The Effects of Seabirds, Rats, and Ecosystem Restoration on Invertebrate Food Webs
Burrowing seabirds that nest on islands transfer nutrients from the sea, disturb the soil through burrowing, damage tree foliage when landing, and thereby modify the surface litter. One of the greatest effects seabirds may have on their recipient ecosystems may be via the nutrient subsidies they transfer onto islands from the sea. How these nutrients effect their recipient ecosystems however, depends on many factors such as water availability. However, seabirds are in decline worldwide, as are their community- and ecosystem-level impacts, primarily due to invasive predatory mammals. Seabird islands are vulnerable to the invasion of predatory mammals such as rats, which can have lasting effects even after these pests are eradicated. Once these islands are restored and seabirds start to return the ecosystems can recover quickly, returning to a pre-disturbance state within as little as 20 years. However, legacy effects of the invasive mammals may occur meaning ecosystems may revert to alternate stable states. The direct and indirect effects of seabirds, their decline and recolonisation on ecosystems are inherently complex. I employed network analysis of invertebrate food webs, as a means of simplifying ecological complexity, to better understand the effects seabirds, their loss, and recolonization, may have on island invertebrate communities. I found that on rat-invaded islands the invertebrate food webs were smaller and less complex than on their seabird-dominated counterparts, likely due to the suppression of seabird derived nutrients and consequent effects on trophic cascades. There was also an interplay between nutrient subsidies and water availability, where invertebrate food webs were larger and more complex as litter water increased and soil C: N slightly decreased. When comparing a restored island to invaded islands and those never invaded I found that the restored island supported some areas that were virtually indistinguishable from an invaded island and it demonstrated strong environmental gradients indicative of a recovering island. Finally when comparing the family richness and missing common families between islands I found that the restored islands had a similar number of missing families to invaded islands and were missing more family groups than islands that had never been invaded when controlling for covariates. Seabird and rat effects on island ecosystems are manifested throughout entire food webs. As seabirds spread across restored islands the areas similar to invaded islands will become fewer as the island starts to fully resemble a burrowing seabird island ecosystem. A key finding was the resilience of the invertebrate food webs, which shrunk to a fraction of their full potential complexity during arid periods then reconstructed themselves with increased water availability. However, the invertebrate food webs were unable to reconstruct fully on restored islands due to the legacy effects of invasive mammal suppression. This may have had negative effects on the nutrient cycling of at least one of the restored islands. I finally conclude that more effort is needed to understand and integrate invertebrate communities into ecosystem restoration in the future.