A Comparative Study of Interpolation for Mapping Soil Physical Properties

Abstract

The choice of an optimal interpolation technique for estimating soil properties at unsampled location is an important issue in site-specific management. The first objective of this study was to evaluate Inverse Distance Weighting (IDW) method, Ordinary Kriging(OK) method to determine the optimal interpolation method for mapping some soil physical properties. The second objective was to analyze the relationships between statistical properties of the data and performance of the interpolation techniques .The relationships between statistical properties of the data and performance of the methods were analyzed using soil test saturated hydraulic conductivity, bulk density and volumetric water content data from a fallow field in Hillia city/Iraq. The results suggest that (OK) method has a preference on (IDW) method in estimating and mapping the soil bulk density. On other hand, the (IDW) method has the preference on the ordinary kriging method in estimating and mapping of saturated hydraulic conductivity and volumetric water content. The accuracy of the method in estimating and mapping the soil physical properties using the goodness of prediction(G) and mean square error(MSE) as criteria, related very well to the regression coefficient (R²) for the fitted line of measured and estimated values of soil physical properties. Regression coefficient(R²) of fitted line for the measured and estimated were higher for (IDW) than that of (OK) for hydraulic conductivity and water content, whereas the (R²) was higher for (OK) then that of (IDW) for bulk density. All studied soil physical parameters were strongly spatially dependent, but the range of spatial dependence varies within the soil physical properties. Bulk density had the shortest range of spatial dependence (6.3 m) and the saturated hydraulic conductivity had the longest range (15.4m).