Issue link: https://read.dmtmag.com/i/53167
L ight detection and ranging (LIDAR) datasets are tricky. Everyone knows they have value, but get- ting that value out is another issue. In 2006, for example, Monroe County, N.Y., as part of a Federal Emergency Management Agency (FEMA) map modernization, purchased a LIDAR flight. The flight deliverable contained all the different returns, but, unknown to the department, not the software knowledge to make the data usable. The raw data comprised 509 1.5-mile square tiles. The first diffi- culty was finding how to use and convert the raw x,y,z,i data to raster format for analysis. After some basic research, a limited size of LIDAR data could be converted to raster using ArcGIS' raster- interpolation tool, such as Natural Neighbor. This tool was a great start and worked fine for many projects. It had a limit of about nine tiles at a time, which worked well for site-area projects. The issues began when LIDAR's usability in hydrological modeling was ques- tioned. The nine-tile limit wasn't able to cover enough of a watershed, and, when attempting to mosaic the larger tiles, a seam was created. A new process had to be found. Initial Projects, Initial Hurdles During the LIDAR delivery, Institute for the Application of Geospatial Technology Inc. (IAGT) provided a quality- analysis/quality-control (QA/QC) report. The county then contracted with IAGT to provide the initial raster and contour coverages for the LIDAR data. The pro- cessing was resource intensive, and the computers used were substantially more powerful than what the county owned. The processing took more than a week, even with the higher-powered machines. The first project included a large-scale redesign of State Route 590 in northern Monroe County. The proj- ect was complicated, as the goal was to convert an expressway to local road with four roundabouts. The consultant was GIS savvy and an Esri business part- ner, and the project area was about 10 square miles. The contour Shapefiles were only 50MB in size. Unfortunately, the project didn't stay in the "GIS world." The data were transferred to a CAD environ- ment, where the same Shapefile turned out to be a Imagery/LIDAR Special Issue 600MB dwg file. Needless to say, this project didn't use the LIDAR data. After taking the processed LIDAR data, the next major hurdle was getting people to use the data. Much of the projects that LIDAR benefited were in the New York Department of Transportation (DOT), but there was a stigma of "that's how we've always done it." The county surveyor saw a benefit in LIDAR, so he helped get LIDAR into the program. The project started small, spot-checking different areas surveyed against LIDAR data. The initial project was for the Attridge Road Bridge Replacement. The LIDAR data fell within the same error as the QA/QC report, but actually per- formed much better than anticipated in open areas. Problem areas were along the bridge and in the tree cover. The edge of the bridge was an issue due to the size of the pixels. The survey measurement was at the top of the bridge, while the LIDAR data averaged the bridge's top and bottom. Another problem was using GIS; points at equal or rand intervals were generated, Figure 1. An analysis for Attridge Road shows three different transect profiles comparing the survey to LIDAR data. The graph with the most differences is next to the bridge, and the other two are in areas around the site. JANUAR Y 2O12 / WWW . GEOPLA CE . COM 15