Karen Beardsley, Richard Barnes

Forest Elephant Distributions in Relation to the Distance from Roads in Gabon

Estimating elephant populations in the rain forests of Gabon is extremely difficult due to the closed tree canopy. Since aerial surveys are impossible under these conditions, the distribution of elephants in Gabon and other central African countries have been estimated using dung dropping counts. After field transects were cut and data were collected, a gradient was observed indicating that elephant densities increase with the distance from the nearest road. Over Gabon as a whole, the distribution of people and roads accounts for 58% of the variance in dung-pile densities.

Using ArcInfo geographic information system, buffer zones were generated around the road network of Gabon and the equations derived from field sampling were used to estimate the dung density in each buffer zone. The forest elephant population was estimated to be 74,842 with asymmetric conference limits from 50,371 to 108,823. We are currently exploring the phenomonen of density gradients shown by other forest mammals. GIS is proving to be a powerful tool in spatial analyses of tropical rainforest communities and the interactions between human and wildlife populations.


Karen Beardsley
Division of Environmental Studies
University of California - Davis
Davis, CA 95616
Telephone: 916-752-0532
Fax: 916-752-3350
Email: kbeardsley@ucdavis.edu

Richard Barnes