Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)


Environmental Science

First Advisor

Thomas P. Husband


The fate of many tropical wildlife species depends on the way in which human-dominated landscapes are managed. Conservation strategies that incorporate agricultural landscapes are imperative for the survival of native fauna. Coffee is a major cash crop worldwide and is grown in tropical regions of the world that are often designated as biodiversity hotspots. Shade coffee, an agroforestry system in which coffee is grown under the shade of intentionally managed and/or native tree cover, has shown promise to provide habitat and support the diversity of bird, bat, and insect communities, but few studies have focused on non-volant mammals living in and around coffee farms. Further research is needed to understand ecosystem dynamics within coffee-dominated landscapes and to evaluate the value of this landscape as habitat for wildlife. The objectives of my research were to assess the mammal diversity within coffee-dominated landscapes, quantify the habitat parameters that are important for mammals, and based on my findings, develop guidance and suggested management practices for ways to enhance mammal diversity on coffee farms.

I hypothesize that increased vegetation diversity within coffee farms leads to an increase in mammalian diversity. Additionally, I hypothesize that native forests support the highest amount of mammal abundance and diversity and that as vegetation complexity decreases within the coffee farms, the abundance and diversity of mammals also decrease.

I assessed mammal diversity within coffee landscapes in Kodagu, India and Turriabla, Costa Rica. In India, I investigated the effects of native and non-native shade trees on mammal diversity and vegetation structure within coffee farms. Twenty farms, each with two 50- x 50-m trap grids, were sampled for five nights throughout the three rainfall zones in the region. Each farm also contained a grid with indirect sampling methods: track plates, hair traps, and camera traps.

In Costa Rica, I surveyed diversity and habitat preference of non-volant mammals across a broader coffee-forest landscape. Three 25-ha sites were selected that contained forest, shade coffee, and sun coffee habitats; and in two of the sites, a mix of other agricultural land uses were present. Each site was set with a 500- x 500-m trap grid, four camera traps, and four track plates. The sites were sampled in four sessions, each approximately two weeks in duration, totaling 46 sampling nights per site.

During the four-month study period in India, I captured 146 small mammals (129 individuals). Eleven species of mammals were detected, six species captured in small mammal traps and five detected from the track plates and camera traps. I found that neither abundance nor richness of small mammals was influenced by the composition of native or non-native trees within the site. On the plot level, small mammal abundances were higher at farms with higher amounts of low vegetation cover and basal area of shade trees, and lower numbers of shade trees within the farm; and species richness was positively associated with low vegetation cover. On a landscape level, the small mammal abundance increased as the distance to forest decreased. Although the amount of non-native trees was not related to mammal diversity, it was significantly related to vegetation structure and composition within the coffee farms. Overall, the coffee farms surveyed in India had relatively high levels of tree species richness and diversity but did not support a high diversity of mammals.

For the study in Costa Rica, I found that small mammals used a variety of agricultural habitats. I captured 1,258 mammals (597 individuals) during the seven-month study period. Sixteen species of mammals were found, thirteen were captured in the traps and camera trap yielded an additional three species. In general, I found the forest habitats had greater species richness and abundance of mammals than shade coffee, which in turn had more species and higher abundances than sun coffee habitats. Habitat type was significantly associated with mammal abundance and richness, but the distance to forest was not. Increased amounts of shade canopy and herbaceous ground cover within the habitats were shown to significantly increase the mammal abundance and richness for the study sites in Costa Rica. While there is no substitute for native forest, the abundance and richness of small mammals within shade coffee rivaled that of forest, whereas abundance and richness within sun coffee was much lower than both shade coffee and forest.

In the habitat preference analysis of five focal species found within the coffee-forest landscapes of Costa Rica, I found that all of the focal species preferred forest habitat over coffee habitats, except one species that had an equal preference for shade coffee and forest. Three species preferred shade coffee over sun coffee, while the remaining two species had no preference between the two coffee habitats.

My findings suggest that, although small mammals may be present in coffee habitats, most do not use shade coffee exclusively and may rely primarily upon forest habitat for survival. Small mammals may require forest tracts surrounding or intersecting coffee agriculture in addition to shade trees within the coffee farm. These habitat requirements should be included in conservation strategies for the promotion of biodiversity and sustainable agriculture. Our conservation strategies may need to be expanded to a broader-landscape scale that incorporates not only the management of shade trees and vegetation complexity within coffee farms, but also that includes forest habitats in the surrounding landscape.