Delineating Hummocky Terrain: A Field-Based Approach

Abstract

A geomorphological study was conducted to determine the most accurate and efficient number of primary data points necessary for modelling hummocky terrain with subtle elevation changes. Primary elevation data were collected on hummocky terrain using an S320 GNSS survey receiver, interpolated using ordinary kriging in ArcGIS, and analyzed using a Monte Carlo simulation in MATLAB. This analysis was done to suggest an ideal number of sample points required to produce a highly accurate digital elevation model. From research conducted in Georgetown, ON, it was determined that 20,000 points should be sampled per square kilometer, but the findings can be altered slightly to best suit future geomorphological and hydrological studies. The findings will increase the understanding of the subtle relationship between topography and interpolation error and will guide future data capture and modelling of terrain with subtle elevation variation of less than 10 m. Results can influence the scope of point collection and the clustering of points or boreholes across terrain with variable elevation changes. There is also a potential for application in regions where placing boreholes would be difficult or costly when trying to understand subsurface geometries.