Ecological restoration promotes zooplankton network complexity in Mediterranean coastal lagoons

Abstract

Ecological recovery following restoration is typically evaluated using metrics based on species diversity and composition. However, increasing evidence suggests the success of long‐term ecological recovery increases when more complex attributes such as biotic interaction networks are targeted. In created lagoons, the influence of nearby habitats can generate early similarities in the community structure, but communities often diverge from surrounding water bodies at later successional stages. These changes have been attributed to the effect of biotic interactions, but few studies have tested this assumption. Here, we analyze the zooplankton community recovery after the creation of new lagoons in a Mediterranean coastal wetland using beta diversity approaches and mixed graphical models to infer interaction networks from abundance data. Increasing differences in the community structure between new and old lagoons were detected the second year after their creation. The overall interaction network was more complex in new than in old lagoons. Interestingly, the network structure in old lagoons increased its complexity during the third and fourth years after restoration. The creation of new lagoons with overall milder environmental conditions contributed to a greater differentiation of the zooplankton community structure between new and old lagoons. Moreover, our results suggest that the creation of a heterogeneous and more connected landscape can increase the interaction network in newly created and pre‐existing habitats, even if environmental conditions remain unchanged. We show how the inclusion of interaction networks for the monitoring of ecosystem recovery reflects unique facets of community complexity, otherwise overlooked when targeting diversity metrics alone.

Habitat restoration and the recovery of metacommunities

“…In the latter case, many pristine sites may be initially inaccessible, and thus, restoring more habitat creates corridors and allows these sites to be recolonised (Gawecka & Bascompte, 2021;Tilman et al, 1997). Differences in the structure and complexity of networks between restored and undisturbed sites have also been observed in restoration projects (Hernández-Carrasco et al, 2023;O'Connell et al, 2022;Ribeiro da Silva et al, 2015).…”

“…A powerful tool for studying ecological interaction is network analysis where nodes represent species and links connecting them represent ecological interactions (Bascompte & Jordano, 2014;Raimundo et al, 2018). When applied to restoration, this network approach has already been shown to provide insights that would have been overlooked if the focus were only on the species (Devoto et al, 2012;Gaiarsa & Bascompte, 2022;Hernández-Carrasco et al, 2023;Mittelman et al, 2021;O'Connell et al, 2022;Pocock et al, 2012).…”