Spatial and Temporal Resource Partitioning of Small Carnivores in the African Rainforest

Abstract

Although African rainforests harbour a high diversity of small carnivores, few studies have been conducted on these species' ecology and interspecific relations. We carried out a camera-trapping survey to examine habitat use and activity patterns of small carnivores in the Moukalaba–Doudou National Park, Gabon. The study area ( ~ 500 km 2 ) consists of various types of vegetation, including forest on dry soils, swamp forest, montane forest and savannah. We detected nine species of small carnivores in the study area. The seven most common carnivores were broadly classified into forest-interior species ( n = 3), savannah/forest-edge species ( n = 3) and aquatic-habitat species ( n = 3), in agreement with observations by other researchers. Occupancy analysis suggested further habitat separation within the small carnivore assemblage: among the savannah/forest-edge species, African civets, Civettictis civetta , more often used the forest edge and less frequently entered the savannah interior compared with Egyptian mongooses, Herpestes ichneumon , and rusty-spotted genets, Genetta maculata . Among the forest-interior species, black-legged mongooses, Bdeogale nigripes , were more closely associated with mature secondary dry forest than were long-nosed mongooses, Xenogale naso , and servaline genets, Genetta servalina . These two forest mongoose species, with similar body size and diet, exhibited different activity patterns. However, their habitat use and activity patterns were not affected by one another's presence, indicating that they had different preferences. Our results show that most pairs of small carnivores in the Moukalaba differ in either habitat use or time of activity, which may promote their coexistence across this region. This suggests that maintenance of habitat heterogeneity may be important for the conservation of these species. The relative proportion of small carnivores over space and time may reflect the degree of degradation of the forest; therefore, long-term monitoring by using camera-traps is highly recommended.