Considering elevation uncertainty for managing probable disasters

Abstract

The existence of elevation errors in the Digital Elevation Models (DEMs) usually is ignored, during spatial analysis of risk assessment and disaster management problems. As a result, conclusions are extracted, decisions are taken and actions are designed and executed, while the problem is examined on a wrong point of view. This paper describes the attempt to introduce a new model, the DEEM (Digital Elevation Error Model), which incorporates elevation uncertainty and accompanies a DEM uniquely. The use of an uncertain DEM, combined with a probabilistic soft decision approach, eliminates the risk of taking decisions that do not imply to the real problem's basis. Research has shown deviations existing in results, up to 20-50% for volume measurements, area measurements, definition of boundaries, visibility calculation, etc., from those of a hard decision approach. The absence of an integrated GIS, able to manage data uncertainty, forces for by-pass approaches of the problem but not the appropriate ones. © 2005 Springer-Verlag Berlin Heidelberg.