Multifractal analysis of 3D images of tillage soil

Recently, X-ray microtomography (CT) has open a new way to study soil pore structures. However, CT data originally comes as gray scale images and the selection of segmentation method to binarized it has an important influence in the structure characterization. Three soil CT 3D images, corresponding to ploughed soil with different tillage tools, were used in this study. A multifractal analysis was applied to the original gray images avoiding image binarization to characterize and differentiate different soil structures. In this analysis we took into account the effect of image resolution and the subdivision method applied, Cube Counting (CC) and Gliding Cube (GC). Comparisons among the multifractal spectrums (MFS) estimated indicated that the reduction in resolution affected more in Molboard image in which the minimum scale included in the analysis was one voxel and it was almost imperceptible in the Chisel image. The three multifractal spectrums were quite distinctive when the maximum resolution was chosen and the GC method was used, meanwhile the CC method obtained the MFS of Chisel and Roller too close to be differentiated. The amplitude and symmetry of the multifractal spectrums point out the influence of each tillage tool in the hierarchy of soil structure. Moldboard creates a higher complexity in soil structure as it physically removes the soil. Chisel tends to destruct the aggregates homogenizing soil structure and presents a weak multifractal nature. Finally, Roller is an intermediate case with a scaling character mainly in the lower gray values of the soil image. (C) 2017 Elsevier B.V. All rights reserved.