The loss of biodiversity is one of the most acute global problems. Nowhere in the world are species more at risk of extinction than on islands. Islands represent only 3.2% of the earth’s land area, yet they contain a disproportionately higher amount of threatened and extinct biota compared to continents. The primary cause of extinction has been linked to introduced species, which occur on at least 80% of the world’s archipelagos. The compounding threats of habitat conversion, exploitation, pollution, and coastal hazards linked to climate change make conservation efforts appear complex and intangible. However, the eradication of invasive species from islands is an increasingly utilized conservation tool with a strong record of successes. Removing invasive species from islands has led to substantial conservation benefits and represents a key baseline activity to undertake to restore islands and increase resilience to global change. Consequently, national and local governments and conservation organizations are advancing eradication programs on islands, and there are ongoing developments in eradication technologies. Nonetheless, the global scale of threats to island biodiversity outweighs the resources available for conservation, and data gaps in the distribution of threatened species, and where they co-occurrence with threats, is lacking for most of the world’s islands.

In this dissertation, I fill major data gaps needed to address biodiversity loss on islands and identify islands where invasive species eradications can directly protect threatened species. I examine the distribution of globally threatened species, island characteristics, and threats from invasives, and analyze opportunities for conservation over short and long time frames in the context of invasive species management and global climate changes. Chapter one examines the global distribution of threatened island vertebrates, a well-studied group of taxa that are highly threatened, and whose conservation can subsequently benefit whole island ecosystems, including lesser known taxa. In Chapter 2, I further investigate the biogeography and conservation opportunities on islands for seabirds, one of the most threatened marine groups and one of the few groups that are almost completely reliant on islands to breed. Finally, in Chapter 3, I adapt existing methods in conservation priority setting to score and rank invasive mammal management opportunities on threatened seabird breeding islands, identifying where an eradication can be implemented and achieve the highest conservation benefits. In this chapter, I also examine the potential impact from climate change related coastal hazards and how flooding may impact the long-term success of conservation efforts. Together, these chapters provide insights into the biogeography of the most threatened island vertebrates and can be used to target conservation planning on the islands where biodiversity is in desperate need of conservation.

Freshwater ecosystems cover less than 0.01% of Earth's surface but are home to nearly one-quarter of all vertebrate diversity. Much of this diversity is concentrated in large tropical rivers, including the Amazon (961 species), Mekong (309 species), Congo (375 species), and the Ogooué River in Gabon (351 species), but these large rivers are threatened by proposed dam development. Globally, freshwater ecosystems are highly threatened and 36% of freshwater fishes are considered endangered, but efforts to protect freshwaters are poorly developed.

With over 400 species of fresh and brackish water fishes, Gabon, represents a freshwater biodiversity hotspot. But Gabon’s freshwater species are threatened by proposed dam development including a total of 38 potential hydropower dam sites across the country, and 28 in the Ogooué watershed. We know comparatively little about the biodiversity and biogeography in Gabon’s freshwater ecosystems, so I have used a combination of field work, taxonomy, and modeling to determine the distribution of freshwater biodiversity in Gabon. We developed a novel application of MaxEnt Modelling and a new metric, species pseudorichness index, to assess the impacts of proposed dam development and provide clear guidance for freshwater conservation in Gabon.

Seabirds are the most threatened group of marine animals, with 29% of species at some risk of extinction. Significant threats to seabirds occur on islands where they breed, but in many cases, effective island conservation can mitigate these threats. To guide island-based seabird conservation actions, we identified all islands with extant or extirpated populations of the 98 globally threatened seabird species recognized by the IUCN Red List and quantified the presence and extent of threatening invasive alien species, protected areas and human populations. We used these data to highlight feasible island conservation opportunities. We identified 1,362 threatened seabird breeding populations on 968 islands. Threatening invasive species and/or lack of adequate protection was identified on 803 of these islands (580 (59.9%) and 609 (63%) islands, respectively). Most identified islands were small (57% were <1km2), uninhabited (74%), and occurred in countries designated as high or middle income (96%). Collectively, these attributes make threatened seabird islands ideal for successful island conservation action. As a result, globally threatened seabirds are in a rare situation whereby some of the most intense threats are feasible to mitigate, providing a significant opportunity for important and effective seabird conservation.

The earth is losing biodiversity and ecosystem services due to anthropogenic impacts, and one way to mitigate this is to establish protected areas. Despite an increase in their global coverage, biodiversity is decreasing, while the world’s human footprint and population density are increasing. Effective coastal conservation requires a better understanding of how human activities on land may affect adjacent marine communities, but empirical work is lacking. My dissertation examines the synergistic benefits of co-locating protected areas and examines connections between terrestrial protection and nearshore marine protection by studying seagrass health, (1) a large-scale, multi-island field study of how human use of terrestrial watersheds directly and indirectly affect recipient marine communities, (2) a field experiment measuring the influence of the flux of sediments from human-impacted vs. non-impacted terrestrial watersheds on three adjacent nearshore island marine communities, (3) an exploration of the ecosystem services provided by seagrass beds, specifically small-scale fisheries. Integrating terrestrial with marine protection can improve coastal conservation efficiency. Using the tropical seagrass system, my dissertation provides evidence that terrestrial protection is more important than marine protection for nearshore marine health. Proper management of shallow-water marine resources should take into account stewardship of the adjacent watersheds and coastlines.

The biosphere is composed of a mosaic of ecosystems that are connected to one another through biological and physical processes. Animals that migrate over large spatial scales can play a significant role in vectoring nutrients and energy from one ecosystem to another. For example, seabirds forage in pelagic ecosystems on fish, krill, and squid and then migrate thousands of kilometers to breeding islands where they deposit large concentrations of nutrient-rich feces, or guano. Over sixty years ago, G.E. Hutchinson hypothesized that nutrients from seabird guano could increase primary productivity in recipient marine ecosystems. The nutrients in seabird guano have been widely documented to dramatically alter recipient terrestrial and freshwater ecosystems, but the influence of guano on nearshore marine ecosystems remains elusive. In the following dissertation, I assess the impact of seabird guano on one of the most biodiverse marine ecosystems on earth; coral reefs. First, I examine the influence of seabird guano subsidies on coral reef benthic community composition and fish biomass across one of the largest marine protected areas on earth, the Pacific Remote Islands Marine National Monument. I demonstrate that coral reefs next to islands with high densities of seabird guano have more herbivorous fishes and significantly different benthic communities. Next, I use stable isotope analysis and benthic percent cover data to evaluate the nitrogen-footprint from seabirds on coral reef macroalgae adjacent to Rose Atoll, American Samoa. I demonstrate that algae adjacent to Rose Island, which is home to a large seabird colony, is enriched in delta15N. However, algal ecological responses to seabirds are taxon-specific: fleshy algal cover increases with increasing proximity to Rose Island, whereas calcareous algal cover decreases with increasing proximity to Rose Island. Finally, I assess relative differences in seawater nutrients and algal nitrogen isotope values in the complex, anthropogenically-modified coral reefs of Oahu, Hawaii to see how seabird community subsidies may alter coral reefs in pristine vs. impacted Pacific islands. Even in the midst of anthropogenic fishing and nutrient enrichment, I demonstrate that islets with large seabird colonies have greater dissolved phosphate values and higher algal delta15N compared to islets with small seabird colonies. Together, these three data chapters are the first test of Hutchinson’s local enrichment hypothesis in coral reef ecosystems, and they provide strong evidence that seabirds significantly influence coral reef ecology and biogeochemistry via cross-ecosystem nutrient subsidies. Coral reefs and seabirds are two of the most threatened marine communities on the planet, but they are currently managed in isolation. The results of this research indicate that seabirds should no longer be ignored as a significant source of nutrients on coral reefs. Instead, seabird and coral reef conservation should be integrated in order to maximize efficient and potentially synergistic management of these two globally-imperiled marine communities.