Comparison of IDW, Kriging and orographic based linear interpolations of rainfall in six rainfall regimes of Ethiopia

Six rainfall regimes of Ethiopia covering an area of 1.13 million km2. Hydrological studies in mountainous terrain require high-resolution spatiotemporal rainfall data. Individual linear interpolation based on elevation, aspect and slope, multivariate linear interpolation,inverse distance weighted, and kriging were employed in six regimes. The performance of the six methods were assessed through cross validation of rain gauge stations and ENACTS data. We adopted different coefficients for six rainfall regimes with respect to orographic liner interpolations. The results confirmed the importance of six rainfall regimes to carry out analysis of the interpolations. This is verified by the fact that performance of the products derived from the approaches differed from regime to regime. Elevation based interpolation is the dominant in regime I and IV. For regime II and III respectively slope and aspect based interpolations are performed well, whereas kriging interpolation is exhibited better in both regime V and VI. It is difficult to get a single best rainfall interpolation method for the Ethiopia heterogeneous land surface. Thus, this study recommends elevation based interpolation to be employed in the mountainous areas where as Kriging can be applied in the area where orographic and local climatic variability are insignificant. For interpolation of rainfall stations, it is essential to apply the methods after classify the study regions based on rainfall variability.