Multifractal properties of porosity as calculated from computed tomography (CT) images of a sandy soil, in relation to soil gas diffusion and linked soil physical properties

Abstract

Relationships between soil porosity and diffusive gas flux are poorly understood, partly because of a difference in measurement scales between the two. The complexity of soil pore systems can be described by multifractal analysis at the microscopic scale, whereas relative soil gas diffusion coefficients (Ds/Do) are usually evaluated at the core scale. The objectives of this study were to (i) define a quantitative ‘pseudo‐macroporositygas’ from high‐resolution X‐ray computed tomography (CT) scanning images and characterize it for 10 intact soil cores, (ii) analyse the frequency distribution of pseudo‐macroporesgas in the columns with a multifractal approach and (iii) assess relationships between Ds/Do measured at the core scale and multifractal parameters describing the pore system heterogeneity within a core. The shape and symmetry of the singularity spectra and the degree of curvilinearity of the Rényi spectra show that the multifractal behaviour of the pseudo‐macroporositygas distribution for a given CT image thresholding varied among soil columns. Correlations found between Ds/Do and some parameter estimates of the singularity spectrum suggest that the distribution of pseudo‐macroporesgas, depending on the CT image thresholding, influenced Ds/Do. In particular, a strong correlation between Ds/Do and the entropy dimension (Dq = 1) indicates that Ds/Do was influenced by the degree of spatial heterogeneity of the pseudo‐macroporositygas distribution. The correlation dimension (Dq = 2) was also linked to Ds/Do, suggesting that a second‐order power law might describe the scaling relationship between pseudo‐macroporositygas distribution and Ds/Do. In conclusion, the multifractal description of soil porosity as calculated from CT images may be regarded as a way to improve our understanding of gas movement in soils at the core scale.

Burrow refilling behavior ofAporrectodea turgida(Eisen) andLumbricus terrestrisL. as revealed by X-ray computed tomography scanning: Graphical and quantitative analyses

“…Following Lafond et al (2012), after trying several thresholds for CT numbers, we considered a voxel with dimensions 0.3 )0.3 )0.3 mm to be (part of) a ''pseudomacropore'' if the associated CT number was below (500 Hounsfield unit (HU), that is, mid-way between the calibrated values for pure air ( (1000 HU) and pure water (0 HU); the threshold of (500 HU theoretically describes space with 50% air'50% water, for example. Thus, the term ''pseudo-macropore'' is justified because CT scanning does not measure porosity directly and soil pores are not restricted to perfect cubes or parallelepiped rectangles.…”

“…The high-resolution X-ray CT scanner in our study was used by Lafond et al (2012). The CT scanning parameter values were: 100 mA (tube current), 120 kV (tube voltage), 1 mm (X-ray beam width), 18 cm (fieldof-view diameter), and 1.2 (zoom factor).…”

A High‐Resolution Surface Mapping Procedure for Soil Cores Using Computed Tomography Scanning Data

“…Of the technologies available for soil porosity evaluation in controlled laboratory studies, which include neutron imaging (e.g., Moradi et al, 2013) and magnetic resonance imaging (e.g., Pohlmeier et al, 2013), X‐ray computed tomography (CT) scanning offers interesting avenues that have already proved useful for the analysis of biopores (e.g., Mooney, 2002; Gregory et al, 2003; Lafond et al, 2012; Jones et al, 2013; Udawatta et al, 2013; Whalen et al, 2015). When the 10‐cm scale of observation (the size of the CT scanned sample) is used in combination with a resolution scale (voxel size) of 100 μm, a CT scanning study is said to be “high resolution” by Ketcham and Carlson (2001), whereas, the appellation microtomography is restricted to observation and resolution scales of 1 mm and 1 μm, respectively.…”

“…A major difficulty in reconstructing porosity in soil cores from CT scanning data comes from the unevenness of the soil surface, which often translates into strong spatial heterogeneity (i.e., large patches of soil pixels interspersed with large patches of air pixels) in the top CT images, which represent, by definition, horizontal sections of the CT scanned soil core. Typically, such CT images are either totally deleted from or totally included in subsequent analyses; this was the case in the studies of Lafond et al (2012) and Whalen et al (2015), to name a few. Although this does not affect the description of the pore space in the volume where it is investigated, when totally deleted it then becomes practically impossible to know how many of those pores are connected to the soil surface and thus allow gas exchange with the atmosphere; when totally included, the result is an overestimation of porosity at the soil–atmosphere interface.…”

The environment qualities evaluation of a degraded area in recuperation and the drying process of a Brazilian soil using nonconventional methods at micrometric scale

“…The version used in this study was a result of partnership between CIP and Embrapa Instrumentation, which was used to calculate a) entropy and b) degree of multifractality and asymmetry of the system (JORGE et al, 2008). The system entropy calculation is derived from multifractal concept, which is appropriate to characterize mass spatial arrangement of complex system, at same time, it is able to solve local density problem LAFOND, et al, 2012).…”