Bwindi Impenetrable National Park, Uganda, is a small protected area (331 km2) within which there is large climatic and altitudinal variation. Therefore we compared habitat types and forest composition between two locations to investigate differences that may influence ecological conditions for large mammals, including endangered mountain gorillas. We demonstrate that there is considerable intersite variation in habitat types and spatial and temporal availability of vegetation resources of which the most apparent are the differences in species composition of plants and fruit availability. Buhoma (the lower altitude site) has a greater diversity of plant species and higher tree and shrub densities, yet has a much lower density of herbaceous vegetation eaten by gorillas compared with Ruhija. Fruit availability is significantly higher throughout the year in Buhoma compared with Ruhija. Both sites exhibit seasonal variation in fruit availability although they do not follow the same seasonal pattern, perhaps because of inherent asynchronous phenological differences amongst individual tree species. The results of this study are important because distribution and abundance of both terrestrial herbaceous vegetation and fruit resources are believed to influence aspects of feeding ecology for large mammals.

Although many animal species consume herbaceous vegetation found in African tropical forests, little is known of the temporal and spatial availability of these plants. From September 2004 to August 2005 we conducted a study that quantified the temporal and spatial biomass availability of 20 species of herbs frequently consumed by endangered mountain gorillas at two locations (Buhoma and Ruhija) in Bwindi Impenetrable National Park, Uganda. In general, the biomass of herbs varied over the study period, but these changes were relatively small. For 12 of 18 and nine of 11 species in Buhoma and Ruhija, respectively, herb biomass differed significantly among habitat types. Of the nine species found in both locations, seven species had a higher biomass at Ruhija, one species had a higher biomass at Buhoma, and one species showed no difference. These results demonstrate that herb biomass varied little temporally but spatial differences in herb biomass were more pronounced. Future studies should investigate the variables that may influence herb phenological patterns such as rainfall, light, soil quality, previous disturbance regimes, and animal foraging and trampling damage.

Tropical Sedimentary Ancient DNA Preservation references in public databases. Of the remaining 7.7%, most of the data (93.0%) derive from Bacteria and Archaea, whereas only 0-5.8% are from Metazoa and 0-6.9% from Viridiplantae, in part due to unbalanced taxa representation in the reference data. The plant DNA record at ordinal level agrees well with local pollen data but resolves less diversity. Our animal DNA record reveals the presence of 41 native taxa (16 orders) including Afrotheria, Carnivora, and Ruminantia at Bwindi during the past 2200 years. Overall, we observe no decline in taxonomic richness with increasing age suggesting that several-thousand-year-old information on past biodiversity can be retrieved from tropical sediments. However, comprehensive genomic surveys of tropical biota need prioritization for sedimentary DNA to be a viable methodology for future tropical biodiversity studies.

No abstract

No abstract

No abstract

No abstract

Elephants are locally concentrated in Bwindi Impenetrable National Park. Vegetation damage attributable to elephants appears to be increasing and may result in the modification of the forest. We examined the implied selectivity of stem damage due to elephants. We followed 26.84 km of recent elephant trails and used 122 plots to document tree damage in relation to species, stem sizes and locations. Of 897 trees (DBH ≥2 cm), 542 (60.4%) were intact, 22 (2.5%) debarked, 274 (30.5%) toppled and 172 (19.2%) had broken branches. Small trees were more likely to be pushed over or have their branches broken, whereas large trees were more commonly debarked. The species most frequently selected for damage included mid‐successional species such as Newtonia buchananii, Myrianthus holstii and Chrysophyllum albidum. These species may be vulnerable to increasing elephant numbers. Our analyses using general linear models indicate that elephants are selective concerning where, how and what tree stems they damage. We found a higher incidence of elephant damage per‐tree stem in open areas than in more closed areas, suggesting feedback in which elephants maintain open habitats that may be conducive for other species such as mountain gorillas. More work is needed to better determine how changing elephant numbers may influence Bwindi's conservation values.