Jill Elder

Evaluating the effects of resampling digital elevation models in a viewshed analysis

ABSTRACT

Digital elevation models (DEM) characterize the topography of the earth's surface through regularly spaced estimates of elevation at horizontal intervals. They are used in applications such as determining surface runoff to estimate non-point pollution, describing hydrologic features such as channel width, measuring soil erosion, describing terrain features and calculating the viewshed in architectural space. To be effective in these application contexts, analysts are frequently required to resample the values to change the horizontal and/or vertical spacing of the elevation estimates. The rationale for changing the spacing between the elevation values include to aid in more efficient rendering and to better match the resolution of affiliated data. The downside of resampling is that additional error is potentially introduced. We investigate the impact of three different resampling techniques on the results of a viewshed analysis. Evaluating the sensitivity between the resampled DEMs will indicate where error can be introduced. Also, comparing multiple viewshed analyses from the resampled DEMs might better point towards how introduced error affects subsequent analyses.