convert data out of the propriety â€˜.lasâ€™ format, the majority of the processing, visualization and analysis work was done using the Open Source Geospatial Foundation project GRASS GIS. Processing included filtering, raster generation through inverse distance weighting and raster generation using regularized splines with tension. Analysis of insolation was accomplished using the GRASS GIS implementation of the SOLPOS 2.0 (SOLar POSition and intensity) algorithm developed and maintained by the National Renewable Energy Laboratory (NREL). This raster data was combined with preexisting building footprint vector data from the campus GIS system to facilitate site selection of solar candidates. The end result was modeled for display using the NVIZ modeling and visualization system within GRASS, and was then supplemented with orthophotos for draping and comparison. The data visualizations of campus solar sites take into account day length, position of the sun, seasonal atmospheric effects, orientation and the local shading effects from trees, buildings and other structures.
This paper explores current methods of LiDAR visualization within the GRASS project. A strong focus is placed on the creation and display of three-dimensional information, specifically on the role that three-dimensional modeled data is coming to play in GIS moving forward. Included in this research is a look at the current barriers to publishing three dimensional datasets such as this on the web, and a look at possible solutions and their subsequent challenges.