Effect of Gridding Method on Digital Terrain Model Profile Data Based on Scattered Data

Abstract

Various forms of data such as profile, cross section, volume, height, and surface area can be derived from the digital terrain model (DTM). Although many research studies were carried out on the accuracy of the DTM, especially on grid data-based DTMs and on the effect of the grid size or grid resolution, the accuracy of profile data derived from a scattered data-based DTM and the effect of the gridding (interpolation) method on the accuracy of the profile data derived are still unfamiliar to many DTM producers and users. Therefore, the purpose of this study was to determine the accuracy of DTM profile data based on scattered data points for several gridding methods. Linear interpolation in triangles and interpolations of Sibson (natural neighbor), nearest neighbor, weighted average, polynomial, multiquadric, and minimum curvature were tested on five theoretical test surfaces. After 1-m grid DTMs were generated using the gridding methods mentioned above, nine different profile data in the y direction for each test surface and for each gridding method were derived from the DTMs. The standard deviation of each profile for each gridding method was derived by differences between true and interpolated profile values. As a result of the case study, the author recommends using interpolations of minimum curvature and multiquadric curvature for more accurate DTM profile data.