Fixed-Point Abundances and the Onset of Extinctions in Multispecies Lotka-Volterra Ecosystems
Abstract
I investigate the average and variance of species abundance distributions for fixed-point solutions to generalized Lotka-Volterra ecosystems. Two different analytical methods are constructed. First, I build up on the perturbative Random Matrix Theory approach developed in my earlier work [Phys. Rev. E. 113, L062202 (2026)], and formalize this method diagrammatically. The abundance average is given by a sum of planar diagrams with non-crossing vertex pairings which can be encoded as Dyck paths. This allows me to express the combinatorial factor multiplying perturbative terms at each order as Catalan and Narayana numbers, following which the abundance average is given by the latter's generating functions. In this way, the infinite perturbation expansion is exactly resummed in a closed-form formula. I have not been able to find a similar resummation for the abundance variance, however diagrammatic Feynman rules are expressed, following which the perturbative series can be calculated to any desirable order. Second, I do the same calculations using the cavity method, which gives closed-form formulas for both the abundance average and variance. For the average, the cavity formula gives the exact same result as the perturbative approach when the average interspecies interaction vanishes. When this is not the case, the two results are functionally different, however they are hardly distinguishable in their parametric range of validity. Comparison with numerical data indicate that the cavity result is more accurate over a larger parameter range.
Related
- A unified fixed point approach for enriched bivariate mappings with applications to volterra systems and fractional-order chaos
- Stationary Measures and Boundary Correlations for Multispecies Half-Line ASEP
- Microscopic sorting in a multispecies exclusion process
- Fastest-Species Selection in Critical Multispecies Competition-Diffusion Systems via an Entropy Approach
- On a Multispecies Electrodiffusion Model in Higher-Dimensional Porous Media
BibTeX
@misc{jacquod2026fixedpoint,
title = {{Fixed-Point Abundances and the Onset of Extinctions in Multispecies Lotka-Volterra Ecosystems}},
author = {Philippe Jacquod},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.22183},
eprint = {2610.22183},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.22183},
}
RIS
TY - UNPB
TI - Fixed-Point Abundances and the Onset of Extinctions in Multispecies Lotka-Volterra Ecosystems
AU - Philippe Jacquod
PY - 2026
DA - 2026/10/15
PB - arXiv
DO - 10.48550/arxiv.2610.22183
UR - https://arxiv.org/abs/2610.22183
AB - I investigate the average and variance of species abundance distributions for fixed-point solutions to generalized Lotka-Volterra ecosystems. Two different analytical methods are constructed. First, I build up on the perturbative Random Matrix Theory approach developed in my earlier work [Phys. Rev. E. 113, L062202 (2026)], and formalize this method diagrammatically. The abundance average is given by a sum of planar diagrams with non-crossing vertex pairings which can be encoded as Dyck paths. This allows me to express the combinatorial factor multiplying perturbative terms at each order as Catalan and Narayana numbers, following which the abundance average is given by the latter's generating functions. In this way, the infinite perturbation expansion is exactly resummed in a closed-form formula. I have not been able to find a similar resummation for the abundance variance, however diagrammatic Feynman rules are expressed, following which the perturbative series can be calculated to any desirable order. Second, I do the same calculations using the cavity method, which gives closed-form formulas for both the abundance average and variance. For the average, the cavity formula gives the exact same result as the perturbative approach when the average interspecies interaction vanishes. When this is not the case, the two results are functionally different, however they are hardly distinguishable in their parametric range of validity. Comparison with numerical data indicate that the cavity result is more accurate over a larger parameter range.
ER -