Abstract
Digital soil mapping of soil particle-size fractions (PSFs) using log-ratio methods is a widely used technique. As a hybrid interpolator, regression kriging (RK) provides a way to improve prediction accuracy. However, there have been few comparisons with other techniques when RK is applied for compositional data, and it is not known if its performance based on different balances of isometric log-ratio (ILR) transformation is robust. Here, we compared the generalized linear model (GLM), random forest (RF), and their hybrid patterns (RK) using different transformed data based on three ILR balances, with 29 environmental covariables (ECs) for the prediction of soil PSFs in the upper reaches of the Heihe River Basin (HRB), China. The results showed that the RF performed best, with more accurate predictions, but the GLM produced a more unbiased prediction. As a hybrid interpolator, RK was recommended because it widened the data ranges of the prediction values, and modified the bias and accuracy of most models, especially the RF. The prediction maps generated from RK revealed more details of the soil sampling points than the other models. Different data distributions were produced for the three ILR balances. Using the most abundant component of the compositional data as the first component of the permutations was not considered to be the right choice because it produced the worst performance. Based on the relative abundance of the components, we recommend that the focus should be on data distribution. This study provides a reference for the mapping of soil PSFs combined with transformed data at the regional scale.
Highlights
Spatial interpolation of soil particle-size fractions (PSFs) has become a focus of soil science researchers
Based on our previous work, the objectives of this study were to: (i) compare the spatial prediction accuracy of soil PSFs using a generalized linear model (GLM) and random forest (RF) combined with environmental covariables (ECs) and isometric log-ratio (ILR) transformed data; (ii) determine whether hybrid interpolators (GLMRK and RFRK) can improve the interpolation performance; and (iii) explore the distributions of different transformed data and the variation law of precision based on different choices of sequential binary partition (SBP)
The distribution of data generated from SBP2 or SBP3 had a mirrored symmetry, with a left-skewed ILR1 of SBP2 and right-skewed ILR2 of SBP3 (Figs. 2c and 2d)
Summary
Spatial interpolation of soil particle-size fractions (PSFs) has become a focus of soil science researchers. More accurately predicted soil PSFs could contribute to a better understanding of hydrological, physical, and environmental processes (Delbari et al, 2011; Ließ et al, 2012; McBratney et al, 2002). Key Laboratory of Land Surface Pattern and Simulation, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China. College of Resources and Environment, University of Chinese Academy of Sciences, Beijing 100049, China. The characteristics of compositional data makes soil PSFs more impressive than other soil properties. Soil PSFs are usually expressed as three components of discrete data – sand, silt, and clay, and carry only relevant percentage information. Soil texture is classified as soil PSFs, which can be demonstrated on a ternary diagram (so-called soil texture triangle).
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
