Author(s): Author Unknown

Linked Author(s):

Keywords: Digital watersheds; Data fusion; Spatial Interpolation; Chin

Abstract:

The distributed hydrological model based on DEM (Digital Elevation Model) is an attractive approach to studying hydrological processes. In order to study the impact of the changing environment and human activity on hydrological processes, the watershed model must be performed on sub-basins that represent different hydrological response units. The grid input module plays an important role in modeling. As the original input of the distributed hydrological model, the precipitation on each grid must be obtained through interpolation based on the limited observed data, thus the studies on spatial interpolation methods become necessary. The fundamental principals and methodologies of spatial interpolation were introduced in this paper. The interpolation methods included Triangle Linear Interpolation Method (TLI), Gradient plus Inverse Distance Squared Method (GRDS), Modified Reversed Distance Squared Method (MRDS) and Distance Direction Weight Average Method (DDWA). In the case study, many different basins were selected. Various interpolation methods were applied to calculate the spatial distribution of daily, monthly and annual precipitation. The cross-validation method was used to examine their precision and applicability of spatial interpolation approaches. The result indicates that the efficiency coefficient increases slightly with the increase of the complexity of the interpolation method. Different interpolation methods were suitable for obtaining precipitation on each grid element in the different basin. Meanwhile, the complex methods spent much more time on interpolation calculation, which means the calculating efficiency decreases obviously. However, whichever method should be adaptive depends on the topographic and climatic traits of watersheds.

DOI:

Year: 2009