Premium
Spatial Dependence and Interpolation of Soil Properties in West Sumatra, Indonesia: I. Anisotropic Variation
Author(s) -
Trangmar B. B.,
Yost R. S.,
Uehara G.
Publication year - 1986
Publication title -
soil science society of america journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.836
H-Index - 168
eISSN - 1435-0661
pISSN - 0361-5995
DOI - 10.2136/sssaj1986.03615995005000060004x
Subject(s) - topsoil , subsoil , anisotropy , soil science , kriging , spatial variability , geology , geostatistics , weathering , multivariate interpolation , mineralogy , mathematics , geomorphology , soil water , statistics , physics , quantum mechanics , bilinear interpolation
Spatial and temporal interaction of soil‐forming factors and processes determine the distribution of soil properties within the landscape. This study relates anisotropic spatial dependence of particle size fractions, pH and 25% HCl‐extractable P to directional differences in the main soil‐forming factors in Sitiung, West Sumatra, Indonesia. Linear geometric anisotropic models were fitted to pooled directional semivariances for each of these properties. Directions of maximum variation coincided with the main southwest to northeast axis of volcanic tuff fallout, deposition of alluvium and the general sequence of soil weathering in the region. Anisotropy ratios ranged from 1.5 for subsoil pH to 5.2 for subsoil HCl‐P. Topsoil textural components and pH were more variable than in subsoils, having larger sample variances and larger anisotropy ratios. Punctual kriging of topsoil sand content using the linear geometric anisotropic model resulted in low estimation variances in densely sampled areas where weighting of neighbor samples by direction as well as distance took most effect. Quantitative analysis of anisotropic spatial dependence of soil properties can aid interpretation of soil genesis. Estimates of anisotropic spatial dependence can also be incorporated into kriging interpolation for optimal, unbiased spatial estimation with minimum variance.