Modelling Soil as a Living System: Feedback between Microbial Activity and Spatial Structure

Riz Fernando Noronha, Kim Sneppen, Kunihiko Kaneko · 2025

Read Modelling Soil as a Living System: Feedback between Microbial Activity and Spatial Structure by Riz Fernando Noronha, Kim Sneppen, Kunihiko Kaneko (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2502-20662.txt.

Introduction

Soil behaves like a living system in a model by Riz Fernando Noronha, Kim Sneppen, and Kunihiko Kaneko in a 2025 paper. Soil is a complex, dynamic material whose physical traits depend on its biological content. The authors propose a cellular automaton where soil aggregates serve as food for microbial species, which in turn produce nutrients that aid their own growth.

That loop creates a cycle of consumption and regeneration across space. Three main results stand out. First, spatial structure supports a stable living state and prevents collapse or runaway growth. Second, the spatial model allows parasitic species to coexist without driving the system extinct. Third, the best growth arises when soil structure varies across diverse length scales. Heterogeneity, the authors argue, is key to resilience.

Catalog record

CREATORRiz Fernando Noronha, Kim Sneppen, Kunihiko Kaneko
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE28 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2502-20662

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Modelling Soil as a Living System: Feedback between Microbial Activity and Spatial Structure Riz Fernando Noronha, Kim Sneppen, Kunihiko Kaneko Abstract Soil is a complex, dynamic material, with physical properties that depend on its biological content. We propose a cellular automaton model for self-organizing soil structure, where soil aggregates and serves as food for microbial species. These, in turn, produce nutrients that facilitate self-amplification, establishing a cyclical dynamic of consumption and regeneration. Our model explores the spatial interactions between these components and their role in sustaining a balanced ecosystem. The main results demonstrate that (1) spatial structure supports a stable living state, preventing population collapse or uncontrolled growth; (2) the spatial model allows for the coexistence of parasitic species, which exploit parts of the system without driving it to extinction; and (3) optimal growth conditions for microbes are associated to diverse length scales in the soil structure, suggesting that heterogeneity is key to ecosystem resilience. These findings highlight the importance of spatio-temporal dynamics of life in soil ecology. I Introduction Soil is of fundamental importance to many facets of life: it forms the basis of agriculture, construction, ecosystems, and much more. It provides plants with the environment and necessary nutrients to grow, thus providing the world with food, oxygen, and energy. Over half of the species on earth are estimated to call the soil their home Anthony et al. 2023 , leading to remarkably complex and interesting ecosystems. Soil is not a physico-chemical material in contrast to sand or rocks, but is shaped through an interplay between organic process and material. Through motility, consumption, and chemical secretion, organisms including a vast number of microorganisms shape the physical structure of soil, whereas such structure affords growth of organisms living in it. Ettema and Wardle Ettema and Wardle 2002 describe how species in soil are neither homogeneously distributed nor random, but instead show spatial heterogeneity and patterning. This spatial heterogeneity plays a role in determining the spatial distribution of plant communities as we observe them above the soil. Soil appears to be a self-organizing substance, with properties that are linked to underlying microbial activity Crawford et al. 2012 ; Neal et al. 2020 ; Feeney et al. 2006 . While a lot of research has investigated how an ecosystem can affect the soil structure, there remains a paucity of studies exploring co-occurring interactions of how soil structure could affect the ecosystem’s survival. It may be possible that species living in soil shape its structure in a way that benefit their continued existence. This paper introduces a simple model to explore the mutual feedback mechanism between physical spatial structure and the microbial ecosystem. A key idea is chemical secretion and cooperation, where a replicating individual needs to feed on nutrients generated by other individuals Baran et al. 2015 ; D’Souza et al. 2018 ; Embree et al. 2015 . II Simple Model and Simulation Soil consists of multiple different species and chemicals. To simplify, we first develop a model with one top-level organism, one nutrient, and a single soil state. The top-level organism is denoted a microbe, and represents some abstract biological entity that could be a protist or some other slightly larger creature. We consider four states: [S]oil, [N]utrient, [M]icrobe, and [E]mpty. The model can be described with the following dynamics: - Microbes convert soil sites into nutrient sites - Microbes consume nutrient sites to reproduce - Microbes die at a constant rate \theta - Soil sites ‘grow’ into neighbouring empty and nutrient sites at rate \sigma The soil ‘growth’ can be justified by considering soil as a living material that includes soil aggregation, as noted by Tisdall and Oades Tisdall and Oades 1982 . In this language, one may think of empty and nutrient states as sites containing ‘dust’ of non-aggregated inert matter. Figure 1: A): Schematic diagram of the model. S,E,N,M represent Soil, Empty, Nutrient, and Microbe states respectfully. Arrows indicate transitions, and the text above them the rates. B): Snapshots of the 2D lattice at different time steps, showing dynamic spatial behaviour. By disregarding spatial heterogeneity, the system can be described with the following mean-field equations: \displaystyle\frac{\mathrm{d}S}{\mathrm{d}t} \displaystyle=\sigma S\left(E+N\right)-MS (1) \displaystyle\frac{\mathrm{d}E}{\mathrm{d}t} \displaystyle=\theta M-\sigma SE (2) \displaystyle\frac{\mathrm{d}N}{\mathrm{d}t} \displaystyle=MS-MN-\sigma SN (3) \displaystyle\frac{\mathrm{d}M}{\mathrm{d}t} \displaystyle=MN-\theta M (4) If we add Equations 1 , 2 , 3 , and 4 , we get \frac{\mathrm{d}S}{\mathrm{d}t}+\frac{\mathrm{d}E}{\mathrm{d}t}+\frac{\mathrm{d}N}{\mathrm{d}t}+\frac{\mathrm{d}M}{\mathrm{d}t}=0 S+E+N+M=\mathcal{C} (5) where \mathcal{C} is a constant, determined by the initial conditions. This is a 3 variable system, as a fourth variable can be easily eliminated through Equation 5 . In addition, this introduces a carrying-capacity set by \mathcal{C} into the system: if the system is composed of mostly soil, then further growth of soil in the system ( \sigma SE ) goes as \sim S(\mathcal{C}-S) , which corresponds to a standard Verhulst or logistic equation \mathrm{d}S/\mathrm{d}t\approx S(1-S) Verhulst 1838 . The same carrying capacity acts on the microbe, as MN\sim M(\mathcal{C}-M) . The model can thus be thought of as a modified predator-prey model Lotka 1920 , with a time delay (as the microbes consume soil in a two-step process: first converting it to nutrients, and then eating them to reproduce) and a carrying capacity. In particular, without the nutrient step (if the microbe directly consumes soil to survive), we see no sustained oscillations. Furthermore, the phase diagram would have no bifurcations, and the fraction of microbes smoothly varies as we change the parameters \sigma and \theta . Figure 2: A): Phase diagram of the attractors of the mean-field model over soil filling rate \sigma and microbe death rate \theta . B): Attractors of the corresponding red points in A), in state space S , E , M represent Soil, Empty and Microbe populations. Fixed points are represented by circles, while limit cycles (oscillations) can be seen as closed loops. Using the above equations, we can construct a mean-field phase diagram, visualized in Figure 2 . The phase diagram contains three bifurcations: a Hopf bifurcation boundary, where the stable limit cycle changes to a stable fixed point, a saddle-node bifurcation on the right, beyond which the system dies out, and a saddle-node on invariant curve (SNIC) bifurcation at the bottom, which again makes survival impossible. Below the SNIC bifurcation, we can observe that the oscillations have too large an amplitude, which leads to the system falling into the pure soil absorbing state 1 1 1 Due to the bifurcations, we can section the phase diagram into four distinct regimes. Note that two fixed points exist everywhere, namely the pure soil absorbing state (as without microbes, soil cannot be destroyed) and the pure empty absorbing state (as without soil, empty states will only proliferate). The first is the area to the right, where there are no fixed points beyond the two trivial ones. Then, we come across the saddle-node bifurcation, which introduces two fixed points: an unstable saddle point, and a stable fixed point (which allows for species survival). Crossing the Hopf bifurcation causes the stable fixed point to swap stability and become unstable, but creates a stable limit cycle. Finally, crossing the SNIC boundary to the bottom causes the stable limit cycle to collide with a nullcline and fall into the trivial pure-soil state, and thus we have only

Read the complete text