Error analysis on grid-based slope and aspect algorithms

Slope and aspect are the most frequently used surface geomorphic parameters in terrain analysis. While derived from grid DEM, the parameters often display noticeable errors due to errors (a) in data, (b) inherent in data structure, and (c) created by algorithms. It has been observed that some controversial results were reported in evaluating the results by various slope and aspect algorithms, largely because of the variety in assessment methodology and the difficulties in separating errors in data and those generated by the algorithms. This paper reports the study that assesses and compares the results from numerous grid-based slope and aspect algorithms using an analytical approach. Tests were made based on artificial polynomial surfaces which can be defined by mathematical formulae, with controllable "added" data errors. By this approach, different algorithms were quantitatively tested and their error components were analyzed. Thus, their suitability and tolerance related to DEM data characteristics can be described.