Award: OCE-1131616

Habitat loss is a major driver of biodiversity decline. In both marine and terrestrial systems, habitat loss, and the fragmentation (i.e. breaking apart) of habitat that accompanies loss, may have strong effects on the diversity of animal communities as well as the abundance and survival of many species. In terrestrial systems, particularly fragmented forests, one challenge is discerning the relative effects of habitat loss vs. elements of landscape context (e.g., productivity, matrix and vegetation type, and patch configuration) on biodiversity. Elements of landscape context often covary with landscape structure and may strongly influence relationships between habitat loss and biodiversity, but there are few opportunities to control for these effects. Additionally, the vast majority of studies on the effects of habitat fragmentation take place in systems in which the fragmentation and loss already have taken place, rather than in experimental systems in which these processes can be controlled. I used seagrass as an experimental model system to test whether an important element of seagrass landscape context, structural complexity, influences the effects of habitat loss on biodiversity, abundance, and survival of seagrass epifauna (small invertebrate and vertebrate animals that use seagrass as a habitat). Seagrasses are submerged marine plants that form critical habitats in shallow coastal waters, but also are highly susceptible to human disturbances that fragment these habitats. The major hypothesis addressed by my work was that structurally complex landscapes would provide more refuge, living space, and other resources for animals, allowing more seagrass habitat to be lost before major declines in biodiversity and abundance occur. To control structural complexity, my team of students and I created 30 small landscapes of artificial seagrass and deployed them in San Diego Bay to be colonized by small animals. Landscapes were 2 meters x 2 meters in size and each was made from 64 small "modules" that could be detached to form different levels of habitat loss. They were deployed in shallow water near existing, naturally occurring seagrass habitat. After colonization (about one month), we actively fragmented these experimental landscapes to form 10 levels of habitat loss. Landscapes varied both in the amount of habitat loss and the level of structural complexity; and because we created these simulated landscapes, we could precisely control these often confounded variables. Specifically, we created a continuum of habitat loss (ten levels from 0 – 90%) for each of three levels of structural complexity. We found that increasing structural complexity promoted diversity and abundance, and altered the effects of habitat loss on epifaunal community structure and survival. Though effects were more consistent at one site than the other, overall, habitat loss resulted in decreased epifaunal diversity, abundance, and survival in sparse landscapes (low structural complexity), but these variables peaked at intermediate levels of habitat loss in dense landscapes (high structural complexity). This supported the hypothesis, though at a second site in quieter water in south San Diego Bay, effects of habitat loss and structural complexity were more variable. Peaks in biodiversity at intermediate levels of habitat loss suggested a strong effect of habitat edge on processes such as recruitment (the movement of animals, in larval or adult stages, to the landscapes), disturbance, and predation. Despite a strong influence of habitat loss on several indices of biodiversity, at all of the sites used in our experiment, community composition was influenced by structural complexity, but not habitat loss. Community composition is defined by the actual species that make up the community, as opposed to an index of biodiversity (such as the number of species). This suggests that habitat loss has strong effects on biodiversity and abundance, b...