Journal of Coastal Research, Volume 30, Issue 5: pp. 942 – 953, September 2014

By Barry W. Eakins and Pamela R. Grothe

“Digital elevation models (DEMs) support a wide variety of uses, including modeling of surface processes, habitat mapping and conservation planning, coastal change and terrain analysis, and Earth visualization and exploration. These models may, however, contain significant deviations from the surface they are intended to represent, which could reduce their usefulness. Additional complexities arise when integrating bathymetric and topographic data to create coastal DEMs. We identify common challenges in building square-cell, coastal DEMs and present some solutions. These challenges are grouped into six general categories: (1) source data, (2) data processing, (3) model development, (4) model assessment, (5) morphologic change, and (6) model uncertainty.

Example of topographic creep introduced by gridding interpolation. (A) Color image of the GEBCO 1-minute global relief model along the northwest coast of Iceland. Note how the data-poor fjords have positive elevation values in the model due to the predominance of surrounding positive topography. (B) Color image of the same area in the ETOPO1 1-minute global relief model. Bathymetric data, in the presence of a coastline (red), were gridded first in order to interpolate negative elevation values into the fjords.

“Some DEM best practices to help improve DEM accuracy and utility include: visual inspection of source data in a geographic information system (GIS) environment; establishing common horizontal and vertical datums; using data buffers and bathymetric presurfaces; assessing DEM accuracy; accounting for morphologic change; and quantifying DEM uncertainty at the cell level.”

• More information