Pattern selection and biomass shifts in a vegetation model with nonlocal competition and Allee activation

arXiv:2611.08215 · nlin.PS, q-bio.QM · Submitted 2026-11-05 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Nonlinear Sciences. The day in nonlinear science: chaos, patterns, networks and adaptation.

Kai: I'm Kai.

Mira: And I'm Mira, here with Lev.

Kai: We're looking at the paper: "Pattern selection and biomass shifts in a vegetation model with nonlocal competition and Allee activation".

Mira: Here's the gist of the paper. Pattern selection and biomass shifts in a vegetation model with nonlocal competition and Allee activation study a single-species vegetation model in which a strong Allee effect acts locally and competition acts through a nonlocal kernel, and we describe the kernel with the one-parameter family of q-Gaussians, which goes from compact support to power-law tails.

Kai: Model Formulation The model considers the biomass density u(r, t) of a single species governed by a logistic growth with a cooperative (Allee) factor and nonlocal competition.

Mira: The local growth equation is given by du/dt = b/(1 - u/K)u + c - d, where the factor u/(u + c) increases with density for u ≲ c, representing facilitation among individuals.

Kai: To build a spatial model, diffusion is represented by D∇2u and the competition factor (1 - u) is replaced by its nonlocal average ¯u = G∗ u.

Paper summary: Kai: So we've seen that the core of this study is about how these two factors—the local Allee activation and the nonlocal competition—work together to select a final pattern shape Mira they use this model to show that you can get patterns even with certain types of nonlocal interaction kernels Lev.

Kai: The paper claims that the quadratic coefficient Q is what really organizes the morphology, meaning whether you get spots or gaps, and also how much the total biomass shifts when a pattern forms Mira.

Kai: They show that for Q greater than zero, you get spots, and for Q less than zero, you get gaps Lev. This means that depending on the parameters of your model—like mortality rates or diffusion coefficients—you can control the resulting spatial structure with this single coefficient Kai.

Mira: That's what matters because it links the microscopic kernel shape to macroscopic pattern selection and biomass distribution Lev it shows that even with nonlocal competition, you still have a mechanism for pattern choice that depends on the local activation strength Kai.

Lev: From my side, what interests me is how they derive those amplitude equations for spots and gaps and then they verify them against steady states using these specific q-Gaussian kernels Mira it's important to see if those coefficients hold up when you move away from the ideal cases, like when the kernel has a power-law tail Lev.

Conclusion: Kai: So to wrap up, this paper by Dimatz and Kuperman looks at how local growth rules, specifically that Allee effect, interact with long-range competition kernels Mira the main point is that the resulting pattern isn't just determined by simple diffusion or simple competition; it's governed by this Q coefficient Lev.

Kai: It tells us that if you tune your model parameters so Q is positive, you get spots and the total biomass actually increases because of how those spots organize Mira. If Q is negative, you get gaps and the total biomass decreases Lev.

Kai: What this means for us outside the paper is that when we think about real systems, like biological growth or even maybe some physical processes in materials science, we can't just assume a simple competition kernel works for everything Mira you have to consider if those interactions are local or nonlocal because that dictates whether you get spots or gaps Lev.

Mira: Exactly. The key insight here is that the local activation isn't just some small detail; it fundamentally changes the stability of the system and allows for these finite-wavelength patterns even when competition is spread out Kai it suggests that when modeling complex spatial systems, you need to explicitly account for how local facilitation affects the long-range competitive forces Lev.

Lev: I'm just thinking about what this means practically—if we were trying to design a system where we wanted a dense, spotted structure versus a sparse one with gaps—the whole outcome hinges on that Q value and not just the raw strength of the competition kernel Kai it shows you need to look at both the local growth dynamics and how that feeds into the nonlocal averaging Mira.

F. Dimatz, M. N. Kuperman

Universidad Nacional del Sur · FiEstIn · Centro Atómico Bariloche · Instituto Balseiro · Consejo Nacional de Investigaciones Científicas y Técnicas

nlin.PS, q-bio.QM

Submitted: 2026-11-05

Updated: 2026-11-05

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Importance score: 74/100

The gist: Pattern selection and biomass shifts in a vegetation model with nonlocal competition and Allee activation study a single-species vegetation model in which a strong Allee effect acts locally and…

Key concepts

Allee Activation
This is a cooperative factor in growth where individuals benefit from higher local density up to a certain threshold. It means that at low densities, the species grows faster due to facilitation among neighbors, which can lead to instability and pattern formation.
Nonlocal Competition Kernel
Competition is not just between immediate neighbors but is averaged over a wider area using a kernel function (like q-Gaussians). This nonlocal interaction means that the density of one part of the vegetation affects growth everywhere else, introducing spatial complexity into the model.
Quadratic Coefficient Q
This coefficient arises from nonlinear terms in the amplitude equations and is crucial for pattern selection. Its sign determines whether the resulting patterns are spots or gaps. Furthermore, it dictates biomass shifts: positive Q leads to increased total biomass (spots), while negative Q leads to decreased total biomass (gaps).
q-Gaussian Family
The nonlocal competition kernel is specifically modeled using a family of q-Gaussians. These functions vary from having compact support (competition only over a small distance) to power-law tails, allowing the model to explore different spatial interaction ranges.

Terminology

Summary

Pattern selection and biomass shifts in a vegetation model with nonlocal competition and Allee activation study a single-species vegetation model in which a strong Allee effect acts locally and competition acts through a nonlocal kernel, and we describe the kernel with the one-parameter family of q-Gaussians, which goes from compact support to power-law tails. p. 0

Model Formulation

The model considers the biomass density u(r, t) of a single species governed by a logistic growth with a cooperative (Allee) factor and nonlocal competition. The local growth equation is given by du/dt = b/(1 - u/K)u + c - d, where the factor u/(u + c) increases with density for u ≲ c, representing facilitation among individuals. p. 1 To build a spatial model, diffusion is represented by D∇2u and the competition factor (1 - u) is replaced by its nonlocal average ¯u = G∗ u. p. 1 The dimensionless parameters derived are α = c/K and β = d/b. p. 1

Linear Stability Analysis

The linear stability analysis yields the growth rate λ(k) = -Dk2 + A + B Gˆ(k). The term A represents a local activation, a direct consequence of the Allee effect, while BGˆ(k) is a nonlocal inhibition due to competition. p. 2 A mode of wavenumber k is unstable when λ(k) > 0, i. p. 2e., when D(u0 + α)k2 + u02 Gˆ(k) < αβ. p. 2 The onset occurs at a critical wavenumber kc where λ(kc) = λ'(kc) = 0.

Amplitude Equations and Morphology Selection

The nonlinear terms up to third order are given by N2 = a2ϕ2 + b2 ϕ (G∗ ϕ) and N3 = a3ϕ3 + b3 ϕ2 (G∗ ϕ). The quadratic coefficient is Q = 2a2 + b2Gˆ(kc). p. 4 This coefficient controls the morphology, as spots, labyrinths and gaps appear with pairwise competition only as Q changes sign. p. 0 The sign of Q depends on whether Gˆ(kc) exceeds the kernel-independent value G∗ (α), resulting in Q > 0 for spots and Q < 0 for gaps. p. 8 p. 8

Total Biomass Shift

The total biomass M = ∫u dr obeys dM/dt = ∫F(u, u¯) dr, and a stationary state requires the spatial average of F to be zero. The mean shift of the density is given by ⟨u⟩ − u0 ≃ Q2λ(0) / 2⟨δu2⟩. p. 5 This prediction increases the total biomass for Q > 0 (spots) and decreases it for Q < 0 (gaps). p. 0

Conclusion

The local activation allows a finite-wavelength instability for any kernel of the q-Gaussian family, including those with G >ˆ 0 and infinite variance. The quadratic coefficient Q organizes both the morphology of the patterns and the change of the total biomass. p. 0 The sign of Q decides between spots and gaps for different normalizations, mortality rates, diffusion coefficients and distances to threshold. p. 8 The strength of the gaps follows Q and σ, and not the tail exponent. p. 8 The total biomass increases for spots and decreases for gaps. p. 0 The parameters for which the sequence of morphologies appears lie close to the fold of the homogeneous state (αc/αs ≳ 0. p. 892 for β = 0.15). The amplitude equations predict the morphology wherever the cubic truncation holds, and the gaps persist, with a skewness close to that of an ideal hexagonal pattern, where it fails. p. 8 The sign of Q still decides between spots and gaps for different normalizations, mortality rates, diffusion coefficients and distances to threshold. p. 8 The total biomass increases for spots and decreases for gaps with a leading-order shift proportional to Q and to the variance of the pattern. p. 8 p. 8 The sign of Q depends on whether Gˆ(kc) exceeds the kernel-independent value G∗ (α), resulting in Q > 0 for spots and Q < 0 for gaps. The pattern selection is determined by the ratio σ/Q2, weighted by the cubic coefficients, and not on Q alone. p. 6 p. 6

Pruning redundant reasoning tokens

The quadratic coefficient of the hexagonal amplitude equations, which we obtained in closed form and verified to better than 1%, results from the balance between the curvature of the local growth, which favors spots, and the competitive inhibition at the pattern scale, which favors gaps.

How the training signal is built

The quadratic coefficient Q is defined by Q = 2a2 + b2Gˆ(kc). This coefficient results from the balance between the curvature of the local growth, which favors spots, and the competitive inhibition at the pattern scale, which favors gaps. p. 0

Where the gains come from

The sign of Q depends on whether Gˆ(kc) exceeds the kernel-independent value G∗ (α), resulting in Q > 0 for spots and Q < 0 for gaps. The pattern selection is determined by the ratio σ/Q2, weighted by the cubic coefficients, and not on Q alone. The total biomass increases for spots and decreases for gaps with a leading-order shift proportional to Q and to the variance of the pattern.

The quadratic coefficient Q is defined by Q = 2a2 + b2Gˆ(kc).

The sign of Q depends on whether Gˆ(kc) exceeds the kernel-independent value G∗ (α), resulting in Q > 0 for spots and Q < 0 for gaps.

The total biomass increases for spots and decreases for gaps with a leading-order shift proportional to Q and to the variance of the pattern.

The sign of Q depends on whether Gˆ(kc) exceeds the kernel-independent value G∗ (α), resulting in Q > 0 for spots and Q < 0 for gaps.

The total biomass increases for spots and decreases for gaps with a leading-order shift proportional to Q and to the variance of the pattern.

The sign of Q depends on whether Gˆ(kc) exceeds the kernel-independent value G∗ (α), resulting in Q > 0 for spots and Q < 0 for gaps.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Improved ability to predict pattern morphology based on kernel properties: The system can determine whether spots or gaps will form by comparing a single quantity, the quadratic coefficient Q, to a threshold G star (G star is defined as G star equals G asterisk (alpha)).

  2. Improved total biomass estimation: The improved system can predict how the patterned state changes the total biomass by using Equation 19, which states that "the pattern increases the total biomass for Q > 0 (spots) and decreases it for Q < 0 (gaps)."

  3. Improved understanding of morphological selection criteria: The AI can predict whether hexagons or stripes will appear in every simulated state where the cubic truncation holds by evaluating the sign of Q, which is defined as Q = 2b2[G star - Gˆ(kc)].

Sources

Related papers