Abstract

Hydrologists usually gain insights into topographic variability by using a high-performance algorithm to estimate the catchment area from a digital elevation model (DEM) image. In the literature, the grid-based algorithms are more popular than the contour-based algorithms; however, the existing ones cannot reduce the error when higher resolution DEM is available. This paper introduces a new contour-based algorithm, formulated with the physically-based concept, to estimate catchment from contour-based DEM data. The formulation was derived from the semi-analytical solution of Laplace's partial differential equation based on the boundary element method (BEM). With this approach, the algorithm can estimate the catchment area along the smooth surface water paths (SWPs), which are delineated from the physically-based algorithm. The proposed algorithm was validated with the standard synthetic surfaces, where the theoretical specific catchment areas (SCAs) for error assessment are exactly known. When estimating the SCAs with ordinary resolution, the average error of the estimated SCAs is between 14.19 and 2.32%, but the average error from the popular grid-based algorithms is between 77.8 and 17.0 %. With higher resolutions, between 12.5 and 0.88 meters, the proposed algorithm significantly reduces the average error from 15 to 0.8%.