Characterizing errors in digital elevation models and estimating the financial costs of accuracy

Digital topographic models are the foundation of more advanced modeling applications and ultimately inform planning and decision making in many fields. Despite this, the error associated with these models and derived attributes is commonly overlooked. Little attention has been given in the scientific literature to the benefits gained from having less error in a model or to the corresponding cost associated with reducing model error by choosing one product over another. To address these gaps in knowledge we evaluated the error associated with five digital elevation models (DEMs) and derived attributes of slope and aspect relative to the same attributes derived from LiDAR data. We also estimated the acquisition and processing costs per square kilometer of the five test models and the LiDAR models. We used three measures to characterize model error: (1) root mean square error, (2) mean error (and standard deviation), and (3) area of significant elevation error. We applied these measures to DEM products that are used extensively across a range of applications for planning and managing natural resources. We depicted the relationship between model accuracy (the inverse of error) and cost in two ways. One was accuracy/cost ratio for each model. The other used separate data on accuracy and cost to better guide potential users in choosing between models or deciding on necessary expenditure on models. The main conclusion of our work was that accounting for error in DEMs can inform choice of models and the need for financial outlays.