Analysis of the behavior of three digital elevation model correction methods on critical natural scenarios

Abstract

Abstract Study region The methods explored in this study were tested in two study areas: Italy and Cuba. Study focus Virtually all Digital Elevation Models (DEM) contain flat areas or depression pixels that may be artifacts or actual landscape representations. These features must be removed before any further hydrological application can proceed. Diverse algorithms have been developed for the purpose of correcting these aspects, differing in how they handle the nature of the depressions, as well as the adopted mathematical procedures. In the present work, the behavior of a standard ( Fill ) and two advanced ( TOPAZ and PEM4PIT ) DEM correction methods on three critical natural scenarios is analyzed. Extensive flat areas, abrupt slope changes and large depressions − expressed in terms of: (1) geomorphological changes (elevation, affected area and slope); (2) flow velocity; (3) river network and width functions (WF) − are affected. New hydrological insights for the region Results confirm improved performance of the advanced methods over the standard method for each case study in Italy and Cuba. The analyzed parameters also show that correction processes are strongly influenced by the relief, the size of the predominating depressions and the neighbouring depressions. There is no one method among those compared which works optimally for every type of correction, and given that the majority of basins have diverse topographical conditions, a different approach to the corrections process and its computational procedures is likely needed.