Pore-scale modeling of water-phase fragmentation in simulated soils with realistic pore geometry

Bacterial diversity in unsaturated soils is very high and is a principle contributor to an array of functional capabilities exhibited by the soil bacterial community. The main causes of this have been the subject of many research works, which mainly focus on chemical and biological factors. Unfortunately, the ways in which spatial heterogeneity of soil contributes to this high diversity are less explored. However, researchers have found compelling evidence that hydraulic and physical isolation of microbial colonies and cells maintains or enables the diverse richness of microbial species in the soil.

Studies have also shown that soil bacteria have abundance distributions with a few dominant species only under saturated conditions in the soil subsurface zone. In unsaturated soils where the water phase is non-uniformly distributed throughout the pore network, they exhibit even abundance distributions for protozoa and bacteria. This indicates coexistence of a good number of taxa in small soil volumes.

Previous research, which forms the basis of the current paper, indicated a bacterial diversity gradient across soil textures, not attributable to chemical composition of the soil or parent components. The authors therefore hypothesized that varying fragmentation of the water phase in variably textured soils had an impact on microbial diversity that the soil could sustain. The current modelling seeks to shed some light on how the pore-scale architecture of soil affects the fragmentation of the water phase, which subsequently affects potential microbial diversity.

Jessica Furrer at Benedict College and Amvrossios Bagtzoglou at University of Connecticut investigated the fine-scale dynamics of water retention at scales relevant to soil microbes. The authors therefore sought to improve upon current pore-scale models through simulation of realistic 3-D soil particles, which resulted in more realistic pore geometries as well as size distributions. The authors then applied the physical domain as boundary conditions in a lattice Boltz-mann fluid dynamic model. Their research work is published in peer-reviewed journal, Granular Matter.

The authors developed a novel methodology for developing irregular rough-surfaced sand and silt particles in silico with realistic particle-particle contacts as well as pore geometries. They then simulated fragmentation of the water phase in unsaturated soils for different textures implementing single-component multiphase lattice Boltzmann method. They quantified the fragmentation by thresholding the continuous fluid density distribution into gas and liquid. This was then followed by counting isolated liquid pockets.

The simulated soil indicated more pockets count in fine textured soils than course-grained soils. The authors found that for intermediated saturations the coarsest soils offered more capacity for microhabitat development as compared to less coarse soils. They also realized that pocket counts for all the soil textures were dominated by filling of small pores developed by domains of high silt, which led to higher pocket counts seen in finer soils.

Lattice Boltzmann method simulations at this resolution didn’t depict fragmentation adequately owing to liquid films on soil grain surfaces in coarser domains. On the other hand, the sensitivity of pocket counts to density threshold in coarser soils indicated the significance of surface roughness related to water pockets, an aspect that could be simulated in idealized particle geometry.