Digital elevation model grid size, landscape representation, and hydrologic simulations

High-resolution digital elevation data from two small catchments in the western United States are used to examine the effect of digital elevation model (DEM) grid size on the portrayal of the land surface and hydrologic simulations. Elevation data were gridded at 2-, 4-, 10-, 30-, and 90-m scales to generate a series of simulated landscapes. Frequency distributions of slope (tan B), drainage area per unit contour length (a), and the topographic index (a/tan B) were calculated for each grid size model. Frequency distributions of a/tan B were then used in O'Loughlin's (1986) criterion for predicting zones of surface saturation and in TOPMODEL (Beven and Kirkby, 1979) for simulating hydrographs. For both catchments, DEM grid size significantly affects computed topographic parameters and hydrographs. While channel routing dominates hydrograph characteristics for large catchments, grid size effects influence physically based models of runoff generation and surface processes. A 10-m grid size provides a substantial improvement over 30- and 90-m data, but 2- or 4-m data provide only marginal additional improvement for the moderately to steep gradient topography of our study areas. Our analyses suggest that for many landscapes, a 10-m grid size presents a rational compromise between increasing resolution and data volume for simulating geomorphic and hydrological processes.