Evaluating island recovery following invasive species removal and seabird restoration

The first component of my dissertation synthesizes ecosystem recovery from disturbances and restoration of seabirds more specifically. The global analysis reviewed 240 studies examining recovery from major disturbances. It provided evidence that ecosystems can recover from major perturbations on timescales of decades to half-centuries. Moreover, I examined worldwide seabird restoration projects, and evaluated their successes and failures. Long-term monetary funding, life history research, and pilot studies are the most important facets of a successful restoration project.;Examining recovery from a different angle, I conducted field research on the Cook Strait islands in New Zealand to ask whether or not rodent eradication is sufficient to restore ecosystems and whether or not seabird restoration can help to speed recovery. Eradication has done little to restore island ecosystems but the results showed some promise for seabird restoration speeding recovery. However, it was still unclear whether or not islands have had sufficient time to recover, or whether they may be entrained in an alternate state in absence of seabirds. The fourth component of my research is a chronosequence that addressed that question for islands off New Zealand's North Island. The study had nine islands that have had from 12--22 years to recover following kiore (R. exulans) eradication. Ecosystem variables will take decades to recover even after centuries-long rat invasion. The fifth component of my research is a model that focuses on the newly restored diving petrels to Mana Island. I used population viability analysis (PVA) to project both bird and nutrient recovery into the future. The models showed that the colony is growing but that the growth is dependent on attracting new immigrants to breed in addition to in-situ production of the restored individuals. The risk of colony extinction would be imminent without such immigration but is low when immigration is accounted for. Moreover, the colony may recover its soil nutrient function in as little as five years but it will take 240 years for it to reach densities of never-invaded islands.;Together, this dissertation provides much hope that the important first step of full island recovery can happen quickly and set the stage for slower response variables such as species interactions and species composition.