Graduation Year

2006

Document Type

Thesis

Degree

M.A.

Degree Granting Department

Geography

Major Professor

Paul Zandbergen, Ph.D.

Co-Major Professor

Robert Brinkmann, Ph.D.

Committee Member

Mark Ross, Ph.D.

Keywords

GIS, LIDAR, DEM, terrain analysis, geomorphometry

Abstract

Terrain attributes computed from Digital Elevation Models (DEMs) are widely used in hydrology and hydrologic modeling. It is important to consider that the values of the attributes can be different depending on the resolution of the DEM from which they are derived. The question arises as to how much exactly the high-resolution DEMs created through LIDAR remote sensing techniques change the values of the terrain attributes when compared to lower resolution DEMs.In this thesis a LIDAR-derived DEM of 20 feet resolution was resampled using a nearest-neighbour algorithm to various coarser resolutions to examine and quantify the effect of DEM resolution upon a series of hydrologically significant terrain attributes including slope, surface curvature, topographic wetness index, stream power index and stream networks. Values for slope and surface curvature are found to be smaller when computed from lower resolution DEMs; values for the topographic wetness index and stream power index are found to increase as DEM cell size increases.The derived stream networks for each resolution were compared in terms of length per stream order, drainage density, bifurcation ratio, and overall accuracy indicating a loss of small detail, but only a modest change in the overall stream network morphometry. This research suggests that it is possible to establish relationships that quantify the effects of DEM resolution upon hydrologically significant terrain attributes, which can then be considered when processing DEMs from various resolutions for the purpose of parameterizing hydrologic models.

Share

COinS