Trophic redundancy, not species diversity, predicts food web structural robustness: edge-level bifurcation analysis across 86 ecosystems

Ecological robustness has long been associated with biodiversity, yet the structural mechanisms that confer resilience in food webs remain poorly understood.Here we introduce an edge-level analytical framework — Topostability — thatcomputes closed-form bifurcation thresholds from local triangle counts, enablingstructural triage at the level of individual trophic interactions rather than entirespecies. Applied to 86 trophic food webs from the GlobalWeb database spanningfreshwater, marine, and terrestrial ecosystems, we show that trophic redundancy —operationalised as the fraction of edges embedded in at least one triangular motif—predicts network-level structural fragility with Pearson r = −0.938 (p < 10−4,n = 86), a result that strengthens to r = −0.948 in species-resolved networks(n = 73). By contrast, species richness and edge density show negligible predictivepower (r = −0.22 and r = −0.13 respectively). Critically, 42% of analysed foodwebs exhibit a Fragility Index exceeding 80, suggesting that most natural ecosystems operate near structural fragility thresholds — a condition consistent with thefrequency and abruptness of observed trophic cascades. The framework provides, foreach trophic interaction, a closed-form collapse threshold that identifies which specific interactions are structurally at risk, independently of species-level abundancedata. These results position trophic redundancy, rather than biodiversity, as theprimary structural predictor of food web robustness, and suggest that interactiontopology — not species richness — is the appropriate unit of analysis for structuralresilience in ecological networks.