Abstract
The objective of this study is to develop an algorithm to compute hydrologically sound, grid-based digital elevation model (DEM) from contour line data. Three commonly used methods, profiling method, bivariate interpolation and inverse distance method, are compared with respect to not only the interpolation accuracy but also occurrence of the spurious pits which make hydrological modeling difficult. From the result of this comparison, a new method is developed based on the profiling method by taking into account the position of valleys and ridges. In delineation of channel network and development the distributed hydrological models, the DEM derived by using this new method reduces the spurious pits significantly.
