Utilisation des isotopes stables pour déterminer la proportion de frugivorie et arthropodivorie chez le tamarin lion noir (Leontopithecus chrysopygus) en milieux fragmentés de la forêt Atlantique brésilienne

Abstract

Although the proportion of fruit consumption by primates can be properly assessed through behavioural observations, the same is not valid for prey consumption that could be underestimated due to the challenges of its quantitative observation in the field. Therefore, stable isotopes represent a promising alternative approach to determine the proportion of arthropodivory and frugivory in the diet of black lion tamarins (Leontopithecus chrysopygus), an endangered primate species endemic of the Brazilian Atlantic Forest. To evaluate the proportion of fruits and arthropods in the diet of black lion tamarins living in forest fragments of different quality, we built a mixing model using MixSIAR, complemented by a trophic position model. Then, we constructed generalized linear mixed models to assess the influence of resource availability and individual characteristics (sex and age) on these dietary proportions. Finally, we tested the influence of forest cover on habitat quality, i.e., resource availability. Models derived from stable isotopes revealed that black lion tamarins consume a higher proportion of arthropods than fruits compared to results derived from behavioural observations and literature. These proportions are similar between individuals in the same group regardless of their age and sex, attesting a strong group cohesion with synchronised behaviours. This species can shift its diet depending on the arthropod and fruit availability, increasing its arthropod consumption in areas with higher arthropod biomass and lower tree basal area. Interestingly, forest cover within a 3 km diameter is positively correlated with arthropod biomass. In conclusion, the black lion tamarin adapts its diet depending on the resource availability, reflecting an ecological and behavioural flexibility in response to habitat changes. Furthermore, our results support that scan sampling seems unsuitable for quantifying elusive and short behaviours such as arthropod consumption. Nevertheless, behavioural observations are essential to complement stable isotope analyses and understand the ecological value of behaviours.