Anisotropic considerations while interpolating river channel bathymetry

摘要: Numerical modeling of flow dynamics in river channels may require the spatial interpolation scattered measurements bathymetry elevation to obtain elevations at computational mesh nodes. This paper examines various methods used interpolating channel bathymetry. Commonly methods, namely inverse distance weighting, spline (tension and regularized), natural neighbor, TopoGrid, ordinary kriging (isotropic anisotropic) are evaluated within ArcGIS environment by using root mean square error (RMSE) criteria. To study anisotropic effects, two different coordinate systems: a Cartesian (x, y) system flow-oriented (s, n) system. Using data from Brazos River Texas, it is shown that neighbor kriging, which do not account anisotropy data, performed better than In system, performance significantly (40% reduction RMSE) compared isotropic methods. A modified version weighting method, called elliptical (EIDW), developed for anisotropy. The RMSE results application EIDW close thus providing simple computationally faster alternative complex