Community-ecosystem interactions control plant biodiversity change before and after mangrove restoration

Restoring biological diversity and ecosystem function requires understanding how introduced species interact with one another and their environments. The most prevalent and challenging scenarios involve multiple invasive species whose traits feed back through ecosystem processes. However, research into these systems often focuses on either community dynamics or ecosystem properties, rather than on their interactions, limiting understanding of what causes biodiversity changes before and after restoration. Leveraging insights from theory and management of single-species invasions driven by feedback between plant litter and germination success, we documented the structure of a disturbed mangrove ecosystem and tested causal hypotheses for community and ecosystem change both in microcosms and across the landscape. Before restoration, competing Schinus terebinthifolia, Casuarina spp., and Cupaniopsis anacardioides generated litter that facilitated the dominance of Cupaniopsis seedlings. After experimental restoration, seedling cover and richness increased only when removing invasive trees and their litter, supporting interacting community and ecosystem effects as the primary drivers of biodiversity change. Effective restoration of multiply invaded ecosystems is possible when simple interventions follow causal hypotheses supported by theoretical mechanisms.