La répartition des chimpanzés à Sebitoli (Parc National de Kibale, Ouganda): influence des facteurs naturels et anthropiques

Currently protected under the National Park status, Kibale forest (Western Uganda) was commercially logged in the 60’s. Human activities are frequent at the edge of the forest and the demographic density can reach 335 inhabitants/km2 in a five kilometres zone around the park (Hartter, 2010). However, about a thousand Eastern Chimpanzees (Pan troglodytes schweinfurthii) live in Kibale (795 km2), with a mean density of 2.2 individuals/km2, which is the highest chimpanzee density known in the world (Ugandan Wildlife Authority, 2005). Therefore, representing geographically the chimpanzee repartition in relation with human activities is a major issue for a better understanding of the stakes of their conservation as this topic is not emphasized in today’s scientific research. Sebitoli, in the northern part of Kibale National Park, is a privileged site to study the edge effect of the repartition of chimpanzees. A tarmac road cuts through the park, while tea and eucalyptus plantations are located at the forest edge as well as numerous crop gardens at the extreme north of the park. Most of crops are experiencing crop-raiding due to wild animals.We analysed GPS records and data collected during two years (February 2009 to February 2011) of chimpanzee habituation at Sebitoli using records of direct observations, vocalizations, and dungs. In addition, we obtained a topographical map that allowed us to digitalize the hydrographical system of Sebitoli. To describe Sebitoli’s landscape, we used a series of high-resolution images extracted from Google Earth Pro that we combined into a single mosaic image using Photoshop CS5. In order to analyse land cover, we conducted non-supervised classification on Landsat image (ETM+) downloaded through GLCF.  We also used Google Earth pro mosaic for an analysis in remote sensing (Idrisi Andes). Finally, using ArcGIS 9.3 we define 200 x 200 meters quadrats in the chimpanzees home range in which we quantify their presence in function of vegetation types, landscape features (river streams, altitude, slopes, transects) and anthropogenic factors (presence of plantations, gardens and houses).In addition, we used MaxEnt software to analyse the influence of environmental variables on chimpanzees’ repartition. We obtained a map representing the probability of chimpanzee’s occurrence regarding natural and anthropogenic factors. It also resulted in statistical curves that describe the weight of each variable in the construction of the chimpanzee’s probability model.We determine the core area of the chimpanzees’s community (around 20 km2 with a more frequented area of 4.5 km2). It appears that chimpanzees do not avoid areas on the forest edge, currently in contact with human activities. But they may concentrate in areas were plantations (tea and eucalyptus) as well as houses and gardens are less frequent. Also chimpanzees most likely occupy high altitude areas (around 1515 meters) and river banks. Chimpanzees use every types of vegetation: dense, scattered and degraded forest. Some areas of their home range seem to be more used than others, as the north side of the road. Also older females tend to occupy smaller core areas than males and their core area are located in either north or south of the road contrary to males who cross the road. Therefore it seems that females are more sedentary than males, especially when they have dependants.Moreover, this study shows that for a better understanding of the influence of landscape properties and plants repartition (food resources) at the forest edge in this area, we should care primarily about interactions between humans, chimpanzees, and the forest, particularly through the connection that the two species have with the landscape. A study of food availability and diversity will help us to describe chimpanzees repartition in function of food phenological status. Finally, studying the perception of nature that local populations have of their environment will lead to a better analysis of spatial and social stakes.