Ecological networks balance connectivity with flexibility to perturbations
Abstract
Food webs become progressively sparser as species richness increases, yet the number of interactions grows faster than linearly, placing communities between the limiting expectations of constant degree and constant connectance. Similar scaling has also been documented in other interaction networks, but whether they share a common architectural regime, and what general principle could generate it, remain unresolved. Here, we show that 1,533 ecological networks spanning different interaction types broadly converge on an intermediate link--species scaling regime, with an overall exponent of 1.43. We interpret this pattern through a thermodynamic trade-off in which increasing network connectivity enhances signal propagation while reducing the diversity of independent modes contributing to the network's response to perturbations. Optimizing this trade-off yields a scaling exponent of 1.45, closely matching the empirical estimate, while randomly adding or removing interactions generally reduces thermodynamic efficiency in empirical networks. These results support a propagation--response-diversity trade-off as a plausible systems-level principle underlying how ecological network complexity grows with diversity.
Related
- Local connectivity balance shapes population dynamics in random recurrent networks
- Growth in early infancy drives optimal brain functional connectivity which predicts cognitive flexibility in later childhood
- Two-State Effective Connectivity Predicts the effect of Anterior Cingulate Cortex Excitatory / Inhibitory Balance on Posterior Cingulate Cortex
- Higher cortical beta oscillatory power before balance perturbations is associated with worse balance in older adults with Parkinson's Disease, but not in age-matched older adults
- Ecological connectivity modelling with WebAssembly
BibTeX
@misc{gimenezromero2026ecological,
title = {{Ecological networks balance connectivity with flexibility to perturbations}},
author = {A. Gimenez-Romero and D. Oro and J. Bascompte and M. Genovart},
howpublished = {bioRxiv},
year = {2026},
month = jun,
doi = {10.64898/2026.06.12.731932},
url = {https://www.biorxiv.org/content/10.64898/2026.06.12.731932v2},
note = {CC BY},
}
RIS
TY - UNPB
TI - Ecological networks balance connectivity with flexibility to perturbations
AU - A. Gimenez-Romero
AU - D. Oro
AU - J. Bascompte
AU - M. Genovart
PY - 2026
DA - 2026/06/12
PB - bioRxiv
DO - 10.64898/2026.06.12.731932
UR - https://www.biorxiv.org/content/10.64898/2026.06.12.731932v2
AB - Food webs become progressively sparser as species richness increases, yet the number of interactions grows faster than linearly, placing communities between the limiting expectations of constant degree and constant connectance. Similar scaling has also been documented in other interaction networks, but whether they share a common architectural regime, and what general principle could generate it, remain unresolved. Here, we show that 1,533 ecological networks spanning different interaction types broadly converge on an intermediate link--species scaling regime, with an overall exponent of 1.43. We interpret this pattern through a thermodynamic trade-off in which increasing network connectivity enhances signal propagation while reducing the diversity of independent modes contributing to the network's response to perturbations. Optimizing this trade-off yields a scaling exponent of 1.45, closely matching the empirical estimate, while randomly adding or removing interactions generally reduces thermodynamic efficiency in empirical networks. These results support a propagation--response-diversity trade-off as a plausible systems-level principle underlying how ecological network complexity grows with diversity.
ER -