Connectivity, non-random extinction and ecosystem function in experimental metacommunities

• Published in: Ecology letters 2010-05, Vol.13 (5), p.543-552
• Main Authors: Staddon, Philip, Lindo, Zoë, Crittenden, Peter D, Gilbert, Francis, Gonzalez, Andrew
• Format: Article
• Language: English
• Subjects: Abundance; Animal and plant ecology; Animal, plant and microbial ecology; Animals; Arthropods - metabolism; Biodiversity; Biological and medical sciences; Carbon - metabolism; Carbon Dioxide - metabolism; Connectivity; dispersal; Ecosystem; Ecosystem biology; ecosystem function; Ecosystems; extinction; Extinction, Biological; fragmentation; Fundamental and applied biological sciences. Psychology; General aspects; metacommunity; Models, Theoretical; Nitrogen - metabolism; trophic cascade
• ISSN: 1461-023X; 1461-0248
• DOI: 10.1111/j.1461-0248.2010.01450.x

Ecology Letters (2010) 13: 543-552 The spatial insurance hypothesis indicates that connectivity is an important attribute of natural ecosystems that sustains both biodiversity and ecosystem function. We tested the hypothesis by measuring the impact of manipulating connectivity in experimental metacommunties of a natural and diverse microecosystem. Isolation led to the extinction of large-bodied apex predators, subsequently followed by increases in prey species abundance. This trophic cascade was associated with significantly altered carbon and nitrogen fluxes in fragmented treatments. The ecosystem impacts were characteristic of a function debt because they persisted for several generations after the initial loss of connectivity. Local extinctions and disruption of ecosystem processes were mitigated, and even reversed, by the presence of corridors in the connected metacommunities, although these beneficial effects were unexpectedly delayed. We hypothesized that corridors maintained grazer movement between fragments, which enhanced microbial activity, and decomposition in comparison to isolated fragments. Our results indicate that knowledge of habitat connectivity and spatial processes is essential to understand the magnitude and timing of ecosystem perturbation in fragmented landscapes.