Ecology, habitat preference, and conservation of Neotropical non-volant mammal communities in Costa Rica's Caribbean lowlands

Biodiversity in the tropics is continually being threatened by anthropogenic disturbances such as habitat degradation and fragmentation. Unfortunately, conservation decisions and management of wildlife are difficult tasks to undertake without adequate scientific knowledge. In the tropics, little is known about the conservation value of secondary forests for mammals, the manner in which fragmentation can affect low-density populations, how mammalian assemblages have changed over time, and even some of the most basic ecology of organisms. Herein, I use multi-species and single-species approaches at different spatial scales to address these topics using the non-volant mammals in the Caribbean lowland rainforests of Costa Rica. I first tested the null hypothesis that primary and secondary forests have equivalent assemblages of non-volant mammals. I found no significant differences in the community composition of non-volant mammals in these two forest types. This community-based study led me to address the population biology of the collared peccary, one of the most common species in these faunal surveys. Using both my data and historical records, I established that this species has been greatly affected by anthropogenic disturbances in the area, and increased after the extirpation of white-lipped peccaries. My work at both the community and population level emphasized that the rodent communities in the Caribbean lowlands are at low densities, as represented by trap success. Fragmentation of tropical forests may therefore have particular consequences for mammalian communities. I thus performed a landscape level study with 15 fragments in Costa Rica's Caribbean lowlands. This work revealed that forest fragments showed differences in species diversity and relative abundance. Larger fragments (>9 km2) had higher relative abundance for all species. I also focused on the population biology of two species, the spiny pocket mouse, 9 km2) had higher relative abundance for all species. I also focused on the population biology of two species, the spiny pocket mouse, Heteromys desmarestianus and the arboreal vesper mouse, Nyctomys sumichrasti. My work on the former species included molecular genetic studies which revealed that the Caribbean lowlands have cryptic diversity that has not been previously explored. In overview, non-volant mammalian communities of the Caribbean lowlands represent a rich assemblage of organisms that are crucial to the health of tropical ecosystems. While anthropogenic disturbances are affecting these populations, a good understanding of the conservation value of secondary forests, how fragmentation can affect populations, and the ecology of organisms is crucial for making data-driven management and conservation decisions