Deciphering EcoFitness of Marine Ecosystems through eDNA Cross-trophic Interactions

Effective assessment of marine ecosystems increasingly requires approaches that go beyond static biodiversity indices to capture cross-trophic organization and dynamic stability based on network topology. In this study, we applied a multi-marker (12S, 18S, COI) eDNA metabarcoding approach to compare the ecological organization of two subtropical ecosystems, Mirs Bay and Daya Bay nearshore and marine habitats, representing contrasting environmental settings. Species co-occurrence networks were reconstructed to examine topological features such as degree distribution, modularity, centrality, and species-pair rank, and to relate those to ecological fitness. Two ecological allometric laws were then evaluated: (1) Taylor’s Power Law (TPL), which quantifies species aggregation; and (2) the Network Kleiber's Law (NKL) attributing relative species abundance (RSA) variance to network hyperconnectivity. Correlations among these ecological indicators and environmental factors were used to explore potential causal determinants of ecological organization. The two ecosystems exhibited strikingly different network architectures. Daya Bay supported higher biodiversity and a larger, more densely connected homogeneous network, while Mirs Bay showed lower diversity and richness but a more hierarchical and compartmentalized structure. TPL described both systems well ($R^2 >$ 0.86), with Mirs Bay showing stronger aggregation (nu=1.697) than Daya Bay (nu=1.460), and a more organized NKL pattern. The TPL exponent nu's sensitivity to several environmental is higher for more organized ecosystems, and increases with depth and average species biomass, suggesting its potential as a systemic ecofitness indicator of multitrophic organization. In contrast, no significant scaling relationship was found between RSA variance and network hyperconnectivity.