Pith. sign in

REVIEW 3 major objections 4 minor 1 cited by

Blurring the Busse balloon: Patterns in a stochastic Klausmeier model

T0 review · 3 major / 4 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Stochastic forcing blurs the boundary of the deterministic Busse balloon in a dryland vegetation model.

desk verdict Noise blurs the Busse balloon in the Klausmeier model, but the pulse-counting definition of exit time needs a robustness check before I'd trust the quantitative claims. read the letter →

arxiv 2411.13238 v2 pith:N43RZTQE submitted 2024-11-20 math.AP

Christian Hamster , Peter van Heijster , Eric Siero This is my paper · ORCID
classification math.AP MSC 35B3635R6060H15
keywords stochasticreaction-diffusionequationsKlausmeiermodelBusseballoonfirstexittimelocalwavenumberperiodicpatternsdrylandvegetationmultiplicativenoise
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper asks what happens to the Busse balloon—the set of wave numbers that are stable periodic patterns in a deterministic reaction-diffusion model—when the model is forced by noise. Using the one-dimensional Klausmeier model for dryland vegetation, it claims that stochastic forcing blurs the balloon's boundary: a wave number outside the deterministic balloon can have a longer average first exit time than wave numbers inside it. The authors introduce two numerical tools, the average first exit time and the local wave number, and show that stability under noise depends strongly on the rainfall parameter, the noise intensity, and the wave number's position inside the balloon. If true, this means deterministic Busse balloons are only a sharp predictor of observable patterns in the zero-noise limit; with realistic noise, pattern persistence is graded rather than binary.

What carries the argument

Busse balloon: the region in parameter space of deterministically stable periodic wave numbers, traced by numerical continuation. First exit time: the first time the pulse number changes, estimated by averaging over many noise realizations; it mimics deterministic stability in the stochastic setting. Local wave number: the predominant mode of a Gaussian-windowed Fourier transform with width $\ell=50$; it captures both the number of pulses and their spatial distribution. The pulse number itself is counted by smoothing the solution and locating extrema with a prominence threshold, a heuristic that defines when destabilization is said to occur.

What would settle it

Run the $a=2.0$, $\sigma=0.25$ stochastic simulations with the pulse-counting thresholds varied (smoothing width and prominence) and check whether wave number 38 still has a longer average first exit time than wave numbers 23 and 24; if the ordering reverses, the claimed blurring is an artifact of the pulse-counting definition.

Watch

Extended reading notes

Core claim

The paper studies the one-dimensional stochastic Klausmeier model (1.1), a reaction-diffusion system for dryland vegetation with multiplicative noise in the mortality rate. It claims that stochastic forcing does not simply destroy the deterministic Busse balloon but blurs its boundary: for fixed rainfall $a=2.0$ and noise $\sigma=0.25$, the average first exit time of the deterministically unstable wave number 38 exceeds that of the deterministically stable wave numbers 23 and 24, so the observed boundary of the balloon is no longer sharp. The blurring is controlled by position within the balloon and noise level: patterns in the centre survive longest, exit times grow exponentially in the rainfall parameter $a$ and obey a power law in $\sigma$, and the boundary sharpens again in the limit $\sigma\to 0$ and $T_{\max}\to\infty$. The paper also introduces local wave numbers to describe what solutions look like after destabilization, showing that typical stochastic solutions are patchworks of different local wave numbers rather than a single global mode.

Load-bearing premise

The definition of first exit time rests on a heuristic pulse-counting procedure (Gaussian smoothing width 64 gridpoints, prominence threshold 0.3); if a different threshold is used, measured exit times and the reported blurring of the balloon boundary could change.

Editorial extensions

If this is right

  • Within the stochastic model, patterns near the centre of the deterministic Busse balloon are the most persistent, but the deterministic boundary loses its predictive power for observability once noise is present.
  • Reducing the noise intensity increases exit times most strongly in the balloon's interior, so the sharp deterministic boundary is recovered only in the double limit $\sigma\to 0$ and $T_{\max}\to\infty$.
  • Typical stochastic solutions are not close to a single periodic pattern: the stationary distribution of local wave numbers is spread across the balloon and shifts with $a$ and $\sigma$.
  • Because exit times are exponentially sensitive to rainfall and power-law sensitive to noise, the relevant pattern persistence timescale can change by orders of magnitude within the parameter range studied.
  • The three measures of stability peak at different wave numbers (31 for exit time, 30 for deterministic selection, 28 for stationary distribution), so 'most stable' depends on which observable is used.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the blurring mechanism is generic, the same pulse-deletion route should blur Busse balloons in other stochastic reaction-diffusion systems, such as Swift-Hohenberg or Gray-Scott; a direct test would be to repeat the exit-time scan for those models.
  • The measured exponential-in-$a$ and power-law-in-$\sigma$ relations for exit times could be turned into a scaling law for pattern persistence in dryland vegetation, once dimensional rainfall and disturbance intensity are calibrated.
  • A Markov chain on pulse numbers with transition rates estimated from short stochastic runs could convert the blurred balloon into a quantitative prediction of how fast patterns degrade, although the paper notes the state definition is delicate.
  • In two-dimensional stripe patterns, the noise is correlated along the stripe direction in the present one-dimensional setup; in true two dimensions, local defects may nucleate differently, so the blurring could appear at lower noise levels.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 4 minor

Summary. The paper studies the one-dimensional stochastic Klausmeier model (1.1) with multiplicative noise in the vegetation mortality. It defines the first exit time Texit as the first time the number of pulses changes, and it introduces local Fourier transforms to assign local wavenumbers to transient and noisy patterns. Numerical experiments show that average exit times vary strongly with rainfall a, wavenumber k, and noise intensity sigma, and that for a=2 and sigma=0.25 the mean exit time at k=38 exceeds that at k=23 and 24 even though k=38 lies outside the deterministic Busse balloon. The authors also compute long-time local-wavenumber distributions, which concentrate around the middle of the balloon, and conclude that stochasticity blurs the boundary of the deterministic Busse balloon while the boundary becomes sharp again in the limit sigma -> 0 and Tmax -> infinity.

Significance. The manuscript addresses a timely question: whether the deterministic Busse balloon remains a useful stability concept under stochastic forcing. The proposed tools, average first exit times and local wavenumber distributions, are natural and well motivated, and the numerical experiments are direct simulations rather than fitted outputs, which avoids circularity. If the central reversal in Figure 8c is robust, the paper gives a concrete, falsifiable statement about stochastic destabilization in a canonical ecological model, and the open-source scripts on GitHub are a strength for reproducibility. The comparison with empirical Busse balloons from Bastiaansen et al. is a useful sanity check. However, the robustness of the pulse-counting definition and the unsupported stationarity label for sigma=0.2 are load-bearing gaps that need to be addressed before the claims can be accepted.

major comments (3)
  1. [Section 2.3, Appendix A.3, Figure 8c] The central reversal in Figure 8c depends on the pulse-counting detector: Texit is the first time the pulse number changes, and in Appendix A.3 the pulse number is obtained by smoothing u with a Gaussian of width 64 gridpoints and counting extrema with MinProminence 0.3, a procedure the authors themselves call heuristic in Section 2.3. Because k=38 has a much shorter wavelength than k=23 or 24, a fixed smoothing width and prominence threshold can have wavenumber-dependent detection lag: a dying short-wavelength pulse at k=38 may be registered as present for a different length of time than a dying long-wavelength pulse, purely because of the detector. The paper provides no test that the ordering E[Texit(38)] > E[Texit(23)], E[Texit(24)] survives variation of the smoothing width or MinProminence, or comparison with a different destabilization criterion. Please add a robustness study of this specific reversal, for example by recomputing Texit with smoothing widths of 32 and 128 gridpoints and MinProminence values of 0.15 and 0.6, and by checking the same ordering with a criterion based on the local wavenumber distribution or on amplitude decay.
  2. [Section 3.2, Appendix A.4, Figures 9b and 10c] The claim that Figures 9b and 10c show stationary distributions is not supported by the paper's own convergence test. Section 3.2 says that Appendix A.4 validates stationarity, but Appendix A.4 explicitly states that for sigma=0.2 two different initial conditions lead to different distributions at T=Tmax=10^4 and that the simulations have not yet reached the stationary distribution. Nevertheless, Figure 9b is labeled as a stationary distribution for sigma=0.2, and Figure 10c uses the same object to compare stability measures. This internal inconsistency affects the observability conclusions. Please either restrict the stationarity claim to sigma=0.25, or present the sigma=0.2 results as finite-time long-time distributions with a quantitative convergence assessment, and state whether the comparison in Figure 10 changes when only verified stationary cases are used.
  3. [Figure 8c and Section 3.1] Figure 8c reports E[Texit] over 50 realizations and displays the standard deviation of the exit times, but the central claim that the mean exit time at k=38 exceeds those at k=23 and 24 is not accompanied by a standard error, a confidence interval, or a pairwise significance test. With heavy-tailed exit-time data and only 50 runs, the observed reversal may be within sampling error, and the caption's note that standard deviations are added to highlight the high variability makes this concern concrete. Additionally, Texit is right-censored at Tmax, so the fraction of runs that reached Tmax for each k must be reported, since censoring biases the estimated mean differently across wavenumbers. Please provide mean uncertainties and censoring counts for the bars in Figure 8c, and test the ordering of the means.
minor comments (4)
  1. [Figure 9 caption] The caption of Figure 9 states Tmax=2500, while Section 3.2 states Tmax=10^4 for the same computations; please reconcile the two values.
  2. [Figure A.4 and Figure 7b] Figure A.4 claims a power-law relation Texit ~ sigma^alpha with alpha approximately 10 based on a visual log-log fit, and Figure 7b claims an exponential relation between a and the maximum exit time from a straight-line fit; report the number of data points, regression standard errors, and residual diagnostics for both fits, and state how the a=0.40 outlier in Figure 7b is treated.
  3. [Appendix A.3] Appendix A.3 mentions a moving median in time to filter short fluctuations but does not specify its window length; since the pulse number and thus Texit can depend on this parameter, please give the value used.
  4. [Section 2.2] The local Fourier transform window width ell=50 is fixed a priori; a brief discussion of the sensitivity of the local wavenumber statistics to ell would help readers assess the observability claims.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the stochastic blurring claim is a direct numerical measurement compared against an independently computed Busse balloon.

full rationale

The paper's central quantities are direct simulation outputs, not fitted inputs renamed as predictions. Texit is defined as the first time the measured pulse number changes (Section 2.3) and is estimated by repeated numerical integration of the stochastic PDE (1.1); the deterministic Busse balloon boundary in Figure 1 is computed independently with the continuation software auto-07p [10]. The blurring conclusion in Section 3.1 is therefore an empirical comparison between two independently generated objects. The scaling laws in Figure 7b and Figure A.4 are descriptive fits to the simulated exit times and are not used to define Texit, so no fitted parameter is called a prediction. The local wave number method follows [38] and the numerical integrator follows [22]; both are methodological citations with independent content, and self-citations such as [6] and [32] supply only a parameter value or contextual comparison. The acknowledged heuristic in pulse counting ('determining the pulse number in a simulation is in some sense heuristic as different settings in the procedure may result in different pulse numbers', Section 2.3; smoothing details in Appendix A.3) is a measurement caveat, not a circular reduction: no equation defines the Busse balloon boundary in terms of Texit, and the measured ordering in Figure 8c is not equal by construction to the deterministic stability classification.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The central claim rests on heuristic numerical definitions (pulse counting, local window width) whose parameters are chosen by hand, and on the assumption that the numerical schemes faithfully approximate the SPDE. No new entities are introduced.

free parameters (5)
  • Local Fourier transform window width ell = 50
    Defines the spatial window for local wave numbers; not derived from the model, and results might depend on it. Table 1 fixes it.
  • Pulse detection smoothing parameters = Gaussian width 64 gridpoints, MinProminence 0.3
    Used in A.3 to count pulses; authors state the procedure is heuristic and settings affect results.
  • Noise spatial correlation length xi = 0.1
    Sets the spatial correlation of the Gaussian noise W^Q; chosen in A.2 and not varied.
  • Power-law exponent alpha in Texit ~ sigma^alpha = ~10
    Fitted from Fig A.4 for (k,a)=(30,2); reported as 'alpha for alpha ≈ 10' with no confidence interval.
  • Exponential decay rate of max exit time vs a = not stated (slope of line in Fig 7b)
    The paper plots log10(max E[Texit]) versus a and asserts an exponential relation, but the slope is not reported and the fit has no uncertainty estimate.
assumptions (4)
  • domain assumption The stochastic Klausmeier model (1.1) with Ito multiplicative noise is an adequate representation of dryland vegetation dynamics.
    The model is taken as given from prior literature [20,32] and the noise structure is assumed; the paper's conclusions are conditional on this.
  • domain assumption The deterministic Busse balloon computed by continuation (pde2path) correctly identifies stable periodic patterns of the deterministic model.
    The continuation results are used as the reference boundary; the paper does not independently verify their correctness.
  • domain assumption Spatial discretization and time-stepping scheme (A.2) accurately approximate the SPDE solutions, including the noise term.
    The numerical scheme from [22] is standard, but no convergence study is presented for the specific parameters used.
  • ad hoc to paper The distribution of local wave numbers at Tmax is stationary for sigma=0.2 and a up to 2.5, even though A.4 shows non-convergence for sigma=0.2.
    Figures 9 and 10 label these distributions as stationary, while Appendix A.4 admits that simulations have not reached the stationary distribution for sigma=0.2.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Blurring the Busse balloon: Patterns in a stochastic Klausmeier model." pith.science (2026). https://pith.science/paper/N43RZTQE

@misc{pith2026241113238,
  author       = {Pith},
  title        = {Pith review of: Blurring the Busse balloon: Patterns in a stochastic Klausmeier model},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/N43RZTQE}},
  note         = {Machine review of arXiv:2411.13238}
}
read the original abstract

We investigate (in)stabilities of periodic patterns under stochastic forcing in reaction-diffusion equations exhibiting a so-called Busse balloon. Specifically, we used a one-dimensional Klausmeier model for dryland vegetation patterns. Using numerical methods, we can accurately describe the transient dynamics of the stochastic solutions and compare several notions of stability. In particular, we show that stochastic stability heavily depends on the model parameters, the intensity of the noise and the location of the wavenumber of the periodic pattern within the deterministic Busse balloon. Furthermore, the boundary of the Busse balloon becomes blurred under the stochastic perturbations.

Figures

Figures reproduced from arXiv: 2411.13238 by the authors.

Figure 1
Figure 1. We show the normalized histogram of 200 runs per [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Single realization of the stochastic RDE (1.1) for [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Figure (a) shows the predominant wave number, the pulse number, and the rounded [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (8 more)
Figure 4
Figure 4. Figure 4: Simulation of (1.1) with σ = 0 and a = 1.5. As initial condition we took the stationary solution with wave number 30, but set it to 0 on the interval [0, λ] for both components, where λ is the wavelength of the periodic pattern. Figure (a) shows that the remaining 29 p…
Figure 5
Figure 5. Figure 5: Stochastic version of Figure 4 with [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 6
Figure 6. Figure 6: We computed the solution to the stochastic RDE (1.1) 50 times with as initial condition [PITH_FULL_IMAGE:figures/full_fig_p009_6.png]
Figure 7
Figure 7. Figure 7: Figure (a) shows the logarithm of the average first exit time log [PITH_FULL_IMAGE:figures/full_fig_p010_7.png]
Figure 8
Figure 8. Figure 8: Average exit time E[Texit] for a = 2.0 and three different values of the noise σ computed over 50 iterations. The red bars indicate the wave numbers within the Busse balloon, and the blue ones are those outside. For each bar, the standard deviation is added to highligh…
Figure 9
Figure 9. Figure 9: For a section of the Busse balloon, we compute the stationary distribution of local wave [PITH_FULL_IMAGE:figures/full_fig_p011_9.png]
Figure 10
Figure 10. Figure 10: Comparison of three different approaches to stability for [PITH_FULL_IMAGE:figures/full_fig_p011_10.png]
Figure 11
Figure 11. Figure 11: For two sites in Somalia, Bastiaansen et al. [6] determined the predominant wave number of stripe patterns in grids of 100 by 100 meters and determined the slope of the terrain in each grid. These figures indicate that for a single value of the slope, there is no pref…

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. A flow-kick model of dryland vegetation patterns: the impact of rainfall variability on resilience

    nlin.PS 2025-01 conditional novelty 6.0 of 10

    In a flow-kick model of dryland vegetation, random storm timing and depth at fixed mean annual rainfall narrows the precipitation window in which banded vegetation patterns form, compared to periodic rainfall.

Reference graph

Works this paper leans on

38 extracted references · 38 canonical work pages · cited by 1 Pith paper

  1. [1]

    B. M. Arani, S. R. Carpenter, L. Lahti, E. H. Van Nes and M. Scheffer (2021), Exit time as a measure of ecological resilience. Science 372(6547), eaay4895

    work page 2021

  2. [2]

    A. Asch, M. Avery, A. Cortez and A. Scheel (2024), Slow passage through the Busse balloon– predicting steps on the Eckhaus staircase. European Journal of Applied Mathematics pp. 1–26

    work page 2024

  3. [3]

    M. E. Assessment (2003), Ecosystems and Human Well-being; A Framework for Assessment . Island Press

    work page 2003

  4. [4]

    Bastiaansen, P

    R. Bastiaansen, P. Carter and A. Doelman (2019), Stable planar vegetation stripe patterns on sloped terrain in dryland ecosystems. Nonlinearity 32(8), 2759

    work page 2019

  5. [5]

    Bastiaansen and A

    R. Bastiaansen and A. Doelman (2019), The dynamics of disappearing pulses in a singularly perturbed reaction–diffusion system with parameters that vary in time and space. Physica D: Nonlinear Phenomena 388, 45–72

    work page 2019

  6. [6]

    Bastiaansen, O

    R. Bastiaansen, O. Ja ¨ ıbi, V. Deblauwe, M. B. Eppinga, K. Siteur, E. Siero, S. Mermoz, A. Bouvet, A. Doelman, and M. Rietkerk (2018), Multistability of model and real dryland ecosystems through spatial self-organization. Proceedings of the National Academy of Sciences 115(44), 11256–11261

    work page 2018

  7. [7]

    Busse (1978), Non-linear properties of thermal convection

    F. Busse (1978), Non-linear properties of thermal convection. Reports on Progress in Physics 41(12), 1929

    work page 1978

  8. [8]

    Couwenberg and H

    J. Couwenberg and H. Joosten (2005), Self-organization in raised bog patterning: the origin of microtope zonation and mesotope diversity. Journal of Ecology pp. 1238–1248

    work page 2005

Show all 38 references
  1. [9]

    Deblauwe, N

    V. Deblauwe, N. Barbier, P. Couteron, O. Lejeune and J. Bogaert (2008), The global biogeo- graphy of semi-arid periodic vegetation patterns. Global Ecology and Biogeography 17(6), 715–723

    2008

  2. [10]

    Doedel, A

    E. Doedel, A. Champneys, T. Fairgrieve, Y. Kuznetsov, B. Sandstede, and X. Wang (1998), AUTO 97:Continuation and bifurcation software for ordinary differential equations. Technical report, Concordia University

    1998

  3. [11]

    Eigentler and J

    L. Eigentler and J. A. Sherratt (2020), Effects of precipitation intermittency on vegetation patterns in semi-arid landscapes. Physica D: Nonlinear Phenomena 405, 132396

    2020

  4. [12]

    Eppinga, P

    M. Eppinga, P. de Ruiter, M. Wassen and M. Rietkerk (2009), Nutrients and Hydrology Indicate the Driving Mechanisms of Peatland Surface Patterning. The American Naturalist 173(6), 803–818

    2009

  5. [13]

    D. R. Foster, G. A. King, P. H. Glaser and H. E. Wright (1983), Origin of string patterns in boreal peatlands. Nature 306(5940), 256–258

    1983

  6. [14]

    Gandhi, S

    P. Gandhi, S. Bonetti, S. Iams, A. Porporato and M. Silber (2020), A fast–slow model of banded vegetation pattern formation in drylands. Physica D: Nonlinear Phenomena 410, 132534

    2020

  7. [15]

    Gandhi, L

    P. Gandhi, L. Liu and M. Silber (2023), A pulsed-precipitation model of dryland vegetation pattern formation. SIAM Journal on Applied Dynamical Systems 22(2), 657–693

    2023

  8. [16]

    Gilad, J

    E. Gilad, J. von Hardenberg, A. Provenzale, M. Shachak and E. Meron (2004), Ecosystem Engineers: From Pattern Formation to Habitat Creation. Physical Review Letters 93, 1–4. 14

    2004

  9. [17]

    Grafke and E

    T. Grafke and E. Vanden-Eijnden (2019), Numerical computation of rare events via large deviation theory. Chaos: An Interdisciplinary Journal of Nonlinear Science 29(6)

    2019

  10. [18]

    Hille Ris Lambers, M

    R. Hille Ris Lambers, M. Rietkerk, F. V. D. Bosch, H. H. T. Prins and H. D. Kroon (2001), Vegetation Pattern Formation in Semi-Arid Grazing Systems. Ecology 82, 50–61

    2001

  11. [19]

    K¨ astner, R

    K. K¨ astner, R. C. Van de Vijsel, D. Caviedes-Voulli` eme and C. Hinz (2024), A scale-invariant method for quantifying the regularity of environmental spatial patterns. Ecological Complexity 60, 101104

    2024

  12. [20]

    C. A. Klausmeier (1999), Regular and irregular patterns in semiarid vegetation. Science 284(5421), 1826–1828

    1999

  13. [21]

    Kondo and T

    S. Kondo and T. Miura (2010), Reaction-diffusion model as a framework for understanding biological pattern formation. Science 329(5999), 1616–1620

    2010

  14. [22]

    G. J. Lord, C. E. Powell and T. Shardlow (2014), An introduction to computational stochastic PDEs. Cambridge University Press

    2014

  15. [23]

    Macfadyen (1950), Vegetation patterns in the semi-desert plains of British Somaliland

    W. Macfadyen (1950), Vegetation patterns in the semi-desert plains of British Somaliland. Geographical Journal 116, 199–211

    1950

  16. [24]

    Martinez-Garcia, C

    R. Martinez-Garcia, C. Cabal, J. M. Calabrese, E. Hern´ andez-Garc ´ ıa, C. E. Tarnita, C. L´ opez, and J. A. Bonachela (2023), Integrating theory and experiments to link local mechanisms and ecosystem-level consequences of vegetation patterns in drylands. Chaos, Solitons & Fr...

    2023

  17. [25]

    J. D. Murray (2003), Mathematical Biology II: Spatial Models and Biomedical Applications , Vol. 18 of Interdisciplinary Applied Mathematics . Springer New York

    2003

  18. [26]

    Rademacher, B

    J. Rademacher, B. Sandstede and A. Scheel (2007), Computing absolute and essential spectra using continuation. Physica D: Nonlinear Phenomena 229, 166–183

    2007

  19. [27]

    Rietkerk, R

    M. Rietkerk, R. Bastiaansen, S. Banerjee, J. van de Koppel, M. Baudena, and A. Doelman (2021), Evasion of tipping in complex systems through spatial pattern formation. Science 374(6564)

    2021

  20. [28]

    Scheffer, J

    M. Scheffer, J. Bascompte, W. Brock, V. Brovkin, S. Carpenter, V. Dakos, H. Held, E. van Nes, M. Rietkerk, and G. Sugihara (2009), Early-warning signals for critical transitions.Nature 461, 53–59

    2009

  21. [29]

    J. A. Sherratt (2013), History-dependent patterns of whole ecosystems. Ecological Complexity 14, 8–20

    2013

  22. [30]

    J. A. Sherratt and G. J. Lord (2007), Nonlinear dynamics and pattern bifurcations in a model for vegetation stripes in semi-arid environments. Theoretical population biology 71(1), 1–11

    2007

  23. [31]

    Siero (2018), Nonlocal grazing in patterned ecosystems

    E. Siero (2018), Nonlocal grazing in patterned ecosystems. Journal of Theoretical Biology 436, 64–71

    2018

  24. [32]

    Siteur, E

    K. Siteur, E. Siero, M. B. Eppinga, J. D. Rademacher, A. Doelman, and M. Rietkerk (2014), Beyond Turing: The response of patterned ecosystems to environmental change. Ecological Complexity 20, 81–96

    2014

  25. [33]

    Uecker (2022), Continuation and bifurcation in nonlinear PDEs–Algorithms, applications, and experiments

    H. Uecker (2022), Continuation and bifurcation in nonlinear PDEs–Algorithms, applications, and experiments. Jahresbericht der Deutschen Mathematiker-Vereinigung pp. 1–38. 15

    2022

  26. [34]

    Uecker, D

    H. Uecker, D. Wetzel and J. D. Rademacher (2014), pde2path-A Matlab package for continu- ation and bifurcation in 2D elliptic systems. Numerical Mathematics: Theory, Methods and Applications 7(1), 58–106

    2014

  27. [35]

    van de Koppel, J

    J. van de Koppel, J. C. Gascoigne, G. Theraulaz, M. Rietkerk, W. M. Mooij, and P. M. Herman (2008), Experimental evidence for spatial self-organization and its emergent effects in mussel bed ecosystems. Science 322(5902), 739–742

    2008

  28. [36]

    van de Koppel, M

    J. van de Koppel, M. Rietkerk, N. Dankers and P. Herman (2005), Scale-dependent feedback and regular spatial patterns in young mussel beds. The American naturalist 165(3), E66–77

    2005

  29. [37]

    van der Stelt, A

    S. van der Stelt, A. Doelman, G. Hek and J. D. Rademacher (2013), Rise and fall of periodic patterns for a generalized Klausmeier–Gray–Scott model. Journal of Nonlinear Science 23, 39–95

    2013

  30. [38]

    Vi˜ nals, E

    J. Vi˜ nals, E. Hern´ andez-Garc ´ ıa, M. San Miguel and R. Toral (1991), Numerical study of the dynamical aspects of pattern selection in the stochastic Swift-Hohenberg equation in one dimension. Physical Review A 44(2), 1123. A Appendix A.1 Computation of the most unstable m...

    1991

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.