TOPOGRID is a useful tool for generating hydrologically correct surfaces from vector data. One advantage of TOPOGRID is its use of all the available data for generating the surface. This is especially useful in areas of sparse data. However, this algorithm requires the whole data matrix to be in active memory during processing, and for efficiency, physical memory. On many workstations, this uses up memory resources rapidly, as the area of processing increases. Besides increased memory resources, processing time increases in proportion to the area size increase. At many sites, many workstations may go idle at night, or even parts of the day, like over noon, breaks, and meetings. Using the recommended 10 cell overlap, and Inter-Application Communication, (IAC), the TOPOGRID surface generation can be divided into multiple smaller problems, instead of one large problem. Using multiple workstations, instead of just one for TOPOGRID will alleviate some processing time and memory problems it has with large areas. Once a matrix size has been picked for processing, the data is divided and sent to the available workstations, each generating a small portion of the surface. Since the surface is an interpolated value, the GRID function MOSAIC is used to put these multiple smaller surfaces together. The result is a surface generated in overall less time and memory, than running it all on a standalone workstation. Because of the dependence on data for processing, the processing area size should be semi-dynamic, as there must be some data present to perform any type of processing.
Terrain relief information is one of the most important fundamental datasets in GIS, and DEM is increasingly required and used in the spatial analysis and visualization of 3D terrain. While there are many data sources for the generation of DEM in ArcInfo, this paper focuses mainly on digitized contours, which are always available in all kinds of GIS. The purpose of this paper is to find the best way to achieve better representation of 3D terrain surface from the most fundamental data sources--digitized contours. Fractal and GRID based analysis as well as other methods are employed for preparing qualified data for the generation of DEM. Various methods are available in ArcInfo for the generation of DEM from contours both in TIN and GRID data structure, we have a comparative studies on these two methods. As 3D visualization of terrain surface is one of the straightforward criteria in the quality evaluation of DEM, we also discuss the visualization techniques of the DEM at the end of the paper.
The province of British Columbia has thousands of newly available tiles of basemap data at 1:20,000 scale. Surface data, such as D.E.M mass points, breaklines, and hydrographic features are available in ArcInfo, but widely untested. Several surface creation and watershed boundary delineation methodologies using tin, grid, and topogrid are being explored and tested on a sample of tiles. The benefits and drawbacks of each methodology will be explored and a choice employed for production. AML-based techniques for efficient production and aggregation will ensure a highly-portable technique.