pith. sign in

arxiv: 2511.04327 · v2 · pith:J55VVZTGnew · submitted 2025-11-06 · 🧬 q-bio.PE · nlin.AO· physics.bio-ph

Feasibility and Single Parameter Scaling of Extinctions in Large Ecological Communities

Pith reviewed 2026-05-21 20:23 UTC · model grok-4.3

classification 🧬 q-bio.PE nlin.AOphysics.bio-ph
keywords Lotka-Volterraecological communitiesfeasibilitystabilityspecies extinctionsrandom matrix theoryscaling law
0
0 comments X

The pith

In large ecological communities, feasibility is broken before stability, with extinctions following a single-parameter scaling law.

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

The paper uses random matrix theory to show that species abundance distributions are Gaussian at equilibrium in the weakly interacting regime of generalized Lotka-Volterra models. This Gaussian property implies that feasibility fails generically before stability when the number of species is large enough. An analytical expression is provided for the probability of n species going extinct, and a single-parameter scaling law is conjectured to describe the extinction process. These analytical results are verified through numerical simulations in a broad range of parameters. A sympathetic reader cares because this offers a predictive framework for understanding species loss in complex ecosystems without needing to track every interaction detail.

Core claim

Multispecies ecosystems modelled by generalized Lotka-Volterra equations exhibit stationary population abundances. Using standard methods of random matrix theory, distributions of species abundances are Gaussian at equilibrium in the weakly interacting regime. One consequence is that feasibility is generically broken before stability for large enough number of species. An analytical expression for the probability that n=0,1,2,... species go extinct is derived and a single-parameter scaling law is conjectured to govern species extinctions.

What carries the argument

Gaussian distributions of equilibrium species abundances obtained from random matrix theory, which determines when feasibility is lost relative to stability.

If this is right

  • Feasibility breaks before stability as the number of species increases.
  • The probability of exactly n species extinctions has a closed-form analytical expression.
  • Extinctions are controlled by a single scaling parameter.
  • These predictions hold across a wide range of system parameters as confirmed by simulations.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • This could mean that larger communities are inherently more prone to extinctions due to feasibility constraints alone.
  • The scaling law might be testable in empirical data from real ecosystems by measuring how extinction rates scale with community size.
  • Similar approaches could be applied to other dynamical systems beyond ecology to predict when feasibility fails.

Load-bearing premise

The system stays in the weakly interacting regime so that random matrix methods produce Gaussian abundance distributions at equilibrium.

What would settle it

Numerical simulations or analytical checks where, for large species counts, stability is lost at lower interaction strengths than feasibility, contradicting the generic breaking of feasibility first.

Figures

Figures reproduced from arXiv: 2511.04327 by Philippe Jacquod.

Figure 2
Figure 2. Figure 2: Normal (a) and semilog (b) plots of species abun [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 4
Figure 4. Figure 4: Normal (a) and semilog (b) plots of the distri [PITH_FULL_IMAGE:figures/full_fig_p004_4.png] view at source ↗
read the original abstract

Multispecies ecosystems modelled by generalized Lotka-Volterra equations exhibit stationary population abundances, where large number of species often coexist. Understanding the precise conditions under which this is at all feasible and what triggers species extinctions is a key, outstanding problem in theoretical ecology. Using standard methods of random matrix theory, I show that distributions of species abundances are Gaussian at equilibrium, in the weakly interacting regime. One consequence is that feasibility is generically broken before stability, for large enough number of species. I further derive an analytical expression for the probability that $n=0,1,2,...$ species go extinct and conjecture that a single-parameter scaling law governs species extinctions. These results are corroborated by numerical simulations in a wide range of system parameters.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript applies standard random matrix theory methods to the generalized Lotka-Volterra equations modeling large ecological communities. It establishes that species abundance distributions at equilibrium are Gaussian in the weakly interacting regime. This implies that feasibility is generically broken before stability for large numbers of species. The paper derives an analytical expression for the probability that a given number n of species go extinct and conjectures a single-parameter scaling law for species extinctions, which is supported by numerical simulations over a wide range of system parameters.

Significance. If the results are confirmed, this work advances the understanding of extinction dynamics in diverse ecosystems by linking random matrix theory to feasibility and stability. The analytical probability expression and the conjectured scaling law offer new tools for predicting biodiversity patterns in large communities. The corroboration with simulations enhances the reliability of the findings in theoretical ecology.

major comments (2)
  1. [Abstract] The statement that Gaussian distributions follow from standard random-matrix methods and that numerical simulations corroborate the scaling conjecture lacks the full derivation or error analysis, making it challenging to evaluate if post-hoc parameter choices influence the central claim that feasibility is broken before stability.
  2. [paragraph on distributions of species abundances] The Gaussian claim relies on the weakly interacting regime where the equilibrium solution remains well inside the positive orthant. However, since the probability that n species have negative abundances grows with N, the effective interaction matrix changes, which may invalidate the RMT-derived Gaussian form and the single-parameter scaling conjecture near the feasibility boundary.
minor comments (1)
  1. Clarify the notation for the interaction matrix and growth rates to ensure consistency throughout the manuscript.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their thoughtful and constructive comments on our manuscript. We address each major comment below and indicate the revisions we will make to strengthen the presentation and clarify the scope of our results.

read point-by-point responses
  1. Referee: [Abstract] The statement that Gaussian distributions follow from standard random-matrix methods and that numerical simulations corroborate the scaling conjecture lacks the full derivation or error analysis, making it challenging to evaluate if post-hoc parameter choices influence the central claim that feasibility is broken before stability.

    Authors: The derivation of the Gaussian form for species abundances is obtained by applying standard random-matrix techniques to the equilibrium condition of the generalized Lotka-Volterra system in the large-N limit; the explicit steps appear in the section titled 'Distributions of species abundances.' We have revised the abstract to include a concise outline of these steps and added a new paragraph in the main text that quantifies the approximation error via direct comparison of the analytic Gaussian to the empirical histograms over the full range of interaction strengths examined. The parameter sweeps reported in the figures were performed on a regular grid rather than selected post hoc; the revised manuscript now states this explicitly and includes the corresponding error bars. revision: yes

  2. Referee: [paragraph on distributions of species abundances] The Gaussian claim relies on the weakly interacting regime where the equilibrium solution remains well inside the positive orthant. However, since the probability that n species have negative abundances grows with N, the effective interaction matrix changes, which may invalidate the RMT-derived Gaussian form and the single-parameter scaling conjecture near the feasibility boundary.

    Authors: We agree that the Gaussian derivation is strictly valid only when the probability of negative abundances remains small. As N grows, this probability increases and extinctions modify the effective interaction matrix. In the revised manuscript we have added an explicit discussion of the regime of validity, together with additional simulations that approach the feasibility boundary from within the weakly interacting regime. These tests show that the single-parameter scaling continues to describe the data well until the fraction of extinctions becomes O(1), at which point the conjecture is no longer expected to hold; we now state this limitation clearly. revision: yes

Circularity Check

0 steps flagged

No circularity: standard RMT derivation with explicit conjecture and numerical checks

full rationale

The paper applies standard random-matrix methods to obtain Gaussian abundance distributions in the stated weakly interacting regime, then derives an analytical extinction probability expression directly from that distribution. The single-parameter scaling is explicitly labeled a conjecture and corroborated by independent numerical simulations across parameter ranges rather than obtained by fitting a subset and relabeling the output. No load-bearing self-citations, self-definitional steps, or ansatz smuggling are present in the provided derivation chain; the central claims remain independent of the target results and rest on externally standard RMT techniques plus direct simulation validation.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claims rest on the applicability of random-matrix theory to the interaction matrix in the weak-coupling limit and on the interpretation of negative abundances as extinctions.

free parameters (1)
  • interaction strength threshold
    The boundary separating the weakly interacting regime from stronger coupling is introduced to define where the Gaussian approximation holds.
axioms (1)
  • domain assumption Species abundances at equilibrium follow a Gaussian distribution when interactions are weak.
    Invoked via standard random-matrix methods applied to the generalized Lotka-Volterra equilibrium equations.

pith-pipeline@v0.9.0 · 5653 in / 1273 out tokens · 38638 ms · 2026-05-21T20:23:04.779980+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

51 extracted references · 51 canonical work pages

  1. [1]

    Wigner, Ann

    E.P. Wigner, Ann. Math.67, 325 (1958)

  2. [2]

    Dyson, J

    F.J. Dyson, J. Math. Phys.3, 140 (1962). [3]The Oxford Handbook of Random Matrix Theory, G. Ake- mann, J. Baik, and P. Di Francesco, Eds., Oxford Uni- versity Press, Oxford (2015)

  3. [3]

    Brody, J

    T.A. Brody, J. Flores, J.B. French, P.A. Mello, A. Pandey, and S.S.M. Wong, Rev. Mod. Phys.53385 (1981)

  4. [4]

    Zelevinsky, B.A

    V. Zelevinsky, B.A. Brown, N. Frazier, M. Horoi, Phys. Rep.276, 85 (1996)

  5. [5]

    Gor’kov and G.M

    L.P. Gor’kov and G.M. Eliashberg, Sov. Phys. JETP21, 940 (1965)

  6. [6]

    Haake, S

    F. Haake, S. Gnutzmann, and M. Ku´ s,Quantum Signa- tures of Chaos, 4th Ed., Springer (2018)

  7. [7]

    Montambaux, D

    G. Montambaux, D. Poilblanc, J. Bellissard, and C. Sire, Phys. Rev. Lett.70, 497 (1993)

  8. [8]

    Jacquod and D.L

    Ph. Jacquod and D.L. Shepelyansky, Phys. Rev. Lett. 79, 1837 (1997)

  9. [9]

    P. Kos, M. Ljubotina, and T. Prosen, Phys. Rev. X8, 021062 (2018)

  10. [10]

    Beenakker, Rev

    C.W.J. Beenakker, Rev. Mod. Phys.69, 731 (1997)

  11. [11]

    Flambaum, A.A

    V.V. Flambaum, A.A. Gribakin, G.F. Gribakina, and M.G. Kozlov, Phys. Rev. A50, 267 (1994). [13]Theoretical Ecology: Principles and Applications, edited by R. May and A.R. McLean (Oxford University Press, 2007)

  12. [12]

    Wootton, J

    J.T. Wootton, J. Sea Res.48, 157 (2002)

  13. [13]

    Werner and S.D

    E.E. Werner and S.D. Peacor, Ecology84, 1083 (2003)

  14. [14]

    Gardini, R

    L. Gardini, R. Lupini, and M.G. Messia, J. Math. Biol. 27, 259 (1989)

  15. [15]

    Fussmann, S.P

    G.F. Fussmann, S.P. Ellner, K.W. Shertzer, and N.G. Hairston, Science290, 1358 (2000)

  16. [16]

    McGehee, N

    E.A. McGehee, N. Schutt, D.A. Vasquez, and E.Peacock- L´ opez, Intl. J. Bif. and Chaos18, 2223 (2008)

  17. [17]

    May, Nature238, 413 (1972)

    R.M. May, Nature238, 413 (1972)

  18. [18]

    Rieger, J

    H. Rieger, J. Phys. A:Math. Gen.22, 3447 (1989)

  19. [19]

    Allesina and S

    S. Allesina and S. Tang, Nature483, 205 (2012)

  20. [20]

    Jansen and G.D

    V.A. Jansen and G.D. Kokkoris, Ecol. Lett.6, 498 (2003)

  21. [21]

    Fyodorov and B.A

    Y.V. Fyodorov and B.A. Khoruzhenko, Proc. Natl. Acad. Sc.113, 6827 (2016)

  22. [22]

    Grilli, M

    J. Grilli, M. Adorisio, S. Suweis, G. Barab´ as, J. Ba- navar, S. Allesina, and A. Maritan, Nature Com.8, 14389 (2017)

  23. [23]

    Fried, N.M

    Y. Fried, N.M. Shnerb, and D.A. Kessler, Phys. Rev. E 96, 012412 (2017)

  24. [25]

    Biroli, G

    G. Biroli, G. Bunin, and C. Cammarota, New J. Phys. 20, 083051 (2018)

  25. [26]

    Barbier, J.-F

    M. Barbier, J.-F. Arnoldi, G. Bunin, and M. Loreau, Proc. Proc. Natl. Acad. Sc.115, 2156 (2018)

  26. [27]

    Galla, Europhys

    T. Galla, Europhys. Lett.123, 48004 (2018)

  27. [28]

    Stone, Sci

    L. Stone, Sci. Rep.8, 8246 (2018)

  28. [29]

    V. Ros, F. Roy, G. Biroli, G. Bunin, and A.M. Turner, Phys. Rev. Lett.130, 257401 (2023)

  29. [30]

    Clenet, F

    M. Clenet, F. Massol, and J. Najim, J. Math. Biology 87, 13 (2023)

  30. [31]

    Mambuca, C

    A.M. Mambuca, C. Cammarota, and I. Neri, Phys. Rev. E105, 014305 (2022)

  31. [32]

    Jacquod and R

    Ph. Jacquod and R. Delabays, Chaos35, 091109 (2025)

  32. [33]

    Allesina and S

    S. Allesina and S. Tang, Popul. Ecol.57, 63 (2015)

  33. [34]

    Akjouj, M

    I. Akjouj, M. Barbier, M. Clenet, W. Hachem, M. Ma¨ ıda, F. Massol, J. Najim, and V.C. Tran, Proc. R. Soc. A480, 20230284 (2024)

  34. [35]

    MacArthur, Ecology36, 533 (1955)

    R.H. MacArthur, Ecology36, 533 (1955)

  35. [36]

    Nunney, Am

    L. Nunney, Am. Naturalist115, 639 (1980)

  36. [37]

    McCann, Nature405, 228 (2000)

    K.S. McCann, Nature405, 228 (2000)

  37. [38]

    A.Hatton, O.Mazzarisi, A

    I. A.Hatton, O.Mazzarisi, A. Altieri, and M. Smerlak, Science383, eadg8488 (2024)

  38. [39]

    Aguad´ e-Gorgori´ o, I

    G. Aguad´ e-Gorgori´ o, I. Lajaaiti, J.-F. Arnoldi, and S. K´ efi, Oikos e10980 (2025)

  39. [40]

    Roberts, Nature251, 607 (1974)

    A. Roberts, Nature251, 607 (1974)

  40. [41]

    Rozdilsky and L

    I.D. Rozdilsky and L. Stone, Ecol. Lett.4, 397(2001)

  41. [42]

    Stone, Nature Com.7, 12857 (2016)

    L. Stone, Nature Com.7, 12857 (2016)

  42. [43]

    Akjouj, J

    I. Akjouj, J. Najim, J Math Biol.85, 66 (2022)

  43. [44]

    Marcus, A.M

    S. Marcus, A.M. Turner, and G. Bunin, J. Th. Biology 608, 112141 (2025)

  44. [45]

    Lotka, J

    A.J. Lotka, J. Phys. Chem.14271 (1910)

  45. [46]

    Volterra, Mem

    V. Volterra, Mem. Acad. Lincei Roma.2, 31 (1926); en- glish translation :Animal Ecology, edited by R.N. Chap- man, McGraw-Hill (1931)

  46. [47]

    For a simple model with a random distribution ofA ij = ±1, the ratio of predator-prey pairs withA ij =−A ji is (1−γ)/2

  47. [48]

    F. Roy, M. Barbier, G. Biroli, and G. Bunin, PLoS Com- put Biol16, e1007827 (2020)

  48. [49]

    Sommers, A

    H.J. Sommers, A. Crisanti, H. Sompolinsky, and Y. Stein, Phys. Rev. Lett.60, 1895 (1988)

  49. [50]

    Durrett,Probability: Theory and Examples, 5th Ed., Cambridge University Press (2019)

    R. Durrett,Probability: Theory and Examples, 5th Ed., Cambridge University Press (2019). 6 Single Parameter Scaling of Species Extinction in Generalized Lotka-V olterra Systems Supplemental Material In this Supplemental Material section I briefly sketch the two Neumann expansions I use to solve Eq (3). The general method is the following. Inserting the Ne...

  50. [51]

    with⟨δA ij⟩= 0 and⟨δA ij δAkl⟩=σ 2(δikδjl +γδ ilδjk)/S

    First expansion The first Neumann expansion takesX=1andY=A=M+δA, withM ij =µ/S, andδAcontaining the zero-average fluctuations of interspecies interactions, i.e. with⟨δA ij⟩= 0 and⟨δA ij δAkl⟩=σ 2(δikδjl +γδ ilδjk)/S. One has N=κ X p X j (−1)p D [(M+δA) p]ij E ,(S11) All terms with even powers ofδAcontribute to the RMT average, as before, and forµ̸= 0, one...

  51. [52]

    Second expansion The second Neumann expansion takesX=1+MandY=δAin Eq. (S10). I obtain, instead of Eqs. (S4) and (S11), N= κ 1 +µ ∞X p=0 X j (−1)p Dh δA 1− M 1 +µ ip ij E ,(S15) where I used (1+M) −1 = (1−M/(1 +µ). Taking into account terms up top= 4, I obtain N=κ (1 +µ) −1 + (1 +µ) −2γσ 2 + [(1 +µ) −2 + (1 +µ) −3]γ2σ4 +O(σ 6, S−1).(S16) Remarkably, I reco...