Multi-site interaction turnover in flea–mammal networks from four continents: Application of zeta diversity concept and multi-site generalised dissimilarity modelling

KRASNOV, BORIS R.; KHOKHLOVA, IRINA S.; KIEFER, MATHIAS S.; KIEFER, DANIEL; LARESCHI, MARCELA; MATTHEE, SONJA; SANCHEZ, JULIANA P.; SHENBROT, GEORGY I.; STANKO, MICHAL; VAN DER MESCHT, LUTHER

ECOLOGICAL ENTOMOLOGY

WILEY-BLACKWELL PUBLISHING, INC

Año: 2023

0307-6946

We studied patterns of changes in host–flea interactions measured as total turnover (TT) which can be partitioned into components, namely species turnover (ST), interaction rewiring (RW), and mixed turnover (MX) in networks from Europe, Asia, Africa, and South America, applying a multi-site interaction turnover metric. We also searched for environmental drivers of TT and its components. We asked whether (a) different components contribute differently to TT in rare versus common interactions (in terms of frequency of interaction occurrence); (b) relative roles of turnover components for rare and common interactions differ between continents; and (c) the environmental drivers of interaction turnover differ between turnover components, rare and common interactions, and/or continental networks. Between-network dissimilarity of interactions increased with an increase in the number of compared networks. Pure ST contributed the most to the turnover of rare interactions, whereas the turnover of common interactions was predominated by MX. The effects of environmental factors, interaction richness, and spatial distance on TT and its components differed between continental networks, turnover components, and rare versus common interactions. Climate and vegetation exerted the strongest effects on (a) ST for rare (except Asia) and, to a lesser degree, common (South America) interactions, (b) RW for both rare and common interactions in Europe/Asia, and (c) MX for both rare and common interactions (except Africa). Interaction richness and spatial distance mainly influenced ST. We conclude that the patterns of interaction turnover and its components were geographically invariant and did not depend on the identity of the interactors, whereas the drivers of the turnover differed between continental networks because of species-specific responses to the environment.