The integrable Volterra system in the case of infinitely many species, either countable or uncountable.

2026

Read The integrable Volterra system in the case of infinitely many species, either countable or uncountable. (2026) 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-2601-15150.txt.

Introduction

A classic model of competing species gets pushed to an endless roster in a 2026 arXiv paper. The Volterra model dates to 1937 and tracks how species grow and compete. Two recent papers showed that an integrable version of the finite-species model is in fact maximally superintegrable.

That is a strong form of order for a dynamical system. The new work extends those results much further. It treats the case of infinitely many species. The count may be countable, like the whole numbers, or uncountable. The integrable structure is shown to survive that leap.

Catalog record

YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE65 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2601-15150

Browse the full library · About the Tower of Babel

Full text

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

The integrable Volterra system in the case of infinitely many species, either countable or uncountable. Abstract In the present paper we derive a further extension of the results contained in two recent articles, both published in Open Communications in Mathematical Physics, where it was shown that the integrable version of the N − N- species Volterra model, introduced by V. Volterra in 1937, is in fact maximally superintegrable. Here we point out that the superintegrability property applies as well to the case of infinitely many competing species, either countable or uncountable. Analytical and numerical results are given. . O. Ragnisco a , F.Zullo b a Dipartimento di Matematica e Fisica, Università degli Studi “Roma TRE" (retired) Via della Vasca Navale 84,00146 Roma e-mail: [email protected] b DICATAM, Università degli Studi di Brescia via Branze, 38 - 25123 Brescia, Italy & INFN, Milano Bicocca Piazza della Scienza 3, Milano, 20126, Italy e-mail: [email protected] Keywords: Volterra integrable model for infinitely many species, Hamiltonian formulation, Superintegrable systems. 1 Introduction The interest on dynamical systems of Lotka-Volterra type has been constantly growing all over the last century, starting from the pioneering results contained in the original articles [ 13 , 14 , 24 , 25 ] . Among the extremely rich literature on the subject (see for instance [ 7 ] , [ 23 ] , [ 16 ] , [ 3 ] , [ 12 ] , [ 4 ] , [ 8 ] , [ 22 ] , [ 9 ] , [ 5 ] , [ 19 ] , [ 6 ] ), one paper deserves in our opinion a very special attention. It is the long essay published by Vito Volterra in 1937 [ 25 ] , where some results of capital importance have been achieved: first, the original predator-prey model has been extended to an arbitrary number of species; second, the N- species model has been framed within the context of Lagrangian and Hamiltonian mechanics; finally, a completely integrable subcase has been identified. Relying on these outstanding results, the authors of the present paper performed some further hopefully non-trivial steps: namely, they discovered that the integrable version introduced in [ 25 ] was in fact maximally superintegrable [ 17 ] , [ 18 ] , and as such reducible to a Hamiltonian system with only one degree of freedom. Here we make yet a further step, investigating the extension to infinitely many competing species, either countable or uncountable, always staying within the integrable Hamiltonian framework, and discover that the maximal superintegrability property is preserved by such generalization. Various different cases of the interaction operator are discussed, and a few periodic numerical examples are displayed. We may thus assert that Infinity collapses to One . To this aim, in Section 2 we briefly recall the results obtained in [ 17 ] , [ 18 ] for the integrable Volterra model in the case of a finite number of species. The new developments are contained in Sections 3 and 4. In particular, in Section 3 we consider the case of a denumerable collection of species, and in Section 4 we grapple with the problem of uncountable many species. Both Sections consist of 3 subsections, where we give first a general outlook and then face two different cases, concerning respectively even and odd kernel of the interaction operators. In section 5 we first present and analyze some analytical examples and then discuss numerical implementations. In Section 6 we critically analyze the results obtained and point out some affordable ways to improve the description of the system. 2 The case of a finite number N of species The equations for the integrable N -species Volterra system have been introduced by V.Volterra in [ 25 ] , and thoroughly investigated in [ 17 ] , [ 18 ] : \frac{dN_{r}}{dt}=\epsilon_{r}N_{r}+\sum_{s\neq r=1}^{N}A_{rs}N_{r}N_{s}\quad\quad r=1,\ldots,N. (2.1) In the equation ( 2.1 ), \epsilon_{r} are the natural growth coefficients of each species and A_{rs} are interaction coefficients between species r and species s that account for the effects of encountering between two individuals. In the integrable N- species case, the N\times N interaction matrix A has the form [ 18 ] : A_{rs}=\epsilon_{r}\epsilon_{s}(B_{r}-B_{s}), (2.2) where all the \epsilon_{r} and the B_{r} are real non-zero numbers, and moreover B_{r}\neq B_{s} if r\neq s . The matrix A has always rank 2 , so {\mathcal{I}} , the Image of A , is 2 -dimensional and {\mathcal{K}} , its Kernel, is (N-2) -dimensional. Moreover {\mathcal{I}} and {\mathcal{K}} are two orthogonal subspaces with respect to the inner product (u,v)\equiv\sum_{j}u_{j}v_{j} and the N -dimensional Euclidean space {\mathbb{E}}_{N} can be written as the direct sum of them: {\mathbb{E}}_{N}={\mathcal{I}}\oplus{\mathcal{K}}. Also, it has been shown in [ 18 ] that {\mathcal{I}} is the linear span of two vectors, which are linearly independent because the coefficients B_{r} are all distinct, namely \epsilon and \eta , of components \epsilon_{r} and \eta_{r}=B_{r}\epsilon_{r} . The Kernel {\mathcal{K}} is naturally defined as the (N-2) -dimensional subspace of \mathbb{E}_{N} orthogonal to {\mathcal{I}} and is such that its generic vector k fulfils the linear conditions (k,\epsilon)=(k,\eta)=0. (2.3) On the other hand, the affine variety of equilibrium configurations {\mathcal{E}} is the set of elements z fulfilling (z,\epsilon)=0,~(z,\eta)=1. (2.4) It is convenient to rewrite the evolution equations ( 2.1 ) In terms of the logarithmic variables defined by y_{j}:=\log N_{j} , getting: \dot{y}_{j}=\epsilon_{j}+\sum_{l=1}^{N}A_{jl}\exp(y_{l})=\epsilon_{j}+(\eta_{j}\sum_{l=1}^{N}\epsilon_{l}-\epsilon_{j}\sum_{l=1}^{N}\eta_{l})\exp(y_{l}) (2.5) It is possible to represent any vector \in{\mathbb{E}}_{N} as a linear combination of a vector \in{\mathcal{I}} and a vector \in{\mathcal{K}} as follows: y_{n}=P\epsilon_{n}+Q\eta_{n}+k_{n}, (2.6) where in ( 2.6 ) k_{n} is a generic element k\in{\mathcal{K}} . 1 1 1 To be precise, with the term k_{n} we mean \sum_{k=1}^{N-2}R_{k}\tau_{m}^{(k)} , i.e. the decomposition of k_{n} along a given orthonormal basis. We adopt here this shorter symbol for easiness of notation. The evolution equations ( 2.5 ) take the form: \dot{P}\epsilon_{n}+\dot{Q}\eta_{n}+\dot{k}_{n}=\epsilon_{n}+\eta_{n}\sum_{k=1}^{N}\epsilon_{k}\exp(y_{k})-\epsilon_{n}\sum_{=1}^{N}\eta_{k}\exp(y_{k}) (2.7) entailing: \displaystyle\dot{P}=1-\sum_{n=1}^{N}\eta_{n}\exp(P\epsilon_{n}+Q\eta_{n}+k_{n}), (2.8) \displaystyle\dot{Q}=\sum_{n=1}^{N}\epsilon_{n}\exp(P\epsilon_{n}+Q\eta_{n}+k_{n}), (2.9) \displaystyle\dot{k}_{n}=0,\quad n=1...N-2 (2.10) we can rewrite the Hamilton equations in the canonical form: \dot{P}=-\frac{\partial{\mathcal{H}}}{\partial Q};~~\dot{Q}=\frac{\partial{\mathcal{H}}}{\partial P} (2.11) with the Hamiltonian explicitly given by: {\mathcal{H}}=\sum_{n=1}^{N}\exp(P\epsilon_{n}+Q\eta_{n}+k_{n})-Q. (2.12) For other properties of this dynamical system, like its Poisson structure, the corresponding associated N-2 Casimirs, existence of equilibrium positions, nature of the orbits and some numerical examples we refer the reader to [ 18 ] . In the next Section we are going to extend the above construction to the infinite chain, i.e. we let the numerosity of the species to be infinite. 3 About the discrete infinite case 3.1 General outlook In the discrete infinite case we extend the involved summation to the whole discrete line, taking care of convergence properties. Here we have to point out that, on one hand, we can look at the operator iA as a self-adjoint separable Hilbert-Schmidt operator [ 11 ] mapping the Hilbert space \ell^{2} of square summable sequences into itself, thanks to the the inner product (x,y)\doteq\sum_{n\in{\mathbb{Z}}}x_{n}y_{n} . As such, its spectrum consists of two opposite real eigenvalues, say \pm\omega , and a null eigenvalue of infinite multiplicity, the Hilbert space \ell^{2} being spanned by the two orthonormal eigenvectors associated to \pm\omega and by the orthogonal subspace Ker (A) of codimension 2. On the other hand, A acts nonlinearly on the space of sequences, according to the law \dot{y}=A\exp(y) , mapping

Read the complete text