Characterization by X-ray μCT of the air-filled porosity of an agricultural soil at different matric potentials

To describe various important soil processes like the release of greenhouse gases or the proliferation of microorganisms, it is necessary to assess quantitatively how the geometry and in particular the connectivity of the air-filled pore space of a soil evolves as it is progressively dried. The availability of X-ray computed microtomography (mu CT) images of soil samples now allows this information to be obtained directly, without having to rely on the interpretation of macroscopic measurements using capillary theory, as used to be the case. In this general context, we present different methods to describe quantitatively the configuration of the air-filled pore space in 3D mu CT images of 20 separate samples of a loamy soil equilibrated at different matric potentials. Even though measures using mu CT on such multi-scale materials strongly depend on image resolution, our results show that in general, soil samples most often behave as expected, for example, connectivity increases with higher negative matric potential, while tortuosity decreases. However, simple correlations could not be found between the evolution of quantitative descriptors of the pore space at the different matric potentials and routinely measured macroscopic soil parameters. A statistical analysis of all soil samples concurrently confirmed this lack of correspondence.

Automated summary

In order to understand soil processes such as greenhouse gas release and microbial proliferation, it is important to quantitatively assess the changes in soil pore space connectivity as the soil dries. X-ray computed microtomography (μCT) can provide direct information on this, eliminating the need for interpretation based on macroscopic measurements using capillary theory. This study presents different quantitative methods to describe the configuration of the air-filled pore space in 3D μCT images of loamy soil samples. Results show that, in general, soil samples behave as expected, with increasing connectivity and decreasing tortuosity as the matric potential becomes more negative. However, no simple correlations were found between quantitative descriptors of pore space and macroscopic soil parameters.