Effect of DEM uncertainty on topographic parameters, DEM scale and terrain evaluation

Digital Elevation Models (DEMs) are representations of topography with inherent errors that constitute uncertainty. DEM data are often used in analyses without quantifying the effects of these errors. DEMs are a commonly utilized standard data source for GIS based studies aimed at evaluating natural processes. This research developed a methodology to quantify uncertainty in the DEM and derived parameters of slope, upslope contributing area, and the topographic index (TI). The methodology was implemented within the ESRI ArcView Spatial Analyst™ GIS environment and utilizes Monte Carlo simulations to quantify DEM uncertainty using random error fields. Different methods for simulating random error were developed. The methodology and resulting tool were applied to investigate: (a) effects of uncertainty on elevation and derived topographic parameters, (b) effects of uncertainty on DEMs of different scale (30m and 10m resolution), and (c) effects of uncertainty in flat and varied terrain. Parameter estimator statistics enabled quantification of DEM error. Results demonstrate the applicability of the methodology in quantifying and depicting the impacts of DEM random error. DEM parameters were affected by random error; this effect varied with the method in which random error was represented. DEM error manifested itself differently in DEMs of different scales. At higher grid resolution, slope and TI were more susceptible to DEM error than elevation and upslope area. In the study that compared uncertainty in flat and varied terrain, slope and upslope area grids were more sensitive to uncertainty than elevation and topographic index grids with a more pronounced effect in flatter areas. Each study demonstrated a bias contributed by sink filling in lower valley areas that could have implications for research that relies on DEMs for parameters in these areas such as agricultural watersheds.