Space partitioning in wild, non-territorial mountain gorillas: the impact of food and neighbours

Abstract

In territorial species, the distribution of neighbours and food abundance play a crucial role in space use patterns but less is known about how and when neighbours use shared areas in non-territorial species. We investigated space partitioning in 10 groups of wild, non-territorial mountain gorillas (Gorilla beringei beringei). Using location data, we examined factors influencing daily movement decisions and calculated the per cent overlap of annual kernel home ranges and core areas among neighbours. We found that the probability that a group chose an area was positively influenced by both food availability and the previous use of that area by the group. Additionally, groups reduced their overall utilization of areas previously used by neighbouring groups. Lastly, groups used their core areas more exclusively than their home ranges. In sum, our results show that both foraging needs and avoidance of competition with neighbours determined the gorillas' daily movement decisions, which presumably lead to largely mutually exclusive core areas. Our research suggests that non-territorial species actively avoid neighbours to maintain core area exclusivity. Together, these findings contribute to our understanding of the costs and benefits of non-territoriality.

Keywords: Gorilla beringei beringei; home range overlap; intraspecific competition; movement decisions; space partitioning; territoriality.

Figure 1.

Annual kernel home ranges of the 10 mountain gorilla groups studied in 2012 and 2013 in Bwindi Impenetrable National Park, Uganda. Home range areas (90% fixed kernel density estimates) in the three general locations of the study groups are depicted in grey and shared areas are indicated by darker grey shading.

Figure 2.

Influence of (a) herbaceous food availability (kcal m⁻², based on herb biomass and nutritional content) and (b) previous use by the group on the probability of choosing a particular area (i.e. a 500 × 500 m grid cell) in Bwindi gorillas. The area of the circles indicates the fourth root of the number of observations. In (a), the largest circle corresponds to 1268 and the smallest circle corresponds to 30 observations, whereas in (b), the largest circle corresponds to 2111 and the smallest circle corresponds to three observations. The dashed and dotted lines indicate the fitted influence of the predictor on the response and its confidence intervals, respectively, with all other predictor variables in the model being at their average.

Figure 3.

Influence of the between-groups effect of previous use by all habituated neighbouring groups on the utilization of a chosen area (quantified as the distance travelled in a chosen 500 × 500 m grid cell) in Bwindi gorillas. The response variable was log-transformed. Between-groups variation is expressed as the mean of the previous use per group. Boxes depict quartiles with the median values indicated as horizontal lines and vertical lines show quantiles (2.5 and 97.5%). The dashed line indicates the fitted influence of the predictor on the response, with all other predictor variables in the model being at their average. The dotted lines depict bootstrapped 95% confidence intervals of the model.

Figure 4.

Annual home range and core area overlap of the Bwindi gorilla groups and herbaceous food availability of their core areas and the rest of their respective home ranges. (a) Per cent overlap of annual home ranges (90% kernel home range) and core areas (50% kernel home range). (b) Herbaceous food availability (kcal m⁻², based on herb biomass and nutritional content) of core areas (50% kernel home range) compared to the rest of the respective home ranges. Dashed lines connect data points from the same respective group. The high home range and core area overlap of three groups (Bu, Mi and Kah) may be due to two group fissions during the study period.